Abstract
AIM: To evaluate the influence of ocular biometric parameters on the performance of actual near-add power in the spectacle plane of multifocal intraocular lenses (MIOLs). METHODS: This retrospective study defined prediction error (PE) as the difference between actual postoperative near-add power and manufacturer-predicted values. Linear regression assessed PE correlations with axial length (AL), mean keratometry (K), pupil size, anterior chamber depth (ACD), lens thickness (LT), ACD+0.5×LT, and back-calculated IOL power. Differences in PE across MIOLs types, AL, K, pupil size, ACD, LT, ACD+0.5×LT, and back-calculated IOL power groups were compared. RESULTS: Totally 250 eyes of 250 patients (116 males and 134 females, mean age 56.22±12.31y) who underwent phacoemulsification with MIOL implantation were reviewed. PE showed no significant correlation with most parameters but had a weak positive correlation with LT. The mean predicted error (MPE) in the AL≤22, 26-27, and 27-28 mm groups were 0.83 (0.51, 1.01) D, 0.78 (0.51, 1.07) D, and 0.72 (0.57, 0.94) D respectively (P<0.001). The MPE in the K between the 45-46 D groups was 0.78 (0.37, 1.07) D (P=0.0004). The MPE in the LT between the 4.5-5.0 mm groups and ≥5 mm was 0.72 (0.50, 1.01) D and 0.72 (0.51, 1.01) D respectively (P<0.001). The MPE in the back-calculated IOL power ≥25 D was 0.86 (0.60, 1.01) D (P<0.001). CONCLUSION: Although there is no significant linear relationship between near-add power and most ocular biometry parameters, and specific segments of AL, K, LT, and IOL power have a significant impact on the near-add power of MIOLs. These factors should be taken into account in preoperative evaluations.