576
Clinical Documentation and Occurrence of Putative Periodontopathic Bacteria in Human Immunodeficiency VirusAssociated Periodontal Disease Mervyn Gornitsky, *f D. Christopher Clark,f Russell Siboo,f Rhonda Amsel,f Irene Iugovaz,f Christine Wooley, * Nathalie Iuliani, * and Eddie CS. Chan*f Human immunodeficiency virus (hiv)-associated gingivitis (HIV-G) and HIVassociated Periodontitis (HIV-P) are two intraoral lesions manifested by patients with HIV infection. Periodontal indices were measured for 87 subjects in 5 study groups: HIV-seropositive patients with healthy periodontium (HIV- ), with HIV-G, or with HIVP; and non-HIV-infected subjects with healthy periodontium (H) or with adult chronic Periodontitis ( ). The quantitative clinical parameters were compared and statistically significant intergroup differences were noted. The mean scores on PI and PD do not discriminate between HIV-seropositive and non-HIV-infected seronegative cohorts, but a significant difference in the GI between HIV- and was noted. When categories of PD and AL are examined, some differences become apparent. Generally, the PD and AL of HIV-P are not as great as those of P. PI correlates well with GI (r 0.86) in P, in of selected but does not (r HIV-P. In the occurrence addition, 0.33) putative periodontopathic bacteria (Porphyromonas gingivalis, spirochetes, and motile eubacteria) in these lesions was determined by brightfield (after staining), darkfield and immunofluorescent microscopy. No difference in microbiological profile in the bacterial groups monitored was found between and HIV-P. Spirochetes were found to be more abundant than P. gingivalis in the lesions of and HIV-P. In marked contrast, P. gingivalis was found to be in highest numbers in samples from the gingival crevice of H as determined by indirect immunofluorescence. J Periodontol 1991; 62:576-585. =
=
Key Words:
HIV
infection; gingivitis/etiology; periodontitis/HIV-etiology; periodontal
index.
Oral lesions due to microbial infections and cancers may be the first manifestation of clinical disease of acquired immune deficiency syndrome (AIDS) caused by the human immunodeficiency virus (HIV), or oral lesions may arise subsequently from the down regulation of the immune system as a consequence of the HIV infection.1,2 The most common and significant oral lesions appear to be Kaposi's sarcoma, candidiasis, hairy leukoplakia, Herpes virus infection, and HIV-associated periodontal disease.3 Periodontal disease in the HIV-seropositive homosexual cohort is the single most important factor in their seeking dental treatment. It is seen in such frequency as to be considered an "opportunistic" disease.4,5 Pain and spontaneous bleeding are the complaints noted; these lead to *Dental
Department,
Sir Mortimer B. Davis-Jewish General
Montreal, PQ.
Faculty of Dentistry,
McGill
University, Montreal, PQ.
Hospital,
difficulty in mastication with a subsequent reduction in the ingestion of solid food for proper nutrition. The lesions of periodontal disease have been divided into two types by Winkler et al.:6 1) HIV-associated gingivitis (HIV-G); and 2) HIV-associated Periodontitis (HIV-P). HIV-G is defined
lesions confined to the soft tissues which showed a distinctive fiery red erythema and edema of the free gingiva, attached gingiva, and alveolar mucosa. Other characteristics include spontaneous gingival bleeding and failure to respond to therapy such as plaque removal and maintenance of good oral hygiene. HIV-P is defined as those sites with the symptoms as described for HIV-G and, in addition, there are extensive soft tissue ulcération and necrosis, severe loss of periodontal attachment with pain, and frank exposure of alveolar bone. Frequently, HIV-P affects only specific areas of the mouth whereas HIV-G is more generalized; besides pain, in HIV-P, mobility of involved teeth and halitosis are evident. as
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI, CHAN
Descriptions of HIV-G and HIV-P in the literature have been almost entirely narrative. Our study provides a quantitative documentation of the periodontal indices of HIVseropositive patients grouped as HIV-G, HIV-P and HIVH (healthy; i.e., no obvious periodontal disease symptoms). Control non-HIV-infected (seronegative) patients with healthy periodontium (H) and with adult chronic Periodontitis ( ) provided data for comparison with the data obtained from patient groups with HIV-G, HIV-P, and HIV-H. In addition, we present data on the relative percentages of Porphyromonas gingivalis, spirochetes, and motile bacteria in periodontally healthy and periodontally diseased sites of subjects, with and without HIV infection, based on cell numbers and not on the number of sites harboring specific bacteria as determined by Murray et al.7 The focus of this part of the study was to determine the occurrence of selected putative periodontopathic bacteria in the lesions of HIV-G and HIV-P sites in comparison with those found in the lesions of adult chronic Periodontitis in healthy patients. Knowledge of the types of microorganisms known to occur in a lesion may facilitate the formulation of improved treatment protocols for oral lesions that may enhance the quality of life in HIV-infected AIDS patients.
577
Figure 1. HIV-infected patient with healthy periodontium (HIV- ). Although the periodontium is asymptomatic for periodontal disease, there is hyperpigmentation (arrows).
MATERIALS AND METHODS Patients and Sampling Sites Patients for the study were received at the Dental Department, Sir Mortimer B. Davis-Jewish General Hospital, which is the major referral center in Montreal for dental and oral treatment of HIV-seropositive patients, 87% of whom were homosexuals. All HIV-infected patients had been tested seropositive for antibody to HIV by enzyme-linked immunosorbent assay (ELISA) at the Royal Victoria Hospital, Montreal, Quebec. All subjects gave informed consent to participate in the study, which was approved by an ethics committee of the Faculty of Dentistry, McGill University. In the selection of patients for the study, the parameters used for Periodontitis included both clinical and radiographic evidence of at least 2 sites with probing pocket depth greater than 5 mm. Radiographic interpretation was used to screen for sites with pockets due to alveolar bone loss. Patients with healthy periodontia had probing pocket depths of less than 3 mm and no evidence of gingivitis. Bleeding on probing to the base of the pocket was considered to be a sign of gingivitis. HIV-G and HIV-P were assessed according to Winkler et al.6 Eighty-seven subjects entered into the clinical documentation study (89 subjects in the microbiology study) were divided into 5 groups: 10 HIV-infected seropositive patients with healthy periodontium (HIV- ) (Fig. 1); 26 HIV-infected seropositive patients with HIV-G (Fig. 2); 24 (25 in microbiology) HIVinfected séropositive patients with HIV-P (Figs. 3 and 4); 9 (10 in microbiology) non-HIV-infected seronegative subjects with healthy periodontium (H), most of whom were
1 Figure 2. HIV-infected patient with gingivitis (HIV-G). There dence of plaque but erythema extends from tooth surface.
is
no
evi-
dental residents while the rest were staff members of the dental clinic; and 18 non-HIV-infected seronegative patients with adult chronic Periodontitis ( ). With each patient, 4 sites with greatest probing depth were selected for microbiological sampling. A site is defined as an interproximal area in either the anterior or posterior region where the proximal surfaces of 2 adjacent teeth are in contact. From each site 2 samples were obtained by inserting sterile paper points. Both paper point samples were placed in a sterile Eppendorf tube and transported to the laboratory where they were processed in a laminar flow biological safety hood. Clinical Evaluation All clinical examinations were carried out by 1 of 2 examiners and the same examiner was used for the same pa-
578
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
J Periodontol 1991
September
(CEJ) measurements were taken at 6 locations per tooth and
every tooth was measured. The 6 locations were at the mesial buccal, distal buccal, mesial lingual, distal lingual line angles, and on the mid-lingual and mid-buccal surfaces, using a set of William's periodontal probes. PD was measured from the gingival margin. The CEJ was employed as a fixed reference point for attachment level (AL), which was the distance between the CEJ and the base of the pocket as described by Clark et al.10 All measurements were to the nearest mm.
Figure 3. HIV-infected patient tion of soft tissue and bone.
with Periodontitis
(HIV-P).
Note destruc-
Figure 4. HIV-infected patient with Periodontitis (HIV-P). tion of soft tissue, bone and mucosa.
Note destruc-
tient
throughout the study.
Both examiners
were
calibrated
(M.G.) to ensure minimal interexaminer variability. Gingival index (GI) and plaque index (PI) were assessed on 6 designated teeth: upper right first molar (16), upper left central (21), upper left first bicuspid (24), lower left first molar (36), lower right central (41), and lower right first bicuspid (44) using mesial buccal, buccal, distal buccal, and lingual surface locations. If 1 of these teeth was missing, then an adjacent tooth or a tooth in the opposite arch was utilized. GI was scored according to the procedure of Löe and Silness;8 PI was assessed by the criteria of Silness and Löe.9 According to these authors, the GI and PI at 6 teeth, representing the 6 segments of the jaws, reliably reflect total mouth scores. Probing pocket depth (PD) and cemento-enamel junction by
one
of the authors
Treatment of Microbiological Samples Each site sample (consisting of two paper points from each site) was suspended in 0.5 ml sterile distilled water and mixed thoroughly with a vortex apparatus. Distilled water instead of buffer was used as suspended fluid for sample plaque material because the presence of salts interfere with the subsequent staining procedure below. Preliminary quantitative studies had shown no lysis of bacterial cells, including spirochetes, in this aqueous suspension. Soluble plaque material probably contributed to the integrity of any fragile cells. Comparable counts of bacterial morphological groups in subgingival plaque samples from 17 periodontally diseased sites support the observation that there was no apparent cell lysis.11 Ten µ aliquots of the sample were placed on glass microscope slides and air-dried; the smears were gently heat-fixed by passing the slides through a flame several times. The slides with unstained smears were stored at room temperature for 1 week (to destroy the infectivity of the virus), and then frozen at -20°C for 1 or 2 weeks until processed for microscopic examination. No evidence of bacterial cell degradation was observed upon such storage of cells. Microscopic examination at 1000 x magnification showed that the bacteria were morphologically intact and both P. gingivalis and oral spirochetes retained their antibody-reactive epitopes on storage. Heat-fixation was sufficient to halt any deterioration of the antibody-reactive
epitopes.
Treatment of Sample Smears The smears were treated with a modified Gray's flagella stain as described previously.11 This method yielded count-
ing results equivalent to that obtained by darkfield microscopy. For instance, flagellated cells did not lose their locomotor organdíes by the staining process. Stained specimens gave counts comparable to that of motile bacteria in the same sample obtained by darkfield microscopy. This staining procedure facilitated enumeration of spirochetes, flagellated (motile-since all motile eubacteria locomote by flagella) eubacteria, coccoid cells and all other bacteria under bright-field microscopy (Fig. 5). An indirect immunofluorescence assay (IFA) was used to detect P. gingivalis. Specific polyclonal antibodies against P. gingivalis ATCC 33277 were produced in New Zealand rabbits. The antiserum was diluted (1:20) in ice-cold (4 °C) 0.01M phosphate buffer, pH 7.2, containing 0.85% NaCl
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI, CHAN
• •
ß
*
579
-
10 UM
mm
Figure 5. Smear from a periodontal pocket sample stained with modified Cray's flagella stain showing spirochetes, flagellated cells, coccoids, and other bacteria under brightfield illumination. Bar 10 µ/ .
Figure 6. Fluorescent cells of P. gingivalis exhibiting typical morphology. Magnification same as Figure 5.
=
and 0.02% Tween 20 (PBST). The diluted antiserum was added to sample smears and allowed to interact with the bacteria for 20 minutes at room temperature and the smears were then washed twice (5 minutes per wash) with ice-cold PBST. Fluorescein-conjugated affinity purified goat antirabbit IgG adsorbed with human serum proteinsf was diluted (1:10) in ice-cold PBST was then added to the washed smears; the interaction was allowed to proceed for 20 minutes at room temperature. The smears were washed twice with cold PBST, as above, and once with cold distilled water for 1 minute. Mounting fluid (9 vol. glycerol: 1 vol. carbonate buffer, pH 9.0) was added to the smears. Cover slips were placed over the smears and sealed to the slides with clear nail varnish. The smears were examined by epifluorescence using a Leitz Dialux microscope at a magnification of 1000. An apple-green fluorescence of the bacterial envelope exhibiting short-rod morphology typical of the bacteroides was scored as positive for P. gingivalis (Fig. 6). The dilutions of the primary and secondary antisera used to enumerate P. gingivalis were determined with smears of gingivalis ATCC 33277 by checkerboard titrations. The P. gingivalis ATCC 33277 antiserum was shown to react with P. gingivalis cells from fresh plaque material; it also reacted with all strains of P. gingivalis in our stock culture. Test experiments showed no cross reaction between anû-P. gingivalis antiserum and Prevotella intermedia, Prevotella melaninogenica, and other species of black-pigmented anerobic rods. Nonspecific binding of the fluorescein-conjugated antiserum to bacteria in smears of subgingival plaque and P. gingivalis ATCC 33277 was determined as described above for the IFA but the primary antiserum was omitted. Nonspecific binding was not observed. Other workers also have used successfully polyclonal anti-P. gingivitis antiserum to enumerate P. gingivalis in smears of subgingival plaque from AIDS patients.7'12
tSigma Chemical Co.,
St.
Louis, MO.
Figure 7. Same field (and magnification) as Figure 6 as seen by darkfield microscopy. Compare with Figure 6 for the relative positions of P. gingivalis cells. Note the presence of spirochetes. For darkfield examination of the same field (Fig. 7) examined previously by epifluorescence, a transmitted quartziodine light source was used with numerical aperture of the oil-immersion objective stepped down by means of an adjustable rotating collar. The count under darkfield illumination represented the total bacterial count for that field. Depending on the cell density, 30 fields or sufficient fields totaling at least 100 cells of each site sample smear were examined alternatively by epifluorescence and darkfield illumination. Smears from 4 sample sites of each subject were examined.
Data Analysis Averages for PI, GI, and PD sites were subjected to oneway analysis of variance (ANOVA). For each subject, the
number of teeth with PD between 1 to 3 mm, 4 to 6 mm, and 7 or more mm were noted. A similar breakdown was
580
J Periodontol 1991
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
September
Table 1: Means of Periodontal Indices (standard error) for
Subject Groups
_Subject Groups*_
_Index_H_ _HIV- _HIV-G_HIV-P Plaque (PI) Gingival (Gl) Pocket depth (PD), *H
mm
0.55(0.09) 0.51(0.11) 2.26(0.07)
1.36(0.12) 1.36(0.10) 3.18(0.17)
0.57(0.10) 1.02(0.03) 2.19(0.07)
0.97(0.07) 1.27(0.03) 2.34(0.04).
1.27(0.10) 1.62(0.08) 2.86(0.10)
non-HIV-infected seronegative patients with healthy periodontium; non-HIV-infected seronegative patients with Periodontitis; HIV-H HIV-infected seropositive patients with healthy periodontium; HIV-infected seropositive patients with gingivitis (HIV-associated gingivitis); HIV-G HIV-P HIV-infected seropositive patients with Periodontitis (HIV-associated Periodontitis). =
=
= =
=
GINGIVAL INDICES
PLAQUE INDICES 2.0
H
Subject groups HIV0 HIV-G
HIV-P
8. Mean
Figure plaque indices of subject groups showing significant inlergroup differences by one way ANOVA. Abbreviations for subject groups defined in "Materials and Methods" in this and following figures. made for AL. These data were examined using one way multivariate analysis of variance and significant results were subjected to univariate analysis. All significant between group differences were examined using Tukey hsd post hoc tests. In addition, the correlation between PI and GI was calculated for each group. The microbiological profile was compared using 1-way independent groups analysis of variance. Microscopic scores of per cent spirochetes, motile cells, coccoids, and other types by staining, per cent P. gingivalis and spirochetes by immunofluorescence and darkfield microscopy, respectively, were averaged over the 4 sites sampled per subject. Post hoc tests were performed using Tukey hsd multiple range statistics. RESULTS The means for PI, GI, and PD for the subject groups are summarized in Table 1. There was a significant group effect for average PI (Fig. 8): the 2 Periodontitis groups (P and HIV-P) scored significantly higher than the 2 periodontally healthy groups (H and HIV-H). Post hoc tests of the significant group effect on the gingival index showed that the H group had significantly lower scores than all other groups and the HIV-H group differed
Tukey hsd HIV-P>H,HIV-H, HIV-G; P>H,HIV-H; HIV-G>H;HIV-H>H(p7 mm Attachment level 1-3 mm 4-6 mm >7 mm
0.78(0.03) 0.22(0.03) 0.00(0.00)
0.42(0.08) 0.43(0.05) 0.16(0.05)
0.81(0.04) 0.18(0.04) 0.01(0.01)
0.71(0.03) 0.28(0.03) 0.01(0.01)
0.56(0.04) 0.08(0.03)
0.91(0.03) 0.09(0.10) 0.00(0.00)
0.24(0.07) 0.53(0.06) 0.23(0.06)
0.82(0.06) 0.18(0.06) 0.00(0.00)
0.71(0.04) 0.26(0.04) 0.03(0.01)
0.40(0.06) 0.45(0.04) 0.15(0.04)
0.37(0.05)
*See Table 1 for for definitions.
MEAN FREQUENCY WITH PD 4-6 MM
POCKET DEPTH 0.6
Tukey hsd HIV-P>H,HIV-H,HIV-G; P>H,HIV-H(p7 MM
o-
.
Subject groups
0 HIV-P m HIV-H M HIV-G Figure 11. Mean frequency of PD 1 to 3 mm among subject groups showing significant intergroup differences by one-way ANOVA. results are shown in Table 3. The high correlation coefficient (0.86) in the subjects is consistent with expectation in periodontal disease in normal (non-HIV-infected) pop-
Subject groups HIV-H
0
HIV-G
HIV-P
Figure 13. Mean frequency of PD > 7 mm among subject groups showing significant intergroup differences by one-way ANOVA. ulation groups. However, there was a lack of correlation between PI and GI in the HIV-P subjects.
582
J Periodontol 1991
September
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
MEAN FREQUENCY WITH AL >7 MM
MEAN FREQUENCY WITH AL 1-3 MM
Subject groups H
0
HIV-H
0
HIV-G
Subject groups HIV-H
HIV-P
Figure 14. Mean frequency ofAL 1 to 3 mm among subject groups showing significant intergroup differences by one-way ANOVA. MEAN FREQUENCY WITH AL 4-6 MM
Tukey hsd P>H,HIV-H,HIV-G; HIV-P>H,HIV-H(p
Table 3. Pearson Correlation Coefficient Between PI and GI For Each Subject Group
Subject Group* HIV-H HIV-G HIV-P
r
.43 .86 .29 .30 .33
9 18 10 26 24
0.13 H, HIV-H); the sampled sites of HIVinfected patients with Periodontitis had significantly greater numbers of stained spirochetes than those of healthy individuals with healthy periodontium (i.e., HIV-P>H). Figure 18 shows the per cent occurrence of bacteria other than spirochetes, motile eubacteria, and coccoid cells in stained smears from the several subject groups. The morphotype "other bacteria" constituted the highest proportion of the 4 morphotypes of bacteria counted in the stained specimens. The results in the figure show that healthy sites of HIV-seropositive patients had significantly more "other bacteria" than diseased sites of Periodontitis subjects without or with HIV infection (HIV-H>P, HIV-P). The relative percentages of P. gingivalis compared to all other bacteria present in the various plaque samples are shown in Figure 19. This species, detected by indirect immunofluorescence, was significantly more numerous in healthy sites of non-HIV-infected subjects than diseased sites of HIV-infected patients with Periodontitis (i.e., H > HIV-P). The data in Figure 20 show the greater proportion of spirochetes in samples of both and HIV-P as seen by darkfield microscopy.
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI,
Table 4: The Mean Per Cent Occurrence of Selected Microbial
Subject Group
Groups (with
CHAN
583
standard error) in Each
_Subject Groups*_
Microbial
Group_H_ _HIV- _HIV-G_HIV-P
Spirochetes (stained)
1.3(0.6) 32.0(4.1) 1.7(0.9) 65.4(3.3) 24.0(5.0) 1.0(0.5)
Coccoids Motile cells Other bacteria P. gingivalis
Spirochetes (darkfield)
16.4(3.2) 23.2(2.5) 2.7(0.7) 58.1(2.6) 13.8(2.1) 14.1(2.5)
4.2(1.8) 23.2(2.2) 1.3(0.3) 71.1(2.5) 18.8(4.9) 2.9(1.7)
7.7(1.6) 23.0(2.0) 2.8(0.6) 66.5(2.5) 18.7(2.0) 5.0(1.5)
11.1(2.0) 27.2(1.9) 2.2(0.5) 59.4(2.2)
11.5(1.2) 9.5(1.8)
*See Table 1 for definitions.
STAINED SPIROCHETES
PORPHYROMONAS GINGIVALIS
Subject groups HIV-P 0 HIV-G Per cent occurrence of stained spirochetes showing significant
HIV-
Figure 17. intergroup differences by one-way ANOVA.
Figure
H 19. Per
!
Subject groups F23 HIV-G HIV-
cent occurrence
HIV-P
of P. gingivalis showing significant
in-
tergroup differences by one-way ANOVA.
SPIROCHETES BY DARKFIELD
OTHER BACTERIA
5
5
Study groups HIV-H 0 HIV-G
H 18. Per
Figure intergroup differences by
of "other bacteria" showing one-way ANOVA.
cent occurrence
Subject groups
W
HIV-P
significant
DISCUSSION The mean scores of PI and PD do not discriminate between seropositive and non-HIV-infected seronegative cohorts. But
Figure
mination
M
HIV-H
HIV-G counted by
HIV-P
of spirochetes darkfield illushowing significant intergroup differences by one-way ANOVA.
20. Per
cent occurrence
there is a significant difference between the GI of HIV-H and H groups (Table 1 and Fig. 9). Thus, even though the periodontia of HIV-H patients may be asymptomatic for
584
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
periodontal disease, there are marginal changes in the gingiva of these patients. When categories of PD and AL are examined (Table 2), some other differences emerge. In intermediate PD 4 to 6 mm, HIV-P is significantly greater
than H, HIV-H, and HIV-G. In PD of 7 mm or more, it is which differs from H, HIV-H, and HIV-G. HIV-P no longer has a significant difference with H, HIV-H, and HIV-G due to the destruction of soft tissue and bone in the HIV-P patient as disease becomes more severe. That is, PD does not seem to be a distinguishing feature of HIV-P because the rapid soft tissue necrosis occurs simultaneously with the loss of bone. This corroborates with the observed clinical picture. In the AL indices frequency distribution, the highest scores occur in the group in the 4 to 6 mm and the >7 mm categories (Table 2); i.e, is significantly greater than H, HIV-H, and HIV-G (Figs. 15 and 16), corroborating the observed clinical picture. In examining the correlations between GI and PI, PI correlates with Gì in patients; this supports the role of plaque in causing periodontal disease in non-HIV-infected subjects. However, in HIV-P patients, no significant correlation exists between PI and GI, which suggests a dissociation between plaque and periodontal disease in these patients. This lack of correlation was also noted by Oshrain et al.13 on renal transplant patients receiving imuran and corticosteroids (prednisone). The immuno-suppression brought about by drugs may explain the decrease in gingival inflammation for a given amount of plaque in immunosuppressed renal transplant patients. However, the clinical picture presented by HIV-associated periodontal disease may be due to an enhanced production of tumor necrosis factor by HIV-infected macrophages. Tumor necrosis factor is known to induce hemorrhagic necrosis and this can be enhanced by bacterial substances in plaque.14 Therefore, in AIDS patients there is gingival erythema with little plaque explaining the lack of correlation between GI and PI in HIV-P. This finding is in agreement with the observation that in patients with HIV-associated periodontal disease, extensive scaling and root planing in conjunction with improved oral hygiene did little to improve free gingival erythema.6 The results in Figure 17, showing samples from > H, HIV-H with respect to stained spirochetes, are in agreement with previous studies on bacterial morphotype distribution in periodontal health and disease in healthy patients. Studies by Listgarten and Hellden,15 Listgarten and Levin,16 and Loesche and Laughon17 have shown that the occurrence of spirochetes is much more abundant in plaque samples from periodontal pockets of patients than those from the gingival crevice of H patients. That HIV-P samples had more stained spirochetes than H samples but not HIV-H samples suggests that the healthy periodontium of HIV-H patients harbored more spirochetes than similar sites in H patients. The results of Figure 18 show that "other bacteria" as a group constituted the highest proportion of the 4 morphotypes of bacteria counted in stained specimens. These results are in conformity with the observation of Listgarten
J Periodontol 1991
September
and Levin16 that "other bacteria" as a morphotype were more abundant than the other morphotypes of bacteria in healthy periodontia than in diseased sites, i.e., samples from HIV-H > and HIV-P samples with respect to the morphotype "other bacteria." The significant finding shown in Figure 19, that samples from H (but not from HIV-H) had greater numbers of P. gingivalis than samples from HIV-P, was surprising. Such a finding is in complete variance with prior reports that this bacterium is associated with marked gingival inflammation and deep periodontal pockets in adult chronic Periodontitis.18-20 This finding was the more surprising since the procedure could apparently be biased towards the detection of P. gingivalis; i.e., specific fluorescing P. gingivalis cells were detected and enumerated first under blue light from a mercury illumination source before switching to darkfield illumination for counting of spirochetes as well as all other bacteria in the same field. The data of Figure 20 show the greater proportion of spirochetes in samples from the lesions of both and HIV-P patients, with samples exhibiting the greatest per cent occurrence of spirochetes (as enumerated under darkfield optics). Similar results were obtained with stained samples as shown in Figure 17. However, Loesche et al.21 also found that subgingival plaque from adult chronic Periodontitis patients had higher proportions of spirochetes and lower proportions of P. gingivalis. The apparent anomaly from the work of others who found more P. gingivalti cells in markedly inflamed gingival tissue and deep periodontal pockets18-20 was explained by the suggestion that a periodontopathic bacterial succession may exist in subgingival plaques—F. gingivalis may be replaced by spirochetes with the progression of disease. Thus P. gingivalis could be the microbial indicator of acute infections and spirochete that of chronic infections. It should be noted that our estimation of the per cent occurrence of P. gingivalis was not by anaerobic culturing technique on bacteriological media, but by indirect immunofluorescence using specific antiserum against a particular strain, namely, P. gingivalis ATCC 33277. This procedural difference could have contributed to our low count in samples from periodontal pockets, but this cannot explain the high counts of the bacterium in gingival crevices in healthy patients. But Rams et al.22 did not detect P. gingivalis in the lesions of HIV-P in 14 patients studied. They ascribed this observation to the effects of prior medical and dental therapies. The predominance of spirochetes in periodontally-diseased sites in is supported by the studies of several groups of workers. Moore et al.23 found that certain specific spirochetes were more closely associated with severe Periodontitis than they were with healthy or gingivitis sites. They suggested that Treponema denticola, a common oral spirochete, may have clinical significance in mature adults and is one of the more likely etiologic agents of severe Periodontitis.24 Furthermore, Simonson et al.25 demonstrated the first quantitative evidence of a positive relationship between a specific spirochete species and severe
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI, CHAN
585
Periodontitis. Our data (Figs. 17 and 20) also show that spirochetes were in higher occurrence in HIV-P sites than in sites of H, HIV-H, and HIV-G. This finding was corroborated by the study of Rams et al.22 who found more spirochetes in HIV-P sites (20.1%) than in HIV-H sites
Murray PA, Grassi M, Winkler JR. The microbiology of HIV-associated periodontal lesions. J Clin Periodontol 1989;16:636-642. 8. Löe H, Silness J. Periodontal disease in pregnancy. I. Prevalence and severity. Acta Odontol Scand 1963;21:533-551. 9. Silness J, Löe H. Periodontal disease in pregnancy. II. Correlation between oral hygiene and periodontal disease, Acta Odontol Scand
Our results show that there is no difference in microbiological profile in the selected putative periodontopathic bacteria chosen for study between chronic Periodontitis in systemically healthy adults and HIV-P. Murray et al.,7 using a different procedural approach and monitoring different bacteria, also reported that the microflora found in HIV-P is similar to that of classical adult Periodontitis. This generalization is reinforced by the study of Zambón and
10. Clark DC, Chin Quee T, Bergeron M-J, Chan ECS, Lauter-Lemay C, De Gruchy K. Reliability of attachment level measurement using the cementoenamel junction and a plastic Stent. / Periodontol
(0.7%).
colleagues12 which indicated that subgingival plaque in AIDS patients with Periodontitis can harbor high proportions of
the same periodontal pathogens as are associated with Periodontitis in non-HIV infected subjects. Rams et al.22 also reported that the predominant subgingival microflora in HIVP is in many ways similar to that of progressing Periodontitis lesions in systemically healthy adults. It should be noted that in our study there was a trend towards a greater occurrence of oral spirochetes in the gingival crevice of HIVH patients than in H subjects (Table 4). This suggests a modification in the tissue response to spirochetes of HIVH, probably because of immunodeficiency, and may presage the development of HIV-associated gingivitis.
Acknowledgment This study was supported by grant No. 6605-2705-A AIDS from the National Health and Development Program, Health and Welfare Canada.
REFERENCES 1. Marcussen DC, Sooy CD. Bacterial profiles of subgingival plaques in Periodontitis. J Periodontol 1985;56:401-405. 2. Phelan JA, Saltzman BR, Friedland GH, Klein RS. Oral findings in patients with acquired immunodeficiency syndrome. Oral Surg Oral
3.
4.
5.
6.
Med Oral Pathol 1987;64:50-56. Epstein JB, Silverman S Jr. Oral manifestations of human immunodeficiency virus infection. Recognition and Diagnosis. Can Dent Assoc J 1988;54:413^119. Silverman J, Migliorati CA, Lozada-Nur F, Greenspan D, Conant M. Oral findings in people with or at risk for AIDS: A study of 375 homosexual males. J Am Dent Assoc 1986;112:187-192. Winkler JR, Murray PA. Periodontal disease: A potential intraoral expression of AIDS may be rapidly progressive Periodontitis. Calif Dent Assoc J 1987;15:20-24. Winkler JR, Grassi M, Murray PA. Clinical description and etiology of HIV-associated periodontal diseases. In: Robertson PB, Greenspan JS, eds. Perspectives on Oral Manifestations of AIDS. Littleton, MA: PSG Publishing Co. Inc.; 1988:49-70.
7.
1964;22:121-135.
1987;58:115-118. 11. Chin Quee T, Bergeron M-J, Amsel R, Chan ECS. Staining method for monitoring subgingival bacteria associated with periodontal disease. / Periodont Res 1986;21:722-727. 12 Zambón JJ, Reynolds HS, Genco RJ. Studies of the subgingival microflora in patients with acquired immunodeficiency syndrome. J Periodontol 1990;61:699-704. 13. Oshrain HI, Mender S, Mandel ID. Periodontal status of patients with reduced immunocapacity. / Periodontol 1979;50:185-188. 14. Rothstein JL, Schreiber H. Synergy between tumor necrosis factor and bacterial products causes hermorrhagic necrosis and lethal shock in normal mice. Proc Nati Acad Sci (USA) 1988;85:607-611. 15. Listgarten MA, Hellden L. Relative distribution of bacteria at clinically healthy and periodontally diseased sites in humans. / Clin Periodontol 1978;5:115-132. 16. Listgarten MA, Levin S. Positive correlation between the proportions of subgingival spirochetes and motile bacteria and susceptibility of human subjects to periodontal deterioration. J Clin Periodontol
1981;8:122-138.
17. Loesche WJ, Laughon BE. Role of spirochetes in periodontal disease. In: Genco RJ, Mergenhagen SE, eds. Host-Parasite Interaction in Periodontal Diseases. Washington, D.C.: American Society for Microbiology; 1982:62-75. 18. Mayrand D, Holt SC. Biology of asaccharolytic black-pigmented Bacteroides species. Microbiol Rev 1988;52:134-152. 19. Spiegel CA, Hayduk SE, Minah GE, Krywolap GN. Black-pigmented Bacteroides from clinically characterized periodontal sites. J Periodont Res 1979;14:376-382. 20. White D, Mayrand D. Association of oral Bacteroides with gingivitis and adult Periodontitis. J Periodont Res 1981;16:259-265. 21. Loesche WJ, Syed SA, Schmidt E, Morrison EC. Bacterial profiles of subgingival plaques in Periodontitis. J Periodontol 1985;56:447456. 22. Rams TE, Andriolo M Jr., Feik D, Abel SN, McGivern TM, Slots J. Microbiological study of HIV-related Periodontitis. / Periodontol
1991;62:74-81.
23. Moore WEC, Holdeman LV, Smibert RM, Hash DE, Burmeister JA, Ranney RR. Bacteriology of severe Periodontitis in young adult humans. Infect Immun 1982;38:1137-1148. 24. Moore WEC, Holdeman LV, Cato EP, Smibert RM, Burmeister JA, Ranney RR. Bacteriology of moderate (chronic) Periodontitis in mature adult humans. Infect Immun 1983;42:510-515. 25. Simonson LG, Goodman CH, Bial JJ, Morton HE. Quantitative relationship of Treponema denticela to severity of periodontal disease. Infect Immun 1988;56:726-728.
Send reprint requests to: Dr. E.C.S. Chan, Department of Oral Biology, Faculty of Dentistry, McGill University, 3640 University Street, Montreal,
PQ H3A 2B2.
Accepted
for
publication
March
12, 1991.
Clinical Documentation and Occurrence of Putative Periodontopathic Bacteria in Human Immunodeficiency VirusAssociated Periodontal Disease Mervyn Gornitsky, *f D. Christopher Clark,f Russell Siboo,f Rhonda Amsel,f Irene Iugovaz,f Christine Wooley, * Nathalie Iuliani, * and Eddie CS. Chan*f Human immunodeficiency virus (hiv)-associated gingivitis (HIV-G) and HIVassociated Periodontitis (HIV-P) are two intraoral lesions manifested by patients with HIV infection. Periodontal indices were measured for 87 subjects in 5 study groups: HIV-seropositive patients with healthy periodontium (HIV- ), with HIV-G, or with HIVP; and non-HIV-infected subjects with healthy periodontium (H) or with adult chronic Periodontitis ( ). The quantitative clinical parameters were compared and statistically significant intergroup differences were noted. The mean scores on PI and PD do not discriminate between HIV-seropositive and non-HIV-infected seronegative cohorts, but a significant difference in the GI between HIV- and was noted. When categories of PD and AL are examined, some differences become apparent. Generally, the PD and AL of HIV-P are not as great as those of P. PI correlates well with GI (r 0.86) in P, in of selected but does not (r HIV-P. In the occurrence addition, 0.33) putative periodontopathic bacteria (Porphyromonas gingivalis, spirochetes, and motile eubacteria) in these lesions was determined by brightfield (after staining), darkfield and immunofluorescent microscopy. No difference in microbiological profile in the bacterial groups monitored was found between and HIV-P. Spirochetes were found to be more abundant than P. gingivalis in the lesions of and HIV-P. In marked contrast, P. gingivalis was found to be in highest numbers in samples from the gingival crevice of H as determined by indirect immunofluorescence. J Periodontol 1991; 62:576-585. =
=
Key Words:
HIV
infection; gingivitis/etiology; periodontitis/HIV-etiology; periodontal
index.
Oral lesions due to microbial infections and cancers may be the first manifestation of clinical disease of acquired immune deficiency syndrome (AIDS) caused by the human immunodeficiency virus (HIV), or oral lesions may arise subsequently from the down regulation of the immune system as a consequence of the HIV infection.1,2 The most common and significant oral lesions appear to be Kaposi's sarcoma, candidiasis, hairy leukoplakia, Herpes virus infection, and HIV-associated periodontal disease.3 Periodontal disease in the HIV-seropositive homosexual cohort is the single most important factor in their seeking dental treatment. It is seen in such frequency as to be considered an "opportunistic" disease.4,5 Pain and spontaneous bleeding are the complaints noted; these lead to *Dental
Department,
Sir Mortimer B. Davis-Jewish General
Montreal, PQ.
Faculty of Dentistry,
McGill
University, Montreal, PQ.
Hospital,
difficulty in mastication with a subsequent reduction in the ingestion of solid food for proper nutrition. The lesions of periodontal disease have been divided into two types by Winkler et al.:6 1) HIV-associated gingivitis (HIV-G); and 2) HIV-associated Periodontitis (HIV-P). HIV-G is defined
lesions confined to the soft tissues which showed a distinctive fiery red erythema and edema of the free gingiva, attached gingiva, and alveolar mucosa. Other characteristics include spontaneous gingival bleeding and failure to respond to therapy such as plaque removal and maintenance of good oral hygiene. HIV-P is defined as those sites with the symptoms as described for HIV-G and, in addition, there are extensive soft tissue ulcération and necrosis, severe loss of periodontal attachment with pain, and frank exposure of alveolar bone. Frequently, HIV-P affects only specific areas of the mouth whereas HIV-G is more generalized; besides pain, in HIV-P, mobility of involved teeth and halitosis are evident. as
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI, CHAN
Descriptions of HIV-G and HIV-P in the literature have been almost entirely narrative. Our study provides a quantitative documentation of the periodontal indices of HIVseropositive patients grouped as HIV-G, HIV-P and HIVH (healthy; i.e., no obvious periodontal disease symptoms). Control non-HIV-infected (seronegative) patients with healthy periodontium (H) and with adult chronic Periodontitis ( ) provided data for comparison with the data obtained from patient groups with HIV-G, HIV-P, and HIV-H. In addition, we present data on the relative percentages of Porphyromonas gingivalis, spirochetes, and motile bacteria in periodontally healthy and periodontally diseased sites of subjects, with and without HIV infection, based on cell numbers and not on the number of sites harboring specific bacteria as determined by Murray et al.7 The focus of this part of the study was to determine the occurrence of selected putative periodontopathic bacteria in the lesions of HIV-G and HIV-P sites in comparison with those found in the lesions of adult chronic Periodontitis in healthy patients. Knowledge of the types of microorganisms known to occur in a lesion may facilitate the formulation of improved treatment protocols for oral lesions that may enhance the quality of life in HIV-infected AIDS patients.
577
Figure 1. HIV-infected patient with healthy periodontium (HIV- ). Although the periodontium is asymptomatic for periodontal disease, there is hyperpigmentation (arrows).
MATERIALS AND METHODS Patients and Sampling Sites Patients for the study were received at the Dental Department, Sir Mortimer B. Davis-Jewish General Hospital, which is the major referral center in Montreal for dental and oral treatment of HIV-seropositive patients, 87% of whom were homosexuals. All HIV-infected patients had been tested seropositive for antibody to HIV by enzyme-linked immunosorbent assay (ELISA) at the Royal Victoria Hospital, Montreal, Quebec. All subjects gave informed consent to participate in the study, which was approved by an ethics committee of the Faculty of Dentistry, McGill University. In the selection of patients for the study, the parameters used for Periodontitis included both clinical and radiographic evidence of at least 2 sites with probing pocket depth greater than 5 mm. Radiographic interpretation was used to screen for sites with pockets due to alveolar bone loss. Patients with healthy periodontia had probing pocket depths of less than 3 mm and no evidence of gingivitis. Bleeding on probing to the base of the pocket was considered to be a sign of gingivitis. HIV-G and HIV-P were assessed according to Winkler et al.6 Eighty-seven subjects entered into the clinical documentation study (89 subjects in the microbiology study) were divided into 5 groups: 10 HIV-infected seropositive patients with healthy periodontium (HIV- ) (Fig. 1); 26 HIV-infected seropositive patients with HIV-G (Fig. 2); 24 (25 in microbiology) HIVinfected séropositive patients with HIV-P (Figs. 3 and 4); 9 (10 in microbiology) non-HIV-infected seronegative subjects with healthy periodontium (H), most of whom were
1 Figure 2. HIV-infected patient with gingivitis (HIV-G). There dence of plaque but erythema extends from tooth surface.
is
no
evi-
dental residents while the rest were staff members of the dental clinic; and 18 non-HIV-infected seronegative patients with adult chronic Periodontitis ( ). With each patient, 4 sites with greatest probing depth were selected for microbiological sampling. A site is defined as an interproximal area in either the anterior or posterior region where the proximal surfaces of 2 adjacent teeth are in contact. From each site 2 samples were obtained by inserting sterile paper points. Both paper point samples were placed in a sterile Eppendorf tube and transported to the laboratory where they were processed in a laminar flow biological safety hood. Clinical Evaluation All clinical examinations were carried out by 1 of 2 examiners and the same examiner was used for the same pa-
578
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
J Periodontol 1991
September
(CEJ) measurements were taken at 6 locations per tooth and
every tooth was measured. The 6 locations were at the mesial buccal, distal buccal, mesial lingual, distal lingual line angles, and on the mid-lingual and mid-buccal surfaces, using a set of William's periodontal probes. PD was measured from the gingival margin. The CEJ was employed as a fixed reference point for attachment level (AL), which was the distance between the CEJ and the base of the pocket as described by Clark et al.10 All measurements were to the nearest mm.
Figure 3. HIV-infected patient tion of soft tissue and bone.
with Periodontitis
(HIV-P).
Note destruc-
Figure 4. HIV-infected patient with Periodontitis (HIV-P). tion of soft tissue, bone and mucosa.
Note destruc-
tient
throughout the study.
Both examiners
were
calibrated
(M.G.) to ensure minimal interexaminer variability. Gingival index (GI) and plaque index (PI) were assessed on 6 designated teeth: upper right first molar (16), upper left central (21), upper left first bicuspid (24), lower left first molar (36), lower right central (41), and lower right first bicuspid (44) using mesial buccal, buccal, distal buccal, and lingual surface locations. If 1 of these teeth was missing, then an adjacent tooth or a tooth in the opposite arch was utilized. GI was scored according to the procedure of Löe and Silness;8 PI was assessed by the criteria of Silness and Löe.9 According to these authors, the GI and PI at 6 teeth, representing the 6 segments of the jaws, reliably reflect total mouth scores. Probing pocket depth (PD) and cemento-enamel junction by
one
of the authors
Treatment of Microbiological Samples Each site sample (consisting of two paper points from each site) was suspended in 0.5 ml sterile distilled water and mixed thoroughly with a vortex apparatus. Distilled water instead of buffer was used as suspended fluid for sample plaque material because the presence of salts interfere with the subsequent staining procedure below. Preliminary quantitative studies had shown no lysis of bacterial cells, including spirochetes, in this aqueous suspension. Soluble plaque material probably contributed to the integrity of any fragile cells. Comparable counts of bacterial morphological groups in subgingival plaque samples from 17 periodontally diseased sites support the observation that there was no apparent cell lysis.11 Ten µ aliquots of the sample were placed on glass microscope slides and air-dried; the smears were gently heat-fixed by passing the slides through a flame several times. The slides with unstained smears were stored at room temperature for 1 week (to destroy the infectivity of the virus), and then frozen at -20°C for 1 or 2 weeks until processed for microscopic examination. No evidence of bacterial cell degradation was observed upon such storage of cells. Microscopic examination at 1000 x magnification showed that the bacteria were morphologically intact and both P. gingivalis and oral spirochetes retained their antibody-reactive epitopes on storage. Heat-fixation was sufficient to halt any deterioration of the antibody-reactive
epitopes.
Treatment of Sample Smears The smears were treated with a modified Gray's flagella stain as described previously.11 This method yielded count-
ing results equivalent to that obtained by darkfield microscopy. For instance, flagellated cells did not lose their locomotor organdíes by the staining process. Stained specimens gave counts comparable to that of motile bacteria in the same sample obtained by darkfield microscopy. This staining procedure facilitated enumeration of spirochetes, flagellated (motile-since all motile eubacteria locomote by flagella) eubacteria, coccoid cells and all other bacteria under bright-field microscopy (Fig. 5). An indirect immunofluorescence assay (IFA) was used to detect P. gingivalis. Specific polyclonal antibodies against P. gingivalis ATCC 33277 were produced in New Zealand rabbits. The antiserum was diluted (1:20) in ice-cold (4 °C) 0.01M phosphate buffer, pH 7.2, containing 0.85% NaCl
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI, CHAN
• •
ß
*
579
-
10 UM
mm
Figure 5. Smear from a periodontal pocket sample stained with modified Cray's flagella stain showing spirochetes, flagellated cells, coccoids, and other bacteria under brightfield illumination. Bar 10 µ/ .
Figure 6. Fluorescent cells of P. gingivalis exhibiting typical morphology. Magnification same as Figure 5.
=
and 0.02% Tween 20 (PBST). The diluted antiserum was added to sample smears and allowed to interact with the bacteria for 20 minutes at room temperature and the smears were then washed twice (5 minutes per wash) with ice-cold PBST. Fluorescein-conjugated affinity purified goat antirabbit IgG adsorbed with human serum proteinsf was diluted (1:10) in ice-cold PBST was then added to the washed smears; the interaction was allowed to proceed for 20 minutes at room temperature. The smears were washed twice with cold PBST, as above, and once with cold distilled water for 1 minute. Mounting fluid (9 vol. glycerol: 1 vol. carbonate buffer, pH 9.0) was added to the smears. Cover slips were placed over the smears and sealed to the slides with clear nail varnish. The smears were examined by epifluorescence using a Leitz Dialux microscope at a magnification of 1000. An apple-green fluorescence of the bacterial envelope exhibiting short-rod morphology typical of the bacteroides was scored as positive for P. gingivalis (Fig. 6). The dilutions of the primary and secondary antisera used to enumerate P. gingivalis were determined with smears of gingivalis ATCC 33277 by checkerboard titrations. The P. gingivalis ATCC 33277 antiserum was shown to react with P. gingivalis cells from fresh plaque material; it also reacted with all strains of P. gingivalis in our stock culture. Test experiments showed no cross reaction between anû-P. gingivalis antiserum and Prevotella intermedia, Prevotella melaninogenica, and other species of black-pigmented anerobic rods. Nonspecific binding of the fluorescein-conjugated antiserum to bacteria in smears of subgingival plaque and P. gingivalis ATCC 33277 was determined as described above for the IFA but the primary antiserum was omitted. Nonspecific binding was not observed. Other workers also have used successfully polyclonal anti-P. gingivitis antiserum to enumerate P. gingivalis in smears of subgingival plaque from AIDS patients.7'12
tSigma Chemical Co.,
St.
Louis, MO.
Figure 7. Same field (and magnification) as Figure 6 as seen by darkfield microscopy. Compare with Figure 6 for the relative positions of P. gingivalis cells. Note the presence of spirochetes. For darkfield examination of the same field (Fig. 7) examined previously by epifluorescence, a transmitted quartziodine light source was used with numerical aperture of the oil-immersion objective stepped down by means of an adjustable rotating collar. The count under darkfield illumination represented the total bacterial count for that field. Depending on the cell density, 30 fields or sufficient fields totaling at least 100 cells of each site sample smear were examined alternatively by epifluorescence and darkfield illumination. Smears from 4 sample sites of each subject were examined.
Data Analysis Averages for PI, GI, and PD sites were subjected to oneway analysis of variance (ANOVA). For each subject, the
number of teeth with PD between 1 to 3 mm, 4 to 6 mm, and 7 or more mm were noted. A similar breakdown was
580
J Periodontol 1991
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
September
Table 1: Means of Periodontal Indices (standard error) for
Subject Groups
_Subject Groups*_
_Index_H_ _HIV- _HIV-G_HIV-P Plaque (PI) Gingival (Gl) Pocket depth (PD), *H
mm
0.55(0.09) 0.51(0.11) 2.26(0.07)
1.36(0.12) 1.36(0.10) 3.18(0.17)
0.57(0.10) 1.02(0.03) 2.19(0.07)
0.97(0.07) 1.27(0.03) 2.34(0.04).
1.27(0.10) 1.62(0.08) 2.86(0.10)
non-HIV-infected seronegative patients with healthy periodontium; non-HIV-infected seronegative patients with Periodontitis; HIV-H HIV-infected seropositive patients with healthy periodontium; HIV-infected seropositive patients with gingivitis (HIV-associated gingivitis); HIV-G HIV-P HIV-infected seropositive patients with Periodontitis (HIV-associated Periodontitis). =
=
= =
=
GINGIVAL INDICES
PLAQUE INDICES 2.0
H
Subject groups HIV0 HIV-G
HIV-P
8. Mean
Figure plaque indices of subject groups showing significant inlergroup differences by one way ANOVA. Abbreviations for subject groups defined in "Materials and Methods" in this and following figures. made for AL. These data were examined using one way multivariate analysis of variance and significant results were subjected to univariate analysis. All significant between group differences were examined using Tukey hsd post hoc tests. In addition, the correlation between PI and GI was calculated for each group. The microbiological profile was compared using 1-way independent groups analysis of variance. Microscopic scores of per cent spirochetes, motile cells, coccoids, and other types by staining, per cent P. gingivalis and spirochetes by immunofluorescence and darkfield microscopy, respectively, were averaged over the 4 sites sampled per subject. Post hoc tests were performed using Tukey hsd multiple range statistics. RESULTS The means for PI, GI, and PD for the subject groups are summarized in Table 1. There was a significant group effect for average PI (Fig. 8): the 2 Periodontitis groups (P and HIV-P) scored significantly higher than the 2 periodontally healthy groups (H and HIV-H). Post hoc tests of the significant group effect on the gingival index showed that the H group had significantly lower scores than all other groups and the HIV-H group differed
Tukey hsd HIV-P>H,HIV-H, HIV-G; P>H,HIV-H; HIV-G>H;HIV-H>H(p7 mm Attachment level 1-3 mm 4-6 mm >7 mm
0.78(0.03) 0.22(0.03) 0.00(0.00)
0.42(0.08) 0.43(0.05) 0.16(0.05)
0.81(0.04) 0.18(0.04) 0.01(0.01)
0.71(0.03) 0.28(0.03) 0.01(0.01)
0.56(0.04) 0.08(0.03)
0.91(0.03) 0.09(0.10) 0.00(0.00)
0.24(0.07) 0.53(0.06) 0.23(0.06)
0.82(0.06) 0.18(0.06) 0.00(0.00)
0.71(0.04) 0.26(0.04) 0.03(0.01)
0.40(0.06) 0.45(0.04) 0.15(0.04)
0.37(0.05)
*See Table 1 for for definitions.
MEAN FREQUENCY WITH PD 4-6 MM
POCKET DEPTH 0.6
Tukey hsd HIV-P>H,HIV-H,HIV-G; P>H,HIV-H(p7 MM
o-
.
Subject groups
0 HIV-P m HIV-H M HIV-G Figure 11. Mean frequency of PD 1 to 3 mm among subject groups showing significant intergroup differences by one-way ANOVA. results are shown in Table 3. The high correlation coefficient (0.86) in the subjects is consistent with expectation in periodontal disease in normal (non-HIV-infected) pop-
Subject groups HIV-H
0
HIV-G
HIV-P
Figure 13. Mean frequency of PD > 7 mm among subject groups showing significant intergroup differences by one-way ANOVA. ulation groups. However, there was a lack of correlation between PI and GI in the HIV-P subjects.
582
J Periodontol 1991
September
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
MEAN FREQUENCY WITH AL >7 MM
MEAN FREQUENCY WITH AL 1-3 MM
Subject groups H
0
HIV-H
0
HIV-G
Subject groups HIV-H
HIV-P
Figure 14. Mean frequency ofAL 1 to 3 mm among subject groups showing significant intergroup differences by one-way ANOVA. MEAN FREQUENCY WITH AL 4-6 MM
Tukey hsd P>H,HIV-H,HIV-G; HIV-P>H,HIV-H(p
Table 3. Pearson Correlation Coefficient Between PI and GI For Each Subject Group
Subject Group* HIV-H HIV-G HIV-P
r
.43 .86 .29 .30 .33
9 18 10 26 24
0.13 H, HIV-H); the sampled sites of HIVinfected patients with Periodontitis had significantly greater numbers of stained spirochetes than those of healthy individuals with healthy periodontium (i.e., HIV-P>H). Figure 18 shows the per cent occurrence of bacteria other than spirochetes, motile eubacteria, and coccoid cells in stained smears from the several subject groups. The morphotype "other bacteria" constituted the highest proportion of the 4 morphotypes of bacteria counted in the stained specimens. The results in the figure show that healthy sites of HIV-seropositive patients had significantly more "other bacteria" than diseased sites of Periodontitis subjects without or with HIV infection (HIV-H>P, HIV-P). The relative percentages of P. gingivalis compared to all other bacteria present in the various plaque samples are shown in Figure 19. This species, detected by indirect immunofluorescence, was significantly more numerous in healthy sites of non-HIV-infected subjects than diseased sites of HIV-infected patients with Periodontitis (i.e., H > HIV-P). The data in Figure 20 show the greater proportion of spirochetes in samples of both and HIV-P as seen by darkfield microscopy.
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI,
Table 4: The Mean Per Cent Occurrence of Selected Microbial
Subject Group
Groups (with
CHAN
583
standard error) in Each
_Subject Groups*_
Microbial
Group_H_ _HIV- _HIV-G_HIV-P
Spirochetes (stained)
1.3(0.6) 32.0(4.1) 1.7(0.9) 65.4(3.3) 24.0(5.0) 1.0(0.5)
Coccoids Motile cells Other bacteria P. gingivalis
Spirochetes (darkfield)
16.4(3.2) 23.2(2.5) 2.7(0.7) 58.1(2.6) 13.8(2.1) 14.1(2.5)
4.2(1.8) 23.2(2.2) 1.3(0.3) 71.1(2.5) 18.8(4.9) 2.9(1.7)
7.7(1.6) 23.0(2.0) 2.8(0.6) 66.5(2.5) 18.7(2.0) 5.0(1.5)
11.1(2.0) 27.2(1.9) 2.2(0.5) 59.4(2.2)
11.5(1.2) 9.5(1.8)
*See Table 1 for definitions.
STAINED SPIROCHETES
PORPHYROMONAS GINGIVALIS
Subject groups HIV-P 0 HIV-G Per cent occurrence of stained spirochetes showing significant
HIV-
Figure 17. intergroup differences by one-way ANOVA.
Figure
H 19. Per
!
Subject groups F23 HIV-G HIV-
cent occurrence
HIV-P
of P. gingivalis showing significant
in-
tergroup differences by one-way ANOVA.
SPIROCHETES BY DARKFIELD
OTHER BACTERIA
5
5
Study groups HIV-H 0 HIV-G
H 18. Per
Figure intergroup differences by
of "other bacteria" showing one-way ANOVA.
cent occurrence
Subject groups
W
HIV-P
significant
DISCUSSION The mean scores of PI and PD do not discriminate between seropositive and non-HIV-infected seronegative cohorts. But
Figure
mination
M
HIV-H
HIV-G counted by
HIV-P
of spirochetes darkfield illushowing significant intergroup differences by one-way ANOVA.
20. Per
cent occurrence
there is a significant difference between the GI of HIV-H and H groups (Table 1 and Fig. 9). Thus, even though the periodontia of HIV-H patients may be asymptomatic for
584
PERIODONTAL FLORA IN HIV-ASSOCIATED PERIODONTAL DISEASE
periodontal disease, there are marginal changes in the gingiva of these patients. When categories of PD and AL are examined (Table 2), some other differences emerge. In intermediate PD 4 to 6 mm, HIV-P is significantly greater
than H, HIV-H, and HIV-G. In PD of 7 mm or more, it is which differs from H, HIV-H, and HIV-G. HIV-P no longer has a significant difference with H, HIV-H, and HIV-G due to the destruction of soft tissue and bone in the HIV-P patient as disease becomes more severe. That is, PD does not seem to be a distinguishing feature of HIV-P because the rapid soft tissue necrosis occurs simultaneously with the loss of bone. This corroborates with the observed clinical picture. In the AL indices frequency distribution, the highest scores occur in the group in the 4 to 6 mm and the >7 mm categories (Table 2); i.e, is significantly greater than H, HIV-H, and HIV-G (Figs. 15 and 16), corroborating the observed clinical picture. In examining the correlations between GI and PI, PI correlates with Gì in patients; this supports the role of plaque in causing periodontal disease in non-HIV-infected subjects. However, in HIV-P patients, no significant correlation exists between PI and GI, which suggests a dissociation between plaque and periodontal disease in these patients. This lack of correlation was also noted by Oshrain et al.13 on renal transplant patients receiving imuran and corticosteroids (prednisone). The immuno-suppression brought about by drugs may explain the decrease in gingival inflammation for a given amount of plaque in immunosuppressed renal transplant patients. However, the clinical picture presented by HIV-associated periodontal disease may be due to an enhanced production of tumor necrosis factor by HIV-infected macrophages. Tumor necrosis factor is known to induce hemorrhagic necrosis and this can be enhanced by bacterial substances in plaque.14 Therefore, in AIDS patients there is gingival erythema with little plaque explaining the lack of correlation between GI and PI in HIV-P. This finding is in agreement with the observation that in patients with HIV-associated periodontal disease, extensive scaling and root planing in conjunction with improved oral hygiene did little to improve free gingival erythema.6 The results in Figure 17, showing samples from > H, HIV-H with respect to stained spirochetes, are in agreement with previous studies on bacterial morphotype distribution in periodontal health and disease in healthy patients. Studies by Listgarten and Hellden,15 Listgarten and Levin,16 and Loesche and Laughon17 have shown that the occurrence of spirochetes is much more abundant in plaque samples from periodontal pockets of patients than those from the gingival crevice of H patients. That HIV-P samples had more stained spirochetes than H samples but not HIV-H samples suggests that the healthy periodontium of HIV-H patients harbored more spirochetes than similar sites in H patients. The results of Figure 18 show that "other bacteria" as a group constituted the highest proportion of the 4 morphotypes of bacteria counted in stained specimens. These results are in conformity with the observation of Listgarten
J Periodontol 1991
September
and Levin16 that "other bacteria" as a morphotype were more abundant than the other morphotypes of bacteria in healthy periodontia than in diseased sites, i.e., samples from HIV-H > and HIV-P samples with respect to the morphotype "other bacteria." The significant finding shown in Figure 19, that samples from H (but not from HIV-H) had greater numbers of P. gingivalis than samples from HIV-P, was surprising. Such a finding is in complete variance with prior reports that this bacterium is associated with marked gingival inflammation and deep periodontal pockets in adult chronic Periodontitis.18-20 This finding was the more surprising since the procedure could apparently be biased towards the detection of P. gingivalis; i.e., specific fluorescing P. gingivalis cells were detected and enumerated first under blue light from a mercury illumination source before switching to darkfield illumination for counting of spirochetes as well as all other bacteria in the same field. The data of Figure 20 show the greater proportion of spirochetes in samples from the lesions of both and HIV-P patients, with samples exhibiting the greatest per cent occurrence of spirochetes (as enumerated under darkfield optics). Similar results were obtained with stained samples as shown in Figure 17. However, Loesche et al.21 also found that subgingival plaque from adult chronic Periodontitis patients had higher proportions of spirochetes and lower proportions of P. gingivalis. The apparent anomaly from the work of others who found more P. gingivalti cells in markedly inflamed gingival tissue and deep periodontal pockets18-20 was explained by the suggestion that a periodontopathic bacterial succession may exist in subgingival plaques—F. gingivalis may be replaced by spirochetes with the progression of disease. Thus P. gingivalis could be the microbial indicator of acute infections and spirochete that of chronic infections. It should be noted that our estimation of the per cent occurrence of P. gingivalis was not by anaerobic culturing technique on bacteriological media, but by indirect immunofluorescence using specific antiserum against a particular strain, namely, P. gingivalis ATCC 33277. This procedural difference could have contributed to our low count in samples from periodontal pockets, but this cannot explain the high counts of the bacterium in gingival crevices in healthy patients. But Rams et al.22 did not detect P. gingivalis in the lesions of HIV-P in 14 patients studied. They ascribed this observation to the effects of prior medical and dental therapies. The predominance of spirochetes in periodontally-diseased sites in is supported by the studies of several groups of workers. Moore et al.23 found that certain specific spirochetes were more closely associated with severe Periodontitis than they were with healthy or gingivitis sites. They suggested that Treponema denticola, a common oral spirochete, may have clinical significance in mature adults and is one of the more likely etiologic agents of severe Periodontitis.24 Furthermore, Simonson et al.25 demonstrated the first quantitative evidence of a positive relationship between a specific spirochete species and severe
Volume 62 Number 9
GORNITSKY, CLARK, SIBOO, AMSEL, IUGOVAZ, WOOLEY, IULIANI, CHAN
585
Periodontitis. Our data (Figs. 17 and 20) also show that spirochetes were in higher occurrence in HIV-P sites than in sites of H, HIV-H, and HIV-G. This finding was corroborated by the study of Rams et al.22 who found more spirochetes in HIV-P sites (20.1%) than in HIV-H sites
Murray PA, Grassi M, Winkler JR. The microbiology of HIV-associated periodontal lesions. J Clin Periodontol 1989;16:636-642. 8. Löe H, Silness J. Periodontal disease in pregnancy. I. Prevalence and severity. Acta Odontol Scand 1963;21:533-551. 9. Silness J, Löe H. Periodontal disease in pregnancy. II. Correlation between oral hygiene and periodontal disease, Acta Odontol Scand
Our results show that there is no difference in microbiological profile in the selected putative periodontopathic bacteria chosen for study between chronic Periodontitis in systemically healthy adults and HIV-P. Murray et al.,7 using a different procedural approach and monitoring different bacteria, also reported that the microflora found in HIV-P is similar to that of classical adult Periodontitis. This generalization is reinforced by the study of Zambón and
10. Clark DC, Chin Quee T, Bergeron M-J, Chan ECS, Lauter-Lemay C, De Gruchy K. Reliability of attachment level measurement using the cementoenamel junction and a plastic Stent. / Periodontol
(0.7%).
colleagues12 which indicated that subgingival plaque in AIDS patients with Periodontitis can harbor high proportions of
the same periodontal pathogens as are associated with Periodontitis in non-HIV infected subjects. Rams et al.22 also reported that the predominant subgingival microflora in HIVP is in many ways similar to that of progressing Periodontitis lesions in systemically healthy adults. It should be noted that in our study there was a trend towards a greater occurrence of oral spirochetes in the gingival crevice of HIVH patients than in H subjects (Table 4). This suggests a modification in the tissue response to spirochetes of HIVH, probably because of immunodeficiency, and may presage the development of HIV-associated gingivitis.
Acknowledgment This study was supported by grant No. 6605-2705-A AIDS from the National Health and Development Program, Health and Welfare Canada.
REFERENCES 1. Marcussen DC, Sooy CD. Bacterial profiles of subgingival plaques in Periodontitis. J Periodontol 1985;56:401-405. 2. Phelan JA, Saltzman BR, Friedland GH, Klein RS. Oral findings in patients with acquired immunodeficiency syndrome. Oral Surg Oral
3.
4.
5.
6.
Med Oral Pathol 1987;64:50-56. Epstein JB, Silverman S Jr. Oral manifestations of human immunodeficiency virus infection. Recognition and Diagnosis. Can Dent Assoc J 1988;54:413^119. Silverman J, Migliorati CA, Lozada-Nur F, Greenspan D, Conant M. Oral findings in people with or at risk for AIDS: A study of 375 homosexual males. J Am Dent Assoc 1986;112:187-192. Winkler JR, Murray PA. Periodontal disease: A potential intraoral expression of AIDS may be rapidly progressive Periodontitis. Calif Dent Assoc J 1987;15:20-24. Winkler JR, Grassi M, Murray PA. Clinical description and etiology of HIV-associated periodontal diseases. In: Robertson PB, Greenspan JS, eds. Perspectives on Oral Manifestations of AIDS. Littleton, MA: PSG Publishing Co. Inc.; 1988:49-70.
7.
1964;22:121-135.
1987;58:115-118. 11. Chin Quee T, Bergeron M-J, Amsel R, Chan ECS. Staining method for monitoring subgingival bacteria associated with periodontal disease. / Periodont Res 1986;21:722-727. 12 Zambón JJ, Reynolds HS, Genco RJ. Studies of the subgingival microflora in patients with acquired immunodeficiency syndrome. J Periodontol 1990;61:699-704. 13. Oshrain HI, Mender S, Mandel ID. Periodontal status of patients with reduced immunocapacity. / Periodontol 1979;50:185-188. 14. Rothstein JL, Schreiber H. Synergy between tumor necrosis factor and bacterial products causes hermorrhagic necrosis and lethal shock in normal mice. Proc Nati Acad Sci (USA) 1988;85:607-611. 15. Listgarten MA, Hellden L. Relative distribution of bacteria at clinically healthy and periodontally diseased sites in humans. / Clin Periodontol 1978;5:115-132. 16. Listgarten MA, Levin S. Positive correlation between the proportions of subgingival spirochetes and motile bacteria and susceptibility of human subjects to periodontal deterioration. J Clin Periodontol
1981;8:122-138.
17. Loesche WJ, Laughon BE. Role of spirochetes in periodontal disease. In: Genco RJ, Mergenhagen SE, eds. Host-Parasite Interaction in Periodontal Diseases. Washington, D.C.: American Society for Microbiology; 1982:62-75. 18. Mayrand D, Holt SC. Biology of asaccharolytic black-pigmented Bacteroides species. Microbiol Rev 1988;52:134-152. 19. Spiegel CA, Hayduk SE, Minah GE, Krywolap GN. Black-pigmented Bacteroides from clinically characterized periodontal sites. J Periodont Res 1979;14:376-382. 20. White D, Mayrand D. Association of oral Bacteroides with gingivitis and adult Periodontitis. J Periodont Res 1981;16:259-265. 21. Loesche WJ, Syed SA, Schmidt E, Morrison EC. Bacterial profiles of subgingival plaques in Periodontitis. J Periodontol 1985;56:447456. 22. Rams TE, Andriolo M Jr., Feik D, Abel SN, McGivern TM, Slots J. Microbiological study of HIV-related Periodontitis. / Periodontol
1991;62:74-81.
23. Moore WEC, Holdeman LV, Smibert RM, Hash DE, Burmeister JA, Ranney RR. Bacteriology of severe Periodontitis in young adult humans. Infect Immun 1982;38:1137-1148. 24. Moore WEC, Holdeman LV, Cato EP, Smibert RM, Burmeister JA, Ranney RR. Bacteriology of moderate (chronic) Periodontitis in mature adult humans. Infect Immun 1983;42:510-515. 25. Simonson LG, Goodman CH, Bial JJ, Morton HE. Quantitative relationship of Treponema denticela to severity of periodontal disease. Infect Immun 1988;56:726-728.
Send reprint requests to: Dr. E.C.S. Chan, Department of Oral Biology, Faculty of Dentistry, McGill University, 3640 University Street, Montreal,
PQ H3A 2B2.
Accepted
for
publication
March
12, 1991.
