Clinical Research

Rapid Method for the Detection of Root Canal Bacteria

in Endodontic Therapy
Kai Soo Tan, PhD,* Victoria Soo Hoon Yu, BDS, MSc,* Samantha Yiling Quah, MSc,*
and Gunnar Bergenholtz, LDS, Dr Odont
Abstract
Introduction: Complete eradication of microorganisms
is essential for successful root canal therapy. However,
current methods to evaluate persistent bacteria after
therapy are not widely practiced. Adenosine triphosphate (ATP) is an indicator of viable cells. The
bioluminescence-based ATP assay is easy to perform,
and results can be obtained in a clinically relevant
time frame of 5 minutes. The aims of this study were
to evaluate the sensitivity of the ATP detection method
and the specificity of this assay for viable cells and to
compare the ATP and culture methods from root canal
samples of patients undergoing endodontic treatment.
Methods: The sensitivity of the ATP assay was determined using bacterial species commonly isolated from
root canals. Bacteria were treated with sodium hypochlorite; after which, culture plating and the ATP assay
were performed. Forty-three root canal samples before
(S1) and after (S2) instrumentation and 36 samples after
the removal of calcium hydroxide dressing (S3) were
collected from patients undergoing root canal treatment
and subjected to ATP assay and anaerobic culture.
Results: The sensitivity of the ATP assay was determined to be between 10 and 100 bacterial cells. This
method of detection also correlated well with the presence of viable bacteria. The ATP readings obtained
allowed clear segregation of anaerobic culture-positive
and -negative samples obtained from infected root
canals of patients. Conclusions: The ATP detection
method can be used as a rapid tool to determine the
presence of viable bacteria during root canal therapy.
This method may be potentially useful as an adjunct
to root canal treatment. (J Endod 2015;-:14)

Key Words
Adenosine triphosphate, bacterial culture, rapid detection, root canal infection

From the *Faculty of Dentistry, National University of

Singapore, Singapore; and The Sahlgrenska Academy University of Gothenburg, Gothenburg, Sweden.
Address requests for reprints to Dr Kai Soo Tan, Faculty of
Dentistry, National University of Singapore, 11 Lower Kent
Ridge Road, Singapore 119260. E-mail address: denkst@nus.
edu.sg
0099-2399/$ - see front matter
Copyright 2015 American Association of Endodontists.
http://dx.doi.org/10.1016/j.joen.2014.11.025

JOE Volume -, Number -, - 2015

he goal of root canal therapy (RCT) is to eliminate the pathogenic effects of bacteria
in the root canal system of teeth. Treatment approaches include chemomechanical
cleaning and filling to prevent microorganisms from infecting or reinfecting root canals
and the periradicular tissues after treatment. The complete removal of bacteria in root
canals is associated with high success rates, whereas substantial amounts of bacteria
remaining after treatment are the major cause of failure (1, 2).
Single-visit RCT has recently become increasingly popular in endodontics
because it is time and cost saving for both the patient and the operator. Single-visit
RCT has also been associated with reduced postoperative pain, flare-up rates, and
risk of interappointment infections (3, 4). Yet, the complete eradication of bacteria
may not be possible in a single-visit treatment because of difficulties in reaching all
sites in the canal (5, 6). Therefore, microorganisms lodged in uninstrumented
areas may provide opportunity for failure (7). In contrast, multiple-visit RCT involves
the placement of an antibacterial medicament such as calcium hydroxide with the aim
of further reducing the bacterial load before permanent root canal filling (8).
Although multiple-visit treatment should logically lead to better healing results, recent
clinical studies comparing these 2 treatment regimens have yielded contradictory findings (9).
The culturing technique, based on samples from treated root canals, was a common method in the past to analyze the extent bacteria persisted in the root canal system
after treatment. Although commonly taught and practiced in dental schools, the method
never gained wide acceptance in the general practice of dentistry (10). An apparent
reason was that culture methods are laborious to conduct, and it takes several days
to weeks to identify anaerobic bacterial species. Furthermore, many species are not
cultivable under laboratory conditions. On the other hand, culture-independent
methods may identify these organisms. Yet, DNA-based identification methods such
as polymerase chain reaction (PCR) suffer from high false-positive readings by the
detection of DNA from dead bacterial cells (11, 12). Recently, the detection of RNA
by reverse-transcriptase PCR has been assumed to be a better alternative to DNA to measure viable bacteria (13). It is argued that RNAs are more labile and possess a shorter
half-life than DNA, thus providing a better indicator of viable cells (14). However, the
detection of viable bacteria by the reverse-transcriptase PCR method is also cumbersome and requires multiple processing steps as well as substantial laboratory instrumentation and setup.
Adenosine triphosphate (ATP) is the main energy source for cellular functions
in living organisms and serves as an indicator of metabolic activity in viable cells.
ATP has been used to estimate the amount of biomass present in groundwater,
drinking water, and biofilms (15, 16). The ATP method has further been
developed to determine the contamination of enteropathogenic bacteria in food
samples and for a sterility check for decontamination of medical devices (17,
18). More recently, the ATP assay was used to quantify oral bacteria in plaque
samples (19). The potential of the ATP method in clinical endodontics to check
persisting bacterial organisms in conjunction with treatment has not been explored.
The procedure is simple to perform, and its results can be obtained in a clinically
relevant time frame of less than 5 minutes. The assay is also sensitive and capable of
detecting as few as 10 bacterial cells (20). Hence, if clinically feasible, the ATP
assay might serve as a valuable adjunct to RCT to control the bacterial status of
treated root canals.

Detection of Root Canal Bacteria

Clinical Research
The specific aims of this study were as follows:
1. To determine the sensitivity of the ATP assay to bacterial species
commonly found in endodontic infections as a proxy for the clinical
presence of bacteria
2. To determine specificity of the ATP assay for viable bacteria
3. To compare ATP and culture methods for the determination of viable
bacteria from root canal samples of patients undergoing endodontic
treatment

Material and Methods

Sensitivity of the ATP Assay
Streptococcus mutans UA159, Streptococcus sanguinis ATCC
10556, Enterococcus faecalis ATCC 29212, Porphyromonas gingivalis W50, Fusobacterium nucleatum ATCC 25586, and Prevotella
intermedia ATCC 25611 obtained from the American Type Culture
Collection (Manassas, VA) were used as an estimate of the sensitivity
of the ATP assay when used for bacterial detection of infected root canal
samples. P. gingivalis, F. nucleatum, and P. intermedia were cultured
anaerobically at 37 C on sheep blood agar plates in an anaerobic
chamber supplemented with 80% N2, 10% H2, and 10% CO2. S. oralis,
S. mutans, S. sanguinis, and E. faecalis were cultured on brain-heart
infusion agar and incubated at 37 C in an incubator supplemented with
5% CO2. The amount of bacteria was determined by serial dilution and
plating on agar plates. Serially diluted microorganisms were subjected
to the ATP assay using BacTiter Glow reagent (Promega, Madison, WI)
according to the manufacturers instructions. Briefly, 100 mL bacterial
suspension was added to 100 mL BacTiter Glow reagent in a 96-well
white plate (Greiner, Monroe, NC) followed by incubation at room temperature at 5 minutes. The luminescence produced was measured with
a luminometer (GloMax, Promega).
Specicity of the ATP Assay for Viable Bacteria
E. faecalis was treated with freshly prepared 1% NaOCl for 2.5, 5,
or 10 minutes; after which, NaOCl was neutralized with sodium thiosulfate to a final concentration of 5%. Bacterial cell viability was determined by culture as described previously. The ATP assay was
performed on an aliquot of the treated sample as described earlier.
Patient Recruitment and Selection Criteria
Ethics approval was obtained from the Domain Specific Review
Board, National Healthcare Group. Patients referred for endodontic treatment at the Faculty of Dentistry graduate clinic, National University Hospital were invited to participate in the study after obtaining written and
informed consent. Inclusion criteria included single-rooted or multirooted teeth with pulp necrosis and apical periodontitis based on clinical
and radiographic findings. Teeth with previously initiated therapy were
accepted if root canal instrumentation was not performed or was incomplete. Teeth with severely broken down coronal tooth structures that
could jeopardize leakage-free sampling conditions, teeth with prior
root canal fillings, teeth with root canals filled with calcium hydroxide,
and teeth presenting with vital or inflamed pulp tissues upon access
were excluded. Patients who had taken antibiotics 4 weeks before sample
collection were excluded from the study as well. A total of 50 S1 samples
were collected; 43 samples fulfilled sterility control requirements.
Clinical Sampling Procedures
Root canal samples included a sterility control before endodontic
treatment, bacterial sampling before (S1) and after (S2) root canal
instrumentation, and bacterial sampling after removal of the calcium hy2

Tan et al.

droxide dressing and before root canal obturation (S3). Plaque, calculus, caries, and defective restorations were removed, and the tooth was
isolated using a rubber dam appropriately secured to ensure leakagefree sampling conditions. After isolation, the tooth was scrubbed with
30% hydrogen peroxide followed by 5% iodine tincture according to
the method described by Moller (21). Disinfectants were deactivated
with 5% sodium thiosulfate, and a sterility control sample was obtained
by scrubbing a sterile cotton pellet on the occlusal surface and transferring it to a vial with thioglycollate broth. To collect S1, canals were
widened with K-files (Flexofiles; Dentsply, York, PA) in the presence
of reduced transport fluid (RTF) to release biofilms on dentinal walls
and to accommodate sampling using paper points (Henry Schein, Melville, NY). Five successive paper points were placed in the canal and
transferred to an RTF vial. Samples were placed on ice until collection
for laboratory processing. Complete root canal instrumentation was
performed using nickel-titanium rotary instruments (ProTaper [Dentsply] or RaCe [FKG Dentaire SA, La Chaux-de-Fonds, Switzerland]) to a
minimum apical size 30 and 1.25% sodium hypochlorite (NaOCl)
(Clorox; The Clorox Company, Oakland, CA) with 2 mL for irrigation
(Monoject Endodontic Irrigation Needles; Kendall Healthcare, Mansfield, MA) per change of instrument. At the end of instrumentation,
1 mL 5% sodium thiosulfate deactivated NaOCl before S2 collection using the same protocol described for S1 collection. In addition, canal
walls were scraped with files to release dentinal shavings by the use
of a sterile K-type file of an ISO size 1 step larger than that of the master
apical file and inserted to the working length. The instrument was
rotated according to the balanced force technique. The apical 5 mm
was cut off with a sterile cutter and transferred to a second RTF vial ac
cording to the technique designed by Orstavik
el al (22). The prepared
and dried canal was filled with nonsetting calcium hydroxide paste (UltraCal XS; Ultradent Products Inc, South Jordan, UT) and the tooth
sealed (IRM Intermediate Restorative Material; Dentsply Caulk, Milford,
DE). After an interappointment period of at least 7 days, the tooth was
isolated as previously described, a sterility control sample was
collected, and the root canal was reaccessed. Calcium hydroxide was
removed using ultrasonically activated sterile saline. Complete removal
of calcium hydroxide was confirmed visually with the aid of a dental
operating microscope (OPMI pico; Carl Zeiss Meditec AG, Oberkochen,
Germany). The S3 sample comprising the apical 5 mm of the last K-file
and 5 paper points was collected in a manner similar to that for S2.

Laboratory Procedures
All samples were transferred to the laboratory on ice for immediate
processing within 1 hour. Sterility control samples were incubated anaerobically at 37 C for 7 days. Tubes containing S1, S2, and S3 were vortexed
at medium speed for 5 minutes to dislodge microorganisms adhering to
the paper points and files. One hundred microliters of the sample were
serially diluted using prereduced brain-heart infusion broth and plated
on prereduced 5% sheep blood agar and incubated anaerobically in an
atmosphere of 80% N2, 10%H2, and 10% CO2 for 7 days. The ATP assay
was performed on 100 mL of each sample in triplicate as described earlier.
Analysis of Data
Results obtained were presented as the mean  standard deviation. R2 values were calculated using GraphPad Prism software, version
6.0 (GraphPad Software, San Diego, CA).

Results
Sensitivity of ATP Assay
The sensitivity of the ATP assay was dependent on several factors
such as efficiency of extraction of ATP from the target organism and
JOE Volume -, Number -, - 2015

Clinical Research
the physiological state of the organism. Reference bacterial strains of 6
commonly found bacterial species in endodontic infections (ie,
S. sanguinis, S. mutans, E. faecalis, F. nucleatum, P. gingivalis,
and P. intermedia) were used as a proxy to determine the sensitivity
of the ATP method when used to detect bacteria from infected root canal
samples. As shown in Figure 1, the minimum amounts of bacteria
required to achieve a signal higher than the background control
were approximately 10 cells for the facultative anaerobes, S. sanguinis,
S. mutans, E. faecalis, and 100 cells for the obligate anaerobes, F. nucleatum, P. gingivalis, and P. intermedia. The luminescent signals,
which were proportional to the amounts of ATP present, showed
good correlations with the number of bacterial cells (R2 = 0.99)
over a wide range for all the bacterial species tested.

Specicity of ATP Assay for Viable Cells

To determine the specificity of the ATP method for viable bacteria,
E. faecalis was exposed to 1% NaOCl for the duration indicated in
Figure 2. Good correlations were obtained between the ATP readings
and bacterial viability as determined by culture (R2 = 0.99).
Comparison Between ATP and Culture
Results of Root Canal Samples
All 43 S1 samples (100%) were positive for bacteria by culture.
The amount of bacteria found in these samples ranged from 102 to
107 colony-forming units. At S2 and S3, 12 of 43 samples (28%) and
3 of 36 samples (8%) were positive for bacteria by culture, respectively.
The amounts of bacteria were 102 to 104 in S2 and 102 colony-forming
units in S3. Using a cutoff luminescence reading of 105, only 1 S2 sample
was ambiguous in terms of the possibility of being positive by culture
(Fig. 3).

Discussion
Even though the goal of RCT is to eliminate bacteria from the root
canal system of infected teeth in endodontic therapy, assessment of the
degree of disinfection is currently not routinely performed partly
because of the lack of a rapid chairside method. The present study
showed that the bioluminescence-based ATP detection method can
be used in this capacity and can help clinicians to determine the degree
of disinfection during the course of treatment. The ATP method then

Figure 1. Sensitivity of the ATP assay. Overnight cultures of S. sanguinis, S.

mutans, E. faecalis, F. nucleatum, P. gingivalis, and P. intermedia were
serially diluted in RTF; after which, known amounts of bacteria were added
into a 96-well plate with an equal volume of bacterial culture and BacTiterGlow Reagent and incubated for 5 minutes at room temperature. Luminescence was read using a luminometer. Luminescence was calculated as the
mean of the signals from bacterial culture the mean of RTF alone. The experiments were performed 3 times, each time in triplicate. Data shown are the
mean of 3 replicates from 3 independent experiments. RLU, relative lights unit.

JOE Volume -, Number -, - 2015

Figure 2. Specificity of the ATP assay for viable cells. E. faecalis was treated
with 1% NaOCl for the indicated time; after this, sodium thiosulfate was added
to neutralize NaOCl. Viability was determined by plating and ATP assay. The
experiments were performed 3 times, each time in triplicate. Data shown
are the mean of 3 replicates from 3 independent experiments. Luminescence
was calculated as the mean of the signals from bacterial culture the mean of
RTF alone. RLU, relative lights unit.

serves as a convenient tool far better than the time-consuming and laborious bacterial culture method. The sampling method is, of course, critical, and in the current study we applied the technology advocated by
rstavik
Moller (21) in all 3 samples plus the method designed by O
et al (22) in S2 and S3. Potential contamination during the sampling
process was controlled by sterility checks. To ensure comparable
microbial load reductions among samples before and after chemomechanical debridement and then again after further chemical disinfection
using an interappointment dressing, each sample was validated by the
currently accepted gold standard of anaerobic bacterial culturing.
Thus, the results of this clinical study could be applied to a clinical
setting using currently accepted aseptic and antiseptic treatment
principles.
Although the ATP method does not allow identification of endodontic microbiota present in an infected root canal, such information
is of limited value because endodontic infections are polymicrobial in
nature, could possess heterogeneous etiology, and may vary between
individuals (23, 24). In nonresolving symptomatic cases in which
aberrant microorganisms can be suspected as a cause, the culturing
method may have to be chosen.
By culture, the amount of bacteria obtained ranged from 102 to
7
10 cells in the S1 samples. The variation is well in line with findings

Figure 3. Comparisons between ATP assay and bacterial culture of root canal
samples at various stages of RCT. Paper point samples obtained from 43 primary infected root canals before irrigation and instrumentation (S1) and postirrigation and instrumentation (S2). Of the 43 cases, S3 samples were
obtained from 36 root canals after calcium hydroxide dressing. All samples
were subjected to anaerobic culture for 7 days and ATP assay. ATP measurements were tabulated as the mean of triplicate readings. Luminescence was
calculated as the mean of the signals from bacterial culture the mean of
RTF alone. RLU, relative lights unit.

Detection of Root Canal Bacteria

Clinical Research
of other studies. Using 6 commonly isolated bacteria, we determined the
sensitivity of the ATP assay to be between 10 and 100 cells (Fig. 1). Thus,
the method affords sufficient sensitivity for the detection of bacteria in
infected root canals. A potential sampling problem is that human cells
can be included, which may generate false-positive results. However,
chances of human cell contamination were remote in the present study
by excluding suppurating cases and avoiding sampling points to penetrate the apical foramen where this would be possible.
E. faecalis is able to survive for long periods under nutrientdeprived conditions (25). Furthermore, biofilms formed by this organism under such conditions are also more difficult to eradicate
(26). Any bacteria left behind in the canal should be regarded as a
risk for failure to attain healthy periodontal conditions. Given that
the amount of ATP varies between different bacterial species and is
affected by the physiological state of the microorganisms, currently
a lack of data exist on whether the ATP assay can be used to successfully detect bacteria in nutrient-deprived states. In this study, all S1
samples were found to be culture positive, and we were able to clearly
distinguish these culture-positive samples by using the ATP assay
(Fig. 3). Although it is impossible to accurately quantify the relationship between ATP luminescence and the actual bacterial counts in the
context of a mixed bacterial infection in the root canal, our data show
that the ATP luminescence readings could allow us to distinguish between culture-positive and -negative root canals within a time frame of
5 minutes.
Bacteria in infected root canals can exist in a viable but nonculturable state (VBNC). This is presumably caused by periods of stress and
starvation. Bacteria in the VBNC state may fail to grow on routine bacteriologic media but are alive and capable of renewed metabolic activity if
appropriately stimulated (27). Interestingly, ATP levels, which decline
rapidly in dead and moribund cells, have been found to remain high in
VBNC cells (28). Because bacteria in the VBNC state when resuscitated
could potentially initiate infection, the ATP method could provide an
edge over the bacteriologic culture method to determine the status of
root canals in conjunction with RCT.
Current sampling techniques may not allow complete sampling of
bacteria dislodged in accessory canals, fins, and dentinal tubules, but it
is still useful to determine bacterial levels that are compatible with healing. Despite some conflicting views (1, 29) on the correlation between
the presence of bacterial infection at the time of root filling and the
outcome of root canal therapy, bacteria and their by-products remain
the primary etiologic agents of apical inflammatory lesions.
This is the first clinical trial testing the utility of the bioluminescence ATP assay on root canal samples as a rapid tool to determine
the presence of viable bacteria. We envision that the availability of the
ATP assay as a rapid chairside diagnostic test kit will be useful as an
adjunct to dentists in RCT. However, further studies will have to be performed to correlate these results with the outcomes of RCT.

Acknowledgments
Supported by a grant (no. R221-000-051-112) from the Ministry of Education Singapore.
The authors deny any conflicts of interests related to this
study.

References
1. Bystrom A, Happonen RP, Sjogren U, Sundqvist G. Healing of periapical lesions of
pulpless teeth after endodontic treatment with controlled asepsis. Endod Dent Traumatol 1987;3:5863.

Tan et al.

2. Fabricius L, Dahlen G, Sundqvist G, et al. Influence of residual bacteria on periapical

tissue healing after chemomechanical treatment and root filling of experimentally
infected monkey teeth. Eur J Oral Sci 2006;114:278285.
3. Roane JB, Dryden JA, Grimes EW. Incidence of postoperative pain after single- and
multiple-visit endodontic procedures. Oral Surg Oral Med Oral Pathol 1983;55:6872.
4. Walton R, Fouad A. Endodontic interappointment flare-ups: a prospective study of
incidence and related factors. J Endod 1992;18:172177.
5. Peters OA, Schonenberger K, Laib A. Effects of four Ni-Ti preparation techniques on
root canal geometry assessed by micro computed tomography. Int Endod J 2001;34:
221230.
6. Wu MK, van der Sluis LW, Wesselink PR. The capability of two hand instrumentation
techniques to remove the inner layer of dentine in oval canals. Int Endod J 2003;36:
218224.
7. Nair PN. Pathogenesis of apical periodontitis and the causes of endodontic failures.
Crit Rev Oral Biol Med 2004;15:348381.
8. Bystrom A, Claesson R, Sundqvist G. The antibacterial effect of camphorated paramonochlorophenol, camphorated phenol and calcium hydroxide in the treatment of
infected root canals. Endod Dent Traumatol 1985;1:170175.
9. Bergenholtz G, Sp
angberg L. Controversies in endodontics. Crit Rev Oral Biol Med
2004;15:99114.
10. Molander A, Reit C, Dahlen G. Reasons for dentists acceptance or rejection of
microbiological root canal sampling. Int Endod J 1996;29:168172.
11. Siqueira JF Jr, R^ocas IN. Exploiting molecular methods to explore endodontic infections: part 2redefining the endodontic microbiota. J Endod 2005;31:
488498.
12. Siqueira JF Jr, R^ocas IN. Exploiting molecular methods to explore endodontic infections: part 1current molecular technologies for microbiological diagnosis.
J Endod 2005;31:411423.
13. R^ocas IN, Siqueira JF Jr. Identification of bacteria enduring endodontic treatment
procedures by a combined reverse transcriptase-polymerase chain reaction and
reverse-capture checkerboard approach. J Endod 2010;36:4552.
14. Keer JT, Birch L. Molecular methods for the assessment of bacterial viability.
J Microbiol Methods 2003;53:175183.
15. Delahaye E, Welte B, Levi Y, et al. An ATP-based method for monitoring the microbiological drinking water quality in a distribution network. Water Res 2003;37:
36893696.
16. Magic-Knezev A, van der Kooij D. Optimisation and significance of ATP analysis for
measuring active biomass in granular activated carbon filters used in water treatment. Water Res 2004;38:39713979.
17. Hunter DM, Lim DV. Rapid detection and identification of bacterial pathogens by
using an ATP bioluminescence immunoassay. J Food Prot 2010;73:739746.
18. Obee PC, Griffith CJ, Cooper RA, et al. Real-time monitoring in managing the decontamination of flexible gastrointestinal endoscopes. Am J Infect Control 2005;33:
202206.
19. Fazilat S, Sauerwein R, McLeod J, et al. Application of adenosine triphosphate-driven
bioluminescence for quantification of plaque bacteria and assessment of oral hygiene in children. Pediatr Dent 2010;32:195204.
20. Promega. BacTiter Glo Microbial Cell Viability Assay Technical Bulletin 2007. Available at: www.promega.com. Accessed December 30, 2014.
21. Moller 
AJ. Microbiological examination of root canals and periapical tissues
of human teeth. Methodological studies. Odontol Tidskr 1966;74(5):Suppl
1380.
rstavik D, Kerekes K, Molven O. Effects of extensive apical reaming and calcium
22. O
hydroxide dressing on bacterial infection during treatment of apical periodontitis:
a pilot study. Int Endod J 1991;24:17.
23. Siqueira JF Jr, R^ocas IN. Diversity of endodontic microbiota revisited. J Dent Res
2009;88:969981.
24. Siqueira JF Jr, R^ocas IN, Rosado AS. Investigation of bacterial communities associated with asymptomatic and symptomatic endodontic infections by denaturing
gradient gel electrophoresis fingerprinting approach. Oral Microbiol Immunol
2004;19:363370.
25. Sundqvist G, Figdor D, Persson S, Sjogren U. Microbiologic analysis of teeth with
failed endodontic treatment and the outcome of conservative re-treatment. Oral
Surg Oral Med Oral Pathol Oral Radiol Endod 1998;85:8693.
26. Liu H, Wei X, Ling J, et al. Biofilm formation capability of Enterococcus faecalis
cells in starvation phase and its susceptibility to sodium hypochlorite. J Endod
2010;36:630635.
27. Oliver JD. Recent findings on the viable but nonculturable state in pathogenic bacteria. FEMS Microbiol Rev 2010;34:415425.
28. Beumer RR, de Vries J, Rombouts FM. Campylobacter jejuni non-culturable
coccoid cells. Int J Food Microbiol 1992;15:153163.
29. Sathorn C, Parashos P, Messer HH. How useful is root canal culturing in predicting
treatment outcome? J Endod 2007;33:220225.

JOE Volume -, Number -, - 2015