Study Population and Design

One-hundred seventy-five patients included in the authors’ database of GTR treated sites from previously published investigations were selected to participate in this study by identifying cases presenting for periodontal maintenance over a 4-month period between February and May 2002. All subjects had previously received comprehensive periodontal treatment including oral hygiene instructions, scaling, prophylaxis, and GTR therapy at at least one intrabony defect at least 2 years before this study. All patients were systemically healthy; did not use medications; and had good oral hygiene at the time of periodontal surgery. Full mouth root planing was performed with or without access flap, as needed. In each patient, one site, presenting with an intrabony defect, was treated with GTR with either non-resorbable or absorbable barrier membranes, as previously described, or a combination of absorbable membranes and alloplastic biomaterials. During the first year after completion of active treatment, all patients were recalled at monthly intervals for re-instruction in oral hygiene procedures and professional tooth cleaning. Only one defect per patient was included in this study. In cases that received GTR treatment at multiple sites, the first defect that received treatment was selected for this study.

At 1 year, all patients were given the opportunity to participate in a periodontal maintenance program that included oral hygiene instructions and professional tooth cleaning every 3 months. Fifty-eight patients did not participate and received community-based care from their general dentist and a visit for oral hygiene and follow up measurements in the authors’ practice at least every second year.

Clinical Measurements

Full-mouth and local plaque scores were recorded as the percentage of total surfaces which revealed the presence of plaque. Bleeding on probing was assessed dichotomously at a force of 0.3 N with a manual pressure sensitive probe. Full-mouth and local bleeding scores were calculated. Probing depths (PD), CAL, and recession of the gingival margin (REC) were recorded to the nearest millimeter. Measurements were taken at baseline before GTR treatment and at 1 year. Thereafter, CAL and PD were measured during maintenance visits, at least every second year. Attempts to minimize potential bias, derived from the fact that the assessor was the same clinician who performed the treatment, included making the patient record unavailable before all follow-up measurements. The width of the radiographic angle of the defect was also measured at baseline on periapical radiographs as previously described.

Evaluation of Patient Characteristics

Each patient was further characterized in terms of additional descriptors known to be of importance in determining the prognosis of conventional periodontal therapy: age, gender, smoking status, and participation in the maintenance program. Smokers were defined as smoking ≥10 cigarettes per day. Participation in the maintenance program was defined as compliance with appointments every 3 months over the study period (up to 16 years). Patients who did not participate in this program received community-based care from their general dentist. A subset of 86 randomly selected subjects was further evaluated in terms of previously reported polymorphisms at the IL-1 gene complex using a commercially available genotyping service.

Data Analysis

Statistical analysis was performed using a statistical software package. The survival analysis was performed using a multivariate approach with the following definitions of events: tooth loss; loss of ≥2 mm of CAL with respect to the baseline pre-GTR CAL; and loss of ≥2 mm of CAL with respect to the 1-year post-GTR CAL.

The Kaplan-Meier method was used to construct survival plots and life-table analyses. The significance of the impact of patient characteristics on survival of GTR treated teeth was assessed constructing multi-variate Cox proportional hazard models.