Clinical Research
Prevalence and Characteristics of the Maxillary
C-shaped Molar
Jorge N.R. Martins, DDS, MSc,* Ant

onio Mata, DMD, PhD, FICD,jj Duarte Marques, DDS, PhD,jj
#
Craig Anderson, DDS, and Jo~ ao Caram^es, DDS, PhD, FICD**
Abstract
Introduction: The anatomy of the maxillary C-shaped
molar has been the subject of several case reports
although no true prevalence research has ever been
conducted. The aim of this observational study was to
evaluate and characterize the incidence of these mor-
phologies using cone-beam computed tomographic
(CBCT) technology. Methods: Patients having presurgi-
cal CBCT examination were selected. CBCT analysis
was performed at 5 different axial levels, and the
teeth were classified as maxillary molar C shapes ac-
cording to the inclusion criteria for the present inves-
tigation. The prevalence of C-shaped anatomy was
calculated. The z test for proportions in independent
groups was used to analyze the differences between
proportions. Intrarater reliability was also tested.
Results: A total of 2227 teeth (928 upper first molars
and 1299 upper second molars) from 895 patients
were included in this research. Five different types of
C-shaped molars were found, which included fusion be-
tween mesiobuccal and palatal roots (type A), mesio-
buccal and distobuccal roots (type B), distobuccal and
palatal roots (type C), 2 possible palatal roots (type
D), and mesiobuccal and palatal roots connecting with
an independent distobuccal root canal at apical or be-
tween distobuccal and palatal roots connecting with
an independent mesiobuccal root canal at apical (type
E). The global prevalence was 1.1% for first molars
and 3.8% for second molars. Differences were observed
between sex, teeth, and some types of C shapes at
P < .05. Conclusions: Maxillary C-shaped molars
have low prevalence but high anatomic complexity. An
understanding of their anatomic configuration and var-
iations is important in ensuring that they are treated
properly. Distinct differences exist between C-shaped
configurations of upper and lower molars. (J Endod
2016;-:17)
From the Departments of Oral Biology, Anatomy and Physiology, and **Implantology, *School of Dental Medicine, and jjOral Biology and Biochemistry Group,
Biomedical and Oral Sciences Research Unit (FCT Unit 4062), Evidence Based Dentistry Center, Lisbon University, Lisbon, Portugal; Private Practice, Lisbon, Portugal;
Department of Biology, New York University, New York, New York; and #Private Practice, Biloxi, Mississippi.
Address requests for reprints to Dr Jorge N.R. Martins, Instituto de Implantologia of Lisbon, Av Columbano Bordalo Pinheiro, 505 e 6 , 1070-064, Lisbon, Portugal.
E-mail address: jnr_martins@yahoo.com.br
0099-2399/$ - see front matter
Copyright 2016 American Association of Endodontists.
http://dx.doi.org/10.1016/j.joen.2015.12.013
JOE Volume -, Number -, - 2016
Key Words
Cone-beam computed tomography, C-shaped root canal, dental morphology, maxillary
molars
F or infection to occur, multiplying bacteria must be present in the root canal system;
these bacteria acquire nutrients and release mediators that can initiate a cascade of
events leading to inflammation and tissue damage (1). The main purpose of the end-
odontic treatment is the cleaning and disinfection of this root canal system (2). Knowl-
edge of the most common root canal configuration as well as their anatomic deviations
allows for more effective cleaning and shaping, which have been associated with higher
success rates (3). This is very important in cases that usually require modifications of
the usual instrumentation, irrigation, and obturation techniques and is especially true
for C-shaped configurations.
C-shaped anatomy has been well documented for several types of teeth. It is most
common in mandibular second molars, with a prevalence of 44.6% in certain Asian
populations (4). These types of morphologies are uncommon in upper molars. The first
authors to describe a maxillary C-shaped molar were Newton and McDonald (5), who,
in the year 1984, described a C-shaped canal appearance connecting both distobuccal
and palatal root canals linked in the pulp chamber floor to the mesiobuccal root canal
by developmental grooves. A few years later, Danker et al (6) and De Moor (7)
confirmed this anatomic condition. In 2006, Yilmaz et al (8) reported a C shape be-
tween the mesiobuccal and distobuccal roots, and, a few years later, Kottoor et al
(9) described a maxillary C-shaped configuration on the palatal root. Several other au-
thors have reported similar clinical conditions (10, 11). There are no true prevalence
studies available regarding the maxillary C-shaped molars. Because of a lack of
information, several studies (11, 12) use the finding reported by De Moor (7) as a
kind of prevalence. De Moor has treated 2 C-shaped teeth in 2175 root canal therapies
performed in the maxillary first molars (0.09%). This information has been used incor-
rectly as the prevalence, and this author has only classified the fusion of the distobuccal
and palatal roots as a maxillary C shape.
Because of the different anatomic characteristics and the difference in the number
of roots between the upper and lower molars, the C-shaped configurations must obvi-
ously be different. As a result, Meltons classification of mandibular C-shaped molars
(13) cannot be applied to the maxillary molars. The definition of a maxillary C-shaped
molar is unclear and lacks standardization.
The purpose of this research was to understand the characteristics of the maxillary
C-shaped molar and investigate its prevalence among a Portuguese population using
cone-beam computed tomographic (CBCT) analysis.
Maxillary C-shaped Molar 1
Clinical Research
Figure 1. UC configuration system used to classify the maxillary molar C-shaped anatomy.
Materials and Methods
CBCT scans performed between May 2011 and May 2014
were collected from the database in the radiology department of
a health center in Lisbon, Portugal. All examinations were per-
formed for diagnostic purposes before implant surgery and
were analyzed retrospectively by a single observer from May
2014 to August 2014.
The criterion for inclusion was the presence of at least 1
maxillary molar in the clinical examinations. All teeth that had
previous endodontic treatments or had artifact images that could
prevent a correct analysis were excluded. Artifacts might have
come from full crown restorations, adjacent implants, or incorrect
radiologist technique.
CBCT examinations were performed using a Planmeca scanner
(Planmeca Promax, Planmeca, Finland) following the instructions
of the manufacturer. All images were obtained at a 0.20-mm voxel
size, 80 kV, and 15 mA with a 12-second exposure time. The images
were analyzed using proper computer visualization software (Plan-
meca Romexis, Planmeca). Cross-sectional images were recon-
structed to be a 0.6-mm-thick slice, and the filter of noise and
artifact reduction was applied.
Maxillary Upper-C Classication System
No classification of the maxillary C-shaped canals is available in the
literature. Therefore, a classification of the lower C-shaped molars (14)
was modified for the purpose of this research in order to fit in the upper
molars. The root canal system was classified as a maxillary C shape as
Figure 2. Classification of the maxillary C-shaped molar according to the position of the fused root canal.
2 Martins et al.
long as they had 2 criteria points: root fusion and 3 consecutive axial
cross sections with an upper-C (UC) 1 or UC2 configuration (see later)
in the fused root.
The UC configuration system (Fig. 1) is a modification for the
upper molars based on Fan et als (14) study regarding the lower
molars. This UC system was established after a preanalysis of 223
upper molars. The UC system has 5 axial root canal system con-
figurations:
1. UC1: Continuous large C-shaped canal system
2. UC2: Continuous C-shaped canal with 2 main canal lumen in the ex-
tremities connected by a large isthmus
3. UC3: 2 separated root canals
4. UC4: A single round or oval root canal
5. UC5: No canal lumen
The axial cross-sectional images were observed at 5 levels of the
root canal system as follows: 2 mm under the canal orifice (coronal),
2 mm above the anatomic apex (apical), middle distance from the
canal orifice and anatomic apex (middle), middle distance between
coronal and middle (one third), and middle distance between
the middle and apical (two thirds) (Fig. 1). The decision of
requiring a UC1 or UC2 at 3 consecutive levels was made to avoid clas-
sifying a simple isthmus as C shaped.
Statistical Analysis
The collected data were introduced in the SPSS software (IBM
SPSS Statistics Version 22; IBM, Armonk, NY), from which absolute
JOE Volume -, Number -, - 2016
 Clinical Research

Figure 3. Examples of each type of maxillary molar C-shape configuration (axial sections).

JOE Volume -, Number -, - 2016 Maxillary C-shaped Molar 3

Clinical Research
Figure 4. Examples of maxillary C-shaped molars viewed from several angles.
counts and percentages for the analyzed groups and subgroups of C
shapes were extracted. The proportion of each group was calcu-
lated, as was the range for the true population proportion, to a con-
fidence level of 95%. The z test for proportions in independent
groups was used to analyze the differences in maxillary C-shaped
molar proportions between sexes, location (left and right side),
and tooth (first or second upper molars). A P value <.05 was
considered significant. Intraobserver reliability was calculated using
the Cohen kappa test.
4 Martins et al.
Results
To determine intrarater reliability, a single observer performed the
evaluation of 223 teeth twice with a 1-month distance between observa-
tions. The kappa coefficient of agreement between both evaluations was
74.7%, with an asymptotic standard error of 17.2%.
A total of 895 CBCT scans from 895 patients (310 men and 585
women) with a mean age of 52 years were included in this study.
From these examinations, it was possible to collect information on
2227 teeth (928 first upper molars and 1299 second upper molars).
JOE Volume -, Number -, - 2016
 Clinical Research
Fifty-nine C-shaped configurations were found in the analyzed TABLE 2. Absolute Counts and Percentages of C Shapes between Sides (all
sample. It was possible to classify 5 different types of maxillary C-shaped teeth combined)
configurations (types A, B, C, D, and E) depending on which roots were C-shaped configuration
fused (Figs. 24):
Not present Present Total
1. Type A: Fusion between the mesiobuccal and palatal roots, forming a Side
semilunar mesiopalatal root canal Right
2. Type B: Fusion between the mesiobuccal and distobuccal root ca- Count 1114 25 1139
nals, forming a semilunar buccal root canal system; the concavity % within right side 97.8 2.2 100.0
of the semilunar shape may be turned to the palatal (subtype B1) Left
Count 1054 34 1088
or buccal (subtype B2) root % within left side 96.9 3.1 100.0
3. Type C: Fusion between the distobuccal and palatal roots, forming a Total
large semilunar distopalatal root canal Count 2168 59 2227
4. Type D: Presence of a large palatal root canal, forming a semilunar % within side 97.4 2.6 100.0
shape; this type has been previously described as a fusion between 2
palatal roots (9) Prevalence of C Shapes According to Tooth
5. Type E: Fusion between the 3 roots; this configuration resembles the From the 928 first molars, 10 of them presented C-shaped
mandibular C-shaped anatomy, with a large semilunar mesiopalatal configurations, which represents a prevalence of 1.1% (95% CI,
canal merging together with an independent distobuccal canal at a 0.43%1.77%); from the 1299 second molars, 49 had C shapes,
single apical foramen (subtype E1) or with a large semilunar disto- which represents a prevalence of 3.8% (95% CI, 2.76%4.84%)
palatal canal connecting with a mesiobuccal canal at a single apical (Table 3). The difference between the teeth was considered signif-
foramen (subtype E2) icant at P < .05.

Prevalence of C Shapes According to Sex Prevalence of C Shapes According to Type

Of a total of 59 C shapes found, 8 were in men (n = 774) with a
prevalence of 1.0% (95% confidence interval [CI], 0.3%1.7%), and
51 were in women (n = 1453) with a prevalence of 3.5% (95% CI,
2.56%4.44%) (Table 1). The difference between the groups was
considered significant at P < .05.
Prevalence of C Shapes According to Location
(Left and Right Side)
Twenty-five cases were from the right side (n = 1139) with a prev-
alence of 2.2% (95% CI, 1.35%3.05%), and 34 were from the left side
(n = 1088) with a prevalence of 3.1% (95% CI, 2.07%4.13%)
(Table 2). The difference between groups was considered nonsignifi-
cant at P < .05.
of C Shape and Tooth
Regarding the 5 different types of maxillary C shapes, type A had a
prevalence of 0.1% (95% CI, 0%0.3%) for the first molars and 0.5%
(95% CI, 0.12%0.88%) for the second molars. Type B had a preva-
lence of 0.1% (95% CI, 0%0.3%) for the first molars and 2.1%
(95% CI, 1.32%2.88%) for the second molars. For type C, the prev-
alence in the first molars was higher than the prevalence in the
second molars, 0.9% (95% CI, 0.29%1.51%) in the first molars
and 0.2% (95% CI, 0%0.44%) in the second molars. Both types D
and E had no expression on the first molars. The prevalence for the
second molar was 0.1% (95% CI, 0%0.27%) for type D and 1.0%
(95% CI, 0.46%1.54%) for type E (Table 4). Except for types A and
D, all the other types showed a difference between teeth considered sig-
nificant at P < .05.

TABLE 1. Absolute Counts and Percentages of C Shapes between Sexes (all

teeth combined) TABLE 3. Absolute Counts and Percentages of C Shapes between Teeth
C-shaped C-shaped
configuration configuration
Not Not
present Present Total present Present Total
Sex Teeth
Male First molar
Count 766 8 774 Count 918 10 928
% within male sex 99.0 1.0 100.0 % within first molar teeth 98.9 1.1 100.0
% within C-shaped teeth 35.3 13.6 34.8 % within C-shaped teeth 42.3 16.9 41.7
% within all teeth combined 34.4 0.4 34.8 % within all teeth combined 41.2 0.4 41.7
Female Second molar
Count 1402 51 1453 Count 1250 49 1299
% within female sex 96.5 3.5 100.0 % within second molar teeth 96.2 3.8 100.0
% within C-shaped teeth 64.7 86.4 65.2 % within C-shaped teeth 57.7 83.1 58.3
% within all teeth combined 63.0 2.3 65.2 % within all teeth combined 56.1 2.2 58.3
Total Total
Count 2168 59 2227 Count 2168 59 2227
% within sex 97.4 2.6 100.0 % within teeth 97.4 2.6 100.0
% within C-shaped teeth 100.0 100.0 100.0 % within C-shaped teeth 100.0 100.0 100.0
% within all teeth combined 97.4 2.6 100.0 % within all teeth combined 97.4 2.6 100.0

JOE Volume -, Number -, - 2016 Maxillary C-shaped Molar 5

Clinical Research
TABLE 4. Absolute Counts and Percentages of C Shapes According to Type of C Shape and Tooth
C-shaped configuration
Not present Type A Type B1
Teeth
First molar
Count 918 1 1
% within teeth 98.9 0.1 0.1
Second molar
Count 1250 6 17
% within teeth 96.2 0.5 1.3
Total
Count 2168 7 18
% within teeth 97.4 0.3 0.8
UC Root Canal Conguration
Overall, the UC1 configuration was the most common at all levels,
excluding the apical level where UC4 was the most common. UC2 anat-
omy occurs more regularly in the middle section, whereas UC3 appears
more at the apical level. The type A configuration appears to be charac-
terized as having a high percentage of UC2 configurations at several
levels, whereas type B anatomy has a higher percentage of UC1. Types
C and E, although there is a higher percentage of UC1 configurations,
have a more balanced distribution between UC1 and UC2 when
compared with types A and B.
Discussion
As discussed in a previous study (10), the fusion between the roots
in a maxillary C-shaped molar may not be a true fusion. Instead, it may
be described as the nondivision of the roots caused by inadequate devel-
opment during the embryologic phase of root formation (15). Because
of the morphologic characteristics of the upper molar, this fusion be-
tween roots may occur in several ways depending on which roots are
involved. Zhang et al (16) performed a microcomputed tomographic
analysis of 187 extracted second upper molars and were able to identify
79 teeth with 6 different types of root fusion. From the original 79 fused
roots, 22 had partial canal merging, and 6 had complete canal merging.
Of the 28 complete or partially merged canals, 16 were between the me-
siobuccal and distobuccal canals, 3 were between the distobuccal and
palatal canals, and 9 showed merging between the 3 root canals.
The terminology maxillary C-shaped molar is usually used to
describe cases with a large semilunar root canal shape that may repre-
sent a complete or partial root canal merging between 2 or more root
canals (510).
The purpose of our research was to investigate the maxillary C-
shaped molar prevalence and study its characteristics. The choice of
an appropriate sample size is crucial in any study design. An increase
in the sample size reduces the error and the range of the CI of the
analyzed proportion, leading to a higher precision and study power.
Regarding the maxillary C-shape prevalence, no study is available,
and the only reasonable data come from the 0.09% reported from
De Moor (7). Therefore, it was easy to understand that a very large num-
ber of cases were required to perform any kind of analysis. The CBCT
scan observation appears to be a good option in performing these kinds
of large sample size in vivo prevalence studies. In this study, a 0.20-mm
voxel size was used, which is a size reported as having the ability to
analyze this type of fusion on other teeth (17, 18).
In our investigation, we included a sample of 2227 maxillary mo-
lars, and we were able to find 59 C-shaped morphologies. It was
possible to detect 5 different root canal fusions depending on which
type of canals were merging (completely or partially).
6 Martins et al.
Type B2 Type C Type D Type E1 Type E2 Total
0 8 0 0 0 928
0.0 0.9 0.0 0.0 0.0 100.0
10 2 1 5 8 1299
0.8 0.2 0.1 0.4 0.6 100.0
10 10 1 5 8 2227
0.4 0.4 0.0 0.2 0.4 100.0
Type A had a prevalence of 0.1% for the first molars and 0.5% for
the second molars, and it is characterized by linking the main mesio-
buccal and palatal canals with a large and deep isthmus, which
represented in our sample, on the average, an extension of 56%
through the entire canal. Although this deep isthmus may go clinically
unnoticed, it may have clinical impact because it may very easily har-
bor microorganisms that may be related to unsuccessful treatment.
Regarding type B, it had a prevalence of 0.1% for first molars and
2.1% for second molars. Excluding some more elongated and angu-
lated B2 types, the majority of the cases tend to move from small semi-
lunar buccal canal systems in the coronal and middle sections to an
oval or round canal configuration in the more apical sections. The
most technically challenging cases might be types C, D, and E. Type
C had a prevalence of 0.9% for first molars and 0.2% for second mo-
lars. Types D and E were not found on the first molars in our sample
and had a prevalence of 0.1% and 1.0% in second molars, respec-
tively. The global prevalence of maxillary C shapes is 1.1% in first
molars and 3.8% in second molars. These root canal configurations
appear to be more common in females and in maxillary
second molars and appear not to be related with either the left or right
side. Type C is more common in maxillary first molars, and the other 4
types are more common in second molars.
The major difference between the upper and lower C-shaped mo-
lars comes from the fact that both teeth have a different number of roots,
which opens to a wider variety of C-shaped types in the upper molars
because of the higher number of combinations of root fusions that
might be made. In the upper molar, the semilunar root canal system
may appear when 2 or 3 roots are fused, whereas in the lower molar
both roots have to be fused.
One limitation of our findings is the low prevalence ratio of this
anatomy. These results came from a population of Western Europe (in
our sample, the whites were the vast majority), which is considered a
geographic location with lower prevalence rates of C-shaped config-
urations when compared with populations from the Middle East and
Asia where the impact of the maxillary C-shaped molar morphologies
is still unknown. The low prevalence makes the study of these mor-
phologies more difficult; we were able to perform a large sample
study (without violating the radiation exposure safety of a large num-
ber of patients) by accessing an already existing CBCT examination
database. These maxillary C-shaped molars may be of low prevalence,
but they are of high anatomic complexity because of large isthmus and
root canal systems connecting root canals that were expected to be
separated. The clinical pulp chamber floor may be completely
different from the usual configuration, which may lead to misinterpre-
tation of the anatomy (Fig. 5AD). Clinically, these cases may require
several technical adjustments in order for the treatments to be
successful.
JOE Volume -, Number -, - 2016

Figure 5. Pulp chamber floor clinical view from several types of upper C-shaped configurations. Note the difference from the usual anatomy. (A) Type A, take note
of the large isthmus connecting mesiobuccal and palatal canals and penetrating deep inside both root canals. (B) Type C, note the large distobuccal-palatal root
canal orifice. (C) Type B, take note of the semilunar buccal root canal orifice. (D) Type E.
Conclusions
The prevalence of the maxillary molar C-shaped configuration is
an uncommon occurrence for the maxillary molars, which presents a
proportion of 1.1% for the first molar and 3.8% for the second one.
There is a higher prevalence in females. Five different types have
been found depending on the position of the fused canal system.
Recommendation for Future Research
As a recommendation for future research, it would be interesting
to analyze these types of configurations with a high-resolution 3-dimen-
sional technology, as is the case of microcomputed tomographic im-
aging, to increase the knowledge of these morphologies. It would also
be interesting to have similar future prevalence studies in Asiatic pop-
ulations in order to understand if these configurations are also higher
among them as happens with the lower molars.
Acknowledgments
The authors deny any conflicts of interest related to this study.
References
1. Sundqvist G, Figdor D. Life as an endodontic pathogen: ecological differences be-
tween the untreated and root-filled root canals. Endod Topics 2003;6:328
2. Friedman S. Prognosis of initial endodontic therapy. Endod Topics 2002;2:5988.
3. Wolcott J, Ishley D, Kennedy W, et al. A 5 yr clinical investigation of second mesio-
buccal canals in endodontically treated and retreated maxillary molars. J Endod
2005;31:2624.
4. Jin G, Lee S, Roh B. Anatomical study of C-shaped canals in mandibular second mo-
lars by analysis of computed tomography. J Endod 2006;32:103.
JOE Volume -, Number -, - 2016
Clinical Research
5. Newton CW, McDonald S. A C-shaped canal configuration in a maxillary first molar.
J Endod 1984;10:3979.
6. Dankner E, Friedman S, Stabholz A. Bilateral C Shape configuration in maxillary first
molars. J Endod 1990;16:6013.
7. De Moor R. C-shaped root canal configuration in maxillary first molar. Int Endod J
2002;35:2008.
8. Yilmaz Z, Tuncel B, Serper A, Calt S. C-shaped root canal in a maxillary first molar: a
case report. Int Endod J 2006;39:1626.
9. Kottoor J, Velmurugan N, Ballal S, Roy A. Four-rooted maxillary fisrt molar having C-
shaped palatal root canal morphology evaluated using cone-beam computerized to-
mography: a case report. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2011;
111:e415.
10. Martins JNR, Quaresma S, Quaresma MC, Frisbie-Teel J. C-Shaped maxillary per-
manent first molara case report and literature review. J Endod 2013;39:
164953.
11. Karanxha L, Kim HL, Hong SO, et al. Endodontic management of a C-shaped maxil-
lary first molar with three independent buccal root canals by using cone-beam
computed tomography. Restor Dent Endod 2012;37:1759.
12. Cleghorn BM, Christie WH, Dong C. Root and root canal morphology of the human
permanent maxillary first molar: a literature review. J Endod 2006;32:81321.
13. Melton D, Krell K, Fuller M. Anatomical and histological features of C-shaped canals
in mandibular second molars. J Endod 1991;17:3848.
14. Fan B, Cheung GSP, Fan M, et al. C-shaped canal system in mandibular second mo-
lars: part Ianatomical features. J Endod 2004;30:899903.
15. Castellucci A. Embryology. In: Castellucci A, ed. Endodontics, 1st ed, vol I. Florence,
Italy: Tridente; 2004:623.
16. Zhang Q, Chen H, Fan B, et al. Root and root canal morphologies in maxillary sec-
ond molar with fused root from a native Chinese population. J Endod 2014;40:
8715.
17. Helvacioglu-Yigit D, Sinanoglu A. Use of cone-beam computed tomography to eval-
uate C-shaped root canal systems in mandibular second molars in a Turkish sub-
population: a retrospective study. Int Endod J 2013;46:10328.
18. Silva EJNL, Nejaim Y, Silva AV, et al. Evaluation of root canal configuration of
mandibular molars in a Brazilian population by using cone-beam computed tomog-
raphy: an in vivo study. J Endod 2013;39:84952.
Maxillary C-shaped Molar 7