Background: Microleakage at the gingival margin of Class II composite restorations remains a key failure pathway, driven by polymerization-shrinkage stress in high configuration-factor (C-factor) cavities and the lower bondability of dentin/cementum. Objective: To compare microleakage scores (primary endpoint) at gingival margins between a nanofilled composite and a conventional microhybrid; secondarily, to compare score distributions and enamel-margin (occlusal) sealing as indicators of marginal sealing ability. Methods: Forty extracted, non-carious human molars were randomized to restoration with a nanofilled composite (Filtek™ Supreme Ultra) or a microhybrid (Filtek™ Z250) using a total-etch adhesive (Adper™ Single Bond 2). Standardized Class II mesio-occlusal cavities extended 1.0 mm apical to the CEJ. After thermocycling (5000 cycles, 5–55 °C), specimens were immersed in 2% methylene blue, sectioned into two halves, and scored (0–4) at gingival and occlusal margins under stereomicroscopy by two blinded examiners. The worst of two sections per tooth was used a priori; sensitivity analysis using the mean of two sections was concordant. Non-parametric analyses (Mann–Whitney U; Fisher–Freeman–Halton exact test) with Holm correction for multiplicity were applied. Inter-examiner reliability was excellent (Cohen’s κ = 0.89); intra-examiner κ = 0.91. Results: Gingival-margin microleakage was lower with nanofilled composite: median (IQR) 1 1–2 vs 2 2–3 for microhybrid; U = 102.5; Z = –2.96; p = 0.003; Rosenthal’s r = 0.47 (medium–large). Minimal/no leakage (scores 0–1) occurred in 65% of nanofilled vs 25% of microhybrid restorations; severe leakage (scores 3–4) occurred in 5% vs 30%, respectively (exact p = 0.028). Occlusal (enamel) margins showed low leakage in both groups with no significant difference (median 0 in both; p > 0.40). Conclusions: Under standardized in vitro conditions, the nanofilled composite reduced gingival-margin microleakage compared with a microhybrid by one full score unit in the median, with a medium–large effect size (r≈0.47) and excellent scoring reliability. Given method and model limitations, these findings suggest improved marginal sealing ability for nanofilled composites in Class II restorations and warrant validation with 3D methods and clinical studies.
Vasudevan et al. (Fri,) studied this question.