International Endodontic Journal PDF

doi:10.1111/j.1365-2591.2008.01446.
REVIEW
Effect of root canal irrigants on cervical dentine

permeability to hydrogen peroxide
P. Surapipongpuntr, W. Duangcharee, S. Kwangsamai & A. Ekka

Restorative Department, Faculty of Dentistry, Naresuan University, Phitsanulok, Thailand
Abstract various solutions, the diffusion of 30% H2O2 was

evaluated through each disc using a plastic-split
Surapipongpuntr P, Duangcharee W, Kwangsamai S,
chamber. H2O2 was applied to the inner-side chamber,
Ekka A. Effect of root canal irrigants on cervical dentine
while the outer-side chamber was filled with de-
permeability to hydrogen peroxide. International Endodontic
ionised water. After 30 min of application of H2O2,
Journal, 41, 821–827, 2008.
the solution in the outer-side chamber was collected
Aim To examine the effects of various root canal to determine the concentration of H2O2 using a
irrigants on cervical dentine permeability by monitor- spectrophotometer.
ing the diffusion of hydrogen peroxide (H2O2). Results The penetration of H2O2 through dentine in
Methodology Sixty cervical dentine discs were group E was significantly highest, followed by groups D,
prepared from human third molar teeth. After C, B and A respectively (one-way anova, P < 0.05).
removal of enamel and cementum, the outer dentine Conclusions Among the irrigants used, 17% EDTA
surface was etched with 17% ethylenediamine tetra- and 5% NaOCl had the greatest effect in increasing
acetic acid (EDTA) for 1 min. The dentine discs were dentinal permeability to H2O2.
randomly assigned to five groups according to the
Keywords: bleaching agent, dentine permeability,
irrigant used: A, saline solution; B, 2.5% sodium
root canal irrigants.
hypochlorite solution (NaOCl); C, 5% NaOCl; D, 17%
EDTA and 2.5% NaOCl; E, 17% EDTA and 5% NaOCl. Received 12 September 2007; accepted 29 April 2008
After irrigation on the inner dentine surface with the
attributed to the toxic nature of 30% H2O2 which

Introduction
diffused through the dentine (Fuss et al. 1989). It has
Hydrogen peroxide (H2O2) is used for intracoronal been postulated that H2O2 denatured dentine at the
bleaching of endodontically treated teeth (Baratieri cervical region (Lado et al. 1983), which further
et al. 1995). However, recent evidence suggests that induced a foreign body reaction. Other authors claimed
this procedure can cause significant complications. Free that the diffusion of H2O2 through dentinal tubules
radicals and active oxygen biologically released from irritated the periodontium, causing an inflammatory
H2O2 have been reported to possess bleaching effects resorptive process and external root resorption (Har-
and to damage human cells and tissues (Kashima- rington & Natkin 1979, Friedman et al. 1988). An
Tanaka et al. 2003). Cervical root resorption was increase in permeability of cervical dentine may influ-
ence the diffusion of H2O2 into the periodontal tissue,
leading to cervical root resorption.
Correspondence: Peraya Surapipongpuntr, Restorative Depart- During root canal treatment, various irrigants are
ment, Faculty of Dentistry, Naresuan University, Amphur
Muang, Phitsanulok 65000, Thailand (Tel.: 61 55 261 934;
used to enhance the efficacy of mechanical cleansing.
fax: 61 55 261 934; e-mail: perayap@nu.ac.th or puapichat@ Sodium hypochlorite (NaOCl), a well-known nonspecific
hotmail.com). proteolytic agent, is the most favoured endodontic
ª 2008 International Endodontic Journal International Endodontic Journal, 41, 821–827, 2008 821
Dentine permeability Surapipongpuntr et al.
irrigant owing to its tissue-dissolving, antibacterial and and polishing the inner dentine into a flat surface using
lubricant properties (Koskinen et al. 1980, Harrison & abrasive papers (numbers 500, 800 and 1000). The
Hand 1981). Furthermore, a combination of irrigants outer dentine was etched with 1 mL of 17% EDTA for
has been reported to improve the potential for cleansing 1 min to remove the smear layer (Crumpton et al.
(Svec & Harrison 1981). Ethylenediamine tetraacetic 2005) and then rinsed with 2 mL of de-ionised water.
acid (EDTA), a chelating agent, has been used to remove The specimens were randomly assigned to five groups
debris accumulating on the instrumented root canal (n = 12) according to the irrigant used: group A
walls (Hottel et al. 1999). Removal of smear layer has (control), physiological saline solution (4 mL for
been suggested to improve the permeation of medica- 2 min); group B, 2.5% NaOCl (4 mL for 2 min); group
ments into dentinal tubules and to improve the sealing C, 5% NaOCl (4 mL for 2 min); group D, 17% EDTA
of root filling materials to dentinal walls (Kennedy et al. (1 mL for 1 min) followed by 2.5% NaOCl (3 mL for
1986, Foster et al. 1993). Application of 15% or 17% 1 min); and group E, 17% EDTA (1 mL for 1 min)
EDTA solution was found to be most effective in followed by 5% NaOCl (3 mL for 1 min). Irrigation was
removing the smear layer both ex vivo and in vivo performed on the inner side of the dentine discs, whilst
(McComb & Smith 1975, Yoshida et al. 1995). More- all dentine discs were rinsed with 2 mL de-ionised
over, investigators have shown that rinsing with 1 mL water. The sample size per group was determined using
of EDTA for 1 min adequately eliminated the smear data from the pilot study. The value of the standardised
layer, opened the dentinal tubules, and provided a clean range was (maximum mean ) minimum mean)/
surface (Crumpton et al. 2005). sigma = (5932.8)12.3)/3208.2 = 1.84 (Woolson &
Dentine permeability may be altered after being Clarke 2002). Using the table provided by Kastenbaum
exposed to NaOCl or the combination of EDTA and et al. (1970), with a = 0.05 and power equal 0.9, the
NaOCl because of the proteolytic effect of NaOCl and group sample size was n = 12.
the chelating effect of EDTA. Previous study has
demonstrated that treatment with 5% NaOCl for 1 h
Hydrogen peroxide penetration test
depleted the dentine of organic compounds and signif-
icantly increased the permeability of dentine (Barbosa Each dentine disc was inserted between two plastic
et al. 1994). In addition, an application of 17% EDTA chambers modified from a previous report and sealed
caused a significantly greater increase in dentine with double O-rings (3 mm in diameter). The inner-side
permeability compared with 1% NaOCl (Carrasco et al. chamber was filled with 1 mL of 30% (W/W) H2O2,
2004). As the irrigation of a root canal is an important whereas the outer-side chamber was filled with 1 mL
step in root canal treatment, this study focused on the de-ionised water (Fig. 1). After 30 min of application of
alteration in the permeability of cervical dentine to H2O2, the solution in the outer-side chamber was
hydrogen peroxide after pre-treatments using various collected to evaluate the amount of diffused H2O2.
endodontic irrigants. Thus, the aim was to examine the
effects of the various root canal irrigants on cervical
dentine permeability by monitoring the diffusion of
H2O2. The hypothesis was that there would be a
difference in dentine permeability among groups.
Materials and methods
Dentine disc preparation

Sixty intact third molar teeth extracted for various
clinical reasons were used. Donor patients, aged
<25 years, gave informed consent for their teeth to
be used. Each tooth was sectioned mesio-distally into
two halves using a diamond blade (Isomet, Buehler,
IL, USA) under copious water. A dentine disc specimen
(1.00 ± 0.05 mm thick) was prepared from the buccal Figure 1 Diagram of the experimental setup for the H2O2
half of each tooth by removing enamel and cementum diffusion test.
822 International Endodontic Journal, 41, 821–827, 2008 ª 2008 International Endodontic Journal
Surapipongpuntr et al. Dentine permeability
Assay for H2O2

The concentration of H2O2 was assayed using the
xylenol orange technique (Jiang et al. 1990). In this
reaction, H2O2 oxidises ferrous (Fe2+) to ferric ion
(Fe3+) in the presence of sorbitol, which acts as a
catalyst. Ferric ion then forms a blue–purple complex
(kmax 560 nm) with xylenol orange. A 200 lL aliquot
of sample was added to 100 lL ferrous ammonium
sulphate (2.5 mmol L)1), prepared fresh daily, 100 lL
sorbitol (0.1 mol L)1), 100 lL sulphuric acid
(0.25 mol L)1) and 100 lL xylenol orange. Reverse
osmosis water was added to make up 1 mL. After Figure 2 Graph showing mean concentration of H2O2 diffus-
shaking the sample for 45 min, the absorbance at ing through cervical dentine discs in each group (n = 12 per
560 nm was determined and compared with a hydro- group).
gen peroxide standard curve, showing a relation
between absorbance and H2O2 concentration for a A, respectively. Statistical analysis of the data showed
dilution series. If the measured absorbance of the significant difference among groups (One-way anova,
sample was higher than the standard curve, the sample P < 0.05). Post hoc test (Dunnett’s T3) showed a
was diluted with de-ionised water and re-examined. significant difference for dentine permeability
to hydrogen peroxide in all pairs except for one pair
Scanning electron microscopic observations (C and D).
The inner surface of the dentine disc was observed by a

scanning electron microscope (SEM; LEO1455vp, LEO SEM observations
Electron Microscopy Ltd, Cambridge, UK). Dentine discs The results of SEM analysis of each group at ·5000
were fixed in 2.5% glutaraldehyde in phosphate- were as the followings.
buffered solution (pH 7.3) at 4 C for 2 h. After
dehydration through ascending concentrations of alco- Group A: irrigation with physiologic saline solution
hol (30, 50, 70 and 90% for 5 min and 100% for The specimen showed an amorphous layer of polishing
10 min), the specimens were dried by the carbon debris, a smear layer (Fig. 3).
dioxide critical-point technique (Polaron CPD7501,
Watford, UK) and coated with gold using the SPI-
Module sputter coater (Structure Probe, Inc., West
Chester, PA, USA).
Data analyses
The amounts of H2O2 diffused through dentine in all
groups were compared by one-way analysis of variance
(anova). A probability value of P < 0.05 was consid-
ered as significant.
Results
The amounts of H2O2, which diffused through dentine
discs, are shown in Fig. 2. All tested irrigants
produced an increase in dentine permeability to
H2O2 compared with the control (group A). The Figure 3 Group A: Irrigation with saline solution: the surface
permeability of dentine measured as H2O2 diffusion was covered with an amorphous smear layer. Bar: 1 lm,
was highest in group E, followed by group D, C, B and ·5000.
Figure 4 Group B: Irrigation with 2.5% NaOCl: some amor- Figure 6 Group D: Irrigation with 17% EDTA followed by
phous debris was removed and a number of dentinal tubules 2.5% NaOCl removed the smear layer and opened dentinal
were observed. Bar: 1 lm, ·5000. tubules. The collagen fibrils and secondary tubules on the
intertubular dentine were observed. Bar: 1 lm, ·5000.
Group B: irrigation with 2.5% NaOCl
The dentine surface was covered with the smear layer. The dentinal tubules were opened and the collagen
Some amorphous debris was removed and dentinal fibrils were exposed (Fig. 6).
tubules were observed (Fig. 4).
Group E: irrigation with 17% EDTA followed by 5% NaOCl
Group C: irrigation with 5% NaOCl This combination removed almost all collagen fibrils
Some superficial organic debris was removed, resulting similar to group D. The secondary tubules on the
in a smoother surface compared to group B. The intertubular dentine were presented and the peritubu-
dentinal tubules were partially obstructed by smear lar dentine was destroyed (Fig. 7).
layer (Fig. 5).
Discussion
Group D: irrigation with 17% EDTA followed by
2.5% NaOCl Irrigants used to improve the efficacy of mechanical
The combination of EDTA and NaOCl completely cleansing during root canal treatment may alter the
removed the smear layer covering the dentine surface. integrity of dentine and increase its permeability. The
Figure 5 Group C: Irrigation with 5% NaOCl produced a Figure 7 Group E: Irrigation with 17% EDTA followed by 5%
smoother surface compared with group B. Dentinal tubules NaOCl removed the collagen fibrils shown in group D. The
were incompletely occluded with the smear layer. Bar: 2 lm, secondary tubules on the intertubular dentine were revealed
·5000. and the peritubular dentine was destroyed. Bar: 1 lm, ·5000.
present results revealed the effects of different root dentine surfaces of group D (17% EDTA and 2.5%
canal irrigants on cervical dentine permeability to 30% NaOCl). Additionally, the secondary tubules on the
H2O2. The dentine permeability was significantly inter-tubular dentine and irregular inter-tubular struc-
highest in group E (17% EDTA and 5% NaOCl), ture were observed (Fig. 7). It may consider that
followed by group D, C, B and A, respectively. deproteinisation after demineralisation is dependent
In group A, physiological saline solution, which has on the concentration of NaOCl (Correr et al. 2006).
no demonstrated demineralising or tissue solvent There is no clear clinical consensus as to whether the
activity, cleaned the surface poorly. SEM images of smear layer should be removed before canal filling
dentine surfaces irrigated with this solution revealed an (Moss et al. 2001). The removal of smear layer may
amorphous layer of polishing debris and occluded improve the permeation of medicaments into dentinal
dentinal tubules. As a result, group A was significantly tubules, the penetration of sealer, and the surface
lower in permeability to 30% H2O2 than the other contact between filling material and the dentine wall
groups. (Kennedy et al. 1986, Foster et al. 1993). On the other
In group B and C, the action of the NaOCl solution on hand, it has been shown that the smear layer can act as
dentine permeability was examined. NaOCl signifi- a barrier, inhibiting bacterial colonisation on dentinal
cantly increased the permeability of the dentine to tubules (Drake et al. 1994). In the present findings, the
30% H2O2 compared with physiologic saline solution. combination of 17% EDTA and 5% NaOCl revealed the
In addition, an increase in dentine permeability seemed greatest alteration of dentinal surfaces and the greatest
to depend on the concentration of NaOCl. The amount increase in dentine permeability. It has been suggested
of diffused H2O2 was higher in group C (5% NaOCl) that the combined use of high concentration of sodium
compared with group B (2.5% NaOCl). This is likely hypochlorite and EDTA during root canal treatment
because of the proteolytic action of concentrated may affect negatively the integrity of the dentine,
sodium hypochlorite on the organic phase of dentine thereby allowing increased access of caustic fluids to
(Barbosa et al. 1994). SEM study of the dentine surface cervical vital tissues and consequently causes cervical
applied with the higher concentration of NaOCl root resorption. Therefore, the concentration of NaOCl
revealed enhanced cleanliness. However, NaOCl alone should be carefully considered, particularly in cases
was not able to remove the smear layer as reported where intra-coronal bleaching may be indicated. The
previously (Yamada et al. 1983). antibacterial and tissue dissolution actions of NaOCl
The effects of the combined use of 17% EDTA increases with its concentration (Yesilsoy et al. 1995,
followed by 2.5% or 5% NaOCl were evaluated in Spano et al. 2001), but the high concentration is
groups D and E. The dentine permeabilities in these accompanied by increased toxicity. Thus, the desirable
groups were significantly higher than those in groups concentration should demonstrate low toxicity, yet
treated with NaOCl alone. The use of EDTA resulted in with adequate antibacterial and tissue dissolving
a considerable increase in dentine permeability and effects. Additionally, previous report has concluded
corroborates previous findings (Carrasco et al. 2004). that 30% H2O2 caused alterations in the chemical
In that study, 17% EDTA significantly increased the structure of pulverised human dentine and cementum
permeability of cervical dentine to copper ion penetra- (Rotstein et al. 1992). The dentine irrigated with EDTA
tion compared with 1% NaOCl. It is possible that the and NaOCl might be more susceptible to hydrogen
mineral element in smear layer can protect the organic peroxide damage. However, in this study no negative
phase in dentine and as a result using NaOCl alone is control group was used to compare the permeability of
not able to remove the smear layer. After first rinsing, dentine before and after application of hydrogen
EDTA removes the mineral phase, thus exposing the peroxide; further study is required to investigate this
organic components lining the dentinal tubules. The issue.
subsequent deproteinisation then exposes the tubules Although bleaching discoloured teeth is usually
(Marshall et al. 2001) therefore increasing dentine performed on anterior teeth, molar teeth were selected
permeability. In this study, the combination of 17% for this study because dentine discs can be readily
EDTA with 5% NaOCl caused the greatest increase in prepared from the cervical part into similar thickness
dentine permeability to 30% H2O2. and area. Intact human third molar teeth from the
SEM images of the dentine surfaces in group E same age range (<25 years old) were used in order to
revealed the removal of smear layer and complete reduce the morphologic difference among groups.
removal of collagen fibrils, which was also evident in However, differences in dentine permeability of each
tooth might influence the results as shown in group C scanning electron microscopy (SEM) evaluation. Journal of
and D. In group D, dentine permeability was higher Dentistry 34, 454–9.
than that in group C but there was no statistical Crumpton BJ, Goodell GG, McClanaban SB (2005) Effects on
difference between them. The group of 12 teeth might smear layer and debris removal with varying volumes of
17% REDTA after rotary instrumentation. Journal of End-
be too small to reach definite statistical conclusions.
odontics 31, 536–8.
In the clinical situation, cervical dentine is covered
Drake DR, Wiemann AH, Rivera EM, Walton RE (1994)
with enamel or cementum, and consequently the Bacterial retention in canal walls in vitro. Journal of
permeability of both mineralised tissues influences the Endodontics 20, 78–82.
diffusion of H2O2. Additionally, it should be kept in mind Foster KH, Kulild JC, Weller RN (1993) Effect of smear layer
that the bleaching agent used in this study was in a removal on the diffusion of calcium hydroxide through
liquid form, which is suitable for the diffusion test. radicular dentin. Journal of Endodontics 19, 136–40.
However, the standard agent for internal bleaching is Friedman S, Rotstein I, Libfeld H, Stabholz A, Heling I (1988)
sodium perborate mixed with water or H2O2 that may Incidence of external root resorption and esthetic results in
reduce the acidity of H2O2 and any adverse effects on the 58 bleached pulpless teeth. Endodontics & Dental Traumatol-
biomechanical properties of the dentine (Chng et al. ogy 4, 23–6.
Fuss Z, Szajkis S, Tagger M (1989) Tubular permeability to
2002, Attin et al. 2003). Using the standard treatment
calcium hydroxide and to bleaching agents. Journal of
method, the amounts of H2O2 diffusing through dentinal
Endodontics 15, 362–4.
tubules may be less than those presented in this study. Harrington GW, Natkin E (1979) External resorption associ-
ated with bleaching of pulpless teeth. Journal of Endodontics
Conclusions 5, 344–8.
Harrison JW, Hand RE (1981) The effect of dilution and
Among the irrigants used, 17% EDTA and 5% NaOCl organic matter on the anti-bacterial property of 5.25%
had the greatest effect on increasing dentinal perme- sodium hypochlorite. Journal of Endodontics 7, 128–32.
ability to H2O2. Hottel TL, el-Refai NY, Jones JJ (1999) A comparison of the
effects of three chelating agents on the root canals of
extracted human teeth. Journal of Endodontics 25, 716–7.
Acknowledgements Jiang ZY, Woollard ACS, Wolff SP (1990) Hydrogen peroxide
production during experimental protein glycation. Federa-
This study was supported by a grant from the Research
tion of European Biochemical Societies Letters 268, 69–71.
Fund of Naresuan University, Faculty of Dentistry. Kashima-Tanaka M, Tsujimoto Y, Kawamoto K, Senda N, Ito K,
Yamazaki M (2003) Generation of free radicals and/or active
References oxygen by light or laser irradiation of hydrogen peroxide or
sodium hypochlorite. Journal of Endodontics 29, 141–3.
Attin T, Paque F, Ajam F, Lennon AM (2003) Review of the Kastenbaum M, Hoel DG, Bowman KO (1970) Sample size
current status of tooth whitening with the walking bleach requirements: One-way analysis of variance. Biometrika 57,
technique. International Endodontic Journal 36, 313–29. 421–30.
Baratieri LN, Ritter AV, Monteiro S Jr, Caldeira de Andrada Kennedy WA, Walker WA, Gough RW (1986) Smear layer
MA, Cardoso Vieira LC (1995) Nonvital tooth bleaching: removal effects on apical leakage. Journal of Endodontics 12,
guidlines for the clinician. Quintessence International 26, 21–7.
597–608. Koskinen KP, Stenvall H, Uitto VJ (1980) Dissolution of bovine
Barbosa SV, Safavi KE, Spangberg SW (1994) Influence of pulp tissue by endodontic solutions. Scandinavian Journal of
sodium hypochlorite on the permeability and structure of Dental Research 88, 406–11.
cervical human dentine. International Endodontic Journal 27, Lado EA, Stanley HR, Weisman MI (1983) Cervical resorption
309–12. in bleached teeth. Oral Surgery, Oral Medicine, and Oral
Carrasco LD, Pecora JD, Froner IC (2004) In vitro assessment Pathology 55, 78–80.
of dentinal permeability after the use of ultrasonic-activated Marshall GW, Yucel N, Balooch M, Kinney JH, Habelitz S,
irrigants in the pulp chamber before internal dental Marshall SJ (2001) Sodium hypochlorite alterations of
bleaching. Endodontics & Dental Traumatology 20, 164–8. dentin and dentin collagen. Surface Science 491, 444–55.
Chng H, Palamara JE, Messer HH (2002) Effect of hydrogen McComb D, Smith DC (1975) A preliminary scanning electron
peroxide and sodium perborate on biomechanical properties microscopic study of root canals after endodontic proce-
of human dentin. Journal of Endodontics 28, 62–7. dures. Journal of Endodontics 1, 238–42.
Correr G, Alonso R, Grando M, Borges A, Puppin-Rontani R Moss HD, Allemang JD, Johnson JD (2001) Philosophies and
(2006) Effect of sodium hypochlorite on primary dentin – A practices regarding the management of the endodontic
smear layer: results from two surveys. Journal of Endodontics Woolson RF, Clarke WR (2002) Statistical Methods for the
27, 537–9. Analysis of Biomedical Data, 2nd edn. New York: John Wiley
Rotstein I, Lehr Z, Gedalia I (1992) Effect of bleaching agents & Sons, Inc., pp. 393–7.
on inorganic components of human dentin and cementum. Yamada RS, Armas A, Goldman M, Lin PS (1983) A scanning
Journal of Endodontics 18, 290–3. electron microscopic comparison of a high volume final
Spano JC, Barbin EL, Santos TC, Guimaraes LF, Pecora JD flush with several irrigating solutions: Part 3. Journal of
(2001) Solvent action of sodium hypochlorite on bovine Endodontics 9, 137–42.
pulp and physico-chemical properties of resulting liquid. Yesilsoy C, Whitaker E, Cleveland D, Phillips E, Trope M (1995)
Brazilian Dental Journal 12, 154–7. Antimicrobial and toxic effects of established and potential
Svec TA, Harrison JW (1981) The effect of effervescence root canal irrigants. Journal of Endodontics 21, 513–5.
on debridement of the apical regions of root canals Yoshida T, Shibata T, Shinohara T, Gomyo S, Sekine I (1995)
in single-rooted teeth. Journal of Endodontics 7, 335– Clinical evaluation of the efficacy of EDTA solution as an
40. endodontic irrigant. Journal of Endodontics 21, 592–3.
doi:10.1111/j.1365-2591.2008.01428.x
Root canal morphology of mandibular permanent

molars at different ages
H. R. D. Peiris1, T. N. Pitakotuwage2, M. Takahashi1, K. Sasaki3 & E. Kanazawa3

1
Department of Anatomy and Physical Anthropology, Nihon University Graduate School of Dentistry at Matsudo, Matsudo, Japan;
2
District Hospital Madulkelle, Madulkelle, Sri Lanka; and 3Department of Anatomy and Physical Anthropology, Nihon University
School of Dentistry at Matsudo, Matsudo, Japan
Abstract Results In both first and second molars, develop-

mental stages of canal morphology amongst age groups
Peiris HRD, Pitakotuwage TN, Takahashi M, Sasaki K,
were significantly different (P < 0.0001). The preva-
Kanazawa E. Root canal morphology of mandibular perma-
lence of inter-canal communications was highly signif-
nent molars at different ages. International Endodontic Journal,
icantly different in the first (P < 0.0001) and less
41, 828–835, 2008.
significant in the second molar (P < 0.05). After
Aim To investigate differences in the root canal completion of the canal differentiation, the mesial roots
morphology of permanent mandibular molar teeth at of first molars had type IV and II canal forms. The
various ages. majority of the mesial roots of second molars had type I
Methodology Four hundred and eighty permanent and III canals. C-shaped canals were found in 3% of
mandibular first and second molars were examined. second molars.
First and second molars were divided into six and five Conclusions Mesial roots of first and second molars
groups, respectively, according to the age of the patient mostly had one large canal until 11 and 15 years of
at the time of extraction. Root canal morphology was age, respectively. In both molars, the canal system was
studied using a clearing technique. The canal mor- completely defined at 30–40 years. The prevalence of
phology of the mesial root was classified into three inter-canal communications was low at young and old
stages depending on its developmental pattern. When ages but high at intermediate ages.
the root canal system was completely differentiated, the
Keywords: age changes, inter-canal communica-
canal classification and the number of lateral canals
tions, permanent mandibular molars, root canal mor-
and inter-canal communications were recorded. Vert-
phology.
ucci’s classification was taken as the main reference.
Canal morphology was compared amongst age groups. Received 30 October 2007; accepted 25 March 2008
often cause extensive differentiation of the root canal

Introduction
system resulting in the development of separate canals
Root canal morphology changes as the teeth develop. and transverse connecting systems (Hess 1925). The
In general, roots in teeth from young individuals have differentiation of a simple root canal into a complex
single large canals. With age, deposition of secondary form occurs most commonly in roots that are flat or
dentine results in the formation of partitions, which which have external grooves (Hess 1925).
Although several investigations have recorded the
canal morphology of the permanent dentition (Walker
Correspondence: H. R. D. Peiris, Department of Anatomy and 1988a,b, Manning 1990a,b, Gulabivala et al. 2001,
Physical Anthropology, Nihon University Graduate School of 2002, Sert et al. 2004, Peiris et al. 2007), age changes
Dentistry at Matsudo, 2-870-1, Sakaecho-Nishi, Matsudo-shi,
Chiba, 271-8587, Japan (Tel.: 81 047 360 9318; fax: 81 047
in canal morphology has rarely been studied in detail.
360 9317; e-mail: roshan.peiris@nihon-u.ac.jp; rdpeiris@ Hess (1925) was the first to undertake a comprehensive
yahoo.com). investigation of the root canal system of the human
Peiris et al. Root canal morphology with age
permanent dentition. That report described the form approved by the Ethics Committee of the Nihon
and number of root canals in a European white University School of Dentistry at Matsudo and con-
population of different age groups ranging from 6 to formed to the provisions of the World Medical Associ-
55 years and concluded that age has an influence upon ation Declaration of Helsinki (revised Washington
the form and number of the root canals in different 2002). Only teeth, which could be verified as mandib-
tooth types. In another study, Thomas et al. (1993) ular permanent first and second molar teeth by crown
reported the canal anatomy of the mesio-buccal root of morphology, were included.
maxillary first molars at various ages using a radio- First and second molars were divided into six and five
graphic technique. They reported that differentiation of groups, respectively, according to the age of the patient
root canals commenced at an early age but that the at the time of extraction (Tables 1 and 2). Teeth were
rate of progression appeared to be variable. Unfortu- washed immediately after extraction and stored in 10%
nately, the limitations of radiographs in studying formalin until the collection was completed. They were
certain features of the root canal system are prone to boiled in 5% NaOH for 5 min and then cleaned with
a wide range of interpretation (Mueller 1936, Pineda & 10% NaOCl for 40 min in an ultrasonic cleaner to
Kuttler 1972). remove surface organic debris on the surface. Any
Information concerning the age-related changes of further deposits calculus and bone fragments were
root canal morphology in the human dentition is removed subsequently. The vacuum injection protocol
insufficient. In addition, more detailed investigations described by Yoshiuchi et al. (1972) was then used to
using a more accurate technique to visualize the root inject ink into the root canal system and make the
canal system is necessary to make a firm conclusion tooth transparent to visualize the canal system. Briefly,
about these changes. A better knowledge of this a conservative coronal access was made into the pulp
changing morphology would assist clinicians in treat- chamber with a carbide bur in a high-speed handpiece.
ing teeth, particularly in young patients. It would be China ink was then injected into the pulp cavity using a
expected that the root canal anatomy of the mandib- vacuum injector two or three times. Teeth were
ular first and second permanent molar teeth would thoroughly cleaned with water to remove surface
change with age. stains. Teeth were then demineralized for five days in
The aim of this study was to investigate the changes
that take place in the canal morphology of permanent Table 1 Age groups and the prevalence of developmental
mandibular molars at different ages. It was hypothe- stages of root canal morphology in the mesial root of
sized that in the mesial root of permanent mandibular mandibular first molars
molar teeth, canal morphology and the prevalence of Root canal stages
Age group of patients
inter-canal communications (ICC) differed amongst at extraction (years) n S1 S2 S3
different age groups.
6–11 19 17 (89.5) 2 (10.5) –
12–15 46 7 (15.2) 20 (43.5) 19 (41.3)
16–20 58 3 (5.2) 9 (31.0) 46 (63.8)
21–30 66 – 4 (15.1) 62 (84.9)
Two hundred and forty permanent mandibular first 31–40 19 – 1 (5.26) 18 (94.73)
41 and over 32 – – 32 (100)
and 240 second molar teeth from patients of known
age and gender were included. Teeth were collected Figures in parentheses denote percentages.
from the patients who attended for extractions as a
result of caries, before prosthodontic treatments, etc. at Table 2 Age groups and the prevalence of developmental
three dental hospitals in the central province of Sri stages of root canal morphology in the mesial root of
Lanka. The patients consisted of Sri Lankan Sinhalese mandibular second molars
and Tamils who are the two major ethnic groups in Sri Age group of patients
Root canal stages
Lanka. It has been reported that the Sinhalese of Sri at extraction (years) n S1 S2 S3
Lanka are genetically similar to the Tamils of Sri Lanka
12–15 30 28 (93.3) 2 (6.7) –
who have always been in proximity with each other 16–20 42 20 (47.6) 6 (14.3) 16 (38.1)
historically, linguistically, geographically and cultur- 21–30 68 8 (11.8) 6 (8.8) 54 (79.4)
ally (Kshatriya 1995, Papiha et al. 1996, Peiris et al. 31–40 48 – – 48 (100)
2006). All subjects enrolled in the project responded to 41 and over 52 – – 52 (100)
an informed-consent protocol, which had been Figures in parentheses denote percentages.
Root canal morphology with age Peiris et al.
5% nitric acid at room temperature (20 C); the nitric

acid solution was changed each day. To learn the
reliability of the demineralization procedure, teeth were
tested for softness by inserting a needle in the coronal
region. After demineralization, the teeth were rinsed in
running water for 24 h and then dehydrated using
ascending concentrations of ethanol (70%, 80%, 90%,
95% and 100%) for 5 days. Finally, the teeth were
rendered transparent by immersion in a solution
containing benzoic acid mixed with benzene and
methylsalycylate for 2–3 days. At the end of this
procedure, all specimens were transparent, and with
no sign of opacity on their surfaces at the end of a
3-day period. Furthermore, it was noticed that the
samples, taken out of the solution lost their transpar-
ency rapidly.
The cleared specimens were examined under a
dissecting microscope at 10· magnification by one
investigator. Depending on the pattern of development, Figure 2 Vertucci’s classification of root canal types (repro-
canal morphology of the mesial (M) root was classified duced from Peiris et al. 2007, with permission).
into three stages (Hess 1925) (Fig. 1): stage 1, single
large canal from pulp chamber to the apex without
secondary dentine deposition; stage 2, commencement
of canal differentiation with the appearance of isolated (a)
secondary dentine depositions and/or secondary
dentine deposition increases, so that bifurcations
appear; stage 3, canal differentiation is completed with
complete division of the root canal system. When the
root canal system was fully differentiated, the type of type I type II type III type IV
canals, the number and position of lateral canals and
ICC were recorded. During the evaluation of the
samples, the classification of Vertucci (1984) was
taken as the main reference. Vertucci (1984) classified
root canal configuration of human permanent teeth
into eight types (Figs 2 and 3a): type 1 (1) – a single
canal extends from the pulp chamber to the apex; type type V type VI type VII type VIII
II (2-1) – two separate canals leave the pulp chamber
(b)
(ad. 1) (ad. 2) (ad. 3)
Figure 3 Transparent root canal appearance of mandibular

permanent molars with completely differentiated canal sys-
tems in this study. (a) Examples for each type of Vertucci’s
classification: type I, 1; type II, 2-1; type III, 1-2-1; type IV, 2;
type V, 1-2; type VI, 2-1-2; type VII, 1-2-1-2; type VIII: 3. (b)
Figure 1 Diagrammatic representation of the different stages Additional canal configurations, commonly found: (ad. 1), 2-
of development of canal morphology in the mesial root of Ml 3; (ad. 2), 1-2-3; (ad. 3), 3-1-2 (reproduced from Peiris et al.
and M2 (modified from Hess 1925). 2007, with permission.).
and join short of the apex to form one canal; type III (1- examine any statistically significant difference and a
2-1) – one canal leaves the pulp chamber, divides into P-value of less than 0.05 was considered significant.
two within the root and then merges to exit as one
canal; type IV (2) – two separate and distinct canals
Results
extend from the pulp chamber to the apex; type V (1-2)
– one canal leaves the pulp chamber and divides short Concordance for the test of consistencies of the observer
of the apex into two separate and distinct canals with in assessing canal morphology was 92% for canal
separate apical foramina; type VI (2-1-2) – two stages and 96% for canal types, indicating that using
separate canals leave the pulp chamber, merge within the present classification, canal morphology could be
the body of the root and redivide short of the apex to scored with high reliability.
exit as two distinct canals; type VII (1-2-1-2) – one The data for different developmental stages of canal
canal leaves the pulp chamber, divides and then rejoins morphology in the mesial root of mandibular molars
within the body of the root and finally redivides into are presented in Tables 1 and 2. In both first and
two distinct canals short of the apex; type VIII (3) – second molars, canal morphology amongst age groups
three separate canals extend from the pulp chamber to was significantly different (P < 0.0001). There was a
the apex. Aberrant forms of root canals that did not fit decline of stage 1 and increase of stage 3 canal forms
Vertucci’s classification were also evaluated. with the advancement of the age. After 30 years of age,
A test of the consistency of the observer in assessing the majority of teeth had stage 3 canal morphology.
developmental stages of canal morphology and canal Furthermore, in first molars, a greater percentage of
types was undertaken by re-examining the mesial teeth in the 6–11 years age group had stage 1 canal
roots of 100 randomly selected first molars and then form. Stage 2 and stage 3 were commonly seen in the
comparing the results with the original canal assess- 12–15 and 16–20 years age groups (Table 1). On the
ment. Sri Lankan Sinhalese and Tamil data were other hand, in second molars, the 12–15 years age
combined for the analysis because no statistically group typically presented with stage 1 canal morphol-
significant difference was observed in canal morphology. A high prevalence of stage 1 and stage 3 was
ogy between these two population groups. In addition, observed in the 16–20 years age group (Table 2).
these parameters did not significantly differ between The distribution of canal differentiation and ICC in
male and female; thus, these groups were also mesial roots of first and second molars is shown in
combined. jmp (Version 3; SAS Institute, North Fig. 4(a,b). Prevalence of ICC between age groups was
Carolina, USA) software was used for the statistical significantly different in first molars (P < 0.0001) and
analysis. The Kruskal–Wallis test was applied to less significant in second molars (P < 0.05). In both
(a) 120
%
100
80 UDCS (stage 1 & 2)
60 0 ICC
CDCS
40 1–5 ICC
(stage 3)
20 >6 ICC
0
6–11y 12–15y 16–20y 21–30y 31–40y >41y
Age gorups
(b) 120
% 100
Figure 4 (a) Prevalence of canal
differentiation and ICC at different age 80
groups in Ml. (b) Prevalence of canal 60 UDCS (stage 1 & 2)
differentiation and ICC at different age 40 0 ICC CDCS
groups in M2. ICC, inter-canal 20 1–5 ICC (stage 3)
communications; UDCS, undifferentiated
0
canal system; CDCS, completely 12–15y 16–20y 21–30y 31–40y >41y
differentiated canal system. Age gorups
first and second molars, with the establishment of canal one or two canals with one apical foramen of type I and
system, the prevalence of mesial roots with 1–5 ICC type III canal morphologies. The distal root commonly
gradually increased with age up to 31–40 years. presented with type I canal configurations. In addition,
Meanwhile, comparatively high and low incidence of C-shaped canals were found in 3% of the cases.
mesial roots without ICC (0 ICC) and 1–5 ICC, Three additional canal configurations beyond Vert-
respectively was observed in >40 years age group. ucci’s classification were identified (Fig. 3b). In one
Furthermore, in first molars, occurrence of ‡6 ICC case, two canals exit the pulp chamber and the buccal
increased with the advancing age up to 21–30 years canal divided into two at the apical third to make three
and then decreased gradually. In this study, 6–11 and separate canals at the apex. In the second case, a wide
12–15 years age groups, respectively, in first and single canal exit of the pulp chamber and separated
second molars did not show any completely differenti- into two in the cervical third. The buccal canal then
ated canal systems and therefore, no ICC. branched in the apical third to exit as three canals at
The results of the evaluation of the root canal system the apex. A third case had three canals leaving the pulp
of first and second molars after completion of canal chamber that merged in the middle third and separated
differentiation (stage 3) are given in Table 3. Mesial into two canals again to exit as two foramina at the
roots of first molars typically had two canals and two apex.
apical foramina of type IV and two canals and one The percentages and location of the ICC and lateral
apical foramen of type II canal configuration. Further- canals are presented in Tables 4 and 5. ICC were more
more, additional canal types were found in 3.4% of first common in the mesial roots of first and second molars;
molars studied. Most of the distal roots of first molars the prevalence was lower in second molars. In both first
had type I canal configurations. The remainder was and second molars, they were seen less frequently in
distributed mainly between type V, type IV and type III. the distal root than in the mesial root. ICC were found
In second molars, the majority of the mesial roots had mostly in the middle third and less frequently in the
apical and cervical third of the root, respectively.
Table 3 Root canal types of mandibular first and second
Moreover, in many cases, they were observed in all
molars
positions simultaneously. In addition, a variable num-
First Second ber of communications was found at each position. For
n 177 165 example, the number of ICC in the middle third varied
Canal Canal
from 1–11 canals. There was an increasing prevalence
configuration type MR DR MR DR of lateral canals towards the apical part of the root. The
1 I 5 (2.8) 127 (71.8)51 (30.9) 159 (96.4)
apical third of the root had a higher prevalence of
2-1 II 44 (24.9) 2 (1.1) 24 (14.6) 1 (0.6) lateral canals than the middle, cervical and furcation
1-2-1 III 6 (3.4) 12 (6.8) 42 (25.5) 2 (1.2) regions combined. Lateral canals were more frequently
2 IV 107 (60.5) 14 (7.9) 25 (15.1) – seen in first rather than in second molars.
1-2 V 3 (1.7) 18 (10.2)20 (12.1) 3 (1.8)
2-1-2 VI 3 (1.7) 2 (1.1) 1 (0.6) –
1-2-1-2 VII 1 (0.5) 1 (0.5) – – Discussion
3 VIII 2 (1.1) – – –
Additional A 6 (3.4) 1 (0.5) 2 (1.2) – It has been suggested that although various techniques
canal have been used to evaluate root canal morphology, the
configurations
most detailed information is obtained by demineraliza-
Figures in parentheses denote percentages. tion and staining (Vertucci 1984, Omer et al. 2004),
Table 4 Number and position of inter-canal communications in mandibular molars
Position of inter-canal communications

Inter-canal
Tooth Root n communi -cations C M A C+M C+A M+A C+M+A
First molar MR 177 137(77.4) 11(6.2) 19(10.7) 7(4.0) 23(13.0) 2(1.1) 16(9.0) 59(33.3)
DR 177 20(11.3) 2(1.1) 8(4.5) 3(1.7) 1(0.6) –. 4(2.3) 2(1.1)
Second molar MR 165 63(38.2) 6(3.6) 16(9.7) 1(0.6) 14(8.5) – 13(7.9) 13(7.9)
DR 165 2(1.2) 1(0.6) 1(0.6) – – – – –
Figures in parentheses denote percentages.

C, cervical third of the root; M, middle third of the root; A, apical third of the root.
Table 5 Number and position of lateral canals in mandibular molars
Position of lateral canals
Tooth Root n Lateral canals C M A C+A M+A
First MR 177 92 (52.0) – 3 (1.7) 80 (45.2) – 9 (5.1)

molar DR 177 60 (33.9) – 6 (3.4) 52 (29.4) – 2 (1.2)
Second MR 165 62 (37.5) – 6 (3.6) 54 (32.7) – 2 (1.2)
molar DR 165 51 (30.9) – 6 (3.6) 44 (26.7) – 1 (0.6)
Figures in parentheses denote percentages.

C, cervical third of the root; M, middle third of the root; A, apical third of the root.
which is excellent for three-dimensional evaluation of type II canal form. Similarly, when two partitions are
root canal morphology. It was anticipated that exam- formed at cervical and apical thirds of the root, type VI
ination of fine details (ICC, lateral canals) would require canal configuration results. However, if numerous
ink penetration; however, the quality of clearing was partitions form, extensive differentiation of the root
sufficient to visualize such details without staining. canal system results in a reticular form in which three
First and second molars erupt around 6–11 years or more vertical canals are present with lateral
and apical closure is completed around 9–14 years of interconnections resulting in type VIII or additional
age, respectively (Hillson 1998). Hess (1925) explained canal types (Fig. 3). This study further confirmed that
that differentiation of the root canals appears only after canal differentiation was completed around
the growth of the root is completed (after the closing of 30–40 years of age in both first and second molars.
the apical foramen). The findings of this study are in However, the rate of progression of secondary dentine
agreement with these previous investigations and deposition was variable as evidenced by the fact that all
confirm that mesial roots of first and second molars developmental stages of root canal morphology were
have most often one large canal until 11 and 15 years represented in younger age groups (up to 30 years of
of age, respectively. In addition, in first and second age).
molars, completion of canal differentiation commences The occurrence of ICC increased with age up to
at about 3–6 and 2–6 years after root completion, 31–40 years in both first and second molars. Hess
respectively. Furthermore, ages 12–20 years in first (1925) noted that extensive differentiation of the root
and 16–30 years in second molars had mixed patterns canal system with age was characterized by the
of canal morphology and therefore, these periods seem appearance of spaces, transverse ICC and especially
to be a transition period for canal differentiation by the division of the root canal into two separate
(Tables 1 and 2). During this period, secondary dentine canals. He further mentioned that after completion of
deposition in the mesio-distal direction within the canal canal development, continuous deposition of secondary
at the cervical, middle and apical thirds causes canal dentine caused numerous intervening stages, which
separation. It has been suggested that the form and resulted in a network of transverse anastomosis and
number of root canals are principally determined by the communications between the separate canals. These
deposition of secondary dentine and these partitions findings are consistent with those of Hess (1925) and
cause extensive differentiation in root canals, confirmed increasing numbers of ICC between 16 and
which were originally in simple form (Hess 1925, 40 years of age in both first and second molars
Thomas et al. 1993). (Fig. 4a,b). Meanwhile, with advancing age further
Thomas et al. (1993) suggested that secondary deposition of secondary dentine, which is a slow
dentine appeared initially to be deposited in the mid continuous process, results in narrowing of canals,
root where the root was constricted. Hess (1925) disappearance of transverse anastomosis and lateral
reported that differentiation might commence either at canals (Hess 1925, Thomas et al. 1993). This finding is
the end or on the middle of the root. Meanwhile, in this further supported by our study and shows a sudden
study, it was noted that islands of secondary dentine decrease in the prevalence of 1–5 ICC and increase in
depositions occurred mostly in the apical and middle occurrence of no ICC in the mesial root of first and
third of the root. It would appear that if a single second molars after 40 years of age.
partition is formed in the cervical third, a single root This investigation confirmed that root canal
canal becomes separated coronally forming Vertucci’s morphology of completely differentiated mandibular
molars is complex. Although the distal root canals et al. 1993, Walton 1997, Basmadjian-Charles et al.
had simple structures, the mesial canals exhibited 2002, Murray et al. 2002, Dammaschke et al. 2003).
many complex configurations. Furthermore, addi- This study further reinforces these findings and shows
tional canal configurations that did not fit Vertucci’s distinct developmental patterns of canal morphology,
classification were more frequently found in mesial especially in young and intermediate age groups.
roots than in distal and in first compared with second Therefore, it is suggested that for successful root canal
molars. For example, 72% of distal root of first treatment, the dentist should be aware of these
molars had type I canal morphology and prevalence developmental variations in the root canal morphol-
of type IV and type II canal configuration in the ogy. Meanwhile, reports of the effect of a patient’s age
mesial root was 61% and 25%, respectively. The on the outcome of endodontic treatment are contra-
result of this study of Sri Lankan first molars dictory. The observation that age made no difference to
compares with those of American Caucasians the outcome was reported by Ingle (1965), Storms
(Vertucci 1984) and Turkish (Sert et al. 2004) (1969) and Swartz et al. (1983). However, Grossman
populations. However, they differ from those of et al. (1964) and Seltzer et al. (1963) both reported
Thai (Gulabivala et al. 2002), Burmese (Gulabivala better outcomes in younger patients.
et al. 2001) and Chinese (Walker 1988a) populations
who had a higher prevalence of two canals and
Conclusions
apical foramina in the distal roots of first molars.
These findings are consistent with the frequent The mesial roots of mandibular first and second molars
occurrence of three-rooted (one mesial and two had mostly one large canal until 11 and 15 years of
distal) first molars in Thai and Burmese populations, age, respectively. In both teeth, the canal system was
which represent a culture mix of people of Chinese completely established at 30–40 years of age. This
and Indian origin (Gulabivala et al. 2002,2001) and findings further support the concept that rate of
in Chinese populations (Walker 1988a). progress of secondary dentine deposition, which differ-
In second molars, 66% of Sri Lankans had three entiate the canal system is variable. The prevalence of
canals and two canals were seen in 31% of the teeth ICC was low at young and old ages but high at
examined. These results are consistent with previous intermediate ages. It is therefore important to be
findings of second molar canal morphology in American familiar with these age-related variations in the root
Caucasians (Vertucci 1984) but differ from those of Thai canal system because such knowledge can aid in the
(Gulabivala et al. 2002), Burmese (Gulabivala et al. location and negotiation of canals as well as their
2001) and Chinese (Walker 1988b) populations. subsequent management in clinical practice.
Walker (1988b) recorded a high prevalence (55%) of
two canals and a low prevalence (45%) of three canals in
Acknowledgements
southern Chinese people. Moreover, Thai (Gulabivala
et al. 2002) and Burmese (Gulabivala et al. 2001) had We are grateful to all members of the Department of
three canals in 55% and 43% of teeth, respectively. In Anatomy and Physical Anthropology, Nihon Univer-
addition, they had one root canal in 3% and 10% of sity School of Dentistry at Matsudo for their kind
teeth, respectively. Studies of canal anatomy of second cooperation in carrying out this work.
molars in Japanese (Kotoku 1985), Chinese (Yang et al.
1988) and Hong Kong Chinese (Walker 1988b)
References
reported a high incidence of C-shaped canals. Mean-
while, amongst Europeans and recently in Sri Lankans, Basmadjian-Charles CL, Farge P, Bourgeois DM, Lebrun T
an absence (Tamse & Kaffe 1981, Vertucci 1984) or (2002) Factors influencing the long-term results of endo-
near absence (Weine et al. 1988, Peiris et al. 2007) of dontic treatment: a review of the literature. International
C-shaped canals in second molars has been described. In Dental Journal 52, 81–6.
Dammaschke T, Steven D, Kaup M, Ott KHR (2003) Long-
this study, the prevalence of 3% in Sri Lankan second
term survival of root canal treated teeth: a retrospective
molars is in agreement with later findings.
study over 10 years. Journal of Endodontics 29, 638–43.
Many investigations have revealed the age-related
Grossman LI, Shepard LI, Pearson LA (1964) Roentgenologic
changes of the root canal system and the importance of and clinical evaluation of endodontically treated teeth. Oral
a better understanding that these changes influence the Surgery, Oral Medicine and Oral Pathology 17, 368–74.
provision of restorative and endodontic care (Smith
Gulabivala K, Aung TH, Alavi A, Ng YL (2001) Root and Seltzer S, Bender IB, Turkenkopf S (1963) Factors effecting
canal morphology of Burmese mandibular molars. Inter- successful repair after root canal therapy. Journal of Amer-
national Endodontic Journal 34, 359–70. ican Dental Association 67, 651–62.
Gulabivala K, Opasanon A, Ng YL, Alavi A (2002) Root and Sert S, Aslanalp V, Tanalp J (2004) Investigation of the root
canal morphology of Thai mandibular molars. International canal configurations of mandibular permanent teeth in the
Endodontic Journal 35, 56–62. Turkish population. International Endodontic Journal 37,
Hess W (1925) The Anatomy of the Root Canals of the Teeth of the 494–9.
Permanent Dentition, 1st edn. London, UK: John Bale, Sons Smith CS, Setchell DJ, Harty FJ (1993) Factors influencing the
and Danielsson Ltd, pp. 4–49. success of conventional root canal therapy – a five-year
Hillson S (1998) Dental Anthropology, 1st edn. Cambridge, UK: retrospective study. International Endodontic Journal 26,
Cambridge University press. 321–33.
Ingle JI (1965) Endodontics, Philadelphia, USA: Lea & Febiger, Storms JL (1969) Factors that influence the success of
pp. 54–76. endodontic treatment. Journal of Canadian Dental Association
Kotoku K (1985) Morphological studies on the roots of 35, 83–97.
the Japanese mandibular second molars. Shikwa Gakuho Swartz DB, Skidmore AE, Griffin JA Jr. (1983) Twenty years of
85, 43–64 (in Japanese). endodontic success and failure. Journal of Endodontics 9,
Kshatriya GK (1995) Genetic affinities of Sri Lanka popula- 198–202.
tions. Human Biology 67, 843–66. Tamse A, Kaffe I (1981) Radiographic survey of the preva-
Manning SA (1990a) Root canal anatomy of mandibular lence of conical lower second molar. International Endodontic
second molars. Part I. International Endodontic Journal 23, Journal 14, 188–90.
34–9. Thomas RP, Moule AJ, Bryant R (1993) Root canal morphol-
Manning SA (1990b) Root canal anatomy of mandibular ogy of maxillary permanent first molar teeth at various ages.
second molars. Part II. C-shaped canals. International International Endodontic Journal 26, 257–67.
Endodontic Journal 23, 40–5. Vertucci FJ (1984) Root canal anatomy of the human
Mueller AH (1936) Morphology of root canals. Journal of permanent teeth. Oral Surgery, Oral Medicine and Oral
American Dental Association 23, 1698–706. Pathology 58, 589–99.
Murray PE, Stanley HR, Matthews JB, Sloan AJ, Smith AJ Walker RT (1988a) Root form and canal anatomy of
(2002) Age-related odontometric changes of human teeth. mandibular first molars in a southern Chinese population.
Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology Endodontics and Dental Traumatology 4, 19–22.
and Endodontics 93, 474–82. Walker RT (1988b) Root form and canal anatomy of
Omer OE, Al Shalabi RM, Jennings M, Glennon J, Claffey NM mandibular second molars in a southern Chinese popula-
(2004) A comparison between clearing and radiographic tion. Journal of Endodontics 14, 325–9.
techniques in the study of the root-canal anatomy of Walton RE (1997) Endodontic considerations in the geriatric
maxillary first and second molars. International Endodontic patient. Dental Clinics of North America 41, 795–816.
Journal 37, 291–6. Weine FS, Pasiewicz RA, Rice RT (1988) Canal configuration
Papiha SS, Mastana SS, Jayasekara R (1996) Genetic varia- of the mandibular second molar using a clinically oriented
tions in Sri Lanka. Human Biology 68, 707–37. in vitro method. Journal of Endodontics 14, 207–13.
Peiris R, Nanayakkara D, Kageyama I (2006) Crown dimen- Yang ZP, Yang SF, Lin YC, Shay JC, Chi CY (1988) C-shaped
sions of the mandibular molars in two ethnic groups in Sri root canals in mandibular second molars in a Chinese
Lanka. Anthropological Sciences 114, 89–92. population. Endodontics and Dental Traumatology 4, 160–3.
Peiris R, Takahashi M, Sasaki K, Kanazawa E (2007) Root and Yoshiuchi Y, Takahashi K, Yokochi C (1972) Studies on the
canal morphology of permanent mandibular molars in a Sri anatomical forms of the pulp cavities with new method of
Lankan population. Odontology 95, 16–23. vacuum injection (II) – accessory canal and apical ramifi-
Pineda F, Kuttler Y (1972) Mesiodistal and buccolingual cation. Japanese Journal of Oral Biology 14, 156–85 (in
roentgenographic investigation of 7,275 root canals. Oral Japanese).
Surgery Oral Medicine and Oral Pathology 33, 101–10.
doi:10.1111/j.1365-2591.2008.01434.x
Effectiveness of different gutta-percha techniques

when filling experimental internal resorptive
cavities
N. Gencoglu1, T. Yildirim2, Y. Garip1, B. Karagenc3 & H. Yilmaz3

1
Department of Endodontics, Faculty of Dentistry, Marmara University, Istanbul; 2Department of Restorative Dentistry, Faculty of
Dentistry, Karadeniz Technical University, Trabzon; and 3Private Practice, Istanbul, Turkey
Abstract gutta-percha and void in the internal resorptive cavi-

ties. All measurements were analysed statistically using
Gencoglu N, Yildirim T, Garip Y, Karagenc B, Yilmaz H.
One-way anova and Newman–Keuls tests.
Effectiveness of different gutta-percha techniques when filling
Results The Microseal technique filled 99% of the
experimental internal resorptive cavities. International End-
artificial resorptive area followed by LC (92%), SystemB
odontic Journal, 41, 836–842, 2008.
(89%), Quick-Fill (88%), Thermafil (74%) and Soft-Core
Aim To determine the quality of root fillings in teeth (73%). Warm gutta-percha compaction techniques
with artificial internal resorptive cavities filled with filled the resorption areas with more gutta-percha than
Thermafil, JS Quick-Fill, Soft Core, System B and Micro- sealer (Microseal 68%, System B 62%) compared to the
seal, and by cold lateral compaction (LC) technique. other techniques (LC 48%, Quick Fill 41%, Soft Core
Methodology Sixty maxillary incisor teeth were 34%, Thermafil 35%). In addition, core techniques left
selected. After access cavity preparation and root canal a considerable volume of voids in the resorptive areas
instrumentation, the roots were sectioned horizontally (Quick-Fill 12%, Thermafil 26%, Soft Core 27%).
and artificial internal resorption cavities were prepared Conclusions Warm gutta techniques filled artificial
on the canal walls. The tooth sections were cemented resorption cavities significantly better than the other
together and the root canals were filled using one of six gutta-percha techniques.
different techniques: Thermafil, JS Quick-Fill, Soft Core,
Keywords: internal resorption, JS Quick-Fill, Micro-
System B and Microseal, and by LC. The roots were
seal, Soft Core, System B, Thermafil.
then divided at the level of the previous section and
each root surface was photographed. Image analysis Received 21 May 2007; accepted 7 January 2008
program was used to calculate the percentage of sealer,
during cleaning, shaping and filling of the root canal

Introduction
(Nyguen 1984).
Complete filling of the root canal with an inert filling Many techniques and materials have been studied ex
material has been proposed as one of the goals of root vivo as possibilities to fill internal resorptive defects
canal treatment (Nyguen 1984). Root canal anatomy (Gutmann et al. 1993, Agarwal et al. 2002, Collins
may display complex irregularities in shape as a result et al. 2006). Gutmann et al. (1993) suggested the use
of pathological processes such as internal resorption. of the Thermafil obturation technique; whilst Agarwal
These irregularities cannot usually be reached by root et al. (2002) reported that the use of ultrasonics to
canal instruments and this leads to some difficulties condense the gutta-percha and the Obtura II system
were superior to the Thermafil and lateral compaction
(LC) techniques. Collins et al. (2006) suggested the use
Correspondence: Dr Tahsin Yildirim, Department of Restorative
Dentistry, Faculty of Dentistry, Karadeniz Technical University,
of warm lateral and warm vertical condensation gutta-
Trabzon, Turkey (Tel.: +90462 377 4733; fax: +90462 325 percha techniques for such cases. Radiographic evalu-
3017; e-mail: tahsiny@hotmail.com, tahsiny@ktu.edu.tr). ation to assess voids is limited and it is difficult to
Gencoglu et al. Filling internal resorptive cavities
differentiate gutta-percha and sealer, even with the

Group 1: lateral compaction
advent of digital radiographic systems.
The objective of this study was to investigate the A gutta-percha master cone was fitted within
quality of root fillings by measuring the sealer/core/ 0.5 mm of the working length. Freshly mixed pulp
voids ratios of artificial internal resorptive areas of root canal sealer (Kerr, Romulus, MI, USA) was applied to
canals filled with the thermo-mechanical (JS Quick-Fill) the root canal walls using a file in a counter-
and thermo-plasticized (Thermafil, Soft Core) gutta- clockwise rotation. The master cone was lightly
percha core techniques, a warm vertical condensation coated with sealer and placed into the root canal.
technique (System B), a warm lateral condensation LC was achieved using standardized finger spreaders
technique (Microseal) or cold LC technique. (Dentsply Maillefer, Ballaigues, Switzerland). When
the points prevented the spreader penetrating beyond
the coronal third of the canal, the canal was
considered to be adequately filled and excess gutta-
Sixty maxillary central incisor teeth were selected and percha was removed with a hot instrument.
adjusted to a length of 20 mm. A conventional
endodontic access was prepared in each tooth and a size
Group 2: warm vertical compaction with System B
10-K file was inserted to determine the location of the
apical foramen. The teeth were instrumented to master In the second group, a System B obturator (EIE/
apical file, size 60, using the step-back technique Analytic Technology, Richmond, WA, USA) was used
combined with 2.5% sodium hypochlorite irrigation. to compact gutta-percha as recommended by the
To create artificial internal resorptive cavities, the manufacturer. Sealer was applied to the canal walls
roots were sectioned horizontally with a fine diamond with a file and then the selected gutta-percha cone was
disc 7 mm from the apex. Semi-circular cavities were placed within 0.5 mm of the working length. A
created using a low speed No. 6 round diamond bur medium large insert tip, which bounds in the canal
around the periphery of the opening of the root canal 3 mm from the working length, was used to condense
of each section (Andreasen et al. 1987, Goldberg et al. the gutta-percha. The System B unit was preset to
2000). Then the sections were cemented together 200 C during the apical condensation of the primary
using Peligom glue (Pelikan cyanoacrylate adhesive; cone (downpack), and to 100 C when adapting and
Istanbul, Turkey) on the dentine surface around the condensing the apical portion of the secondary gutta-
cavities (Fig. 1). Each tooth was embedded in a plaster percha cone, and finally to 250 C to soften the rest of
cast. Then, sixty teeth were randomly assigned to six the secondary cone prior to vertical condensation
groups of ten. (Silver et al. 1999).
(a) (b)
Figure 1 (a,b) Radiographic appearance

of simulated resorptive cavity. Radio-
graph of root filling specimen.
Filling internal resorptive cavities Gencoglu et al.
manufacturer’s instructions. Sealer (Kerr) was applied

Group 3: Microseal system
to the canal walls with a file and a size 50 Quick-Fill
In the third group, root canals were filled using the obturator (two sizes smaller than the last file used to
Microseal system. An appropriate sized master cone prepare the apical third of the canal) was lightly coated
was selected to achieve tug back. The appropriate with the sealer and positioned in the canal until a slight
spreader was selected to compact the gutta-percha resistance was felt. Rotation was used until plasticity of
master cone within 1.0 mm of the working length. gutta-percha was observed, and slight pressure was
Finally, the appropriate mechanical compactors were applied apically until it reached working length and
selected according to the manufacturer’s instruction. then the obturator was left in the canal. Then, an
Sealer (Kerr) was applied to the canal walls using a file inverted cone bur was used to cut through the shank of
and the master gutta-percha cone coated in sealer was each carrier.
positioned. Then the spreader was inserted alongside
the master cone at the appropriate length for compac-
Group 6: Soft Core
tion. Upon withdrawal of the spreader from the canal, a
tapered void was formed between the compacted gutta- Based on the information obtained from a ‘size verifier’,
percha cone and the root canal walls. The appropriate a size 60 Soft-Core obturator was selected and heated.
compactor was inserted in the heated gutta-percha Sealer (Kerr) was applied to the canal walls using a file
cartridge and was coated with a uniform layer of and the plasticized Soft-Core device was inserted to the
material. The gutta-percha-coated compactor was then apical stop. After the filling was completed, the handle
immediately placed in the void, and placed as close to and insertion pin were removed, and the excess plastic
the working length as possible, avoiding rotation whilst core was removed with a small inverted cone bur (Soft
being inserted. Whilst resisting the backing-out motion, Core Manual, Copenhagen, Denmark).
but without using apical pressure, the rotation of the Following filling, the teeth were stored for 7 days at
compactor was initiated at a speed of 6000 rpm. After room temperature to ensure all materials had set. The
approximately 2 s, the compactor was removed slowly, plaster cast was removed and radiographs of the teeth
whilst being gently pushed against one side of the were taken (Fig. 1). Then, each tooth was sectioned
canal. Rotation did not stop until the compactor was with a rotary saw 7 mm from the apex at the level of
removed fully from the canal. If the canal was not the previous cut, and under cold water to minimize
completely filled, more gutta-percha was placed on the gutta-percha smearing.
compactor. That is, the process continued until the Photographs of both surfaces of the sectioned area
canal was perceived to be completely filled. Then, were taken by using a Nikon Coolpix 885 digital
excess gutta-percha and sealer were removed from the camera (Nikon Imaging Japan Inc., Tokyo, Japan),
access cavity (Korzen 1997). which was mounted on a stereomicroscope ocular eye
Group 4: Thermafil obturation

The canals were filled using the metal carrier Thermafil
system (Dentsply Maillefer). Sealer (Kerr) was applied to
the canal walls with a file and a size 60 Thermafil
obturator was warmed in a Thermaprep oven (Therma
Prep Plus, Dentsply Maillefer, Ballaigues, Switzerland)
for a minimum of 10 s in accordance with the
manufacturer’s recommendations. The heated obtura-
tor was slowly inserted into the canal within 0.5 mm of
the working length. An inverted cone bur was used to
cut through the shank of each carrier.
Group 5: JS Quick-Fill
Figure 2 Photograph of tooth sectioned at the middle of the
JS Quick-Fill obturators (JS Dental Manufacturing Inc, simulated internal resorptive cavity and filled using the
Ridgefeild, CT, USA) were used in accordance with the Thermafil obturation technique.
Figure 3 Photograph of tooth sectioned at the middle of the Figure 5 Photograph of tooth sectioned at the middle of the
simulated internal resorptive cavity and filled using the Soft simulated internal resorptive cavity and filled using the lateral
Core obturation technique. compaction technique.
Figure 4 Photograph of tooth sectioned at the middle of the Figure 6 Photograph of tooth sectioned at the middle of the
simulated internal resorptive cavity and filled using the JS simulated internal resorptive cavity and filled using the
Quick-Fill obturation technique. Microseal obturation technique.
Results
(Fig. 2–7). The photographs were transferred to a
computer and an image analysis program (image- Percentages of artificial resorptive area filled by gutta-
pro plus; Media Cybernetics, Inc., Silver Spring, MD, percha, sealer or voids are given in Table 1.
USA) was used to calculate the percentage of sealer,
gutta-percha and voids. In the Thermafil, JS Quick-Fill
Total filled area
and Soft-Core specimens, metal or plastic carriers were
regarded as gutta-percha whilst calculating the ratio of Statistical analyses revealed that the warm lateral
sealer, gutta-percha and voids (Gencoglu et al. 2002, condensation technique (Microseal) was associated
Gencoglu 2003). Whilst filling procedures were per- with the greatest total filled area (gutta-percha and
formed by one individual, the evaluations were per- sealer) (P < 0.05). Warm lateral (Microseal) and ver-
formed independently by someone blinded with respect tical (System B) condensation techniques had more
to the experimental group. All measurements were gutta-percha and sealer content than core techniques
statistically analysed by using One-way anova and (P < 0.05). Cold LC had more gutta-percha and sealer
Newman–Keuls tests. content than core techniques (P > 0.05).
Discussion
The objective of filling a root canal is to provide an
environment that prevents growth of residual bacteria
and inhibits the introduction of new bacteria. However,
there is little data available on whether any particular
technique is superior for filling canals with resorptive
defects.
In previous studies, radiographic methods were used
to analyse the filling of resorptive areas. Both Stamos &
Stamos (1986) and Wilson & Barnes (1987) reported
excellent results radiographically when using the
Obtura system combined with vertical compaction in
the filling of root canals with internal resorptive
Figure 7 Photograph of tooth sectioned at the middle of the cavities. However, Goldberg et al. (2000) stated that
simulated internal resorptive cavity and filled using the System the buccolingual radiograph was limited for document-
B obturation technique. ing filling of defects; they claimed that adding a
mesiodistally directed radiograph is more useful for
this purpose. As it is difficult to assess voids and almost
Gutta-percha area
impossible to differentiate gutta-percha and sealer
When the proportion of gutta-percha in the artificial radiographically, radiographic evaluation is inadequate
resorptive areas were analysed, Microseal had a statis- for evaluating the quality of root filling for internal
tically greater proportion of gutta-percha than other resorptive cavities.
techniques (P < 0.05), except for System B. System B In the present study, an image analysis program
had a great proportion of gutta-percha than core (image-pro plus 4.5) was used to evaluate the quality
techniques (Thermafil, Quick-Fill and Soft-Core) of the root filling by calculating the percentage of gutta-
(P < 0.05); cold LC also contained more gutta-percha percha, sealer or voids. The warm lateral condensation
than Thermafil and Soft-Core techniques (P < 0.05). technique, Microseal, resulted in the few voids whereas
When the gutta and sealer ratios were compared, all the thermoplasticized core techniques were ineffective
condensation techniques (Microseal, System B and LC) in filling the resorptive area with gutta-percha. How-
had more gutta-percha than sealer. However, all core ever, it must be remembered that no vertical conden-
techniques (Thermafil, Soft-Core, Quick-Fill) had more sation was included in the core technique, a technique
sealer than gutta-percha. that could have compacted gutta-percha into the
resorptive cavities more than using the passive tech-
nique used in the present study.
Voids
It is generally accepted (Peters 1986, Wesselink
When the area of voids was compared, Microseal had 1990, Miserendino 1991, Kontakiotis et al. 1997,
the lowest proportion of voids (P < 0.05) followed by Gencoglu et al. 2002, Gencoglu 2003) that root fillings
LC and System B. Core techniques had more voids than should contain more gutta-percha and less sealer. This
the other techniques. Amongst the core techniques, may be more important when filling root canals with
Quick-Fill had more gutta-percha and less voids con- resorptive lacunas. In resorptive lacunas, it is difficult
tent than the Thermafil and Soft-Core techniques. to remove all bacteria and their products from the
Table 1 Percentages of canal area filled

Technique GP + Sealer Gutta-percha Sealer Voids
with gutta-percha/sealer by filling tech-
Mean(%) ± SD nique
Microseal 98.59 ± 3.58 67.98 ± 11.11 30.60 ± 11.39 1.27 ± 3.55
Lateral compaction 91.55 ± 10.20 47.54 ± 10.89 44.01 ± 14.13 8.45 ± 10.52
System B 89.17 ± 13.02 62.05 ± 9.32 27.12 ± 9.39 10.82 ± 4.65
Quick-Fill 87.91 ± 11.31 41.39 ± 10.48 47.85 ± 16.62 12.09 ± 11.31
Thermafil 74.36 ± 17.92 34.30 ± 10.09 40.06 ± 14.09 25.67 ± 18.60
Soft Core 73.21 ± 23.78 35.47 ± 8.01 37.73 ± 21.45 26.79 ± 23.78
dentinal tubules. Dense compacted gutta-percha may However, in the LC technique, the sealer proportion
block dentinal tubules, and this may lead to better in the resorption area was 48%. This high sealer
entombment of microorganisms. proportion in LC is associated with inability of cold
Gencoglu (2003) found that core techniques con- gutta-percha cones to be compacted into the cavities
tained more gutta-percha than warm condensation and the diffusion of the sealer into the resorption
(Microseal), vertical condensation (System B) and cold area during condensation.
lateral condensation techniques when filling regular
root canals. Wilson & Barnes (1987) reported that
Conclusions
thermoplasticized gutta-percha techniques were
acceptable for filling of internal resorption cavities in The Microseal technique filled artificial resorptive
a case report. In addition, Gutmann et al. (1993) cavities better than other gutta-percha techniques.
concluded that the use of Thermafil obturation tech-
nique ex vivo was acceptable. However, these results
References
were not confirmed in the present study with regard to
filling internal resorptive cavities. Goldberg et al. Agarwal M, Rajkumar K, Lakshminarayanan L (2002)
(2000) and Agarwal et al. (2002) also reported that Obturation of internal resorption cavities with 4 different
the Thermafil technique did not fill resorptive cavities. techniques: an in-vitro comparative study. Endodontology
Indeed, in the present study, all core techniques 14, 3–8.
Andreasen FM, Sewerin I, Mandel U, Andreasen JO (1987)
(Thermafil, Quick-Fill, Soft-Core) were less effective for
Radiographic assessment of simulated root resorption cav-
filling resorptive cavities, whereas both Microseal and
ities. Endodontics & Dental Traumatology 3, 21–7.
System B filled the resorptive cavities to a greater extent
Collins J, Walk er MP, Kulild J, Lee C (2006) A comparison of
than core techniques, mainly with gutta-percha. In three gutta-percha obturation techniques to replicate canal
addition, all core techniques had more voids than the irregularities. Journal of Endodontics 32, 762–5.
condensation techniques. Gencoglu N (2003) Comparison different gutta-percha tech-
JS Quick-Fill is a mechanically thermoplasticized niques (Part II); Thermafil, JS Quick-Fill, Soft Core, Micro-
(thermomechanic) gutta-percha filling method in seal, System B and lateral condensation techniques. Oral
which a titanium carrier is covered with gutta- Surgery Oral Medicine Oral Pathology Oral Radiology Endo-
percha. Rotation of the carrier within the root canal dontics 96, 1–5.
generates frictional heat, and the plasticized gutta- Gencoglu N, Garip Y, Bas M, Samani S (2002) Comparison of
different gutta-percha root filling techniques: Thermafil,
percha is moved apically. Rotation of the carrier
Quick-Fill, System B and lateral condensation. Oral Surgery
within the root canal and heat may increase the
Oral Medicine Oral Pathology Oral Radiology Endodontics 93,
amount of gutta-percha in the resorptive area. But it
333–6.
appeared that warmed gutta-percha around the core Goldberg F, Massone EJ, Esmoris M, Alfie D (2000) Compar-
material was not adequate in filling the hollow space ison of different techniques for obturating experimental
in the resorbtive area. In each core technique, the internal resorptive cavities. Endodontics & Dental Traumatol-
amount of gutta-percha surrounding the core mate- ogy 16, 116–21.
rial and the characteristics of fluidity is different and Gutmann JL, Saunders WP, Saunders EM, Nguyen L (1993)
this may affect the filling quality of the core An assessment of the plastic Thermafil obturation tech-
techniques in resorbtive area. nique. Part 2. Material adaptation and sealability. Inter-
All condensation techniques were found to be more national Endodontic Journal 26, 179–83.
Kontakiotis EG, Wu MK, Wesselink PR (1997) Effect of sealer
successful than core techniques in the filling of
thickness on long-term sealing ability: a 2-year follow-up
resorptive cavities. Amongst the condensation tech-
study. International Endodontic Journal 30, 307–12.
niques, Microseal was observed to be the most
Korzen BH (1997) Endodontic obturation using the Microseal
successful in comparison with System B and LC. technique. Oral Health 10, 67–73.
Using the Microseal technique prevented voids and Miserendino LJ (1991) Instruments, materials and devices. In:
produced more homogenous gutta-percha content in Cohens S, Burns RC, eds. Pathways of the Pulp, 5th edn. St
the artificial resorptive area; System B and LC Louis, MO: Mosby-Year Book Inc, pp. 388–432.
techniques revealed similar results (System B 89%, Nyguen NT (1984) Obturation of the root canal system. In:
LC (92%). Whilst the resorption area was extensively Cohens S, Burns RC, eds. Pathways of the pulp, 3rd edn. St
filled with gutta-percha in the System B technique, Louis, MO: Mosby Co, pp. 205–99.
the sealer proportion in resorption areas was 27%.
Peters DD (1986) Two year in vitro solubility evaluation of Wesselink PR (1990) Conventional root canal therapy. III:
four guttapercha sealer obturation techniques. Journal of root filling. In: Harty FJ, ed. Endodontics in Clinical Practice,
Endodontics 12, 139–45. 3rd edn. London, UK: Wright, pp. 186–223.
Silver GK, Love RM, Purton DG (1999) Comparison of two Wilson PR, Barnes IE (1987) Treatment of internal root
vertical condensation obturation techniques. International resorption with thermoplasticized gutta-percha. A case
Endodontic Journal 32, 287–95. report. International Endodontic Journal 20, 94–7.
Stamos DE, Stamos DG (1986) A new treatment modality for
internal resorption. Journal of Endodontics 12, 315–9.
doi:10.1111/j.1365-2591.2008.01435.x
Modification of mineral trioxide aggregate. Physical

and mechanical properties
J. Camilleri
Department of Building and Civil Engineering, Faculty of Architecture and Civil Engineering; Faculty of Dental Surgery, University
of Malta, Malta
Abstract materials were compared with proprietary brand

glass–ionomer cement.
Camilleri J. Modification of mineral trioxide aggregate.
Results The setting time of the CSA and CFA cements
Physical and mechanical properties. International Endodontic
was less than 6 min. The use of an admixture to
Journal, 41, 843–849, 2008.
improve the handling properties tended to retard
Aim To evaluate the physical and mechanical prop- setting. CSA was stronger then CFA in both compres-
erties of two cements with a similar chemical compo- sion and flexure (P < 0.001). Addition of granite
sition to mineral trioxide aggregate and to attempt to increased the flexural strength of both cements but
improve their properties by producing a composite reduced the compressive strength (P < 0.01). CFA
material. absorbed more water then CSA. Addition of granite
Methodology Two cement types were used: a mix- reduced the water uptake of both cements.
ture of calcium sulpho-aluminate cement and Portland Conclusions Both CSA and CFA cements had ade-
cement (CSA) and calcium fluoro-aluminate cement quate setting times and compressive strength values
(CFA) in conjunction with an admixture to improve the when compared with proprietary brand glass–ionomer
handling characteristics. Cements were mixed with an cement. CSA was superior to CFA and had more
inert filler to produce cement composite. The setting promise as a prospective dental material.
time of the cements was evaluated using an indentation
Keywords: accelerated cements, compressive stre-
technique. The flexural and uni-axial compressive
ngth, flexural strength, Portland cement, setting time.
strengths and solubility of the cements and cement
composites were evaluated. The properties of the Received 31 January 2008; accepted 4 April 2008
be suitable for use as a pulp capping agent (Pitt Ford et al.

Introduction
1996, Bakland 2000); as a dressing over pulpotomies of
Mineral trioxide aggregate (MTA) is composed of Port- permanent (Holland et al. 2001) and primary teeth
land cement and bismuth oxide (Torabinejad & White replacing the formocresol pulpotomy procedure
1995). MTA was developed as a root-end filling material (Eidelman et al. 2001); for obturation of retained
following apicectomy and to repair root perforations (Lee primary teeth (O’Sullivan & Hartwell 2001), and
et al. 1993, Pitt Ford et al. 1995). The material has been permanent immature (Hayashi et al. 2004) and mature
used successfully in this regard and indeed, its use has teeth (Vizgirda et al. 2004); for single visit apexification
been extended. For example, MTA has been reported to procedures for immature teeth with necrotic pulps
(Whiterspoon & Ham 2001), thus acting as an apical
barrier material (Shabahang & Torabinejad 2000); and
Correspondence: Dr Josette Camilleri, PhD, Department of as a root canal sealer cement (Holland et al. 1999,
Building and Civil Engineering, Faculty of Architecture and
Civil Engineering, University of Malta, Malta (Tel.: 00356
Geurtsen 2000). Unfortunately, MTA has limited
2340 2870; fax: 00356 21330190; e-mail: josette.camilleri applications in operative dentistry because of its long
@um.edu.mt). setting time and low compressive strength compared
MTA modification Camilleri
with other materials (Torabinejad et al. 1995). The • 3 parts calcium aluminate (Lafarge Special Cements,
extended endodontic uses and its potential use in Nottingham, UK)
operative dentistry necessitate the development of a • 8 parts white Portland cement (Lafarge Asland,
new formulation of the material based on Portland Valencia, Spain)
cement, which optimizes both its strength and its setting • 1 part synthetic anhydrite (Lafarge Special Cements,
time without compromising its biocompatibility. Nottingham, UK)
Portland cement takes a minimum of 4 h to achieve 2. CFA: calcium flouro-aluminate cement (Italcementi
a final set (Taylor 1997). The rate of reaction is SPA, Bergamo, Italy).
controlled by the calcium sulphate (gypsum) which is A super-plasticizing admixture (Degussa Construc-
added by the manufacturer for that purpose. The tion Chemicals, Manchester, UK) was added to the
calcium sulphate reacts with the tricalcium aluminate mixing water to increase the workability of the mix and
in Portland cement to produce a high-sulphate calcium also to reduce the amount of water required by the
sulpho-aluminate (ettringite). In the absence of calcium cement during mixing. A cement composite made up of
sulphate, the tricalcium aluminate forms a hydrated the cement mixed with granite (Halmann, Iklin, Malta)
calcium aluminate of cubic structure leading to a flash was also tested (CSAG, CFAG). The granite used was
set. Thus, the calcium sulphate is added by the obtained by grinding a slab to very fine particles
manufacturer to avoid the formation of the hydrated retained on a 150 lm sieve. The proportion of cement
calcium aluminate thus delaying the hydration mech- and granite was calculated using the Theory of Particle
anism (Taylor 1997). Admixtures (Dewar 1999). Glass–ionomer cement
The setting time of Portland cement can be reduced (GIC; Ketac Molar, 3M Seefeld, Germany) was used as
by addition of an accelerator, the removal of gypsum the control.
from the cement and addition of calcium aluminate
cement (Neville 1981). Removal of gypsum results in a
Setting time of cements
flash set which can be controlled by superplasticizing
admixtures (Camilleri et al. 2005). When calcium The setting time of the cements with and without the
aluminate cement is added to Portland cement the addition of the super-plasticizing admixture was tested
former reacts to the calcium sulphate (gypsum) in the using the Vicat apparatus (British Standard Institution
Portland cement leaving the tricalcium aluminate free 2005b). Four hundred grams of the cement was mixed
to flash set, thus resulting in a shorter setting time with water on a moistened aluminum tray with a small
(Neville & Wainwright 1975). Cements produced by trowel at a water/cement ratio of 0.25 as suggested by
addition of calcium aluminate cement to Portland the manufacturer. The super-plasticizing admixture
cement have a reduced compressive strength (Camilleri was added at the end of the mixing process. The dosage
et al. 2006). The compressive strength of the cement was adjusted to 0.8 L per 100 kg of cement, again as
can be improved by reduction in the water to cement suggested by the manufacturer. The mix was then
ratio. This is achieved through the use of super- compacted in a two-part brass mould and initial and
plasticing admixtures. The plasticizing action of final setting times were determined. The Vicat assembly
water reducers is related to their adsorption and was placed in an incubator at 37 C. The test was
dispersing effects in the cement–water system (Hattori repeated thrice.
1978).
The aim of the study was to test the suitability of two
Flexural and compressive strength
cement types based on Portland cement and to improve
their properties by admixing them with an inert The flexural and compressive strengths of the cements
material. The physical and mechanical properties of were tested according to British Standard Institution
these materials were evaluated and compared with (2005a). A total of 270 g of cement was mixed
proprietary brand glass–ionomer cement. thoroughly with water at the pre-determined water/
cement ratio for 2 min. The cement was then loaded
into a prism mould 40 · 40 · 160 mm in size previ-
ously coated with mould oil (Separol, Sika, Switzer-
Two cement types were used:. land). The cement was compacted in two layers and a
1. CSA: calcium sulpho-aluminate cement mixed in vibrating table was used during the compaction to
the following proportions avoid air entrapment in the cement mix. After casting
Camilleri MTA modification
was complete, the moulds were covered with a plastic particle packing and optimal physical properties were
sheet to avoid loss of moisture and the cements were thus calculated. The cement composite was tested in
allowed to cure for 24 h. Twenty-four prisms were cast both flexure and compression in the same way as
of each cement type. They were then removed from the described for the cements.
mould and cured at 37 C in water. Flexural strength
as performed after 1 day, and then after 7, 28 and
Solubility
56 days. The prisms were mounted in a flexure testing
jig. This jig allowed for 3-point bending to be The solubility of CSA, CFA, granite admixed cements
performed. The distance between the lower rollers (CSAG, CFAG) and glass–ionomer cement was calcu-
was 100 mm and the upper roller was placed in the lated by measuring the water uptake of the cements.
middle. The flexure jig was mounted between the Twelve samples each 10 mm in diameter and 2 mm
platens of a compression machine (Controls spa, thick were made for each cement type. Six were cured
Bergamo, Italy). A 15 kN load cell was used and the in water and the rest were cured at 100% humidity at
load applied was accurate to 0.1%. The cements were 37 C. The weight of the samples was recorded after 1,
loaded at 50 N s)1 until they failed. Flexural strength 7, 28 and 56 days.
was calculated using the following formula:
1:5 Ff l Statistical analysis
Rf ¼
b3
For all tests carried out, the data were evaluated using
where Rf is the flexural strength in N mm)2, Ff is the
statistical package for the social sciences (SPSS
load applied to the middle of the prism at fracture, l is
Inc., Chicago Ill, USA) software. The distribution was
the distance between the supports in mm and b is the
first evaluated to determine the appropriate statistical
side of the square section in mm.
test. The data were plotted and the distribution curve
After flexural strength was determined the prism
was analysed together with the Kolmogorov–Zmirnov
halves were tested in compression at 14.4 kN s)1. A
test with P = 0.05. The P-value of >0.05 signified a
250 kN load cell was used for this test with an
normal distribution and thus parametric tests were
accuracy up to 0.1%. Compressive strength was
performed. With normally distributed data, analysis
calculated using the following formula:
of variance (anova) with P = 0.05 was performed
Fc to evaluate variations between the means. Then
Rc ¼
1600 two-tailed independent sample t-tests at 95% confidence
where Rc is the compressive strength in N mm)2, Fc is level with P = 0.05 was used to analyse the data.
the maximum load at fracture in N,
1600 = 40 mm · 40 mm is the area of the platens in
Results
mm2.
In addition to the cements, cement composite was
Setting time of cements
also tested. The composite was produced by addition of
granite to the two cement types. The setting times of the cements is shown in Table 1.
Preliminary tests to establish the relative density Both cement types had a very rapid set, with CFA
(British Standard Institution 1999b), the loose bulk setting faster than CSA; however, this difference was
density (British Standard Institution 1998), the mois-
ture content (British Standard Institution 1999a) and Table 1 Means of initial and final setting times of different
the water absorption (British Standard Institution cement types ± SD (n = 3)
2000) of the granite were performed. Tests on cement
Admixture %
were carried out to establish the relative density (British Cement Water weight of Initial set Final set
Standard Institution 1992), the moisture content type % fine material minutes minutes
(British Standard Institution 1999a) and the fineness CSA 28 0.5 2.41 ± 0.37 4.05 ± 0.09
(British Standard Institution 1992). After performing CFA 27 0.5 1.83 ± 0.29 3.08 ± 0.14
the preliminary tests on the cements and aggregate, the CSA and 25 0.5 3.67 ± 0.14 5.25 ± 0.25
proportion of granite to cement was deduced using the admixture
CFA and 25 0.5 3.05 ± 0.18 5.42 ± 0.14
Theory of Particle Admixtures (Dewar 1999). Quanti-
admixture
ties of each material required to achieve the best
not statistically significant. The use of a super-plasti- flexural strengths of the cement and the cement
cizing admixture retarded the final setting time of the composite were the same at 56 days (P > 0.05). The
cements, reduced the water required by the cements compressive strength of the calcium fluoro-aluminate
and improved the handling properties. cement did not change markedly with addition of
granite. All the cements tested failed in the normal
mode.
Flexural and compressive strength
The results for the physical properties of the cements
Solubility
and aggregate are shown in Table 2. The moisture
content of the cements was minimal, thus they were The results for the water uptake and loss of the cements
used without desiccation. The quantities of material and cement composites is shown in Table 4. The CFA
required to produce the cement composite as calculated cement absorbed more water than the CSA cement at
by the Theory of Particle Admixtures (Dewar 1999) are all ages (P < 0.01). Both cement composites absorbed
shown in Table 3. less water in comparison with the parent cement
The results for the flexural strength testing of the (P < 0.001). The glass–ionomer cement did not show a
cements are shown in Fig. 1a and for the compressive progressive water uptake or loss with time with the
strength tests in Fig. 1b. The CSA cement was stronger peak water absorption being at 7 days of curing. The
than the CFA cement both in compression and in CSA cement and composite were quite stable and both
flexure at all curing times (P < 0.001) except for lost minimal amounts of water when cured at 100%
flexural strength at 1 day where no difference between humidity (P > 0.05). Both the CFA cement and com-
the two cements was demonstrated (P > 0.05). Addi- posite exhibited extensive water loss at all ages.
tion of granite to the CSA cement resulted in an
increase in flexural strength at 28 and 56 days but a
Discussion
reduction in the compressive strength at all ages
(P < 0.001). Addition of granite to CFA increased the Compressive strength testing of dental cements is usually
flexural strength up to 28 days (P < 0.01) but the performed according to British Standard Institution
(2003). In a previous paper (Camilleri et al. 2006)
testing various proprietary brands, it was concluded that
Table 2 Results of preliminary tests carried out on (a)
granite ± SD; (b) cements ± SD compressive strength testing depends on specific test
conditions. The cements were not susceptible to changes
Test Result
in the compressive strength testing procedure at 1 and
(a) 7 days, but at 28 days all the fast setting cements had a
Relative density kg m)3 2.68 ± 0.56
significantly higher strength with a particular testing
Loose bulk density g ml)1 1.11 ± 0.1
Moisture content % 0.02 ± 0.01
mode. Abnormal failure occurred when testing accord-
Water absorption % 1.4 ± 0.19 ing to British Standard Institution (2003) and the
standard deviations for the tests carried out were large.
Cement type
Compressive strength testing of prototype dental
Test CSA CFA OPC cements was thus performed according to British
(b) Standard Institution (2005a).
Relative density kg m)3 2.95 ± 0.08 2.85 ± 0.04 2.94 ± 0.01 The setting time of the cement was controlled by
Moisture content % 0.08 ± 0.06 0.06 ± 0.01 0.54 ± 0.05
addition of calcium aluminate. The gypsum present in
Fineness m2 kg)1 450 450 421 ± 0.9
the Portland cement system reacts with the calcium
The fineness of the CSA and CFA was determined by the
aluminate to produce ettringite as the main by-product
manufacturer.
of the hydration reaction. Once the gypsum is depleted,
the calcium silicate hydration reaction is accelerated.
Table 3 Percentage ratio of materials used in the mix design
This has already been investigated in a previous paper
Cement type % cement % granite % water % admixture (Camilleri et al. 2006) where Portland cement was
CSA 100 0 25.00 0.5 admixed with calcium aluminate cement; This resulted
CFA 100 0 25.00 0.5 in poor compressive strengths. In the present experi-
CSAG 58.25 41.60 15.19 0.5
ment, excess sulphate ions provided by the synthetic
CFAG 58.64 41.36 15.28 0.5
anhydrite reacted with the calcium hydroxide produced
(a) 18 1 day
7 days
16 28 days
56 days
Flexural strength N mm-2

14
12
10
0
CSA CFA CSAG CFAG
Cement type
(b)140 1 day
7 days
28 days
120
56 days
Compressive strength N mm-2
100
80
60
40
20
0
CSA CFA CSAG CFAG
Cement type
Figure 1 (a) Mean flexural strength testing of cements and cement composites over a period of 56 days ± SD (n = 6). (b) Mean
compressive strength testing of cements and cement composites over a period of 56 days ± SD (n = 12).
Table 4 Percentage water absorption/shrinkage of cements cured at 100% humidity or immersed in water over a period of
56 days ± SD (n = 6)
%water absorption of cements cured in water %water absorption of cements cured at 37 C and
at 37 C/days 100% humidity/days
Cement type 1 7 28 56 1 7 28 56
CSA 4.16 ± 0.6 4.81 ± 0.5 5.60 ± 0.6 5.90 ± 0.8 )0.80 ± 0.1 )1.09 ± 0.1 )1.62 ± 0.2 )1.97 ± 0.3
CSAG 0.99 ± 0.1 1.29 ± 0.1 2.17 ± 0.1 2.72 ± 0.1 )0.63 ± 0.2 )1.14 ± 0.1 )1.21 ± 0.1 )1.37 ± 0.2
CFA 5.61 ± 0.3 7.57 ± 0.3 7.76 ± 0.2 8.32 ± 0.2 )7.27 ± 0.5 )7.37 ± 0.6 )6.64 ± 0.8 )6.43 ± 0.8
CFAG 2.22 ± 0.6 3.23 ± 0.8 3.22 ± 0.7 3.57 ± 0.7 )5.10 ± 0.2 )5.14 ± 0.1 )4.83 ± 0.2 )4.81 ± 0.2
GIC 1.32 ± 0.0 5.05 ± 2.1 2.92 ± 0.1 3.32 ± 0.2 )2.17 ± 0.3 )3.36 ± 0.4 )3.19 ± 0.4 )3.87 ± 0.5
by the hydration of the tricalcium silicate that increased development seemed to show a maximum within the
the ettringite deposition and resulted in an increase in first few days. A qualitative explanation of this behaviour
initial strength (Knöfel & Wang 1993). Strength is based on a combined process of expansive cracking and
self-healing, the balance of which appears to determine glass–ionomer cement. Addition of granite to the
the course of expansion and strength development cements reduced the water exchange within the
(Bentur et al. 1975). The extent of ettringite-related material. Both calcium sulpho-aluminate cement and
expansion depends on the conditions of curing. It its composite lost less water when cured at 100%
increases with increasing amounts of water taken up humidity than glass–ionomer cement. The results
from the environment while ettringite is formed, indi- obtained for the prototype dental cements are compa-
cating the participation of swelling phenomena in the rable with those of proprietary brand glass–ionomer
expansion process (Odler & Colán-Subauste 1999). cement (McCabe et al. 1990, Pereira et al. 2002,
Curing in water enhances early ettringite formation Piwowarczyk et al. 2002).
and increase in strength. The fluoro-aluminate cement
contains sufficient tricalcium silicate to achieve high
Conclusions
early strengths (Costa et al. 2000). Utilization of calcium
fluoro-aluminate leads to slow ettringite formation, The CSA and CFA cements had adequate setting times
which could explain the low compressive strength as and compressive strength values when compared with
compared with the sulpho-aluminate cement. proprietary brand glass–ionomer cement. CSA was
Attempts at improving the material strength were superior to CFA and had more promise as a prospective
made by using a superplacticizing admixture and by dental material. The improved properties of the modified
addition of inert filler. The superplasticizing admixture cement could allow for a reduction in the chair-side time
causes particle dispersion and allows workability at low and amount of visits per treatment when the cement is
water/cement ratios. Maximum particle packing is used for apexification procedures and pulpotomies, and
attained by dispersion of the cement grains at low enable the material to withstand packing pressures of
water/cement ratios with the use of high levels of a restorative materials if used as a pulp capping agent.
dispersing agent. There is a practical limit to the
reduction of water/cement ratio as the mixture would
Acknowledgements
become unworkable. The flocculated nature of the
cement particles does not permit fluid suspensions The University of Malta for funding. Dr J. Newman of
unless they are dilute (Malhotra et al. 1978). In the Department of Civil and Environmental Engineer-
addition, the cement was mixed with an inert material ing, Imperial College London for his help with the mix
of different particle size in an attempt to improve its design, Mr Gavin Gartshore at Lafarge Special Cements
flexural strength. In a particle admixture macroscopic section, Nottingham, UK and Dr Umberto Costa of
porosity remains as a result of air entrainment and Italcementi SPA, Italy for providing the cement sam-
defective particle packing. To attain the ideal structure ples, Mr Roy Jones of Degussa Construction Chemicals,
of strong cement paste, maximum particle packing with Manchester, UK for providing the admixture and 3M
minimal total porosity and elimination of macroscopic Healthcare Limited, Leicestershire, UK for providing the
voids is necessary. The use of cement on its own to glass–ionomer cement. Mr Nicholas Azzopardi for his
produce any kind of material is difficult because of help with the laboratory work.
development of microcracks. The addition of an aggre-
gate is thus mandatory. The strength of a composite
References
material is influenced by the ratio of cement to mixing
water, the ratio of cement to aggregate, the grading, Bakland LK (2000) Management of traumatically injured
surface texture, shape, strength, and stiffness of aggre- pulps in immature teeth using MTA. Journal of the Californian
gate particles, and the maximum size of the aggregate Dental Association 28, 855–8.
(Gilkey 1961). The Theory of Particle Admixtures Bentur A, Ish-Shalom M (1975) Properties of type K expansive
cement of pure components IV. Hydration of mixtures of C3S
(Dewar 1999) utilizes parameters calculated by phys-
with pure expansive component. Cement and Concrete
ical tests carried out on the raw materials to calculate
Research 5, 597–606.
the percentages of materials required to produce
British Standard Institution (1992) Methods for testing
optimal particle packing for maximum strength. cements: determination of fineness. BS EN 196-6.
Glass–ionomer cements exhibited hygroscopic British Standard Institution (1998) Tests for mechanical and
expansion on immersion in water. This expansion physical properties of aggregates. Part 3: Determination of
reached its peak after 7 days immersion. Both proto- loose bulk density and voids. BS EN 1097-3.
type cements showed higher water absorption than
British Standard Institution (1999a) Tests for mechanical and Lee SJ, Monsef M, Torabinejad M (1993) Sealing ability of a
physical properties of aggregates. Part 5: Determination of mineral trioxide aggregate for repair of lateral root perfora-
water content by drying in a ventilated oven. BS EN 1097-5. tions. Journal of Endodontics 19, 541–4.
British Standard Institution (1999b) Tests for mechanical Malhotra VM, Berry EE, Wheat TA (1978) Developments in
and physical properties of aggregates. Part 7: Determina- the use of superplasticizers. Proceedings from the 5th Inter-
tion of particle density of filler: pyknometer method. BS national Symposium on Superplasicizers in Concrete. Farming-
EN 1097-7. ton Hills MI, USA: American Concrete Institute.
British Standard Institution (2000) Tests for mechanical and McCabe JF, Watts DC, Wilson HJ, Worthington HV (1990) An
physical properties of aggregates. Part 6: Determination of investigation of test-house variability in the mechanical
particle density and water absorption. BS EN 1097-6. testing of dental materials and the statistical treatment of
British Standard Institution (2003) Dentistry: water-based results. Journal of Dentistry 18, 90–7.
cements. Part 1: Powder/liquid acid–base cements. BS EN Neville AM (1981) Properties of Concrete, 3rd edn. UK:
ISO 9917-1. Longman Scientific and Technical Essex.
British Standard Institution (2005a) Methods for testing Neville AM, Wainwright PJ (1975) High Alumina Cement
cements: determination of strength. BS EN 196-1. Concrete. London: The Construction Press.
British Standard Institution (2005b) Methods of testing cement. O’Sullivan SM, Hartwell GR (2001) Obturation of a retained
Determination of setting time and soundness. BS EN 196-3. primary mandibular second molar using mineral trioxide
Camilleri J, Montesin FE, Di Silvio L, Pitt Ford TR (2005) The aggregate: a case report. Journal of Endodontics 27, 703–5.
constitution and biocompatibility of accelerated Portland Odler I, Colán-Subauste J (1999) Investigations on cement
cement for endodontic use. International Endodontic Journal expansion associated with ettringite formation. Cement and
38, 834–42. Concrete Research 29, 731–5.
Camilleri J, Montesin FE, Curtis RV, Pitt Ford TR (2006) Pereira LC, Nunes MC, Dibb RG, Powers JM, Roulet JF,
Characterization of Portland cement for use as a dental Navarro MF (2002) Mechanical properties and bond
restorative material. Dental Materials 22, 569–75. strength of glass–ionomer cements. Journal of Adhesive
Costa U, Cucitore R (2000) Rapid hardening cements: new Dentistry 4, 73–80.
applications. Proceedings of the 7th NCB International Seminar Pitt Ford TR, Torabinejad M, McKendry DJ, Hong CU,
on Cement and Building Materials, New Delhi: National Kariyawasam SP (1995) Use of mineral trioxide aggregate
Council of Building Materials. for repair of furcal perforations. Oral Surgery, Oral Medicine,
Dewar JD (1999) Computer Modeling for Concrete Mixtures. Oral Pathology, Oral Radiology and Endodontics 79, 756–63.
London: E and FN Spon. Pitt Ford TR, Torabinejad M, Abedi HR, Bakland LK, Kariy-
Eidelman E, Holan G, Fuks AB (2001) Mineral trioxide awasam SP (1996) Using mineral trioxide aggregate as a
aggregate vs. formocresol in pulpotomized primary molars: pulp-capping material. Journal of the American Dental Asso-
a preliminary report. Paediatric Dentistry 23, 15–8. ciation 127, 1491–4.
Geurtsen W (2000) Biocompatibility of root canal filling Piwowarczyk A, Ottl P, Lauer HC, Buchler A (2002) Labora-
materials. Australian Dental Journal 27, 12–21. tory strength of glass ionomer cements, compomers and
Gilkey HJ (1961) Water/cement ratio versus strength – resin composites. Journal of Prosthetics 11, 86–91.
another look. Journal of the American Concrete Institution Shabahang S, Torabinejad M (2000) Treatment of teeth with
58, 18511–78. open apices using mineral trioxide aggregate. Practical
Hattori K (1978) Superplasticizers in concrete. Proceedings from Periodontics Aesthetic Dentistry 12, 315–20.
the International Symposium, Ottawa, Canada, Hawthorn, Taylor HFW (1997) Cement Chemistry, 2nd edn. London:
Australia: Gordon and Breach Science Publishers, 49–86. Thomas Telford.
Hayashi M, Shimizu A, Ebisu S (2004) MTA for obturation of Torabinejad M, White DJ (1995) Tooth filling material and
mandibular central incisors with open apices: case report. use. US patent number 5,769,638.
Journal of Endodontics 30, 120–2. Torabinejad M, Hong CU, McDonald F, Pitt Ford TR (1995)
Holland R, de Souza V, Nery MJ, Otoboni Filho JA, Bernabe PF, Physical and chemical properties of a new root-end filling
Dezan E Jr (1999) Reaction of dogs’ teeth to root canal filling material. Journal of Endodontics 21, 349–53.
with mineral trioxide aggregate or a glass ionomer sealer. Vizgirda PJ, Liewehr FR, Patton WR, McPherson JC, Buxton
Journal of Endodontics 25, 728–30. TB (2004) A comparison of laterally condensed gutta-
Holland R, de Souza V, Murata SS et al. (2001) Healing percha, thermoplasticized gutta-percha, and mineral triox-
process of dog dental pulp after pulpotomy and pulp ide aggregate as root canal filling materials. Journal of
covering with mineral trioxide aggregate and Portland Endodontics 30, 103–6.
cement. Brazilian Dental Journal 12, 109–13. Whiterspoon DE, Ham K (2001) One-visit apexification:
Knöfel D, Wang JF (1993) Early hydration of quick cements in technique for inducing root-end barrier formation in apical
the system C11A7ÆCaF2–C2S–C2(A,F). Advances in Cementing closures. Practical Procedures and Aesthetic Dentistry 13,
Materials 1, 77–82. 455–60.
doi:10.1111/j.1365-2591.2008.01436.x
Neuropeptide Y Y1 receptor in human dental pulp

cells of noncarious and carious teeth
I. A. El Karim, P.-J. Lamey, G. J. Linden & F. T. Lundy

Oral Science Research Centre, School of Medicine and Dentistry, Queen’s University, Belfast, Northern Ireland, UK
Abstract levels in noncarious, moderately carious and grossly

carious teeth.
El Karim IA, Lamey P-J, Linden GJ, Lundy FT. Neuropep-
Results Neuropeptide Y Y1 receptor immunoreactiv-
tide Y Y1 receptor in human dental pulp cells of noncarious and
ity was detected on the walls of blood vessels in pulp
carious teeth. International Endodontic Journal, 41, 850–855,
tissue from noncarious teeth. In carious teeth NPY Y1
2008.
immunoreactvity was observed on nerve fibres, blood
Aim To determine the distribution of the NPY Y1 vessels and inflammatory cells. Western blotting indi-
receptor in carious and noncarious human dental pulp cated the presence and confirmed the variability of NPY
tissue using immunohistochemistry. A subsidiary aim Y1 receptor protein expression in solubilised membrane
was to confirm the presence of the NPY Y1 protein preparations of human dental pulp tissue from carious
product in membrane fractions of dental pulp tissue from and noncarious teeth.
carious and noncarious teeth using western blotting. Conclusions Neuropeptide Y Y1 is expressed in
Methodology Twenty two dental pulp samples were human dental pulp tissue with evidence of increased
collected from carious and noncarious extracted teeth. expression in carious compared with noncarious teeth,
Ten samples were processed for immunohistochemistry suggesting a role for NPY Y1 in modulation of caries
using a specific antibody to the NPY Y1 receptor. induced pulpal inflammation.
Twelve samples were used to obtain membrane extracts
Keywords: caries, dental pulp, human, NPY, NPY
which were electrophoresed, blotted onto nitrocellulose
Y1, receptor.
and probed with NPY Y1 receptor antibody. Kruskal–
Wallis one-way analysis of variance was employed to Received 11 September 2005; accepted 10 April 2008
test for overall statistical differences between NPY Y1
ropeptides cannot cross cell membranes and therefore

Introduction
to exert their biological effects they must bind to
Data is emerging that supports the view that neuro- selected receptors on target cell membranes. More
peptides have a role in the pulpal response to caries. It than 80% of neuropeptide receptors are G protein
has been shown that the levels of neuropeptides such coupled receptors, so-called because they bind guano-
as substance P (SP), neurokinin A (NKA), calcitonin sine triphosphate and act as intermediaries between
gene-related peptide (CGRP), vasoactive intestinal the receptor and several second messenger systems.
polypeptide (VIP) and neuropeptide Y (NPY) are Neuropeptides usually have several different iso-recep-
elevated in the dental pulps of carious compared with tors or receptor subtypes that can be distinguished by
noncarious teeth (Rodd & Boissonade 2000, Awawdeh specific agonists, antagonists or antibodies. The con-
et al. 2002, El Karim et al. 2006a,b). However, neu- cept of multiple types of receptors explains subtle
differences in neuropeptide effects within various
tissues.
Correspondence: Fionnuala Lundy, School of Medicine and
Dentistry, Queen’s University, Grosvenor Road, Belfast BT12
Neuropeptide Y exerts its actions through at least five
6BP, Northern Ireland, UK (Tel.: +44 28 9063 3875; fax: +44 different receptors subtypes (Michel 1991). To date the
28 90 438861; e-mail: f.lundy@qub.ac.uk). most important and widely studied NPY receptor
El Karim et al. NPY Y1 in human dental pulp
subtypes are NPY Y1 and NPY Y2. Both NPY Y1 and with fine tweezers within 1–2 min. Pulp tissue
NPY Y2 have been reported to mediate the vascular required for immunohistochemistry was immediately
effects of NPY (Nilsson et al. 1996, Rump et al. 1997) fixed in 4% paraformaldehyde in phosphate buffered
and NPY Y1 is known to have a widespread distribu- saline (PBS). The pulp tissue required for western
tion in the peripheral circulation (Uddman et al. 2002). blotting was placed in a pre-weighed Eppendorf tube,
In addition NPY Y1 receptors are believed to be immediately frozen in liquid nitrogen and stored at
involved in the pain modulatory effects of NPY (Zhang )70 C. The split halves of the teeth were used to
et al. 2000, Wang et al. 2001). visually assess the extent of caries using a dissecting
Neuropeptide Y receptor subtypes have been dem- microscope at 20· magnification. Each tooth was
onstrated in a variety of different tissues, where they categorised as: (1) noncarious, i.e. no colour change
have been localised to blood vessel walls, nerve fibres, indicative of caries within dentine; (2) mild/moderate
epithelium and endocrine-like cells (Jackerott & Larsson caries, i.e. the colour change did not extend beyond
1997, Mannon et al. 1999, Matsuda et al. 2002, half the dentine thickness; (3) advanced/gross caries,
Uddman et al. 2002). Although mRNA encoding the i.e. the colour change extended beyond half the
NPY Y1 receptor has been demonstrated in human dentine thickness.
dental pulp (Uddman et al. 1998), evidence for the
expressed NPY Y1 protein product and its cellular
Immunohistochemistry
localisation remain to be reported.
The aim of this study was to determine the Pulp tissue from carious (n = 5) and non carious
distribution of the NPY Y1 receptor in carious and (n = 5) teeth was processed for immunohistochemis-
noncarious human dental pulp tissue using immu- try. Pulp tissue which had been fixed in 4%
nohistochemistry. A subsidiary aim was to confirm paraformaldehyde in PBS for 4 h at 4 C, was
the presence of the NPY Y1 protein product in washed in 5% sucrose for 20 min and cryoprotected
membrane fractions prepared from dental pulp tissue in 30% sucrose until submerged. Specimens were
of carious and noncarious teeth using western frozen onto stubs using embedding media (O.C.T.
blotting. compound) and sections (10 lm) were cut using a
cryostat and collected onto 3-aminopropyltriethoxysi-
lane (APES) coated slides. Specimens were then
processed for immunohistochemical staining for
NPY Y1 receptor using the Dako Envision system
Clinical samples
(Dako Co., Carpinteria, CA, USA). After rehydration
The study was approved by the local Research Ethics in PBS for 15 min, sections were incubated in 10%
Committee, Queen’s University Belfast, UK and all normal goat serum for 30 min, permeablised in 0.2%
subjects (n = 22) gave their informed written consent. Triton X-100 for 1 h and then treated with perox-
Carious teeth (range from superficial to deep caries) idase block reagent (0.03% hydrogen peroxidase
were collected from patients requiring permanent tooth containing sodium azide) for 5 min. The sections
extraction for various therapeutic reasons. Exclusion were washed in PBS and incubated with polyclonal
criteria for carious teeth were teeth from patients with rabbit anti-human NPY Y1 (DiaSorin Ltd, Woking-
clinically significant medical problems, teeth that had ham, UK) at a dilution of 1/100 in PBS overnight at
undergone any form of endodontic therapy, teeth with 4 C. Following a washing step in PBS, specimens
associated pathosis other than dental caries and teeth were incubated with secondary anti-rabbit peroxi-
with periapical pathosis suggestive of pulp necrosis. dase-labelled polymer (peroxidase labelled polymer
Control teeth with no caries were obtained from conjugated to goat anti-rabbit immunoglobulins in
patients having extractions for orthodontic reasons or Tris-HCl buffer) for 1 h at room temperature. The
having wisdom teeth removed. substrate diaminobenzidine (DAB) was added to
All extractions were performed under local anaes- visualise bound secondary antibody. The slides were
thesia (2% lignocaine 1 : 80 000 epinephrine) and incubated with DAB substrate solution for 2–3 min
were uncomplicated, as only fully erupted teeth were and then washed in distilled water before being
included in the study. Immediately following extrac- counterstained with Harris’ haematoxylin (BDH,
tion, each tooth was split in a vice fitted with a cutting Poole, UK) for 1 min. Sections were dehydrated in
edge (Lilja 1979) and the pulp tissue was removed degrading alcohols, cleared in xylene and mounted in
NPY Y1 in human dental pulp El Karim et al.
DPX (Searle Diagnostic, UK). Pre-immune rabbit

Results
serum was used as a negative control in place of
the primary antibody.
Immunohistochemistry
Neuropeptide Y Y1-immunoreactivity (Ir) was detected
Western blotting
in the dental pulp of both noncarious and carious teeth.
Since the NPY Y1 receptor is an integral membrane In the dental pulp of noncarious teeth NPY Y1-Ir was
protein, it was necessary to prepare membrane demonstrated in the walls of small and medium sized
extracts from the dental pulp tissue prior to electro- vessels in the coronal, subodontoblastic and central
phoresis. Membrane extracts were prepared from the pulp regions. The endothelial cells of many small blood
dental pulps of noncarious (n = 4), moderately carious vessels were intensely stained (Fig. 1a). In the dental
(n = 4) and grossly carious teeth (n = 4) using the pulp of carious teeth there was variable expression of
Mem-PER eukaryotic membrane protein extraction NPY Y1-Ir, where NPY Y1-Ir was detected in walls of
reagent kit (Pierce, Rockford, IL, USA). The hydropho- blood vessels as well as in a subpopulation of nerve
bic membrane protein phase obtained from the mem- fibres scattered throughout the subodontoblastic layer
brane extraction kit was further prepared for and pulp region proper (Fig. 1b,c). In some of the
electrophoresis using PAGEprep clean up and enrich- grossly carious teeth positive staining was also
ment kit (Pierce, Rockford, IL, USA). Equal volumes observed in inflammatory cells with large and /or
of samples (15 lL) were electrophoresed and blotted multi-lobed nuclei resembling lymphocytes and poly-
onto nitrocellulose as previously described (Lundy & morphonuclear leukocytes (Fig. 1d). The specificity of
Wisdom 1992). The blot was blocked for 2 h in a the antibody-antigen interaction was confirmed by
solution of 5% non fat milk in tris-buffered saline (TBS; absence of immunoreactivity in samples incubated with
0.02 mol L)1 Tris-HCL buffer, pH 7.4, containing pre-immune rabbit serum (results not shown).
0.15 mol L)1 NaCl). After washing in TBS the blot was
incubated with polyclonal rabbit anti-human NPY Y1
Western blotting
(Abcam, Cambridge, UK) at a dilution of 1/2000
overnight at room temperature. Subsequent detection The presence of NPY Y1 receptor proteins in solubilised
of bound primary antibody was achieved by incuba- membrane preparations of human dental pulp tissue
tion for 2 h with anti-rabbit immunoglobulin-alkaline was confirmed by western blotting. A major immuno-
phosphatase conjugate (Sigma Chemical Co., Gilling- reactive band was detected at approximately 55 kDa in
ham, UK) at a dilution of 1/1000. Bound alkaline all samples derived from both intact and carious teeth.
phosphatase was detected as previously described There was considerable variability in the density of the
(Lundy & Wisdom 1992). Samples which were immunoblots within the groups of healthy or carious
electrophoresed, blotted onto nitrocellulose and probed teeth (Fig. 2). There was a trend for the mean blot
as described above in the absence of primary antibody density to increase from a low value of 0.06 (SD 0.04)
served as controls for nonspecific binding. The inten- in healthy through an intermediate value in moder-
sity of the immunoreactive bands was measured using ately carious (0.11 SD 0.05) to the highest value in
a BioRad GS-670 Imaging Densitometer. Each band membrane extracts of pulp tissue from grossly carious
was scanned in triplicate and the intensity of the teeth (0.17 SD 0.11) (Fig. 3). These differences, how-
background was subtracted before the result was ever, were not statistically significant. Control blots
recorded. incubated in the absence of primary antibody showed
no immunoreactive bands (results not shown).
Statistical analysis
Discussion
Data were analyzed using the statistical package SPSS
version 15.0. The level of statistical significance for all The immunohistochemical localisation and distribution
tests was set at P < 0.05. Kruskal–Wallis one-way of NPY Y1 receptors has previously been reported in
analysis of variance was employed to test for overall various tissues but to date, not in the dental pulp. The
statistical differences between NPY Y1 levels in present study demonstrated that NPY Y1 receptor
noncarious, moderately carious and grossly carious protein is present in human dental pulp and its
teeth. expression varies in health and disease. NPY Y1-Ir
(a) (b)
(c) (d)
Figure 1 NPY Y1 receptor expression in (a) a small blood vessel [BV] in noncarious pulp tissue, (b) a blood vessel [BV] and nerve
fibres [NFs] in moderately carious pulp tissue, (c) nerve fibres [NFs] in grossly carious pulp tissue, (d) inflammatory cells [ICs] in
grossly carious pulp tissue.
(a)
55 kDa 0.2500
0.2000
(b)
NPYY1
0.1500
Figure 2 Representative western blots of NPY Y1 receptor

proteins in membrane extracts prepared from human dental 0.1000
pulp of carious and noncarious teeth. (a) noncarious pulp
samples. (b) Carious pulp samples.
0.0500
was shown to be localised in the smooth muscles cells

of smaller arteries as compared to large arteries and Non carious Moderate caries Gross caries
veins suggesting an important role for NPY in small Pulpstatus
resistance vessels (Matsuda et al. 2002, Uddman et al.
Figure 3 Scattered plot showing intensity of NPY Y1-Ir band
2002). The fact that the largest vessels found in the
staining according to pulp status.
human dental pulp are arterioles, together with the
currently reported localisation of NPY Y1 in pulpal et al. 2006a) support an important role for NPY acting
blood vessels and the previous evidence for an associ- via Y1 receptors in pulpal haemoregulation. In carious
ation of NPY fibres with pulpal blood vessels (El Karim teeth NPY Y1 receptor was found to be localised not
NPY Y1 in human dental pulp El Karim et al.
only to the walls of blood vessels but to be distributed Another proposed mechanism for NPY Y1-mediated
more widely in nerves fibres and in grossly carious inhibition of inflammation in immune cells has been
teeth in inflammatory cells as previously described in reported. In Y(1)-deficient mice Wheway et al. (2005)
other tissues (Mannon et al. 1999, Uddman et al. demonstrated a bimodal role for Y1 in immune cells.
2002). Whereas its presence is essential for the function of the
Neuropeptide Y Y1 has been shown to play an antigen presenting cells, NPY Y1 was also found to
essential role in mediating the inhibitory effects of NPY inhibit T cell response. Thus, signalling via the Y1
in neurogenic inflammation. Using Y(1)-deficient mice receptor is necessary for the proper functioning of
an essential role for NPY Y1 in mediating plasma antigen presenting cells as priming elements for T cells
extravasation and antinociception has been reported and at the same time it appears to be protective against
(Naveilhan et al. 2001). Deletion of NPY Y1 receptors excessive inflammation by inhibiting hyperresponsive-
has also been shown to be associated with increased ness of T cells.
release of SP and CGRP (Shi et al. 2006) as did the use
of NPY Y1 antagonists (Gibbs et al. 2006). It has been
Conclusion
shown previously that NPY co-localises with SP in
nerves in the dental pulp of carious teeth (El Karim Neuropeptide Y Y1 was expressed in human dental
et al. 2006a) and the present localisation of NPY Y1 to pulp with evidence of increased expression in carious
the nerve fibres reported in present study lends support compared with noncarious teeth. NPY Y1 receptors
to the hypothesis that NPY acting via NPY Y1 receptors were localised to nerve fibres and inflammatory cells in
in the dental pulp of carious teeth may could mediate the dental pulp of carious teeth. Taken together with
inhibition of SP and CGRP release and subsequent the prevailing evidence from animal studies on the
inhibition of neurogenic inflammation and pain. function of NPY in pain and inflammation these results
In the present study there was a trend for increased are consistent with a role for this neuropeptide acting
NPY Y1 receptor expression with the degree of caries, through the NPY Y1 receptor in the modulation of
however there was considerable sample variation and pulpal inflammation. Further studies are warranted to
this trend did not reach statistical significant. It has improve understanding of the functional role of NPY
been reported previously that increased expression of acting through the NPY Y1 receptor in pulpal pain and
NPY occurs in the dental pulp of carious teeth (El inflammation.
Karim et al. 2006a) and that there is a tendency for
increased NPY expression in moderately carious teeth.
Acknowledgement
This would fit with an increased role for NPY acting
through the Y1 receptor in the suppression of pulpal The authors are indebted to Ms Catherine Fulton and
inflammation resulting from dental caries. In the Mrs Kathy Pogue, School of Medicine and Dentistry,
present study there was variability in receptor expres- Queen’s University Belfast, for their expert technical
sion which was particularly evident within the grossly assistance.
carious group, which had not only the highest levels of
NPY Y1 but also one very low value similar to that
References
recorded from the healthy pulp samples. This may
reflect the state of the pulp in the grossly carious teeth. Awawdeh L, Lundy FT, Shaw C, Lamey P-J, Linden G J,
Inflammation of the dental pulp commences early Kennedy JG (2002) Quantitative analysis of substance P,
during the caries process, as reversible pulpitis. During neurokinin A and calcitonin gene-related peptide in pulp
a neurogenic inflammatory response in a low compli- tissue from painful and healthy human teeth. International
Endodontic Journal 35, 30–6.
ance environment such as the dental pulp, increased
El Karim IA, Lamey P-J, Linden GJ, Lundy FT (2006a) Caries
plasma extravasation and oedema formation associated
induced changes in the expression of pulpal neuropeptide Y
with SP and CGRP release, can lead to irreversible
(NPY). European Journal of Oral Sciences 114, 133–7.
inflammation and pulpal necrosis. As inflammation El Karim IA, Lamey P-J, Linden GJ, Lundy FT (2006b)
increases then so does NPY Y1 expression. The pulp Vasoactive intestinal polypeptide (VIP) and VAPC1 receptors
tissue harvested from grossly carious teeth may in some in relation to caries. Archives of Oral Biology 51, 849–55.
cases be partly necrotic and therefore immunoblots Gibbs JL, Flores CM, Hargreaves KM (2006) Attenuation of
derived from membrane extracts could as a result have capsaicin evoked mechanical allodynia by peripheral neuro-
low density. peptide Y Y1 receptors. Pain 124, 167–74.
Jackerott M, Larsson LI (1997) Immunocytochemical locali- Rodd HD, Boissonade FM (2000) Substance P expression in
zation of the NPY/PYY Y1 receptor in enteric neurons, human tooth pulp in relation to caries and pain experience.
endothelial cells, and endocrine-like cells of the rat intestinal European Journal of Oral Sciences 108, 467–74.
tract. Journal of Histochemistry and Cytochemistry 45, 1643– Rump LC, Riess M, Schwertfeger E, Michel MC, Bohmann C,
50. Schollmeyer P (1997) Prejunctional neuropeptide Y recep-
Lilja J (1979) Innervation of different parts of the predentin tors in human kidney and atrium. Journal of Cardiovascular
and dentin in young human premolars. Acta Odontologica Pharmacology 29, 656–61.
Scandinavica 37, 339–46. Shi TJ, Li J, Dahlstrom A et al. (2006) Deletion of the
Lundy FT, Wisdom GB (1992) The determination of asialogly- neuropeptide Y Y1 receptor affects pain sensitivity, neuro-
coforms of serum glycoproteins by lectin blotting with Ricinus peptide transport and expression, and dorsal root ganglion
communis algutinin. Clinica Chimica Acta 105, 187–95. neuron numbers. Neuroscience 140, 293–404.
Mannon PJ, Kanungo A, Mannon RB, Ludwig KA (1999) Uddman R, Kato J, Cantera L, Edvinsson L (1998) Localization
Peptide YY/neuropeptide Y Y1 receptor expression in the of neuropeptide Y Y1 receptor mRNA in human tooth pulp.
epithelium and mucosal nerves of the human colon. Archives of Oral Biology 43, 389–94.
Regulatory Peptides 83, 11–9. Uddman R, Moller S, Nilsson T, Nyström S, Ekstrand J,
Matsuda H, Brumovsky PR, Kopp J, Pedrazzini T, Hökfelt T Edvinsson L (2002) Neuropeptide Y Y1 and neuropeptide Y
(2002) Distribution of neuropeptide Y Y1 receptors in Y2 receptors in human cardiovascular tissues. Peptides 23,
rodent peripheral tissues. Journal of Comparative Neurology 927–34.
449, 390–404. Wang JZ, Lundeberg T, Yu LC (2001) Anti-nociceptive effect of
Michel MC (1991) Receptors for neuropeptide Y: multiple neuropeptide Y in periaqueductal grey in rats with inflam-
subtypes and multiple second messengers. Trends in Phar- mation. Brain Research 893, 264–7.
macological Sciences 12, 389–94. Wheway J, Mackay CR, Newton RA et al. (2005) A funda-
Naveilhan P, Hassani H, Lucas G et al. (2001) Reduced mental bimodal role for neuropeptide Y1 receptor in the
antinociception and plasma extravasation in mice lacking a immune system. Journal of Experimental Medicine 202,
neuropeptide Y receptor. Nature 409, 513–7. 1527–38.
Nilsson T, Erlinge D, Cantera L, Edvinsson L (1996) Contractile Zhang Y, Lundeberg T, Yu L (2000) Involvement of
effects of neuropeptide Y in human subcutaneous resistance neuropeptide Y and Y1 receptor in antinociception in
arteries are mediated by Y1 receptors. Journal of Cardiovas- nucleus raphe magnus of rats. Regulatory Peptides 95,
cular Pharmacology 28, 764–8. 109–13.
doi:10.1111/j.1365-2591.2008.01437.x
Substance P influenced gelatinolytic activity via

reactive oxygen species in human pulp cells
F.-M. Wang1, T. Hu1, R. Cheng1, H. Tan2 & X.-D. Zhou1

1
State Key Laboratory of Oral Diseases, Sichuan University, Chengdu, China; and 2Department of Operative Dentistry and
Endodontics, West China College of Stomatology, Sichuan University, Chengdu, China
Abstract evaluate the roles of ROS pathway in mediating the

impact of SP on cellular gelatinolytic activity. Data were
Wang F-M, Hu T, Cheng R, Tan H, Zhou X-D. Substance P
analysed using analysis of variance with Bonferroni
influenced gelatinolytic activity via reactive oxygen species in
correction for multiple comparisons or an unpaired
human pulp cells. International Endodontic Journal, 41, 856–
Student’s t-test.
862, 2008.
Results Substance P, at levels above 1 lmol L)1,
Aim To investigate the effects of substance P (SP) on remarkably enhanced MMP-2 activity reflected by the
gelatinolytic activity of matrix metalloproteinases band migrating at 66 kDa (P < 0.05). A gelatinolytic
(MMPs) in human pulp cells. band at approximately 44 kDa appeared to be intensi-
Methodology Human dental pulp cells were isolated fied in a SP dose-dependent manner. In addition, it was
and cultured. Subconfluent cells, between the third and demonstrated that SP could induce ROS production in
sixth passages, were maintained under serum depriva- pulp cells and ROS scavenger NAC was further found to
tion for 18 h followed by the treatment of varying doses significantly reduce MMP-2 activity (P < 0.05), as well
of SP (1 pmol L)1, 100 pmol L)1, 10 nmol L)1, as other bands of gelatinolytic proteinases.
1 lmol L)1 and 100 lmol L)1). Conditioned media Conclusion Substance P can influence gelatinolytic
were then subjected to gelatin zymography using 8% activity in human pulp cells via ROS pathway.
sodium dodecyl sulphate polyacrylamide gel electropho-
Keywords: gelatinolytic activity, human pulp cells,
resis minigels containing 1.5 g L)1 gelatin. The effect of
matrix metalloproteinases, reactive oxygen species,
SP on intracellular reactive oxygen species (ROS) was
substance P.
also examined by confocal microscopy. ROS scavenger
N-Acetyl-l-cysteine (NAC, 5 mmol L)1) was utilized to Received 7 September 2007; accepted 10 April 2008
reported to have the ability to release SP when

Introduction
stimulated by bacterial virulence factor RgpB
Dental pulp inflammation is a complex process, which (Tancharoen et al. 2005). In painful pulpitis, tissue
is thought to be neurogenically mediated, involving a level of SP was found to be increased (Awawdeh et al.
variety of nervous and vascular reactions that could 2002, Bowles et al. 2003). SP receptor expression in
lead to pulp necrosis (Byers et al. 1990, Kim 1990). A pulp tissue is also significantly increased during
number of different neuropeptides, including substance inflammatory phenomena such as acute irreversible
P (SP), are known to be present in the nerve fibres of pulpitis (Caviedes-Bucheli et al. 2007). Moreover, SP is
dental pulp (Casasco et al. 1990). Pulp cells were also released following dental procedures (e.g. cavity
preparation), and its expression may contribute to
inflammation and pain. (Caviedes-Bucheli et al. 2005).
Correspondence: Xue-Dong Zhou, DDS, PhD, West China Recently, SP has been documented to induce the pro-
College of Stomatology, Sichuan University, 3rd section 14#,
Renmin South Road, Chengdu 610041, China (Tel.:
duction of inflammatory cytokines such as interleukin
86 28 85501439; fax: 86 28 85582167; e-mail: zhouxd.scu (IL)-1b, IL-6, IL-8 and tumour necrosis factor-alpha in
@gmail.com). human pulp cell cultures (Patel et al. 2003, Park et al.
Wang et al. SP affects gelatinolytic activity via ROS
2004, Yamaguchi et al. 2004). SP can also enhance (Hyclone, South Logan, UT, USA), 2 mmol L)1 l-gluta-
expression of lipopolysaccharide-induced inflammatory mine, 100 units mL)1 penicillin and 100 lg mL)1
factors in pulp cells (Tokuda et al. 2004). These streptomycin. Confluent cells were detached with
cytokines are thought to be important in the patho- 0.20% trypsin and 0.02% ethylenediaminetetraacetate
genesis of pulpitis. Therefore, SP may initiate, exacer- (EDTA). Immunocytochemical staining using anti-
bate or maintain the inflammatory processes in vimentin antibody was performed for tissue origin
pulpitis. characterization. Fibroblast-like pulp cells between the
A notable feature of suppurative pulpitis is the third and sixth passages were used in the following. Cells
degradation of the extracellular matrix (ECM) (Gusman were pre-incubated with serum-free DMEM for 18 h
et al. 2002, Shin et al. 2002, Wahlgren et al. 2002, Tsai before the addition of other factors.
et al. 2005). Matrix metalloproteinases (MMPs) are a
family of zinc-dependent endopeptidases responsible for
Detection of intracellular ROS by confocal
degrading structural proteins of ECM and for cleaving
microscopy
other non-ECM molecules ranging from growth factor
precursors, cytokines and binding proteins, to cell The formation of intracellular ROS was determined by
surface receptors (Sternlicht & Werb 2001). In pulp the 2¢,7¢-dichlorodihydrofluorescein diacetate (DCFDA)
cell cultures, stimulation of both inflammatory cyto- fluorescence method (Wang et al. 2007). For experi-
kines and bacterial extracts may induce production of ments to determine the generation of ROS induced by
MMPs and cell mediated collagen degradation (Panaga- SP, cells (6 · 104 per dish) were plated and allowed to
kos et al. 1996, O’Boskey & Panagakos 1998, Lin et al. adhere on 35-mm glass-bottomed culture dishes (WPI
2001, Chang et al. 2002, Wisithphrom & Windsor Inc., Berlin, Germany) for 24 h. After the starvation,
2006, Wisithphrom et al. 2006). However, little infor- the serum-deprived cells were stabilized in serum-free
mation is currently known regarding the effects of SP on DMEM without phenol red for at least 30 min. Then
MMPs activity of pulp cells. cells were loaded with a cell-permeant probe H2DCFDA
Reactive oxygen species (ROS) have emerged (Sigma, St. Louis, MO, USA) at 10 lmol L)1 for
as essential signalling molecules in the regulation 30 min. Afterwards, the medium was aspirated, and
of inflammation processes (Springer et al. 2005, the cells were washed twice in media followed by the
McCubrey et al. 2006, Nakano et al. 2006). Meantime, treatment with or without (control) 1 lmol L)1 SP for
intracellular ROS can translocate into the ECM and 30 min. Where indicated, the cells were treated with
oxidize latent MMP-2 complexes (Wang et al. 2005). To 5 mmol L)1 N-Acetyl-l-cysteine (NAC) (Alexis, San
date, SP has been shown to induce ROS generation in Diego, CA, USA) for 1 h before the treatment of SP. To
rat microglia (Block et al. 2006) and A549 airway assess ROS-mediated oxidation of H2DCFDA to the
epithelial cells (Springer et al. 2005). The purpose of fluorescent compound DCF, cells were immediately
this study was to investigate the effects of SP on observed with a laser scanning confocal microscope
gelatinolytic activity of MMPs in human pulp cells and (Leica TCS SP2, Leica, Wetzlar, Germany). Images
the possibility of ROS involvement. (20 · objective) were recorded at an excitation wave-
length of 488 nm and an emission wavelength
530 nm.
Isolation and culture of human pulp cells Gelatin zymography

Normal human impacted third molars were obtained Cells were seeded at a density of 5 · 104 per well in
from healthy donors, with informed consent, at the West 24-well plates (Corning, Corning, NY, USA) and
China Stomatology Hospital of Sichuan University. The grown to confluence. After the starvation, cells was
study was reviewed and approved by the ethical board of washed once in fresh serum-free media and treated
the hospital. Coronal pulp tissue was gently separated different doses of SP (1 pmol L)1, 100 pmol L)1,
and minced with scalpels. Dental pulp cells were 10 nmol L)1, 1 lmol L)1 and 100 lmol L)1) for
enzymatically released as previously described (Wang 48 h. Where indicated, the cells were treated with
et al. 2006), then cultured in Dulbecco’s modified 5 mmol L)1 NAC for 1 h before the treatment of SP.
Eagle’s medium (DMEM) (Gibco, Grand Island, NY, Then conditioned media were collected and stored at
USA) supplemented with 10% foetal calf serum )20 C until assayed.
SP affects gelatinolytic activity via ROS Wang et al.
The gelatinolytic activity of the conditioned media

Results
was assayed by gelatin zymography (Gusman et al.
2002) using 8% sodium dodecyl sulphate (SDS) poly-
SP increased levels of ROS
acrylamide gel electrophoresis minigels containing
1.5 g L)1 gelatin (Merck, Darmstadt, Germany). The The basic levels of ROS were very low (Fig. 1a). As
samples (each 20 lL) were mixed 4 : 1 with 5 · shown in Fig. 1b, it was obvious that the intracellular
nonreducing sample buffer containing 0.25 mol L)1 ROS levels increased after the stimulation of SP.
Tris–HCl (pH 6.8), 50% glycerol, 10% SDS and 0.005% Conversely, significant inhibition of ROS generation
bromophenol blue, before loaded onto the gels. After was observed when the cells were pre-treated by the
electrophoresis, gels were washed twice for 20 min antioxidant NAC before the exposure of SP (Fig. 1c).
in 2.5% Triton X-100 on a rotary shaker to remove
SDS and to allow proteins to renature. The gels were
SP influenced gelatinolytic activity of pulp cells
then incubated in a buffer containing 50 mmol L)1
Tris–HCl (pH 7.5), 100 mmol L)1 NaCl, 5 mmol L)1 The gelatin zymography indicated that there were
CaCl2, ZnCl2 and 0.002% NaN3 for 24 h at 37 C. several kinds of gelatinolytic proteinase in the condi-
Zymographic gels were stained with 0.1% Coomassie tioned medium of pulp cells (Fig. 2). The major
Brilliant Blue R-250 and de-stained. Gels were proteinase evident in the gels migrated at approxi-
photographed with the Gel Doc 2000 system mately 66 kDa, which corresponded to the active form
(Bio-Rad, Hercules, CA, USA) and analysed with of MMP-2. Interestingly, a less evident gelatinolytic
Image-Pro Plus 6.0 (Media Cybernetics, Silver Spring, band was visualized at about 116 kDa. A band with
MD, USA). molecular masses between 66 and 97 kDa, presenting
MMP-9, was very weak.
Following 1 or 100 lmol L)1 SP stimulation, MMP-2
activity was significantly enhanced (anova, P < 0.05)
All assays were repeated at least three times to (Fig. 3). In this experiment, a gelatinolytic band at
ensure reproducibility. Values in zymography assay approximately 44 kDa was detected for the first time. Its
were expressed as mean values and standard devia- intensity appeared to be SP dose dependent (Fig. 2). As
tions of integrated optical density (per cent of shown in Figs 2 and 3, following the pre-treatment of
control). All values obtained were analysed using cells with antioxidant NAC, MMP-2 activity was signif-
analysis of variance (anova) with correction for icantly inhibited (Student’s t-test, P < 0.05) and the
multiple comparisons (Bonferroni) or an unpaired latent form of MMP-2 was evident compared with other
Student’s t-test (n = 6), with a significance level of lane (Fig. 2). In addition to MMP-2, other gelatinolytic
P < 0.05. bands also turned weaker after NAC treatment (Fig. 2).
Control SP SP + NAC
(a) (b) (c)
Figure 1 Substance P (SP) increased intracellular levels of reactive oxygen species (ROS) in human pulp cells. Human pulp cells
cultured on glass-bottomed dishes were loaded with 10 lmol L)1 2¢,7¢-dichlorodihydrofluorescein diacetate (DCFDA) for 30 min
and then stimulated without (a) or with (b) 1 lmol L)1 SP for 30 min. Pre-treatment with 5 mmol L)1 N-Acetyl-l-cysteine (NAC)
for 1 h before the exposure of SP inhibited ROS generation (c). Cells were imaged by confocal microscopy using 20 · objective.
Bar, 150 lmol L)1. Results were representative of three independent experiments.
C
SP
SP
L –1 +
L –1 P
NA
S
ol SP
SP
SP
L –1
L –1
–1
1
L –1
m L–
ol
ol
ol
ol
µm
pm
5 ol
ol
ol
tr
m
µm
pm
pm
m
on
0
10
10
10
C
1
200 kDa
116 kDa
97 kDa MMP-9
Figure 2 Gelatin zymography of condi-
tioned media from pulp cells incubated
Latent MMP-2
for 48 h with no treatment (control) and
66 kDa Active MMP-2
different doses of substance P (SP)
(1 pmol L)1, 100 pmol L)1,
10 nmol L)1, 1 lmol L)1 and
100 lmol L)1); 5 nmol L)1 N-Acetyl-l-
cysteine (NAC) was also used as antiox-
idant to assess the role of reactive oxygen 44 kDa
species (ROS) in gelatinolytic activity.
Results were representative of three
independent experiments.
Control is known that dental pulp tissue is innervated with a

Integrated optical density (% of control)
1 pmol L–1 SP
100 pmol L–1 SP subpopulation of sensory neurons containing neuro-
*
150 10 nmol L–1 SP peptides, such as SP. These neuropeptides are associ-
1 µmol L–1 SP
100 µmol L–1 SP * ated with the development of neurogenic inflammation.
1 µmol L–1 SP+ 5 mmol L–1 NAC
**
The present study, therefore, was designed to investi-
100 gate the mechanisms of ECM degradation in the aspect
of SP.
In this study, intracellular ROS levels were measured
using the fluorescence dye H2DCFDA, which is a
50 nonpolar compound that is converted into a nonfluo-
rescent polar derivative (H2DCF) by cellular esterases
after incorporation into cells. H2DCF is membrane
0 impermeable and rapidly oxidized to the highly fluo-
rescent DCF in the presence of intracellular ROS
Figure 3 Matrix metalloproteinase (MMP)-2 activity of condi-
(Gomes et al. 2005). It was shown that after SP
tioned media from pulp cells in response to 48 h treatment of
stimulation, ROS generation was induced in human
substance P (SP) or in the presence of 5 mmol L)1 N-Acetyl-l-
pulp cells (Fig. 1b). ROS scavenger NAC significantly
cysteine (NAC). Values were expressed as means + SD of
integrated optical density (IOD) (per cent of control). Statisti-
reduced the intracellular ROS level of pulp cells in
cally significant: *P < 0.05, compared with control; response to SP (Fig. 1c). These findings suggest that
**P < 0.05, compared with treatment of 1 lmol L)1 SP. ROS may mediate SP signalling, which is consistent
with previous studies using other cell lines (Springer
et al. 2005, Block et al. 2006).
Gelatin zymography has been extensively used to
Discussion
examine gelatinolytic activity of pulp cells. The results
Toothache is the most common orofacial pain and often suggest that in the conditioned medium from pulp cells,
involves pulpitis. However, the pathological mecha- MMP-2 is the major gelatinase and present mostly in
nisms of painful pulpitis still remain to be elucidated. It active form migrating at 66 kDa whilst, only minimal
amounts of MMP-9 could be detectable. These data are MMP-2 to its active form (Wang et al. 2005). As is
in agreement with several previous studies examining suggested, ROS work through disturbance of the
conditioned medium using gelatin zymography cysteine–zinc interaction of the cysteine switch in
(Panagakos et al. 1996, O’Boskey & Panagakos 1998, inactive zymogens (Hannas et al. 2007). Taken
Ueda & Matsushima 2001). When cells were exposed to together, the present results imply that ROS play an
SP at levels above 1 lmol L)1, MMP-2 activity was important role in ECM degradation by proteinase
significantly enhanced. However, SP at levels below secreted from pulp cells. Therefore, it can be proposed
1 lmol L)1 did not influence MMP-2 activity signifi- that the reduction of ROS levels by using antioxidants
cantly, although these levels of SP can activate p38 and might be of potential value in preventing ECM degra-
ERK signalling in cultured pulp cells (Tokuda et al. dation during pulpitis.
2005). This phenomenon might be due to high basic
levels of active MMP-2 in pulp cells cultured ex vivo.
Conclusions
However, the levels of immunoreactive SP in inflamed
pulp are around 0.1 nmol L)1 (Bowles et al. 2003). It In summary, neuropeptide SP can influence gelatin-
means that native pulp cells might be more sensitive to olytic activity in pulp cells, which is mediated by a
SP and the supra-physiological concentration range of ROS-dependent pathway. Nevertheless, the present
SP (>1 lmol L)1), which has been selected in many study only examined effects of SP on gelatinolytic
experiments, is not suitable to clinical pulp inflamma- activity in conditioned medium from pulp cells. Except
tion. In the pathogenesis of clinical pulpitis, MMPs can for gelatinases, MMPs also include collagenases,
be secreted by infiltrating cells such as polymorphonu- stromelysins, membrane-type MMPs, minimal-domain
clear leukocytes and macrophages (Gusman et al. MMPs and others. In vivo MMPs activities are always
2002, Shin et al. 2002, Wahlgren et al. 2002). Com- regulated by their interaction with the tissue inhibi-
plicated interaction between those infiltrating cells and tors of matrix metalloproteinases (TIMPs). Thus,
pulp cells might affect the sensitivity of pulp cells to SP future investigation should be undertaken to clarify
regarding to their MMPs production. As ex vivo culture whether SP also influence other MMPs activities as
models can mimick the natural environment, the well as TIMPs.
present results at least indicated that the production
of MMP-2 in pulp cells might be affected by tissue SP
Acknowledgements
levels. Recently, it has been indicated that inflamma-
tory cytokines stimulate the production of elevated The authors greatly acknowledge Chao-Liang Zhang
levels of MMP-2 in pulp cells (O’Boskey & Panagakos for technical assistance on confocal microscopy. The
1998, Chang et al. 2001, Wisithphrom & Windsor authors are also grateful to James L. Gutmann and
2006). As SP has been shown to up-regulate inflam- Anthony J. Smith for critically reading the manuscript.
matory cytokines (Patel et al. 2003, Park et al. 2004, This study was supported by the Research Found
Yamaguchi et al. 2004), one may speculate that for Young Teachers from Sichuan University, China
cytokines might directly or indirectly mediate the (No. 06059).
enhancement of MMP-2 activity. Additionally, it is
interesting to note that a band migrating at 44 kDa
References
appeared to be intensified in a SP-dose-dependent
manner (Fig. 2). Detailed information of the corre- Awawdeh L, Lundy FT, Shaw C, Lamey PJ, Linden GJ,
sponding proteinase needs further investigation. Kennedy JG (2002) Quantitative analysis of substance P,
According to the gelatin zymography results, inhibi- neurokinin A and calcitonin gene-related peptide in pulp
tion of ROS production significantly reduced MMP-2 tissue from painful and healthy human teeth. International
activity of pulp cells exposed to SP. The band migrating
Block ML, Li G, Qin L et al. (2006) Potent regulation of
at 44 kDa as well as those bands between 116 and
microglia-derived oxidative stress and dopaminergic neuron
200 kDa were all attenuated by NAC treatment. It was
survival: substance P vs. dynorphin. FASEB Journal 20,
notable that NAC pre-treatment caused appearance of 251–8.
latent MMP-2 in the conditioned medium, which Bowles WR, Withrow JC, Lepinski AM, Hargreaves KM (2003)
suggests that ROS might be essential for the activation Tissue levels of immunoreactive substance P are increased
of latent MMP-2. Interestingly, previous study also in patients with irreversible pulpitis. Journal of Endodontics
reported that ROS was required for conversion of latent 29, 265–7.
Byers MR, Taylor PE, Khayat BG, Kimberly CL (1990) Effects Park SH, Hsiao GY, Huang GT (2004) Role of substance P and
of injury and inflammation on pulpal and periapical nerves. calcitonin gene-related peptide in the regulation of interleu-
Journal of Endodontics 16, 78–84. kin-8 and monocyte chemotactic protein-1 expression in
Casasco A, Calligaro A, Casasco M et al. (1990) Peptidergic human dental pulp. International Endodontic Journal 37,
nerves in human dental pulp. An immunocytochemical 185–92.
study. Histochemistry 95, 115–21. Patel T, Park SH, Lin LM, Chiappelli F, Huang GT (2003)
Caviedes-Bucheli J, Correa-Ortı́z JA, Garcı́a LV, López-Torres R, Substance P induces interleukin-8 secretion from human
Lombana N, Muñoz HR (2005) The effect of cavity dental pulp cells. Oral Surgery, Oral Medicine, Oral Pathology,
preparation on substance P expression in human dental Oral Radiology, and Endodontics 96, 478–85.
pulp. Journal of Endodontics 31, 857–9. Shin SJ, Lee JI, Baek SH, Lim SS (2002) Tissue levels of matrix
Caviedes-Bucheli J, Gutierrez-Guerra JE, Salazar F, Pichardo D, metalloproteinases in pulps and periapical lesions. Journal of
Moreno GC, Muñoz HR (2007) Substance P receptor Endodontics 28, 313–5.
expression in healthy and inflamed human pulp tissue. Springer J, Pleimes D, Scholz FR, Fischer A (2005) Substance P
International Endodontic Journal 40, 106–11. mediates AP-1 induction in A549 cells via reactive oxygen
Chang YC, Yang SF, Hsieh YS (2001) Regulation of matrix species. Regulatory Peptides 124, 99–103.
metalloproteinase-2 production by cytokines and pharma- Sternlicht MD, Werb Z (2001) How matrix metalloproteinases
cological agents in human pulp cell cultures. Journal of regulate cell behavior. Annual Review of Cell and Develop-
Endodontics 27, 679–82. mental Biology 17, 463–516.
Chang YC, Lai CC, Yang SF, Chan Y, Hsieh YS (2002) Tancharoen S, Sarker KP, Imamura T et al. (2005) Neuro-
Stimulation of matrix metalloproteinases by black-pig- peptide release from dental pulp cells by RgpB via protein-
mented Bacteroides in human pulp and periodontal liga- ase-activated receptor-2 signaling. Journal of Immunology
ment cell cultures. Journal of Endodontics 28, 90–3. 174, 5796–804.
Gomes A, Fernandes E, Lima JLFC (2005) Fluorescence probes Tokuda M, Miyamoto R, Nagaoka S, Torii M (2004) Substance
used for detection of reactive oxygen species. Journal of P enhances expression of lipopolysaccharide-induced
Biochemical and Biophysical Methods 65, 45–80. inflammatory factors in dental pulp cells. Journal of End-
Gusman H, Santana RB, Zehnder M (2002) Matrix metallo- odontics 30, 770–3.
proteinase levels and gelatinolytic activity in clinically Tokuda M, Miyamoto R, Sakuta T, Nagaoka S, Torii M (2005)
healthy and inflamed human dental pulps. European Journal Substance P activates p38 mitogen-activated protein kinase
of Oral Sciences 110, 353–7. to promote IL-6 induction in human dental pulp fibroblasts.
Hannas AR, Pereira JC, Granjeiro JM, Tjäderhane L (2007) Connective Tissue Research 46, 153–8.
The role of matrix metalloproteinases in the oral environ- Tsai CH, Chen YJ, Huang FM, Su YF, Chang YC (2005) The
ment. Acta Odontologica Scandinavica 65, 1–13. upregulation of matrix metalloproteinase-9 in inflamed
Kim S (1990) Neurovascular interactions in the dental pulp in human dental pulps. Journal of Endodontics 31, 860–2.
health and inflammation. Journal of Endodontics 16, 48–53. Ueda L, Matsushima K (2001) Stimulation of plasminogen
Lin SK, Wang CC, Huang S et al. (2001) Induction of dental activator activity and matrix metalloproteinases of human
pulp fibroblast matrix metalloproteinase-1 and tissue inhib- dental pulp-derived cells by tumor necrosis factor-alpha.
itor of metalloproteinase-1 gene expression by interleukin- Journal of Endodontics 27, 175–9.
1alpha and tumor necrosis factor-alpha through a prosta- Wahlgren J, Salo T, Teronen O, Luoto H, Sorsa T,
glandin-dependent pathway. Journal of Endodontics 27, 185– Tjäderhane L (2002) Matrix metalloproteinase-8 (MMP-
9. 8) in pulpal and periapical inflammation and periapical
McCubrey JA, Lahair MM, Franklin RA (2006) Reactive root-canal exudates. International Endodontic Journal 35,
oxygen species-induced activation of the MAP kinase 897–904.
signaling pathways. Antioxidants & Redox Signaling 8, Wang L, Clutter S, Benincosa J, Fortney J, Gibson LF (2005)
1775–89. Activation of transforming growth factor-beta1/p38/Smad3
Nakano H, Nakajima A, Sakon-Komazawa S, Piao JH, Xue X, signaling in stromal cells requires reactive oxygen species-
Okumura K (2006) Reactive oxygen species mediate cros- mediated MMP-2 activity during bone marrow damage.
stalk between NF-kappaB and JNK. Cell Death and Differen- Stem Cells 23, 1122–34.
tiation 13, 730–7. Wang FM, Qiu K, Hu T, Wan CX, Zhou XD, Gutmann JL
O’Boskey FJ Jr, Panagakos FS (1998) Cytokines stimulate (2006) Biodegradable porous calcium polyphosphate scaf-
matrix metalloproteinase production by human pulp cells folds for the three-dimensional culture of dental pulp cells.
during long-term culture. Journal of Endodontics 24, 7–10. International Endodontic Journal 39, 477–83.
Panagakos FS, O’Boskey JF Jr, Rodriguez E (1996) Regulation Wang X, Wang Y, Kim HP, Choi AM, Ryter SW (2007) FLIP
of pulp cell matrix metalloproteinase production by cyto- inhibits endothelial cell apoptosis during hyperoxia by
kines and lipopolysaccharides. Journal of Endodontics 22, suppressing Bax. Free Radical Biology & Medicine 42,
358–61. 1599–609.
Wisithphrom K, Windsor LJ (2006) The effects of tumor collagen degradation by pulp fibroblasts. Journal of Endodon-
necrosis factor-alpha, interleukin-1beta, interleukin-6, and tics 32, 186–92.
transforming growth factor-beta1 on pulp fibroblast medi- Yamaguchi M, Kojima T, Kanekawa M, Aihara N, Nogimura
ated collagen degradation. Journal of Endodontics 32, 853– A, Kasai K (2004) Neuropeptides stimulate production of
61. interleukin-1 beta, interleukin-6, and tumor necrosis factor-
Wisithphrom K, Murray PE, Windsor LJ (2006) Interleukin-1 alpha in human dental pulp cells. Inflammation Research 53,
alpha alters the expression of matrix metalloproteinases and 199–204.
doi:10.1111/j.1365-2591.2008.01438.x
Pre-clinical endodontics: a survey amongst German

dental schools
D. Sonntag1, R. Bärwald1, M. Hülsmann2 & V. Stachniss1

1
Department of Operative Dentistry, Philipps-University, Marburg; and 2Department of Operative Dentistry, Preventive
Dentistry and Periodontology, Georg-August-University, Göttingen, Germany
Abstract Results With feedback from 27 of 28 dental schools,

the response rate was 96%. Pre-clinical endodontic
Sonntag D, Bärwald R, Hülsmann M, Stachniss V. Pre-
education at German universities varied considerably.
clinical endodontics: a survey amongst German dental schools.
Theory classes ranged from 5 to 30 h (13.3 h mean),
International Endodontic Journal, 41, 863–868, 2008.
practical classes from 12.5 to 60 h (45.4 h mean). The
Aim To evaluate the state and level of pre-clinical student to staff ratio varied between 9 : 1 and 30 : 1
endodontic education in German dental schools and to (16 : 1 mean). Forty-eight per cent of the universities
evaluate differences with regard to intensity and extent had a specialist in endodontics or a teacher with a
of teaching, time devoted to teaching pre-clinical special interest. A dental microscope was available for
endodontics, personnel resources in teaching and pre-clinical teaching purposes in 38% of the universi-
technical equipment. ties. The majority (63%) of universities taught root
Methodology Twenty-eight questionnaires were canal preparation with rotary nickel titanium instru-
e-mailed to those in charge of pre-clinical endodontic ments.
education in German dental schools. The extent of Conclusion Pre-clinical endodontic education var-
education, the student–teacher ratio, the teaching ied considerably between German universities because
content as well as the application of teaching materials of differences in programme design, staff and course
and technologies were asked. If, after 4 weeks, no content.
response had been received, the questionnaire was sent
Keywords: endodontics, pre-clinical, survey, teach-
out by e-mail again. In the absence of a reply, a phone
ing, undergraduate education.
call was made to the corresponding university to
conduct the survey by phone. Received 3 September 2007; accepted 10 April 2008
25% of the teeth with root fillings had signs of

Introduction
persistent apical pathosis. Furthermore, a retrospective
In Germany, as well as in other European countries, the cohort study carried out by Weiger et al. (1997)
standard of root canal treatment is not high (Kirkevang revealed that 78.6% of root fillings in maxillary molars
et al. 2001, Boucher et al. 2002, Eriksen et al. 2002, and 67.6% of root fillings in mandibular molars ended
Boltacz-Rzepkowska & Pawlicka 2003, Loftus et al. more than 2 mm short of the radiographic apex.
2005). Schulte et al. (1998) reported that 56.8% of the Although many factors could account for these tech-
patients examined at the Clinical Department of nical deficiencies, it is possible that the quality and
Operative Dentistry, Phillips-University in Marburg, quantity of pre-clinical and clinical endodontic educa-
Germany, had root fillings that were short and over tion has an impact.
The quality of endodontic treatments performed
during clinical and pre-clinical education has been
Correspondence: Dr David Sonntag, Department of Operative evaluated in only a few studies. Radiographic homo-
Dentistry, Philipps-University Marburg, Georg-Voigt-Str. 3,
35033 Marburg, Germany (Tel.: 0049 6421 286 3194; fax:
geneity and length of root fillings performed with
0049 6421 286 3245; e-mail: sonntag@staff.uni-marburg. lateral compaction by dental students were shown to be
de). sufficient in no more than 62.7% of cases (Eleftheriadis
Pre-clinical endodontics Sonntag et al.
& Lambrianidis 2005). According to Barrieshi-Nusair end of their undergraduate education were reported
et al. (2004), 61.3% of the students managed to reach (Plasschaert et al. 2006).
the desired working length. Hayes et al. (2001) con- Rotary nickel titanium instruments are an example
sidered only 13% teeth root filled by students during that the general use of new materials and techniques is
their clinical experience were satisfactory, whereas a slow process (Parashos & Messer 2006) even though
82% of the teeth had root fillings with voids and were they may have an influence on the outcome of
either long or short. treatment. Students achieve significantly better results
The German Medical Licensure Act (‘Deutsche in root canal preparation with rotary instruments than
Approbationsordnung’) of the dental undergraduate with stainless steel hand instruments (Namazikhah
programme schedules a preliminary dental examina- et al. 2000, Garip & Gunday 2001, Gluskin et al. 2001,
tion after the fifth term of the dental undergraduate Peru et al. 2006). Despite this fact, rotary instruments
programme. Apart from general medical subjects such are still not being used for training purposes by all
as anatomy, physiology and physiological chemistry, universities (Arbab-Chirani & Vulcain 2004).
the only required dental subject is prosthodontics The purpose of this survey was to assess the current
(Bundesgesetzblatt (BGBI) 1999). Restorative dentistry status of pre-clinical endodontic education and to
including inlays, partial crowns, restorative treatment determine whether quality and quantity of pre-clinical
with composite and amalgam as well as the field of endodontic education in German Dental Schools are
endodontics are only taught from the sixth term (3rd comparable with regard to personnel and technical
year) onward, just before students start undertaking equipment, supervision and complexity of the training.
root canal treatment on patients during the seventh
term. During the sixth term, students have to complete
an average of 180 h of practical work in restorative
dentistry to allow them to treat patients in various Twenty-eight questionnaires were e-mailed to directors
courses that follow. of 3rd year pre-clinical courses (manikin, phantom
Qualtrough et al. (1999) in their international head) in the 28 German departments or clinics for
survey observed that the average period of pre-clinical restorative dentistry running their own independent
endodontic training in Western Europe of 38 h was pre-clinical course. The questions were answered
low. This explains why the undergraduate curriculum either by ticking boxes (yes/no decision) or by short
guidelines, established that most dental schools in written responses giving detailed data (e.g. information
Europe included insufficient endodontic training (Euro- about student–teacher ratio, number of teeth treated,
pean Society of Endodontology 2001). Despite the preparation techniques, etc). Sufficient space was
existence of EU directives and guidelines, which were provided on the sheets to include additional com-
devised to assure comparable and acceptable standards ments. If the course directors and/or their e-mail
of dental education and competence gained from addresses were not known, the information was
training programmes, there are major differences requested from the clinic’s secretarial department.
amongst countries. The differences are greater in those Four weeks after the first mailing, the questionnaires
countries about to join an expanding EU (Shanley were sent out again by e-mail if no response had been
2004). The first version of the undergraduate curric- received. In the repeated absence of feedback, a phone
ulum guidelines published by the European Society of call was made, if possible, to the corresponding course
Endodontology (ESE) in 1992 had a detailed require- director to conduct the survey by phone. In doing so,
ment profile for dental education, whereas the current responses were received from 27 of 28 clinics, giving a
guidelines leave much more scope for interpretation. response rate of 96%.
However, this does not necessarily lead to more
consistent standards of endodontic education (Euro-
Results
pean Society of Endodontology 1992, 2001). The
General Assembly of the Association for Dental Educa-
Student–teacher ratio
tion in Europe (ADEE) published a draft survey entitled
‘Profile and competences for the European dentist’ in A mean total of 41 students attended pre-clinical
2004 (Plasschaert et al. 2006). Additional to the range endodontic training courses at the same time. The
of core dental competencies, the specific competencies student–teacher ratio varied between 9 : 1 and 30 : 1,
in endodontics students should have acquired at the the average ratio being 16 : 1 (Tables 1 and 2).
Sonntag et al. Pre-clinical endodontics
Table 1 Number of students attending pre-clinical classes Table 4 Practical training lessons (at 45 min) for endodontic
as well as number of universities and corresponding group sizes education
Students Universities (%) Practical training (h) Universities (%)
11–20 2 (7.4) 10–20 5 (18.5)

21–30 3 (11.1) 21–30 4 (14.8)
31–40 9 (33.3) 31–40 8 (29.6)
41–50 6 (22.2) 41–50 5 (18.5)
51–60 3 (11.1) 51–60 5 (18.5)
60–80 4 (14.8)
Table 5 Minimum number of artificial canals, incisors/

Table 2 Student–faculty ratio for pre-clinical endodontic canines and molars/pre-molars to be prepared
training as well as number of universities by ratio
Type of canals/teeth Number Universities (%)
Students–staff ratio Universities (%)
Artificial canals n/a 3 (11.1)
9–12 : 1 7 (25.9) Artificial canals 1–2 18 (66.6)
13–16 : 1 11 (40.7) Artificial canals 3–4 6 (22.2)
17–20 : 1 5 (18.5) Incisors/canines n/a 2 (7.4)
21–24 : 1 1 (3.7) Incisors/canines 1–2 23 (85.2)
25–28 : 1 2 (7.4) Incisors/canines 3–4 2 (7.4)
29–31 : 1 1 (3.7) Molars/pre-molars n/a 0
Molars/pre-molars 1–2 14 (51.8)
Extent of education
n/a, not applicable.
An average of 13.3 h was spent on didactic teaching in
endodontology and an average of 45.5 h on training.
Table 6 Minimum number of canals to be prepared and filled
Overall, the practical endodontic exercises were con-
ducted on 1.7 artificial root canals, 1.5 incisors and/or Total number of canals Universities (%)
canines and 2.5 pre-molars and/or molars. Each 1–5 4 (14.8)
student, therefore, treated 10 canals (mean value). 6–10 9 (33.3)
Forty-eight per cent of the universities (n = 13) had a 11–15 12 (44.4)
16–20 1 (3.7)
specialist endodontist leading the pre-clinical endodon-
21–25 1 (3.7)
tic training.
Observing a clinical root canal treatment performed
by a staff member was mandatory in 48% (n = 13) of
[e.g. rubber dam placement, access cavity, radiographic
the universities, voluntary in 37% (n = 10) of the
length determination, mastercone fit, etc.). To assess
universities and not required in 15% (n = 4) of the
theoretical knowledge, written examinations were
universities (Tables 3–6). Three universities offered
used. A practical test such as preparing an access
problem-based learning units.
cavity in a maxillary first molar was carried out in 30%
(n = 8) of the universities, an additional individual
Assessment and teaching material theory examination was carried out in 27% (n = 7) of
the schools. A total of 14 different textbooks were
To assess students’ practical work, dentists supervised
recommended, three of which were in English. In
root canal treatment continuously in all universities
addition to text books, 41% of the universities offered
lecture hand-outs as well as intranet and internet
Table 3 Theory lessons (at 45 min) for endodontic education
access to the relevant subjects.
Theory lessons (h) Universities (%)
6–10 10 (37.0)
Content of teaching
11–15 8 (29.6)
16–20 7 (25.9) The stepback technique was the most commonly
21–25 1 (3.7)
taught (70%) manual preparation method. Preparation
26–30 1 (3.7)
of root canals with rotary nickel titanium instruments
was taught by 63% (n = 17) of the universities. All Table 10 Use of microscopes (multiple answers possible).
universities teaching rotary preparation used instru- The percentages refer to the universities possessing a
ments with torque controlled motors. Lateral compac- microscope (n = 10)
tion was the filling method taught by the majority of Intended use Universities (%)
schools (Tables 7–9). Demonstrations 6 (60)
Evaluation of access cavity + root canal 5 (50)
orifices
Equipment and costs Hands on training 3 (30)
A minority of universities (37%, n = 10) had a dental

microscope for pre-clinical endodontic training pur- root canals are prepared and filled. The department did
poses. The costs borne by the students for consumable not have a microscope and did not teach rotary
material such as root canal instruments, X-ray films, instrumentation. Furthermore, the department had
silver points, gutta-percha, sealer, etc. varied consider- no endodontic specialist.
ably and were on average approximately 145€. Six The clinic with the overall most favourable education
universities gave no feedback regarding material costs. situation had a student to teacher ratio of 10 : 1.
The answers given in several questionnaires and in direct Theory was taught for 13 h and there were 45 h of
interviews made clear that the programme directors are practical experience. A total of 15 root canals are
unsure about the exact material costs (Tables 8–10). prepared and filled. The department had several dental
The overall most unfavourable staff–student ratio microscopes and taught preparation with rotary instru-
was 20 : 1. Time for didactic teaching was 6 h and ments. Furthermore, two endodontic specialists were
there were 20 h of practical experience. A total of seven working in the department.
Table 7 Teaching materials at the students’ disposal (the

percentages refer to n = 27 universities) Discussion
Teaching materials Yes (%) No (%) Partially (%) Multiple national and international surveys on end-
Endodontic script 10 (37%) 17 (62.9) 0 odontic education have been carried out (Shovelton
Lecture print-outs 11 (40.7) 7 (25.9) 9 (33.3) 1979, Qualtrough & Dummer 1997, Qualtrough et al.
CD hand-outs with 2 (7.4) 22 (81.5) 3 (11.1)
1999). Sending out questionnaires has proven to be an
information
Videos/files available 11 (40.7) 16 (59.3) 0
effective method for capturing data on educational
via intranet/internet issue. Whilst Dummer (1991) reported a response rate
of 100%, the present survey did not receive a response
from one school; the person in charge of courses was
Table 8 Preparation methods taught in class (multiple unavailable. Other national surveys achieved response
answers possible) rates of 87% (Cruz et al. 2000) and of 91% (Cailleteau
Manual preparation Universities (%) & Mullaney 1997). Based on questionnaires sent out by
Crown-down/step-down 7 (25.9) e-mail and on selective interviews by phone, this survey
Step-back 19 (70.4) managed to achieve a response rate of 96%. The survey
Balanced force 4 (14.8) results available, thus give a reliable reflection of pre-
Coronal crown-down, then step-back 4 (14.8) clinical endodontic education in Germany.
Standardized technique 3 (11.1)
The education level varied significantly from clinic to
clinic. Theory lessons vary from 5 to 30 academic
Table 9 Rotary preparation techniques with nickel titanium hours (equivalent to 45 min), practical training from
instruments (multiple answers possible) 12.5 to 60 h. This reflects amongst other things the
different focus on pre-clinical education. Many clinics
Preparation system User (%)
consider pre-clinical education to be mainly for the
FlexMaster, VDW 15 (88.2)
teaching of manual skills. The underlying theory comes
ProFile, Dentsply 4 (23.5)
ProTaper, Dentsply 3 (17.6)
second and is often only taught in subsequent terms.
Mtwo, VDW 1 (5.8) The total number of hours for pre-clinical theory
GT, Dentsply 1 (5.8) varied in Germany around the factor of 5, and for
The percentages apply to the number of universities using NiTi practical training around the factor of 6. In a compar-
instruments (n = 17). ison in Western Europe, results varied more widely:
Sonntag et al. Pre-clinical endodontics
around a factor of 25 (Qualtrough et al. 1999). On it preserves the original canal geometry with only
average, 45 h were spent on practical training in minimal canal straightening (Peters 2004, Hülsmann
Germany compared with 78 h in France, which is 1.5 et al. 2005). Even practitioners with little experience or
times more. The ESE guidelines, however, reveal that ‘it practice can preserve the canal geometry when using
is neither expected nor desirable that all courses in rotary instruments (Namazikhah et al. 2000, Garip &
endodontics are given before students start clinical Gunday 2001, Gluskin et al. 2001, Peru et al. 2006).
work on patients’. ‘It is therefore not intended to give Only three universities had a dental microscope available
definite directions. Innovative ways in which endodon- for pre-clinical exercises. None of the universities pos-
tic curricula can be progressed are to be encouraged’ sesses a sufficient amount of equipment to enable every
(European Society of Endodontology 2001). student to practice using dental microscopes. Since
The ratio between students and teachers varied 1997, the US requirements for postgraduate endodontic
between 9 : 1 and 30 : 1. The average faculty to student training include the use of microscopes (Selden 2002).
ratio of 16 : 1 in Germany is worse than the correspond- However, there is no information available as to what
ing ratio in the UK (12 : 1), which came out the worst in extent these requirements have been implemented. The
an European comparison. The same survey conducted in surveys with regard to endodontic education do not
1999 revealed that Eastern Europe has the best super- reveal any information nor have European guidelines
vision ratio: 6 : 1. A national survey in France, however, made any statements regarding the use of dental
revealed an even worse student to faculty ratio (18 : 1) microscopes.
than the ratio established in the present survey (Arbab- Pan-European surveys on the situation of endodontic
Chirani & Vulcain 2004). The actual outcome-oriented education have repeatedly revealed considerable differ-
guidelines completely refrain from mentioning a tutor- ences (Qualtrough et al. 1999). Shanley et al. (1997)
ing ratio, whereas the guidelines in 1992 still claimed already stated important differences 10 years ago with
that an adequate faculty to student ratio was essential for regard to resources, qualification of staff, availability of
a successful implementation of the endodontic curricu- adequate training facilities as well as to research and
lum (European Society of Endodontology 1992, 2001). patient care within the European Union.
Before undertaking clinical practice on patients
during the seventh term, students performed root canal
Conclusion
treatment on an average of four extracted natural
teeth. Comparable figures for Europe are not available. Pre-clinical endodontic education at German universi-
In the Philippines, however, the average was four teeth, ties varied considerably because of differences in
the same as in Germany (Cruz et al. 2000). The programme design, staff and course content. Students’
undergraduate curriculum guidelines advise ‘…to gain experiences before the treatment of their first patient
the requisite experience on twenty teeth (…including varied substantially. Only a national conference of
extracted teeth)’ during pre-clinical and clinical under- university staff, responsible for restorative dentistry and
graduate education (European Society of Endodontol- in the position to take decisions, will allow for a
ogy 2001). The remaining 16 teeth would, therefore, consistent academic education. Specific and detailed
have to be prepared in clinical courses on patients. recommendations, together with outcome-oriented
Based on the number of patients and on the amount of guidelines, could assist decision makers to improve
time available, it seems highly improbable that this academic education and facilitate step-by-step imple-
objective of education can be reached by the majority of mentation at the various locations.
German universities.
Only 63% of the universities taught root canal
Acknowledgements
preparation with rotary nickel titanium instruments.
In 1999, 27% of the Western European universities We would like to thank all those colleagues of the 27
interviewed stated that rotary preparation methods were clinics interviewed, who took time to reply to our
taught. In Scandinavia, the percentage was 25% (Qualt- questionnaires.
rough et al. 1999). According to a survey conducted in
2004, 81% of the French universities were teaching
References
rotary instrumentation during pre-clinical and clinical
education (Arbab-Chirani & Vulcain 2004). The great Arbab-Chirani R, Vulcain JM (2004) Undergraduate teaching
majority of surveys on rotary preparation confirms that and clinical use of rotary nickel–titanium endodontic
instruments: a survey of French dental schools. International Hülsmann M, Peters OA, Dummer PMH (2005) Mechanical
Endodontic Journal 37, 320–4. preparation of root canals: shaping goals, techniques and
Barrieshi-Nusair KM, AlOmari MA, Al Hiyasat AS (2004) means. Endodontic Topics 10, 30–76.
Radiographic technical quality of root canal treatment Kirkevang LL, Hörstedt-Bindslev P, Ørstavik D, Wenzel A
performed by dental students at the Dental Teaching Center (2001) A comparison of the quality of root canal
in Jordan. Journal of Dentistry 32, 301–7. treatment in two Danish subpopulations examined
Boltacz-Rzepkowska E, Pawlicka H (2003) Radiographic 1974–75 and 1997–98. International Endodontic Journal
features and outcome of root canal treatment carried out 34, 607–12.
in the Lodz region of Poland. International Endodontic Journal Loftus JJ, Keating AP, McCartan BE (2005) Periapical status
36, 27–32. and quality of endodontic treatment in an adult Irish
Boucher Y, Matossian L, Rilliard F, Machtou P (2002) population. International Endodontic Journal 38, 81–6.
Radiographic evaluation of the prevalence and technical Namazikhah MS, Mokhlis HR, Alasmakh K (2000) Compar-
quality of root canal treatment in a French subpopulation. ison between a hand stainless-steel K file and a rotary NiTi
International Endodontic Journal 35, 229–38. 0.04 taper. Journal of the Californian Dental Association 28,
Bundesgesetzblatt (BGBI) (1999) Approbationsordnung für 421–6.
Zahnärzte. Gesetz zur Ausübung der Zahnheilkunde. Bun- Parashos P, Messer HH (2006) Rotary NiTi instrument
desgesetzesblatt Teil III, Gliederungsnummer. 50, 2123-1, fracture and its consequences. Journal of Endodontics 32,
2175–85. 1031–43.
Cailleteau JG, Mullaney TP (1997) Prevalence of teaching Peru M, Peru C, Mannocci F, Sherriff M, Buchanan LS, Pitt
apical patency and various instrumentation and obturation Ford TR (2006) Hand and nickel–titanium root canal
techniques in United States dental schools. Journal of instrumentation performed by dental students: a micro-
Endodontics 23, 394–6. computed tomographic study. European Journal of Dental
Cruz EV, Jimena ME, Puzon EG, Iwaku M (2000) Endodontic Education 10, 52–9.
teaching in Philippine dental schools. International Endodon- Peters OA (2004) Current challenges and concepts in the
tic Journal 33, 427–34. preparation of root canal systems: a review. Journal of
Dummer PMH (1991) Comparison of undergraduate endo- Endodontics 30, 559–67.
dontic teaching programmes in the United Kingdom and in Plasschaert AJ, Lindh C, McLoughlin J et al. (2006) Curricu-
some dental schools in Europe and the United States. lum structure and the European Credit Transfer System for
International Endodontic Journal 24, 169–77. European dental schools: part I. European Journal of Dental
Eleftheriadis GI, Lambrianidis TP (2005) Technical quality of Education 10, 123–30.
root canal treatment and detection of iatrogenic errors in an Qualtrough AJE, Dummer PMH (1997) Undergraduate end-
undergraduate dental clinic. International Endodontic Journal odontic teaching in the United Kingdom: an update.
38, 725–34. International Endodontic Journal 30, 234–9.
Eriksen HM, Kirkevang LL, Petersson K (2002) Endodontic Qualtrough AJE, Whitworth JM, Dummer PMH (1999)
epidemiology and treatment outcome: general consider- Preclinical endodontology: an international comparison.
ations. Endodontic Topics 2, 1–9. International Endodontic Journal 32, 406–14.
European Society of Endodontology (1992) Undergraduate Schulte A, Pieper K, Charalabidou O, Stoll R, Stachniss V (1998)
curriculum guidelines for endodontology. International End- Prevalence and quality of root canal fillings in a German adult
odontic Journal 25, 169–72. population. A survey of orthopantomograms taken in 1983
European Society of Endodontology (2001) Undergraduate and 1992. Clinical Oral Investigation 2, 67–72.
curriculum guidelines for endodontology. International End- Selden HS (2002) The dental-operating microscope and its
odontic Journal 34, 574–80. slow acceptance. Journal of Endodontics 28, 206–7.
Garip Y, Gunday M (2001) The use of computed tomography Shanley DB (2004) Convergence towards higher standards in
when comparing nickel–titanium and stainless steel files international dental education. New York State Dental
during preparation of simulated curved canals. International Journal 70, 35–9.
Endododontic Journal 34, 452–7. Shanley DB, Barna S, Gannon P et al. (1997) Undergraduate
Gluskin AH, Brown DC, Buchanan LS (2001) A reconstructed training in the European Union. Convergence or diver-
computerized tomographic comparison of Ni–Ti rotary GT gence? European Journal of Dental Education 1, 35–43.
files versus traditional instruments in canals shaped by Shovelton DS (1979) The teaching of endodontics in the
novice operators. International Endodontic Journal 34, United Kingdom and Ireland. Journal of the British Endodontic
476–84. Society 12(Suppl.), 9–16.
Hayes SJ, Gibson M, Hammond M, Bryant ST, Dummer PMH Weiger R, Hitzler S, Hermle G, Löst C (1997) Periapical status,
(2001) An audit of root canal treatment performed by quality of root canal fillings and estimated endodontic
undergraduate students. International Endodontic Journal 34, treatment needs in an urban German population. Endo-
501–5. dontics & Dental Traumatology 13, 69–74.
doi:10.1111/j.1365-2591.2008.01440.x
Glucose reactivity with filling materials as a

limitation for using the glucose leakage model
H. Shemesh1, E. M. Souza2, M.-K. Wu1 & P. R. Wesselink1

1
Department of Cariology Endodontology Pedodontology, Academic Center for Dentistry Amsterdam (ACTA), Amsterdam, the
Netherlands; and 2Department of Dentistry and Endodontics, Araraquara Dental School, São Paulo State University, São Paulo,
Brazil
Abstract glucose was evaluated using an enzymatic reaction

after 1 week. Statistical analysis was performed with
Shemesh H, Souza EM, Wu M-K, Wesselink PR. Glucose
the anova and Dunnett tests at a significant level of
reactivity with filling materials as a limitation for using the
P < 0.05.
glucose leakage model. International Endodontic Journal, 41,
Results Portland cement, MTA, Ca(OH)2 and sealer
869–872, 2008.
26 reduced the concentration in the test tube of glucose
Aim To evaluate the reactivity of different endodontic significantly after 1 week (P < 0.05). Calcium sulphate
materials and sealers with glucose and to asses the reduced the concentration of glucose, but the difference
reliability of the glucose leakage model in measuring in concentrations was not significant (P = 0.054).
penetration of glucose through these materials. Conclusions Portland cement, MTA, Ca(OH)2 and
Methodology Ten uniform discs (radius 5 mm, sealer 26 react with a 0.2 mg mL)1 glucose solution.
thickness 2 mm) were made of each of the following Therefore, these materials should not be evaluated for
materials: Portland cement, MTA (grey and white), sealing ability with the glucose leakage model.
sealer 26, calcium sulphate, calcium hydroxide
Keywords: glucose, leakage model, optical density,
[Ca(OH)2], AH26,Epiphany, Resilon, gutta-percha and
reactivity.
dentine. After storing the discs for 1 week at 37C and
humid conditions, they were immersed in 0.2 mg mL)1 Received 29 January 2008; accepted 10 April 2008
glucose solution in a test tube. The concentration of
later to concentration units. The advantages of this

Introduction
model are the relative ease of assembly and operation,
A new leakage model was suggested recently (Xu et al. the availability of the materials and equipment and the
2005) using glucose as the tracer. Several studies were great sensitivity of the test (Shemesh et al. 2006).
published using this model (Shemesh et al. 2006, As leakage studies are constantly being scrutinized
2007, Kaya et al. 2007, van der Sluis et al. 2007, for their clinical relevance and reproducibility (Editorial
Zou et al. 2007, Xu et al. 2007, Ozok et al. 2008) and Board of the Journal of Endodontics, 2007), a critical
more are being conducted. This model is based on approach to the various leakage models is required.
measurements of glucose concentrations in an apical One of the problems of using a tracer for evaluating
chamber using a sensitive enzymatic reaction. A leakage is that the tracer itself could chemically react
coloured substance is produced and optical density with some materials challenging the reliability of such
(OD) is determined by a spectrophotometer, translated a test in evaluating sealing ability. For example,
methylene blue, which is often used for dye-leakage
studies, was previously shown to discolour when in
Correspondence: H. Shemesh, Department of Cariology En- contact with certain filling materials (Wu et al. 1998),
dodontology Pedodontology, Academic Center for Dentistry
Amsterdam (ACTA), Louwesweg 1, 1066 EA, Amsterdam, the
whilst root canal fillings might directly inhibit bacteria
Netherlands (Tel.: +31 20 5188651; fax: +31 20 6692881; in bacterial leakage studies (Pitout et al. 2006). The
e-mail: hshemesh@acta.nl). purpose of this study was to determine the reactivity of
Reliability of glucose leakage model Shemesh et al.
glucose solution with different dental materials to percha cones (Dentsply De Trey, Konstanz, Germany).
assess whether glucose is an acceptable tracer for all Ten moulds were filled with each material. An
tested materials. additional 10 dentine discs of similar dimensions were
created from extracted molar teeth. All moulds and
discs were maintained at 37C and humid conditions.
After 1 week, the set materials were carefully removed
The following materials were mixed according to the from the moulds, forming round discs. Each of the discs
manufacturer’s instructions and inserted into round was then inserted to a small test tube with 4 mL of
plastic moulds 2 mm deep with a diameter of 5 mm : glucose 0.2 mg mL)1 solution. Additional 10 test tubes
Portland cement (Gamma, Leusden, the Netherlands), were used as controls and contained only 4 mL of
MTA grey (Angelus, Londrina, Brazil), MTA white glucose 0.2 mg mL)1 solution. All test tubes were kept
(Angelus), Sealer 26 (Dentsply, Petrópolis, Brazil), at 37C. A sample of 0.1 mL of the solution was
calcium hydroxide powder (Merck, Darmstadt, Ger- removed after 1 week from each test tube and was
many), calcium sulphate (Sigma-Aldrich, Steinheim, analysed with a Glucose kit (Megazyme, Wicklow,
Germany), AH26 (Dentsply-Maillefer, Tulsa, OK, USA) Ireland) in a spectrophotometer (Spectra max 384 plus;
and Epiphany sealer (Pentron Clinical Technologies, Molecular Devices, Sunnyvale, CA, USA) at a wave-
Wallingford, CT, USA). The following thermoplastic length of 340 nm. Results were expressed as OD.
materials were adapted into similar moulds with a Statistical analysis was performed with the anova and
warm instrument: Resilon cones (Pentron), Gutta- the Dunnett test at significance level of P < 0.05 (spss
version 15.0, SPSS Inc., Chicago, IL, USA).
Table 1 Average optical density (OD), SD and P-value of the

OD measured after 1-week immersion with different materials Results
Average P-value (compared with
The results are presented in Table 1 and Fig. 1.
Material OD ± SD glucose solution)
The glucose kit used has a detectable threshold of OD
Water 0 ± 0.005 <0.001 0.05, so all readings lower than 0.05 were ignored. The
Glucose stand. 0.290 ± 0.010 –
glucose test tubes with no test material showed the
Dentine 0.293 ± 0.007 0.999
Portland cement 0 ± 0.007 <0.001 expected OD from a 0.2 mg mL)1 glucose solution (OD
MTA-grey 0 ± 0.004 <0.001 0.3 ± 0.05).
MTA-white 0 ± 0.007 <0.001 Amongst the materials tested, MTA, Portland
Sealer 26 0 ± 0.027 <0.001 cement, sealer 26 and Ca(OH)2 lowered significantly
Ca(OH)2 0 ± 0.005 <0.001
the OD of the solution after 1 week (P < 0.05) .
Calcium sulphate 0.275 ± 0.007 0.054
AH26 0.282 ± 0.007 0.645 Dentine, Resilon, Epiphany, AH26 and gutta-percha
Epiphany 0.290 ± 0.017 1.000 did not significantly alter the concentration of glucose.
Resilon 0.300 ± 0.015 0.382 Calcium sulphate reduced the concentration of glucose
Gutta-percha 0.297 ± 0.008 0.781 but the difference in concentrations was not significant
The kit used has a detectable threshold of OD 0.05. (P = 0.054).
Calculated absorption from 0.2

0.35 mg mL–1 glucose solution according to
the glucose kit used
0.3
0.25
OD values
0.2
0.15
0.1
0.05
0 Figure 1 The measured optical density
(OD) of 0.2 mg mL)1 glucose solution
er
se
TA .
y
Se hite
a( 6
Su 2
te
Ep 26
ny
ta ilon
a
M dc
M -gre
Po ntin
ch
lfa
at
H
co
ha
AH
er
es
an
-w
er
W
lu
ip
al
-p
R
TA
D
rtl
after 1-week immersion with different

G
c.
C
al
ut
C
materials.
G
Shemesh et al. Reliability of glucose leakage model
strate a large range of concentrations (Shemesh et al.

Discussion
2006). In this experiment, a glucose concentration of
After 1 week in contact with glucose solution, a 0.2 mg mL)1 (20 lg per assay) was chosen as it lies
number of the materials tested reduced significantly within the linear range of detection of the glucose kit
the OD of the glucose reaction to the level of water used and could give more accurate information on the
(Fig. 1). variations in glucose concentrations after immersion
As all published glucose leakage tests checked the with different materials .
penetration of glucose for periods ranging from 20 days
(Zou et al. 2007) to 56 days (Shemesh et al. 2006) an
Conclusion
observation period of 1 week seems relevant when
assessing the reactivity of different materials with the The observation that certain materials react with
tracer. glucose suggests that evaluating these materials with
Glucose (an aldohexose) in an alkaline solution is the glucose leakage model might lead to erroneous
slowly oxidized by oxygen, forming gluconic acid: conclusions. An investigation of the influence of tested
1 materials on glucose should always precede glucose
CH2 OH-(CHOH)4 CHO þ O2 leakage tests to validate the conclusions from such
2
! CH2 OH-(CHOH)4 CO2 H studies.
In the presence of sodium hydroxide, for example,

gluconic acid is converted to sodium gluconate (Sow- References
den & Schaffer 1952). This means that glucose will not Camilleri J (2007) Hydration mechanisms of mineral trioxide
be detected by the glucose kit and could thus mask aggregate. International Endodontic Journal 40, 462–70.
leakage in the glucose leakage model. Duarte MA, Demarchi AC, Giaxa MH, Kuga MC, Fraga SC, de
Interestingly, Sealer 26 releases Ca(OH)2 (Duarte Souza LC (2000) Evaluation of pH and calcium ion release of
et al. 2000), whilst MTA contains Ca(OH)2 (Camilleri three root canal sealers. Journal of Endodontics 26, 389–90.
2007). It seems that Ca(OH)2 containing products react Editorial Board of the Journal of Endodontics (2007) Wanted:
a base of evidence. Journal of Endodontics 33, 1401–2.
directly with glucose. This reaction could also be
Kaya BU, Kececi AD, Belli S (2007) Evaluation of the sealing
initiated by traces of Ca(OH)2 left in the canal when
ability of gutta-percha and thermoplastic synthetic polymer-
used as an intra-canal medication. Thus, the leakage
based systems along the root canals through the glucose
measured by glucose concentrations could be influ- penetration model. Oral Surgery, Oral Medicine, Oral Pathol-
enced by this chemical reaction. ogy, Oral Radiology and Endodontics 104, 66–73.
Zou et al. (2007) used the glucose leakage model and Ozok AR, van der Sluis LWM, Wu MK, Wesselink PR (2008)
concluded that calcium sulphate significantly improved Sealing ability of a new polydimethylsiloxane-based root
the sealing ability of 1 mm perforations repaired with canal filling material. Journal of Endododontics 34, 204–7.
composite resin. Although the statistical tests used in Pitout E, Oberholzer TG, Blignaut E, Molepo J (2006) Coronal
this study resulted in a nonsignificant influence of leakage of teeth root-filled with gutta-percha or Resilon root
calcium sulphate on glucose (P = 0.054), the P-value canal filling material. Journal of Endodontics 32, 879–81.
Shemesh H, Wu MK, Wesselink PR (2006) Leakage along
is small and it cannot be excluded that calcium
apical root fillings with and without smear layer using two
sulphate might have an influence. Thus, calcium
different leakage models: a two-month longitudinal ex vivo
sulphate could be better evaluated for its sealing
study. International Endodontic Journal 39, 968–76.
properties by another tracer or model. Shemesh H, van den Bos M, Wu MK, Wesselink PR (2007)
In the glucose leakage model, a 1 mol L)1 glucose Glucose penetration and fluid transport through coronal
solution is used as the tracer and placed in the upper root structure and filled root canals. International Endodontic
chamber of the model assembly (Xu et al. 2005). This Journal 40, 866–72.
concentration is much higher than the linear range van der Sluis LW, Shemesh H, Wu MK, Wesselink PR (2007)
detected by the glucose kits: 4–80 lg of glucose per An evaluation of the influence of passive ultrasonic irriga-
assay (Glucose–HK assay procedure, Megazyme Inter- tion on the seal of root canal fillings. International Endodontic
national Limited, 2004). However, the range of glucose Journal 40, 356–61.
Sowden JC, Schaffer R (1952) The reaction of d-Glucose,
concentrations measured in the apical chamber during
d-Mannose and d-Fructose in 0.035 N Sodium Hydroxide at
a leakage test is dependent on the amount of glucose
35. Journal of the American Chemical Society 74, 499–504.
that penetrated through the canal and can demon-
Reliability of glucose leakage model Shemesh et al.
Wu MK, Kontakiotis EG, Wesselink PR (1998) Decoloration of by using a glucose leakage test. Oral Surgery, Oral
1% methylene blue solution in contact with dental filling Medicine, Oral Pathology, Oral Radiology and Endodontics
materials. Journal of Dentistry 26, 585–9. 104, 109–13.
Xu Q, Fan MW, Fan B, Cheung GS, Hu HL (2005) A new Zou L, Liu J, Yin SH et al. (2007) Effect of placement of calcium
quantitative method using glucose for analysis of endodon- sulphate when used for the repair of furcation perforations
tic leakage. Oral Surgery, Oral Medicine, Oral Pathology, Oral on the seal produced by a resin-based material. International
Radiology and Endodontics 99, 107–11. Endodontic Journal 40, 100–5.
Xu Q, Ling J, Cheung GS, Hu Y (2007) A quantitative
evaluation of sealing ability of 4 obturation techniques
doi:10.1111/j.1365-2591.2008.01445.x
Dentinal tubule invasion and adherence by

Enterococcus faecalis
P. Chivatxaranukul1, S. G. Dashper2 & H. H. Messer1

1
School of Dental Science, University of Melbourne, Melbourne, Australia and 2CRC for Oral Health Science, Molecular Science and
Biotechnology Institute, Melbourne, Australia
Abstract using scanning electron microscopy. Bacterial adhesion

to tubule walls versus fractured OD was calculated as
Chivatxaranukul P, Dashper SG, Messer HH. Dentinal
number of cells per 100 lm2.
tubule invasion and adherence by Enterococcus faecalis. Inter-
Results The strain of E. faecalis used in this study
national Endodontic Journal, 41, 873–882, 2008.
showed moderate to heavy tubule invasion after
Aim To investigate dentinal tubule invasion and the 8 weeks. In the adhesion studies, significantly more
predilection of Enterococcus faecalis for dentinal tubule bacteria adhered to fractured OD than to dentinal
walls. tubule walls (anova, P < 0.001). With respect to
Methodology The invasion of dentinal tubules in the tubule wall, adherence was greater in inner
extracted human teeth by E. faecalis was measured versus outer dentine (P = 0.02) and greater when
ex vivo after 8 weeks of incubation. The canal walls of bacterial adhesion was tested in chemically defined
16 root sections were either intact or instrumented medium than in phosphate-buffered saline (anova,
with or without smear layer present. Extent and P < 0.001).
maximum depth of tubule invasion were assessed Conclusions Although E. faecalis readily invaded
histologically and compared between groups. In the tubules, it did not adhere preferentially to tubule walls.
adherence study, 44 vertically split root samples were Initial colonization of dentinal tubules by E. faecalis
prepared to expose longitudinally aligned dentinal may depend primarily on other factors.
tubules and fractured orthodentine (OD). Surfaces were
Keywords: adherence, dentinal tubules, Enterococ-
exposed to E. faecalis (erythromycin resistant strain,
cus faecalis, invasion.
JH2-2 carrying plasmid pGh9:ISS1) and incubated
aerobically for 2 h. Samples were processed for analysis Received 27 November 2007; accepted 29 April 2008
located in inaccessible areas such as complicated root

Introduction
canal anatomy and dentinal tubules, and it is difficult
Because apical periodontitis is usually caused by to deliver antibacterial agents to these locations
bacteria, a major objective of root canal treatment is (Haapasalo & Ørstavik 1987, Berutti et al. 1997, Wu
to eliminate bacteria from infected root canals. et al. 2006). Bacteria may survive and recolonize the
Although bacterial infection can be substantially root canal space whenever there is opportunity, and
reduced by standard intracanal procedures (Byström this may become a focal source for persistent infection.
& Sundqvist 1981, Byström & Sundqvist 1983, Bys- Bacteria are commonly found within dentinal
tröm et al. 1985, Dalton et al. 1998), it is difficult to tubules of clinically infected canals (Ando & Hoshino
render the root canal free from bacteria. Bacteria are 1990, Love & Jenkinson 2002, Siqueira et al. 2002).
Amongst these bacteria, Enterococcus faecalis is of
interest because it is the most frequently detected
Correspondence: Professor Harold H Messer, Faculty of Med- species in root filled teeth with persistent lesions
icine, Dentistry & Health Sciences, School of Dental Science,
University of Melbourne, 720 Swanston Street, Melbourne,
(Molander et al. 1998, Sundqvist et al. 1998, Peciuliene
Australia (Tel.: +613 9341 1472; fax: +613 9341 1595; et al. 2000, Peciuliene et al. 2001, Rocas et al. 2004).
e-mail: hhm@unimelb.edu.au). Some possible factors facilitating its long-term survival
Enterococcus faecalis and dentinal tubules Chivatxaranukul et al.
in the root canal system are its ability to invade overnight from glycerol stock cultures. Brain heart
dentinal tubules (Ørstavik & Haapasalo 1990, Love infusion broth (Oxoid Ltd, Hampshire, UK) diluted in
2001), where it can survive for a prolonged period deionized water (1 : 5) was used as growth medium
under adverse conditions such as starvation (Hartke throughout the invasion study whilst chemically
et al. 1998, Figdor et al. 2003, Sedgley et al. 2005) and defined medium (CDM) adapted from Pichereau et al.
the high pH of calcium hydroxide medication (Evans (1999) was used in the binding study. CDM lacks any
et al. 2002, Gomes et al. 2002). added proteins or peptides, and consists of amino acids,
Although the mechanism of bacterial invasion is not vitamins and minerals plus glucose. Erythromycin
completely understood, bacterial adhesion to dentinal (6.5 lg mL)1, Sigma, St Louis, MO, USA) was added
tubule walls (TWs) is a logical early step in the process. to media immediately before use. An overnight broth
Collagen is widely considered to be the primary culture was diluted 1 : 10 into 200 mL of fresh
substrate for specific binding of E. faecalis to dentine, medium and incubated at 37 C.
and the collagen binding protein of E. faecalis (Ace) and In all experiments, the purity of all bacterial cultures
a serine protease (Spr) have been proposed to play was determined by both Gram staining and by culture
significant roles in binding to the root canal wall analysis using Brain Heart Infusion (BHI) agar (Oxide
(Hubble et al. 2003). Ace also promotes E. faecalis Ltd., Hampshire, UK) under both aerobic and anaerobic
binding to collagen type I (Nallapareddy et al. 2000a,b) conditions.
and in vitro ace gene expression at 37 C was enhanced
in the presence of collagen (Nallapareddy & Murray
Tubule invasion study
2006). However, the interaction of E. faecalis specifi-
cally with dentinal tubules has never been investigated. Sample preparation
Understanding the pattern and mechanisms of initial Sixteen premolar teeth with oval-shaped canals were
colonization of dentinal tubules by E. faecalis may assist decoronated at the cemento-enamel junction (CEJ), and
the development of more effective clinical procedures 2 mm of the apical root was removed with a diamond
for bacterial management and elimination. The aim of saw. One side of the canal wall was prepared with
this study was to investigate ex vivo tubule invasion Gates–Glidden drills up to two sizes larger than the
and adhesion to TWs by E. faecalis. Dentinal tubules original canal size, whilst the other side of the canal
within intact dentine are normally inaccessible for was left intact (Fig. 1), providing both natural canal
systematic investigation. Thus, for the adherence wall and prepared canal wall in each root segment.
study, human tooth roots were split longitudinally to
expose tubule surfaces. It was hypothesized that the
dentinal TW may be a selective site for bacterial
adherence and that alteration of dentine components
may influence initial colonization.
Teeth
Root samples were prepared from extracted, intact
human teeth from young subjects, with complete root
formation, which had been stored in 1% (w/v) chlor-
amine T. Ethical approval was obtained from the
Health Sciences Human Ethics Subcommittee of the
University of Melbourne (HERC Project No. 050100).
Bacterial strain and growth conditions Figure 1 Histological sample prepared for Enterococcus faecalis
invasion study. One side of the root canal wall was enlarged
Enterococcus faecalis (erythromycin resistant strain, using Gates–Glidden drills (E), whilst the other side remained
JH2-2 carrying plasmid pGh9:ISS1, derived from the intact (I). Prepared and intact root canal walls were divided
parental strain JH2) (Jacob & Hobbs 1974) was grown into three segments each for scoring purposes.
Chivatxaranukul et al. Enterococcus faecalis and dentinal tubules
During canal preparation, root samples were divided sharp blade and prepared for scanning electron micros-
into two groups. In one group (n = 8), root canals were copy (SEM).
irrigated with normal saline solution throughout the
preparation, leaving the smear layer intact. Root canals Sample processing for light and scanning electron
in the other group were irrigated with 1% NaOCl and microscopy
17% ethylenediaminetetraacetic acid (EDTA) was left in The samples for light microscopy were processed
the canal for 3 min before final rinsing with 3 mL of according to standard histological procedures. Tripli-
1% NaOCl solution. Thus, prepared root segments had cate transverse 3-lm decalcified sections from the
smear layer present or absent from the prepared canal coronal third were obtained at the level of 2–3 mm
wall. Then, two dentine blocks per root (3 and 5 mm in from the CEJ and stained with a modified Brown and
height) were obtained by transverse sectioning at levels Brenn staining technique for observing any bacteria
of 3 and 8 mm from the CEJ, using a low speed located in dentinal tubules (Love 2001). Slides were
diamond saw under copious coolant (Fig. 2). A bucco- scanned using an automated scanner (ScanScope,
lingual groove 1.5 mm in depth was prepared at the Aperio Technologies Inc., Vista, CA, USA), which
apical end of the 5 mm block to facilitate vertical provided a digital image for analysis using Image-
splitting after incubation. Prepared samples from each Scope software (Aperio Technologies Inc.).
tooth were individually stored in moist conditions and Samples for SEM were fixed in 2.5% glutaraldehyde
sterilized by gamma irradiation (25 kGy). in PBS, immersed in serial dilutions of acetone, critical
point dried and gold-sputter-coated for examination
Dentinal tubule invasion by Enterococcus faecalis under field emission SEM (FESEM, model XL30 FEG,
Sterilized root canal specimens were inoculated with Philips, Eindhoven, the Netherlands).
5 mL E. faecalis culture grown in diluted BHI broth
with added erythromycin (6.5 lg mL)1) and incubated Scoring of dentinal tubule invasion
(37 C) for 8 weeks. During the incubation period, The scoring system for tubule invasion was modified
4.5 mL of the bacterial suspension was replaced with from the tubule invasion index (TI) (Love 2001). An
fresh medium twice a week. At the end of the imaginary line from the junction of the prepared and
incubation period, the 3-mm dentine blocks were natural root canal wall was drawn from one side of the
removed from the bacterial suspension, washed with root canal wall to the opposite side (Fig. 1). The centre
phosphate-buffered saline (PBS) and processed for of this line was identified and from this position, the
histological examination using light microscopy. The canal wall was divided into six segments including
5-mm dentine blocks were initially fixed with 2.5% three equal lengths of prepared canal sites and three
glutaraldehyde in PBS and longitudinally split with a equal lengths of natural, unprepared canal wall. The
extent of bacterial invasion within each of these lengths
was determined by the number of tubules containing
bacteria (400· magnification). The extent of tubule
invasion by E. faecalis was scored 0, 1, 2 or 3 when 0,
1–20, 20–50 or >50 tubules were infected in the
segment, respectively. The score 0 was interpreted as
nil, 1 as mild, 2 as moderate and 3 as heavy invasion
(Love 2001). The categorical data were then statisti-
cally analysed to compare the effect of smear layer on
the extent of tubule invasion using Pearson’s chi-
squared test, adjusted for clustering within tooth
samples, at 95% significance level.
The maximum depth of tubule invasion in lm was
Figure 2 Diagram of sample preparation for invasion study.
also measured in each of the six compartments, as
Teeth were decoronated and 2 mm of the apical root was
removed. One side of the root canal wall was left intact, whilst mentioned previously, using the imaging processing
the other side was enlarged using Gates–Glidden drills. After software. The maximum penetration depth of prepared
sectioning, 3 and 5-mm dentine blocks were obtained. A 1.5- and natural canal sites from all sections were added
mm bucco-lingual groove was made as the guide for root and averaged for the maximum penetration depth of
splitting. the group. The difference in maximum penetration
depth amongst groups was statistically analysed using samples were removed, washed three times (shaking at
a generalized linear model approach at 95% signifi- 120 rpm in PBS, 5 min) to remove nonadherent
cance level. bacteria and processed for SEM.
To investigate the effect of culture medium on
adhesion, paired dentine samples were exposed to
Adhesion to split dentine surface
bacterial suspensions in either CDM or PBS. Bacteria
Tooth preparation resuspended in PBS were grown in CDM to mid-
Intact, noncarious premolar teeth were decoronated exponential phase (optical density = 0.6, 570 nm)
and split vertically into two halves to expose longitu- centrifuged (8000 g, 20 min, 4 C), and resuspended
dinally aligned dentinal tubules and fractured OD in PBS to the same cell density as the bacteria in CDM.
(Fig. 3). Although fractured OD is not normally acces- Cell density of bacteria in both groups was determined
sible to bacteria, the fractured surface of OD was used by optical density measurement at a wavelength of
as a control for adhesion relative to the TW. The 570 nm before exposure to tooth samples.
fractured surfaces were likely to have abundant
exposed mineralized collagen. Paired samples were Evaluation of adherence to dentinal tubules
utilized to compare a range of treatment effects on At each location (e.g. coronal inner dentine), 10
bacterial adhesion, including sterilization methods randomly selected nonoverlapping fields within the
(autoclaving and gamma irradiation vs. untreated overall area were imaged, beginning at a specific point
freshly split surfaces) and nutrient availability. Eight (e.g. 100 lm from the pulpal surface). Scanning elec-
split samples were assigned to each experimental tron micrographs of inner and outer root dentine were
group. Split samples were sterilized by either autoclav- captured at 100–300 lm from root canal walls and
ing (121 C, 20 min) or gamma irradiation (25 kGy), within 1 mm of the external root surface, respectively.
whilst nonsterilized fresh samples were copiously Bacterial adhesion was calculated as the average
washed with sterile de-ionized water. number of bacterial cells per unit area of dentine
(100 lm2), based on 10 separate fields. Bacteria
Adherence conditions attached to different dentine surfaces (TW vs. fractured
Split dentine samples from each tooth were individually OD) were counted, and the surface area of the two
immersed in 5 mL E. faecalis culture grown in CDM to structures (TW and OD) in each section was measured,
log phase (0.5–0.6 · 108cells per mL), except where using a specific biological imaging program (Olysia
indicated, and incubated aerobically (37 C, 2 h) Imaging Software, Olympus Corporation, Tokyo, Japan)
without shaking. CDM was used throughout this part (Fig. 3).
of the study to avoid any potential interference of The number of bacteria per 100 lm2 and the ratio of
exogenous proteins with binding. After incubation, TW/OD adherence at each location were calculated
and average values from each sample were compared
within and amongst groups with different variables.
Finally, adhesion to different dentine structures, loca-
tions and depths and the effect of sterilization methods
and growth activity on adhesion were statistically
analysed using anova (General Linear Model) at 95%
significance level.
Results
Bacterial invasion of tubules

The strain of E. faecalis used was able to invade dentinal
tubules to a heavy extent. Overall, both the extent of
tubule invasion and the maximum tubule penetration
Figure 3 Scanning electron micrograph showing Enterococcus into the natural root canal wall were significantly
faecalis adherence to two exposed structures, orthodentine greater than into the prepared root canal surface
(OD) and tubule walls (TW) of split dentine surface (9000·). (P < 0.001). There was a low extent of tubule invasion
Table 1 Extent of tubule invasion (median value) and average

of the maximum penetration depths (lm) at different areas of
the canal surface irrigated with either saline (smear layer
present) or 1% NaOCl and 15% EDTA (smear layer removed)
after 8 weeks incubation at 37 C
Extent of tubule Maximum penetration
invasiona depth (meanb ± SEc) (lm)
Unprepared Prepared Unprepared Prepared

canal canal canal canal
Saline 3 0* 202.4 ± 16.2 31.3 ± 12.3**

NaOCl 3 3 193.9 ± 15.3 156.2 ± 25.3
and
EDTA
a
Modified from tubule invasion index (TI) of Love (2001), where
0 = no invasion, 3 = heavy invasion.
b
Average of 81 values from eight roots irrigated with saline or
93 values from eight roots replicates irrigated with NaOCl and
Figure 4 Scanning electron micrograph showing random
EDTA.
c
Robust standard errors which adjust estimates for clustering of tubule invasion by Enterococcus faecalis. Whilst some tubules
values from tooth samples. were densely packed, the adjacent tubule was empty. Pleo-
*Significantly different from all other groups (P < 0.001). morphic change was observed for bacteria densely packed in
**Significantly less than the average maximum penetration dentinal tubules.
depth values of other groups (P < 0.001).
(median value: 0) with shallow bacterial penetration in location ranged from 7.4 to 8.9 and 29.9 to 30.5 cells
prepared canals irrigated with normal saline (i.e. with per 100 lm2, respectively (Table 2). Bacterial adhesion
smear layer present) (Table 1). Depth of tubule pene- to TWs was significantly and consistently less than to
tration by E. faecalis in root canals with smear layer fractured OD (P < 0.001) in all experimental condi-
removed was not significantly different in both tions applied (Table 2). Even when collagen fibres
prepared (156.2 ± 25.3 lm) and unprepared areas of lining the inner TW were clearly visible, little adher-
the canal (193.9 ± 15.3 lm) (anova, P = 0.07). ence occurred (Fig. 7). The average ratio of adherent
Using SEM, bacterial invasion of tubules was irreg- cells to TW versus OD at coronal inner dentine from all
ular, with some tubules containing densely packed cells experiments ranged from 0.3 to 0.5. The ratios were
whilst adjacent tubules were empty (Figs 4 and 5). not affected by most test conditions applied, including
Morphological change was observed in densely packed sterilization method, location and media.
bacteria that appeared to be pleomorphic (Fig. 4),
whereas less densely packed bacteria (Fig. 5) retained Sterilization method
a typical round or oval shape. The surface texture of Whilst the extent of E. faecalis adhesion was similar in
tubules varied considerably with some tubules having freshly split and gamma-irradiated samples (anova,
numerous exposed collagen fibres, whilst adjacent P = 0.6), significantly less adhesion (approximately
tubules had a smooth surface (Fig. 5). Presence or one-third) was found with autoclaved samples
absence of intratubular bacteria was not dependent on (P < 0.001) (Table 2).
the tubule surface. When the smear layer was present,
large numbers of bacteria attached to the smear layer Location
with very few bacterial cells present in underlying Significantly greater adherence of E. faecalis to dentinal
tubules (Fig. 6, Table 1). TWs occurred at inner versus outer dentine (P = 0.02)
(Table 2). Adherence to TWs of coronal dentine was
not significantly greater than in apical dentine
Bacterial adherence
(P = 0.4) (Table 2).
Dentine surfaces
For freshly split or gamma-irradiated dentine exposed Medium
to bacterial cultures in CDM, mean values of bacterial Dentine specimens exposed to bacterial cultures grown in
cells adhering to TW and OD at the coronal inner CDM had approximately double the bacterial adherence
(a)
(b)
Figure 6 Scanning electron micrograph showing root canal

wall covered with smear layer. Bacteria can be seen associated
with the smeared surface and no tubule invasion was present.
recovery in the presence of serum (Figdor et al. 2003).

It was also shown to survive long term in the root canal
even without additional nutrients (Sedgley et al. 2005).
The E. faecalis strain used here has been widely used in
other endodontic studies that investigated the effects of
antimicrobial agents (Turner et al 2004, Rossi-Fedele &
Roberts 2007), starvation (Appelbe & Sedgley 2007),
Figure 5 Scanning electron micrograph showing the pattern resistance to high pH (Evans et al 2002) and most
of tubule invasion. (a) Dentinal tubules at the intact pulpal importantly invasion of dentinal tubules (Love 2001).
surface were randomly invaded by Enterococcus faecalis cells. This study has confirmed the invasion of dentinal
(b) Colonization of E. faecalis in dentinal tubules approximately
tubules by this bacterium. The depth of tubule invasion
100 lm from the pulpal surface. Some tubules were occupied
in the present study is slightly higher than the reported
by bacteria, whilst adjacent tubules were empty.
depth of bacterial penetration in clinical samples with
periapical lesions, which ranged between 10 and
compared to specimens exposed to bacteria resuspended 150 lm (Sen et al. 1995). The heavy extent of tubule
in PBS (anova, P < 0.001) (Table 2). invasion is comparable to results from a previous
experimental study of E. faecalis in which the TI was
used as a scoring system (Love 2001). These results
Discussion
demonstrate that the strain is suitable as a representative
In this study, E. faecalis was chosen as the test organism of bacteria with the ability to invade dentinal tubules.
because it is known to invade dentinal tubules in vitro Significantly less tubule invasion was found in
(Haapasalo & Ørstavik 1987, Love 2002, Weiger et al. prepared canals irrigated with normal saline solution,
2002) and appears to be associated with persistent compared with other groups. This was caused by the
periapical pathosis (Molander et al. 1998, Siqueira & presence of smear layer and smear plugs that impede
Rocas 2004, Stuart et al. 2006). Examples of factors tubule invasion (Love 2002). Greater invasion was
facilitating its presence in filled root canals with observed in prepared canals when the smear layer was
persistent lesions include the ability to invade dentinal removed by irrigation with NaOCl and EDTA. The
tubules (Haapasalo & Ørstavik 1987, Love 2002, similar extent of tubule invasion in natural canal wall
Weiger et al. 2002), ability to tolerate the high pH of and prepared canal with smear layer removed showed
calcium hydroxide (Evans et al. 2002) and ability to that E. faecalis was able to invade tubules in conditions
withstand long periods of starvation with subsequent of both primary infection and persistent infection. This
Table 2 Mean and standard deviation of

Bacterial adherencea (cells per
bacterial adherence and tubule wall 100 lm2)
(TW)/orthodentine (OD) ratio values of
samples in groups with different sterili- Tubule wall Orthodentine
Factors Subgroup (mean ± SD) (mean ± SD) TW/OD ratio
zation methods, locations or media
Sterilization Gamma irradiated 8.9 ± 5.7 29.9 ± 12.0 0.3 ± 0.1
methoda (n = 16)
Autoclaved (n = 16) 3.2 ± 2.0 9.2 ± 6.4 0.4 ± 0.1
Freshly split (n = 16) 7.4 ± 4.2 30.5 ± 27.2 0.3 ± 0.2
Location Coronal inner 9.1 ± 3.6 28.7 ± 5.6 0.3 ± 0.1
(n = 12) Coronal outer 5.0 ± 2.3 23.1 ± 3.4 0.2 ± 0.1
Apical inner 6.1 ± 4.0 23.4 ± 11.8 0.3 ± 0.1
Apical outer 2.9 ± 1.4 19.8 ± 2.7 0.1 ± 0.1
Media CDM 8.7 ± 4.4 24.8 ± 8.9 0.36 ± 0.1
(n = 8) PBS 3.6 ± 1.9 9.5 ± 5.1 0.5 ± 0.4
CDM, chemically defined medium; PBS, phosphate-buffered saline.

a
Paired-split surface samples from the eight roots were used to compare two treatment
methods (gamma irradiation vs. autoclaving, gamma irradiation vs. freshly split and
autoclaving vs. freshly split group). Therefore 48 split root samples from 24 roots were
included in this experiment.
might support the recent finding that equal amounts of BHI and CDM were found to bind readily to collagen-
E. faecalis were detected in both situations using PCR coated wells in a separate study (data not shown).
techniques (Gomes et al. 2006). Whilst the extent of bacterial adhesion to the dentine
The geometry of dentinal tubules with a narrow surface may be influenced by nutrient conditions, it
lumen (1–2 lm) and considerable length (2–3 mm) cannot account for the low adherence of bacteria to
complicates experimental studies of dentine binding. TW relative to OD.
Therefore, vertically split root samples were utilized for Adherence is considered to be the first step for bacterial
the adherence study. TWs throughout their length colonization of host tissue, including tubule invasion,
were equally exposed to bacteria, making systematic and is mediated by bacterial specific cell-surface compo-
investigation possible. Using split root samples also nents (adhesins) (Patti et al. 1994). It has been shown
provided access to fractured OD, which is not normally that Ace (collagen-binding protein of enterococci) pro-
exposed. Bacterial adherence to fractured OD could be moted E. faecalis adhesion to some extracellular matrix
used as a reference for the less mineralized OD surface proteins including collagen type I (Nallapareddy et al.
compared with more highly mineralized peritubular 2000a). As collagen type I is the main organic compo-
dentine of the TW. nent of dentine, it is widely considered to be a major
Although moderate to heavy tubule invasion was substrate for E. faecalis binding to dentine. A study of
found after 8 weeks of incubation, E. faecalis had only mutant strains of E. faecalis also showed that Ace and
limited tendency to adhere to the surface of dentinal serine protease (Spr) played significant roles in E. faecalis
tubules. These observations suggest that E. faecalis has binding to root canal walls (Hubble et al. 2003). In that
only a low predilection for tubules and that coloniza- study, however, the canal walls had been subjected to
tion may be influenced by other factors such as instrumentation with endodontic files, and the resultant
adhesin-related gene expression, nutrient conditions, smear layer (Drake et al. 1994, Hülsmann et al. 2003,
amount of peritubular dentine deposition and ability of Teixeira et al. 2005) was not removed prior to the
bacteria to proliferate as chains (Love et al. 1997). In measurement of binding. Hence, their results may reflect
this study, adherence was evaluated using bacteria bacterial adherence to smeared dentine rather than
grown in CDM rather than in BHI (as was used for binding to natural root canal wall.
tubule invasion). For adherence, it is essential to avoid Most studies of E. faecalis Ace function were per-
the presence of any exogenous proteins or peptides that formed in vitro at 46 C because of the low level of ace
may interfere with binding. It is possible that the gene expression at 37 C (Nallapareddy et al. 2000a,
expression of cell surface receptors is affected by the Nallapareddy & Murray 2006). However, it has been
growth medium used; however, cells grown in both shown in both in vivo and in laboratory studies that
enced by stress such as low nutrient conditions during

the incubation period or the presence of unmineralized
collagen. However, other receptors, substrates, not yet
identified binding sites, bacterial hydrophobicity and
electrostatic charges that may play roles in initial and
nonspecific dentine binding must still be considered.
When dentine specimens were exposed to bacteria
suspended in PBS (i.e. non-nutrient conditions), there
was less bacterial adhesion to the dentine surface
compared with bacterial cultures in CDM. One possible
explanation for this difference is that metabolic activity
may be required for the expression of specific proteins
(receptors) on the cell surface that is expressed mainly
in the presence of an inducer (i.e. collagen type I).
Bacterial cells suspended in PBS would be unable to
synthesize new protein because of lack of an energy
source. However, other explanations cannot be ex-
cluded such as differences in other constituents of the
two media that might influence adherence. In this
regard, CDM rather than a protein containing medium
(such as BHI) was used so that no exogenous proteins
were present during dentine adhesion.
In the present study, less bacterial adherence
occurred to autoclaved dentine, compared with gam-
ma-irradiated and freshly split dentine. As collagen is
generally considered to be the substrate for binding to
dentine (Love and Jenkinson 2002, Hubble et al. 2003),
less E. faecalis adherence to autoclaved teeth may be
caused by alteration of dentine components such as
collagen by autoclaving (White et al. 1994).
The different levels of bacterial adherence at different
Figure 7 Two scanning electron micrographs showing more
locations may result from varying extents of peritubu-
adherence of Enterococcus faecalis to fractured orthodentine
(OD) compared to tubule wall surface (TW) (9000·). This
lar dentine deposition, with its higher mineral and
characteristic was observed whether collagen fibres were lower collagen content in outer root dentine compared
clearly visible on the tubule wall (CF) or whether the tubule with inner dentine. Love (1996) also reported low
had a smooth wall (SW). bacterial adherence (Streptococcus gordonii) to outer
sclerotic dentine. This difference seems to support the
bacterial colonization of dentinal tubules does occur at significance of organic matrix components in bacterial
37 C, including this study. Evidence of anti-Ace binding. The deposition of more mineralized dentine
antibodies detected in human sera collected from with increasing age may explain the variations in
patients with E. faecalis endocarditis has also been adherence amongst tooth samples, which is in agree-
reported (Nallapareddy et al. 2000b). Moreover, in vitro ment with an earlier study (Siqueira et al. 2002).
expression of Ace was shown to be induced during The ability of E. faecalis to grow as chains has been
some specific conditions such as early growth phase suggested as another explanation for the moderate to
(Hall et al. 2007) and the presence of collagen type I or high extent of tubule invasion (Love & Jenkinson
serum (Nallapareddy & Murray 2006) or elevated 2002), in spite of its low affinity for dentinal tubules. It
temperature (Xiao et al. 1998, Hubble et al. 2003). is possible that only adherence to the tubule orifice is
Thus, the regulation of ace gene expression in vivo may required for invasion. After initial attachment to the
be enhanced by specific conditions or the presence of poorly/nonmineralized pre-dentine at the tubule ori-
substrate and it may play a role in E. faecalis binding in fices, deeper penetration may not require specific
vivo. Considerable tubule invasion may also be influ- binding as invasion may result from intratubular cell
growth (Love & Jenkinson 2002). This somewhat Appelbe OK, Sedgley CM (2007) Effects of prolonged exposure
speculative suggestion has not been investigated sys- to alkaline pH on Enterococcus faecalis survival and specific
tematically but deserves further study. gene transcripts. Oral Microbiology and Immunology 22,
In this study, scanning electron micrographs showed 169–74.
Berutti E, Marini R, Angeretti A (1997) Penetration ability of
densely packed bacteria within tubules, with irregular
different irrigants into dentinal tubules. Journal of Endodon-
shape (Fig. 4), whilst less densely packed cells retained
tics 23, 725–7.
normal morphology (Fig. 5). It is possible that bacteria Byström A, Sundqvist G (1981) Bacteriologic evaluation of the
themselves may occupy dentinal tubules and impede efficacy of mechanical root canal instrumentation in end-
access of medium to bacteria, leading to nutrient odontic therapy. Scandinavian Journal of Dental Research 89,
deprived conditions of bacteria in the deeper part of the 321–8.
tubule. SEM in a previous study showed that alteration Byström A, Sundqvist G (1983) Bacteriologic evaluation of the
in shape of some E. faecalis JH2-2 cells was observed effect of 0.5 percent sodium hypochlorite in endodontic
within 3 to 7 weeks after incubation in oligotrophic therapy. Oral Surgery, Oral Medicine, Oral Pathology 55,
conditions (Hartke et al. 1998). This extreme starved 307–12.
condition may be applicable to explain the pleomorphic Byström A, Claesson R, Sundqvist G (1985) The antibacterial
effect of camphorated paramonochlorophenol, camphorated
shape of packed cells in dentinal tubules with possibly
phenol and calcium hydroxide in the treatment of infected
low nutrient condition in this study.
root canals. Endodontics and Dental Traumatology 1, 170–5.
Dalton BC, Ørstavik D, Phillips C, Pettiette M, Trope M (1998)
Conclusion Bacterial reduction with nickel-titanium rotary instrumen-
tation. Journal of Endodontics 24, 763–7.
Under the experimental conditions, a strain of E. faecalis Drake DR, Wiemann AH, Rivera EM, Walton RE (1994)
readily invaded dentinal tubules even though the Bacterial retention in canal walls in vitro: effect of smear
organism seemed to have low affinity for TWs. Dentine layer. Journal of Endodontics 20, 78–82.
components may include selective factors for bacterial Evans M, Davies JK, Sundqvist G, Figdor D (2002) Mechanisms
adherence and invasion as implied by the effect of involved in the resistance of Enterococcus faecalis to calcium
autoclave sterilization. Also, adherence was higher in hydroxide. International Endodontic Journal 35, 221–8.
Figdor D, Davies JK, Sundqvist G (2003) Starvation survival,
the less mineralized inner dentine than outer dentine.
growth and recovery of Enterococcus faecalis in human
However, low adherence was observed regardless of the
serum. Oral Microbiology and Immunology 18, 234–9.
appearance of the TW surface. Therefore, the initial Gomes BP, Ferraz CC, Garrido FD et al. (2002) Microbial
colonization of dentinal tubules by E. faecalis may be susceptibility to calcium hydroxide pastes and their vehicles.
primarily dependent on other factors such as the Journal of Endodontics 28, 758–61.
environmental conditions inducing surface receptor Gomes BP, Pinheiro ET, Sousa EL et al. (2006) Enterococcus
related gene expression. The further investigation of faecalis in dental root canals detected by culture and by
such factors that may play roles in tubule colonization polymerase chain reaction analysis. Oral Surgery, Oral
and possible competitive agents is still necessary. Under- Medicine, Oral Pathology, Oral Radiology, and Endodontics
standing these mechanisms may lead to development of 102, 247–53.
novel clinical strategies to inhibit bacterial colonization Haapasalo M, Ørstavik D (1987) In vitro infection and
disinfection of dentinal tubules. Journal of Dental Research
and to manage these infections in the future.
66, 1375–9.
Hall AE, Gorovits EL, Syribeys PJ et al. (2007) Monoclonal
Acknowledgements antibodies recognizing the Enterococcus faecalis collagen-
binding MSCRAMM Ace: conditional expression and bind-
This work was funded by the Australian Society of ing analysis. Microbial Pathogenesis 43, 55–66.
Endodontology and the Australian Dental Research Hartke A, Giard JC, Laplace JM, Auffray Y (1998) Survival of
Foundation. We would also like to thank Geoff Adams Enterococcus faecalis in an oligotrophic microcosm: changes
for helpful discussions on statistical analysis of data. in morphology, development of general stress resistance,
and analysis of protein synthesis. Applied and Environmental
Microbiology 64, 4238–45.
References Hubble TS, Hatton JF, Nallapareddy SR, Murray BE, Gillespie
MJ (2003) Influence of Enterococcus faecalis proteases and
Ando N, Hoshino E (1990) Predominant obligate anaerobes
the collagen-binding protein, Ace, on adhesion to dentin.
invading the deep layers of root canal dentin. International
Oral Microbiology and Immunology 18, 121–6.
Hülsmann M, Heckendorff M, Lennon A (2003) Chelating inhibits salt-induced cross-tolerance towards lethal treat-
agents in root canal treatment: mode of action and indications ment in Enterococcus faecalis. Microbiology 145, 427–35.
for their use. International Endodontic Journal 36, 810–30. Rocas IN, Siqueira JF Jr, Santos KR (2004) Association of
Jacob AE, Hobbs SJ (1974) Conjugal transfer of plasmid-borne Enterococcus faecalis with different forms of periradicular
multiple antibiotic resistance in Streptococcus faecalis var. diseases. Journal of Endodontics 30, 315–20.
zymogenes. Journal of Bacteriology 117, 360–72. Rossi-Fedele G, Roberts AP (2007) A preliminary study
Love RM (1996) Adherence of Streptococcus gordonii to investigating the survival of tetracycline resistant Entero-
smeared and nonsmeared dentine. International Endodontic coccus faecalis after root canal irrigation with high
Journal 29, 108–12. concentrations of tetracycline. International Endodontic Jour-
Love RM (2001) Enterococcus faecalis – a mechanism for its role nal 40, 772–7.
in endodontic failure. International Endodontic Journal 34, Sedgley CM, Lennan SL, Appelbe OK (2005) Survival of
399–405. Enterococcus faecalis in root canals ex vivo. International
Love RM (2002) The effect of tissue molecules on bacterial Endodontic Journal 38, 735–42.
invasion of dentine. Oral Microbiology and Immunology 17, Sen BH, Piskin B, Demirci T (1995) Observation of bacteria
32–7. and fungi in infected root canals and dentinal tubules by
Love RM, Jenkinson HF (2002) Invasion of dentinal tubules by SEM. Endodontics and Dental Traumatology 11, 6–9.
oral bacteria. Critical Reviews in Oral Biology & Medicine 13, Siqueira JF Jr, Rocas IN, Lopes HP (2002) Patterns of microbial
171–83. colonization in primary root canal infections. Oral Surgery,
Love RM, McMillan MD, Jenkinson HF (1997) Invasion of Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics
dentinal tubules by oral streptococci is associated with 93, 174–8.
collagen recognition mediated by the antigen I/II family of Siqueira JF Jr, Rocas IN (2004) Polymerase chain reaction-
polypeptides. Infection and Immunity 65, 5157–64. based analysis of microorganisms associated with failed
Molander A, Reit C, Dahlen G, Kvist T (1998) Microbiological endodontic treatment. Oral Surgery, Oral Medicine, Oral
status of root-filled teeth with apical periodontitis. Interna- Pathology, Oral Radiology, and Endodontics 97, 85–94.
tional Endodontic Journal 31, 1–7. Stuart CH, Schwartz SA, Beeson TJ, Owatz CB (2006)
Nallapareddy SR, Qin X, Weinstock GM, Hook M, Murray BE Enterococcus faecalis: its role in root canal treatment failure
(2000a) Enterococcus faecalis adhesin, Ace, mediates attach- and current concepts in retreatment. Journal of Endodontics
ment to extracellular matrix proteins collagen type IV and 32, 93–8.
laminin as well as collagen type I. Infection and Immunity 68, Sundqvist G, Figdor D, Persson S, Sjogren U (1998) Microb-
5218–24. iologic analysis of teeth with failed endodontic treatment
Nallapareddy SR, Singh KV, Duh RW, Weinstock GM, Murray and the outcome of conservative re-treatment. Oral Surgery,
BE (2000b) Diversity of ace, a gene encoding a microbial Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics
surface component recognizing adhesive matrix molecules, 85, 86–93.
from different strains of Enterococcus faecalis and evidence for Teixeira CS, Felippe MC, Felippe WT (2005) The effect of
production of Ace during human infections. Infection and application time of EDTA and NaOCl on intracanal smear
Immunity 68, 5210–7. layer removal: a SEM analysis. International Endodontic
Nallapareddy SR, Murray BE (2006) Ligand-signaled upregu- Journal 38, 285–90.
lation of Enterococcus faecalis ace transcription, a mechanism Turner SR, Love RM, Lyons KM (2004) An in-vitro investi-
for modulating host-E. faecalis interaction. Infection and gation of the antibacterial effect of nisin in root canals and
Immunity 74, 4982–9. canal wall radicular dentine. International Endodontic Journal
Ørstavik D, Haapasalo M (1990) Disinfection by endodontic 37, 664–71.
irrigants and dressings of experimentally infected dentinal Weiger R, de Lucena J, Decker HE, Lost C (2002) Vitality status
tubules. Endodontics and Dental Traumatology 6, 142–9. of microorganisms in infected human root dentine. Interna-
Patti JM, Allen BL, McGavin MJ, Hook M (1994) MSCRAMM- tional Endodontic Journal 35, 166–71.
mediated adherence of microorganisms to host tissues. White JM, Goodis HE, Marshall SJ, Marshall GW (1994)
Annual Review of Microbiology 48, 585–617. Sterilization of teeth by gamma radiation. Journal of Dental
Peciuliene V, Balciuniene I, Eriksen HM, Haapasalo M (2000) Research 73, 1560–7.
Isolation of Enterococcus faecalis in previously root-filled canals Wu MK, Dummer PM, Wesselink PR (2006) Consequences
in a Lithuanian population. Journal of Endodontics 26, 593–5. of and strategies to deal with residual post-treatment root
Peciuliene V, Reynaud AH, Balciuniene I, Haapasalo M (2001) canal infection. International Endodontic Journal 39, 343–
Isolation of yeasts and enteric bacteria in root-filled teeth 56.
with chronic apical periodontitis. International Endodontic Xiao J, Hook M, Weinstock GM, Murray BE (1998) Conditional
Journal 34, 429–34. adherence of Enterococcus faecalis to extracellular matrix
Pichereau V, Bourot S, Flahaut S, Blanco C, Auffray Y, proteins. FEMS Immunology and Medical Microbiology 21,
Bernard T (1999) The osmoprotectant glycine betaine 287–95.
doi:10.1111/j.1365-2591.2008.01449.x
Influence of cyclic torsional loading on the fatigue

resistance of K3 instruments
M. G. A. Bahia1, M. C. C. Melo1 & V. T. L. Buono2

1
Department of Restoration Dentistry, Faculty of Dentistry, Federal University of Minas Gerais, Belo Horizonte, MG; and
2
Department of Metallurgical and Materials Engineering, School of Engineering, Federal University of Minas Gerais, Belo Horizonte,
MG, Brazil
Abstract in torsion and flexural fatigue. Data obtained were

subjected to a one way analysis of variance (anova) at
Bahia MGA, Melo MCC, Buono VTL. Influence of cyclic
95% confidence level.
torsional loading on the fatigue resistance of K3 instruments.
Results Cyclic torsional loading caused no significant
International Endodontic Journal, 41, 883–891, 2008.
differences in maximum torque or in maximum angular
Aim To evaluate the influence of cyclic torsional deflection of the instruments analysed, but comparative
loading on the flexural fatigue resistance and torsional statistical analysis between measured NCF values of new
properties of rotary NiTi instruments. and previously cycled K3 instruments showed signifi-
Methodology Twelve sets of new K3 instruments, cant differences for all tested instrument. Longitudinal
sizes 20, 25 and 30 with an 0.04 taper, and sizes 20 cracks, that is, cracks apparently parallel to the long axis
and 25 with an 0.06 taper, were torsion tested until of the instruments cycled in torsion was observed.
rupture, to establish their mean values of maximum Conclusions Cyclic torsional loading experiments in
torque and angular deflection. Twelve new K3 instru- new K3 rotary endodontic instruments showed that
ments of each of the following dimensions, size 30, torsional fatigue decreased the resistance of these
0.04 taper and sizes 20 and 25 with 0.06 taper, were instruments to flexural fatigue, although it did not
tested to failure by rotation bending in a fatigue test affect their torsional resistance.
device. Cyclic torsional loading was performed in 20
Keywords: cyclic loading, endodontic instruments,
cycles from zero angular deflection to 180 and then
fatigue resistance, longitudinal cracks, nickel–titanium.
return to zero applied torque. After cyclic loading, the
same number of instruments were tested until rupture Received 15 February 2008; accepted 16 May 2008
narrow root canals whilst maintaining the original

Introduction
anatomy is the major characteristics of these instru-
The peculiar properties of shape memory alloys are ments (Schäfer & Florek 2003).
related to a reversible solid-to-solid phase transforma- During root canal shaping, rotary NiTi instruments
tion, the martensitic transformation, which can be undergo flexural and torsional cyclic loads simulta-
thermally or stress-induced (Otsuka & Wayman 1998). neously. These time-varying stresses can lead to fatigue
One of these properties is superelasticity, which has and failure (Pruett et al. 1997). Cyclic loading is one of
allowed for the development of NiTi endodontic instru- the generic characteristic features of many of the
ments. Facilitating the preparation of curved and present applications of NiTi shape memory alloys. It is
well accepted that the behaviour of different engineer-
ing materials under cyclic loading depends on material
Correspondence: Prof. Vicente T. L. Buono, Department of strength, microstructure, surface quality and fatigue
Metallurgical and Materials Engineering, Federal University of
Minas Gerais, Rua Espı́rito Santo 35 room 206, 30160-030
loading type (Eggeler et al. 2004). Cyclic loading is
Belo Horizonte, MG, Brazil (Tel.: +55 31 3238 1859; fax: +55 associated with structural and functional fatigue, both
31 3238 1815; e-mail: vbuono@demet.ufmg.br). limiting the service life of moving components.
Cyclic torsional loading of K3 instruments Bahia et al.
Structural fatigue relates to the microstructural dam- the material, this type of loading significantly decreased
age that accumulates during cyclic loading, which flexural fatigue resistance.
eventually leads to fatigue failure. The term functional There is currently little information available on
fatigue indicates that during cyclic loading, the mate- torsional fatigue itself. Best et al. (2004) assessed the
rial generally suffers a decrease in functional properties endurance limit of a size 30, 0.06 taper ProFile
(Eggeler et al. 2004). instrument, revealing that 106 cycles were completed
In rotational bending or torsional fatigue, the mate- without instrument fracture at an angular deflection
rial fails after being subjected to repeated cycling at of 2.5.
strain levels below those which cause failure upon This study was performed with the objective of
monotonic loading. So far, the majority of fatigue clarifying the effects of cyclic torsional loading on the
experiments have been performed under uniaxial or mechanical behaviour and fatigue resistance of rotary
rotational bending conditions (Predki et al. 2006). NiTi NiTi endodontic instruments. Cyclic torsional loading
rotary instruments, however, are generally subjected to was carried out in new K3 instruments (SybronEndo,
multiaxial loading, that is, tension, compression and Orange, CA, USA) and its influence was evaluated in
shear, during curved root canal preparation. torsion and flexural fatigue tests.
Each time the continually rotating NiTi instrument
meets resistance, it undergoes torsional loading. The
Material and methods
load is higher whenever the dentine is hard or the
canal diameter is small. Acting on the instrument A total of 192 new K3 instruments (SybronEndo), in
surface, this torsional load can prevent its rotation to a sizes and tapers 20/0.04, 25/0.04, 30/0.04, 20/0.06
greater or lesser extent. Although this is the principle and 25/0.06, were evaluated.
by which dentine can be removed, in extreme cases,
when the resistance is so high that it constrains the
Torsion tests
instrument, it may fracture (Berutti et al. 2003). In
addition, the repeated torsional loading and unloading Twelve new K3 instruments of each size and taper
applied to rotary NiTi instruments during clinical use considered, totalling 60 instruments, were torsion
can lead to torsional fatigue. tested until rupture to establish their mean values of
Rotational bending fatigue of NiTi rotary instru- torque to failure and maximum angular deflection. The
ments has been assessed extensively (Pruett et al. torsion tests were performed based on International
1997, Haı̈kel et al. 1999, Yared et al. 2000, Bahia & Organization for Standardization ISO 3630-1 (1992)
Buono 2005) and appears to have a cumulative using a torsion machine described in detail elsewhere
effect on instruments, causing a weakening over (Bahia et al. 2006). In brief, torque values were
time. It is affected by the angle and radius of canal assessed by measuring the force exerted on a small
curvature and the diameter of the instrument at the load cell by a lever arm linked to the torsion axis.
point of maximum flexure in the canal (Pruett et al. Measurement and control of the rotation angle were
1997), indicating that the fatigue resistance of the performed by a resistive angular transducer connected
NiTi instruments is inversely proportional to the to a process controller. The rotation speed was set
maximum tensile strain amplitude to which they clockwise to 2 rpm. Before testing, each instrument
were submitted (Bahia & Buono 2005, Melo et al. handle was removed at the point where the handle is
2008). Moreover, recent reports indicated that flex- attached to the shaft. The end of the shaft was clamped
ural loads, developed during curved root canal into a chuck connected to a reversible geared motor.
shaping, may decrease the torsional resistance of Three millimetres of the instrument’s tip were clamped
the instruments (Yared et al. 2003a,b, Ullmann & in another chuck with brass jaws to prevent sliding.
Peters 2005, Bahia et al. 2006). Continuous recording of torque and angular deflection
Conversely, the influence of torsional loads on as well as measurements of the maximum torque and
flexural fatigue has received very little attention. The angular deflection were provided by a specifically
influence of previously applied monotonically torsional designed computer program (Analógica, Belo Horizon-
loading on the flexural fatigue behaviour has also te, MG, Brazil).
recently been evaluated by Barbosa et al. (2007) in size For cyclic torsional loading testing, the machine was
25, 0.06 taper K3 instruments. These authors observed programmed to repeatedly rotate from zero angular
that even with torsional loads below the elastic limit of deflection to 180 and then return to zero applied
Bahia et al. Cyclic torsional loading of K3 instruments
torque. Each rotation was defined as one cycle. Twelve Secondary electron images were recorded at length
new K3 instruments of each size and taper, totalling 60 intervals that enabled the observation of the cutting
instruments were subjected to 20 cycles of torsional sections perpendicular to the electron beam, from the
loading as described. Subsequently, these instruments instrument tip onwards, up to approximately 4.0 mm
were tested in torsion until rupture, under the same from the tip, thus encompassing the area submitted to
conditions previously applied to the other set of new the most severe conditions of cyclic straining during
instruments. the tests.
The fracture surfaces of three instruments of each
size and taper, randomly selected after the torsion tests
Fatigue tests
to failure, were analysed by SEM to evaluate the
New K3 instruments of size 30, 0.04 taper and sizes 20 features associated with the failure process. Before SEM
and 25, 0.06 taper, 12 of each type, totalling 36 evaluation, instruments were ultrasonically cleaned to
instruments, were tested to failure by means of rota- remove debris.
tional bending in a fatigue test bench device, to
determine their mean number of cycles to failure
(NCF). These specific types of instruments were chosen
because of their larger diameters, which make them Data obtained in the torsion and fatigue tests were
more prone to fatigue failure during clinical use. The subjected to a one-way analysis of variance (anova).
tests were carried out in a bench device described by Significance was determined at a 95% confidence level.
Bahia & Buono (2005), in which the files rotate freely
inside an artificial canal made up of AISI H13 tool steel,
Results
consisting of an arch whose angle of curvature was
45, with a radius of 5 mm and a guide cylinder of
Torsional behaviour
10 mm in diameter, made of the same material. The
artificial canal geometry was chosen in accordance Typical cyclic torsional loading curves of a new size 30,
with previously determined mean values of angle and 0.04 taper K3 instrument submitted to 20 loading
radius of curvature (Bahia & Buono 2005, Martins cycles are illustrated in Fig. 1. These curves show that
et al. 2006, Vieira et al. 2008). The chosen geometry because of the residual deformation, cycling must be
placed the area of maximum tensile strain amplitude performed between 180 and zero applied torque. It can
approximately 3 mm from the tip of the instrument. also be observed that there is a tendency towards
After machining, the artificial canal was quenched to stabilization of the cyclic torsional behaviour after cycle
prevent wear by friction with the rotating files. During number 2.
the tests, friction was minimized by the use of a mineral The mean values of maximum torque and angular
oil as a lubricant. The time to fracture was recorded deflection at fracture of new K3 instruments and of
using a digital chronometer and converted to NCF by those previously submitted to 20 cycles of torsional
multiplying it by the rotation speed (300 rpm). The
point of fracture in relation to the tip of the instrument
was determined by measuring the fractured instrument
with an endodontic rule.
Thirty-six other new K3 instruments of size 30, 0.04
taper and sizes 20 and 25, 0.06 taper, 12 of each type,
were subjected to the same 20 cycles of torsional
loading, as described above and then tested until
rupture in fatigue through rotational bending. The
same conditions applied to the new instruments were
also used so as to assess the effect of cyclic straining in
torsion on the fatigue resistance of these instruments.
Before and after testing, three instruments of each
size and taper, randomly selected, were examined by
scanning electron microscopy (SEM) (Jeol 6360LV, Figure 1 Typical cyclic torsional loading curves of a new size
Tokyo, Japan) to evaluate their surface characteristics. 30, 0.04 taper K3 instrument submitted to 20 loading cycles.
(a)
(b) Figure 3 Mean values (standard deviations shown as error

bars) of number of cycles to failure (NCF) of new K3
instruments and of instruments previously cycled in torsion.
analysis between measured NCF values of new and

previously cycled K3 instruments showed significant
differences for all tested instrument pairs (P < 0.05),
indicating that cyclic torsional loading decreased the
fatigue resistance of these instruments.
The location of fracture, expressed as the distance
relative to the instrument tip, presented a mean value
of 3.0 ± 0.1 mm, with no statistically significant
difference amongst the different sizes or between new
and cycled instruments.
Figure 2 Mean values (standard deviations shown as error
bars) of (a) maximum torque and (b) angular deflection at
fracture measured in new K3 instruments and in instruments Surface characteristics of cycled instruments
previously cycled in torsion.
As illustrated in Fig. 4, the lateral surfaces of two size
20, 0.06 taper K3 instruments submitted to 20 cycles
loading are illustrated in Fig. 2. As usual, torsional of torsional loading exhibited longitudinal cracks, that
resistance increased as the diameter of the instru- is, cracks apparently parallel to the long axis of the
ments increased, but no apparent correlation existed instrument. Remarkable in these images is the fact
between angular deflection and instrument diameter. that the two crack patterns showed a strong similarity
Statistical analysis of the influence of cyclic loading between each other. Figure 5 shows examples of the
in torsion showed no significant differences in max- cracks found on the fracture surface of instruments
imum torque or in maximum angular deflection tested in flexural fatigue: in Fig. 5a, where the fracture
between pairs of same size and taper instruments surface of a new instrument is presented, most of the
assessed. cracks are perpendicular to the radius of the instru-
ment (dark arrows); in Fig. 5b, showing the fracture
surface of an instrument previously submitted to 20
Fatigue behaviour
cycles of torsional loading, a crack running along the
The mean NCF values determined in the fatigue tests of radius of the instrument’s cross-section can be
new K3 instruments and of those submitted to cyclic observed (white arrows). Cracks of this type were
loading in torsion are shown in Fig. 3. Fatigue resis- frequently found on fracture surfaces of instruments
tance of new instruments decreased as the diameter of cycled in torsion but were not found in fatigue-tested
the instruments increased. Comparative statistical new instruments.
(a) (a)
(b) (b)
Figure 4 Longitudinal cracks (arrowed) in two size 20, 0.06 Figure 5 Border of the fracture surface of size 25, 0.06 taper
taper K3 instruments submitted to 20 torsional loading cycles. K3 instruments tested in flexural fatigue: (a) new instrument,
Similar crack patterns were found in different instruments, (a) (b) instrument previously submitted to 20 cycles of torsional
and (b). loading.
Flexural fatigue and torsional overload have been

Discussion
identified as the main reasons for rotary nickel–
According to American Society for Testing of Mate- titanium instrument fracture (Sattapan et al. 2000).
rials ASTM E 1823 (2005) designation, fatigue is the Torsional overload generally occurs when a substantial
process of progressive localized permanent structural area of an instrument encounters excessive friction on
change occurring in a material subjected to condi- a canal wall, when the instrument tip is larger than the
tions that produce fluctuating stresses and strains at canal, or when excessive pressure is placed on the
some point or points and that may culminate in handpiece. Under these situations, the tip may lock,
cracks or complete fracture after a sufficient number leading to large increases in torsional stress. The torque
of fluctuations. Endodontic instruments are subjected developed by the motor may exceed a critical level, thus
to both torsional and flexural stresses during root causing the instrument to undergo plastic deformation
canal preparation and these types of stress can lead and failure (Gambarini 2000).
to metal fatigue and failure. In a recent review, The results of the torsion tests shown in Fig. 2
Parashos & Messer (2006) reported that rotary NiTi indicated that the resistance of K3 instruments to
instrument fracture is most likely a rare occurrence torsional loads significantly increased in accordance
in clinical practice, but structural damage caused by with instrument diameter and are in agreement with
fatigue is frequently observed in endodontic instru- other reports for K3 (Yared et al. 2003a,b, Melo et al.
ments (Bahia & Buono 2005, Peng et al. 2005, Vieira 2008) and other instruments such as ProFile
et al. 2008). (Wolcott & Himel 1997, Svec & Powers 1999, Peters
& Barbakow 2002, Bahia et al. 2006) and ProTaper without instrument fracture at an angular deflection
(Peters et al. 2003) (ProFile/ProTaper; Dentsply-Maille- of 2.5.
fer, Ballaigues, Switzerland). In fact, a definite correla- Peters et al. (2003) established that torque was
tion was found between the maximum torque of K3 correlated not only with the apically exerted force,
instruments and the diameter and cross-sectional area but also with the preoperative canal volume. Hence,
at 3 mm from the instrument tip (Melo et al. 2008). the shaping of narrow and constricted canals can
Concerning the fatigue resistance of the K3 instru- subject rotary instruments to higher torsional loads
ments analysed, the results revealed a tendency of NCF and high apically directed forces. Thus, each time the
to decrease as the diameter of the instrument increased continually rotating NiTi instrument meets resistance,
(Fig. 2), in agreement with the literature. Previous it also undergoes torsional loading, whose extent
studies demonstrated that canal curvature and instru- depends on dentine hardness and canal diameter.
ment diameter at the point of maximum curvature are Acting on the instrument surface, this torsional cyclic
important parameters: the smaller the radius of curva- loading can lead to torsional fatigue.
ture and the greater the diameter, the shorter the In this study, a specific loading cycle was used to
lifespan of the rotary instrument (Pruett et al. 1997, simulate torsional fatigue in rotary instruments during
Haı̈kel et al. 1999, Gambarini 2001, Melo et al. 2002, clinical use. The loading cycle consisted of 20 repeti-
Bahia & Buono 2005). In addition, evaluation of the tions of torsion from zero angular deflection to 180
tensile strain amplitudes on the surface of rotary and back to zero load. The last step in this cycle
instruments, taking into account the instrument diam- guaranteed that nonrecovered strains remained
eter and the radius of curvature of the canal, indicated unchanged for the next cycle (Fig. 1). The occurrence
that fatigue resistance varies inversely with the max- of this type of strain is a common feature in superelastic
imum tensile strain amplitude to which the instru- NiTi alloys, which is associated with the generation of
ments are submitted in the root canal (Bahia & Buono dislocation and the presence of untransformed mar-
2005, Cheung & Darvell 2007). tensite variants near grain boundaries, as discussed in
Flexural fatigue of NiTi rotary instruments has been the findings from Bahia et al. (2005). The fact that
extensively evaluated in the literature (Pruett et al. changes on the load–unload curves of Fig. 1 decreased
1997, Haı̈kel et al. 1999, Gambarini 2001, Melo et al. as the number of cycles increased is related to the
2002, Peters et al. 2003, Bahia & Buono 2005, Grande saturation of these internal defects. The number of
et al. 2006, Cheung & Darvell 2007) and, as previously cycles employed in the load–unload tests was selected
reported by various authors for a variety of instrument based on the assumption that an instrument requires
types (Gambarini 2001, Fife et al. 2004, Bahia & an average of 24 revolutions in three strokes to shape
Buono 2005, Plotino et al. 2006, Vieira et al. 2008), one root canal (Peters & Barbakow 2002). Thus, 20
appears to have a cumulative effect, leading to a torsional strain cycles should be roughly equivalent to
reduction in the remaining fatigue life of clinically used the instrument’s use in six root canals, which repre-
instruments. Furthermore, recent reports indicated that sents approximately half of the recommended number
flexural loads, developed during curved root canal of uses for rotary instruments (Yared et al. 2000,
shaping, may decrease the instrument’s torsional Gambarini 2001, Foschi et al. 2004, Bahia & Buono
resistance (Yared et al. 2003a,b, Ullmann & Peters 2005). The maximum angular deflection of 180 was
2005, Bahia et al. 2006). chosen based on results from the torsional behaviour of
The influence of previously applied monotonical K3 instruments (Melo et al. 2008), showing that this
torsional loading on flexural fatigue has recently been deflection is within the range of superelastic straining
evaluated by Barbosa et al. (2007) in size 25, 0.06 in torsion.
taper K3 instruments. The authors observed that even The longitudinal cracks observed in instruments
with torsional loads below the elastic limit of the submitted to cyclic deformation in torsion analysed by
material, this type of loading significantly decreased SEM (Fig. 4) have been described previously (Alapati
flexural fatigue resistance. However, the actual con- et al. 2003, Peng et al. 2005, Tripi et al. 2006, Vieira
tribution of cyclic loading in torsion to potential et al. 2008). It has been suggested that these types of
instrument failure has not yet been reported in the cracks reflect the orientation of the stress on the surface
literature. Best et al. (2004) evaluated the torsional of the instrument under torsional load. During cyclic
endurance limit of size 30, 0.06 taper ProFile instru- torsion, planes with a maximum shear stress are either
ments, revealing that 106 cycles were completed perpendicular or parallel to the longitudinal axis, whilst
the normal stress component on the slip plane is zero. prevent torsional fatigue. As a matter of fact,
Microscopic investigations have shown that micro- the problem may even be worsened if the motors are
cracks nucleate in a slip band under cyclic torsion and programmed to reverse the rotational motion when
then grow further in a direction perpendicular to the the instrument cannot advance further, considering
main stress. In a cylindrical bar, this direction makes the fact that this can make the longitudinal cracks
an angle of 45 with the axis of the bar. As a propagate even further.
consequence, cracks in a round axle under cyclic Finally, it is important to observe that the findings of
torsion grow as a spiral around its surface (Schijve this work suggest that torsional fatigue can play an
2001). The longitudinal appearance of the cracks important role in the failure of rotary endodontic
observed in endodontic instruments is because of the instruments, decreasing their resistance to flexural
fact that the instruments have helical shapes and that fatigue. Although this was not tested here, it is also
the cracks, being rather small in size, require large reasonable to expect that if the instrument becomes less
magnifications to be observed. resistant to flexural fatigue, its torsional resistance may
Vieira et al. (2008) observed that, after clinical use in also decrease (Yared et al. 2003a,b, Ullmann & Peters
five and eight molars, ProTaper instruments had 2005, Bahia et al. 2006).
microcracks transverse to the cutting edge, probably
associated with instrument bending in the curved
Conclusions
section of the root canals and longitudinal cracks,
similar to those shown in Fig. 4, suggesting that they Cyclic torsional loading experiments in new K3 rotary
reflect the direction of stress on the surface of the endodontic instruments revealed that torsional fatigue
instrument under torsional load. In fact, these cracks decreased the resistance of these instruments to flexural
are possibly the result of both repeated flexural and fatigue, although it did not affect their torsional
torsional loading, indicating that instrument fracture resistance. This behaviour is probably associated with
may take place as a result of triaxial stresses. the generation of longitudinal cracks during torsional
The results of this study showed that torsional cyclic loading cycles, which can act as nucleation sites for
loading did not affect torsional resistance but did flexural fatigue cracks but introduce no appreciable
decrease the flexural fatigue resistance of K3 instru- change in the cross-sectional area of the instruments.
ments. The longitudinal cracks generated during tor- These results suggest that the failure of rotary NiTi
sional cyclic loading can be responsible for this endodontic instruments results concomitantly from the
behaviour because they can act as nucleation sites for processes of torsional overload and flexural and
flexural fatigue cracks, as the image in Fig. 5b seems to torsional fatigue.
indicate. On the other hand, because of their orienta-
tion with respect to the instrument cross-section,
Acknowledgements
running from the surface to the centre along the
radius, and not perpendicular to it as do the flexural This work was partially supported by Fundação de
fatigue cracks, the longitudinal cracks should not Amparo à Pesquisa do Estado de Minas Gerais –
significantly affect the torsional resistance because they FAPEMIG, Belo Horizonte, MG, Brazil, and Conselho
did not significantly reduce the cross-sectional area. Nacional de Desenvolvimento Cientı́fico e Tecnológico –
The fracture by torsional overload has been dealt CNPq, Brası́lia, DF, Brazil.
with using low-torque endodontic motors, which can
prevent the application of a higher torque than that
References
which each instrument can bear without failing. This
strategy presents two problems: first, the flexural loads, Alapati SB, Brantley WA, Svec TA, Powers JM, Mitchell JC
developed during curved root canal shaping, result in a (2003) Scanning electron microscope observations of new
statistically significant reduction of the instrument’s and used nickel–titanium rotary files. Journal of Endodontics
torsional resistance (Yared et al. 2003a,b, Ullmann & 29, 667–9.
American Society for Testing of Materials ASTM E 1823
Peters 2005, Bahia et al. 2006). Therefore, low-torque
(2005) Relating to Fatigue and Fracture Testing. USA:
motors can only prevent torsional failures when the
American Society for Testing of Materials International.
selected torque value corresponds to that associated
Bahia MGA, Buono VTL (2005) Decrease in fatigue resistance
with the instrument’s stage of usage. The second of nickel–titanium rotary instruments after clinical use in
important point is that low torque motors cannot
curved root canals. Oral Surgery, Oral Medicine, Oral International Organization for Standardization ISO 3630-1
Pathology, Oral Radiology and Endodontology 100, 249–55. (1992) Dental Root-canal Instruments. Part 1: Files, Reamers,
Bahia MGA, Martins RC, Gonzalez BM, Buono VTL (2005) Barbed Broaches, Rasps, Paste Carriers, Explorers and Cotton
Physical and mechanical characterization and the influence Broaches. Switzerland: International Organization for Stan-
of cyclic loading on the behaviour of nickel–titanium wires dardization.
employed in the manufacture of rotary endodontic instru- Martins RC, Bahia MGA, Buono VTL (2006) The effect of
ments. International Endodontic Journal 38, 795–801. sodium hypochlorite on the surface characteristics and
Bahia MGA, Melo MCC, Buono VTL (2006) Influence of fatigue resistance of nickel–titanium endodontic instru-
simulated clinical use on the torsional behavior of nickel– ments. Oral Surgery, Oral Medicine, Oral Pathology, Oral
titanium rotary endodontic instruments. Oral Surgery, Oral Radiology and Endodontology 102, e99–105.
Medicine, Oral Pathology, Oral Radiology and Endodontology Melo MCC, Bahia MGA, Buono VTL (2002) Fatigue resistance
101, 675–80. of engine-driven rotary nickel–titanium endodontic instru-
Barbosa FOG, Ponciano JAC, Araújo MCP (2007) Influence of ments. Journal of Endodontics 28, 765–9.
previous angular deformation on flexural fatigue resistance Melo MCC, Pereira ESJ, Viana ACD, Fonseca AMA, Buono
of K3 nickel–titanium rotary instruments. Journal of Endo- VTL, Bahia MGA (2008) Dimensional characterization and
dontics 33, 1477–80. mechanical behaviour of K3 rotary instruments. Inter-
Berutti E, Chiandussi G, Gaviglio I, Ibba A (2003) Comparative national Endodontic Journal 41, 329–38.
analysis of torsional and bending stresses in two mathe- Otsuka K, Wayman CM (1998) Shape Memory Materials, 1st
matical models of nickel–titanium rotary instruments: edn. Cambridge, UK: Cambridge University Press, pp. 26–
ProTaper versus ProFile. Journal of Endodontics 29, 15–9. 96.
Best S, Watson P, Pilliar R, Kulkarni GGK, Yared G (2004) Parashos P, Messer HH (2006) Rotary NiTi instrument
Torsional fatigue and endurance limit of a size 30.06 fracture and its consequences. Journal of Endodontics 32,
ProFile rotary instrument. International Endodontic Journal 1031–43.
37, 370–3. Peng B, Shen Y, Cheung GSP, Xia TJ (2005) Defects in
Cheung GSP, Darvell BW (2007) Fatigue testing of a NiTi ProTaper S1 instruments after clinical use: longitudinal
rotary instrument. Part 1: Strain-life relationship. Interna- examination. International Endodontic Journal 38, 550–7.
tional Endodontic Journal 40, 626–32. Peters OA, Barbakow F (2002) Dynamic torque and apical
Eggeler G, Hornbogen E, Yawny A, Heckmann A, Wagner M forces of ProFile.04 rotary instruments during preparation
(2004) Structural and functional fatigue of NiTi shape of curved canals. International Endodontic Journal 35, 379–
memory alloys. Material Science and Engineering A 378, 89.
24–33. Peters OA, Peters CI, Schönenberger K, Barbakow F (2003)
Fife D, Gambarini G, Britto LR (2004) Cyclic fatigue testing of ProTaper rotary root canal preparation: assessment of
ProTaper NiTi rotary instruments after clinical use. Oral torque and force in relation to canal anatomy. International
Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontic Journal 36, 93–9.
Endodontics 97, 251–6. Plotino G, Grande NM, Sorci E, Malagnino VA, Somma F
Foschi F, Nucci C, Montebugnoli L et al. (2004) SEM evalu- (2006) A comparation of cyclic fatigue between used and
ation of canal wall dentine following use of Mtwo and new Mtwo Ni–Ti rotary instruments. International Endo-
ProTaper NiTi rotary instruments. International Endodontic dontic Journal 39, 716–23.
Journal 37, 832–9. Predki W, Klönne M, Knopik A (2006) Cyclic torsional loading
Gambarini G (2000) Rationale for the use of low-torque of pseudoelastic NiTi shape memory alloys: damping and
endodontic motors in root canal instrumentation. Endo- fatigue failure. Materials Science and Enginnering A 417,
dontics Dental Traumatology 16, 95–100. 182–9.
Gambarini G (2001) Cyclic fatigue of ProFile rotary instru- Pruett JP, Clement DJ, Carnes DL (1997) Cyclic fatigue testing
ments after prolonged clinical use. International Endodontic of nickel–titanium endodontic instruments. Journal of Endo-
Journal 34, 386–9. dontics 23, 77–85.
Grande NM, Plotino G, Pecci R, Bedini R, Malagnino VA, Sattapan B, Palamara JE, Messer HH (2000) Torque during
Somma F (2006) Cyclic fatigue resistance and three- canal instrumentation using rotary nickel–titanium files.
dimensional analysis of instruments from two nickel– Journal of Endodontics 26, 156–60.
titanium rotary systems. International Endodontic Journal Schäfer E, Florek H (2003) Efficiency of rotary nickel–titanium
39, 755–63. K3 instruments compared with stainless steel hand
Haı̈kel Y, Serfaty R, Bateman G, Senger B, Allemann C (1999) K-Flexofile. Part 1. Shaping ability in simulated curved
Dynamic and cyclic fatigue of engine-driven rotary nickel– canals. International Endodontic Journal 36, 199–207.
titanium endodontic instruments. Journal of Endodontics 25, Schijve J (2001) Fatigue of Structures and Materials, 1st edn.
434–40. Dordrecht, the Netherlands: Kluwer Academic Publishers.
Svec TA, Powers JM (1999) Effects of simulated clinical Wolcott J, Himel V (1997) Torsional properties of nickel–
conditions on nickel–titanium rotery files. Journal of Endo- titanium versus stainless steel endodontic files. Journal of
dontics 25, 759–60. Endodontics 23, 217–20.
Tripi TR, Bonaccorso A, Condorelli GG (2006) Cyclic fatigue of Yared GM, Bou Dagher FE, Machtou P (2000) Cyclic fatigue of
different nickel–titanium endodontic rotary instruments. Profile rotary instruments after clinical use. International
Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology Endodontic Journal 33, 204–7.
and Endodontics 102, e106–14. Yared G, Kulkarni GK, Ghossayn F (2003a) An in vitro study
Ullmann CJ, Peters OA (2005) Effect of cyclic fatigue on static of the torsional properties of new and used K3 instruments.
fracture loads in ProTaper nickel–titanium rotary instru- International Endodontic Journal 36, 764–9.
ments. Journal of Endodontics 31, 183–6. Yared G, Kulkarni GK, Ghossayn F (2003b) Torsional prop-
Vieira EP, França EC, Martins RC, Buono VTL, Bahia MGA erties of new and used rotary K3 NiTi files. Australian
(2008) Influence of multiple clinical use on fatigue resis- Endodontic Journal 29, 75–8.
tance of ProTaper rotary nickel–titanium instruments.
International Endodontic Journal 41, 163–72.
doi:10.1111/j.1365-2591.2008.01450.x
Effect of canal preparation and residual root filling

material on root impedance
A. Al-bulushi, M. Levinkind, M. Flanagan, Y.-L. Ng & K Gulabivala

UCL Eastman Dental Institute, University College London, London, UK
Abstract Nyquist plots and comparisons made within each

tooth, between measurement points along the length
Al-bulushi A, Levinkind M, Flanagan M, Ng Y-L, Gula-
of the canal, as well as under the different canal
bivala K. Effect of canal preparation and residual root filling
conditions. Equivalent circuits were modelled for
material on root impedance. International Endodontic Journal,
different test conditions.
41, 892–904, 2008.
Results The impedance decreased from the coronal
Aim To investigate the effect of root canal preparation to the apical levels in all canal conditions in a
and residual root filling material on the impedance characteristic way, with an exaggerated drop at the
characteristics of extracted human roots. apical terminus. Impedance decreased after chemo-
Methodology Thirty extracted, human single- mechanical preparation, but gave higher values com-
rooted teeth were mounted in a custom-made appa- pared with before or after instrumentation, once canal
ratus that allowed strict temperature control. Imped- filling had been removed. Equivalent circuits remained
ance measurements of the roots were made with a file consistent at the tested positions within the canal,
acting as the internal electrode, using a frequency regardless of canal condition, but the circuit compo-
response analyser. The measurements were made nent values changed with the impedance.
under three canal conditions: (i) before chemo- Conclusions Impedance was influenced by corono-
mechanical preparation; (ii) after chemo-mechanical apical position, chemo-mechanical preparation and
preparation; (iii) after root filling removal to residual root canal filling material.
re-establish patency (following placement of root
Keywords: electronic apex locator, impedance, root
filling). The measurements were taken at 0, 0.5, 1,
apex, root filling.
2, 3, 4, 5, 6, 7, 8, 9 and 10 mm coronal to the apical
terminus and also at 0.5 and 1 mm past the apical Received 19 September 2006; accepted 21 May 2008
terminus. Impedance values were viewed using
Ng et al. 2008) but the effect of length of canal

Introduction
preparation per se has not been determined. Good
It is well established that successful root canal treat- length control demands accurate knowledge of location
ment is strongly influenced by length control dur- of apical terminus of the root canal system, as well as
ing root canal treatment (Ng et al. 2007, Ng et al. refined instrument manipulation skills. Traditionally,
2008). Both over-filling and under-filling decrease several methods have been used to determine the
the success rate of root canal treatment (Smith et al. location of canal terminus, the most common being the
1993, Basmadjian-Charles et al. 2002, Ng et al. 2007, radiographic approach. Compression of three-dimen-
sional data to produce a two-dimensional radiographic
image brings inherent inaccuracies. The apical canal
Correspondence: Prof. K Gulabivala, Unit of Endodontology, terminus does not always coincide with the anatomical
UCL Eastman Dental Institute, University College London, 256
Grays Inn Road, London WC1X 8LD, UK (Tel.: 020 7915
apex of the root and frequently is not apparent on
1033; fax: 020 7915 2371; e-mail: k.gulabivala@ radiographs (Green 1960, Levy & Glatt 1970, Burch &
eastman.ucl.ac.uk). Hulen 1972, Pineda & Kuttler 1972, Dummer et al.
Al-bulushi et al. Effect of canal preparation and residual root filling material on root impedance
1984). The electronic apex locator was a significant icantly different from such determination during root
advancement because it was better able to deter- canal retreatment involving the removal of gutta-
mine the apical terminus of the root canal and percha root filling material. The aim of this study was
potentially reduced the need for irradiation to the to examine the effect of: (i) chemo-mechanical canal
patient (Nekoofar et al. 2006). preparation (during the primary phase); and (ii) resid-
Electronic apex locators (EALs), from their earliest ual root canal filling material (during the retreatment
designs have proved sensitive to root canal contents, phase), on root impedance measured over a wide range
particularly the electrolyte composition (Suchde & of frequencies using a frequency response analyser.
Talim 1977, Arora & Gulabivala 1995). The currently
available multi-frequency impedance-type apex loca-
tors operate accurately despite the presence of intra-
canal electrolytes such as irrigating solutions, pus,
Sample
blood or serum exudate (Fouad et al. 1990, Mayeda
et al. 1993, Arora & Gulabivala 1995, Pilot & Pitts A total of 30 freshly extracted human teeth were used
1997, Dunlap et al. 1998, Pagavino et al. 1998, for the study following acquisition of informed consent.
Jenkins et al. 2001, Meares & Steiman 2002, Welk Tooth selection was guided by the following criteria:
et al. 2003, Hoer & Attin 2004). EALs rely on single-rooted intact teeth, with absence of caries, cracks
impedance change along the length of the root to or tooth surface loss. The teeth were accessed to achieve
determine the position of the apical canal terminus canal patency with a size 10 Flexofile (Dentsply,
(Meredith & Gulabivala 1997, Pilot & Pitts 1997, Weybridge, UK) and then stored in 0.1 mmol potas-
Rambo et al. 2007). Amongst the various factors sium chloride solution for 12 weeks for equilibration.
influencing tooth impedance (Levinkind et al. 1990,
Meredith & Gulabivala 1997, Gordon & Chandler
Experimental setup
2004), are the size of the dentinal tubules (Levinkind
et al. 1992) and the smear layer (Levinkind et al. 1992, Each tooth was individually suspended in a tempering
Eldarrat et al. 2004). There is little published data on beaker (Cole-Parmer, Hanwell, UK), with an inlet and
the effect of canal preparation or residual root canal an outlet to facilitate circulation of temperature-
filling material when performing root canal re-treat- controlled water (25 C) via plastic tubing. The tem-
ment, on the electrical characteristics of the root dental perature of the solution was maintained by way of an
tissue. One ex vivo study assessed the accuracy of the tri immersion electric heater (Galenkamp thermo-stirrer,
auto ZX (J Morita Corporation, Osaka, Japan) during re- Leicester, UK) in a fixed volume water tank, from which
treatment (Alves et al. 2005) and another laboratory a water pump (Maxi jet; Aquatic design, London, UK)
study evaluated conductivity pathways (Levinkind propelled the heated water through plastic tubing. The
et al. 1994). The latter study examined the effect of room and the circulating water temperatures were
external root surface insulation on its electrical char- monitored using a digital thermometer (ETI, Worthing,
acteristics. Repeated electrical measurements following UK) accurate to ±1 C, whilst the temperature of the
the application of a beeswax coating over the entire KCl solution in the beaker was monitored by a high
root surface (except for the apical 1 mm) showed a accuracy (±0.1 C) glass mercury thermometer (Bran-
gradual drop in impedance as the instrument was nan, Cleater Moor, UK). The teeth were suspended at
inserted deeper into the canal, followed by a further their cemento–enamel junction in the KCl solution in
dramatic decrease as it approached and passed the the tempering beaker using custom-made apparatus
canal terminus. This inferred a significant lateral that allowed accurate positioning of the teeth (Fig. 1).
conductivity pathway through the root dentine but To further minimize temperature fluctuation, the tem-
that the apical pathway became more dominant with pering beaker was placed in a thick polystyrene box
the proximity of the instrument to the apex. The data (Promega, Southampton, UK) with a custom window to
indicated the potential for a complex interaction enable tooth visualization.
between the two conductivity pathways in contributing
to the electrical characteristics of the root, a supposition
Impedance measurement
confirmed by Križaj et al. (2004). It is likely that
electronic canal length measurement during primary The impedance measurements were taken using a
root canal preparation and treatment would be signif- frequency response analyser controlled by dedicated
Effect of canal preparation and residual root filling material on root impedance Al-bulushi et al.
Figure 1 Tempering beaker with tooth positioning apparatus (photo and schematic).
computer software (Voltech TF2000; Voltech instru- attached to a micrometer screw gauge (L.S. Starrett Co,
ments Ltd, Didcot, UK) and supplied by a stabilized Athol, MA, USA) and rotating a dial on the micrometer
electric power supply (Farnell Electronics Ltd, Leeds, (accurate to 0.04 mm) controlled the position of the file
UK) to eliminate potential fluctuation in mains electri- tip relative to the apical terminus (Fig. 1). The circuit
cal supply. The current flow through the KCl solution layout is shown in Fig. 2.
was achieved via a clip-type electrode attached to a No.
10 K-file (Flexofile; Dentsply) inserted into the root
Stages of impedance measurement
canal, with a counter electrode consisting of a gold-
coated perspex disk at the bottom of the beaker. Radiographic and photographic views were obtained
Impedance measurements were taken at 0, 0.5, 1, 2, from two angles (buccal and proximal) for all the teeth
3, 4, 5, 6, 7, 8, 9 and 10 mm coronal to the apical at the following stages:
terminus and also at 0.5 and 1 mm past the apical • Preoperatively (status A);
terminus. The position of the apical terminus was • Post-chemo-mechanical preparation (status B);
established by inserting a file into the root canal until • Post-root filling-material removal and re-establish-
the tip was just visible through the apical foramen, it ment of patency (status C).
was then withdrawn until it just disappeared into the
canal, this was taken to be the ‘0’ position and used as Preoperative measurements
the reference point. The file length at the reference Measurements were made on un-instrumented teeth as
position was measured and recorded. The file was described above.
were then stored in distilled water for 10 days at room

temperature with changes of the water, twice daily.
The gutta-percha was removed from the canals using
a combination of heat (System B) and chloroform
(Chloroform BP; JM Lovridge Ltd, Southampton, UK)
until apical patency was re-established (Fig. 3). Again,
the teeth were stored in distilled water as before,
followed by storage in 0.1 mmol KCl solution for
5 days. The impedance measurements were then
repeated as before.
Data analysis
Figure 2 Experiment circuit layout. The data were collected through a modified computer
program (Voltech Instruments Ltd) in raw DOS format;
Post-chemo-mechanical-preparation measurements they were converted for use in a Windows-based
The teeth were instrumented using rotary GT nickel– software package (Zview; Scribner & Associates,
titanium instruments (Dentsply) to an apical size 30 Southern Pines, NC, USA) by custom-written Java
and 0.8 taper. A solution of 2.5% sodium hypochlorite programs (Flanagan 2002; Flanagan M, Personal
solution (Teepol bleach; Teepol products, Egham, UK) communication). The impedance values were plotted
was used for irrigation; the concentration was verified to compare different canal conditions and the results
by titration (British Pharmacopoeia 1973). The canals presented in Nyquist plots (see Appendix). Equivalent
were also irrigated with nonproprietary 17% EDTA circuit modelling was performed using custom-written
solution, which was prepared and used on the same programs (Flanagan 2002: Flanagan M, Personal
day. Following this, the teeth were stored in distilled communication).
water for 5 days with frequent changes of the water
twice daily, after which they were stored in 0.1 mmol
Results
of KCl solution for 1 week. The impedance measure-
ments were repeated as described above. The results were presented and analysed through:
• Analysis of the Nyquist plots;
Post-root filling-removal measurements • Equivalent circuit modelling.
The root canals were obturated with gutta-percha
(AutoFit; SybronEndo, Orange, CA, USA) and Roth
Analysis of Nyquist plots
801 sealer (Roth International Ltd, Englewood, NJ,
USA), using the Continuous-Wave technique with a The impedance measurements were assessed from two
System B heat source (SybronEndo) and Obtura aspects:
(Spartan, Fenton, MO, USA). The sealer was mixed 1. The change in impedance as a function of the
according to manufacturer’s recommendations on a coronal–apical position of the file within the canal under
glass slab at room temperature and was applied to the different canal conditions (representative examples of
canal using the master gutta-percha point. The teeth changes within one tooth are shown in Figs 4–6);
Figure 3 Example of a tooth used in the

study showing uninstrumented canals
(a–d), obturated canals (e, f) and
canals after removal of root filling and
re-establishment of patency (g, h).
Figure 4 Change in impedance as a function of file position in

an uninstrumented canal (Z¢ and Z¢¢ = real and imaginary
a canal after removal of root filling and re-establishment of
impedance in ohms – see Appendix for explanation).
patency (Z¢ and Z¢¢ = real and imaginary impedance in ohms –
see Appendix for explanation).

a canal after instrumentation (Z¢ and Z¢¢ = real and imaginary
Figure 7 Change in impedance at the apical terminus as a
impedance in ohms – see Appendix for explanation).
function of canal status. (a) uninstrumented canal, (b) after
2. The change in impedance as a function of the canal chemo-mechanical instrumentation, (c) after root filling
removal and re-establishment of patency (Z¢ and Z¢¢ = real
condition, at the apex and at 1 mm and at 4 mm
and imaginary impedance in ohms – see Appendix for
coronal to the canal terminus. (representative examples
explanation).
of changes related to canal condition within one tooth
are shown in Figs 7–9).
The Nyquist plots for all teeth showed that the appeared translucent and had a yellow to brown colour
impedance decreased from the coronal to the apical and lower in roots which appeared to be lighter in
levels in a gradual and consistent pattern and showed a colour), presence of wide canals with thin root walls
dramatic decrease at the apical terminus in all canal and canal division.
conditions (Figs 4–6). There were some variations in a As a rule, canal instrumentation decreased the
small number of teeth, which were apparently related impedance, whilst the measurements following the
to the colour of the tooth (being higher for roots which removal of root filling material increased the impedance
Equivalent circuit modelling

Equivalent circuit modelling is a common practice
(Levinkind et al. 1992, Meredith & Gulabivala 1997)
designed to assess the electrical characteristics of a
circuit (in this case the root dentine with its surround-
ing tissues), whereby the circuit characteristics are
expressed in electrical components and their relative
arrangement. These tests, performed using custom-
made software, revealed the equivalent circuit to be
unique for each root.
Data analysis was performed to evaluate the effect of
the electrode position and the canal content on the
equivalent circuit. Thirty-two proposed circuits (which
closely matched the circuit layout) were analysed for fit
to the data obtained from the impedance measure-
Figure 8 Change in impedance at 1 mm coronal to the apical
ments, and out of these only five circuits closely
terminus as a function of canal status. (A) uninstrumented modelled the teeth used in this study.
canal, (B) after chemo-mechanical instrumentation, (C) after
root filling removal and re-establishment of patency (Z¢ and
Effect of electrode position in the canal on equivalent
Z¢¢ = real and imaginary impedance in ohms – see Appendix
circuit
for explanation).
The equivalent circuit was the same at every point of
measurement in any given tooth. Thus, the electrode
position had no effect in changing the equivalent
circuit, the only change was in the circuit component
values; this is illustrated in Figs 10–12 with the
equivalent circuit shown in Fig. 13.
Effect of canal contents on equivalent circuit

The equivalent circuit at any particular position in any
given canal remained the same in spite of changing the
canal status, only the circuit component values
changed as shown in the example in Figs 14 and 15
with its equivalent circuit shown in Fig. 16.
Discussion
Figure 9 Change in impedance at 4 mm coronal to the apical

The final methodology was refined through initial pilot
terminus as a function of canal status. (A) uninstrumented studies and revealed the need to control accuracy of: (i)
canal, (B) after chemo-mechanical instrumentation, (C) after suspension of the teeth in solution to ensure repeatable
root filling removal and re-establishment of patency (Z¢ and positioning through the measurement stages; and (ii)
Z¢¢ = real and imaginary impedance in ohms – see Appendix control of KCl solution temperature to ensure reliable
for explanation). impedance measurements. To address these require-
ments, custom-made apparatus was designed and
refined to facilitate accurate and repeatable tooth
(Figs 7–9). The impedance changes were more positioning. Temperature fluctuation significantly
pronounced within the last 0.5 mm of the apex, affects impedance measurements; higher temperatures
particularly in status C (after root filling removal and decrease impedance and vice versa. Considerable time
re-establishment of patency). was therefore devoted to maintain the solution and
Figure 10 Equivalent circuit-impedance

fitting graph at the apical terminus in
uninstrumented canal (imaginary versus
real impedance values). The circles rep-
resent the impedance of the tooth, whilst
the solid line represents the impedance of
the equivalent circuit.

fitting graph at 1 mm coronal to the
apical terminus in uninstrumented
canal. The circles represent the imped-
ance of the tooth, whilst the solid line
represents the impedance of the
equivalent circuit.

fitting graph at 10 mm coronal to the
apical terminus in uninstrumented
canal. The circles represent the
impedance of the tooth, whilst the solid
line represents the impedance of the
equivalent circuit.
ambient room temperature at 25 C; as it was easier to would require a far more elaborate and cost-prohibitive
control. Measurements were not recorded if the tem- experimental set-up.
perature was unsteady. A temperature of 37 C would A solution of 0.1 mmol KCl solution was used for
have been desirable to mimic the mouth conditions but storage, equilibration and measurements because the
treating this experimental set-up, represented as a

single impedance (ZS), and a 40-kW series resistance
(impedance ZREF) as a voltage divider (see Appendix for
basic definitions). On applying a sinusoidal voltage (VO)
to this series combination and measuring the voltage
appearing across the set up (VS), the sample impedance
ZS was obtained using the relationship:
Figure 13 Equivalent circuit for the tooth evaluated in

ZS ¼ V S ZREF =ðV S þ V O Þ
Figs 10–12. As the voltage, VS, is phase-shifted with respect to
the excitation voltage, both its phase and magnitude
small K+ size at this low concentration allows their were recorded. These were used to calculate the
freer motion within dentinal tubules [compared with observed voltage as a complex number to obtain the
Na+ (Križaj et al. 2004)], giving a better flow of current resultant impedance, ZS in the complex form. This
(Levinkind et al. 1992). A long storage time in KCl was facilitated the following regression procedure as its real
used to ensure good dentine hydration and equilibra- and imaginary parts were of comparable magnitude.
tion. The size 10 file was used for the measurements The best estimates of the component parameters of a
throughout the study to control this variable. range of possible equivalent circuits were obtained by
The complex impedance values of the probe (size 10 fitting the variation of ZS with the frequency of the
file) and tooth experimental set-up were determined by applied excitation to these equivalent circuits using a

fitting graph at the apical terminus of
uninstrumented canal. The circles rep-
resent the impedance of the tooth, whilst
the solid line represents the impedance of
the equivalent circuit.

fitting graph at the apical terminus after
removal of root filling and re-establish-
ment of patency. The circles represent
the impedance of the tooth, whilst the
solid line represents the impedance of the
equivalent circuit.
and p2 (p2 > p1), the extra sum of squares rule was
used to obtain an F-ratio,
ðSS1 SS2 Þðn p2 Þ
F¼
SS2 ðp2 p1 Þ
that then enabled the calculation of the probability that

Figure 16 Equivalent circuit for the tooth evaluated in Figs 14 the two models fell within the sampling variability of
and 15. each other, i.e. whether the more complex model could
be validly chosen (Bevington & Robinson 2003).
nonlinear regression procedure based on the Nelder & Using this method, impedance was found to decrease
Mead (1965) simplex method as implemented by in a corono-apical direction and reached its lowest
Flanagan (2002). The real and imaginary parts were value at the apical terminus; this applied to all canal
both included within the same regression procedure, states. This is probably related to greater dentine
their comparable magnitudes leading to a simpler and thickness and longer dentinal tubules more coronally.
more numerically stable regression procedure than The impedance characteristics varied between the
would be obtained using the phases and magnitudes teeth, which can be correlated with the size of the
directly. The components (resistances and capaci- canal and thickness of dentine (Križaj et al. 2004).
tances) of the equivalent circuits were restricted to Dentine sclerosis (as judged from the colour and
those known to occur in biological systems, such as translucency of the external root surface) seemed to
those occurring at solid/liquid interfaces and cell- increase the impedance (Križaj et al. 2004), probably as
membranes, and constant phase elements characteris- a result of occluded dentinal tubules, with reduced flow
tic of diffusion-limited paths such as occurs in porous of electrolyte and hence current (Tagami et al. 1992).
materials and in heterogeneous electrode surfaces The results showed that impedance decreased after
(McDonald 1987). Estimates of the standard devia- chemo-mechanical preparation of the root canal in
tion of each circuit parameter were obtained by the most roots, except those that exhibited very low
conventional procedure of assuming that the regression impedance, where there was an increase in impedance,
surface about the minimum, at which the parameter possibly because of unremoved smear layer (Levinkind
estimates were taken, approximates to a quadratic et al. 1992, Eldarrat et al. 2004). A representative
surface, i.e. the linear approximation. These were used series of graphs were presented to depict the results
to eliminate clearly over-parameterized models, i.e. because there is no simple means of producing
ones with one or more parameters in which their summary statistics from such complex data; as may
standard deviations were at least twice the value of the be the convention with simple numerical data. An
component parameter, e.g. resistance. This commonly intuitive approach to synthesis of the collective data
reduced the number of equivalent circuits that might from the 30 teeth is called for and is given descriptively
validly represent the experimental arrangement to a above. The described decrease in impedance is most
small number. If these all contained the same number likely related to a reduction in dentine thickness and
of components they were ranked according to the therefore length of dentinal tubules. It has been
reduced sum of squares of the regression residuals suggested that such decrease in impedance may reduce
the accuracy of the final length determination using
X
n
ðZS;real;obs;i ZS;real;model;i Þ2
SS ¼ EALs because the change in impedance as a function of
i¼1
ðn pÞ distance from apical terminus would be smaller and
X
n
ðZS;imag;obs;i ZS;imag;model;i Þ2 more difficult to interpret. The use of EALs utilizing the
þ
ðn pÞ ratio method (Nekoofar et al. 2006) may partly over-
i¼1
come this problem (Križaj et al. 2004).
where the subscript real indicates the real part of the In contrast, the impedance increased significantly
complex impedance, ZS; imag, the imaginary part of the along the length of the canal after removal of the root
impedance; ZS, measured at n frequencies; obs, the filling and re-establishment of patency, probably related
observed value; model, the theoretical value calculated to dentinal tubule occlusion by root filling material and
for the model equivalent circuit with p component consequently reduced canal conductivity. This is
parameters. In cases where there were two potential consistent and in agreement with previous studies
models with differing numbers of circuit parameters, p1 (Pilot & Pitts 1997, Alves et al. 2005). From the above
discussion, it may be expected that such impedance in cases of root canal retreatment. The blockage of
change may potentially improve the accuracy of conduction pathways formed by the dentinal tubules,
location of the apical terminus. Alves et al. (2005), as well as lateral and apical canal exits, may
however, found that the residual gutta-percha pre- profoundly influence apex locator function. The
vented signal transmission, until the file was beyond the behaviour of apex locators in retreatment cases where
filling material. As the filling material was sometimes root filling material removal leaves residues on the
extruded in their study, the lengths during retreatment canal wall may therefore be different. The difference in
were longer than those determinations prior to root behaviour compared with a previously untreated tooth
filling. Such was not a finding in this study as the may be manifested as a more abrupt but well-defined
material remained confined to the root canal system. ‘count-down’ to the ‘zero reading’, assuming that the
The impedance in all three canal states (unprepared, apical canal exits are patent. If not, the apex locator
prepared and post-filling removal) decreased as the tip may show a ‘count-down’ towards the ‘zero reading’
of the measuring electrode (No. 10 file) approached the as the file reaches the apex, without actually reaching
apical terminus, consistent with other studies (Mere- this value. This latter behaviour may be indicative of
dith & Gulabivala 1997, Pilot & Pitts 1997, Križaj et al. patent apical exits that could not be negotiated by the
2004, Rambo et al. 2007). file as a result of mechanical obstruction in its
Initial equivalent circuit modelling indicated that the pathway. This leads to the clinical idea of having
advancement of file position in the canal was not electrical (and therefore actual canal patency) but not
associated with a change in the composition of the mechanical patency. An absence of any EAL reading
equivalent circuit, but it was associated with a change would denote an absence of a circuit; this could be
in the values of the electrical components of that circuit, interpreted as an absence of actual canal patency.
as was the case with the canal status. This would Under these circumstances, the chances of reaching
indicate that the impedance measurements at each any apical infection through irrigation may be com-
position in the canal reflected the geometry and promised. A better understanding of the physics
composition of the root dentine in the vicinity of the underpinning the function of apex locators and each
instrument electrode in the canal. Furthermore, it is commercially available model’s limitations may en-
unlikely that the root filling material would have hance the practitioners’ handling and interpretation of
induced a change in the structure of the root and apex locator readings.
probably merely acted by occlusion of the dentinal
tubules to act as an insulating layer. These speculative
Conclusions
observations seem intuitively obvious but it is not
always easy to explain the impedance behaviour based Measured root impedance decreased with apical posi-
on physical structures as there may not always be tion of the file in the canal, as well as with chemo-
obvious physical counterparts (Levinkind et al. 1990). mechanical preparation; it increased with residual root
The difference in the structure of the equivalent circuits canal filling material in the canal system. The findings
between the roots was minimal and the characteristics will help to better explain the behaviour of commercial
of all the roots were explained by one of five circuits in apex locators before, during and after canal prepara-
the sample used in this study. This may explain the tion, as well as during root canal retreatment.
relatively high accuracy of the apex locators but also
offers an explanation for potential variability in accu-
Acknowledgements
racy (Gordon & Chandler 2004, Nekoofar et al. 2006)
because the calibration characteristics of commercial The authors are grateful to Dr Michael Flanagan of the
EALs are likely to be pre-selected. Department of Electronic and Electrical Engineering for
Another way of analysing these impedance data was his help with the data recording and analysis.
proposed by Levinkind (1994). This was based on using
a neural network computing technique to evaluate the
References
changes in the complex impedance data sets. It is likely
that this may be one of the techniques that may be Alves AM, Felippe MC, Felippe WT, Rocha MJ (2005) Ex vivo
being used by the current EALs. evaluation of the capacity of the Tri Auto ZX to locate the
The findings of this study may explain the function apical foramen during root canal retreatment. International
of, as well as inform the use of electronic apex locators Endodontic Journal 38, 718–24.
Arora RK, Gulabivala K (1995) An in vivo evaluation of the Levinkind M, Vandernoot TJ, Elliott JC (1992) Evaluation of
ENDEX and RCM Mark II electronic apex locators in root smear layers on serial sections of human dentin by means of
canals with different contents. Oral Surgery, Oral Medicine, electrochemical impedance measurements. Journal of Dental
Oral Pathology, Oral Radiology and Endodontics 79, 497–503. Research 71, 426–33.
Basmadjian-Charles CL, Farge P, Bourgeois DM, Lebrun T Levinkind M, Wadge R, Ahlberg K (1994) Evaluation of
(2002) Factors influencing the long-term results of end- conductivity pathways that affect electrical determination of
odontic treatment: a review of the literature. International root canal working length. International Endodontic Journal
Dental Journal 52, 81–6. 27, 102.
Bevington PR, Robinson DK (2003) Data Reduction and Error Levy AB, Glatt L (1970) Deviation of the apical foramen from
Analysis for the Physical Sciences. 3rd edn. New York, NY, the radiographic apex. Journal of New Jersey State Dental
USA: McGraw Hill. Society 41, 12–3.
British Pharmacopoeia (1973) Appendix IA. In: HMSO for Mayeda DL, Simon JHS, Aimar DF, Finley K (1993) In vivo
Department of Health and Social Security, British Pharmaco- measurement accuracy in vital and necrotic canals with the
poeia. 12th edn. London, UK: HMSO. Endex apex locator. Journal of Endodontics 19, 545–8.
Burch JG, Hulen S (1972) The relationship of the apical McDonald JR (1987) Impedance Spectroscopy. New York, NY,
foramen to the anatomic apex of the tooth root. Oral USA: John Wiley.
Surgery, Oral Medicine and Oral Pathology 34, 262–8. Meares WA, Steiman HR (2002) The influence of sodium
Dummer PM, McGinn JH, Rees DG (1984) The position and hypochlorite irrigation on the accuracy of the Root ZX
topography of the apical canal constriction and apical electronic apex locator. Journal of Endodontics 28, 595–8.
foramen. International Endodontic Journal 17, 192–8. Meredith N, Gulabivala K (1997) Electrical impedance mea-
Dunlap CA, Remeikis NA, BeGole EA, Rauschenberger CR surements of root canal length. Endodontics and Dental
(1998) An in vivo evaluation of an electronic apex locator Traumatology 13, 126–31.
that uses the ratio method in vital and necrotic canals. Nekoofar MH, Ghandhi MM, Hayes SJ, Dummer PMH (2006)
Journal of Endodontics 24, 48–50. The fundamental operating principles of electronic root
Eldarrat AH, High AS, Kale GM (2004) In vitro analysis of canal length measurement devices. International Endodontic
‘smear layer’ on human dentine using AC-impedance Journal 39, 595–609.
spectroscopy. Journal of Dentistry 32, 547–54. Nelder JA, Mead R (1965) A simplex method for function
Flanagan MT (2002) Java On-line Scientific Library (Available minimization. Comput. J. 7, 308–13.
at: http://www.ee.ucl.ac.uk/~mflanaga/java/Regression. Ng Y-L, Mann V, Rahbaran S, Lewsey J, Gulabivala K (2007)
html). Outcome of primary root canal treatment: systematic review
Fouad AF, Krell KV, McKendry DJ, Koorbusch GF, Olson RA of the literature – Part 1 Effects of study characteristics on
(1990) Clinical evaluation of five electronic root canal length probability of success. International Endodontic Journal 40,
measuring instruments. Journal of Endodontics 16, 446–9. 921–39.
Gordon MP, Chandler NP (2004) Electronic apex locators. Ng Y-L, Mann V, Rahbaran S, Lewsey J, Gulabivala K (2007)
International Endodontic Journal 37, 425–37. Outcome of primary root canal treatment: systematic review
Green D (1960) Stereomicroscopic study of 700 root apices of of the literature – Part 2 Influence of clinical factors.
maxillary and mandibular posterior teeth. Oral Surgery, Oral International Endodontic Journal 41, 6–31 (doi: 10.1111/
Medicine and Oral Pathology 13, 728–33. j.1365-2591.2007.01323.x).
Hoer D, Attin T (2004) The accuracy of electronic working Pagavino G, Pace R, Baccetti T (1998) A SEM study of in vivo
length determination. International Endodontic Journal 37, accuracy of the Root ZX electronic apex locator. Journal of
125–31. Endodontics 24, 438–41.
Jenkins JA, Walker WA III, Schindler WG, Flores CM (2001) Pilot TF, Pitts DL (1997) Determination of impedance changes
An in vitro evaluation of the accuracy of the root ZX in the at varying frequencies in relation to root canal file position
presence of various irrigants. Journal of Endodontics 27, 209– and irrigant. Journal of Endodontics 23, 719–24.
11. Pineda F, Kuttler Y (1972) Mesiodistal and buccolingual
Križaj D, Jan J, Valenčič V (2004) Modelling AC current roentgenographic investigation of 7275 root canals. Oral
conduction through a human tooth. Bioelectromagnetics 25, Surgery, Oral Medicine and Oral Pathology 33, 101–10.
185–95. Rambo MVH, Gamba HR, Ratzke AS et al. (2007) In vivo
Levinkind M (1994) A dental application of neural network determination of the frequency response of the tooth root
computing: classification of complex electrical impedance canal impedance versus distance from the apical foramen.
measurements to aid root canal treatment. Neural Computing Conference proceedings: Annual International Conference of the
and Applications 2, 209–15. IEEE Engineering in Medicine and Biology Society IEEE
Levinkind M, Vandernoot TJ, Elliott JC (1990) Electrochemical Engineering in Medicine and Biology Society 2007, 570–3.
impedance characterisation of human and bovine enamel. Smith CS, Setchell DJ, Harty FJ (1993) Factors influencing the
Journal of Dental Research 69, 1806–11. success of conventional root canal therapy – a five-year
retrospective study. International Endodontic Journal 26, imaginary unit (instead of i, to avoid confusion with
321–33. current) and x is the angular frequency of the signal.
Suchde RV, Talim ST (1977) Electronic ohmmeter. An The impedances of components then combine accord-
electronic device for the determination of the root canal ing to the same formulas used for resistances, using
length. Oral Surgery, Oral Medicine and Oral Pathology 43,
complex number arithmetic.
141–50.
The imaginary component of an impedance is called
Tagami J, Hosoda H, Burrow MF, Nakajima M (1992) Effect of
aging and caries on dentin permeability. Proceedings of
reactance, X. In general,
Finnish Dental Society 88, 149–54. Z ¼ R þ jX
Welk AR, Baumgartner JC, Marshall JG (2003) An in vivo
Note that the reactance depends on the frequency f of
comparison of two frequency-based electronic apex locators.
Journal of Endodontics 29, 497–500.
the applied voltage: the higher the frequency, the
smaller the capacitive reactance XC and the larger the
inductive reactance XL.
Appendix
Source: http://www.consultrsr.com Fixed frequency signals
If the applied voltage is periodically changing with a
fixed frequency f, according to a sine curve, it is
Impedance spectroscopy – theory, experiment, and represented as the real part of a function of the form:
applications, 2nd edition edited by Barsoukov &
uðtÞ ¼ ue2pjft
Macdonald
where u is a complex number that encodes the phase and
Impedance
amplitude. If the current is represented in an analogous
Impedance is a measure for the manner and degree by
manner as the real value of a function i(t), then the
which a component resists the flow of electrical current
relation between current and voltage is given by
if a given voltage is applied. It is denoted by the symbol
Z and is measured in ohms. Impedance differs from uðtÞ
Z¼
simple resistance in that it takes into account possible iðtÞ
phase offset. an equation quite similar to Ohm’s law.
Constant signals Variable frequency signals

If the applied signal is constant, such as a DC signal, the If the voltage is not a sine curve of fixed frequency, then
circuit is in a steady state. In this state, capacitors are one first has to perform Fourier analysis to find the
considered ‘open’ and inductors are considered ‘closed’. signal components at the various frequencies. Each one
This makes sense because capacitors in steady-state are is then represented as the real part of a complex
modelled as two plates separated by a gap and inductors function and divided by the impedance at the respective
are a coil of wire (which has negligible resistance); the frequency. Adding the resulting current components
former allows for no current flow, whereas the latter yields a function i(t) whose real part is the current.
does. The impedance is then caused by resistors alone If the internal structure of a component is known, its
and is a real number equal to the component’s resistance impedance can be computed using the same laws that
R. The notion of impedance remains useful in such are used for resistances: the total impedance of
a circuit, to study what happens at the instant when subcomponents connected in series is the sum of the
the constant voltage is switched on or off: generally, subcomponents’ impedances; the reciprocal of the total
inductors cause the change in current to be gradual, impedance of subcomponents connected in parallel is
whilst capacitors can cause large peaks in current. the sum of the reciprocals of the subcomponents’
impedances. These simple rules are the main reason for
Sine function signals using the formalism of complex numbers.
If the applied signal is a sine function, or a sum of many
such functions, then capacitors and inductors can Magnitude
produce a phase shift (time delay) from the source. This Often it is enough to know only the magnitude of the
effect can be modelled by assigning those components impedance:
imaginary impedance values. Inductors have an pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
impedance of jxL and capacitors 1/jxC, where j is the jZj ¼ R2 þ X 2
It is equal to the ratio of RMS voltage (VRMS) to RMS originally used the term ‘imaginary number’ to mean
current (IRMS): what is currently meant by the term ‘complex number’,
VRMS the term today specifically means a complex number
jZj ¼ with real part equal to 0, i.e. a number of the form ib.
IRMS
Note that technically, 0 is considered to be a purely
The word ‘impedance’ is often used for this magni- imaginary number: 0 is the only complex number
tude; however, it is important to realize that to compute which is both real and purely imaginary. The Nyquist
this magnitude, one first computes the complex imped- plots consist of imaginary impedance (Z¢) plotted
ance as explained above and then takes the magnitude against real impedance (Z¢¢).
of the result. There are no simple rules that allow one
to compute |Z| directly. Real number
In mathematics, the real numbers are intuitively
Constant phase element defined as numbers that are in one-to-one correspon-
The constant phase element is a nonintuitive circuit dence with the points on an infinite line – the number
element that was discovered (or invented) whilst line. The term ‘real number’ is a retronym coined in
looking at the response of real-world systems. In some response to ‘imaginary number’. Real numbers may be
systems, the Nyquist plot (also called the Cole–Cole plot rational or irrational; algebraic or transcendental; and
or complex impedance plane plot) was expected to be a positive, negative, or zero. Real numbers measure
semicircle with the centre on the x-axis. However, the continuous quantities. They may in theory be ex-
observed plot was indeed the arc of a circle, but with pressed by decimal fractions that have an infinite
the center some distance below the x-axis. These sequence of digits to the right of the decimal point;
depressed semicircles have been explained by a number these are often mis-represented in the same form as
of phenomena, depending on the nature of the system 324.823211247... (where the three dots express that
being investigated. However, the common thread there would still be more digits to come, no matter how
amongst these explanations is that some property of many more might be added at the end). Measurements
the system is not homogeneous or that there is some in the physical sciences are almost always conceived
distribution (dispersion) of the value of some physical as approximations to real numbers. Writing them as
property of the system. These electrical components are decimal fractions (which are rational numbers that
usually associated with rough or porous surfaces, such could be written as ratios, with an explicit denomina-
as dentine and cementum. tor) is not only more compact, but to some extent
conveys the sense of an underlying real number. The
Imaginary number real numbers are the central object of study in
In mathematics, an imaginary number (or purely real analysis. A real number is said to be computable
imaginary number) is a complex number whose square if there exists an algorithm that yields its digits.
is negative or zero. The term was coined by René Because there are only countable many algorithms,
Descartes in 1637 in his La Géométrie and was meant but an uncountable number of reals, most real
to be derogatory: obviously, such numbers were numbers are not computable. Some constructivists
thought not to exist. Any complex number can be accept the existence of only those reals that are
written as a + ib, where a and b are real numbers and i computable. The set of definable numbers is broader,
is the imaginary unit with the property that but still only countable.
i2 ¼ 1
The number a is the real part of the complex number
and b is the imaginary part. Although Descartes
doi:10.1111/j.1365-2591.2008.01412.x
CASE REPORT
Treatment of inflammatory internal

root resorption with mineral trioxide
aggregate: a case report
M. Jacobovitz & R. K. P. de Lima

Paulista Association of Dental Surgeons, Regional de São Carlos, São Carlos, Brazil
Abstract
Jacobovitz M, de Lima RKP. Treatment of inflammatory internal root resorption with mineral
trioxide aggregate: a case report. International Endodontic Journal, 41, 905–912, 2008.
Aim To report the treatment and follow-up of a maxillary central incisor with internal
resorption managed by root canal treatment, white mineral trioxide aggregate (MTA)
repair and a fibre-glass post for reinforcement.
Summary This study presents a case of extensive internal root resorption affecting tooth
11 (FDI) in a 28-year-old male patient, with a history of trauma in that region. The
substantial loss of tooth structure, including cementum, and a perforation with lateral
periodontal communication were complicating factors. Despite a guarded prognosis but
encouraged by a healthy periodontal condition, treatment based on reconstructing the
tooth with white MTA and a fibre-glass core was carried out. Follow-up radiographs over
20 months demonstrated the maintenance of a functional tooth. The tooth did, however,
discolour after MTA treatment.
Key learning points
• Because of its insidious pathology, internal root resorption can extend to significant
dimensions before being recognized.
• MTA plus the fitting of a fibre-glass post provided adequate, functional rehabilitation of a
compromised tooth for 20 months.
• Despite the favourable biological and mechanical properties of white MTA, considerable
tooth discolouration may occur.
Keywords: fibre-glass post, internal inflammatory root resorption, mineral trioxide

aggregate, root canal.
Received 25 may 2007; accepted 8 February 2008
Introduction
Internal root resorption originating from pulp inflammation is always pathological. Internal
root resorption is established after necrosis of odontoblasts and is associated with chronic
Correspondence: Prof. Dr Marcos Jacobovitz, Rua Isis Fernandes 255 ap. 61 São Carlos, SP CEP
13561-040, Brazil (Tel.: 0055 16 33715496; fax: 005 16 33721314; e-mail: marcosjacobovitz@
yahoo.com.br).
partial pulp inflammation and partial pulpal necrosis (Consolaro 2002). Three-dimensional
CASE REPORT
imaging (Lyroudia et al. 2002) has shown such defects to be circumscribed and oval-
shaped. Internal resorption is usually asymptomatic and discovered by chance on routine
radiographic examinations (Trope & Chivian 1997) or by the clinical sign of a ‘pink spot’ on
the crown (Lyroudia et al. 2002). Tissue loss can be extensive and often unrestorable.
After considering the differential diagnosis, including external root resorption, treatment
must aim at complete removal of the resorptive tissue from the root canal system, in an
attempt to prevent further loss of hard tissue. However, selecting suitable restorative
materials for these cases remain a challenge, especially if tooth loss is extensive;
extraction is the only realistic option in some cases.
Mineral trioxide aggregate (MTA) has satisfactory properties for solving many
endodontic problems, including: biocompatibility (Torabinejad et al. 1995b, Holland et al.
2002), favourable sealing ability (Torabinejad et al. 1993), mechanical strength (Torabinejad
et al. 1995a) and a capacity to promote periradicular tissue healing (Torabinejad & Chivian
1999).
Originally developed as a surgical root-end filling material, MTA has been used
successfully in several clinical applications such as pulp capping (Pitt Ford et al. 1996),
pulpotomy (Salako et al. 2003, Menezes et al. 2004), perforation repair (Main et al. 2004),
treatment of traumatized teeth with immature apices (Bakland 2000, Simon et al. 2007)
and for treatment of root resorptions (White & Bryant 2002, Hsien et al. 2003).
This case report describes the 20-month follow-up of extensive internal inflammatory
root resorption in a maxillary central incisor, where MTA and a fibre-glass post, were
employed to restore function.
Clinical report
A periapical radiograph of an asymptomatic 28-year-old male patient with a history of

trauma revealed evidence of extensive internal root resorption affecting both maxillary
central incisors (Fig. 1).
Because of the degree of tissue loss, both teeth were initially condemned for
extraction, but owing to the favourable periodontal condition (Fig. 2), the possibility of
conserving tooth 11 was considered through a combination of root canal treatment,
internal MTA repair and fibre-post reinforcement.
Root canal treatment was accomplished by established methods. The tooth was
isolated with a rubber dam without a clamp, to avoid the possibility of horizontal fracture.
After coronal access, the pulp tissue was removed. A small communication between the
resorption cavity and the lateral periodontium was observed as a haemorrhagic area
(Fig.3).
After working length determination, the canal was prepared by the EndoEze oscillatory
system (Ultradent Products Co., South Jordan, UT, USA). An apical stop was created with a
rotary nickel–titanium size 60, .04 Profile file (Dentsply Maillefer, Ballaigues, Switzerland).
This was accompanied by copious irrigation with freshly prepared 1% sodium hypochlorite
(Farmácia Amazon, São Carlos, SP, Brazil) and a calcium hydroxide paste dressing (Calen,
SSW Artigos Dentários Ltda., Rio de Janeiro, RJ, Brazil) was placed to alkalinize the
environment, remove remaining pulp tissue and control bleeding at the perforation.
After 7 days, the dressing material was removed with 1% sodium hypochlorite
irrigation and aspiration. The root canal was irrigated with a 3-min rinse with a buffered
solution EDTA of pH 7.4 (Odahcam; Dentsply, Petrópolis, RJ, Brazil) under instrument
agitation. The internal portion of resorbed space was filled with white MTA (MTA Angelus,
Londrina, PR, Brazil), inserted by means of an amalgam carrier and condensed with a size
40 Gutta Condensor (Dentsply Maillefer) aiming to remove air inclusions from the material.
CASE REPORT
Figure 1 Preoperative radiograph. Extensive internal root resorption in both maxillary central incisors.
Figure 2 Remaining central incisor and healthy gingiva.
After this, the apical part of the root canal was re-established by the introduction of a D
finger plugger (Dentsply Maillefer), and immediately afterwards, an auxiliary gutta-percha
point (Dentsply Maillefer, Ballaigues, Switzerland), was introduced to the working length
of the root canal (Fig.4a) and fitted to its apical preparation to preserve the original root
canal space. All excess filling material was removed from the coronal pulp chamber, which
was promptly sealed with a provisional cement.
At the third clinical session, successful hardening of the MTA was verified (Fig. 4b). The
‘guide’ gutta-percha point was removed Fig. 4c,d) and the empty root canal space was
filled with EndoRez Dual cement (Ultradent Products Co., South Jordan, UT, USA) and a
gutta-percha point, together with a fibre-glass post (Reforpost; Angelus) bonded to a
length of 14 mm with a chemically activated cement (Cement Post; Angelus). Finally, the
coronal chamber was restored with composite. An immediate postoperative radiograph
was taken (Fig. 5), confirming satisfactory filling of the root canal and resorptive defect.
CASE REPORT
Figure 3 Lateral periodontium bleeding evidencing pathological perforation.
(a) (b)
(c) (d)
Figure 4 (a) Use of a ‘guide’ gutta-percha point, to preserve the apical access throughout the root
canal. (b) Successful curing of white MTA. (c) Removing of the ‘guide’ gutta-percha after cement set.
(d) Preservation of root canal space for root canal filling and post installation.
Clinical and radiographic follow-up was conducted for 20 months (Figs 6 and 7),
demonstrating a functional tooth with no endodontic pathosis. Despite the tooth’s
retention in a functional condition, it discoloured to a grey tone (Fig. 8).
Discussion
Internal inflammatory root resorption is an insidious pathological process, initiated within

the pulp space and associated with loss of dentine. It is often described as an oval shaped
enlargement of the root canal space and is usually asymptomatic and detectable by
radiographs. When diagnosed, immediate removal of the causative agent must be
CASE REPORT
Figure 5 Immediate post-treatment radiograph (August 2005).
Figure 6 Twelve-month post-treatment radiograph (August 2006).
considered, aiming to arrest the cellular activity responsible for the resorptive activity
(Trope & Chivian 1997, Trope 2002).
According to Culbreath et al. (2000), the treatment for internal resorption can include
several materials such as gutta-percha, zinc oxide eugenol and amalgam alloy. However,
CASE REPORT
Figure 7 Twenty-month post-treatment radiograph (April 2007).
Figure 8 Post-treatment clinical aspect after 20 months, showing tooth discoloration (April 2007).
these materials do not provide strength to the tooth structure and may be responsible for
considerable tooth discolouration.
The use of a root canal dressing with a material based on calcium hydroxide between
sessions was aimed at dissolving remaining pulpal debris, alkalinizing the environment and
controlling periodontal bleeding (Siqueira & Lopes 2004). In the following session, the
defect in the canal was filled with white MTA to seal the perforation and fill the resorbed
area. For this case, MTA was selected because of its known abilities as a repair material,
along with its sealing ability and mechanical strength (Torabinejad et al. 1995a).
CASE REPORT
The fibre-glass posts were used to enhance the fracture resistance of the tooth.
Maccari et al. (2003) concluded that compromised teeth could be strengthened with
distribution of functional forces through the tooth’s long axis.
Clinical use of MTA in humans has demonstrated their applicability in wet environ-
ments, preventing bacterial microleakage and alkalinizing the medium. On account of the
predominant presence of calcium oxide in its formula (Camilleri & Pitt Ford 2006), its
biological properties show similarity to those of calcium hydroxide, making it useful for
tissue healing. The relative roles of MTA and the fibre post in mechanically reinforcing the
current tooth are unclear. Little evidence is also available on the bonding of composites
and other materials to MTA.
White MTA was introduced as a low-iron, nonstaining formula. Despite this, the cement
in this case discoloured the tooth perhaps as a result of oxidation of iron in the product
formulation: tetracalcium aluminoferrite (Camilleri & Pitt Ford 2006). This has not been
reported before and warrants further investigation.
Conclusion
Mineral trioxide aggregate and a fibre-glass post were employed to restore a severely
weakened, internally resorbed tooth, with satisfactory follow-up of 20 months. The
tooth discoloured despite the use of white MTA, highlighting the need for further
research.
Acknowledgements
The authors thank Professor Paulo Tambasco de Oliveira from FORP-USP Brazil, for his
helpful comments on the manuscript.
Disclaimer
Whilst this article has been subjected to Editorial review, the opinions expressed, unless
specifically indicated, are those of the author. The views expressed do not necessarily
represent best practice, or the views of the IEJ Editorial Board, or of its affiliated Specialist
Societies.
References
Bakland LK (2000) Management of traumatically injured pulps in immature teeth using MTA. Journal of
the California Dental Association 28, 855–8.
Camilleri J, Pitt Ford TR (2006) Mineral trioxide aggregate: a review of the constituents and biological
properties of the material. International Endodontic Journal 39, 747–54.
Consolaro A (2002) Reabsorções dentárias nas especialidades clı́nicas, 1st edn. Maringá, PR: Dental
Press.
Culbreath TE, Davis GM, West NM, Jackson A (2000) Treating internal resorption using a syringeable
composite resin. Journal of the American Dental Association 131, 493–5.
Holland R, de Souza V, Nery MJ et al. (2002) Reaction of rat connective tissue to implanted dentin
tubes filled with a white mineral trioxide aggregate. Brazilian Dental Journal 13, 23–6.
Hsien HC, Cheng YA, Lee YL, Lan WH, Lin CP (2003) Repair of perforating internal resorption with
mineral trioxide aggregate: a case report. Journal of Endodontics 29, 538–9.
Lyroudia KM, Dourou VI, Pantelidou OC, Labrianidis T, Pitas IK (2002) Internal root resorption studied
by radiography, stereomicroscope, scanning electron microscope, and computerized 3D recon-
structive method. Dental Traumatology 18, 148–52.
Maccari PC, Conceição EN, Nunes MF (2003) Fracture resistance of endodontically treated teeth
CASE REPORT
restored with three different prefabricated esthetic posts. Journal of Esthetics and Restorative
Dentistry 15, 25–30.
Main C, Mirzayan N, Shabahang S, Torabinejad M (2004) Repair of root perforations using mineral
trioxide aggregate: a long-term study. Journal of Endodontics 30, 80–3.
Menezes R, Bramante CM, Letra A, Carvalho VG, Garcia RB (2004) Histologic evaluation of
pulpotomies in dog using two types of mineral trioxide aggregate and regular and white Portland
cements as wound dressings. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and
Pitt Ford TR, Torabinejad M, Abedi HR, Bakland LK, Kariyawasam SP (1996) Using mineral trioxide
aggregate as a pulp-capping material. Journal of the American Dental Association 127, 1491–4.
Salako N, Joseph B, Ritwik P, Salonen J, John P, Junaid TA (2003) Comparison of bioactive glass,
mineral trioxide aggregate, ferric sulfate and formocresol as pulpotomy agents in rat molar. Dental
Traumatology 19, 314–20.
Simon S, Rilliard F, Berdal A, Machtou P (2007) The use of mineral trioxide aggregate in one-visit
apexification treatment: a prospective study. International Endodontic Journal 40, 186–97.
Siqueira JF Jr, Lopes HP (2004) Medicação intracanal. In: Siqueira JF Jr, Lopes HP, eds. Endodontia:
biologia e técnica, 2nd edn. Rio de Janeiro: Guanabara Koogan, pp. 581–618.
Torabinejad M, Chivian N (1999) Clinical applications of mineral trioxide aggregate. Journal of
Torabinejad M, Watson TF, Pitt Ford TR (1993) Sealing ability of a mineral trioxide aggregate when
used as a root end filling material. Journal of Endodontics 19, 591–5.
Torabinejad M, Hong CU, McDonald F, Pitt Ford TR (1995a) Physical and chemical properties of a new
root-end filling material. Journal of Endodontics 21, 349–53.
Torabinejad M, Hong CU, Pitt Ford TR, Kaiyawasam SP (1995b) Tissue reaction to implanted super-
EBA and mineral trioxide aggregate in the mandible of guinea pigs: a preliminary report. Journal of
Trope M (2002) Root resorption due to dental trauma. Endodontic Topics 1, 79–100.
Trope M, Chivian N (1997) Reabsorção radicular. In: Cohen S, Burns RC, eds. Caminhos da polpa, 6th
edn. Rio de Janeiro, BR: Guanabara Koogan, pp. 488–514.
White C Jr, Bryant N (2002) Combined therapy of mineral trioxide aggregate and guided tissue
regeneration in the treatment of external root resorption and an associated osseous defect. Journal
of Periodontology 73, 1517–21.
doi:10.1111/j.1365-2591.2008.01414.x
CASE REPORT
Complex endodontic treatment of an

immature type III dens invaginatus.
A case report
E. R. Fregnani1,2, L. F. B. Spinola2, J. R. O. Sônego2, C. E. S. Bueno2 &

A. S. De Martin2
1
Department of Stomatology, A. C. Camargo Cancer Hospital, São Paulo; and 2São Leopoldo
Mandic Dental Research Centre, Campinas, São Paulo, Brazil
Abstract
Fregnani ER, Spinola LFB, Sônego JRO, Bueno CES, De Martin AS. Complex endodontic
treatment of an immature type III dens invaginatus. A case report. International Endodontic Journal,
41, 913–919, 2008.
Aim To report the endodontic treatment of an immature maxillary central incisor with
dens invaginatus.
Summary Dens invaginatus is a rare malformation of teeth, probably resulting from an
infolding of the dental papilla during tooth development. The present case describes the
complex endodontic treatment of a type III dens invaginatus in an immature maxillary
central incisor with a necrotic pulp and abscess formation. The initial treatment goal was
to achieve apexification of the pseudocanal root and conservative root canal treatment in
the main canal. Following 1-year of treatment with calcium hydroxide dressings,
radiography revealed a healing response, but no sign of a hard tissue barrier at the apex.
Periapical surgery with the placement of a zinc oxide cement (IRM) root-end filling was
considered successful at the 4-year follow-up.
Key learning points
• The complexity of the canal system and open apex in dens invaginatus present a
challenge to endodontic treatment.
• Correct diagnosis and treatment planning are fundamental to treatment of dens
invaginatus.
• Periapical surgery is indicated in cases of unsuccessful apexification in immature teeth
with dens invaginatus and nonvital pulp.
Keywords: calcium hydroxide, dens invaginatus, endodontic treatment.

Received 8 July 2007; accepted 4 February 2008
Correspondence: Dr Eduardo Fregnani, Rua Alves Guimarães, 150/1307, São Paulo,

SP 05410-000, Brazil (Tel.: 55 11 2189-5129; fax: 55 11 2189 -5100; e-mail: erfreg@terra.com.br).
CASE REPORT
Introduction
Dens invaginatus is a dental developmental abnormality in which the enamel organ is

disorganized, leading to an invagination in the tooth crown before the calcification phase
(Chen et al. 1998, de Sousa & Bramante 1998). The cavity formed retains external
communication through the foramen coecum. The aetiology remains uncertain and its
prevalence varies substantially, ranging from 0.25 to 10%, depending on the type of
classification (Jung 2004). Most authors use Oehlers (1957) classification based on the
extent of the invaginated dental tissue. Type I, characterized by the invagination confined
within the crown, extending only to the cementoenamel junction; type II, characterized by
the invagination extending apically beyond the cementoenamel junction, where connec-
tion between the invagination and the pulp is possible; and type III, characterized by the
invagination extending beyond the cementoenamel junction and exhibiting a second
foramen into the lateral periodontal ligament or periradicular tissue. Various techniques
and approaches to treating dens invaginatus have been reported in the literature, including
preventive sealing or filling of the invagination, root canal treatment either associated or
not with endodontic apical surgery, intentional replantation and extraction (Hata & Toda
1987, de Sousa & Bramante 1998, De Martin et al. 2005).
Case report
A 9-year-old boy with severe pain and facial swelling was referred for endodontic
treatment of his maxillary right central incisor. An extensive intra-oral swelling was
observed in the buccal area above the affected incisor, which showed clinical signs of
morphological alteration, with a pronounced ‘dens invaginatus’ on the palatal surface, but
no visible foramen coecum. The tooth had no obvious decay; there was no history of oral–
facial trauma and the colour of the tooth did not differ from that of the adjacent teeth. The
total number of teeth in the anterior region was normal. Vitality tests were performed with
refrigerant spray (dichlorotetrafluoroethane, Endo-Frost; Roeko, Langenau, Germany) and
there was no response from the right maxillary central incisor.
Radiographic examination revealed the presence of an immature Oehlers type III dens
invaginatus and an extended radiolucent area in the apical region. A foramen coecum was
visible as a radiolucency on the crown (Fig. 1). A large invagination extended from the
crown to the apex of the root (pseudocanal), with no evident communication with the
main canal. Moreover, the main canal was associated with complete root formation, but
as seen on Fig. 1, the pseudocanal apex was not completely shaped.
With the diagnosis of acute alveolar abscess and after local anaesthesia, surgical soft
tissue drainage was performed and systemic analgesics were prescribed (Paracetamol
500 mg, 6/6 h.s. for 2 days, Jenssen-Cilag Farmacêutica, São José dos Campos, Brazil).
After 1 week, there were no signs of acute inflammation and conservative nonsurgical
root canal treatment was the treatment of choice for the main canal (with full root
formation). Figure 2 shows the clinical signs of morphological alteration in the maxillary
right central incisor at that time. After isolating the tooth, access preparation was
completed and two canal orifices were located. Preparation of the main canal was
accomplished with the balanced-force technique of instrument manipulation and 5.25%
sodium hypochlorite irrigant. Working length was determined with an apex locator (Root
ZX; J. Morita Europe GmbH, Dietzenbach, Germany) and a radiograph. After canal
preparation, the canal was irrigated with 17% EDTA for 3 min followed by irrigation with
5.25% sodium hypochlorite, then dried and filled with Gutta-percha and AH-plus root canal
sealer (Dentsply-De Trey, Konstanz, Germany), using lateral compaction and Gutta-
condensers (Dentsply-Maillefer, Ballaigues, Switzerland).
CASE REPORT
Figure 1 Periapical radiograph shows a type III dens invaginatus. Main canal (MC) and pseudo-canal
(PC) are indicated. Apical radiolucence and open apex are also observed.
Figure 2 Clinical signs of morphological alteration in the maxillary right central incisor.
The pseudocanal was filled with calcium hydroxide and an apexification treatment was
initiated (Fig. 3). The calcium hydroxide dressing was renewed every month for 1 year and
although there was radiographic evidence of healing, no hard tissue barrier was formed.
The patient failed to attend for 2 years, and when a follow-up periapical radiograph was
taken, it revealed no hard tissue barrier formation (Fig. 4). A decision was made at this
stage to complete treatment by a combined nonsurgical and surgical approach.
First, the coronal root filling material was condensed with warm hand instrument and
the coronal cavity was sealed with a composite restoration. The radiographs showed
CASE REPORT
Figure 3 Periapical radiograph immediately after main canal filling and calcium hydroxide treatment
for pseudo canal apexicification.
Figure 4 Periapical radiograph shows healing of the periapical lesion, but no evidence of hard tissue
barrier was found.
inadequate root filling with poor apical adaptation. Surgical access followed at the same
appointment and a triangular flap was raised before removing the extruded root filling
material. An immature root apex with defective dentine walls was revealed and its surface
CASE REPORT
Figure 5 A 4-year follow-up periapical radiograph.
was corrected with appropriate ultrasonic tips, a zinc oxide cement (IRM) was placed into
the root-end cavity. The flap was sutured for 7 days; an antibiotic (Amoxicillin 250 mg each
8 h for 7 days; Medley S/A Indústria Farmacêutica, Campinas, Brazil), analgesic (Parac-
etamol 500 mg, 6/6 h.s. for 2 days; Jenssen-Cilag Farmacêutica, São José dos Campos,
Brazil) and 2% chlorhexidine mouth rinse were prescribed.
After 4 years of clinical and radiographic follow-up, the tooth was asymptomatic, normal
clinical conditions were observed and the periapical appearance on the radiograph was
considered healthy (Fig. 5).
Discussion
Endodontic treatment in teeth with dens invaginatus, which extends up to the apical
region and associated with apical pathosis, generally involves complicated procedures that
require accurate diagnosis and appropriate treatment planning (Chen et al. 1998, De Martin
et al. 2005). There is a high incidence of pulp infection and degeneration associated with
this anomaly, and because of this, Oehlers (1957) emphasized the importance of early
diagnosis to prevent pulp necrosis and periapical inflammation. However, in many cases
of conventional root canal treatment, surgical or combined treatments need to be carried
out (Ferguson et al. 1980, Holtzman & Lezion 1996, Pai et al. 2004).
The complexity of the canal system and possible open apex present a challenge to root
canal treatment (da Silva Neto et al. 2005, Sübay & Kayatas 2006, Soares et al. 2007).
Furthermore, the morphology of the main canal may be irregular, circular or narrow and a
possible explanation is that the invaginated tissue occupied and compressed the main
canal at different levels (De Smit & Demaut 1982, Soares et al. 2007). As observed in
the present case, pulp necrosis and acute periradicular abscess are frequently found
in cases of dens invaginatus (Ferguson et al. 1980, Rotstein et al. 1987, Yeh et al. 1999,
De Martin et al. 2005).
Nonsurgical root canal treatment should be attempted first. Irrespective of the size of
CASE REPORT
the periradicluar lesion, surgical treatment is the second option and is only indicated when
nonsurgical root canal treatment has failed (Yeh et al. 1999, Steffen & Splieth 2000,
Pai et al. 2004, Subay & Kayatas 2006). In this case, the option was to attempt
apexification with the use of calcium hydroxide, as other reports have indicated this to be
successful (Ferguson et al. 1980, Tarjan & Rozsa 1999, Yeh et al. 1999, Jung 2004).
However, the technical result obtained was not satisfactory and further surgical treatment
was indicated.
Although MTA has been widely evaluated and is the preferred root-end material, IRM
has been extensively studied with favourable results (Chong et al. 2003). As evidenced
with this case and others, it can be considered as an alternative material (Bernabé et al.
2005, Lindeboom et al. 2005).
The use of zinc oxide cement may lead to the formation of a fibrous barrier that has
been shown to reduce the initial toxicity of the cement enhancing tissue repair (Olsen
et al. 1994). Pitt Ford et al. (1994) reported that tissue response to IRM was one of
toleration rather than biocompatibility. So, the radiolucency that can be observed in Fig. 5
suggests scar formation that tends to hinder complete periapical regeneration.
Periapical surgery is indicated in cases in which conventional root canal therapy has
failed, and in teeth that present anatomic variations that do not allow access to and
cleaning of all the parts of the canal system, such as many cases of dens invaginatus type
III with periapical lesions (Hata & Toda 1987, da Silva Neto et al. 2005, Sübay & Kayatas
2006, Soares et al. 2007).
Conclusion
With the complications presented in this case of dens invaginatus type III, including an
open apex and acute periapical abscess, a combination of conservative and surgical
treatments preserved a functional and aesthetic tooth during a 4-year follow-up.
Disclaimer
Societies.
References
Bernabé PF, Holland R, Morandi R et al. (2005) Comparative study of MTA and other materials in
retrofilling of pulpless dogs’ teeth. Brazilan Dental Journal 16, 149–55.
Chen YH, Tseng CC, Harn WM (1998) Dens invaginatus. Review of formation and morphology with 2
case reports. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 86,
347–52.
Chong BS, Pitt Ford TR, Hudson MB (2003) A prospective clinical study of mineral trioxide aggregate
and IRM when used as root-end filling materials in endodontic surgery. International Endodontic
Journal 36, 520–6.
De Martin AS, da Silveira Bueno CE, Sandhes Cunha R, Aranha de Araujo R, Fernandes de Magalhaes
Silveira C (2005) Endodontic treatment of dens invaginatus with a periradicular lesion: case report.
Australian Endodontic Journal 31, 123–5.
De Smit A, Demaut L (1982) Nonsurgical endodontic treatment of invaginated teeth. Journal of
CASE REPORT
Ferguson F, Friedman S, Frazetto V (1980) Successful apexification technique in an immature tooth
with dens in dente. Oral Surgery, Oral Medicine, and Oral Pathology 49, 356–9.
Hata G, Toda T (1987) Treatment of dens invaginatus by endodontic therapy, apicocurettage, and
retrofilling. Journal of Endodontics 13, 469–72.
Holtzman L, Lezion R (1996) Endodontic treatment of maxillary canine with dens invaginatus and
immature root. Oral Surgery, Oral Medicine, Oral Patholology, Oral Radiology, and Endodontics 82,
452–5.
Jung M (2004) Endodontic treatment of dens invaginatus type III with three root canals and open
apical foramen. International Endodontic Journal 37, 205–13.
Lindeboom JA, Frenken JW, Kroon FH, van den Akker HP (2005) A comparative prospective
randomized clinical study of MTA and IRM as root-end filling materials in single-rooted teeth in
endodontic surgery. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics
100, 495–500.
Oehlers FAC (1957) Dens invaginatus (dilated composite odontome). I. Variations of the invagination
process and associated anterior crown forms. Oral Surgery, Oral Medicine, and Oral Pathology 10,
1204–18.
Olsen FK, Austin BP, Walia H (1994) Osseous reaction to implanted ZOE retrograde filling materials in
the tibia of rats. Journal of Endodontics 20, 389–94.
Pai SF, Yang SF, Lin LM (2004) Nonsurgical endodontic treatment of dens invaginatus with large
periradicular lesion: a case report. Journal of Endodontics 30, 597–600.
Pitt Ford TR, Andreasen JO, Dorn SO, Kariyawasam SP (1994) Effect of IRM root end fillings on
healing after replantation. Journal of Endodontics 20, 381–5.
Rotstein I, Stabholz A, Heling I, Friedman S (1987) Clinical considerations in the treatment of dens
invaginatus. Endodontics & Dental Traumatology 3, 249–54.
da Silva Neto UX, Hirai VH, Papalexiou V et al. (2005) Combined endodontic therapy and surgery in the
treatment of dens invaginatus type 3: case report. Journal of Canadian Dental Association 71, 855–8.
Soares J, Santos S, Silveira F, Nunes E (2007) Calcium hydroxide barrier over the apical root-end of a
type III dens invaginatus after endodontic and surgical treatment. International Endodontic Journal
40, 146–55.
de Sousa SM, Bramante CM (1998) Dens invaginatus: treatment choices. Endodontics & Dental
Traumatology 14, 152–8.
Steffen H, Splieth C (2000) Conventional treatment of dens invaginatus in maxillary lateral incisor with
sinus tract: one year follow-up. Journal of Endodontics 31, 130–3.
Sübay RK, Kayatas M (2006) Dens invaginatus in an immature maxillary lateral incisor: a case report of
complex endodontic treatment. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and
Tarjan I, Rozsa N (1999) Endodontic treatment of immature tooth with dens invaginatus: a case report.
International Journal of Paediatric Dentistry 9, 53–6.
Yeh SC, Lin YT, Lu SY (1999) Dens invaginatus in the maxillary lateral incisor: treatment of 3 cases.
Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 87, 628–31.
doi:10.1111/j.1365-2591.2008.01423.x
CASE REPORT
Foreign body in the apical portion of a

root canal in a tooth with an immature
apex: a case report
A. R. Prabhakar, S. Namineni & H. N. Subhadra

Department of Pedodontics and Preventive Dentistry, Bapuji Dental College and Hospital,
Davangere, India
Abstract
Prabhakar AR, Namineni S, Subhadra HN. Foreign body in the apical portion of a root canal in a
tooth with an immature apex: a case report. International Endodontic Journal, 41, 920–927, 2008.
Aim To describe the successful retrieval of a foreign object located in the apical portion of
an immature root canal by simple orthograde techniques, avoiding the need for surgery or
intentional reimplantation.
Summary A radio-opaque foreign object lodged in the apical portion of an immature root
canal was discovered on radiographic examination of a patient with a complicated crown
fracture. Attempts to retrieve it resulted in displacement into the periapical area.
Eventually, the object was retrieved by a simple technique, followed by successful
apexification, root canal filling and jacket crown placement.
Key learning points
• Foreign bodies in root canals should be carefully evaluated to determine their nature,
position, size and the degree of difficulty that may be encountered during retrieval.
• Patience, care and appropriate techniques may be helpful in retrieving foreign bodies
and avoiding periapical surgery.
• Complicated crown fractures should be managed promptly, and prolonged open
drainage avoided in children if the risks of foreign body impaction are to be minimized.
Keywords: apical portion, foreign body, immature apex, non surgical technique.
Received 3 August 2007; accepted 27 February 2008
Introduction
Root canal treatment can be challenging in children and occasionally clinicians may
encounter bizarre situations that require both skill as well as perseverance. Children have
the habit of placing foreign objects in the oral cavity which can cause both hard and soft
Correspondence: Dr A.R. Prabhakar, MDS, Professor and Head, Department of Pedodontics and
Preventive Dentistry, Bapuji Dental College and Hospital, Davangere 577004, Karnataka, India
(Tel.: 91 8192 220575; fax: 91 8192 220578; e-mail: attiguppeprabhakar@yahoo.com).
CASE REPORT
tissue injuries. At times, these objects can get lodged inside the pulp chamber or root
canal of a tooth. This is more likely to occur in a tooth with an open pulp chamber caused
by trauma, during root canal procedures in which canals are left open for drainage, and in
the case of open carious lesions. Such foreign objects may become a potent source of
pain and focus of infection for the patient. These objects can be retrieved with some ease
if they are located within the pulp chamber, but once the object has been pushed apically,
their retrieval may be complicated. Surgery or intentional reimplantation may sometimes
be unavoidable.
The report describes the case of a foreign object impacted into the apical third of an
immature maxillary central incisor which was finally retrieved by simple, intracanal means.
Case report
A 12-year-old male reported to the Department of Pedodontics and Preventive Dentistry,

Bapuji Dental College and Hospital, India with a 1-month history of pain in an upper front
tooth. He had suffered dental trauma 2½ years previously.
Intra-oral examination revealed a complicated enamel–dentine fracture with a slit-like
opening into the pulp chamber of tooth 11 (Federation Dentaire Internationale). The tooth
exhibited the following clinical features:
• grade I mobility;
• tenderness in the buccal sulcus;
• pain on percussion and
• a draining sinus on the attached gingiva.
An intra-oral periapical radiograph revealed the presence of a linear radio-opaque object
in the root canal, extending from the middle third to the immature apex of the root (Fig. 1).
After taking the clinical and radiographic findings into consideration, it was decided that
root canal treatment should be initiated, with an attempt to retrieve the foreign object and
thereafter complete the root canal treatment.
A conventional access cavity was prepared and the pulp chamber was cleared of debris by
copious irrigation with saline solution. Attempts were made to retrieve the object using 40
size K-files (Mani, Inc., Nakaakutso, Japan) using a simple filing action; this was
unsuccessful. A second intra-oral periapical radiograph was then obtained with a slightly
different horizontal angulation. The object took on a different shape and was partly extruded
into the periapical region (Fig. 2). Moreover, the second radiograph confirmed the presence
of the object within the root canal rather than periodontal ligament space. As the root canal
was large, a decision was made to retrieve the foreign object using a size 120 K-file by
attempting to engage the object between the file and canal walls and then by pulling it out
coronally. Exploration along the palatal portion of the canal was successful in retrieving the
object towards the pulp chamber, which was then grasped with tweezers and removed.
The retrieved foreign object appeared golden in colour and measured approximately
4 mm in length (Fig. 3). The patient did not know what it was, and denied having inserted the
object within the tooth. The patient’s mother felt it could be a fractured piece of an ornament.
Following retrieval of the foreign object, an intracanal calcium hydroxide medicament
was placed and apical closure was achieved in five and a half months (Fig. 4). The root
canal was filled using the rolled cone technique (Gutmann & Heaton 1981) (Fig. 5),
followed by core build up and a jacket crown.
Discussion
A number of cases have been cited in the literature describing various foreign objects
being lodged in the pulp chamber or root canal. Most of the cases arose when the pulp
CASE REPORT
Figure 1 Radio-opaque object extending from the middle third of the root canal to the immature apex
of tooth 11 with a complicated crown fracture.
chamber was wide open. These objects have ranged from pencil leads (Hall 1969), darning
needles (Nernst 1972), metal screws (Prabhakar et al. 1998), to beads (Reddy & Mehtha
1990), paper clips (Cataldo 1976) and stapler pins (Macauliffe et al. 2005). Grossman
(1974) reported retrieval of indelible ink pencil tips, brads, a tooth pick, adsorbent points
and even a tomato seed from the root canals of anterior teeth left open for drainage. Toida
et al. (1992) have reported a plastic chopstick embedded in an unerupted supernumerary
tooth in the pre-maxillary region of a 12-year-old Japanese boy.
A common procedure employed during emergency root canal treatment involves
leaving the pulp chamber open where pus continues to discharge through the canal and
cannot be dried within a reasonable period of time (Cohen & Brown 2002). Such a
procedure may place the patient at risk of foreign body lodgement in the canal. Numerous
reports on foreign bodies being detected within the open pulp chamber and canal may
question the safety of such procedures. Alternatively, Weine (2004) recommends that the
patient remains in the office with a draining tooth for an hour or even more and finally
ending the appointment by sealing the access cavity. With the access cavity closed, no
new strains of microorganism systems are introduced and food debris and foreign body
lodgement within the tooth can be avoided (Nair 2006).
If a clinician decides to leave the pulp chamber open following access cavity preparation,
the patient and parents should be warned about the risks of any foreign object being lodged
in the open canal. However, the clinician should always consider the benefits and risks
associated with leaving the pulp chamber open for prolonged periods of time.
922 ª 2008 International Endodontic Journal, 41, 920–927, 2008 ª 2008 International Endodontic Journal
CASE REPORT
Figure 2 Periapical film at a different angulation: the object had extruded into the periapical region.
Foreign bodies in root canals may act as obstructions for the smooth passage of
endodontic instruments. A radiograph can be of diagnostic significance especially if the
foreign body is radio-opaque. Specialized radiographic techniques such as radiovisiogra-
phy, 3D CAT (computerized axial tomography) scans can play a pivotal role in the
localization of the exact position of these foreign objects inside the root canal.
Foreign bodies in root canals can act as focus of infection. Actinomycosis following
placement of piece of jewellery chain into a maxillary central incisor has been reported
(Goldstein et al. 1972). Foreign bodies pushed through root canal into the sinus are one of
the causes of chronic maxillary sinusitis of dental origin (Costa 2006). Hence, prompt
attempts at their retrieval should be initiated.
Retrieval of foreign objects lying in the pulp chamber or canal using ultrasonic
instruments (Meidinger & Kabes 1985), the Masserann kit (Williams & Bjorndal 1983),
modified Castroviejo needle holders (Fros & Berg 1983) have been described in the
literature. There is also a description of an assembly of a disposable injection needle and
thin steel wire loop formed by passing the wire through the needle being used. This
assembly was used along with a mosquito haemostat to tighten the loop around the
object (Roig-Greene 1983).
CASE REPORT
Figure 3 Retrieved foreign object.
The use of an operating microscope is also beneficial. The microscope gives light and
illumination inside the canal and provides the clinician with the ability to visualize any
intraradicular obstruction and locate its position in relation to surrounding root canal walls.
Nehme (2001) has recommended the use of operating microscope along with ultrasonic
filing to eliminate intracanal metallic obstructions.
Nonetheless, retrieval of the object may become difficult when it is lodged in the
periapical region. Srivastava & Vineeta (2001) have suggested periapical surgery or
intentional reimplantation to remove such objects. They reported retrieval of a straight pin
lodged in the periapical area of maxillary central incisor by periapical surgery. Zillich &
Pickens (1982) also resorted to the surgical approach for removing the apical portion of a
hat pin lodged in a maxillary lateral incisor.
In the present case, the foreign object was located within the root canal and retrieved
successfully by a simple nonsurgical technique. It is essential that the dentist, when faced
with retrieval of a foreign body, obtains a thorough history, carries out a detailed
examination and necessary investigation to determine the position, size, likely composi-
tion, and degree of difficulty that will be encountered during its retrieval.
As a foreign object can act as a source of pain and cause difficulty in the elimination of
infection from the root canal, prompt but cautious attempts should be made to retrieve it
first by simple nonsurgical means. Finally, when the foreign object resists all efforts for
removal and when a strong possibility of failure exists, a surgical procedure may be the
only viable alternative. This technique however eliminates the possibility of apex closure in
the case of an immature apex. Numerous reports on foreign bodies detected within the
open pulp chamber and canal question the safety of open drainage during endodontic
treatment.
CASE REPORT
Figure 4 Intracanal calcium hydroxide dressing following retrieval of the foreign object.
Conclusions
A radio-opaque foreign object lodged in the apical portion of an immature root canal was
discovered on radiographic examination of a patient with a complicated crown fracture.
Attempts to retrieve it resulted in displacement into the periapical area. Eventually, the
object was retrieved by a simple technique, followed by successful apexification, root
canal filling and jacket crown placement.
Disclaimer
Societies.
References
Cataldo E (1976) Unusual foreign object in pulp canal. Oral Surgery Oral Medicine Oral Pathology 42,
851.
Cohen AS, Brown DC (2002) Orofacial dental pain emergencies: Endodontic diagnoses and
management. In: Cohen S, Burns RC, ed. Pathways of the Pulp, 8th edn. St. Louis, MO: Mosby,
pp. 31–75.
CASE REPORT
Figure 5 Root canal filled with gutta-percha cones following successful apexification 80 · 104 mm
(600 · 600 DPI).
Costa F, Robiony M, Toro C, Sembronio S, Politi M (2006) Endoscopically assisted procedure for
removal of a foreign body from the maxillary sinus and contemporary endodontic surgical treatment
of the tooth. Head and Face Medicine 2, 37.
Fros UGH, Berg JO (1983) A method for the removal of broken endodontic instruments from root
canals. Journal of Endodontics 9, 156–9.
Goldstein BH, Scuibba JJ, Laskin DM (1972) Actinomycosis of the maxilla: a review of literature and
report of a case. Journal of Oral Surgery 30, 362–6.
Grossman LI (1974) Endodontic case reports. Dental Clinics of North America 18, 509–27.
Gutmann JL, Heaton JF (1981) Management of the open (immature) apex. 2. Non-vital teeth.
International Endodontic Journal 14, 173–8.
Hall JB (1969) Endodontics – patient performed endodontics. Journal of Dentistry for Children 36,
213–5.
Macauliffe N, Drage NA, Hunter B (2005) Staple diet: a foreign body in a tooth. International Journal of
Paediatric Dentistry 15, 468–71.
Meidinger DL, Kabes BJ (1985) Foreign object removal utilizing the cavi-endo ultrasonic instrument.
Journal of Endodontics 11, 301–4.
Nair PNR (2006) On the causes of persistent apical periodontitis: a review. International Endodontic
Journal 39, 249–81.
Nehme WB (2001) Elimination of intracanal metallic obstructions by abrasion using an operational
microscope and ultrasonics. Journal of Endodontics 27, 365–7.
Nernst H (1972) Foreign body in a root canal. Quintessence International 3, 33–4.
Prabhakar AR, Basappa N, Raju OS (1998) Foreign body in a mandibular permanent molar – a case
report. Journal of Indian Society of Pedodontics and Preventive Dentistry 16, 120–1.
CASE REPORT
Reddy VVS, Mehtha DS (1990) Beads. Oral Surgery Oral Medicine Oral Pathology 69, 769–70.
Roig-Greene JL (1983) The retrieval of foreign objects from root canals: a simple aid. Journal of
Srivastava N, Vineeta N (2001) Foreign body in the periradicular area. Journal of Endodontics 27, 593–
4.
Toida M, Ichihara H, Okutomi T, Nakamura K, Ishimaru JI (1992) An unusual foreign body in an
unerupted supernumerary tooth. British Dental Journal 173, 345–6.
Weine FS (2004) Endodontic Therapy, 6th edn. St. Louis, MO: Mosby.
Williams VD, Bjorndal AM (1983) The Masserann technique for the removal of fractured posts in
endodontically treated teeth. Journal of Prosthetic Dentistry 49, 46–8.
Zillich RM, Pickens TN (1982) Patient induced blockage of the root canal. Report of a case. Oral
Surgery Oral Medicine Oral Pathology 54, 689–90.

International Endodontic Journal PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

International Endodontic Journal PDF

Uploaded by

Copyright:

Available Formats

doi:10.1111/j.1365-2591.2008.01446.

Effect of root canal irrigants on cervical dentine

P. Surapipongpuntr, W. Duangcharee, S. Kwangsamai & A. Ekka

Abstract various solutions, the diffusion of 30% H2O2 was

attributed to the toxic nature of 30% H2O2 which

Materials and methods

Dentine disc preparation

Assay for H2O2

The inner surface of the dentine disc was observed by a

Root canal morphology of mandibular permanent

H. R. D. Peiris1, T. N. Pitakotuwage2, M. Takahashi1, K. Sasaki3 & E. Kanazawa3

Abstract Results In both first and second molars, develop-

often cause extensive differentiation of the root canal

an informed-consent protocol, which had been Figures in parentheses denote percentages.

5% nitric acid at room temperature (20 C); the nitric

(ad. 1) (ad. 2) (ad. 3)

Figure 3 Transparent root canal appearance of mandibular

Table 4 Number and position of inter-canal communications in mandibular molars

Position of inter-canal communications

Figures in parentheses denote percentages.

Table 5 Number and position of lateral canals in mandibular molars

Position of lateral canals

Tooth Root n Lateral canals C M A C+A M+A

First MR 177 92 (52.0) – 3 (1.7) 80 (45.2) – 9 (5.1)

Figures in parentheses denote percentages.

Effectiveness of different gutta-percha techniques

N. Gencoglu1, T. Yildirim2, Y. Garip1, B. Karagenc3 & H. Yilmaz3

Abstract gutta-percha and void in the internal resorptive cavi-

during cleaning, shaping and filling of the root canal

differentiate gutta-percha and sealer, even with the

Figure 1 (a,b) Radiographic appearance

manufacturer’s instructions. Sealer (Kerr) was applied

Group 4: Thermafil obturation

Table 1 Percentages of canal area filled

Modification of mineral trioxide aggregate. Physical

Abstract materials were compared with proprietary brand

be suitable for use as a pulp capping agent (Pitt Ford et al.

Flexural strength N mm-2

Neuropeptide Y Y1 receptor in human dental pulp

I. A. El Karim, P.-J. Lamey, G. J. Linden & F. T. Lundy

Abstract levels in noncarious, moderately carious and grossly

ropeptides cannot cross cell membranes and therefore

DPX (Searle Diagnostic, UK). Pre-immune rabbit

Figure 2 Representative western blots of NPY Y1 receptor

was shown to be localised in the smooth muscles cells

Substance P influenced gelatinolytic activity via

F.-M. Wang1, T. Hu1, R. Cheng1, H. Tan2 & X.-D. Zhou1

Abstract evaluate the roles of ROS pathway in mediating the

reported to have the ability to release SP when

Isolation and culture of human pulp cells Gelatin zymography

The gelatinolytic activity of the conditioned media

Control is known that dental pulp tissue is innervated with a

Pre-clinical endodontics: a survey amongst German

D. Sonntag1, R. Bärwald1, M. Hülsmann2 & V. Stachniss1

Abstract Results With feedback from 27 of 28 dental schools,

25% of the teeth with root fillings had signs of

11–20 2 (7.4) 10–20 5 (18.5)

Table 5 Minimum number of artificial canals, incisors/

A minority of universities (37%, n = 10) had a dental

Table 7 Teaching materials at the students’ disposal (the

Glucose reactivity with filling materials as a

H. Shemesh1, E. M. Souza2, M.-K. Wu1 & P. R. Wesselink1

Abstract glucose was evaluated using an enzymatic reaction