Comparative Study of Three IOL Power Calculation Formulae for Asian Eyes Shorter Than 22mm or Longer Than 25mm

Overview

Implantation of an intraocular lens is the gold standard in modern day cataract surgery. The appropriate lens power needed to achieve the desired refractive outcome can be calculated with a whole variety of formulas. To date, there has been no prospective study conducted to evaluate the accuracy of the IOL power calculation formulae commonly in use. It is well established that the frequently used IOL formulas do not show significant differences when used in eyes of average axial length (i.e. between 22mm and 25mm) and it is at the extremes of axial lengths where discrepancies arise. Our aim is to find the most appropriate formula(e) for these 'long' and 'short eyes' particularly in our population where there is a significant proportion of high myopes.

Our primary objective is to compare three IOL power calculation formulae and determine the most appropriate formula for accurate prediction of postoperative refractive error in Asian eyes with axial length less than 22mm and greater than 25mm that are undergoing phacoemulsification cataract surgery. This will be a randomized prospective study in patients with axial lengths less than 22mm and greater than 25mm undergoing phacoemulsification cataract surgery and IOL implantation. Each patient will be randomly assigned to have their IOL power calculated using one of three IOL power calculation formulae (SRK-T, Hoffer Q and Holladay 2). A randomization list will be used to decide the IOL formula allocation. The primary outcome measure of this study is the mean absolute error (MAE) which is calculated by subtracting the intended formula-derived preoperative refractive error from the actual postoperative refractive error. Based on the estimates from a previous study, it is postulated that the MAE for Hoffer Q, Holladay 2, and SRK-T are +0.36 D, +0.53 D \& +0.74 D, respectively. The interest here is pair-wise comparisons amongst SRK-T, Hoffer Q and Holladay 2. The sample size calculations for each pair-wise comparison is based on a SD of 0.57 D with a power of 80% and a 2-sided test of 5%. Using equal randomization, a total of 176 subjects would be sufficient to achieve statistical differences between Hoffer Q over the Holladay 2 \& SRK-T and also the Holladay 2 over SRK-T (standardized effect difference of 30% between Hoffer Q \& Holladay 2; 67% \& 37% effect size for Hoffer Q vs SRK-T and Hollady 2 vs SRK-T, respectively). Factoring in an attrition rate of 20%, a total sample size of 210 eyes will be recruited for the study. Preoperatively, Snellen visual acuity will be assessed and all patients will undergo a non-cycloplegic autorefraction, keratometry measurement and axial length measurement with the Zeiss IOL Master (see below Visit 1). All patients will undergo phacoemulsification and 'in-the-bag' IOL implantation (Visit 2). Surgery will be performed through a 2.75mm temporal clear corneal incision and only the Alcon MA60BM or MA60MA 3-piece acrylic foldable IOL will be implanted. A standard postoperative topical antibiotic and anti-inflammatory regime consisting of tobramycin and beclomethasone will be administered (Visit 3-5). Patients will be examined at the following intervals: Screening Within 90 days prior to surgery Preoperative (Visit 1) Within 30 days prior to surgery Operative (Visit 2) Day of surgery Postoperative (Visit 3) 1 day after surgery Postoperative (Visit 4) 3-5 days after surgery Postoperative (Visit 5) 4-5 weeks after surgery Postoperative (Visit 6) 3-4 months after surgery Postoperatively, logMAR visual acuity will be recorded at every visit and non-cycloplegic refraction will be carried out at Visit 5 and Visit 6.