Porphyromonas gingivalis

Copyright C Blackwell Munksgaard 2002

J Clin Periodontol 2002: 29: 929–934 Printed in Denmark . All rights reserved

0303-6979

Presence and antibiotic resistance of Porphyromonas gingivalis, Prevotella intermedia, and Prevotella nigrescens in children

Yasaman Sanai1, G. Rutger Persson1, Jacqueline R. Starr2, Henrique S. Luis3, Mario Bernardo3, Jorge Leitao3 and Marilyn C. Roberts4 Department of 1Periodontics, 2Epidemiology and 4Pathobiology, University of Washington, Seattle, Washington, USA; 3Faculty of Medical Dentistry, University of Lisbon, Lisbon, Portugal

Sanai Y, Persson GR, Starr JR, Luis HS, Bernardo M, Leitao J, Roberts MC. Presence and antibiotic resistance of Porphyromonas gingivalis, Prevotella intermedia, and Prevotella nigrescens in children. J Clin Periodontol 2002;29: 929– 934. Abstract Background/aims: Only limited information exists about the prevalence in children of pathogens associated with periodontitis. The aim of the present study was to determine by culture whether 8–11-year-old children carry Porphyromonas gingivalis, Prevotella intermedia, and/or P. nigrescens in samples from the gingiva and/or the buccal mucosa taken before, and after caries treatment and oral hygiene instruction. A second aim was to assess the proportion of subjects who had gram-negative anaerobes carrying the tet(Q) and erm(F) genes, suggesting antibiotic resistance to tetracycline or erythromycin. Method: A total of 150 children provided gingival and buccal swab bacterial samples that were cultured for P. gingivalis, P. intermedia, and P. nigrescens. The species was verified using DNA–DNA hybridization with species-specific probes made from the variable region of the 16S rRNA sequences. Antibiotic-resistant genes, tet(Q) and erm(F), were identified using specific DNA–DNA hybridization with specific DNA probes. Results: A total of 116 isolates of black-pigmented bacteria were cultured from 47 (31%) of 150 children. Five isolates were identified as P. gingivalis, 29 as P. intermedia, 33 as P. nigrescens, and 49 as other species. In general, the bacteria were not culturable at more than one time period. We found that 55% of these 47 children harbored black pigmented bacteria that carried either one or both of the two antibiotic-resistant genes studied (tet(Q), and erm(F)). Conclusion: The present study demonstrated that children not exposed to regular dental treatment carry bacteria outside the gingival sulcus that have been associated with periodontitis, and that standard treatment procedures may not clear the presence of the putative pathogens. In addition, antibiotic-resistant genes are common in identifiable gram-negative anaerobes, including putative pathogens.

Cross-sectional studies of oral health and the prevalence of gingivitis in young children have suggested that the prevalence of gingivitis in children is high (Akpata & Jackson 1979, Addy

et al. 1987, al-Banyan et al. 2000). With similar levels of plaque accumulation, the tendency to develop gingivitis is lower in preschool children than in young adults (Matsson & Goldberg

Key words: antibiotic resistance; children; erm(F); P. gingivalis; P. intermedia; P. nigrescens; tet(Q) Accepted for publication 1 November 2001

1985). It appears that gingivitis is a poor predictor of periodontitis in subjects younger than 30 years. Attempts to prevent periodontitis in young adults through antigingivitis measures such as

930

Sanai et al.

plaque control may therefore be unsuccessful (Prayitno et al. 1993). Periodontitis differs from many other types of infections in that it does not appear to be caused by a single bacterium but by a group of bacteria. Over 300 different types of bacteria have been detected in the mouth (Haffajee & Socransky 1994). Most of these bacteria are thought to be an indigenous part of the normal flora and not associated with oral disease. However, Actinobacillus actinomycetemcomitans, Bacteroides forsythus, Prevotella intermedia, Porphyromonas gingivalis, and Treponema denticola have been implicated as pathogens associated with the development and progression of periodontitis (Socransky & Haffajee 1992, XimenezFyvie et al. 2000). Prevotella nigrescens has recently been speciated from P. intermedia, but its role in periodontitis has not been clearly defined (Shah & Gharbia 1992). Several microbiological tests and clinical parameters have been used to predict and identify active sites of periodontal breakdown (Machtei et al. 1997, 1999, Grossi et al. 1994). However, none of these tests have proven accurate in predicting future sites of attachment loss. Similarly, there are no good measures to determine which persons may be at a higher risk for developing periodontitis as they age. Therefore, better indicators of disease activity in a patient of young age would be of great benefit. One possible way to determine who is at increased risk for disease would be to follow a person, microbiologically, from childhood into adulthood to see if there are any changes in the composition or the enumeration of the oral bacteria and if these changes can account for the breakdown of the periodontium. However, investigations regarding the oral carriage of periodontal pathogens in children have been limited (Kononen et al. 1992, 1994, Frisken et al. 1990). Studies of bacterial transmission from mothers to their children have shown that P. intermedia was not recovered from children but nearly exclusively from mothers with periodontitis, whereas P. nigrescens was found in both mothers and children (Kononen et al. 2000). Primary acquisition of the oral microflora in infants is most often from the mother or primary care giver (Socransky & Manganiello 1971, Kononen et al. 1992). It is not clear when the pathogens associated with periodontitis first colonize the oral

environment and whether these pathogens can be identified from other oral locations than the gingival area. Studies have indicated that the normal flora may act as a reservoir for antibiotic resistance genes (Cohen 1992). Studies of erm(F) and tet(Q) genes in bacterial samples of older subjects with either gingivitis or periodontitis have demonstrated that 81% of samples studied carried antibiotic-resistant genes to erythromycin or tetracycline (Chung et al. 2002). The regional widespread use of antibiotics is reflected in the level of resistance of the subgingival microflora of adult patients with periodontitis (van Winkelhoff et al. 2000). Whether children who may not have been exposed to tetracycline or erythromycin carry drug-resistant pathogens in the oral cavity is poorly studied. The purpose of this study was to determine, by culture, whether 8–11-yearold children carried the gram-negative anaerobes P. gingivalis, P. intermedia, and P. nigrescens in samples from the gingiva and/or the buccal mucosa. In addition, a second aim was to assess the proportion of subjects who had gramnegative anaerobes carrying the tet(Q) and erm(F) genes, suggesting antibiotic resistance to tetracycline or erythromycin.

Material and methods Patient population

A subsample of 150 children was randomly identified among the 500 children who participated in a longitudinal National Institute of Health (NIH) supported study, in Lisbon, Portugal. The Institutional Review Boards (IRB) at the University of Lisbon, Portugal and the University of Washington, Seattle had approved the study. Consent by guardian family member as well as minority subject assent was obtained. Of the 150 children, 77 were male and 73 female. Four children were 8 years old, 38 were 9 years old, 72 were 10 years old, and 33 were 11 years old. The ethnic and racial minorities included those of African descent (17%), and Indonesian (2%), the majority being Caucasian (81%). At the baseline examination all subjects had significant clinical findings of caries and gingivitis. Study exclusion criteria were systemic diseases, periodontitis, previous dental fillings, or no cavities. Forty percent of the children had received any previous

dental treatment, and nearly all reported no history of toothbrushing. Bacterial samples

Bacterial plaque samples were obtained at three different occasions: (i) at baseline; (ii) at 3–7 months after dental treatment, at which time oral hygiene instructions for daily oral care were given and prophylaxis and caries control carried out as needed; (iii) 1 year after treatment. CultureSwabTM Transport System swabs (Difco Laboratories, Detroit, MI, USA) were used to obtain two sets of intraoral bacterial samples from each participant in Lisbon. One sample was taken from the buccal gingiva (mandibular right quadrant), and the other was obtained from the adjacent buccal mucosa. The swabs were placed in transport media and shipped at room temperature to the microbiology laboratories at the University of Washington in Seattle, Washington. The usual shipping time was 50 h, though some samples arrived 96 h after they were taken. However, in the late shipments, anaerobic bacteria were almost never recovered and black-pigmented species were not recovered. Once in Seattle, the swabs were placed into 1.0 mL PRAS buffer (0.038  NaCl, 1.073 m KCl, 2.05 m sodium thiosulfate, 1 m resazurin, 23 m cysteine) and vortexed for 1 min. Serial dilutions in PRAS were made and plated on Brucella agar (Difco) supplemented with 5% sheep blood, 5 mg/L heme and 0.05 mg/L vitamin K under anaerobic conditions (5% CO2, 10% H2 and 85% N2) at 36.5 æC for 5–7 days. The black-pigmented colonies were picked and regrown until pure. The detection level for culturing was defined at ⱖ103 colony forming units (CFU). The isolates were stored at ª70 æC in sterile skim milk until needed. Ten colonies were chosen from each sample when possible, otherwise as many as were available was taken. Each isolate was treated separately and most samples had between 5 and 10 colonies examined. DNA–DNA hybridization for identification

Each isolate was grown on supplemented blood agar plates under anaerobic conditions and incubated at 36.5 æC for 5–7 days. Whole-cell bacterial dots were prepared on GeneScreen membranes (New England Nuclear,

P. gingivalis, P. intermedia, and P. nigrescens in children Table 1. Oligonucleotide probe sequences Probe Name

a

PG 005 PI 011 PN 002

DNA Sequence (5ƒ to 3ƒ)

Wash Temp (æC)

CCG ATG CTT ATT CTT ACG GTA CAT GGA GTC AAC ATC TCT GTA TCC TGC GTC T CGT GCG CCA ATT TAT TCC CAC ATA

67 79 69

a

PG 005 P. gingivalis; PI 011 P. intermedia; PN 002 P. nigrescens (Dix et al. 1990).

Table 2. Comparison of the prevalence of the black-pigmented organisms in buccal and gingival sites* Initial Organism P. gingivalis P. intermedia P. nigrescens Other

G 0 8 6 10

B 3 11 16 10

After treatment

1-year follow-up

G 0 4 5 9

G 1 2 0 6

B 0 4 4 8

B 1 0 2 6

Numbers represent the number of children from whom black-pigmented bacteria were isolated. * No statistically significant differences were found between buccal (B) and gingival (G) samples as determined by the Wilcoxon signed ranks test (p ⬎ 0.05).

Boston, MA, USA) as previously described (Roe et al. 1995). The filters were lysed and stored at room temperature until needed as previously described (Chung et al. 1999). Labeling of oligonucleotides and hybridization

Oligonucleotide probes (Gibco BRL, Life Technologies, Grand Island, NY, USA) for identification of bacterial species, P. gingivalis, P. intermedia, and P. nigrescens (Chung et al. 2002), were derived from 16S rRNA sequences of the bacteria (Moncla et al. 1990)(Table 1). Each DNA probe was end labeled with 32 P-ATP using 10 U of T4 polynucleotide kinase (Promega, Madison, WI, USA) as previously described (Roe et al. 1995). DNA–DNA hybridization was done as previously described (Roe et al. 1995). Clinical P. gingivalis and P. intermedia identified with biochemical tests were used to verify the specificity of these two probes (authors’ unpublished observations). The X-ray films were developed and read visually for presence or absence of hybridization. Positive and negative controls were run with each assay. The DNA probes detected between 102 and 103 CFU. If black-pigmented isolates did not hybridize with any of the three speciesspecific 16S rRNA sequences, they were labeled as ‘other species’.

Data analysis

At each time point, patients provided two pooled samples, one from the buccal area (B) and one from the gingival (G) (Table 2). The B and G samples were pooled together for each patient. Thus samples from the gum and cheek from the same patient were placed together for analysis. The genetic relatedness of P. gingivalis, P. intermedia, and P. nigrescens isolates from the same patient sample has previously been shown using pulse-field gel electrophoresis (PFGE), even though these isolates differed with their carriage of erm(F) and tet(Q) genes (Chung et al. 2002). Therefore, to be conservative, we assumed that all the isolates of the same species, from a single patient, were genetically related. However, to look at the variability in antibiotic-resistant gene carriage, we did the following. If two isolates were the same species and had the same antibiotic-resistant genes from the same patient, then they were con-

931

sidered one genotype; if two isolates from one patient were the same species, but had different antibiotic-resistant genes, each genotype was counted as a separate genotype. Descriptive statistics were used to present prevalence and frequency distribution of putative pathogens in the samples. The Wilcoxon signed rank test was used to define statistical differences comparing buccal and gingival samples. The Kruskal–Wallis  for non-parametric data was used to compare bacterial findings between the three sample periods. Statistical analysis was carried out using the SPSS statistical package (version 10.0 for Windows, Chicago, IL, USA).

Results Prevalence and frequency distribution of black-pigmented organisms

The levels of black-pigmented bacteria ranged from one isolate on 10ª6 dilution to 80% of the colonies on the 10ª5 dilution plate. A total of 432 samples were examined (150 samples at the initial examination, 146 samples after treatment, and 138 samples at the 1-year follow-up). Positive cultures with at least 104 black-pigmented bacteria in their samples were found in 47 (31.3%) of the 150 children. From the 47 children (31% of study population) who had black-pigmented bacteria, a total of 232 bacterial isolates were further studied. At no time could statistically significant differences in the prevalence of black-pigmented bacteria in buccal vs. gingival be identified (Table 2). Five (4.3%) out of the 116 black-pigmented organisms were identified as P. gingivalis (Table 3); these were detected in four (8.5%) out of the 47 children. Twenty-nine (25%) P. intermedia were identified in 15 (31.9%) individuals. Thirty-three (28.4%) P. nigrescens were identified in 20 (42.6%) of the children. The majority of the children (32/47, 68.1%) harbored 49 (42.2%) other black-pigmented organisms (Prevotella

Table 3. Distribution of the 116 black-pigmented bacteria Time

P. gingivalis

P. intermedia

P. nigrescens

Other species

Initial After treatment 1-year follow-up Total no. isolates

3 (2.6%) 0 2 (1.7%) 5 (4.3%)

19 8 2 29

22 9 2 33

20 17 12 49

(16.4%) (6.9%) (1.7%) (25%)

(19.0%) (7.6%) (1.7%) (28.4%)

(17.2%) (14.7%) (10.3%) (42.2%)

932

Sanai et al.

Table 4. Frequency of detection of the target black-pigmented bacteria at different time intervals No. of target bacteria

Initial

After treatment

1-year follow-up

0 1 2 3

30 13 2 2

32 (68.1%) 15 (31.9%) 0 0

42 (89.4%) 5 (10.6%) 0 0

(63.8%) (27.7%) (4.3%) (4.3%)

Table 5. Distribution of the tet(Q) and erm(F) genes for black-pigmented bacteria erm(F) π

erm(F) –

Organism

tet(Q) π

tet(Q) –

P. gingivalis P. intermedia P. nigrescens Other P. gingivalis P. intermedia P. nigrescens Other

3/5 4/29 6/33 8/49 0 6/29 5/33 6/49

0 6/29 5/33 7/49 2/5 13/29 17/33 28/49

and Porphyromonas species), labeled as ‘other’. At baseline, the percentage of oral black-pigmented organisms present ranged from 2.6% (P. gingivalis) to 19.0% (P. nigrescens) (Table 3). After treatment, the prevalence ranged from 0% (P. gingivalis) to 14.7% (other). At the 1-year follow-up, the presence of these putative pathogens ranged from 1.7% (P. gingivalis, P. intermedia, P. nigrescens) to 10.3% (other species) (Table 3). Statistical analysis failed to demonstrate an effect of treatment on the prevalence of bacteria studied at the three time points. However, the trend was downward in the number of children carrying P. intermedia, P. nigrescens and other species (Table 3). The samples at the post-treatment examination and at the 1-year follow-up also showed a reduction in the number of target bacteria isolated. By the 1-year follow-up, 89.4% of the children had none of the three target bacteria, with the remaining 10.6% carrying only one of the target bacteria (Table 4). None of the children had P. gingivalis isolated from samples from more than one time point. Three (6.4%) of the children had P. intermedia cultured from the baseline and after treatment samples, but not at the 1-year time point. One child (2.1%) had P. nigrescens cultured at the initial examination and at 1-year follow-up. Other blackpigmented species were found at more than one time point in four (8.5%) of the children.

(60%) (13.7%) (18.2%) (16.3%) (21%) (15.2%) (12.2%)

(21%) (15.2%) (14.3%) (40%) (44.8%) (51.5%) (57.1%)

Distribution of erm(F) and tet(Q) genes

In previous work with multiple P. gingivalis, P. intermedia and P. nigrescens isolates from a single patient, we found that all isolates from any one patient were genetically related to each other, using PFGE analysis (Chung et al., 2002). Therefore, to be conservative, we also assumed that this was the case with these isolates. We also found that twothirds of the patients carried erm(F) and/or tet(Q) but that the distribution of these two genes varied within the same species isolated from the same patient (Chung et al. 2002). Taking this into consideration, for analysis, we counted two isolates of the same species as 1 when they carried the same antibiotic-resistant genes but as 2 if they carried different antibiotic-resistant genes or neither gene. Thus two P. nigrescens from one child who did not carry either erm(F) or tet(Q) was counted as the same genotype, while two P. nigrescens from one child where one isolate carried the erm(F) gene and the other isolated carried the tet(Q) gene were counted as two different genotypes for this study. From the 232 isolates, we identified 116 different genotypes from the 47 children. This method could have underrepresented the variability of the target bacteria found in each patient, but not the distribution of the isolates over the three time periods (Tables 2 and 3). Isolates that did not hybridize with one of the three species-specific probes were

grouped together as ‘other’ black-pigmented species. Sixteen (34%) out of the 47 children had bacteria that carried the tet(Q) gene, and 24 (51%) had black-pigmented isolates that carried the erm(F) gene. Twenty-one children (45%) carried bacteria that were negative for both the tet(Q) and erm(F) genes. Among the five P. gingivalis, three (60%) carried both tet(Q) and erm(F) genes (Table 5). P. intermedia had a variable pattern with 13.7% carrying antibiotic resistance genes, 21% of the isolates carried the erm(F) gene, and 21% the tet(Q) gene only. Six (18.2%) of the isolates out of the 33 P. nigrescens had both genes, 15.2% of the isolates carried erm(F) only, and 15.2% carried tet(Q) only. A similar distribution exists for the other black-pigmented species, with 16.3% of the isolates carrying both genes, 14.3% carrying the erm(F) gene alone, and 12.2% of the isolates carrying the tet(Q) gene alone. No pattern or combination was found in the carriage of specific black-pigmented bacteria and presence of the tet(Q) or erm(F) genes in these samples. Discussion

In the present study, P. gingivalis, P. intermedia, P. nigrescens, and other black-pigmenting bacteria that could not be identified by standard methods were cultured from the gingiva and buccal mucosa of children 8–11 years of age from Lisbon, Portugal. Although none of the children was diagnosed with periodontitis, gram-negative anaerobes associated with periodontitis were cultured from approximately one-third of the children. These findings are consistent with results reported by other groups specifically for P. gingivalis and P. intermedia (Kamma et al. 2000, Okada et al. 2000). In the present study, transportation of the samples may have resulted in loss of viable P. gingivalis, P. intermedia, and P. nigrescens isolates in the samples. However, given the similarities in bacterial presence reported by other research groups and with other transportation media, the likelihood of such losses in the present study should have been limited and not significantly influenced the results. The microbiological findings, in this study, are in contrast to the common belief that obligate anaerobic bacteria such P. gingivalis and P. intermedia require as deep periodontal pockets for

P. gingivalis, P. intermedia, and P. nigrescens in children growth. However, our data agrees with a recent report that P. gingivalis, P. intermedia, and P. nigrescens could be cultured from older subjects with gingivitis or periodontitis (Schlegel-Bregenzer et al. 1998). In the another study by Umeda et al. (1998), using microbiological methods that do not require live bacteria such as DNA–DNA hybridization, pathogens associated with periodontitis were found more often in the whole saliva rather than from the pooled periodontal pocket samples. Previously, P. gingivalis has been found, almost exclusively, in active sites of deep pockets of periodontally diseased patients, while P. intermedia has been cultured in both healthy and diseased sites, with P. nigrescens predominating in healthy sites in adults (Maeda et al. 1998). This is similar to the distribution of these bacteria in the children from the present study. Data published by Kononen et al. (1992) have suggested that various anaerobic bacterial species readily colonize the edentulous mouth in infants and that their mothers may have exposed these infants to these pathogens. A majority of the oral bacteria usually considered, as putative pathogens in periodontitis may in fact be indigenous. Although the children in our study had their teeth cleaned and were given fluoride treatment and instructions in oral hygiene, the treatment only provided a trend toward a reduction in the detection, by culture, of all black-pigmented bacteria in the children. These changes were not confirmed by statistical methods. The explanations for this may either be that the preventive measures failed to eliminate the bacteria studied because of inadequate procedures, or that the subjects were re-infected prior to the next sampling. It should be recognized that the study population included young children with no or very limited previous exposure to dental care. Longitudinal studies of bacterial samples from these children using both culture and/or more sensitive techniques, such as direct detection using DNA probes, should be explored. This might answer the question of whether the presence of selected bacteria, preferably identified also by genotype and serotype, at a young age may increase the risk of developing periodontitis later in life. A most noticeable finding inthe present study was that 55%, of the 47 children harboring black pigmented

bacteria, carried strains with one or two antibiotic resistant genes. The findings are similar to those found in adults by Chung et al. (2002) and also consistent with studies antibiotic resistance testing (van Winkelhoff et al. 2000). Whether the bacteria carrying the tet(Q) and erm(F) genes are able to transfer the antibiotic-resistant genes to other bacteria in vivo needs further studies. In the present study, no antibiotics were prescribed. It seems appropriate to hypothesize that these subjects will most likely not respond to future therapy that includes tetracycline or erythromycin to clear infections of P. gingivalis or P. intermedia. and if such resistant bacteria can serve as a pool for antibiotic resistance and transfer the resistant genes to other bacteria, the consequences can be significant. In conclusion, the present study demonstrated that children who have not been exposed to dental care including preventive procedures (i) carry bacteria that have been associated with periodontitis; (ii) such bacteria can be located outside the gingival sulcus; (3) standard procedures including caries treatment, oral hygiene instructions and cleanings failed to clear putative pathogens, though there was a trend downward, and (ii) antibiotic-resistant genes are common in identifiable gram-negative facultative anaerobes. Acknowledgments

This study was supported by NIDCR grant DE-11894 and contract DE72623 from the National Institutes of Health. J. R. Starr was supported by NIH Training Grant T32 DE07227. Zusammenfassung Vorhersage marginaler Knochen- und Zahnverluste – eine prospektive Studie über 20 Jahre Zielsetzung: Das Ziel der vorliegenden Studie war die Untersuchung potentieller Risikovorhersagefaktoren / Risikofaktoren auf longitudinale marginale Knochen- und Zahnverluste über 20 Jahre. Material und Methodik: In den Jahren 1970 und 1990 wurde eine klinische und röntgenologische Untersuchung einer 513 Probanden umfassenden Gruppe durchgeführt. Zur Berechnung des Einflusses potentieller Risikovorhersagefaktoren/Risikofaktoren auf longitudinale marginale Knochen- und Zahnverluste wurde eine stufenweise mehrfache Regressionsanalyse durchgeführt. Ergebnisse: Im Jahre 1970 standen der parodontale Index nach Russell sowie Rauchge-

933

wohnheiten in einer statistisch signifikanten und positiven Korrelation mit longitudinalen Knochenverlusten. Der parodontale Index nach Russell, der marginale Knochenverlust und der Plaque-Index zu Baseline standen in einer positiven und statistisch signifikanten Korrelation mit dem longitudinalen Knochenverlust, wogegen Alter, Anzahl der fehlenden Zähne im Jahre 1970 sowie Anzahl der Schuljahre eine signifikante negative Korrelation zur Anzahl der zwischen 1970 und 1990 verlorenen Zähne aufwiesen. Schlussfolgerung: Als signifikanter Risikofaktor für marginale Knochenverluste konnte das Rauchen identifiziert werden; Plaque wurde als Risikofaktor für Zahnverluste identifiziert. Die Regressionsmodelle jedoch erklärten einen geringen Prozentsatz der Gesamtvarianz, insbesondere für marginale Knochenverluste.

Re´sume´ Pre´vision de perte osseuse marginale et de perte dentaire: e´tude prospective sur 20 ans But: Cette e´tude avait pour objectif d’analyser l’influence des pre´dicteurs de risque/ facteurs de risque potentiels sur la perte osseuse marginale longitudinale et la perte dentaire sur une dure´e de 20 ans. Mate´riaux et me´thodes: L’examen clinique et radiologique d’un e´chantillon de 513 individus a e´te´ re´alise´ en 1970 et 1990. Des analyses de re´gression multiple pas a` pas ont e´te´ utilise´es pour calculer l’influence des pre´dicteurs de risque/ facteurs de risque potentiels sur la perte osseuse marginale longitudinale et la perte dentaire. Re´sultats: L’indice parodontal de Russell en 1970 et le tabagisme e´taient significativement et positivement associe´s a` la perte osseuse longitudinale. L’indice parodontal de Russell, la perte osseuse marginale et l’indice de plaque initiaux e´taient positivement et significativement associe´s a` la perte dentaire longitudinale, tandis que l’aˆge, le nombre de dents manquantes en 1970 et le nombre d’anne´es scolaires indiquaient une corre´lation ne´gative significative avec le nombre de dents perdues entre 1970 et 1990. Conclusions: Le tabagisme s’est ave´re´ eˆtre un facteur de risque significatif de la perte osseuse marginale, alors que la plaque est un facteur de risque de la perte dentaire. Les mode`les de re´gression ont toutefois explique´ le faible pourcentage de variance totale, notamment pour la perte osseuse marginale.

References Abraham, J., Stiles, H. M., Kammerman, L. A. & Forrester, D. (1990) Assessing periodontal pathogens in children with varying levels of oral hygiene. American Society of the Dental Caries. Journal of Dentistry for Children 57, 189–193. Akpata, E. S. & Jackson, D. (1979) The prevalence and distribution of gingivitis and gingival recession in children and

934

Sanai et al.

young adults in Lagos, Nigeria. Journal of Periodontology 50, 79–83. al-Banyan, R. A., Echeverri, E. A., Narendran, S. & Keene, H. J. (2000) Oral health survey of 5–12-year-old children of National Guard employees in Riyadh, Saudi Arabia. International Journal of Paediatric Dentistry 10, 39–45. Chung, W. O., Gabanay, J., Persson, G. R. & Roberts, M. C. (2002) Distribution of erm (F) and tet (Q) genes in four oral bacterial species and genotypic variation between resistant and susceptible isolates. Journal of Clinical Periodontology. Chung, W. O., Young, K., Leng, Z. & Roberts, M. C. (1999) Mobile elements carrying ermF and tetQ genes in gram-positive and gram-negative bacteria. Journal of Antimicrobial Chemotherapy 44, 329–335. Cohen, M. (1992) Epidemiology of drug resistance: implications for a post-antimicrobial era. Science 257, 1050–1055. Dix, K., Watanabe, S. M., McArdle, S., Lee, D. I., Randolph, C., Moncla, B. & Schwartz, D. E. (1990) Species-specific oligodeoxynucleotide probes for the identification of periodontal bacteria. Journal of Clinical Microbiology 28, 319–323. Frisken, K. W., Higgins, T. & Palmer, J. M. (1990) The incidence of periodontopathic microorganisms in young children. Oral Microbiology and Immunology 5, 43–45. Grossi, S. G., Zambon, J. J., Ho, A. W., Koch, G., Dunford, R. G., Machtei, E. E., Norderyd, O. M. & Genco, R. J. (1994) Assessment of risk for periodontal disease. I. Risk indicators for attachment loss. Journal of Periodontology 65, 260–267. Haffajee, A. D. & Socransky, S. S. (1994) Microbial etiological agents of destructive periodontal diseases. Periodontology 2000 (5), 78–111. Kamma, J. J., Diamanti-Kipioti, A., Nakou, M. & Mitsis, F. J. (2000) Profile of subgingival microbiota in children with mixed dentition. Oral Microbiology and Immunology 15, 103–111. Kononen, E., Asikainen, S. & JousimiesSomer, H. (1992) The early colonization of gram-negative anaerobic bacteria in edentulous infants. Oral Microbiology and Immunology 7, 8–31.

Kononen, E., Asikainen, S., Saarela, M., Karjalainen, J. & Jousimies-Somer, H. (1994) The oral gram-negative anaerobic microflora in young children: longitudinal changes from edentulous to dentate mouth. Oral Microbiology and Immunology 9, 136–141. Kononen, E., Wolf, J., Matto, J., Frandsen, E. V., Poulsen, K., Jousimies-Somer, H. & Asikainen, S. (2000) The Prevotella intermedia group organisms in young children and their mothers as related to maternal periodontal status. Journal of Periodontal Reserach 35, 329–334. Machtei, E. E., Dunford, R., Hausmann, E., Grossi, S. G., Powell, J., Cummins, D., Zambon, J. J. & Genco, R. J. (1997) Longitudinal study of prognostic factors in established periodontitis patients. Journal of Clinical Periodontology 24, 102– 109. Machtei, E. E., Hausmann, E., Dunford, R., Grossi, S., Ho, A., Davis, G., Chandler, J., Zambon, J. & Genco, R. J. (1999) Longitudinal study of predictive factors for periodontal disease and tooth loss. Journal of Clinical Periodontology 26, 374–380. Maeda, N., Okamoto, M., Kondo, K., Ishikawa, H., Osada, R., Tsurumoto, A. & Fujita, H. (1998) Incidence of Prevotella intermedia and Prevotella nigrescens in periodontal health and disease. Microbiology and Immunology 42, 583–589. Matsson, L. & Goldberg, P. (1985) Gingival inflammation reaction in children at different ages. Journal of Clinical Periodontology 12, 98–103. Moncla, B. J., Braham, P., Dix, K., Watanabe, S. & Schwartz, D. (1990) Use of synthetic oligonucleotide DNA probes for the identification of Bacteroides gingivalis. Journal of Clinical Microbiology 28, 324– 327. Okada, M., Hayashi, F. & Nagasaka, N. (2000) Detection of Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in dental plaque samples from children 2–12 years of age. Journal of Clinical Periodontology 27, 763–768. Prayitno, S. W., Addy, M. & Wade, W. G. (1993) Does gingivitis lead to periodontitis in young adults. Lancet 342, 471–472. Roe, D. E., Braham, P. H., Weinberg, A. & Roberts, M. C. (1995) Characterization of

tetracycline resistance in Actinobacillus actinomycetemcomitans. Oral Microbiology and Immunology 10, 227–232. Schlegel-Bregenzer, B., Persson, R. E., Lukehart, S., Braham, P., Oswald, T. & Persson, G. R. (1998) Clinical and microbiological findings in elderly subjects with gingivitis or periodontitis. Journal of Clinical Periodontology 25, 897–907. Shah, H. N. & Gharbia, S. E. (1992) Biochemical and chemical studies on strains designated Prevotella intermedia and proposal of a new pigmented species, Prevotella nigrescens sp. International Journal of Systematic Bacteriology 42, 542–546. Socransky, S. S. & Haffajee, A. D. (1992) The bacterial etiology of destructive periodontal disease: current concepts. Journal of Periodontology 63, 322–331. Socransky, S. S. & Manganiello, S. D. (1971) The oral microbiota of man from birth to senility. Journal of Periodontology 42, 485– 496. Umeda, M., Contreras, A., Chen, C., Bakker, I. & Slots, J. (1998) The utility of whole saliva to detect the oral presence of periodontopathic bacteria. Journal of Periodontology 69, 828–833. van Winkelhoff, A. J., Herrera Gonzales, D., Winkel, E. G., Dellemijn-Kippuw, N., Vandenbroucke-Grauls, C. M. & Sanz, M. (2000) Antimicrobial resistance in the subgingival microflora in patients with adult periodontitis. A comparison between The Netherlands and Spain. Journal of Clinical Periodontology 27, 79–86. Ximenez-Fyvie, L. A., Haffajee, A. D. & Socransky, S. S. (2000) Microbial composition of supra- and subgingival plaque in subjects with adult periodontitis. Journal of Clinical Periodontology 27, 722–732. Address: Marilyn C. Roberts Department of Pathobiology Box 357238 School of Public Health and Community Medicine University of Washington Seattle, WA 98195 USA e-mail: marilynr/u.washington.edu