Andrei Georgescu, Sorin Andrian, Cristina-Angela Ghiorghe, Galina Pancu, Claudiu Topoliceanu, Ionuţ Tărăboanţă, Irina Nica, Răzvan Brânzan, Cristina Dascălu, Maria-Alexandra Mârţu, Mihaela Sălceanu
Background/Objectives: Caries adjacent to restorations or sealants (CARS) is among the most frequent biological causes of composite restoration failure, resulting from the interplay between the patient's cariogenic risk and restoration-related factors such as marginal adaptation. The study aimed to investigate whether cariogenic risk modifies the effect of marginal adaptation on the occurrence of caries adjacent to posterior composite restorations (CARS), using risk-stratified univariate analysis and a clustering-adjusted approach, to identify associated clinical, socio-demographic, and local risk factors. Methods: This retrospective cross-sectional study evaluated 453 posterior direct composite resin restorations (Class I, Class II) in 88 patients at the Faculty of Dental Medicine, "Grigore T. Popa" University, Iași, Romania. CARS and marginal adaptation were assessed using FDI Criteria. Patients were classified as high or low cariogenic risk using a multifactorial institutional protocol. Univariate analysis and Generalized Estimating Equations (GEE) accounting for within-patient clustering were performed. Results: CARS prevalence was 27.9% in high cariogenic risk patients versus 3.15% in low cariogenic risk patients. Univariate analysis identified high cariogenic risk (OR = 11.904), deficient marginal adaptation (OR = 25.929-69.475), restoration age of 3-5 years (OR = 3.045), and Class II configuration (OR = 2.597) as significant risk factors. GEE analysis, adjusted for clustering of restorations within patients, confirmed high cariogenic risk (OR = 12.597) and deficient marginal adaptation (OR = 121.320) as independent correlates of CARS. A significant interaction between the two (p < 0.001) showed that marginal adaptation remained the dominant local risk factor in low cariogenic risk patients, while its association with CARS was largely attenuated in high-risk patients. Conclusions: GEE analysis confirmed high cariogenic risk and deficient marginal adaptation as independent correlates of CARS, with marginal adaptation acting as the dominant local risk factor only in low-risk patients.