Microbiological status of root canal after unsuccessful endodontic treatment

Summary Introduction The main objective of endodontic treatment is to eliminate infection from root canal and prevent reinfection by three-dimensional hermetic obturation of the canal system. Endodontic failure can occur due to inability of complete control and elimination of infection from the root canal. The aim of this study is to investigate, by PCR technique, microbiological status of previously endodontically unsuccessfully treated teeth immediately after the removal of obturation material. Material and Methods The analysis included 30 teeth indicated for endodontic retreatment. After removing previous root canal filling material, the bacteriological sample was taken by sterile instrument (# 15) and paper points. Standard PCR technique was used to analyze the incidence of E.faecalis, P.micros, P.intermedia, P. endodontalis and A.actinomycetemcomitans. Results Positive bacteriological findings were registered in 80% of cases, while bacteria were not identified in 20% of all samples (all taken from the root canals without significant changes in periapical tissue). From 24 canals with identified bacteria, 17 had affected apical periodontium. The most dominant microbe in root canals with positive bacteriological finding was E.faecalis (83.3% of the canals) and P.intermedia (75%). In case of teeth with chronic periapical changes, the most common was E. faecalis (94%) and P.intermedia (82.3%). Conclusion The presence of periapical lesions significantly affects microbiological status of endodontically treated teeth. The presence of bacteria was confirmed in most teeth with periapical lesions, while the most frequently identified bacteria were E. faecalis, P.intermedia and P.micros.


INTRODUCTION
The main objective of endodontic treatment is to eliminate infection from the root canal and prevent reinfection by three-dimensional hermetic obturation of the canal system. However, sometimes even properly conducted endodontic treatment can fail. It has been confirmed that the outcome of endodontic treatment depends largely on the quality of endodontic procedures and possibility of eliminating infection from root canal system before obturation [1,2]. Due to the complexity of canal systems their cleaning is difficult, therefore bacteria may remain in inaccessible parts of the canal, especially in the apical portion.
Persistant infection in the apical third is most often result of inadequate completion of endodontic treatment i.e. non-aseptic conditions with insufficiently extended and poorly designed access cavities, insufficient dimension of instrumentation, inadequate hermetic obturation or microleakage due to inadequate temporary or definitive restorations [3]. However, number of infections may persist as asymptomatic periapical radiolucency even if endodontic procedure is properly implemented. The reason is usually complex anatomy of the root canal system with regions that cannot be adequately treated and obturated by existing instruments, materials and techniques [4]. Some studies have shown that certain parts of the root canal space remain untouched during chemo-mechanical instrumentation regardless of the preparation technique or instruments used [5]. Untreated parts of the root canal can contain bacteria and necrotic tissues even when obturation seems to be radiographically correct [4]. It is believed that endodontic failure is caused by inability of complete control and elimination of infection in the root canal.
The aim of this study was, using PCR technique, to investigate microbiological status of endodontically treated teeth with persistent infection immediately after the removal of obturation material.

METHODS
Material for microbiological tests was obtained by taking samples from 30 patients who had root canal treatment done earlier but they needed retreatment. After obtaining dental history, and taking periapical radiographs, patients were clinically examined and failure of the initial endodontic treatment was diagnosed.
Microbiological study included 30 teeth (8 multirooted and 22 single-rooted) indicated for endodontic retreatment. Primary endodontic therapy was performed 12 months ago in 2 cases, 1-5 years in 8 cases, while for 20 teeth, primary endodontic treatment was done > 5 years ago. All 30 teeth had inadequate obturation and that was the failure criterion for which patients needed retreatment. Poor quality of obturation was assessed as short filling (in 22 teeth), as "forgotten" canal (in 7 teeth), or separated instrument (in 1 tooth). Adequate restoration or prosthetic restorations (crowns) were observed in 13 teeth, 7 teeth were without coronal restoration for a longer period of time and 10 teeth did not have proper restoration. The presence of symptoms such as pain, swelling, presence of a fistula, sensitivity to percussion or pain when biting were observed in 10 patients, while the remaining 20 had no clinical signs or symptoms.
Status of apical periodontal tissue was evaluated by PAI index, where completely healthy periodontium (PAI 1) was radiographically registered in 6 cases; PAI 2 score (small changes in the bone structure that are not pathognomonic for apical periodontitis) was recorded in 7 cases; PAI 3 score (which includes changes in bone structure with decalcification characteristic for apical periodontitis) was registered in 13 teeth; PAI 4 (which represents periodontitis with clearly defined zone of radiolucency) was noted in 3 cases; the highest score of PAI 5 (advanced periodontitis with signs of exacerbation and expansion of the bone), was registered only in 1 case. Microbiological process involved first the removal of hard and soft deposits from teeth, restorations and decay and placing a rubber dam. Disinfection of the operative field was done with 30% hydrogen peroxide solution and 2.5% sodium hypochlorite solution, which is then inactivated with 5% sodium thiosulfate. Previous root canal filling material was removed without the use of any solvents, fillers, lubricants or irrigants. Gates Glidden drills were used to remove gutta-percha from the first two-thirds of the root canal, and Hedstrom files were used for the apical third of the root canal. During radiographic determination of working length, the quality of removal of obturation material was checked. Working length was determined with apex locator. In cases of forgotten canals, access cavity was extended, oriphicies located, and root canals were in-strumented by modified double-cone technique -Gates Glidden drills, hand Hedstrom and K-Flex files to full working length. The size of apical preparation (ISO # 25 or # 30) depended on initial diameter and curvature of the root canal. Then after small amount of sterile saline was placed in the root canal and further instrumented in order to scrape the material from the canal walls. The sample was taken using sterile canal instrument -pulp extirpation type (# 15) or Hedstrom files (# 15) with the help of paper points that were used to dry the canal. The paper points were placed with sterile forceps and left in the canal for 60 seconds and then placed in sterile microtubes together with canal instruments whose handle was cut off with sterile forceps. The micro-tubes were stored at the temperature of -20 °C until microbiological analysis.
With the aim to isolate DNA, 100 ml of redistilled water was added in each micro-tube with paper points. Isolation of total bacterial DNA (Gram positive and Gram negative bacteria) was performed using commercial kit QIAamp DNA Mini Kit (Qiagen). After application of isolation protocol, bacterial DNA was dissolved in 100 ml of elution buffer.
The incidence of the following bacteria was analyzed by classical PCR: Enterococcus faecalis, Peptostreptococcus micros, Porphyromonas endodontalis and Actinobacillus actinomycetemcomitans. The sequences of used primers; the temperature profile and length of PCR products are shown in Table 1. In each PCR reaction, simultaneously with the test samples, positive and negative controls were used to avoid false positive and negative results. Reaction mixture of PCR with 25 ml volume was made up of the following components: 13 ml dH 2 O shall, 2.5 ml PCR buffer, 1.5 ml of 25 mM MgCl 2 , 1 ml of dNTP, per 1 ml of F and R primers, 0.2 ml of Taq polymerase and 5 ml bacterial DNA.
The effectiveness of PCR reaction was measured by electrophoresis on a vertical 8% polyacrylamide gel (PAA) in 1XTBE buffer, at constant voltage of 200 V for a period of 30 min. Visualization of PCR products was performed by staining with ethidium bromide.

RESULTS
PCR technique was used to analyze the presence of the following microorganisms: E. faecalis, P. micros, P. inter- Positive bacteriological findings were registered in 80% of cases, while the bacteria were not identified in 20% of samples. All negative samples were taken from the root canal without significant changes in the apical periodontal tissue (PAI 1, 2) while 17 out of 24 canals with identified bacteria belonged to the teeth with damaged apical periodontium (PAI 3, 4, 5) ( Table 2). All samples taken from the root canals with chronic periapical lesions were positive for bacteria (100%).
The most dominant microorganism in root canals with positive bacteriological findings was E.faecalis (83.3%), followed by P. intermedia (75%) and P.micros (58.3%) (Table 4). In the group of teeth with healthy apical pericapical tissue, all bacterial species (except A. actinomycetemcomitans which was not detected in any of the samples) were equally represented or were identified in 57.7% of canals. In the case of teeth with chronic periapical changes, the most common isolation was E. faecalis identified in 94% of the canals, then P.intermedia that was present in 82.3% of samples (Table 4).
In regards to the number of bacterial species contained in a single sample, monoinfection was registered in 13.3% of cases (E.faecalis was presented in half of the canals), while the most common bacteria identified in 33.3% of canals were 3 bacterial species per one canal E. faecalis, P. intermedia and P.micros (Table 5). It was observed that the samples taken from the tooth with healthy periodontal tissues around the root tip, showed mainly absence of bacteria or the presence of 1 or 2 bacterial species, while in the case of chronic periapical changes, in more than a half of samples, the presence of 3 or more bacterial species was identified (Table 6).     Correlation between the presence of certain symptoms after the initial endodontic treatment and findings of specific bacterial species in root canal are shown in Table 7. E. faecalis and P.intermedia were detected in all patients with pain, teeth sensitive to percussion and pain when biting, as well as the half of the samples taken from the root canal with sinus tract and swelling. P.micros was identified in 66.6% of patients with pain, 50% of patients with swelling, sinus tract and percussion sensitivity and 33.3% of patients with pain when biting. The presence of P.endodontalis was confirmed in 33.3% of the canals registered with spontaneous pain and 25% with sensitivity to percussion. A. actinomycetemcomitans was identified in 50% of the canals with sinus tract, 33.3% of patients with pain when biting and 12.5% of the samples taken from the root canal of patients with sensitivity to percussion (Table 7).

DISCUSSION
The analysis included 30 teeth that needed retreatment regardless of its causes -prosthetic indications, incidental finding or the patient reported having problems. In order to prevent contamination of canal samples, strictly aseptic conditions were conducted by the current protocol for disinfection of working fields [6][7][8][9]. This involved decontamination and disinfection of the operative field with 30% hydrogen peroxide solution and 2.5% sodium hypochlorite followed by inactivation of 5% sodium thiosulfate so that its remains would not affect the sample taken [7]. Also, complete removing of the root canals filings was done purely mechanically without use of any solvents, lubricants or irrigants. The micro-tubes with microbiological material were stored at -20º C, for no longer than a month. PCR (Polymerase Chain Reaction) is a modern, fast (it takes a few hours), and a simple method for identification of microorganisms. It is extremely sensitive and highly specific. Theoretically, it is possible to demonstrate the presence of only single bacterial cell (living or dead) in the sample, although the number of 10 cells is considered to be the lowest limit for detection (e.g. 100 viable cells are necessary for one method of bacteria cultivation) [9]. Identification is based on in vitro amplification of target DNA fragment, which can be repeated up to a billion times. Because its detection works on the basis of genotypic structure, rather than phenotypic characteristics of microorganisms, identification is very reliable and precise. It is possible to identify those bacterial species that cannot be cultivated in vitro or cases where the number of cells in examined material is very small [9]. However, the PCR method of identification of microorganisms has its limitations. Firstly, the species that are not targeted or detected by other methods cannot be identified. Precisely determined conditions are required for reaction with specific primer pair that can detect only particular and specific bacterial species. It should be noted that PCR method can not determine if identified cell is alive or dead. Another limitation of the conventional PCR method is its inability to quantify the number of bacteria. This can be overcome by "real-time" PCR method where the number of bacterial cells can be determined [10].
Positive bacteriological findings were registered in 80% of cases, while no bacteria were identified in 20% of samples. All samples taken from root canals with chronic periapical lesions were positive for presence of bacteria (100%). Also, all negative samples were taken from root canal with healthy periapical tissue. This confirmed the hypothesis that without bacteria, there is no infection in periapical tissues, and consequently there is no failure of endodontic treatment. Similar results were obtained by Siqueira et al. [5], Gomes et al. [11,12], Roca et al. [13], Sedgley et al. [14] and Sakamoto et al. [15].
However, 23% of positive bacteriological samples were taken from root canals where periapical tissue had PAI score 2 (periapical radiolucency) with no pathognomonic signs of chronic inflammation. Kaufman et al. [16] and Olette et al. [17] reported the presence of bacteria in the root canal without any periapical changes detected radiographically. The explanation may lie in the fact that two-dimensional radiography is not sufficiently accurate to diagnose apical periodontitis with less destroyed bone tissue [18]. Also, it could be that small number and low virulent bacteria are present. Only if present in higher numbers and pathogenic bacteria persisting after primary endodontic treatment can cause or maintain periradicular inflammation [2]. However, there is a dilemma whether bacteria remained after primary endodontic treatment (persistent infection) or they are the result of re-infection (secondary infection). In the recent years, research has pointed out the importance of proper coronal restora- tion in preventing re-infection of endodontic space with opinion that secondary infection is important cause of endodontic treatment failure [1,19]. The most common microorganism in the canal system with positive bacteriological findings was E. faecalis followed by P. intermedia and P. micros accounting for 58.3%. E. faecalis was identified in 94% of root canals with chronic periapical lesions (PAI 3, 4, 5) which is similar to findings of Sedgley et al. who also used PCR identification method. They used 48 samples and showed that the incidence of E. faecalis finding was 90% [14]. Gomes et al. used PCR to analyze microbiological status of previously filled canals with periapical lesions and came to the conclusion that E. faecalis was present in 90% of bacteriologically positive root canals followed by P. micros 59%, P.gingivalis 41%, P. endodontalis 26% and P. intermedia 13% [12]. This high percentage of E. faecalis was probably the result of its numerous and diverse virulence factors and extraordinary ability to survive. This microorganism is small; it easily penetrates into dentinal tubules and has a good adherence to collagen [20]. It is resistant to calcium hydroxide [21] and has the ability to survive (as a single species) in dentinal tubules without the support of other bacteria [22]. Also, it has the ability to survive without nutrients and recover easily when they become available in the form of serum (the origin of alveolar bone and periodontal ligament) [23]. The results of this study confirmed the presence of E. faecalis in 57.7% of root canals without changes in apical periodontal tissue. However, P.intermedia, P. endodontalis and P. micros were also identified in 57.7% of root canals which leads to the conclusion that E.faecalis is not the only microorganism responsible for failure of endodontic treatment. Williams et al. showed that E. faecalis can survive all stages of endodontic treatment because RT-PCR detected its presence in samples taken immediately after instrumentation and irrigation as well as after medication [24].
Our research indicated that remaining microorganisms could be present in root canals due to inadequate coronal seal. In 70% of teeth with healthy periodontal apical tissue, coronal restoration did not show satisfactory quality that opened door for secondary infection of endodontic space. In addition, although present in root canal, the bacteria had no effect on periapical tissue. They probably remained blocked and trapped in dentinal tubules or root canal filling material blocked further progress to periapical tissues (microleakage).
Patients who took part in our study came from general dental practice and health centers where it was quite difficult to adhere to contemporary standards and good endodontic practice, therefore the incidence of residual bacteria after primary endodontic treatment was high. Prevotella spp and Porphiromonas spp (previously classified as Bacteroides spp) belong to the group of "blackpigmented bacteria" because in contact with agar, they form shiny, smooth colonies of gray or black. According to the new taxonomy, saharolytic Bacteroides spp species are classified as genus Prevotella, while asaharolytic species belong to genus Porphiromonas. Types of P. intermedia, P.melaninogenica, P.denticola and P.dentalis belong to gram-negative obligate anaerobes. Although they have limited ability of fermentation of amino acids and require the presence of hemin and menadione for growth, they can be observed in different parts of body (oral cavity, upper respiratory and urogenital system) [25]. Ruan et al. found that P. intermedia originating from oral cavity represents potential opportunistic microorganism associated with periodontal disease but also apical periodontitis due to its adhesiveness and competitiveness with surrounding microorganisms [25]. P. endodontalis is gram-negative microorganism associated with periodontitis, endodontic infections and gingivitis, and more frequently with symptomatic than asymptomatic infections in oral cavity [26].
One-third of patients who took part in the current study showed some clinical symptoms, but the most common symptom was sensitivity to percussion. The most common microorganisms present in the samples taken from such teeth were E.faecalis and P.intermedia (100%), followed by P.micros (50%). Gomes et al. also found statistically significant relationship between P.micros and tooth sensitivity to percussion [12] while Pinheiro et al. noticed association between P.intermedia and the presence of these symptoms [27]. E.faecalis and P.intermedia were detected in all patients with pain, tenderness to percussion and pain when biting, as well as one half of the samples taken from root canals of patients with swelling and fistula. P. micros was identified in 66.6% of patients with pain that is in accordance with the study of Pinheiro et al. who reported involvement of bacterial species P.intermedia and P.micros in teeth where pain was present [27].