To compare the accuracy of the intraocular lens (IOL) power calculation of the Zeiss IOLMaster versus conventional automated keratometry and contact acoustic biometry after personalized optimization.Methods
Three hundred twenty eyes of 249 patients consecutively receiving phacoemulsification and IOL implantation with the sutureless clear cornea approach were enrolled. Preoperative biometry was derived from the Zeiss IOLMaster and an acoustic device (Alcon OcuScan RxP), and keratometry was measured by the Zeiss IOLMaster and a conventional automated keratometer (Topcon KR-8800). One month after surgery, refraction was measured and the predicted refractive errors were calculated with personalized optimization.Results
For eyes responsive to all devices, IOLMaster biometry + IOLMaster keratometry had the best predictability for postoperative refraction, with a mean absolute error (MAE) of 0.38 ± 0.28D, followed by OcuScan RxP biometry + IOLMaster keratometry (MAE, 0.49 ± 0.34D) and OcuScan RxP biometry + KR-8800 keratometry (MAE, 0.54 ± 0.37D) (P < 0.05 for all paired comparisons). For eyes that could not be measured by IOLMaster biometry, the MAE was smaller with IOLMaster keratometry (0.62 ± 0.56D) than with KR-8800 keratometry (0.57 ± 0.52D) (P = 0.03). The variables of age, diabetes mellitus, severity of cataract, axial length, and corneal curvature were unrelated to the predictability of postoperative refraction.Conclusions
The Zeiss IOLMaster yielded more accurate refractive outcomes than the conventional automated keratometry and contact acoustic biometry after personalized optimization. For eyes irresponsive to axial length measurement by the IOLMaster, keratometry of the IOLMaster was still superior to conventional automated keratometry.