Abstract
Purpose :
The accuracy of the Holladay 2 (H2) formula has been well documented to improve refractive outcomes for cataract surgery. The derivations of such formulae remain unspecified due their proprietary nature. Use of additional biometric parameters to enhance outcomes is reported to be one source of improvement over previous formulae. This exploratory study looks to identify relative importance of various biometric parameters on intraocular lens calculations and also cite places for potential human error input as more variables are included in every increasingly complex formulae.
Methods :
Four adult volunteers were imaged with IOL Master 500 (Zeiss) containing H2 Formula. We selected the TECNIS ZCB00 lens. The study population included one high myope with history of bilateral photorefractive keratectomy, one high myope with mild cataract, one mild myope and one hyperopic volunteer. Attention was given to seven modifiable biometric variables of interest: Axial Length (AL), Keratometry (K), Horizontal White To White (WTW), Anterior Chamber Depth (ACD), Manifest Refraction (MRx), Lens Thickness (LT) and Age of subject (A). Each of these variables were manually and independently modified with a subsequent derivation of change in recommended IOL averaged across all patients.
Results :
For each one mm change of AL, there was 1.625D change in recommended IOL power. For each 1D change of average K, there was a 1.12D change of IOL power. For the allowed range of manual input of ACD (1-8 mm) there was an average maximum effect of 1.18D. For the allowed range of input of WTW (8.5-15 mm) there was an average maximum 0.79D potential error. For the allowed range of MRx max/min (-25D to +15D) spherical equivalent had a maximum average 2.64D change. For the allowed range or LT (1-9mm) had a maximum 0.7D of effect. On average, changing the age by 50 years changed the recommended IOL power by -0.345D.
Conclusions :
Each input variable is important and has potential to alter recommended IOL power. Independently, the order of significance would be AL, K, ACD, WTW, MRx, LT and A respectively. In the context of increasingly complex newer generation formulae, this analysis gives general insight on the significance of various bioparameters independently. Further in depth analyses with changing multiple variables would further elucidate the importance of accuracy in measurement devices as well as user input.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.