Accuracy of SS-OCT biometry compared with partial coherence interferometry biometry for combined phacovitrectomy with internal limiting membrane peeling

Purpose To evaluate the accuracy of a swept-source optical coherence tomography (SS-OCT) biometer compared with a standard partial coherence interferometry (PCI) biometer in terms of phacovitrectomy for vitreomacular pathology and cataract. Setting Department of Ophthalmology, Ludwig-Maximilians-University Munich, Germany. Design Prospective, nonrandomized, single center consecutive case series. Methods Patients with age-related cataract, regular corneal astigmatism less than 1.25 diopters (D) and a clinically significant vitreomacular pathology (epiretinal membrane [ERM], macular hole, or vitreomacular traction [VMT] syndrome) who had combined phacovitrectomy with internal limiting membrane peeling were enrolled. Objective and manifest refraction, corrected distance visual acuity, SS-OCT biometry, and PCI biometry were obtained preoperatively and 3 months postoperatively. The refractive prediction error and mean absolute error (MAE) were calculated for each biometric device with the Haigis formula. The percentage of eyes with a prediction error within ±0.50 D or less were calculated. Results The study comprised 79 eyes of 79 patients. The mean refractive prediction error of all eyes was −0.27 D ± 0.76 (SD) for the PCI biometer and −0.28 ± 0.57 D for the SS-OCT biometer ( P  = .802), and the MAE was 0.61 ± 0.53 D and 0.46 ± 0.43 D, respectively ( P  = .012). A prediction error within ±0.50 D or less was observed in 55 eyes (70%) with the SS-OCT biometer compared with 42 eyes (53%) with the PCI biometer ( P  = .0001). The MAE with the SS-OCT biometer was statistically significantly lower in eyes with ERMs or macular holes compared with the PCI biometer ( P  = .043 and P  = .028, respectively), whereas no statistically significant difference was observed in eyes with VMT syndrome ( P  = .286). Conclusions The SS-OCT biometer showed a statistically significant lower MAE compared with the PCI biometer, and it is recommended for precise intraocular lens power calculations in combined phacovitrectomy for cataract and vitreomacular diseases.