Objective: Dental restorations may be plaque retentive. The aim of this study was to evaluate the long-term association between proximal restorations and the incidence and progression of periodontal disease in well-maintained patients. Method and Materials: Probing pocket depths (PPD), bleeding on probing (BOP), and radiographic status of proximal restorations were retrospectively retrieved from files of patients attending a specialist periodontal office. Ill-fitting margins were recorded. The association between these parameters was evaluated at baseline examination (T0), after cause-related therapy (T1) and after ≥ 10 years from T0 (T2), during which supportive periodontal therapy (SPT) was administered, using descriptive statistics, ANOVA-Bonferroni, and chi-square analyses. Results: 1,301 teeth were examined. Mean PPD in unrestored surfaces was 3.7 ± 1.7 mm, 3.1 ± 1.3 mm, and 2.8 ± 1 mm at T0, T1, and T2, respectively. Deeper pockets were found in restored surfaces at those time points with PPD values of 4.4 ± 1.8 mm, 3.6 ± 1.4 mm, and 3.2 ± 1.1 mm, respectively (P .001). Higher PPD values were found in restored surfaces exhibiting inadequate restorations when compared to restored surfaces with adequate restorations at all time points. These values were 4.9 ± 1.9 mm, 4.1 ± 1.5 mm, and 4 ± 1.7 mm vs 4.3 ± 1.8 mm, 3.6 ± 1.4 mm, and 3.1 ± 1.1 mm, respectively (P .001). Conclusion: The present study confirmed that restorations might be detrimental to periodontal health. A significant association between the presence of proximal restorations and the incidence of periodontal disease was observed. This association was more pronounced for inadequate restorations while becoming less significant over time in patients receiving routine SPT. Keywords: dental restorations, longitudinal study, periodontal disease