OCT for early quality evaluation of tooth-composite bond in clinical trials

Objectives To evaluate early quality of composite restorations with a universal adhesive in different application modes clinically and with optical coherence tomography (OCT). Methods: 22 patients with four non-carious cervical lesions each received composite restorations (Filtek Supreme (TM) XTE, 3 M). The universal adhesive Scotchbond Universal (TM)(SBU, 3 M) was applied with three etching protocols: self-etch (SE), selective-enamel-etch (SEE) and etch-and-rinse (ER). The etch-and-rinse adhesive OptiBond (TM) FL (OFL, Kerr) served as a control. Restorations were imaged by OCT (Thorlabs) directly after application (t(0)). After 14 days (t(1)) and 6 month (t(2)) OCT imaging (interfacial adhesive defects) was repeated combined with clinical assessment (FDI criteria). Groups were compared by Friedman-/Wilcoxon- and McNemar-Test. Results: No differences were seen clinically between groups (p(i) >= 0.500). OCT assessment revealed more adhesive defects at the enamel interface with SBU/SE at t0-t2 compared to all groups (pi <= 0.016). OFL showed more defects than SBU/ER (t(1): p = 0.01; t(2): p = 0.083). At dentin/cementum interface OFL exhibited more adhesive defects than SBU with all conditioning modes (t(0), t(1), p(i) <= 0.003) and at t(2) to SBU/SE and SBU/ER (p < 0.001). Since t(1) defects with SBU were detected more frequently in the SE and SEE modes compared to ER (p(i) <= 0.037). In contrast to SBU defects increased with OFL up to t(2) (p(i) <= 0.007). Conclusions: In contrast to clinical evaluation, OCT revealed subtle adhesive defects directly after application that might interfere with clinical success. It was demonstrated that ER does not decrease initial adhesion of SBU to dentin.