The aim of this study was to evaluate a clinical and a microbiological test for monitoring tissue condition during supportive periodontal therapy (SPT) and to compare their diagnostic characteristics at implant and tooth sites. Twelve female (age: 37-72 years) and 7 male patients (age: 26-83 years) were evaluated in this study on the basis of availability to follow a rigid SPT program. Patients had received a complete periodontal examination at 1 and 5 years after implant placement. This included standardized radiographs obtained at implants and matching control teeth. One implant site and one tooth site per patient were followed during the last 2 years of the SPT program. At each recall visit microbiological samples were analyzed according to DNA/RNA analysis identifying periodontal pathogens (IAI Pado Test 4.5, Institute for Applied Immunology, Zuchwil, Switzerland). Presence or absence of bleeding on probing at these sites was also noted using a standardized probing force of 0.25 N (Audio Probe, ESRO, Thalwil ZH, Switzerland). The percentage number of recall visits with positive bacteriological test results and positive BOP scores were calculated. Disease progression at the sites was defined if the annual increase in probing depth was > or = 0.5 mm/year (2.5 mm in 5 years) or if the annual decrease in CADIA values (Computer Assisted Densitometric Image Analysis) was more than -0.7 per year (-3.5 in 5 years). Changes below these values were considered as negative test results indicating stability of the sites. The diagnostic characteristics (sensitivity, specificity, positive and negative predictive values) of BOP and microbiological tests alone or in combination were then calculated using two-by-two tables. By application of increasing thresholds of BOP frequencies set for definition of positive test outcome (BOP > or = 10% > or = 20% > or = 25% > or = 50% > or = 75% > or = 90% or the combined BOP > or = 75%, but DNA positive > or = 10%, > or = 25% > or = 34% > or = 50% > or = 67% > or = 90%) receiver operator characteristics curves (ROC) were constructed for teeth and implants. The areas under the ROC curves were calculated and compared by means of chi-square tests. The results indicated statistically significant better diagnostic characteristics of both tests at implants compared to teeth. The inclusion of an additional microbiological test significantly enhanced the diagnostic characteristics of BOP alone at teeth as well as at implants.