Clinical Comparison of 4D Phaco and Torsional Ultrasound Phacoemulsification Platforms: Association with Cumulative...

Background/Objectives: Four-dimensional Phaco and torsional ultrasound represent distinct approaches to phacoemulsification, and both platforms are compatible with a hybrid polymer-coated tip designed to reduce posterior capsule contact risk. Clinical comparative data evaluating both ultrasound efficiency and postoperative inflammatory response remain limited. Methods: We retrospectively compared 149 eyes of 139 patients undergoing phacoemulsification with the UNITY (4D Phaco) or CENTURION (torsional ultrasound) platform, each combined with a standard metal or hybrid polymer-coated tip, evaluating cumulative dissipated energy (CDE), postoperative aqueous flare, endothelial cell loss, visual acuity, and surgical time. Because CENTURION and UNITY were used during non-overlapping periods, the device comparison reflects a historical rather than concurrent design; tip-use periods overlapped substantially for CENTURION but only briefly (approximately three days) for UNITY. A 2 × 2 factorial regression model with heteroscedasticity-consistent (HC3) robust standard errors estimated device and tip effects and their interaction. Results: UNITY was associated with 52.8% lower geometric mean CDE (GMR 0.472; 95% CI, 0.413-0.539; p < 0.0001) and 31.6% lower 1-month aqueous flare (GMR 0.684; 95% CI, 0.580-0.806; p < 0.0001) than CENTURION. The hybrid tip was associated with 35.9% higher CDE (GMR 1.359; 95% CI, 1.190-1.551; p < 0.0001) but no significant flare effect. Endothelial cell loss, day-1 flare, visual acuity, and surgical time did not differ significantly by device or tip, with no device × tip interaction for any outcome (all p ≥ 0.073). Conclusions: UNITY was associated with substantially lower CDE and lower postoperative aqueous flare than CENTURION, without a demonstrable difference in endothelial cell loss or visual outcome, and without significant modification by tip type. Because the device comparison was historical rather than concurrent, these findings should be regarded as hypothesis-generating pending confirmation with concurrent device allocation.