Frequency distributions of periodontal attachment loss. Clinical and microbiological features.

The present investigation attempted to determine if the pattern of past periodontal destruction could be concisely summarized, and related to other clinical and microbiological parameters. 61 subjects between the ages of 12 and 61 years with destructive periodontal disease were evaluated at 6 sites per tooth for redness, plaque, suppuration, bleeding on probing, pocket depth, and attachment level. The frequency distribution of baseline attachment level measurements was computed for each individual. A curve fitting algorithm was used to fit the frequency distribution to 1-, 2-, and 3-term normal distributions. The parameters of the fit could be used to summarize concisely all of the frequency distributions. 3 major patterns of attachment loss could be distinguished. Pattern I required a two-term distribution with localized destruction at less than 34% of sites and was further divided into 3 groups, depending on average attachment loss at diseased sites. The means of the second peak for the subgroups were 2.7, 5.3, and 8.6 mm, respectively. Pattern II exhibited more widespread disease (greater than 33% of sites affected) with multiple peaks in the frequency distribution requiring a 3-term distribution for satisfactory fit. However, a significant proportion of sites was not affected. Pattern III exhibited a single-peaked normal distribution in which virtually all sites were affected. Mean attachment levels of the peaks in this group ranged from 2.7 to 8.4 mm. 23 of the 61 subjects showed significant attachment loss at 1 or more sites during the course of bi-monthly monitoring, as determined by the tolerance method of analysis. Subgingival plaque samples were taken from these active sites and matched with control sites prior to therapy. The proportions of Fusobacterium nucleatum, Streptococcus intermedius, and Eikenella corrodens were significantly elevated in active and control sites of subjects in groups II and III combined (the widespread disease groups), and proportions of Actinobacillus actinomycetemcomitans and Propionibacterium acnes were elevated in active and control sites of the more localized disease group I subjects. Group I subjects showed a 13- to 15-fold decrease in hazard rates of periodontal sites after Widman flap surgery and systemic tetracycline, whereas groups II and III subjects showed 2-to 6-fold decreases.