The association between phenytoin-induced gingival hyperplasia (PIGH) and five major periodontopathic bacteria was investigated by real-time PCR in 12 subjects (30.6 +/- 7.95 years of age) with severe motor and intellectual disabilities. The subjects had not received antibiotics or periodontal therapy within three months. A healthy gingival site or a site with the lowest gingival hyperplasia score (minimum site), and a site with the maximum score of gingival hyperplasia (maximum site) on each subject were selected for investigation. The clinical parameters were measured after microbial sampling. The following results were obtained: 1. The detection rates of both Treponema denticola and Porphyromonas gingivalis were 75% (9/12) at the maximum sites, and that of Tannerella forsythia was 33.3% (4/12). The detection rate of Prevotella intermedia was 58.3% (7/12), but no significant correlation was observed between this bacterium and the other types of bacteria in quantitative analysis. 2. The detection rate of P. gingivalis with type II fimA was 25% (1/4) at the minimum sites, and 83.3% (5/6) at the maximum sites. 3. The samples were divided into three groups; namely, those with no detection of P. gingivalis at either the minimum sites or the maximum sites, no detection at the minimum sites but detection at the maximum sites, and detection at both sites. In conclusion, 1) the detection rates of T. denticola and P. gingivalis were the highest at the maximum sites. In particular, the detection rate of P. gingivalis with type II fimA was high in the present study, and 2) the subgingival microflora was diverse because of the diverse severity of gingival inflammation associated with PIGH.