Enterococcus spp. isolated from root canals with persistent chronic apical periodontitis in a Chilean population

Aim: To isolate and identify in a Chilean population, Enterococcus spp. from root canals with persistent chronic apical periodontitis (CAP) and to investigate the potential correlation between the bacteria and the observed clinical features. Methods: Twenty patients with indication for endodontic retreatment due to persistent CAP were selected. Data from patient general health and dental clinical history were recorded. During retreatment, a microbial sample was obtained from the root canal and inoculated in a selective Enterococcus medium. Using bacterial cultivation methods, bacterial isolates belonging to the genus Enterococcus were identified. The relationship between the number of colony-forming units of Enterococcus spp. and patient clinical data was assessed statistically by the Pearson Chi square and Fisher exact tests. Finally, a Polymerase Chain Reaction (PCR) assay to determine the most prevalent species of Enterococcus spp. was conducted in the clinical samples, and the results were analyzed by a proportion comparison test. Results: Enterococcus spp. strains were isolated in 70% of the patients . Most of them (98.8%) accounted for Enterococcus faecalis and only 1.2% for Enterococcus faecium. A high frequency of E. faecalis was found in teeth with inadequate endodontic treatment or dental crown restorations. Conclusions: This study concluded that E. faecalis is prevalent in root canals with persistent CAP in a Chilean population. E. faecium as found in a single case with the poorest root canal filling. Further studies are still required to investigate the presence of other species, which may be linked to persistent chronic apical periodontitis.


Introduction
One of the prevalent causes of lesions that appear or remain after endodontic treatment is the presence of microorganisms or contamination in the root canal system 1 .Several authors reported that bacteria could survive within the root canal even after careful mechanical-chemical preparation 2 .Such bacterial presence may lead to clinical symptoms of persistent chronic apical periodontitis (CAP), which has high prevalence in some countries, ranging from 40% to 61% [3][4] .The treatment of persistent CAP aims the removal of root canal filling material, disinfection of canals with irrigants and use of intra canal pastes such as calcium hydroxide.Even so, sometimes the microorganisms persist even after treatment 5 .
In cases of primary endodontic infections, there is a higher bacterial load than in secondary infections [6][7] .After biomechanical preparation of root canals, the bacterial load is reduced by at least 95% and the remaining viable bacteria may be the cause of refractory lesions 8 .Teeth with endodontic failure have been associated with the presence of Enterococcus species [9][10] .These bacteria probably reach the root canal system after treatment, but the source is still unclear.Enterococci do not seem to be colonizers of the oral cavity, unless there is a preferable site, such as a necrotic or obturated root canal without sealing.Probably, the transient microorganisms in the oral cavity and changes in the microenvironment create favorable conditions for secondary or refractory endodontic infection 11 .Enterococcus faecalis is a commensal organism, part of the normal microbiota of the digestive system, but it can also be an opportunistic human pathogen.Moreover, E. faecalis has the ability to maintain its viability for twelve months without additional nutrients.This bacteria is most frequently isolated in CAP cases after endodontic tooth treatment, but it may also be found in cases of necrotic pulps in very few cases 12 .Viable E. faecalis strains could be trapped inside the root canal in the filling material and provide a niche for subsequent infection in the long term 13 .
In Chile, there are yet no studies about the presence of this bacterial genus in CAP.The aim of this study was to identify Enterococcus spp.from the root canals of teeth that had previous endodontic treatment with diagnosis of CAP in a Chilean population and to correlate its presence with the patient's clinical features.

Patient Selection
Twenty adult patients, aged eighteen and older, were chosen.All were treated at the Faculty of Dentistry, University of Concepción, Chile, in November 2013, with endodontic retreatment indicated in the mandibular and maxillary canine, premolar, first or second molar.Teeth that had endodontic treatment two or more years before and showed radiographic evidence of periapical disease were considered to require retreatment.The radiolucent area size was recorded, and classified as widened periodontal ligament space (<1 mm) or periapical lesion (≥1 mm).Pre-treatment quality was recorded after radiographic examination.Root canal fillings with clearly visible voids in the filling material were classified as inadequate.Patients who had antibiotic treatment during the preceding three months and whose teeth could not be fully isolated with a rubber dam and/or with periodontal disease in the selected or adjacent teeth were excluded.The following characteristics for each patient and correlated with the microbial findings were recorded: age, sex, tooth type, and time when the previous treatment was performed.Clinical signs and symptoms included the quality of the dental crown restoration, spontaneous pain, pain history, sensitivity to percussion, pain on palpation, mobility, history of antibiotic therapy or any other relevant medication 9 .
The Ethics Committee of the Faculty of Dentistry at the University of Concepción approved the study protocol (C.I.Y.B.Nº08/14).Before sampling, patients understood and gave written informed consent, according to the Declaration of Helsinki.

Sampling Procedure
A single trained operator collected samples under strict asepsis, as described below.The tooth surface was cleaned with pumice stone powder and isolated with a rubber dam.The tooth and the operation field were disinfected with 3 mL of 5.25% sodium hypochlorite solution, which was then absorbed with sterile gauze.The solution was inactivated with 5% sodium thiosulphate in order to avoid interference with the bacteriological sampling 14 .
Coronal restorations were removed using high-speed sterile carbide drills.To expose the root filling, the access cavity was prepared with sterile drills (Denstply, Maillefer, Switzerland) without water spray but under manual irrigation with sterile saline solution (B.Braun, Barcelona, Spain).Then, the operating field, including the pulp chamber, rubber dam and clamp, was cleaned with 2.5% sodium hypochlorite solution (Clorox Regular Bleach, Oakland, CA, USA), and inactivated with 5% sodium thiosulphate (Hash, Loveland, CO, USA).To control the effectiveness of disinfection, samples from the operating field were obtained and the tooth crown was disinfected with two F1 paper points (Denstply, Maillefer, Switzerland).Next, they were transferred to a tube containing 3mL of trypticase soy broth (TSB) (Difco, Detroit, MI, USA) and incubated for 72 h at 35 °C.The root canal filling was removed with Hedström Files, K and/or H type, and Gates-Glidden drills (Denstply, Maillefer, Switzerland).ProTaper nickel-titanium retreatment instruments D1 -D3 (Denstply, Maillefer, Switzerland) were also used.Solvent was not used in order to avoid harm to bacterial cells.The working length was set at 1 mm from the radicular apex.Radiographs were taken to ensure that all the filling material had been removed.Subsequently, 5 TSB drops were placed in the root canal and a sterile endodontic file was introduced at a level of approximately 1 mm from the apex of the tooth.The content from the root canal was absorbed with 4 sterile paper points.Each paper point was maintained in position inside the duct at the established working length for one minute and then transferred into a tube with 1 mL TSB 15 .From this sample, an aliquot was taken to perform bacterial counts and the remainder was incubated for 24-48 h.If the plate count was negative, the remainder of the sample was allowed to seed on a KF Streptococcus agar plate (Merck, D-61 Darmstadt, Germany) to definitively rule out the presence of Enterococcus spp. 5,16.

Bacterial Count
The bacterial count was carried out in triplicate by seeding onto the surface of KF Streptococcus agar plates (Merck, D-61 Darmstadt, Germany).Plates were incubated for 72 h at 37 °C.Subsequently, six presumptive Enterococcus spp.colonies were chosen from each patient sample.

Metabolic and biochemical strain characterization
To identify the presumptive Enterococcus spp.colonies, Braz J Oral Sci.14(3):240-245 Enterococcus spp.isolated from root canals with persistent chronic apical periodontitis in a Chilean population conventional tests, such as Gram staining, growth in 6.5% NaCl broth, esculin hydrolysis and the arabinose fermentation were carried out 17 .Each Enterococcus spp.strain was stored at -80 ºC in a TSB and 50% glycerol mixture at a ratio of 2:1.

Data Analysis
Prevalence of Enterococcus spp. was recorded as percentage of positive cases.The relationship between the presence of E. faecalis and the clinical variables was analysed with the Pearson, chi-square and Fisher's exact tests using SPSS version 15 (SPSS Inc., Chicago, IL, USA).The results of PCR assay to detect E. faecalis in clinical samples were analysed with the proportion comparison z-test, adopting a 0.05 significance level.

Results
Presumptive colonies of enterococci were found on KF Streptococcus agar plate in 14 out of 20 samples obtained from the root canals.The colonies were circular with convex elevation and smooth margins, deep red and 1 mm diameter (Figure 1).The bacterial counts ranged from zero CFU/mL to 2.1x10 5 CFU/mL.They represent the distribution of E. faecalis countings over the samples (Table 1).After regrowth time, aliquots seeded again in KF agar, reporting some colonies (few) not characteristic to Enterococcus, presumably positive for other microorganism.Phenotypic identification indicated that 83/84 (98%) selected strains corresponded to E. faecalis, which was confirmed by PCR, as amplified bands of 941 bp were obtained with the specific primers for this species in all assays (Figure 2).E. faecium (2%) was the only strain with different characteristics identified through phenotypic and molecular tests.In this case, a 550 bp amplification product was obtained using specific primers for this species (Figure 3).
Regarding the clinical aspects, no significant differences were found between gender, age and medical history variables.Most teeth with presence of Enterococcus spp.were maxillary and posterior.For the variable related to time after previous endodontic treatment, increased presence of Enterococcus spp. was observed in the range of two to five years.For the variable regarding quality of root canal filling and dental crown restorations, a high microorganism frequency was found in treatments labelled as "inadequate".In relation to the clinical evaluation, there was a direct association between the absence of signs and symptoms and  the presence of Enterococcus spp.For the periapical variable, there was no association between the lesion size and the count.However, no statistically significant relationship was found between the clinical variables and bacterial count.

Discussion
Due to the physical limitations of the root canal system, obtaining a representative sample from this site is not an easy task, and it is even more difficult in patients with an indication for retreatment, in which the number of microorganisms may be low and a significant number of microbial cells can be lost during root canal desobturation procedures.As a result, the number of bacterial cells in the sample can fall below the detection rate of the isolation method and a particular species' prevalence may be underestimated 20 .This was one limitation of the present study, in which the impossibility of using liquid solvents made the surgical procedure last for three hours in order to completely remove all root canal filling material.This could have influenced those cases that were negative for enterococci.The appearance of some colonies (few) not characteristic to Enterococcus can be explained because in the KF Streptococcus agar may grow other species like Listeria monocytogenes, Streptococcus bovis group, Pediococcus and staphylococci.Other organisms (e.g., micrococci, Candida, corynebacteria and gram-negative bacteria) may appear as clear small colonies or produce growth traces.
A study carried out to compare the detection rate of E. faecalis from cultivation and PCR test from the same sample found that 46.6% of the sampled canals had E. faecalis 14 .In a Lithuanian population study, 56% of the cases were reported to be positive for E. faecalis, using only culture techniques for identification 21 .In this study, 45% of the samples were positive for Enterococcus spp.using culture techniques.However, performing sample regrowth and subsequent seeding again on KF Streptococcus agar plates, the detection percentage yielded up to 70%, depending on the total number of samples, which was confirmed by the PCR test; which is consistent with previous studies.It should be underscored that the sample size in the present case is smaller than in other studies and it may have influenced the higher percentage obtained.The authors know that sample size calculation may determine the exact number of samples is required to strictly determine the prevalence of Enterococcus spp.The authors developed a descriptive, cross-sectional study.Relative frequency Enterococcus spp isolated from root canals was evaluated in a limited time.In this case, the sample was not probabilistic.It was taken depending on availability of filed cases in a given time.This limited the number of sampled patients.
The prevalence of E. faecalis and its association with clinical findings among patients in Beijing, China, showed that root canal treatments displaying unsatisfactory obturation presented a greater diversity of cultivable species than those with satisfactory obturation 22 .Those canals probably provided more space and nutrition than wellobturated canals.In fact, in that study, E. faecalis was more frequent in the canals than in saliva, but this was not the aim of our study.On the contrary, the well-obturated canals maintained an obliged anaerobic environment, which does not favor the survival and growth of E. faecalis.In that Chinese study, as well as in the present study, no association was found between these clinical features and the prevalence of E. faecalis in radicular canals, which is also attributed to the low sensitivity of Fisher's exact test in small samples 22 .E. faecalis has been commonly found in filled canals with radiographic evidence of peri-radicular periodontitis or without injury than in those with a radiolucent area, suggesting that the entry of bacteria may occur after filling, which is consistent with the findings of this study 23 .The observation of similar percentages of E. faecalis leakage, regardless of the obturation technique, reinforces this idea 24 .
Confirmation of the identity of E. faecalis by PCR has proven to be more sensitive than conventional E. faecalis culture methods.In this study, the use of a selective medium (KF Streptococcus agar) and some biochemical tests enabled us to focus on the identification of this bacterium in relation to other species.In this case, there was no presence of E. faecalis in 30% of the samples in comparison with other reports of 6.8% 25 to 20% 21 of teeth.Combining molecular and culture techniques, including biochemical profile is most likely the best available method to provide a comprehensive information on the bacterial populations associated with endodontic infections.Ultimately, the observation that 98.8% of the strains were E. faecalis versus the 1.2% that turned out to be E. faecium was a probable outcome, considering the results of a study on non-dental clinical isolates from a hospital environment in a Chilean population 18 .It should be stressed that the most relevant clinical characteristic of one of the patients, from whom the strain of E. faecium was isolated, was that he presented the worst quality of obturation, which was evident in the root canal filling material, the absence of dental crown restorations and sensitivity to palpation.However, his initial count was <10 CFU/mL as an estimated value.
Clearly, further studies are required to analyse other enterococci populations that may be linked to the presence and establishment of refractory chronic apical periodontitis during or after endodontic treatment, and which may form a multi-species biofilm.E. faecalis has been regarded for more than 30 years as a leading cause of endodontic failures.However, this assertion was placed in doubt based on recent publications reporting that other species, such as streptococci, may be predominant pathogens associated with persistent endodontic infections.The difference may be due to different methodological approaches for sampling and detection, different clinical conditions or socio-geographical differences in the studied subjects 6,18,26 .Finally, it is necessary to advocate for a comprehensive approach to find new antibacterial substances that completely eradicate the microorganisms from the root canal system, which in its genesis is formed to maintain sterility 27 .
This study has shown that there is a high prevalence of E. faecalis in root canal-treated teeth in a Chilean population.Further studies are required to analyse the presence and role of other species, which may be linked to refractory chronic apical periodontitis.