Basic Research—Biology
Antibiotic Resistance and Capacity for Biofilm Formation of Different Bacteria Isolated from Endodontic Infections Associated with Root-filled Teeth Ali Al-Ahmad, PD,* Hawnaz Ameen,* Klaus Pelz, MD,† Lamprini Karygianni, DMD,* Annette Wittmer,† Annette C. Anderson, Dr hum med,* Bettina Spitzm€ uller,* and Elmar Hellwig, DMD* Abstract Introduction: To date, a variety of microbial species have been isolated from endodontic infections. However, endodontic clinical bacterial isolates have not been sufficiently characterized with regard to their capacity for antibiotic resistance and biofilm formation. In this study, antibiotic resistance and biofilm formation of 47 different aerobic and anaerobic bacterial isolates, belonging to 32 different species previously isolated from infected filled root canals, were studied. Methods: Antibiotic sensitivity to 11 antibiotics including penicillin G, amoxicillin, clindamycin, gentamicin, vancomycin, tetracycline, doxycycline, fosfomycin, rifampicin, ciprofloxacin, and moxifloxacin was tested using the standardized Etest method (Bio Merieux, Marcy-1’Etoile, France). The antibiotic sensitivity of 4 control strains was also estimated in parallel. Additionally, the capacity to form biofilms was quantified using the microtiter plate test. Results: Different aerobic and anaerobic bacterial species were either resistant against a number of antibiotics or showed high minimal inhibitory concentrations against clinically relevant antibiotics. Five aerobic and 2 anaerobic isolates, including Enterococcus faecalis, Streptococcus mutans, Lactobacillus fermentum, Actinomyces naeslundii, Actinomyces viscosus, Prevotella buccae, and Propionibacterium acidifaciens, were characterized as being high biofilm producers, whereas 8 aerobic and 3 anaerobic isolates were found to be moderate biofilm producers. Most isolates with resistance or markedly high minimal inhibitory concentration values were also either moderate biofilm producers or high biofilm producers. Conclusions: These results suggest that the clinical significance of endodontic infections could include that they serve as a reservoir for antibiotic resistance. Furthermore, endodontic treatment should consider the adhesion and biofilm formation by a variety of bacteria. (J Endod 2014;40:223–230)
Key Words Antibiotic resistance, apical periodontitis, biofilm, endodontic infection
T
he persistence of microorganisms in root canal infections is the most widely accepted cause of endodontic treatment failure and is accompanied by the continuing presence of periradicular lesions (1, 2). In cases of endodontic retreatment, microorganisms have been isolated from 35%–100% of the root canal– treated teeth after removal of the obturation material (3–13). Many studies have highlighted the diversity of the microbial populations isolated from secondary endodontic infections as described in an earlier report from our group (13). However, very few studies have characterized the recovered clinical bacterial isolates from secondary endodontic infections with regard to their antibiotic resistance or capacity to form biofilm. Both characteristics are of major clinical concern. Preethee et al (14) recently found evidence that efaA, a potent Enterococcus faecalis virulence gene associated with infective endocarditis, can be found in E. faecalis strains detected in therapy-resistant infected root canals. Because E. faecalis has been frequently recovered from persistent and secondary endodontic infections (13) and microorganisms such as streptococci found in infected root canals are known to be potential agents of endocarditis, the characterization of antibiotic resistance of endodontic bacteria remains an important focus of microbiological research. Another clinically important property of endodontic microorganisms is their ability to form biofilms. In treated and untreated root canals, apical periodontitis can be classified as a biofilm-induced disease (15, 16). To the best of our knowledge, among different clinical bacterial isolates recovered from endodontic infections, E. faecalis is the only species that has been widely studied for its capacity to form biofilm (17, 18). If bacteria participate in gene exchange within a biofilm via horizontal gene transfer, processes leading to a spread of antibiotic resistance genes between different clinically relevant species can be accelerated. As summarized by Madson et al (19), horizontal gene transfer rates are typically higher in biofilm communities compared with those in planktonic niches. Thus, there is a connection between biofilm formation and horizontal gene transfer. In addition to this, the persistence of endodontic bacteria via biofilm formation underlines the necessity for more effective methods not only to completely eliminate bacteria during endodontic retreatment but also to isolate all of the existing microorganisms during the microbiological sampling from infected root canals. It should also be kept in mind that the complex anatomy of the root canal poses further difficulties because biofilms of persistent microorganisms within root canals may also be located on the walls of
From the Departments of *Operative Dentistry and Periodontology and †Hygiene and Microbiology, Albert-Ludwigs-University, Freiburg, Germany. Address requests for reprints to Dr Ali Al-Ahmad, Department of Operative Dentistry and Periodontology, University School and Dental Hospital, Hugstetter Straße 55, D-79106 Freiburg, Germany. E-mail address: [email protected] 0099-2399/$ - see front matter Copyright ª 2014 American Association of Endodontists. http://dx.doi.org/10.1016/j.joen.2013.07.023
JOE — Volume 40, Number 2, February 2014
Antibiotic Resistance and Capacity of Bacteria
223
Basic Research—Biology ramifications and isthmuses (15). The aim of the present study was to analyze the capacity for biofilm formation and a broad pattern of antibiotic sensitivity for a diverse array of clinical bacterial isolates obtained from patients with secondary endodontic infections in asymptomatic teeth with apical periodontitis.
Material and Methods Clinical Bacterial Isolates and Strain Maintenance A total of 21 patients were referred to the University Clinic and Dental Hospital, University of Freiburg, Freiburg, Germany, for endodontic retreatment over a period of 2 years. After microbiological sampling, 47 different bacterial isolates belonging to 32 different species were isolated from the infected root canals. All patients gave their written informed consent to the study protocol, which had been approved by the Ethics Committee (no. 140/09, University of Freiburg). During the pretreatment examination, the following clinical parameters were evaluated by an endodontist: sex and age of the patient, endodontic history, tooth type (ie, incisor, premolar, or molar in upper/lower jaw), clinical signs (ie, the presence of sinus tract and pus), type of previous endodontic treatment, and radiographic appearance. Indeed, preoperative radiographs were taken, all radiographs were digitized, and the radiographic quality of the pre-existing obturation was estimated. Endodontic treatment of all teeth had been completed at least 2 years before the study, and all of the teeth exhibited apical periodontitis on radiographic examination. In each of the cases, retreatment was indicated, and previous root canal treatment was considered a failure. In our attempt to focus on bacterially induced apical infections and to eliminate other causative factors, we assumed that mechanical microbial access to the apex is a prerequisite for the contamination of periapical tissues. In this context, teeth with obturation material that did not reach within 4 mm of the radiographic apex or that could not be isolated with a rubber dam were excluded from the study. No direct exposure of the root canal filling material to the oral cavity was evident. All of the teeth were asymptomatic. Patients who used antibiotics within the last 30 days before commencement of the study were excluded. A detailed description of the patients’ collective characteristics, the sampling procedure used, the isolation, and the identification of the recovered clinical isolates has been reported in earlier studies (11–13). Briefly, the tooth and surrounding area were cleaned with 30% hydrogen peroxide (H2O2) and swabbed with a 3% sodium hypochlorite solution (NaOCl). Endodontic access was achieved with a sterile high-speed carbide bur until the root filling was exposed. Subsequently, the tooth and the adjacent rubber dam were disinfected a second time using 30% H2O2 and 3% NaOCl. The cavity was swabbed with a 5% sodium thiosulfate solution to inactivate the NaOCl. To assess the efficacy of the disinfection, a sterile foam pellet was moistened in a sterile 0.9% NaCl solution and was used to swab the access cavity and the tooth surface. If bacterial growth occurred in these quality control samples, the tooth was excluded from the study. Coronal gutta-percha was removed with Gates Glidden drills. The working length was established radiographically and with the aid of an electronic apex locator (Raypex 5; VDW, Munich, Germany). The canal was enlarged from 0.5–2 mm from the radiographic apex with ProTaper NiTi instruments (Dentsply Maillefer, Ballaigues, Switzerland). Teeth that could not be instrumented to this length were excluded from the study. No solvent was used at any time. After introducing approximately 40 mL sterile saline solution (0.9% NaCl) into the canal with a sterile syringe, 3 sequential sterile paper points (ISO 25, taper 04; ROEKO, Langenau, Germany) were placed to the working length to soak 224
Al-Ahmad et al.
up the fluid. Each paper point was kept inside the canal for 1 minute and then transferred into a sterile vial containing 0.75 mL reduced transfer fluid. Finally, conventional retreatment was finished after root canal disinfection, and the root canal was filled by using vertical compaction. The long-term storage of all clinical isolates was accomplished at 80 C in basic growth medium containing 15% (v/v) glycerol as described by Jones et al (20).
Testing Antibiotic Sensitivity Using the Etest The testing of antibiotic sensitivity using the Etest method (Bio Merieux, Marcy-1’Etoile, France) was conducted according to the manufacturer’s instructions and as described in an earlier report (3). The standardized Etest was applied to reveal any antibiotic sensitivity for the different clinical bacterial isolates recovered from the infected root canals. Thus, the antibiotic sensitivity of aerobic and facultative anaerobic bacterial isolates was tested against 11 antibiotics including penicillin G, amoxicillin, clindamycin, gentamicin, vancomycin, tetracycline, doxycycline, fosfomycin, rifampicin, ciprofloxacin, and moxifloxacin. The antibiotic sensitivity of anaerobic bacteria was tested using 12 antibiotics, including those previously mentioned and metronidazole. For aerobic and facultative anaerobic bacteria, the Etest was conducted on Mueller-Hinton agar for E. faecalis isolates and the control strain E. faecalis (American Type Culture Collection [ATCC] 29212) or on Mueller-Hinton agar containing 5% horse blood for Streptococcus mutans, Streptococcus oralis, Streptococcus sanguinis, Streptococcus constellatus, Lactobacillus fermentum, Lactobacillus rhamnosus, Lactobacillus casei, Lactobacillus gasseri, Actinomyces naeslundii, Actinomyces viscosus, Moraxella osloensis, Eikenella corrodens, and both control strains Haemophilus influenzae (National Collection of Type Cultures [England] [NCTC] 8468) and Streptococcus pneumoniae (ATCC 49619). Colonies of each isolate were picked from an overnight agar plate and suspended in sterile saline (0.9%) to an inoculum turbidity of McFarland 0.5. Using sterile nontoxic spun swabs, each isolate was streaked on agar plates. The Etest strip was then applied to the agar surface using sterile tweezers. The anaerobic isolates belonged to the following species: Prevotella intermedia, Prevotella buccae, Fusobacterium nucleatum, Tannerella forsythia, Porphyromonas gingivalis, Campylobacter rectus, Campylobacter gracilis, Veillonella dispar, Dialister pneumosintes, Parvimonas micra, Atopobium rimae, Propionibacterium propionicum, Propionibacterium acidifaciens, Pseudoramibacter alactolyticus, Solobacterium moorei, Mogibacterium pumilum, Mogibacterium timidum, Bifidobacterium sp, and the control strain Bacteroides fragilis (ATCC 25285). For P. gingivalis, the Etest was conducted on yeast cysteine blood agar. For all other anaerobic isolates, Wilkins-Chalgren agar with 5% horse blood was used (anaerobic jar [AnaeroCultA; Merck, Darmstadt, Germany]). Colonies of each isolate were picked from a 48-hour subculture and suspended in sterile saline (0.9%) to an inoculum turbidity of 0.5–1 McFarland. Four milliliters of each suspended isolate were pipetted on the agar plate, excessive solution was then removed, and the plates were dried. The Etest strip was then applied to the agar surface using sterile tweezers. Nitrocefin discs (Becton Dickinson, Sparks, MD) were used for the detection of beta-lactamases. If available, the breakpoints according to the European Committee on Antimicrobial Susceptibility Testing and the epidemiologic cut-off values were used to compare the results. When European Committee on Antimicrobial Susceptibility Testing and epidemiologic cut-off values did not exist, minimal inhibitory concentration (MIC) values for similar strains were derived from the literature and used to compare the antibiotic sensitivity of the endodontic clinical isolates. JOE — Volume 40, Number 2, February 2014
JOE — Volume 40, Number 2, February 2014
0.19 0.094 0.016 0.023 0.032 1 NCTC, National Collection of Type Cultures (England); ND, not determined for facultative anaerobic bacterial isolates.
16 >1024 >1024 0.75 8
6
ND 0.006 0.25 0.38 0.094 0.004 192 0.125 3 4 0.38
0.19
ND 0.064 0.032 0.064 0.094 0.25 0.5 8 12 0.094 0.38
0.047
ND 0.125 0.500 4.000 16.000 0.380 4.000 32.000 8.000 0.500 6.000 2.000
E. faecalis ATCC 29212 S. pneumoniae ATCC 49619 H. influenzae NCTC 8468 B. fragilis ATCC 25285
TABLE 1. Antibiotic Sensitivity of All Tested Control Strains
Results Antibiotic Sensitivity Testing A total of 47 different bacterial isolates belonging to 32 different species were screened for antibiotic sensitivity. Table 1 shows the values of antibiotic sensitivity of all 4 control strains (E. faecalis, S. pneumoniae, H. influenzae, and B. fragilis) tested in parallel. The results of antibiotic sensitivity testing of aerobic, facultative anaerobic, and anaerobic bacteria are depicted in Tables 2 and 3, respectively. Only 2 E. faecalis isolates (nos. 3 and 6 in Table 2) showed MICs of 64 mg/L for tetracycline and 12 mg/L for doxycycline and can therefore be considered resistant to these antibiotics. Attention should be paid to S. mutans (no. 8 in Table 2) exhibiting high MIC (256 mg/L) for fosfomycin. All other Streptococcus strains as well as A. naeslundii and Actinomyces viscosus showed high MIC values in the range of 24–192 mg/L for fosfomycin. Lactobacilli presented high antibiotic resistance against fosfomycin (MIC >1024 mg/L). With an MIC higher than 256 mg/L, L. fermentum, L. rhamnosus, and L. casei also showed high antibiotic resistance against vancomycin. L. gasseri showed intermediary MIC values in the range of 1–1.5 mg/L for vancomycin but was resistant to ciprofloxacin (MIC >32 mg/L). Taking a look at the anaerobic isolates (Table 3), P. intermedia proved to be penicillin resistant (MIC 16 mg/L) and showed a positive beta-lactamase result. However, P. intermedia and Prevotella buccae revealed high MIC values for tetracycline and doxycycline. Compared with the break points found for aerobic bacteria, these 2 species could
Species
Biofilm Plate Assay The biofilm formation test was conducted as previously described (21). In brief, aerobic bacterial strains were grown in tryptic soy broth (Merck, Darmstadt, Germany) overnight at 37 C under aerobic conditions with 5% CO2 (capnophilic conditions), whereas anaerobic bacteria were grown in gas chromatography Hewlett-Packard (GCHP, Wilmington, DE) broth overnight under anaerobic conditions (anaerobic jars [GENbox anaer; bioMerieuxsa, Marcy l’Etoile, France]). The log10 of the colony-forming unit of each overnight culture as determined on Columbia blood agar or yeast cysteine blood agar was in the range of 108 colony-forming unit/mL. Polystyrene 96well tissue culture plates (Greiner Bio-One, Frickenhausen, Germany) were filled with either 180 mL fresh tryptic soy broth for aerobic bacteria or with GC-HP broth for anaerobic bacteria; afterward, 20 mL of the overnight culture were added to each well. The plates were incubated for 48 hours at 37 C in an aerobic atmosphere with 5% CO2 to cultivate biofilms of aerobic bacteria or under anaerobic conditions in anaerobic jars (AnaeroCult A) for biofilm formation of the anaerobic isolates. The culture medium was discarded, and the wells were washed 3 times with 300 mL phosphate-buffered saline to remove nonadherent bacteria. The plates were air dried and stained with 0.1% crystal violet (Median Diagnostics GmbH, Dunningen, Switzerland) for 10 minutes. Excess stain was removed by washing 3 times with 200 mL distilled water. The plates were dried for 10 minutes at 60 C. Fifty microliters of alcohol (99.9%, absolute for analysis, Merck) were added to each well for resolubilization of the dye. The optical density was measured at 595 nm with a Tecan Infinite 200 plate reader (Tecan, Crailsheim, Germany). All tests were performed in quadruplicate, whereas the experiments were conducted twice, and the mean values were determined. Three biofilm-forming categories were established according to 2 different cut-off values, which were fixed to define the following categories: biofilm nonproducer or C1, biofilm moderate producer or C2, and biofilm high producer or C3. A low cut-off value was fixed by adding 3 standard deviations of the blank to the negative control. The high cut-off value was defined as 3 times the low cut-off value.
Penicillin G Clindamycin Amoxicillin Gentamicin Fosfomycin Vancomycin Ciprofloxacin Tetracycline Doxycycline Rifampicin Moxifloxacin Metronidazol
Basic Research—Biology
Antibiotic Resistance and Capacity of Bacteria
225
Moxifloxacin
0.125 0.190 0.190 0.125 0.125 0.125 0.064 0.125 0.19 0.094 0.023 0.023 0.5 0.125 0.125 0.023 0.75 0.38 0.125 0.032 0.032 1.000 1.500 2.000 1.500 0.380 1.000 0.047 0.012 0.012 0.064 0.064 0.003 0.047 0.38 0.064 0.094 0.19 0.002 0.002 0.125 0.5 0.125 0.190 12.000 0.190 0.750 12.000 0.25 0.5 0.25 0.125 32 1.5 1 0.032 0.032 1.500 2.000 1.500 2.000 1.000 3.000 0.75 0.5 1 0.75 1 1 >256 >256 >256 >256 1 0.5 0.19 3 48
Vancomycin
3.000 8.000 6.000 8.000 4.000 8.000 2 0.75 4 4 0.75 0.5 0.125 2 3 0.5 3 0.38 0.125 0.047 3 0.094 1.000 0.750 0.750 1.000 0.750 0.023 0.016 0.064 0.032 0.047 0.023 0.125 2 0.5 0.064 0.25 0.032 0.016 0.047 0.25
Fosfomycin Gentamicin Amoxicillin
Discussion
6.000 16.000 6.000 2.000 3.000 8.000 0.023 0.023 0.094 0.032 0.032 256
Clindamycin
Biofilm Formation Besides testing the antibiotic sensitivity patterns for all of the aerobic and facultative anaerobic isolates and the anaerobic endodontic isolates, their capacity for biofilm formation was also examined. The different clinical isolates were divided into 3 categories including nonproducers of biofilm or C1, moderate biofilm producers or C2, and high biofilm producers or C3. The corresponding optical density (OD) cut-off values varied between 0.086 and 0.258. OD values lower than 0.086 represented biofilm nonproducers, values ranging between 0.086–0.258 were associated with moderate biofilm producers, and OD values higher than 0.258 corresponded to high biofilm producers. Figures 1 and 2 show the results concerning biofilm formation by aerobic and anaerobic bacteria, respectively. Among the aerobic bacteria (Fig. 1), 7 isolates comprising the species E. faecalis, S. anginosus, S. oralis, M. osloensis, and E. corrodens failed to form biofilms. Eight isolates belonging to the species E. faecalis, S. mutans, S. constellatus, L. rhamnosus, L. casei, and Lactobacillus gasseri were found to be biofilm moderate producers. Five isolates of the species E. faecalis. S. mutans, L. fermentum, A. naeslundii, and A. viscosus were characterized as biofilm high producers. Considering the anaerobic isolates (Fig. 2), 2 isolates comprising the species P. buccae and Propionibacterium acidifaciens were found to be high biofilm producers. Three isolates of the species F. nucleatum, V. dispar, and Propionibacterium propionicum were characterized as moderate biofilm producers. All other anaerobic bacterial isolates depicted in Figure 2 failed to form biofilms and were categorized as biofilm nonproducers. However, 1 isolate (no. 14) of the species Parvimonas micra tended to be a moderate biofilm producer.
0.500 4.000 4.000 3.000 3.000 3.000 0.008 0.012 0.047 0.032 0.016 0.016 0.25 2 0.5 0.094 0.094 0.016 0.006 0.5 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Penicillin G Species
E. faecalis E. faecalis E. faecalis E. faecalis E. faecalis E. faecalis S. mutans S. mutans S. mutans S. oralis S. anginosus S. constellatus L. fermentum L. rhamnosus L. rhamnosus L. casei L. gasseri A. naeslundii A. viscosus M. osloensis E. corrodens
No.
TABLE 2. Antibiotic Sensitivity of All Tested Aerobic and Facultative Anaerobic Endodontic Bacterial Isolates
226
be considered resistant to tetracycline and doxycycline. Tannerella forsythia exhibited resistance against ciprofloxacin (MIC >32 mg/ mL) and rifampicin (MIC = 32 mg/mL). Veillonella spp and Dialister spp showed typical resistance to gentamycin and vancomycin. P. alactolyticus and Mogibacterium timidum were resistant to tetracycline (MIC 8 and 4 mg/L, respectively) and doxycycline (MIC = 6 and 3 mg/mL, respectively). Solobacterium moorei presented a noticeably high MIC (32 mg/L) for rifampicin.
6.000 48.000 24.000 48.000 48.000 24.000 94 256 64 24 24 64 >1024 >1024 >1024 >1024 >1024 64 192 24 >1024
Doxycycline
Rifampicin
Basic Research—Biology
In various recent reports, a variety of bacteria have been isolated and identified from secondary endodontic infections. However, considering the broad assortment of bacterial species that are causing endodontic infections, very few reports have depicted their capacity for biofilm formation or sensitivity to different antibiotics. Both topics are of major clinical significance for general oral health and particularly for the prognosis of endodontic treatment. In this study, we present for the first time a characterization of the capacity of 47 clinical isolates belonging to 32 different species, which were gained entirely from secondary/persistent endodontic infections to form biofilms as well as their antibiotic sensitivity patterns. Antibiotic sensitivity testing not only constitutes an essential requirement for the treatment of infectious bacteria but is also an effective method to characterize them. The Etest is a standard test widely used in medical microbiology and shows acceptable antibiotic sensitivity results for most medically important antimicrobial drugs as well as different pathogens (22–24). For aerobic bacteria, a total of 11 different antibiotics were tested, whereas 12 were used for the anaerobic bacterial isolates. Furthermore, the comparison of antimicrobial susceptibility of different strains using this test was JOE — Volume 40, Number 2, February 2014
JOE — Volume 40, Number 2, February 2014
TABLE 3. Antibiotic Sensitivity of All Tested Anaerobic Endodontic Bacterial Isolates Species
1 2 3 4 5 6 7 8 9 10 11
P. intermedia P. buccae F. nucleatum F. nucleatum F. nucleatum F. nucleatum T. forsythia P. gingivalis C. rectus V. dispar Dialister pneumisintes P. micra P. micros P. micros Atopobium rimae A. rimae P. alactolyticus P. alactolyticus P. alactolyticus S. moorei M. pumilum M. timidum C. gracilis Bifidobacterium sp P. propionicum P. acidifaciens
12 13 14 15
Antibiotic Resistance and Capacity of Bacteria
16 17 18 19 20 21 22 23 24 25 26
Penicillin G Clindamycin Amoxicillin Gentamicin Fosfomycin Vancomycin Ciprofloxacin Tetracycline Doxycycline Rifampicin Moxifloxacin Metronidazol
Resistant bacteria are marked in bold.
16 0.047 0.064 0.016 0.008 0.002 0.002 0.002 0.008 0.125 0.19
0.008 0.012 0.016 0.016 0.016 0.016 0.016 0.008 256 >256
0.5 0.75 1.5 0.5 0.38 0.5 >32 0.25 0.047/0.25 0.064 0.047
3 8 0.25 0.064 0.047 0.016 0.32 0.047 0.023 0.064 0.064
3 3 0.19 0.047 0.047 0.012 0.32 0.094 0.023 0.094 0.094
0.047 0.125 0.5 0.125 0.125 0.047 32 0.003
Antibiotic Resistance and Capacity for Biofilm Formation of Different Bacteria Isolated from Endodontic Infections Associated with Root-filled Teeth Ali Al-Ahmad, PD,* Hawnaz Ameen,* Klaus Pelz, MD,† Lamprini Karygianni, DMD,* Annette Wittmer,† Annette C. Anderson, Dr hum med,* Bettina Spitzm€ uller,* and Elmar Hellwig, DMD* Abstract Introduction: To date, a variety of microbial species have been isolated from endodontic infections. However, endodontic clinical bacterial isolates have not been sufficiently characterized with regard to their capacity for antibiotic resistance and biofilm formation. In this study, antibiotic resistance and biofilm formation of 47 different aerobic and anaerobic bacterial isolates, belonging to 32 different species previously isolated from infected filled root canals, were studied. Methods: Antibiotic sensitivity to 11 antibiotics including penicillin G, amoxicillin, clindamycin, gentamicin, vancomycin, tetracycline, doxycycline, fosfomycin, rifampicin, ciprofloxacin, and moxifloxacin was tested using the standardized Etest method (Bio Merieux, Marcy-1’Etoile, France). The antibiotic sensitivity of 4 control strains was also estimated in parallel. Additionally, the capacity to form biofilms was quantified using the microtiter plate test. Results: Different aerobic and anaerobic bacterial species were either resistant against a number of antibiotics or showed high minimal inhibitory concentrations against clinically relevant antibiotics. Five aerobic and 2 anaerobic isolates, including Enterococcus faecalis, Streptococcus mutans, Lactobacillus fermentum, Actinomyces naeslundii, Actinomyces viscosus, Prevotella buccae, and Propionibacterium acidifaciens, were characterized as being high biofilm producers, whereas 8 aerobic and 3 anaerobic isolates were found to be moderate biofilm producers. Most isolates with resistance or markedly high minimal inhibitory concentration values were also either moderate biofilm producers or high biofilm producers. Conclusions: These results suggest that the clinical significance of endodontic infections could include that they serve as a reservoir for antibiotic resistance. Furthermore, endodontic treatment should consider the adhesion and biofilm formation by a variety of bacteria. (J Endod 2014;40:223–230)
Key Words Antibiotic resistance, apical periodontitis, biofilm, endodontic infection
T
he persistence of microorganisms in root canal infections is the most widely accepted cause of endodontic treatment failure and is accompanied by the continuing presence of periradicular lesions (1, 2). In cases of endodontic retreatment, microorganisms have been isolated from 35%–100% of the root canal– treated teeth after removal of the obturation material (3–13). Many studies have highlighted the diversity of the microbial populations isolated from secondary endodontic infections as described in an earlier report from our group (13). However, very few studies have characterized the recovered clinical bacterial isolates from secondary endodontic infections with regard to their antibiotic resistance or capacity to form biofilm. Both characteristics are of major clinical concern. Preethee et al (14) recently found evidence that efaA, a potent Enterococcus faecalis virulence gene associated with infective endocarditis, can be found in E. faecalis strains detected in therapy-resistant infected root canals. Because E. faecalis has been frequently recovered from persistent and secondary endodontic infections (13) and microorganisms such as streptococci found in infected root canals are known to be potential agents of endocarditis, the characterization of antibiotic resistance of endodontic bacteria remains an important focus of microbiological research. Another clinically important property of endodontic microorganisms is their ability to form biofilms. In treated and untreated root canals, apical periodontitis can be classified as a biofilm-induced disease (15, 16). To the best of our knowledge, among different clinical bacterial isolates recovered from endodontic infections, E. faecalis is the only species that has been widely studied for its capacity to form biofilm (17, 18). If bacteria participate in gene exchange within a biofilm via horizontal gene transfer, processes leading to a spread of antibiotic resistance genes between different clinically relevant species can be accelerated. As summarized by Madson et al (19), horizontal gene transfer rates are typically higher in biofilm communities compared with those in planktonic niches. Thus, there is a connection between biofilm formation and horizontal gene transfer. In addition to this, the persistence of endodontic bacteria via biofilm formation underlines the necessity for more effective methods not only to completely eliminate bacteria during endodontic retreatment but also to isolate all of the existing microorganisms during the microbiological sampling from infected root canals. It should also be kept in mind that the complex anatomy of the root canal poses further difficulties because biofilms of persistent microorganisms within root canals may also be located on the walls of
From the Departments of *Operative Dentistry and Periodontology and †Hygiene and Microbiology, Albert-Ludwigs-University, Freiburg, Germany. Address requests for reprints to Dr Ali Al-Ahmad, Department of Operative Dentistry and Periodontology, University School and Dental Hospital, Hugstetter Straße 55, D-79106 Freiburg, Germany. E-mail address: [email protected] 0099-2399/$ - see front matter Copyright ª 2014 American Association of Endodontists. http://dx.doi.org/10.1016/j.joen.2013.07.023
JOE — Volume 40, Number 2, February 2014
Antibiotic Resistance and Capacity of Bacteria
223
Basic Research—Biology ramifications and isthmuses (15). The aim of the present study was to analyze the capacity for biofilm formation and a broad pattern of antibiotic sensitivity for a diverse array of clinical bacterial isolates obtained from patients with secondary endodontic infections in asymptomatic teeth with apical periodontitis.
Material and Methods Clinical Bacterial Isolates and Strain Maintenance A total of 21 patients were referred to the University Clinic and Dental Hospital, University of Freiburg, Freiburg, Germany, for endodontic retreatment over a period of 2 years. After microbiological sampling, 47 different bacterial isolates belonging to 32 different species were isolated from the infected root canals. All patients gave their written informed consent to the study protocol, which had been approved by the Ethics Committee (no. 140/09, University of Freiburg). During the pretreatment examination, the following clinical parameters were evaluated by an endodontist: sex and age of the patient, endodontic history, tooth type (ie, incisor, premolar, or molar in upper/lower jaw), clinical signs (ie, the presence of sinus tract and pus), type of previous endodontic treatment, and radiographic appearance. Indeed, preoperative radiographs were taken, all radiographs were digitized, and the radiographic quality of the pre-existing obturation was estimated. Endodontic treatment of all teeth had been completed at least 2 years before the study, and all of the teeth exhibited apical periodontitis on radiographic examination. In each of the cases, retreatment was indicated, and previous root canal treatment was considered a failure. In our attempt to focus on bacterially induced apical infections and to eliminate other causative factors, we assumed that mechanical microbial access to the apex is a prerequisite for the contamination of periapical tissues. In this context, teeth with obturation material that did not reach within 4 mm of the radiographic apex or that could not be isolated with a rubber dam were excluded from the study. No direct exposure of the root canal filling material to the oral cavity was evident. All of the teeth were asymptomatic. Patients who used antibiotics within the last 30 days before commencement of the study were excluded. A detailed description of the patients’ collective characteristics, the sampling procedure used, the isolation, and the identification of the recovered clinical isolates has been reported in earlier studies (11–13). Briefly, the tooth and surrounding area were cleaned with 30% hydrogen peroxide (H2O2) and swabbed with a 3% sodium hypochlorite solution (NaOCl). Endodontic access was achieved with a sterile high-speed carbide bur until the root filling was exposed. Subsequently, the tooth and the adjacent rubber dam were disinfected a second time using 30% H2O2 and 3% NaOCl. The cavity was swabbed with a 5% sodium thiosulfate solution to inactivate the NaOCl. To assess the efficacy of the disinfection, a sterile foam pellet was moistened in a sterile 0.9% NaCl solution and was used to swab the access cavity and the tooth surface. If bacterial growth occurred in these quality control samples, the tooth was excluded from the study. Coronal gutta-percha was removed with Gates Glidden drills. The working length was established radiographically and with the aid of an electronic apex locator (Raypex 5; VDW, Munich, Germany). The canal was enlarged from 0.5–2 mm from the radiographic apex with ProTaper NiTi instruments (Dentsply Maillefer, Ballaigues, Switzerland). Teeth that could not be instrumented to this length were excluded from the study. No solvent was used at any time. After introducing approximately 40 mL sterile saline solution (0.9% NaCl) into the canal with a sterile syringe, 3 sequential sterile paper points (ISO 25, taper 04; ROEKO, Langenau, Germany) were placed to the working length to soak 224
Al-Ahmad et al.
up the fluid. Each paper point was kept inside the canal for 1 minute and then transferred into a sterile vial containing 0.75 mL reduced transfer fluid. Finally, conventional retreatment was finished after root canal disinfection, and the root canal was filled by using vertical compaction. The long-term storage of all clinical isolates was accomplished at 80 C in basic growth medium containing 15% (v/v) glycerol as described by Jones et al (20).
Testing Antibiotic Sensitivity Using the Etest The testing of antibiotic sensitivity using the Etest method (Bio Merieux, Marcy-1’Etoile, France) was conducted according to the manufacturer’s instructions and as described in an earlier report (3). The standardized Etest was applied to reveal any antibiotic sensitivity for the different clinical bacterial isolates recovered from the infected root canals. Thus, the antibiotic sensitivity of aerobic and facultative anaerobic bacterial isolates was tested against 11 antibiotics including penicillin G, amoxicillin, clindamycin, gentamicin, vancomycin, tetracycline, doxycycline, fosfomycin, rifampicin, ciprofloxacin, and moxifloxacin. The antibiotic sensitivity of anaerobic bacteria was tested using 12 antibiotics, including those previously mentioned and metronidazole. For aerobic and facultative anaerobic bacteria, the Etest was conducted on Mueller-Hinton agar for E. faecalis isolates and the control strain E. faecalis (American Type Culture Collection [ATCC] 29212) or on Mueller-Hinton agar containing 5% horse blood for Streptococcus mutans, Streptococcus oralis, Streptococcus sanguinis, Streptococcus constellatus, Lactobacillus fermentum, Lactobacillus rhamnosus, Lactobacillus casei, Lactobacillus gasseri, Actinomyces naeslundii, Actinomyces viscosus, Moraxella osloensis, Eikenella corrodens, and both control strains Haemophilus influenzae (National Collection of Type Cultures [England] [NCTC] 8468) and Streptococcus pneumoniae (ATCC 49619). Colonies of each isolate were picked from an overnight agar plate and suspended in sterile saline (0.9%) to an inoculum turbidity of McFarland 0.5. Using sterile nontoxic spun swabs, each isolate was streaked on agar plates. The Etest strip was then applied to the agar surface using sterile tweezers. The anaerobic isolates belonged to the following species: Prevotella intermedia, Prevotella buccae, Fusobacterium nucleatum, Tannerella forsythia, Porphyromonas gingivalis, Campylobacter rectus, Campylobacter gracilis, Veillonella dispar, Dialister pneumosintes, Parvimonas micra, Atopobium rimae, Propionibacterium propionicum, Propionibacterium acidifaciens, Pseudoramibacter alactolyticus, Solobacterium moorei, Mogibacterium pumilum, Mogibacterium timidum, Bifidobacterium sp, and the control strain Bacteroides fragilis (ATCC 25285). For P. gingivalis, the Etest was conducted on yeast cysteine blood agar. For all other anaerobic isolates, Wilkins-Chalgren agar with 5% horse blood was used (anaerobic jar [AnaeroCultA; Merck, Darmstadt, Germany]). Colonies of each isolate were picked from a 48-hour subculture and suspended in sterile saline (0.9%) to an inoculum turbidity of 0.5–1 McFarland. Four milliliters of each suspended isolate were pipetted on the agar plate, excessive solution was then removed, and the plates were dried. The Etest strip was then applied to the agar surface using sterile tweezers. Nitrocefin discs (Becton Dickinson, Sparks, MD) were used for the detection of beta-lactamases. If available, the breakpoints according to the European Committee on Antimicrobial Susceptibility Testing and the epidemiologic cut-off values were used to compare the results. When European Committee on Antimicrobial Susceptibility Testing and epidemiologic cut-off values did not exist, minimal inhibitory concentration (MIC) values for similar strains were derived from the literature and used to compare the antibiotic sensitivity of the endodontic clinical isolates. JOE — Volume 40, Number 2, February 2014
JOE — Volume 40, Number 2, February 2014
0.19 0.094 0.016 0.023 0.032 1 NCTC, National Collection of Type Cultures (England); ND, not determined for facultative anaerobic bacterial isolates.
16 >1024 >1024 0.75 8
6
ND 0.006 0.25 0.38 0.094 0.004 192 0.125 3 4 0.38
0.19
ND 0.064 0.032 0.064 0.094 0.25 0.5 8 12 0.094 0.38
0.047
ND 0.125 0.500 4.000 16.000 0.380 4.000 32.000 8.000 0.500 6.000 2.000
E. faecalis ATCC 29212 S. pneumoniae ATCC 49619 H. influenzae NCTC 8468 B. fragilis ATCC 25285
TABLE 1. Antibiotic Sensitivity of All Tested Control Strains
Results Antibiotic Sensitivity Testing A total of 47 different bacterial isolates belonging to 32 different species were screened for antibiotic sensitivity. Table 1 shows the values of antibiotic sensitivity of all 4 control strains (E. faecalis, S. pneumoniae, H. influenzae, and B. fragilis) tested in parallel. The results of antibiotic sensitivity testing of aerobic, facultative anaerobic, and anaerobic bacteria are depicted in Tables 2 and 3, respectively. Only 2 E. faecalis isolates (nos. 3 and 6 in Table 2) showed MICs of 64 mg/L for tetracycline and 12 mg/L for doxycycline and can therefore be considered resistant to these antibiotics. Attention should be paid to S. mutans (no. 8 in Table 2) exhibiting high MIC (256 mg/L) for fosfomycin. All other Streptococcus strains as well as A. naeslundii and Actinomyces viscosus showed high MIC values in the range of 24–192 mg/L for fosfomycin. Lactobacilli presented high antibiotic resistance against fosfomycin (MIC >1024 mg/L). With an MIC higher than 256 mg/L, L. fermentum, L. rhamnosus, and L. casei also showed high antibiotic resistance against vancomycin. L. gasseri showed intermediary MIC values in the range of 1–1.5 mg/L for vancomycin but was resistant to ciprofloxacin (MIC >32 mg/L). Taking a look at the anaerobic isolates (Table 3), P. intermedia proved to be penicillin resistant (MIC 16 mg/L) and showed a positive beta-lactamase result. However, P. intermedia and Prevotella buccae revealed high MIC values for tetracycline and doxycycline. Compared with the break points found for aerobic bacteria, these 2 species could
Species
Biofilm Plate Assay The biofilm formation test was conducted as previously described (21). In brief, aerobic bacterial strains were grown in tryptic soy broth (Merck, Darmstadt, Germany) overnight at 37 C under aerobic conditions with 5% CO2 (capnophilic conditions), whereas anaerobic bacteria were grown in gas chromatography Hewlett-Packard (GCHP, Wilmington, DE) broth overnight under anaerobic conditions (anaerobic jars [GENbox anaer; bioMerieuxsa, Marcy l’Etoile, France]). The log10 of the colony-forming unit of each overnight culture as determined on Columbia blood agar or yeast cysteine blood agar was in the range of 108 colony-forming unit/mL. Polystyrene 96well tissue culture plates (Greiner Bio-One, Frickenhausen, Germany) were filled with either 180 mL fresh tryptic soy broth for aerobic bacteria or with GC-HP broth for anaerobic bacteria; afterward, 20 mL of the overnight culture were added to each well. The plates were incubated for 48 hours at 37 C in an aerobic atmosphere with 5% CO2 to cultivate biofilms of aerobic bacteria or under anaerobic conditions in anaerobic jars (AnaeroCult A) for biofilm formation of the anaerobic isolates. The culture medium was discarded, and the wells were washed 3 times with 300 mL phosphate-buffered saline to remove nonadherent bacteria. The plates were air dried and stained with 0.1% crystal violet (Median Diagnostics GmbH, Dunningen, Switzerland) for 10 minutes. Excess stain was removed by washing 3 times with 200 mL distilled water. The plates were dried for 10 minutes at 60 C. Fifty microliters of alcohol (99.9%, absolute for analysis, Merck) were added to each well for resolubilization of the dye. The optical density was measured at 595 nm with a Tecan Infinite 200 plate reader (Tecan, Crailsheim, Germany). All tests were performed in quadruplicate, whereas the experiments were conducted twice, and the mean values were determined. Three biofilm-forming categories were established according to 2 different cut-off values, which were fixed to define the following categories: biofilm nonproducer or C1, biofilm moderate producer or C2, and biofilm high producer or C3. A low cut-off value was fixed by adding 3 standard deviations of the blank to the negative control. The high cut-off value was defined as 3 times the low cut-off value.
Penicillin G Clindamycin Amoxicillin Gentamicin Fosfomycin Vancomycin Ciprofloxacin Tetracycline Doxycycline Rifampicin Moxifloxacin Metronidazol
Basic Research—Biology
Antibiotic Resistance and Capacity of Bacteria
225
Moxifloxacin
0.125 0.190 0.190 0.125 0.125 0.125 0.064 0.125 0.19 0.094 0.023 0.023 0.5 0.125 0.125 0.023 0.75 0.38 0.125 0.032 0.032 1.000 1.500 2.000 1.500 0.380 1.000 0.047 0.012 0.012 0.064 0.064 0.003 0.047 0.38 0.064 0.094 0.19 0.002 0.002 0.125 0.5 0.125 0.190 12.000 0.190 0.750 12.000 0.25 0.5 0.25 0.125 32 1.5 1 0.032 0.032 1.500 2.000 1.500 2.000 1.000 3.000 0.75 0.5 1 0.75 1 1 >256 >256 >256 >256 1 0.5 0.19 3 48
Vancomycin
3.000 8.000 6.000 8.000 4.000 8.000 2 0.75 4 4 0.75 0.5 0.125 2 3 0.5 3 0.38 0.125 0.047 3 0.094 1.000 0.750 0.750 1.000 0.750 0.023 0.016 0.064 0.032 0.047 0.023 0.125 2 0.5 0.064 0.25 0.032 0.016 0.047 0.25
Fosfomycin Gentamicin Amoxicillin
Discussion
6.000 16.000 6.000 2.000 3.000 8.000 0.023 0.023 0.094 0.032 0.032 256
Clindamycin
Biofilm Formation Besides testing the antibiotic sensitivity patterns for all of the aerobic and facultative anaerobic isolates and the anaerobic endodontic isolates, their capacity for biofilm formation was also examined. The different clinical isolates were divided into 3 categories including nonproducers of biofilm or C1, moderate biofilm producers or C2, and high biofilm producers or C3. The corresponding optical density (OD) cut-off values varied between 0.086 and 0.258. OD values lower than 0.086 represented biofilm nonproducers, values ranging between 0.086–0.258 were associated with moderate biofilm producers, and OD values higher than 0.258 corresponded to high biofilm producers. Figures 1 and 2 show the results concerning biofilm formation by aerobic and anaerobic bacteria, respectively. Among the aerobic bacteria (Fig. 1), 7 isolates comprising the species E. faecalis, S. anginosus, S. oralis, M. osloensis, and E. corrodens failed to form biofilms. Eight isolates belonging to the species E. faecalis, S. mutans, S. constellatus, L. rhamnosus, L. casei, and Lactobacillus gasseri were found to be biofilm moderate producers. Five isolates of the species E. faecalis. S. mutans, L. fermentum, A. naeslundii, and A. viscosus were characterized as biofilm high producers. Considering the anaerobic isolates (Fig. 2), 2 isolates comprising the species P. buccae and Propionibacterium acidifaciens were found to be high biofilm producers. Three isolates of the species F. nucleatum, V. dispar, and Propionibacterium propionicum were characterized as moderate biofilm producers. All other anaerobic bacterial isolates depicted in Figure 2 failed to form biofilms and were categorized as biofilm nonproducers. However, 1 isolate (no. 14) of the species Parvimonas micra tended to be a moderate biofilm producer.
0.500 4.000 4.000 3.000 3.000 3.000 0.008 0.012 0.047 0.032 0.016 0.016 0.25 2 0.5 0.094 0.094 0.016 0.006 0.5 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Penicillin G Species
E. faecalis E. faecalis E. faecalis E. faecalis E. faecalis E. faecalis S. mutans S. mutans S. mutans S. oralis S. anginosus S. constellatus L. fermentum L. rhamnosus L. rhamnosus L. casei L. gasseri A. naeslundii A. viscosus M. osloensis E. corrodens
No.
TABLE 2. Antibiotic Sensitivity of All Tested Aerobic and Facultative Anaerobic Endodontic Bacterial Isolates
226
be considered resistant to tetracycline and doxycycline. Tannerella forsythia exhibited resistance against ciprofloxacin (MIC >32 mg/ mL) and rifampicin (MIC = 32 mg/mL). Veillonella spp and Dialister spp showed typical resistance to gentamycin and vancomycin. P. alactolyticus and Mogibacterium timidum were resistant to tetracycline (MIC 8 and 4 mg/L, respectively) and doxycycline (MIC = 6 and 3 mg/mL, respectively). Solobacterium moorei presented a noticeably high MIC (32 mg/L) for rifampicin.
6.000 48.000 24.000 48.000 48.000 24.000 94 256 64 24 24 64 >1024 >1024 >1024 >1024 >1024 64 192 24 >1024
Doxycycline
Rifampicin
Basic Research—Biology
In various recent reports, a variety of bacteria have been isolated and identified from secondary endodontic infections. However, considering the broad assortment of bacterial species that are causing endodontic infections, very few reports have depicted their capacity for biofilm formation or sensitivity to different antibiotics. Both topics are of major clinical significance for general oral health and particularly for the prognosis of endodontic treatment. In this study, we present for the first time a characterization of the capacity of 47 clinical isolates belonging to 32 different species, which were gained entirely from secondary/persistent endodontic infections to form biofilms as well as their antibiotic sensitivity patterns. Antibiotic sensitivity testing not only constitutes an essential requirement for the treatment of infectious bacteria but is also an effective method to characterize them. The Etest is a standard test widely used in medical microbiology and shows acceptable antibiotic sensitivity results for most medically important antimicrobial drugs as well as different pathogens (22–24). For aerobic bacteria, a total of 11 different antibiotics were tested, whereas 12 were used for the anaerobic bacterial isolates. Furthermore, the comparison of antimicrobial susceptibility of different strains using this test was JOE — Volume 40, Number 2, February 2014
JOE — Volume 40, Number 2, February 2014
TABLE 3. Antibiotic Sensitivity of All Tested Anaerobic Endodontic Bacterial Isolates Species
1 2 3 4 5 6 7 8 9 10 11
P. intermedia P. buccae F. nucleatum F. nucleatum F. nucleatum F. nucleatum T. forsythia P. gingivalis C. rectus V. dispar Dialister pneumisintes P. micra P. micros P. micros Atopobium rimae A. rimae P. alactolyticus P. alactolyticus P. alactolyticus S. moorei M. pumilum M. timidum C. gracilis Bifidobacterium sp P. propionicum P. acidifaciens
12 13 14 15
Antibiotic Resistance and Capacity of Bacteria
16 17 18 19 20 21 22 23 24 25 26
Penicillin G Clindamycin Amoxicillin Gentamicin Fosfomycin Vancomycin Ciprofloxacin Tetracycline Doxycycline Rifampicin Moxifloxacin Metronidazol
Resistant bacteria are marked in bold.
16 0.047 0.064 0.016 0.008 0.002 0.002 0.002 0.008 0.125 0.19
0.008 0.012 0.016 0.016 0.016 0.016 0.016 0.008 256 >256
0.5 0.75 1.5 0.5 0.38 0.5 >32 0.25 0.047/0.25 0.064 0.047
3 8 0.25 0.064 0.047 0.016 0.32 0.047 0.023 0.064 0.064
3 3 0.19 0.047 0.047 0.012 0.32 0.094 0.023 0.094 0.094
0.047 0.125 0.5 0.125 0.125 0.047 32 0.003
