Basic ResearchBiology
Viability of Intratubular Bacteria after Chemomechanical
Caries Removal
Hamdi H. Hamama, BDS, MDS, PhD,* Cynthia K. Yiu, BDS, MDS, PhD,
and Michael F. Burrow, BDS, MDS, DDSc, PhD
Abstract
Introduction: The aim of this study was to assess the ef-
fect on bacterial viability within dentinal tubules after the
application of sodium hypochlorite (NaOCl) or enzyme-
based chemomechanical caries removal agents.
Methods: Twenty-five caries-free dentin discs prepared
from 25 sound maxillary premolars were used. The discs
were then infected with Streptococcus mutans sus-
pension and randomly divided into the following 6 groups
according to the dentin treatments: the negative control
group: noninfected sound dentin discs, the positive con-
trol group: infected discs were left untreated; the NaOCl
group: treated with 5% NaOCl solution, the chlorhexidine
(CHX) group: 2% CHX solution, the Carisolv group: Cari-
solv gel (Medi Team Dentalutveckling AB, Savedalen,
Sweden), and the Papacarie group: Papacarie gel (For-
mula & Acao, S~ao Paulo, Brazil). All the agents were
applied for 5 minutes. The dentin discs were fractured
into 2 halves and stained with fluorescent LIVE/DEAD
Stain (Molecular Probes, Eugene, OR). Each specimen
was observed using confocal laser scanning microscopy
at 5 different randomly selected sites. Results: The re-
sults of 1-way analysis of variance revealed that 5%
NaOCl solution achieved the highest intratubular antibac-
terial effect, whereas Carisolv gel had the lowest antibac-
terial effect (P < .05). No significant difference in
antibacterial effect was observed between the Papacarie
gel and 2% CHX solution (P > .05). Conclusions: The
enzyme-based chemomechanical caries removal (Papa-
carie) was effective in the reduction of residual cariogenic
bacteria in the dentinal tubules of coronal dentin.
(J Endod 2014;40:19721976)
Key Words
Carisolv, chemomechanical caries removal, confocal
laser microscopy, Papacarie, Streptococcus mutans
From the *Faculty of Dentistry, Paediatric Dentistry and
Orthodontics, and Oral Diagnosis and Polyclinics, University
of Hong Kong, Hong Kong SAR, China.
Address requests for reprints to Prof Cynthia K. Yiu, Paedi-
atric Dentistry and Orthodontics, Faculty of Dentistry, The Uni-
versity of Hong Kong, Prince Philip Dental Hospital, 34 Hospital
Road, Hong Kong SAR, China. E-mail address: ckyyiu@hkucc.
hku.hk
0099-2399/$ - see front matter
Copyright 2014 American Association of Endodontists.
http://dx.doi.org/10.1016/j.joen.2014.07.025
1972 Hamama et al.
T he removal of infected tissue and the reduction of residual cariogenic bacteria are
the ultimate goals of cavity preparation (1). Residual bacteria remaining in the cavity
after caries excavation may allow the progression of secondary caries in the event that
the tooth restoration seal fails. Chemomechanical caries removal agents can be classi-
fied into 2 groups: sodium hypochlorite (NaOCl) and enzyme-based agents. Currently,
Carisolv (Medi Team Dentalutveckling AB, Savedalen, Sweden) and Papacarie
(a papain-based gel) (Formula & Acao, S~ao Paulo, Brazil) are the only available prod-
ucts on the market representing the NaOCl- and enzyme-based chemomechanical caries
removal agents, respectively.
The majority of studies have reported that the chemomechanical caries removal
method preserved caries-affected dentin (2, 3). However, clinically, it is very difficult
to differentiate between caries-affected dentin and deep layers of caries-infected
dentin, which may contain residual bacteria (4). The presence of residual bacteria,
especially those in the dentinal tubules, may cause inflammatory reactions in the
pulp, leading to pulpal complications. Hence, chemomechanical caries removal
would be even more effective if it could also kill or reduce the number of remaining
bacteria in the cavity.
Few studies have evaluated the antibacterial effects of NaOCl-based chemomechan-
ical caries removal agents on superficial excavated dentin surface (57). These studies
used microbial assay methods to compare pre- and postexcavation samples, which
could have been affected by various factors, such as lesion depth, oral condition,
and operating field (5, 7). At the time of this study, no literature has evaluated the
antibacterial effect of enzyme-based chemomechanical caries removal agents. Further-
more, no studies have evaluated the viability of bacteria remaining inside the dentinal
tubules after chemomechanical caries removal.
Recently, a new method has been reported that centrifuged bacteria such as
Enterococcus faecalis into the dentinal tubules (8, 9). So far, this technique has
been used to infect radicular dentin to evaluate the antibacterial effectiveness of
contemporary root canal irrigating solutions (8, 9). This methodology was
modified for use in coronal dentin using Streptococcus mutans. The aim of this
study was to evaluate the viability of intratubular bacteria in coronal dentin after a
5-minute treatment using either NaOCl- or enzyme-based chemomechanical caries
removal agents to simulate the effect on residual bacteria after caries excavation.
The null hypothesis tested was that there was no difference in the viability of intra-
tubular bacteria after the application of the 2 chemomechanical caries removal
agents.
Materials and Methods
Preparation of Dentin Discs
Thirty sound caries-free maxillary premolars stored in 0.5% chloramine T solu-
tion at 4
C were used within 6 months after extraction. The study was approved by a
local institutional review board (ref no: UW 11-355). The occlusal enamel was removed
using a low-speed water-cooled diamond saw (IsoMet; Buehler Ltd, Lake Bluff, IL). A 2
mm-thick dentin disc was prepared from each premolar. The discs were ultrasonically
cleaned in 5.25% NaOCl and 6% citric acid solutions, respectively, for 4 minutes each to
remove the smear layer. Each disc was stabilized at the bottom of a 50-mL centrifugation
tube using a resin composite restorative material (Filtek Z250; 3M ESPE, St Paul, MN
JOE Volume 40, Number 12, December 2014

[batch #N347713]). The spaces between the untested surfaces of the
disc and the wall of the centrifugation tube were carefully blocked
with the same resin material. However, the blocking resin did not cover
any part of the tested dentin surface.
Infection of the Dentin Discs with S. Mutans
In the current study, the coronal dentin discs were infected
following the protocol described by Ma et al (8), except that S. mutans
was used instead of E. faecalis. S. mutans was previously isolated from
soft human carious lesions and suspended in brain-heart infusion
broth. The bacterial suspension was standardized to 3  106 colony-
forming unit/mL using a spectrophotometer (DU 530; Life Science
UV/Vis, Beckman, IN). Five hundred microliters of the bacterial suspen-
sion was added to each tube and centrifuged at 1400g, 2000g, 3600g,
and 5600g, respectively, for 5 minutes each. Each centrifugation cycle
was repeated twice. At the end of each cycle, a new 500 mL bacterial
suspension was added after discarding the old one. All tubes were
incubated at 37
C to allow bacterial recovery; afterward, the discs
were removed from the tubes for the following treatment and staining
procedures.
Dentin Treatment
The infected dentin discs were randomly divided into the following
6 groups according to the treatment regimen (5 discs per group): the
negative control group: 5 noninfected sound dentin discs, the positive
control group: the discs were left untreated, the NaOCl group: the discs
were treated with 5.25% NaOCl solution, the chlorhexidine (CHX)
group: 2% CHX solution freshly prepared from 1,10 -hexamethylenebis
(5-[P-chloro-phenyl] biguanide) dihydrochloride (C22 H30 Cl2 N10.
2Hcl) (batch 65H0778; Sigma-Aldrich, St Louis, MO), the Carisolv
group: Carisolv gel (batch no. 10489), and the Papacarie group: Papa-
carie gel (batch no. 8996). Following the protocol of Ma et al (8), the
untreated surface of each disc (opposite side to treated surface) was
covered with 2 layers of nail varnish to ensure that the bacteria were
maintained within the dentinal tubules. All the treatments were passively
applied on the dentin surface for 5 minutes. The treated dentin surface
was then thoroughly rinsed with deionized water for 1 minute. The
dentin discs were fractured in 2 halves and stained with fluorescent
LIVE/DEAD Baclight Bacterial Viability Stain (Molecular Probes,
Eugene, OR).
Observation of the Fractured Specimens
with Confocal Laser Scanning Microscopy
Each specimen was observed using confocal laser scanning micro-
scopy (Fluoview 1000; Olympus Corp, Shinjuku-Ku, Tokyo, Japan) at 5
different randomly selected sites. The confocal scanning microscopic
(CLSM) software was adjusted to use a multiline argon laser (488-
nm wavelength) for the first channel and a helium-neon (gas) laser
(543-nm wavelength) for the second one. Observations were per-
formed under a 20 objective lens at a resolution of 512  512 pixels.
Ten-micrometer-deep scans (0.5-mm step size, 20 slices/scan) were
taken for each site. The sections were reconstructed to a 3-
dimensional model using FV 10-ASW (V 1.7a, Olympus) software.
The red/green fluorescence (dead/live) volume percentage was calcu-
lated using the color segmentation method for the characterization of
viability and physiological activity of biofilms using BiolmageL software
(V2, developed by Dr. de Paz L, Malmo, Sweden) (10). The calculated
percentages were tabulated and subjected to 1-way analysis of variance
followed by the Tukey post hoc multiple comparison tests. The data
were statistically analyzed using SPSS software (V.20; IBM Corp, Ar-
monk, NY).
JOE Volume 40, Number 12, December 2014
Basic ResearchBiology
Examination of Representative Fractured Specimens
with Scanning Electron Microscopy
Two additional infected dentin discs prepared in the same manner
as mentioned in the previous section were examined under scanning
electron microscopy to ensure the validity of the infection protocol
(ie, the penetration of bacteria into the dentinal tubules). The discs
were fractured, gold sputter coated, and examined with field-
emission scanning electron microscopy (LEO 1530; LEO Elektronen-
mikroskopie GmbH, Oberkochen, Germany) operated in the secondary
electron detection mode with an 8-mm working distance and a 5-kV
accelerating voltage for confirmation of penetration of the bacteria
into the dentinal tubules.
Measuring the pH of the Solutions and Gels
Used in the Study
The pH of the 5% NaOCl solution, the 2% CHX solution, and the
Carisolv and Papacarie gels were measured using a digital pH meter
(CyberScan pH 500; Thermo Fisher Scientific Inc, Waltham, MA).
Results
The common fluorescence patterns after each dentin surface treat-
ment are shown in Figure 1AF. The green regions represent the fluo-
rescence from live bacteria, whereas the red regions represent the
fluorescence from dead bacteria. All the CLSM micrographs of the
noninfected sound discs (ie, the negative control group) showed no
bacteria inside the tubules as well as a faint green dentin background
fluorescence (Fig. 1A). Moreover, CLSM observations revealed that
the vitality of most of the intratubular bacteria was preserved after a
5-minute dentin treatment with Carisolv gel. Conversely, a high percent-
age of dead bacteria was observed after a 5-minute dentin treatment
with Papacarie gel and NaOCl and CHX solutions, indicating that these
solutions exhibited antibacterial activity.
The mean percentages of dead and live bacteria for the different
dentin surface treatments are shown in Table 1. The results of 1-way
analysis of variance and Tukey HSD post hoc multiple comparison tests
showed that the highest antibacterial activity was observed after a
5-minute treatment with 5.25% NaOCl solution (50.1%  16.8%
dead bacteria) followed by Papacarie gel (31.9%  11.9% dead bac-
teria) and 2% CHX solution (29.8%  8.4% dead bacteria). No signif-
icant difference in the mean percentages of dead bacteria was observed
between Papacarie gel and 2% CHX solution (P > .05). Conversely, the
5-minute dentin treatment with Carisolv gel showed the lowest antibac-
terial effect among all the treatments (7.8%  6.3% dead bacteria),
which was not significantly different from the negative control group
(2.3%  2.8% dead bacteria) (P > .05). The antibacterial effects of
all the other groups were significantly different from the negative control
group (P < .05). The SEM micrographs of fractured specimens showed
the presence of S. mutans inside the dentinal tubules. This confirms the
viability of this protocol in the current study for use in coronal dentin
(Fig. 2). The pH levels of 5.25% NaOCl solution, 2% CHX solution, Car-
isolv gel, and Papacarie gel were 12.4, 6.8, 10.1, and 9.2, respectively.
Discussion
Microbial assay for pre- and postexcavation dentin samples (using
culture method) is the most commonly used method for quantifying the
amount of residual bacteria remaining in dentin after caries excavation
(6, 7). The lack of standardization among the different excavation sites
is 1 of the most important limitations of this methodology. Furthermore,
contamination of the specimens may also occur during transfer from
the patients mouth to the culturing medium.
Viability of Intratubular Bacteria 1973
Basic ResearchBiology

Figure 1. Confocal laser scanning microscopy images of (A) noninfected sound dentin (negative control) and (B) infected untreated dentin (positive control).
Also, it shows the viability of intratubular bacteria after 5-minute dentin treatment with (C) 5% NaOCl solution, (D) 2% CHX solution, (E) Carisolv gel, and
(F) Papacarie gel. The highest percentage of the dead intratubular bacteria was observed after treating the dentin with 5% NaOCl solution followed by 2% CHX
solution and Papacarie gel. However, the vitality of most of the intratubular bacteria was preserved after 5-minute dentin treatment with Carisolv chemomechanical
caries removal gel.

The viability of intratubular bacteria in radicular dentin has been fugation method. In the current study, the protocol described by Ma
studied using confocal scanning microscopy, and this seems to be an et al (8) was modified for testing in coronal dentin. The coronal
accepted methodology (11). However, the infection protocols used in dentin was infected with S. mutans (an anaerobic, gram-positive bac-
these early studies were nonstandardized. Recently, Ma et al (8) teria) instead of E. faecalis to simulate the clinical condition for
developed a standardized noninvasive infection model using a centri- caries. The rationale for using S. mutans was based on findings
from previous studies, which identified S. mutans as 1 of the major
contributors of causing dental caries (12). Both S. mutans and E.
TABLE 1. 1-way Analysis of Variance Results Comparing the Viability of the
Intratubular Bacteria among the Different Dentin Treatments Groups* faecalis have similar diameters of 0.70.9 mm (13), which facilitates
their penetration into dentinal tubules (1.52 mm in diameter) (1).
Bacterial activity This was confirmed by scanning electron microscopic observation of
the fractured specimens (Fig. 2). This study aimed to simulate the
Group n % Dead bacteria % Live bacteria SD
presence of residual bacteria in dentin and determine whether the
Control 5 2.3C 97.7c 2.8 chemomechanical caries removal agents (ie, Carisolv and Papacarie)
NaOCl 5 50.1A 49.9a 16.8
CHX 5 29.8B 70.2b 8.4 were able to kill residual bacteria that remain after excavation of
Carisolv 5 7.8C 92.2c 6.3 carious tissue.
Papacarie 5 31.9B 68.1b 11.9 By comparing the CLSM micrographs of the noninfected sound
CHX, chlorhexidine; NaOCl, sodium hypochlorite; SD, standard deviation.
dentin discs (the negative control) with the infected untreated discs
The lowercase letters indicate the multiple comparison results of percentage live bacteria, which
(the positive control) groups, the validity of the centrifugal method
cannot be compared to percentage of dead bacteria; uppercase letters were used for dead group. used in this study to infect the coronal dentin was confirmed
*Values are means  standard deviation. (Fig. 1). Furthermore, the observation of a faint green fluorescence

Groups identified by different superscripts were significantly different at P < .05. of the noninfected sound dentin was in agreement with the previous

Standard deviation of both dead and live bacteria is exactly the same. CLSM observations reported by Banerjee et al (14).

1974 Hamama et al. JOE Volume 40, Number 12, December 2014

Figure 2. High-magnification scanning electron microscopic micrograph of
a representative infected specimen confirming the presence of S. mutans in-
side the dentinal tubules. The hand pointer showed S. mutans in ongoing cell
division. DT, dentinal tubule.
The 5.25% NaOCl solution was selected as a positive control anti-
microbial agent because it contains almost the same concentration of
NaOCl (5%) as in Carisolv gel but without the additional amino acids.
The chlorine reacts with the amino groups of the bacterial cell wall,
leading to the formation of chloramines, which inhibit the bacterial
metabolic activity. Furthermore, chlorine itself (considered a potent
oxidizing agent) oxidizes the sulfhydryl group, which is essential for
the synthesis of bacterial enzymes (15). The potent antibacterial effect
of 5.25% NaOCl solution may also be attributed to its alkalinity (pH =
12.4) (16). The 2% CHX was selected as a second positive control anti-
bacterial agent because it has been reported as the gold standard anti-
bacterial agent for toileting cavities before the application of the
restorative materials (17). Furthermore, the outcomes of recent studies
revealed that a CHX-containing root canal irrigation solution (eg, QMIX;
Dentsply Tulsa Dental, Tulsa OK [2% CHX, EDTA, and surfactant] and
Alexidine; Gentaur, Kampenhout, Belgium [ethyl-end group derivative
of CHX]) showed potent antimicrobial activities (1821). The
antibacterial action of CHX is attributed to its high penetration power
through bacterial membranes, leading to destruction of the bacterial
cell wall and extrusion of the bacterial intracellular components (22).
Although Carisolv gel contains 5% NaOCl, surprisingly, it showed the
least antibacterial activity. This may be because of the viscosity of the gel,
which possibly limits its penetration into the dentinal tubules. Further-
more, the difference in antibacterial activity between Carisolv and
5.25% NaOCl may also be related to the amino acids (glutamic, leucine,
and lysine) in Carisolv gel, which have been added to reduce the reactivity
of the NaOCl against sound dental tissues (23). Moreover, Carisolv gel is
less alkaline (pH = 10.1) when compared with the 5.25% NaOCl solution
(pH = 12.4), which may influence its antibacterial properties.
Papacarie (papain-based) gel showed an antibacterial activity
similar to 2% CHX solution. Although Papacarie gel (pH = 9.2) is
less alkaline than Cariosolv gel (pH = 10.1), it contains the papain
enzyme (similar to human pepsin), which has been considered as a
potent antibacterial and anti-inflammatory agent (24). Papain is widely
used in the medical field as a wound healing dressing (25) and in the
treatment of tumors (26). It destroys peptidoglycans, which cover the
bacterial cell wall, leading to exposure of the cell membrane as a result
of osmotic shock and exudation of the cytoplasmic contents (27).
Furthermore, Papacarie gel is less viscous than Carisolv gel, which
may facilitate its penetration into the dentinal tubules.
JOE Volume 40, Number 12, December 2014
Basic ResearchBiology
Finally, in light of the results of this study, the null hypothesis that
there was no difference in the viability of intratubular bacteria after
application of the 2 chemomechanical caries removal agents was
rejected.
Conclusion
The enzyme-based chemomechanical caries removal gel (Papa-
carie) is effective in the reduction of residual bacteria in the dentinal
tubules of coronal dentin and has a stronger antibacterial effect than
NaOCl-based (Carisolv) gel. This may be useful in reducing the number
of residual bacteria when excavating deep lesions where infected dentin
remains in order to avoid pulp exposure.
Acknowledgments
The authors deny any conflicts of interest related to this study.
References
1. Heymann H, Swift EJ, Ritter AV, Sturdevant CM. Sturdevants art and science of oper-
ative dentistry. In: Boushell LW, Roberson TM, Walter R, eds. Fundamentals of
Tooth Preparation an Pulp Protection, 6th ed. St Louis, Mo: Elsevier/Mosby;
2013:1425.
2. Banerjee A, Kidd EA, Watson TF. In vitro evaluation of five alternative methods of
carious dentine excavation. Caries Res 2000;34:14450.
3. Hamama HH, Yiu CK, Burrow MF, King NM. Chemical, morphological and micro-
hardness changes of dentine after chemomechanical caries removal. Aust Dent J
2013;58:28392.
4. Kidd EA. How clean must a cavity be before restoration? Caries Res 2004;38:
30513.
5. Lager A, Thornqvist E, Ericson D. Cultivatable bacteria in dentine after caries exca-
vation using rose-bur or carisolv. Caries Res 2003;37:20611.
6. Azrak B, Callaway A, Grundheber A, et al. Comparison of the efficacy of chemome-
chanical caries removal (Carisolv) with that of conventional excavation in reducing
the cariogenic flora. Int J Paediatr Dent 2004;14:18291.
7. Subramaniam P, Babu KL, Neeraja G. Comparison of the antimicrobial efficacy of
chemomechanical caries removal (Carisolv) with that of conventional drilling in
reducing cariogenic flora. J Clin Pediatr Dent 2008;32:2159.
8. Ma J, Wang Z, Shen Y, Haapasalo M. A new noninvasive model to study the effective-
ness of dentin disinfection by using confocal laser scanning microscopy. J Endod
2011;37:13805.
9. Wang Z, Shen Y, Ma J, Haapasalo M. The effect of detergents on the antibacterial
activity of disinfecting solutions in dentin. J Endod 2012;38:94853.
10. Chavez de Paz LE. Image analysis software based on color segmentation for charac-
terization of viability and physiological activity of biofilms. Appl Environ Microbiol
2009;75:17349.
11. Nagayoshi M, Kitamura C, Fukuizumi T, et al. Antimicrobial effect of ozonated water
on bacteria invading dentinal tubules. J Endod 2004;30:77881.
12. Ryan KJ, Ray CG, Sherris JC. Sherris Medical Microbiology: An Introduction to In-
fectious Diseases, 4th ed. New York: McGraw-Hill: Medical Publishing Division;
2004:27396.
13. Samaranayake LP. Essential Microbiology for Dentistry, 4th ed. Edinburgh:
Churchill Livingstone Elsevier; 2012.
14. Banerjee A, Sherriff M, Kidd EA, Watson TF. A confocal microscopic study relating
the autofluorescence of carious dentine to its microhardness. Br Dent J 1999;187:
20610.
15. Estrela C, Estrela CR, Barbin EL, et al. Mechanism of action of sodium hypochlorite.
Braz Dent J 2002;13:1137.
16. Tirali RE, Bodur H, Ece G. In vitro antimicrobial activity of sodium hypochlorite,
chlorhexidine gluconate and octenidine dihydrochloride in elimination of microor-
ganisms within dentinal tubules of primary and permanent teeth. Med Oral Patol
Oral Cir Bucal 2012;17:e51722.
17. Lafuente D. SEM analysis of hybrid layer and bonding interface after chlorhexidine
use. Oper Dent 2012;37:17280.
18. Morgental RD, Singh A, Sappal H, et al. Dentin inhibits the antibacterial effect of new
and conventional endodontic irrigants. J Endod 2013;39:40610.
19. Stojicic S, Shen Y, Haapasalo M. Effect of the source of biofilm bacteria, level of bio-
film maturation, and type of disinfecting agent on the susceptibility of biofilm bac-
teria to antibacterial agents. J Endod 2013;39:4737.
20. Agrafioti A, Tzimpoulas NE, Kontakiotis EG. Influence of dentin from the root canal
walls and the pulp chamber floor on the pH of intracanal medicaments. J Endod
2013;39:7013.
Viability of Intratubular Bacteria 1975
Basic ResearchBiology
21. Barrios R, Ferrer-Luque CM, Arias-Moliz MT, et al. Antimicrobial substantivity of
alexidine and chlorhexidine in dentin. J Endod 2013;39:14135.
22. Borges FM, de Melo MA, Lima JP, et al. Antimicrobial effect of chlorhexidine digluc-
onate in dentin: In vitro and in situ study. J Conserv Dent 2012;15:226.
23. Hannig M. Effect of Carisolv solution on sound, demineralized and denatured
dentin-an ultrastructural investigation. Clin Oral Investig 1999;3:1559.
24. Maini S, Rudralingam M, Zeitoun H, Osbourne JE. Aspiration pneumonitis following
papain enzyme treatment for oesophageal meat impaction. J Laryngol Otol 2001;
115:5856.
1976 Hamama et al.
25. Shi L, Ermis R, Kiedaisch B, Carson D. The effect of various wound dress-
ings on the activity of debriding enzymes. Adv Skin Wound Care 2010;23:
45662.
26. Li S, Tang Y. Accurate determination of internalization for target binding anti-
body using papain digestion and flow cytometry. Hybridoma (Larchmt) 2010;
29:1339.
27. Seenivasan R, Roopa L, Geetha S. Investigations on purification, characterization and
antimicrobial activity of enzyme papain from Carica papaya Linn. J Pharm Res 2010;
3:10925.
JOE Volume 40, Number 12, December 2014