The prognostic value of several periodontal factors measured as radiographic bone level variation: A 10-year retrospective multilevel analysis of treated and maintained periodontal patients

Background: Assigning a prognosis to a periodontal patient is one of the greatest challenges in clinical practice. Many different factors can affect the result of periodontal therapy. The purpose of this study was to evaluate the prognostic value of some clinical, genetic, and radiographic variables in predicting bone level variation in periodontal patients (aged 40 to 60) treated and maintained for 10 years. Methods: Sixty consecutive non-smoking patients (mean age 46.77 +/- 4.96) with moderate to severe chronic periodontitis were treated with scaling and root planing (SRP). Some patients also underwent additional surgical treatments. All patients were maintained in the same private practice for 10 years. At baseline (TO) and at least 10 years later (T2), the following clinical variables were evaluated: probing depth (PD), tooth mobility (TM), presence of prosthetic restorations (PR), and molar teeth (MT). In addition, radiographic measurements were taken of the mesial and distal distances from the cemento-enamel junction (CEJ) to the bottom of the defect (BD), to the bone crest (BC), and to the root apex (RA). At T2, a genetic test to determine the IL-1 genotype and genetic susceptibility for severe periodontal disease was performed for all 60 patients. Based on the results of this assay, the patients were categorized as IL-1 genotype positive (G+) or negative (G-). The differences between the bone levels measured at TO and T2 (DeltaBD), indicating the bone level variation, was used as the outcome variable. Different predictor variables were then tested using a 3-level statistical model (multilevel statistical analysis; patient, tooth, and site level). At the patient level these were: age, gender, and interaction between mean bone loss and the IL-1 genotype (mean CEJ-BDT0 X IL-1 genotype). At the tooth level the variables were: TMT0, PRT0, MTT0; and at the site level the evaluated factors were: the infrabony component of the defect (CEJ-BDT0-CEJ-BCT0), PDT0, bone level (CEJ-BDT0), and the residual supporting bone (BD-RA(T0)). Results: Among the considered predictor parameters, the following were significantly associated with the outcome variable: 1) mean CEJ-BDT0 x IL-1 genotype (P = 0.0019); 2) TMT0 (P <0.0000); 3) CEJ-BDT0 (P<0.0000); 4) CEJ-BDT0 - CEJ-BCT0 (P<0.0000); 5) PDT0 (P = 0.0010). Deeper probing depths at a site and tooth mobility at baseline were associated with worst prognosis. Greater CEJ-BDT0 distance and infrabony component at a site at baseline were associated with a better prognosis. The interaction between mean CEJ-BD measurement at baseline and IL-1 genotype was significantly associated both with a good or a poor prognosis. The other parameters evaluated - age, gender, presence of molars and prosthetic restorations, and residual supporting bone - were not significantly associated with bone level variation. Conclusions: Within the scope of this study design, many traditional prognostic factors were ineffective in predicting future bone level variation and therefore were of no prognostic value. Conversely, a few specific factors at each level emerged as valuable prognostic factors. At the patient level, the prognostic factor was initial mean bone level in conjunction with a positive IL-1 genotype. At the tooth level, the prognostic factor was tooth mobility. At the site level, the significant prognostic factors were initial bone level at a site, the infrabony component of a defect, and initial probing depth at a site. The use of these factors may be of value to clinicians as predictors of bone level variation when assigning a prognosis to a patient, a tooth, or a site.