Search

PURPOSE To compare the accuracy of the Barrett Universal II (BUII) five-variable formula to previous generation formulae in calculating intraocular lens (IOL) power following paediatric cataract extraction. METHODS Retrospective study of consecutive paediatric patients who underwent uneventful cataract extraction surgery along with in-the-bag IOL implantation between 2012 and 2018 in the Hospital for Sick Children, Toronto, Ontario, Canada. The accuracy of five different IOL formulae, including the BUII, Sanders-Retzlaff-Kraff Theoretical (SRK/T), Holladay I, Hoffer Q and Haigis, was evaluated. Constant optimization was performed for each IOL and for each formula separately. Mean prediction error (PE) and the mean and median absolute PE (APE) were calculated for the five different IOL formulae investigated. RESULTS Sixty-six eyes of 66 children (59% males) with a median age at surgery of 6.2 years (IQR, 3.2-9.2 years) were included in the study. The mean IOL power that was implanted was 23.3 ± 5.1 D (range; 12.0-39.0 D). Overall, the BUII had a comparable median APE to the Hoffer Q, Holladay I, SRK/T and Haigis formulae (BUII: 0.49D versus 0.48D, 0.61D, 0.74D and 0.58D respectively; p = 0.205). The BUII, together with Hoffer Q, produced better predictability within 0.5D from target refraction compared with the SRK/T formula (BUII:51.5%, Hoffer Q:51.5% versus SRK/T:31.8%, p = 0.002 for both). CONCLUSION The BUII formula had comparable accuracy to other tested formulae and outperformed the SRK/T formula, when calculating IOL power within the 0.5D range from target refraction in paediatric eyes undergoing cataract surgery with in-the-bag IOL implantation.

Background To evaluate the accuracy of predicted refraction using multifocal intraocular lenses (IOLs) with power calculation based on two biometric devices and multiple IOL power calculation formulas. Design A retrospective study conducted in a private practice ophthalmology clinic. Participants Seventy-three eyes of 48 patients were evaluated. Methods Consecutive cases of eyes that had undergone successful cataract surgery with an implantation of a multifocal IOL (SN6AD1, Alcon Laboratories, Inc., Fort Worth, TX, USA) by a single surgeon were enrolled. Patients were meticulously screened for suitability for a multifocal IOL implantation according to our clinic guidelines. Two biometric measurement devices (IOLMaster-500 [Carl Zeiss Meditec AG, Jena, Germany] and Lenstar-LS900 [Haag-Streit AG, Koeniz, Switzerland]) were used under strict validation criteria to evaluate the predicted refraction errors for the Holladay 1, SRK/T, Hoffer Q, Haigis, Holladay 2, Barrett Universal II and Olsen formulas. Main Outcome Measures Predicted refractive sphere equivalent (RSE) errors. Results The measurements obtained from the two biometric devices were highly correlated. The standard deviation of the error in predicted RSE and the median absolute error were similar for the IOLMaster and the Lenstar devices using all formulas, ranging from 0.27 dioptres (D) to 0.31D and from 0.15D to 0.21D, respectively. A high percentage of eyes had an error in predicted RSE within ±0.5D from target refraction, ranging from 86.3% to 93.2%. Conclusions High accuracy can be achieved in multifocal IOL power calculations by appropriate patient selection, precise biometry and appropriate IOL calculation formulas.