Survival analysis of periodontal sites before and after periodontal therapy

Periodontal diseases appear to progress with bursts of destructive activity at individual sites. One effect of treatment might be to diminish the frequency of such bursts. Survival analysis was employed to seek such effects on the periodontal sites of 16 individuals with prior evidence of destructive periodontal disease. The subjects were monitored at bi-monthly intervals and actively breaking down sites were detected using attachment level measurements and the tolerance method of analysis. When active sites were detected, control sites of equal pocket depth and attachment loss were selected and microbiological and immunological samples were taken. The subjects were treated by modified Widman flap surgery and systemically administered tetracycline. On completion of therapy, bi-monthly monitoring was reinstituted. Life tables were constructed for periodontal sites in each of the 16 subjects prior to and after therapy. A site losing more than 3 mm of attachment at any time interval was considered to have relapsed or "died". Survivor functions were calculated for each time period indicating the % of sites which survived at any time. The subjects were divided into 3 categories on the basis of post-therapy survivor functions. The annual hazard rate in 9 good treatment response subjects (group 1) was reduced from 0.10/year to 0.01/year. The hazard rate of 5 intermediate treatment response subjects (group 2A) was reduced from 0.16/year to 0.04/year and that of 2 poor treatment response subjects (group 2B) from 0.15/year to 0.07/year. Group 2A and 2B individuals were combined and differences were sought in clinical, microbiological and immunological parameters between the good and poor treatment response groups. 5 out of 7 of the poor responding individuals showed elevated humoral antibody responses to 3 or more gram-negative subgingival species tested. Many of the elevated responses in this group were to organisms which are widely distributed and return quickly after therapy such as Fusobacterium nucleatum, Eikenella corrodens, Bacteroides intermedius and Capnocytophaga sputigena. The predominant cultivable microbiota in subgingival samples taken prior to therapy from the good responding group had significantly greater proportions of Actinobacillus actinomycetemcomitans, C. ochracea and B. intermedius than the poor responding group. The latter group showed significantly elevated proportions of F. nucleatum, Peptostreptococcus micros and Streptococcus intermedius.