Microbial Analysis of Canals of Root-filled Teeth withPeriapical Lesions Using Polymerase Chain ReactionBrenda P.F.A. Gomes, PhD, MSc, BDS,* Ericka T. Pinheiro, PhD, MSc, BDS,†

Rogério C. Jacinto, PhD, MSc, BDS,† Alexandre A. Zaia, PhD, MSc, BDS,†

Caio C.R. Ferraz, PhD, MSc, BDS,† and Francisco J. Souza-Filho, PhD, MSc, BDS†

Abstract

The objective of the present study was to investigatethe presence of nine bacterial species in root-filledteeth associated with periapical lesions using a poly-merase chain reaction analysis and to correlate thesespecies with clinical features of the cases. DNA wasextracted from 45 canal samples of root-filled teethwith periapical lesions. A PCR assay using species-specific primers of 16S rDNA and the downstreamintergenic spacer region was used for microbial detec-tion. Enterococcus faecalis was the most prevalentspecies, detected in 77.8% of the study teeth, followedby Peptostreptococcus micros, detected in 51.1%. Por-phyromonas gingivalis, Porphyromonas endodontalis,Prevotella intermedia, and Prevotella nigrescens weredetected in 35.6%, 22.2%, 11.1%, and 11.1% of thesampled teeth, respectively. Moreover, PCR detectedFilifactor alocis in 26.7%, Treponema denticola in24.4%, and Tannerella forsythia in 4.4% of the sam-ples. T. denticola and P. micros were statistically asso-ciated with tenderness to percussion (p 0.05). P.nigrescens was associated with the presence of spon-taneous pain and abscess (p 0.05). P. endodontalisand P. nigrescens were associated with purulent exu-dates (p 0.05). Synergistic relationship was alsoobserved between some species. The results of thisstudy indicated that E. faecalis was the most frequentlyidentified test species by PCR in teeth with failingendodontic treatment. (J Endod 2008;34:537–540)

Key Words

Bacteria, endodontic failure, polymerase chain reaction

Persistent intraradicular or secondary infections seem to be the major causes of rootcanal treatment failure, which is usually characterized by radiographic evidence ofperiapical lesions (1). Culture and molecular methods have been used to detect bac-terial species from canals of root-filled teeth with persistent periapical lesions.

Enterococcus faecalis has been the most frequently identified species in canals ofroot-filled teeth with periapical lesions either by culture or molecular methods (2–11).Although E. faecalis possesses several virulence factors, its ability to cause periradicu-lar disease stems from its ability to survive the effects of root canal treatment and persistas a pathogen in the root canals and dentinal tubules of teeth and survive starvation (12,13). However, its prevalence has diverged among the studies, 12% to 77% of the cases(2–11, 14, 15). Moreover, some authors have failed to detect E. faecalis in retreatmentcases using culture ormolecularmethods (16, 17). These discrepanciesmay be causedby differences in patient selection, sampling and microbial methodology, sample size,or geographic location (18).

Anaerobic species, especially some difficult-to-grow bacteria, have been moreregularly detected in canals of teeth with failed endodontic treatment by molecularmethods than by culture. Pseudoramibacter alactolyticus, Propionibacteriumpropionicus, Filifactor alocis, and Dialister pneumosintes have been frequentlydetected by polymerase chain reaction (PCR) in a Brazilian study (8). However, asimilar study has failed to detect the latter species in retreatment cases in a SouthKorean population (19).

Porphyromonas gingivalis, Porphyromonas endodontalis, Prevotella interme-dia, and Prevotella nigrescens, have been found in relatively low prevalence valuesfrom teeth with failing endodontic treatment despite the use of more sensitive PCRtechniques (8, 20). Similarly, Treponema denticola has not been frequently reportedin such cases (8).

Culture-based studies have shown that the microbial flora of cases with failedendodontic treatment is limited to a small number of predominantly facultative gram-positive species (2–7). On the other hand, molecular-based studies have revealed amore complex microbiota in cases of endodontic failure (8–10). Therefore, not onlyspecies frequently detected by culture in retreatment cases such as E. faecalis and P.micros but also species not detected so often by culture such as P. intermedia, P.nigrescens, P. gingivalis, P. endodontalis, and species that are difficult to grow onagar plates but can be easily detected by PCR such as T. denticola, F. alocis, and T.forsythia,might be involved in the pathogenesis of endodontic failures, especially if theydevelop a synergistic relationship.

Data concerning the molecular detection of microorganisms from canals of teethwith failed endodontic treatment are limited and, particularly for some species, varywidely. Therefore, the aim of this research was to investigate the presence of ninebacterial species in root-filled teeth associated with periapical lesions using a PCRanalysis and to correlate these species with clinical features of the cases.

Materials and MethodsSubjects

Forty-five root canal samples were taken from patients who needed nonsurgicalendodontic retreatment and attended the Piracicaba Dental School, Sao Paulo, Brazil. Adetailed medical and dental history was obtained from each patient. Patients who had

From the *Department of Restorative Dentistry, Endodon-tic Division, Piracicaba Dental School, State University ofCampinas-UNICAMP, and the †Department of RestorativeDentistry, Endodontic Division, State University of Campinas-UNICAMP, Piracicaba, SP, Brazil.Supported by the Brazilian agencies FAPESP (04/05743-2,

05/51653-8, 06/60500-3), CNPq (304282/2003-0), and CAPES(PRODOC).Address requests for reprints to Dr Brenda P.F.A. Gomes,

Endodontia, Faculdade de Odontologia de Piracicaba, FOP-UNICAMP, Avenida Limeira, 901, Piracicaba, Sao Paulo 13414-900, Brazil. E-mail address: bpgomes@fop.unicamp.br.0099-2399/$0 - see front matterCopyright © 2008 by the American Association of

Endodontists.doi:10.1016/j.joen.2008.01.016

Clinical Research

JOE — Volume 34, Number 5, May 2008 Microbial Analysis of Root-filled Teeth 537

received antibiotic treatment during the last 3 months or had a generaldisease were excluded from the study. The human Volunteers Researchand Ethics Committee of the Dental School of Piracicaba approved thestudy, and all patients signed an informed consent.

All teeth had endodontic therapy performed more than 4 years butshowed radiographic evidence of apical periodontitis. The followingfeatures were recorded for each patient: tooth type, clinical symptoms,presence or absence of a sound coronal restoration (ie, a permanentrestoration that clinically and radiographically appeared sealed), car-ies, sinus, swelling of periodontal tissues, tenderness to percussion,mobility, periodontal status of the tooth, status of the root canal in termsof whether dry or wet (the term “wet canal” in this study means pres-ence of exudate), and the radiographic quality of the root canal filling(21). Teeth presenting mobility, periodontal pockets, or root fractureswere excluded from the study.

Sampling Procedure

All coronal restorations, posts, and carious defects were removed.After access cavity preparation, the teeth were individually isolated fromthe oral cavity with a rubber dam, and disinfection was performed using5.25% sodium hypochlorite. The sterility of the operation field waschecked after inactivation of the solution with 5% sodium thiosulfate.The sterility was checked by taking a swab sample of the cavity surfaceand streaking on to blood agar plates with subsequent incubation at37°C in both aerobic and anaerobic conditions, which presented neg-ative results in all cases. Aseptic techniques were used throughout rootcanal treatment and sample acquisition. The root filling was removed byusing Gates Glidden drills (Dentsply Maillefer, Ballaigues, Switzerland)and endodontic files without the use of chemical solvents. Irrigationwith sterile saline solution was performed in order to remove any re-maining materials and to moisten the canal before sample collection.

For microbial sampling, a sterile paper point was introduced intothe full length of the canal (as determined with a preoperative radio-graph) and kept in place for 60 seconds. In the cases that have beenpreviously irrigated with saline, as many paper points as possible wereused to absorb all liquid or fluid inside the canal. The paper points wereimmediately transferred to a test tube containing 1 mL of the VMGA IIItransport medium (Anaerobic Systems, Morgan Hill, CA) (22), whichwas frozen at!70°C.

Microbial Identification

For PCR detection of the target species, DNA was extracted (23)and was first amplified with prokaryotic universal ribosomal 16S and23S primers (785 and 422, respectively) (24, 25). Amplification wasthen performed of the 16S rDNA and the downstream intergenic spacerregion (ISR). Inclusion of the ISR provided an additional check of thespecificity of primers because the length of the ISR region varies among

bacterial species. E. faecalis, P.micros, P. gingivalis, P. endodontalis,P. intermedia, P. nigrescens, T. denticola, F. alocis, and T. forsythiaspecies were then identified by a second, nested amplification withspecies-specific 16S primers paired with a universal primer located inthe 23S gene (L189). Table 1 shows the primer sequences. All primerswere synthesized by Biosynthesis, Lewisville, TX.

PCR reactions were performed in a total volume of 50�L contain-ing 1.25 U TaqDNA polymerase (Perkin-Elmer, Foster City, CA), 5�L of10� PCR buffer plus 3mmol/LMgCl2, 0.25mmol/L of each primer, and0.2 mmol/L (each) deoxynucleoside triphosphates. For each sample,0.5 �L of extracted DNA was added to the reaction mixture. PCR wasalso performed by using a positive control (0.5 �L of DNA extractedfrom American Type Culture Collection [ATCC] species) and severalnegative controls.

For the first amplification, samples were subjected to 22 cyclesof denaturation at 94°C for 1 minute, annealing at 42°C for 2 min-utes, primer extension at 72°C for 3 minutes, and a final extensionof 72°C for 10 minutes. For the second amplification, the PCR re-action conditions were 30 cycles of 94°C for 1 minute, 52°C for 2minutes, and 72°C for 3 minutes. PCR amplification was performedin an automated Perkin-Elmer Cetus DNA Thermal Cycler PCR (Sci-entific Support, Hayward, CA).

PCR products were analyzed by 1% agarose gel electrophoresis,stained with ethidium bromide, and viewed under ultraviolet transillu-mination. A positive or negative identificationwas based on the presenceof clear bands of the expectedmolecular size using a 21-kb lambda DNAladder (Invitrogen Corporation, Carlsbad, CA).

PCR Primer Specificity

The species-specific primer in the 16S rDNA coding region wasselected based on previous investigation of endodontic bacteria by clon-ing and sequencing of the bacterial 16S gene (26) and on sequencesavailable in GenBank. The species specificity was further confirmed bysequencing at least one PCR product from a clinical sample for thespecific primer in an ABI Prism 310 automated sequencer (AME Bio-science Ltd, London, UK) (27).

Statistical Analysis

The data collected for each case (clinical features) were typedonto a spreadsheet and statistically analyzed using SPSS for Windows(SPSS Inc, Chicago, IL). The Pearson chi-square test or the one-sidedFisher exact test, as appropriate, was chosen to examine the null hy-pothesis that there is no relationship between endodontic symptomsand signs, and the presence of specific bacterial species. Fisher’s Exacttest was applied to test the null hypothesis that there was no relationshipbetween any pair of bacterial species recovered from the root canal(P"0.05). When there was a significant difference, indicating that

TABLE 1. Primers Used in This Study

Primers Specificity/Location/Orientation Sequence

Sm785 Universal primer/16S/785 bp from 5’end/forward GGATTAGATACCCTGGTAGTC422 Universal primer/23S/422 bp from 5’end/forward GGAGTATTTAGCTTL189 Universal primer/23S/forward GGTACTTAGATGTTTCAGTTCENFE E. faecalis/16S/forward GTCGCTAGACCGCGAGGTCATGAPmic Peptostreptoccus micros/16S/forward AACGAGAAGCGAGATAGAGATGTTAPRIN P. intermedia/16S/forward TGTTAGCGCCTTGCGCTAPRNI P. nigrescens/16S/forward CGTTGGCCCTGCCTGCGGPg13 P. gingivalis/16S/forward CATCGGTAGTTGCTAACAGTTTTCPe1 P. endodontalis/16S/forward TTTAGATGATGGCAGATGAGAGTdent T. denticola/16S/forward CAACAGCAATGAGATATGGFalo F. alocis/16S/forward ACATAACCAATGACAGCCTTTTTAABF4R Tannerella forsythia/16S/forward TGCGATATAGTGTAAGCTCTACAG

Clinical Research

538 Gomes et al. JOE— Volume 34, Number 5, May 2008

the relationship between the pair of species was present, the oddsratio was calculated. Positive associations were those with an aver-age odds ratio "2.0 and negative associations were those with anaverage odds ratio 0.5.

ResultsIn general, the target species were detected in 39 of 45 cases

(86.6%). Of the species studied, E. faecalis was the most prevalentspecies in canals of root-filled teeth associated with periapical lesionsdetected in 35 of 45 cases (77.8%). The prevalence values of the othermicrobial species were as follows: P. micros (23/45, 51.1%), P. gin-givalis (16/45, 35.6%), F. alocis (12/45, 26.7%), T. denticola (11/45,24.4%), P. endodontalis (10/45, 22.2%), P. intermedia and P. nigre-

scens (5/45, 11.1% both), and T. forsythia (2/45, 4.4%). The micro-bial findings and clinical features of each case are presented in Table 2.An average of 2.67 species per canal was found. T. denticola and P.microswere statistically associated with tenderness to percussion (p 0.05). P. nigrescens was associated with the presence of spontaneouspain and abscess (p 0.05). P. endodontalis and P. nigrescens wereassociated with purulent exudates (p 0.05). Positive associationswere found between P. endodontalis and F. alocis (odds ratio,14.000), P. gingivalis and P. endodontalis (odds ratio, 36.140), P.endodontalis and T. denticola (odds ratio, 9.000), F. nucleatum andG. morbilorum (odds ratio, 6.670), P. gingivalis and T. denticola(odds ratio, 8.600), and P. gingivalis and G. morbilorum (odds ratio,6.600).

TABLE 2. Clinical and Radiographic Features and Bacterial Findings in 45 Root-filled Teeth with Periapical Lesions

Patients T TAT R P HP TTP Sw W/D Si RCF PCR

1. 22 "4 NR N N N N D N PE —2. 41 "4 TR N N N N D N PE —3. 12 20 PR N N N N D N PE —4. 22 "2 TR N N N N W Y PE E. faecalis, P. gingivalis, P. endodontalis, P. nigrescens5. 11 "4 NR Y Y Y N D N GE E. faecalis, P. micros6. 22 "4 TR N N N N D N PE —7. 45 5 PR N N N N D N PE E. faecalis, P. gingivalis8. 11 "4 GR N N N N D N PE E. faecalis, P. micros, P. gingivalis, P. endodontalis,

P. intermedia, T. denticola, F. alocis9. 45 "4 NR Y Y Y N W N GE E. faecalis, P. gingivalis, P. endodontalis, P. intermedia,

P. nigrescens, T. forsythia, T. denticola, F. alocis10. 22 "2 GR N N Y N D N PE E. faecalis, P. gingivalis, P. micros, T. denticola11. 46 "2 NR N Y Y N W N PE P. gingivalis, P. endodontalis, P. nigrescens, P. micros,

T. denticola, F. alocis12. 12 5 PR N N N N D N PE E. faecalis, P. gingivalis, P. micros13. 12 20 PR N Y N N D Y PE E. faecalis, P. gingivalis, P. endodontalis, P. nigrescens,

T. forsythia, F. alocis14. 12 "2 TR N N Y N D N PE E. faecalis, P. gingivalis, P. endodontalis, P. micros15. 37 20 PR N N Y N D N GE E. faecalis, P. gingivalis, P. endodontalis, T. denticola, F. alocis,

P. micros16. 32 5 GR N N N N D N PE E. faecalis17. 22 5 PR N Y Y N D N GE E. faecalis18. 12 10 PR N Y Y N D Y PE E. faecalis, P. micros19. 15 "4 PR N N N N D N PE P. micros20. 11 10 PR N N N N D N PE E. faecalis, T. denticola21. 22 10 PR N N N N D N GE E. faecalis22. 21 5 NR N Y Y N D Y GE E. faecalis, T. denticola, F. alocis, P. micros23. 46 7 NR N N Y N D N PE P. micros24. 45 7 TR N Y N N D N GE E. faecalis, P. micros25. 37 5 PR N Y Y N D N GE E. faecalis, P. micros26. 11 8 PR N N N N D N GE —27. 35 "4 GR N N Y N D N PE E. faecalis28. 22 6 NR N N N N D N PE E. faecalis, P. gingivalis, P. micros29. 11 8 NR N Y N N D N PE E. faecalis30. 33 30 PR N N N N D N GE —31. 21 12 TR N N N N D N PE E. faecalis32. 23 5 PR N N N N D N PE E. faecalis33. 45 5 NR N Y N N D N GE E. faecalis34. 32 30 PR N N N N D N GE E. faecalis, P. gingivalis, P. endodontalis, F. alocis35. 37 4 PR N Y Y N D N GE E. faecalis, P. gingivalis, P. intermedia, P. micros, T. denticola36. 21 5 TR N Y N N D N GE E. faecalis, F. alocis37. 34 12 PR N N N N D N PE E. faecalis, P. gingivalis, P. endodontalis, P. intermedia,

P. nigrescens, P. micros, T. denticola38. 22 5 GR N N N N D N PE E. faecalis39. 12 15 GR N Y Y N D N PE E. faecalis, P. gingivalis, P. intermedia, P. micros, T. denticola,

F. alocis40. 21 11 GR N N N N D N PE E. faecalis, P. gingivalis, P. micros41. 14 11 NR N N N N D N PE E. faecalis, P. micros, F. alocis42. 46 "4 GR N N N N D N PE E. faecalis, P. micros, F. alocis43. 47 "4 GR N N N N D N PE E. faecalis, P. gingivalis, P. endodontalis, P. micros, F. alocis44. 41 8 PR N N N N D N PE P. micros45. 46 "2 GR N Y Y N D N PE E. faecalis, P. micros

T� tooth; TAT� time after treatment (y); R� restoration; GR� good restoration; PR� poor restoration; TR� temporary restoration; NR� no restoration; P� pain; HP� history of pain; TTP� tenderness

to percussion; Sw� swelling; W/D� wet (W) or dry (D) canal; Si� sinus tract; RCF� root canal filling; GE� good endodontic filling; PE� poor endodontic filling; Y� yes; N� no.

Clinical Research

JOE — Volume 34, Number 5, May 2008 Microbial Analysis of Root-filled Teeth 539

DiscussionPCR-based analyses have reported differences in the prevalence of

E. faecalis in cases of endodontic failure. This species was detected in77% (8) and 67% (9) of root-filled teeth with apical periodontitis in aBrazilian population and in 64% in a South Korean population (19). E.faecalis was also the most prevalent species detected in the presentstudy (77.8%). However, this prevalence is higher than that reported ina North American population, 22% (14) and 12.1% (15). The lowincidence of E. faecalis in the study of Kaufman et al. (15) could beexplained by factors such as patient selection and time after root filling.The high prevalence of E. faecalis may also be influenced by the geo-graphic location of the study population. Besides, such a differencecould be related to the presence of coronal leakage. In the presentstudy, microleakage (by defective coronal restorations, old temporaryrestorativematerials, or nonrestored teeth) was detected in themajorityof the teeth (35/45) and may have influenced the microbial findings.Nevertheless, primer sensitivity might influence the results of E. faecalisprevalence studies because certain primersmight have higher sensitivitythan others. Ke et al. (28), for example, developed a PCR-based assaythat improved the diagnosis of enterococcal infections by targeting thetuf gene, which encodes elongation factor EF-Tu.

PCR-based identification, not being dependent on bacterial viabil-ity, may not be as technique-sensitive as anaerobic culture. Despite theuse of VMGA III as a transport medium, which contains agar and otheringredients that could affect the DNA extraction, the amplification withuniversal primers detected the presence of DNA in all samples. The PCRmethod used in this study only detects targeted microbial species. Theresults showed that the target species were not detected in six cases.However, it is possible that species other than those studied could havebeen present in the examined teeth.

P. micros was the most frequently anaerobic species detected byPCR; it was present in half of the samples. All cases with positive PCRreactions harbored at least 1 gram-positive species, E. faecalis or P.micros. On the other hand, the prevalence of gram-negative anaerobicbacteria, particularly the black-pigmented species, was low when com-pared with primary root canal infections (16, 18, 20). These findingsconfirmed those reported in a previousmolecular study (8), despite thefact that P. micros, P. gingivalis, P. endodontalis, P. intermedia, andP. nigrescens, were detected in a higher prevalence in the present studyand association with clinical features could be observed.

This study confirmed that molecular techniques allow a more re-liable identification of diverse bacterial species, especially some diffi-cult-to-grow bacteria. Fastidious bacterial species were detected in theroot-filled canals studied by PCR. F. alocis, T. forsythia, and T. denti-cola were detected in, respectively, 26.7%, 24.4%, and 4.4% of thecases. However, a higher percentage of F. alocis (48%) and T. denti-cola (14%) have been previously reported in retreatment cases (8).

In conclusion, the results of this study showed that E. faecalis is acommon member of the microbiota of teeth with failing endodontictreatment detected by molecular technique. Similarly, gram-negativeanaerobic bacteria species commonly found in cases of primary end-odontic infections, such as P. gingivalis, P. endodontalis, P. interme-dia, P. nigrescens, F. alocis, T. forsythia, and T. denticola, are notfrequently detected in cases of endodontic failure but can be associatedwith clinical features and exert a synergistic relationship that might beinvolved in the pathogenesis of failed endodontic treatments.

AcknowledgmentWe would like to thank Drs Eugene J. Leys and Ann L. Griffen,

from the Ohio State University College of Dentistry, for providing thefacilities. We are also thankful to Dr Purnima S. Kumar and Ms

ZhengWang for all their help and Francisco J. Martinez and Adailtondos Santos Lima for technical support.

References1. Siqueira JF Jr. Aetiology of root canal treatment failure: whywell-treated teeth can fail.Int Endod J 2001;34:1–10.

2. Sundqvist G, Figdor D, Persson S, Sjogren U. Microbiology analysis of teeth withendodontic treatment and the outcome of conservative retreatment. Oral Surg OralMed Oral Pathol Oral Radiol Endod 1998;85:86–93.

3. Molander A, Reit C, Dahlen G, Kvist T. Microbiological status of root-filled teeth withapical periodontitis. Int Endod J 1998;31:1–7.

4. Peciuliene V, Balciuniene I, Eriksen HM, Haapasalo M. Isolation of Enterococcusfaecalis in previously root-filled canals in a Lithuanian population. J Endod2000;26:593–5.

5. Peciuliene V, Reynaud AH, Balciuniene I, HaapasaloM. Isolation of yeasts and entericbacteria in root-filled teeth with chronic apical periodontitis. Int Endod J2001;34:429–34.

6. Pinheiro ET, Gomes BP, Ferraz CC, Sousa EL, Teixeira FB, Souza-Filho FJ. Microor-ganisms from canals of root filled teeth with periapical lesions. Int Endod J2003;36:1–11.

7. Pinheiro ET, Gomes BP, Ferraz CC, Teixeira FB, Zaia AA, Souza Filho FJ. Evaluation ofroot canal microorganisms isolated from teeth with endodontic failure and theirantimicrobial susceptibility. Oral Microbiol Immunol 2003;18:100–3.

8. Siqueira JF Jr, Roças IN. Polymerase chain reaction-based analysis of microorgan-isms associated with failed endodontic treatment. Oral Surg Oral Med Oral PatholOral Radiol Endod 2004;97:85–94.

9. Roças IN, Siqueira JF Jr, Santos KRN. Association of Enterococcus faecalis withdifferent forms of periradicular diseases. J Endod 2004;30:315–20.

10. Schirrmeister JF, Liebenow AL, Braun G, Wittmer A, Hellwig E, Al-Ahmad A. Detectionand eradication of microorganisms in root-filled teeth associated with periradicularlesions: an in vivo study. J Endod 2007;33:536–40.

11. Hancock HH 3rd, Sigurdsson A, Trope M, Moiseiwitsch J. Bacteria isolated afterunsuccessful endodontic treatment in a North American population. Oral Surg OralMed Oral Pathol Oral Radiol Endod 2001;91:579–86.

12. Stuart CH, Schwartz SA, Beeson TJ, Owatz CB. Enterococcus faecalis: Its role in rootcanal treatment failure and current concepts in retreatment. J Endod 2006;32:93–8.

13. Davis JM, Maki J, Bahcall JK. An in vitro comparison of the antimicrobial effects ofvarious endodonticmedicaments on Enterococcus faecalis. J Endod 2007;33:567–9.

14. Fouad AF, Zerella J, Barry J, Spångberg LS. Molecular detection of Enterococcusspecies in root canals of therapy-resistant endodontic infections. Oral Surg Oral MedOral Pathol Oral Radiol Endod 2005;99:112–8.

15. Kaufman B, Spångberg L, Barry J, Fouad AF. Enterococcus spp. In endodonticallytreated teeth with and without periradicular lesions. J Endod 2005;31:851–6.

16. Rolph HJ, Lennon A, Riggio MP, et al. Molecular identification of microorganismsfrom endodontic infections. J Clin Microbiol 2001;39:3282–9.

17. Cheung GS, Ho MW. Microbial flora of root canal-treated teeth associated with asymp-tomatic periapical radiolucent lesions. Oral Microbiol Immunol 2001;16:332–7.

18. Baumgartner JC, Siqueira JF Jr, Xia T, Roças IN. Geographical differences in bacteriadetected in endodontic infections using polymerase chain reaction. J Endod 2004;30:141–4.

19. Roças IN, Jung IY, Lee CY, Siqueira JF Jr. Polymerase chain reaction identification ofmicroorganisms in previously root-filled teeth in a South Korean population. J Endod2004;30:504–8.

20. Gomes BP, Jacinto RC, Pinheiro ET, et al. Porphyromonas gingivalis, Porphyromo-nas endodontalis, Prevotella intermedia and Prevotella nigrescens in endodonticlesions detected by culture and by PCR. Oral Microbiol Immunol 2005;20:211–5.

21. Ray HA, Trope M. Periapical status of endodontically treated teeth in relation to thetechnical quality of the root filling and the coronal restoration. Int Endod J1995;28:12–8.

22. Dahlen G, Pipattanagovit P, Rosling B, Möller AJR. A comparison of two transportmedia for saliva and subgingival samples. Oral Microbiol Immunol 1993;8:375–82.

23. Leys EJ, Griffen AL, Strong SJ, Fuerst PA. Detection and strain identification of Actinoba-ciluys actinomycetemcomitans by nested PCR. J Clin Microbiol 1994;32:1288–94.

24. McClellan DL, Griffen AL, Leys EJ. Age and prevalence of Porphyromonas gingivalisin children. J Clin Microbiol 1996;34:2017–9.

25. Kumar PS, Griffen AL, Barton JA, Paster BJ, Moeschberger ML, Leys EJ. New bacterialspecies associated with chronic periodontitis. J Dent Res 2003;82:338–44.

26. Gomes BPFA, Griffen AL, Kumar PS, et al. Molecular and culture analysis of bacteriain infected root canals. J Dent Res 2004;83:1040SL.

27. Rumpf RW, Griffen AL, Wen BG, Leys EJ. Sequencing of the ribosomal intergenicspacer region for strain identification of Porphyromonas gingivalis. J ClinMicrobiol1999;37:2723–5.

28. Ke D, Picard FJ, Martineau F, et al. Development of a PCR assay for rapid detection ofenterococci. J Clin Microbiol 1999;37:3497–503.

Clinical Research

540 Gomes et al. JOE— Volume 34, Number 5, May 2008