Effect of extraction of partly erupted third molars on subgingival microorganisms Ari Rajasuo, DDS,a Jukka H. Meurman, MD, DDS, PhD,b Heikki Murtomaa, DDS, PhD, MPH,’ and Heini Torkko, MSC,~ Vekaranjiirvi, Kuopio, and Helsinki, Finland VALKEALA
MILITARY
HOSPITAL,
UNIVERSITY
OF KUOPIO,
AND
UNIVERSITY
OF HELSINKI
This study was made to investigate the effect of extraction of third molars on subgingival microbes in 39 generally and gingivally healthy men with an average age of 20.2 years (SD 0.9). Microbial samples were taken from the pericoronal space of symptom-free partly erupted lower third molars and from the adjacent gingival pockets of the second molars. The samples were cultivated anaerobically. All partly erupted third molars were extracted from 20 subjects. A control group of 19 subjects was lleft untreated. Microbe sampling was repeated 2 and 5 months postoperatively with highly significant results. It was shown that at baseline the number of black-pigmented gram-negative bacteria and Fusobacterium species was more frequent in third molar than in second molar sites. Tlie total bacterial count decreased significantly at the second molar sites after extraction of the third molars when compared with the control group. Before the extractions, black-pigmented gram-negative bacteria were detected in 45% of the test subjects and Actinobacillus actinomycetemcomitans in 20%. The respective postoperative figures were 30% for black-pigmented gram-negative bacteria and 10% for Actinobacillus actinomycetemcomitans. Capnocytophaga species were not affected by the extractions. The findings suggest that erupting third molars may harbor harmful bacteria that can be reduced by eradicating the foci. (ORAL SURG ORAL MED
ORAL PATHOL
1992;74:431-6)
P
revious studieson the indications for extraction of third molars have shown that in more than one half of the cases no preloperative symptoms had been recorded’ or that the first indication for extraction was “no opposingtooth” or malposition.2 The indications have been classified as prophylactic in 20% to 27% of the cases.2’3 Pathologic changeswere seenin 60% of the third molars.3 Pericoronitis, in particular, was the first diagnosisin 11% to 32% of the cases.‘?2 Caries was found only in 13% of the cases.’ Supported by the Medical Department of General Headquarters of Finnish Defence Forces. aCaptain, Valkeala Military Hospital, Finnish Defence Forces. bProfessor, Department of Preventive Dentistry and Cariology, Faculty of Dentistry, University of Kuopio. CAssociate Professor and Head, Department of Dental Public Health, University of Helsinki. dMicrobiologist, Laboratory of Oral Microbiology, University of Helsinki. 7112137541
Partly erupted lower third molars increase dental plaque on mandibular teeth up to the region of the second premolar.4 The removal of symptom-free partly erupted lower third molars decreasesthe number of salivary cariogenic bacteria, mutans streptococci and lactobacilli.5 Gingivally healthy persons may harbor presumptive periodontopathogens, Prevotella intermedia, Porphyromonas gingivalis and Actinobacillus actinomycetemcomitans (Aa),6-8 under the gingival flap of partly erupted lower third molars, and in the gingival pockets of the adjacent second molars.9 The purpose of this study was to investigate the effeet of the extraction of symptom-free, partly erupted lower third molars on the total bacterial count, partitularly on periodontopathogens in the lower retromolar region. We compared periodontopathogenic bacteria in the follicular area of the lower third molar with those in the gingival pocket of the adjacent secondmolar, before and after extraction.
431
432
Rajasuo et al.
ORALSURG
ORAL
3.4~~
C~ALPATHOL
October 1992
Table 1. Basic data, oral health indexes, data of the third molars, and antibiotic therapies of the test and control subjects at baseline examination Study
Fig. 4. Bacteria! samplingsites:spacebetweengingival flap and crown of lowerthird molar (arrow) and distal gingival pocket of adjacentsecondmolar (arrow).
MATERIAL AND METHODS Subjects and controls A group of 4.5 male conscripts who served at the garrison of Valkeala in southeastern Finland were invited to undergo a dental examination. The final study group consisted of 39 men with a mean age of 20.2 years (SD, 0.9, Table I). They were followed up for 5 months. Of the six conscripts who did not complete the study, two had been ordered to serve at other military bases,one had to terminate his service for health reasons, and three subjects were excluded because they had received more than two antibiotic treatments during the follow-up period. Twenty subjects with a mean age of 20.4 years (SD 1 year), who had at least one lower third molar occlusal surface partly visible, were selectedto be test subjects. Their lower third molars were considered to not have the potential for full eruption becauseof a lack of spaceor abnormal position. These teeth were extracted at one appointment. The following antibiotics were given postoperatively to 16 of the 20 test subjects for 7 days: phenoxymethyl penicillin (1 million IU, three times daily) or erythromycin acistrate (400 mg three times daily). An anti-inflammatory analgesic (tolfenamic acid, 100 mg three times daily) wasgiven for 2 days postoperatively and after that for a maximum of 3 days if needed. Nineteen conscripts with a mean age of 19.9 years (SD 0.6 year) were studied as a control group. The control group had similar partly erupted lower third molars but removal was not indicated because,on the basisof clinical judgment, the teeth were expected to erupt normally. The subjects and the controls were
group
No. of patients Age (SD) No. of smokers CP index,* No. of healthy gingivai sextants? Depth of sampled gingival pockets of second molars Depth of sampled follicle space of third molars (SD)
Test
Control
20 20.4 yr (1.0) 8 3.8
19 19.9 yr (0.6) 8 2.8
53.0 mm
53.0 mm
7.7 mm (1.8)
7.3 mm (1.2)
‘Community periodontal index for treatment tiainamo J. Bay I. Int Dent J 1975;25:229-35.
need.
otherwise in good general and oral health and none of them received antibiotics more than twice during the 5-month follow-up period. The check-ups were performed at least 1 month after the last doseof antibiotic had been taken. The antibiotic therapies were verified afterward from the medical records of the patients. Ethical consideration The informed-consent principle was followed throughout the study. The study plan had been approved by the ethical committee of the Medical Department of General Headquarters of Finnish Defence Forces. Clinical visits and samplings Dental check-ups and both lower third and adjacent second molar subgingival bacteria samplings were made at baseline before the extractions, and 2 and 5 months later from the secondmolar sites. The same protocol was used for the control group, with additional triple samplesfrom both the second and third molar sites. No teeth were extracted from the controls, and the control group, of course, did not receive any antibiotics for that reason. Dental plaque was taken from the distal gingival pocket of the lower secondmolar by scraping with a sterile curette with care to avoid saliva and supragingival plaque. Third molar sampleswere taken by a paper-point method under the gingival flap of each symptom-free, partly erupted tooth with no visible exudate (Fig. 1). 9,lo One 0.8 mm thick sterile endodontic paper point was placed as deep as possiblebetween the crown and gingival flap of the partly
Volume 74 Number 4
erupted lower third molar for 30 seconds. At the same time, the retromolar gingiva was pressed to get follicle exudate if present. Saliva and supragingival plaque were carefully removed from the region with a suction and cotton swabs. The samples were placed into a nonnutritious medium (viability-preserving medium of the University of Gothenburg No. III, VMG III).’ 1 This anaerobic medium has been shown to be suitable for long transport times.12, t3 Microbiologic methods
The plaque samples were transported at room temperature to a bacteriologic laboratory within 22 hours. In the laboratory, the samples were dispersed by Vortex mixing for 30 seconds and then serially diluted. Aliquots of 0.1 ml were plated on selective and nonselective media. ‘Nondiluted sample and dilution 10-l were plated on TSBV agar, which is a selective medium for Aa. l4 Dilutions 10m3and 1OW4were plated on nonselective Brucella medium. supplemented with 5% horse blood (BBL Microbiology Systems, Becton Dickinson & Co., Cockeysville, Md.). The plates were incubated anaerobically at $37” C, the TSBV plates were in a glove box for 3 days, and the Brucella plates were in anaerobic jars for 7 days. All plates were examined under a stereomicroscope. The colonies on TSBV agar that were suspected to be Aa on the basis of their morphology, were tested for a positive catalase reaction. The percentage of Aa was determined by a comparison of the counts of Aa on TSBV medium with the total bacterial counts on the nonselective Brucella medium. Presumptive Capnocytophaga sp., based on colony morphology and Gra.m stain, were also recorded from TSBV medium. The bacteria were further confirmed to give a negative reaction in the catalase, oxidase, and indole reactions,. The colonies were counted on the nonselective Bruce/la medium, and the total viable count was determined as the weig,hed average of the colony counts. The amount of black-pigmented gram-negative bacteria (BPG-) was determined by counting the darkpigmented colonies (BPG- wa.s previously named black-pigmented bacteroides, or BPBs, and included suspected periodontopathogens, Porphyromonas gingivalis [formerly Bacteroides gingivalis] and Prevotella intermedia [formerly Bacteroides intermedius]). The presumptive colonies of P. gingivaZis were tested for a positive CAAM15 and negative MUG reaction.16 The presence of Fusobacterium-like organisms was determined by colony morphology, Gram stain, and a positive indole reaction.
Periodontopathogens in 3rd molar region
433
Statistics
Statistical analyses were made with two-way analyses of variance and procedures of the Statistical Analyses System. l7 The statistically significant differences were accepted on the 5% level. RESULTS
The basic data of the test subjects and controls are shown in Table I. Gf those molars extracted, a vertical position was observed in two of three cases (+lO degrees to - 10 degrees around vertical),‘* whereas in the other cases the teeth were in mesioangular or distoangular position. A total of 49 third molars were extracted, 32 taken from the mandible and 17 taken from the maxilla. All molars in the lower jaw were partly erupted, whereas all but three of the extracted upper third molars were fully erupted. In the mean, the test group had 1.45 partly erupted mandibular and 0.15 maxillary third molars per subject vs. 1.47 and 0.11 in the control subjects, respectively. The test subjects received 26 antibiotic treatments (1.3 per subject) compared with 8 treatments in the control subjects (0.42 per subject) during the 5-month follow-up period. Of the 20 subjects who had molars extracted, 13 received phenoxymethyl penicillin for 1 week postoperatively, and three subjects received erythromycin. Ten antibiotic treatments (0.5 per subject) were recorded during the follow-up period because of general illness. Four of the test subjects had doxycycline prescriptions compared with three of the controls. In addition, six prescriptions for phenoxymethyl penicillin, three for erythromycin and one sulfa-trimethoprim and one kefalexin prescription had ‘been given. Fig. 2 shows the microbial counts at baseline examination in the lower third and second molar sites. A total of 56% of the third molars were positive for BPG-, which corresponded to 3.7% of the total bacterial count, whereas BPG- were found in only one second molar site (1.9%). The difference was statistically highly significant (p < 0.001). If Aa was detected at the third molar site, then 40% of the adjacent second molar sites were Au-positive as well. In the cases of Fusobacterium sp. detected at the second molar sites, only 2 of 13 adjacent third molars were negative for this bacterium. At baseline, the mean total bacterial count was 2.7 X lo7 in the third molar sites. The count was 1.4 X lo7 in the second molar sites. This difference was also statistically highly significant (p < 0.001). In the control subjects, the total bacterial count in the second lower molar sampling sites increased sig-
434
Rajaswo
et al.
~RA~SCRG
ORAL
!&ED
URAL
PAT~+OL
October 1992
NO. OF BACTERIA X I MILLION
CAPNO
TOTAL
Fig. 2. Distribution of positivesampiesof cultivated microorganisms in third and.secondmolar sampling sitestaken at baseline.Actinobacillus actinomycetemcomitans (Aa), black-pigmentedgram-negativebacteria (BPG-), Fusobacterium (FUSO) and Capnocytophaga (CAPNO) speciesareshown. The numberof total bacterial count is shownseparately.***Highly significantdifferencein the third molar sampleswhencomparedwith the secondmolars(p < 0.001). TOTAL
BACTERIAL
COUNT
AT
X lo6
BASELINE
AFTER
5 MONTHS
Fig. 3. Effect of extraction of lower third molarson total bacterial count in secondlower molar sampling site. Differencewasstatistically significant (p = 0.044). nificantly from the baseline examination to the 5-month follow-up (p = 0.044), whereas a decrease was observed in the test subjects (Fig. 3). Table II shows that the species of BPG- in the controls increased significantly (p = 0.016) during the follow-up period, whereas Fusobacterium sp. increased in the test subjects (JJ = 0.033). No difference was observed in Aa cultivations between the groups. DISCUSSION Our longitudinal findings support the results of Mombelli et a1.,gwho found periodontopathogens in gingivally healthy persons in the area of partly erupted lower third molars. Our hypothesis was that the extraction of partly
erupted third molars may cause a decreasein pathogenic microbes in a wider area than the third molar site. The total bacterial count decreasedin the gingival pockets of the second lower molars (Fig. 3), whereas the number of Fusobacterium speciesand BPG- bacteria increased (Table II). Aa was not very common in either the third or second molar sites. Our 5-month follow-up period might have been too brief for the thorough healing of the extraction wounds. Retention sites are often found more than 5 months after lower third molar surgery. Thus the increasednumber of somebacteria in the secondmolar sitesmight have been causedby microbes harbored in the extraction wound. i9 The number of bacteria could have been decreased in a longer follow-up period,
Periodontopathogens
in 3rd molar region
Volume Number
74 4
Table
II. Effect of extraction of lower third molars on bacterial flora in second lower molar sampling site Test group Follow-up Bacteria
AA BPGFUSO CAPNO
/
Baseline
1
Control
period 2 mo
2* 0 5 13
*Number of culture positive patients. lStatistica1 difference between the test and control NS = Not significant. Aa = Actinobacillus actinomycetemcomitans. BPG= Black pigmented gram-negative bacteria. FUSO = Fusobacterium species. CAPNO = Capnocytophaga species.
2 4 14 11
group
Follow-up 1
5 mo 2 6 15 14
Baseline
435
period 2 1120
0 1 8 12
0 6 10 11
5mo 0 6 10 I
Signi$cance
f
NS p = 0.016 p = 0.033 p = 0.064
subjects.
which practical reasons prevented in the present study protocol. Another explanation for the increased microbial counts in the distal gingival pockets of the second molars might be that the postoperative curettage sampled more subgingival microbes because it was easier to reach the sampling sites after the third molars had been extracted. However, increased counts were also observed in the controls and thus the decreased total bacterial count in the test subjects would not support any difference because of the sampling technique. The paper-point method has been shown to give an acceptable degree of reproducibility for the recovery of Aa and P. gingivalis, but not for P. intermedia.1° The paper-point technique was compatible with the curette-scraping method when the total viable counts were compared with these two alternative methods.12 Systemic antibiotics affect the counts of oral microbes, and their use:must naturally be taken into account in the interpretation of microbial findings.20-22 The study by Asikainen et al. observed the efficacy of doxycycline in juvenile periodontitis.21 There was a decrease in the amount of P. intermedia and F. nucleatum during the administration of the drug, but recovery was observed in 6 weeks. The proportions of other periodontopathogenic bacteria were quite stable in 5 to 8 mm deep gingival pockets in spite of the antibiotics. However, we cannot totally rule out the effect of antibiotics on our results. Ethical reasons prevented us from prescribing the same antibiotics for the control subjects as for the test subjects. During the Vietnam war, pericoronitis was the second most frequent dental emergency of the U.S. Navy and Marine personnel, after dental caries.23 The highest incidence of complications with the eruption of third molars and pericoronitis around these teeth
occurs around 20 years of age.24 Nitzan et a1.25 showed that 8 1% of the pericoronitic patients were 20 to 29 years old and that the pathologic condition occurred in the lower third molar region in 95% of the cases. Extraction of third molars has long been recommended if there is no possibility for the crown to erupt. For example, third molar extractions were indicated in 82% of university students aged 20 to 21 years, and the average was 2.5 third molars per student.24 The extraction should be performed before the patient’s twenty-fifth birthday. In older patients the healing is more complicated because of the intrabony defects on the distal aspect of the lower second molars when examined 2 to 4 years after surgery.26 In conclusion, the present study showed that the presumptive periodontopathogens should be taken into account when the indications for removal of third molars are being considered. The follicle space or gingival pocket in the distal area that surrounds the lower third molar may indeed be a reservoir for a variety of microorganisms. We thank Mrs. Kirsti Kari, MSc, Laboratory of Oral Microbiology of the University of Helsinki, for her guidance and financial support in the microbial cultivations. We also thank Mr. Jussi Metteri, MSc, General Headquarters of Finnish Defence Forces, for statistics. REFERENCES 1. Lysell L, Rohlin the mandibular 1988;17:161-4. 2. Schroeder DC, of third molars 3. Nordenram A, tions for surgical of 2,630 cases.
M. A study of indications used for removal of third molar. Int .I Oral Maxillofac Surg Cecil JC, Cohen ME. Retention and extraction in naval personnel. Milit Med 1983;148:50-3. Hultin M, Kjellman 0, Ramstrom G. Indicaremoval of the mandibular third molar: study Swed Dent J 1987;11:23-9.
436
Rajasuo
et al.
~Ck4iSb~G
ORAL
MEDORM
PATHOL
October 4. Ylipaavalniemi P, Turtola L, Rytijmaa I, Helminen S, Jauhiainen L. Effect of position of wisdom teeth on the visible plaque index and gingival bleeding index. Proc Finn Dent Sot 1982;78:47-9. 5. Rajasuo A, Meurman JH, Metteri J, Ankkuriniemi 0. Effect of extraction of partly erupted third molars on salivary microbial counts in conscripts. Caries Res 1990;24:273-8. 6. Slots J, Listgarten MA. Bacteroides gingivalis, bacteroides intermedius and actinobacillus actinomycetemcomitans in human periodontal diseases. J Clin Periodontol1988;15:85-93. 7. Listgarten MA, Slots J, Rosenberg J, Nitkin L, Sullivan P, Oler J. Clinical and microbiological characteristics of treated periodontitis patients on maintenance care. J Periodontol 1989;60:452-9. 8. Choi JI, Nakagawa T, Yamada S, Takazoe I, Okuda K. Clinical, microbiological, and immunological studies on recurrent periodontal disease. J Clin Periodontol 1990;17:426-34. 9. Mombelli A, Buser D, Lang NP, Berthold H. Suspected periodontopathogens in erupting third molar sites of periodontally healthv individuals. J Clin Periodontol 1990:17:48-54. M, Egelberg’J. Reproducibil10. Dahlei G, Renvert S, Wikstriim ity of microbiological samples from periodontal pockets. J Clin Periodonto 1990;17:73-7. 11. Mijller AJR. Microbiological examination of root canals and periapical tissues of human teeth. Odontol Tidskr 1966;74:1380. 12. Dahlen G, Manji F, Baelum V, Fejerskov 0. Black-pigmented Bacteroides species and Actinobacillus actinomycetemeomitans in subgingival plaque of adult Kenyans. J Clin Periodonto1 1989;16:305-10. P, Neut C, Charon J, Romond C. In 13. Deveaux E, Delplanque vitro survival of Bacteroides intermedius in five transport media, alone or in the presence of Streptococcus sanguis. Oral Microbial Immunol 1990;5:320-3. 14. Slots J. Selective medium for isolation of Actinobacillus actinomycetemcomitans. J Clin Microbial 1982; 15:606-9. 15. Slots J. Detection of colonies of Bacteroides gingivalis by a rapid fluorescence assay for trypsin-like activity. Oral Microbiol Immunol 1987;2:139-41.
16. Slots
17.
18. 19. 20.
21.
22.
23.
24.
25.
26.
1992
J. Rapid identification of important periodontal microorganisms by cultivation. Oral Microbial Immunol 1986;1:4855. Freund RJ, Littell RC, Spector PC. SAS system for linear models. Cary, NC: SAS Institute Inc., 1986. Tetsch P, Wagner W. Operative extraction of wisdom teeth. London: Wolfe Medical Publications Ltd, 1985:1&-21. MacGregor AJ, Hart P. Bacteria of the extraction wound. J Oral Surg 1970;28:885-7. Baker PJ, Evans RT, Slots J, Genco RJ. Antibiotic susceptibility of anaerobic bacteria from the human oral cavity. J Dent Wes 1985;64:1233-44. Asikainen S, Jousimies-Somer H, Kanervo A, Saxen L. The immediate efficacy of adjunctive doxycycline in treatment of localized juvenile periodontitis. Arch Oral Biol 1990;35(suppl):23 l-4. Kinder SA, Holt SC, Korman KS. Peniciliin resistance in the subgingival microbiota associated with adult periodontitis. J Clin Microbial 1986;23:1127-33. Ludwick WE, Pogas JA, Gendron EG, Weldon AL. Dental emergencies occurring among Navy-Marine personnel serving in Vietnam. Milit Med 1974;139:121-3. Ylipaavalniemi P, Turtola L, Murtomaa H, Rytiimaa I. Evaluation of the need for third molar removals among 20- to 21year-old Finnish university students. Proc Finn Dent Sot 1985;81:222-5. Nitzan DW, Tal 0, Sela MN, Shteyer A. Pericoronitis: a reappraisal of its clinical and microbiologic aspects. J Oral Maxillofac Surg 1985;43:510-6. Kugelberg CF. Periodontal healing two and four years after impacted lower third molar surgery: a comparative retrospeciive study. Int J Oral Maxillofac Surg 1990;19:341-5.
Reprint requests: Ari Rajasuo, DDS Valkeala Military Hospital PO Box 5, SF-46141 Vekaranjgrvi, Finland
MILITARY
HOSPITAL,
UNIVERSITY
OF KUOPIO,
AND
UNIVERSITY
OF HELSINKI
This study was made to investigate the effect of extraction of third molars on subgingival microbes in 39 generally and gingivally healthy men with an average age of 20.2 years (SD 0.9). Microbial samples were taken from the pericoronal space of symptom-free partly erupted lower third molars and from the adjacent gingival pockets of the second molars. The samples were cultivated anaerobically. All partly erupted third molars were extracted from 20 subjects. A control group of 19 subjects was lleft untreated. Microbe sampling was repeated 2 and 5 months postoperatively with highly significant results. It was shown that at baseline the number of black-pigmented gram-negative bacteria and Fusobacterium species was more frequent in third molar than in second molar sites. Tlie total bacterial count decreased significantly at the second molar sites after extraction of the third molars when compared with the control group. Before the extractions, black-pigmented gram-negative bacteria were detected in 45% of the test subjects and Actinobacillus actinomycetemcomitans in 20%. The respective postoperative figures were 30% for black-pigmented gram-negative bacteria and 10% for Actinobacillus actinomycetemcomitans. Capnocytophaga species were not affected by the extractions. The findings suggest that erupting third molars may harbor harmful bacteria that can be reduced by eradicating the foci. (ORAL SURG ORAL MED
ORAL PATHOL
1992;74:431-6)
P
revious studieson the indications for extraction of third molars have shown that in more than one half of the cases no preloperative symptoms had been recorded’ or that the first indication for extraction was “no opposingtooth” or malposition.2 The indications have been classified as prophylactic in 20% to 27% of the cases.2’3 Pathologic changeswere seenin 60% of the third molars.3 Pericoronitis, in particular, was the first diagnosisin 11% to 32% of the cases.‘?2 Caries was found only in 13% of the cases.’ Supported by the Medical Department of General Headquarters of Finnish Defence Forces. aCaptain, Valkeala Military Hospital, Finnish Defence Forces. bProfessor, Department of Preventive Dentistry and Cariology, Faculty of Dentistry, University of Kuopio. CAssociate Professor and Head, Department of Dental Public Health, University of Helsinki. dMicrobiologist, Laboratory of Oral Microbiology, University of Helsinki. 7112137541
Partly erupted lower third molars increase dental plaque on mandibular teeth up to the region of the second premolar.4 The removal of symptom-free partly erupted lower third molars decreasesthe number of salivary cariogenic bacteria, mutans streptococci and lactobacilli.5 Gingivally healthy persons may harbor presumptive periodontopathogens, Prevotella intermedia, Porphyromonas gingivalis and Actinobacillus actinomycetemcomitans (Aa),6-8 under the gingival flap of partly erupted lower third molars, and in the gingival pockets of the adjacent second molars.9 The purpose of this study was to investigate the effeet of the extraction of symptom-free, partly erupted lower third molars on the total bacterial count, partitularly on periodontopathogens in the lower retromolar region. We compared periodontopathogenic bacteria in the follicular area of the lower third molar with those in the gingival pocket of the adjacent secondmolar, before and after extraction.
431
432
Rajasuo et al.
ORALSURG
ORAL
3.4~~
C~ALPATHOL
October 1992
Table 1. Basic data, oral health indexes, data of the third molars, and antibiotic therapies of the test and control subjects at baseline examination Study
Fig. 4. Bacteria! samplingsites:spacebetweengingival flap and crown of lowerthird molar (arrow) and distal gingival pocket of adjacentsecondmolar (arrow).
MATERIAL AND METHODS Subjects and controls A group of 4.5 male conscripts who served at the garrison of Valkeala in southeastern Finland were invited to undergo a dental examination. The final study group consisted of 39 men with a mean age of 20.2 years (SD, 0.9, Table I). They were followed up for 5 months. Of the six conscripts who did not complete the study, two had been ordered to serve at other military bases,one had to terminate his service for health reasons, and three subjects were excluded because they had received more than two antibiotic treatments during the follow-up period. Twenty subjects with a mean age of 20.4 years (SD 1 year), who had at least one lower third molar occlusal surface partly visible, were selectedto be test subjects. Their lower third molars were considered to not have the potential for full eruption becauseof a lack of spaceor abnormal position. These teeth were extracted at one appointment. The following antibiotics were given postoperatively to 16 of the 20 test subjects for 7 days: phenoxymethyl penicillin (1 million IU, three times daily) or erythromycin acistrate (400 mg three times daily). An anti-inflammatory analgesic (tolfenamic acid, 100 mg three times daily) wasgiven for 2 days postoperatively and after that for a maximum of 3 days if needed. Nineteen conscripts with a mean age of 19.9 years (SD 0.6 year) were studied as a control group. The control group had similar partly erupted lower third molars but removal was not indicated because,on the basisof clinical judgment, the teeth were expected to erupt normally. The subjects and the controls were
group
No. of patients Age (SD) No. of smokers CP index,* No. of healthy gingivai sextants? Depth of sampled gingival pockets of second molars Depth of sampled follicle space of third molars (SD)
Test
Control
20 20.4 yr (1.0) 8 3.8
19 19.9 yr (0.6) 8 2.8
53.0 mm
53.0 mm
7.7 mm (1.8)
7.3 mm (1.2)
‘Community periodontal index for treatment tiainamo J. Bay I. Int Dent J 1975;25:229-35.
need.
otherwise in good general and oral health and none of them received antibiotics more than twice during the 5-month follow-up period. The check-ups were performed at least 1 month after the last doseof antibiotic had been taken. The antibiotic therapies were verified afterward from the medical records of the patients. Ethical consideration The informed-consent principle was followed throughout the study. The study plan had been approved by the ethical committee of the Medical Department of General Headquarters of Finnish Defence Forces. Clinical visits and samplings Dental check-ups and both lower third and adjacent second molar subgingival bacteria samplings were made at baseline before the extractions, and 2 and 5 months later from the secondmolar sites. The same protocol was used for the control group, with additional triple samplesfrom both the second and third molar sites. No teeth were extracted from the controls, and the control group, of course, did not receive any antibiotics for that reason. Dental plaque was taken from the distal gingival pocket of the lower secondmolar by scraping with a sterile curette with care to avoid saliva and supragingival plaque. Third molar sampleswere taken by a paper-point method under the gingival flap of each symptom-free, partly erupted tooth with no visible exudate (Fig. 1). 9,lo One 0.8 mm thick sterile endodontic paper point was placed as deep as possiblebetween the crown and gingival flap of the partly
Volume 74 Number 4
erupted lower third molar for 30 seconds. At the same time, the retromolar gingiva was pressed to get follicle exudate if present. Saliva and supragingival plaque were carefully removed from the region with a suction and cotton swabs. The samples were placed into a nonnutritious medium (viability-preserving medium of the University of Gothenburg No. III, VMG III).’ 1 This anaerobic medium has been shown to be suitable for long transport times.12, t3 Microbiologic methods
The plaque samples were transported at room temperature to a bacteriologic laboratory within 22 hours. In the laboratory, the samples were dispersed by Vortex mixing for 30 seconds and then serially diluted. Aliquots of 0.1 ml were plated on selective and nonselective media. ‘Nondiluted sample and dilution 10-l were plated on TSBV agar, which is a selective medium for Aa. l4 Dilutions 10m3and 1OW4were plated on nonselective Brucella medium. supplemented with 5% horse blood (BBL Microbiology Systems, Becton Dickinson & Co., Cockeysville, Md.). The plates were incubated anaerobically at $37” C, the TSBV plates were in a glove box for 3 days, and the Brucella plates were in anaerobic jars for 7 days. All plates were examined under a stereomicroscope. The colonies on TSBV agar that were suspected to be Aa on the basis of their morphology, were tested for a positive catalase reaction. The percentage of Aa was determined by a comparison of the counts of Aa on TSBV medium with the total bacterial counts on the nonselective Brucella medium. Presumptive Capnocytophaga sp., based on colony morphology and Gra.m stain, were also recorded from TSBV medium. The bacteria were further confirmed to give a negative reaction in the catalase, oxidase, and indole reactions,. The colonies were counted on the nonselective Bruce/la medium, and the total viable count was determined as the weig,hed average of the colony counts. The amount of black-pigmented gram-negative bacteria (BPG-) was determined by counting the darkpigmented colonies (BPG- wa.s previously named black-pigmented bacteroides, or BPBs, and included suspected periodontopathogens, Porphyromonas gingivalis [formerly Bacteroides gingivalis] and Prevotella intermedia [formerly Bacteroides intermedius]). The presumptive colonies of P. gingivaZis were tested for a positive CAAM15 and negative MUG reaction.16 The presence of Fusobacterium-like organisms was determined by colony morphology, Gram stain, and a positive indole reaction.
Periodontopathogens in 3rd molar region
433
Statistics
Statistical analyses were made with two-way analyses of variance and procedures of the Statistical Analyses System. l7 The statistically significant differences were accepted on the 5% level. RESULTS
The basic data of the test subjects and controls are shown in Table I. Gf those molars extracted, a vertical position was observed in two of three cases (+lO degrees to - 10 degrees around vertical),‘* whereas in the other cases the teeth were in mesioangular or distoangular position. A total of 49 third molars were extracted, 32 taken from the mandible and 17 taken from the maxilla. All molars in the lower jaw were partly erupted, whereas all but three of the extracted upper third molars were fully erupted. In the mean, the test group had 1.45 partly erupted mandibular and 0.15 maxillary third molars per subject vs. 1.47 and 0.11 in the control subjects, respectively. The test subjects received 26 antibiotic treatments (1.3 per subject) compared with 8 treatments in the control subjects (0.42 per subject) during the 5-month follow-up period. Of the 20 subjects who had molars extracted, 13 received phenoxymethyl penicillin for 1 week postoperatively, and three subjects received erythromycin. Ten antibiotic treatments (0.5 per subject) were recorded during the follow-up period because of general illness. Four of the test subjects had doxycycline prescriptions compared with three of the controls. In addition, six prescriptions for phenoxymethyl penicillin, three for erythromycin and one sulfa-trimethoprim and one kefalexin prescription had ‘been given. Fig. 2 shows the microbial counts at baseline examination in the lower third and second molar sites. A total of 56% of the third molars were positive for BPG-, which corresponded to 3.7% of the total bacterial count, whereas BPG- were found in only one second molar site (1.9%). The difference was statistically highly significant (p < 0.001). If Aa was detected at the third molar site, then 40% of the adjacent second molar sites were Au-positive as well. In the cases of Fusobacterium sp. detected at the second molar sites, only 2 of 13 adjacent third molars were negative for this bacterium. At baseline, the mean total bacterial count was 2.7 X lo7 in the third molar sites. The count was 1.4 X lo7 in the second molar sites. This difference was also statistically highly significant (p < 0.001). In the control subjects, the total bacterial count in the second lower molar sampling sites increased sig-
434
Rajaswo
et al.
~RA~SCRG
ORAL
!&ED
URAL
PAT~+OL
October 1992
NO. OF BACTERIA X I MILLION
CAPNO
TOTAL
Fig. 2. Distribution of positivesampiesof cultivated microorganisms in third and.secondmolar sampling sitestaken at baseline.Actinobacillus actinomycetemcomitans (Aa), black-pigmentedgram-negativebacteria (BPG-), Fusobacterium (FUSO) and Capnocytophaga (CAPNO) speciesareshown. The numberof total bacterial count is shownseparately.***Highly significantdifferencein the third molar sampleswhencomparedwith the secondmolars(p < 0.001). TOTAL
BACTERIAL
COUNT
AT
X lo6
BASELINE
AFTER
5 MONTHS
Fig. 3. Effect of extraction of lower third molarson total bacterial count in secondlower molar sampling site. Differencewasstatistically significant (p = 0.044). nificantly from the baseline examination to the 5-month follow-up (p = 0.044), whereas a decrease was observed in the test subjects (Fig. 3). Table II shows that the species of BPG- in the controls increased significantly (p = 0.016) during the follow-up period, whereas Fusobacterium sp. increased in the test subjects (JJ = 0.033). No difference was observed in Aa cultivations between the groups. DISCUSSION Our longitudinal findings support the results of Mombelli et a1.,gwho found periodontopathogens in gingivally healthy persons in the area of partly erupted lower third molars. Our hypothesis was that the extraction of partly
erupted third molars may cause a decreasein pathogenic microbes in a wider area than the third molar site. The total bacterial count decreasedin the gingival pockets of the second lower molars (Fig. 3), whereas the number of Fusobacterium speciesand BPG- bacteria increased (Table II). Aa was not very common in either the third or second molar sites. Our 5-month follow-up period might have been too brief for the thorough healing of the extraction wounds. Retention sites are often found more than 5 months after lower third molar surgery. Thus the increasednumber of somebacteria in the secondmolar sitesmight have been causedby microbes harbored in the extraction wound. i9 The number of bacteria could have been decreased in a longer follow-up period,
Periodontopathogens
in 3rd molar region
Volume Number
74 4
Table
II. Effect of extraction of lower third molars on bacterial flora in second lower molar sampling site Test group Follow-up Bacteria
AA BPGFUSO CAPNO
/
Baseline
1
Control
period 2 mo
2* 0 5 13
*Number of culture positive patients. lStatistica1 difference between the test and control NS = Not significant. Aa = Actinobacillus actinomycetemcomitans. BPG= Black pigmented gram-negative bacteria. FUSO = Fusobacterium species. CAPNO = Capnocytophaga species.
2 4 14 11
group
Follow-up 1
5 mo 2 6 15 14
Baseline
435
period 2 1120
0 1 8 12
0 6 10 11
5mo 0 6 10 I
Signi$cance
f
NS p = 0.016 p = 0.033 p = 0.064
subjects.
which practical reasons prevented in the present study protocol. Another explanation for the increased microbial counts in the distal gingival pockets of the second molars might be that the postoperative curettage sampled more subgingival microbes because it was easier to reach the sampling sites after the third molars had been extracted. However, increased counts were also observed in the controls and thus the decreased total bacterial count in the test subjects would not support any difference because of the sampling technique. The paper-point method has been shown to give an acceptable degree of reproducibility for the recovery of Aa and P. gingivalis, but not for P. intermedia.1° The paper-point technique was compatible with the curette-scraping method when the total viable counts were compared with these two alternative methods.12 Systemic antibiotics affect the counts of oral microbes, and their use:must naturally be taken into account in the interpretation of microbial findings.20-22 The study by Asikainen et al. observed the efficacy of doxycycline in juvenile periodontitis.21 There was a decrease in the amount of P. intermedia and F. nucleatum during the administration of the drug, but recovery was observed in 6 weeks. The proportions of other periodontopathogenic bacteria were quite stable in 5 to 8 mm deep gingival pockets in spite of the antibiotics. However, we cannot totally rule out the effect of antibiotics on our results. Ethical reasons prevented us from prescribing the same antibiotics for the control subjects as for the test subjects. During the Vietnam war, pericoronitis was the second most frequent dental emergency of the U.S. Navy and Marine personnel, after dental caries.23 The highest incidence of complications with the eruption of third molars and pericoronitis around these teeth
occurs around 20 years of age.24 Nitzan et a1.25 showed that 8 1% of the pericoronitic patients were 20 to 29 years old and that the pathologic condition occurred in the lower third molar region in 95% of the cases. Extraction of third molars has long been recommended if there is no possibility for the crown to erupt. For example, third molar extractions were indicated in 82% of university students aged 20 to 21 years, and the average was 2.5 third molars per student.24 The extraction should be performed before the patient’s twenty-fifth birthday. In older patients the healing is more complicated because of the intrabony defects on the distal aspect of the lower second molars when examined 2 to 4 years after surgery.26 In conclusion, the present study showed that the presumptive periodontopathogens should be taken into account when the indications for removal of third molars are being considered. The follicle space or gingival pocket in the distal area that surrounds the lower third molar may indeed be a reservoir for a variety of microorganisms. We thank Mrs. Kirsti Kari, MSc, Laboratory of Oral Microbiology of the University of Helsinki, for her guidance and financial support in the microbial cultivations. We also thank Mr. Jussi Metteri, MSc, General Headquarters of Finnish Defence Forces, for statistics. REFERENCES 1. Lysell L, Rohlin the mandibular 1988;17:161-4. 2. Schroeder DC, of third molars 3. Nordenram A, tions for surgical of 2,630 cases.
M. A study of indications used for removal of third molar. Int .I Oral Maxillofac Surg Cecil JC, Cohen ME. Retention and extraction in naval personnel. Milit Med 1983;148:50-3. Hultin M, Kjellman 0, Ramstrom G. Indicaremoval of the mandibular third molar: study Swed Dent J 1987;11:23-9.
436
Rajasuo
et al.
~Ck4iSb~G
ORAL
MEDORM
PATHOL
October 4. Ylipaavalniemi P, Turtola L, Rytijmaa I, Helminen S, Jauhiainen L. Effect of position of wisdom teeth on the visible plaque index and gingival bleeding index. Proc Finn Dent Sot 1982;78:47-9. 5. Rajasuo A, Meurman JH, Metteri J, Ankkuriniemi 0. Effect of extraction of partly erupted third molars on salivary microbial counts in conscripts. Caries Res 1990;24:273-8. 6. Slots J, Listgarten MA. Bacteroides gingivalis, bacteroides intermedius and actinobacillus actinomycetemcomitans in human periodontal diseases. J Clin Periodontol1988;15:85-93. 7. Listgarten MA, Slots J, Rosenberg J, Nitkin L, Sullivan P, Oler J. Clinical and microbiological characteristics of treated periodontitis patients on maintenance care. J Periodontol 1989;60:452-9. 8. Choi JI, Nakagawa T, Yamada S, Takazoe I, Okuda K. Clinical, microbiological, and immunological studies on recurrent periodontal disease. J Clin Periodontol 1990;17:426-34. 9. Mombelli A, Buser D, Lang NP, Berthold H. Suspected periodontopathogens in erupting third molar sites of periodontally healthv individuals. J Clin Periodontol 1990:17:48-54. M, Egelberg’J. Reproducibil10. Dahlei G, Renvert S, Wikstriim ity of microbiological samples from periodontal pockets. J Clin Periodonto 1990;17:73-7. 11. Mijller AJR. Microbiological examination of root canals and periapical tissues of human teeth. Odontol Tidskr 1966;74:1380. 12. Dahlen G, Manji F, Baelum V, Fejerskov 0. Black-pigmented Bacteroides species and Actinobacillus actinomycetemeomitans in subgingival plaque of adult Kenyans. J Clin Periodonto1 1989;16:305-10. P, Neut C, Charon J, Romond C. In 13. Deveaux E, Delplanque vitro survival of Bacteroides intermedius in five transport media, alone or in the presence of Streptococcus sanguis. Oral Microbial Immunol 1990;5:320-3. 14. Slots J. Selective medium for isolation of Actinobacillus actinomycetemcomitans. J Clin Microbial 1982; 15:606-9. 15. Slots J. Detection of colonies of Bacteroides gingivalis by a rapid fluorescence assay for trypsin-like activity. Oral Microbiol Immunol 1987;2:139-41.
16. Slots
17.
18. 19. 20.
21.
22.
23.
24.
25.
26.
1992
J. Rapid identification of important periodontal microorganisms by cultivation. Oral Microbial Immunol 1986;1:4855. Freund RJ, Littell RC, Spector PC. SAS system for linear models. Cary, NC: SAS Institute Inc., 1986. Tetsch P, Wagner W. Operative extraction of wisdom teeth. London: Wolfe Medical Publications Ltd, 1985:1&-21. MacGregor AJ, Hart P. Bacteria of the extraction wound. J Oral Surg 1970;28:885-7. Baker PJ, Evans RT, Slots J, Genco RJ. Antibiotic susceptibility of anaerobic bacteria from the human oral cavity. J Dent Wes 1985;64:1233-44. Asikainen S, Jousimies-Somer H, Kanervo A, Saxen L. The immediate efficacy of adjunctive doxycycline in treatment of localized juvenile periodontitis. Arch Oral Biol 1990;35(suppl):23 l-4. Kinder SA, Holt SC, Korman KS. Peniciliin resistance in the subgingival microbiota associated with adult periodontitis. J Clin Microbial 1986;23:1127-33. Ludwick WE, Pogas JA, Gendron EG, Weldon AL. Dental emergencies occurring among Navy-Marine personnel serving in Vietnam. Milit Med 1974;139:121-3. Ylipaavalniemi P, Turtola L, Murtomaa H, Rytiimaa I. Evaluation of the need for third molar removals among 20- to 21year-old Finnish university students. Proc Finn Dent Sot 1985;81:222-5. Nitzan DW, Tal 0, Sela MN, Shteyer A. Pericoronitis: a reappraisal of its clinical and microbiologic aspects. J Oral Maxillofac Surg 1985;43:510-6. Kugelberg CF. Periodontal healing two and four years after impacted lower third molar surgery: a comparative retrospeciive study. Int J Oral Maxillofac Surg 1990;19:341-5.
Reprint requests: Ari Rajasuo, DDS Valkeala Military Hospital PO Box 5, SF-46141 Vekaranjgrvi, Finland
