Update on Intraocular Lens Power Calculation Study Protocols: The Better Way to Design and Report Clinical Trials.

It was almost 40 years ago when one of the authors (K.J.H.) published an organized system to quantify the accuracy of intraocular lens (IOL) power calculation formulas, methods, and instruments. At the behest of the editor of the American Journal of Ophthalmology, the IOL Power Club (along with a statistician) published an editorial in 2015 modernizing and quantifying the proper protocols for these studies. Over the past decade, so many new optical biometers, formulas, and methods (whose accuracies have yet to be completely tested) have been introduced that we were asked to modernize and update these guidelines yet again to help others design and report correctly the results of clinical studies on IOL power calculation and biometry for 2020. We evaluated guidelines to enroll patients, including visual acuity minimums, exclusion of bilateral eyes, sample size issues, demographics (age, gender, and ethnicity), and whether such studies should not be performed using the same data that were used to develop the formula being tested. We showed the absolute need for constant optimization, which formulas should be tested for comparison, refraction measurement (testing distance), as well as the analysis of the prediction error (median and mean absolute errors; standard deviation; range of errors; percentage of eyes with a prediction within ±0.25 diopter [D], ±0.50 D, ±0.75 D, and ±1.00 D; and interquartile displays) and statistical methods of analyses. We present methods of ranking formula accuracy, including the new Haigis IOL Formula Performance Index. We also point out the issues of who programmed the formulas being tested, that all formulas used in the study must be referenced, and the software version number of all instruments used in the study should be stated clearly. The definition of anterior chamber depth should be stated as measured from the corneal epithelium to the lens. We hope that these recommendations will help researchers to improve the validity and accuracy of their studies with the ultimate goal to improve the accuracy of IOL power calculation.