Modeling the relationship between clinical, microbiologic, and immunologic parameters and alveolar bone levels in an elderly population.

A cross-sectional periodontal study of 74 subjects aged 65 to 75 years was performed. Clinical data were collected and related to microbiological and immunological data. A statistical model (step-wise multiple regression) of factors related to bone loss was created initially using clinical data only; then by adding either the microbiologic or immunologic data; and then by using clinical, microbiologic, and immunologic data together. When only clinical data were considered, three factors were found to have significant positive correlations with bone loss. Tooth mobility accounted for 17% of the variability in the alveolar bone level measurements, probing depth for 12%(r2), and plaque index for 3%, for a total of 32% of the variability explained by these clinical factors. Tooth mobility and probing depth were clinical factors which remained significant in the model when the microbiological data were also considered. As percentages of the total cultivable microbiota, E. corrodens (r2 = 14%) and black-pigmenting Prevotella intermedia (r2 = 4%) correlated positively with alveolar bone loss. The addition of the microbiologic data only increased the r2 to 33%. When immunological data were considered with the clinical data, pocket depth and tooth mobility were the clinical parameters which remained in the model. IgG antibody levels to P. gingivalis W83 and/or 381 (r2 = 24%) A. actinomycetemcomitans 627 (r2 = 2%) were the significant immunologic measures having a positive correlation with bone loss. Anti-F. nucleatum levels had a significant negative correlation. A total of 50% of the variability in alveolar bone level was accounted for in the model by the addition of specific serum antibody levels to subgingival plaque microorganisms. When clinical, microbiological, and immunological measurements were all considered together, antibody to P. gingivalis W83 and/or 381 (r2 = 42%), percentage of B-lymphocytes (r2 = 3%), probing depth (r2 = 4%), anti-E. corrodens levels (r2 = 2%), and anti-P. gingivalis 33277 levels (r2 = 4%) all had significant positive correlation with loss of alveolar bone. The number of enteric bacteria, anti-F. nucleatum levels, and anti-P. intermedia levels each had a significant negative correlation with alveolar bone heights. The r2 for this model was 75%. These results indicated that antibody levels to subgingival plaque microorganisms and tooth mobility were the best predictors of bone loss in the elderly patients tested in this study.