Abstract
Purpose :
Human error has been examined in the preoperative setting in multiple studies, however little attention has been dedicated to the role of human error in the input of biometric variables used in intraocular lens calculation formulas. The aim of this study is to determine the impact of deviance in biometric variables used in newer generation IOL formulae, and to examine which types of IOL calculation errors due to human input have the greatest effect on target lens selection and expected refraction using a test model eye.
Methods :
A standard model eye used to test the IOL Master 500 (Carl Zeiss Meditec AG, Germany) with the Holladay II formula was examined and multiple biometric variables were adjusted. The variables of base sphere (BS), vertex distance (VTX), white to white distance (WTW), and lens thickness (LT) were adjusted for multiple target refractions (TMRx). BS and TMRx of -10, -5, 0, +5, and +10 diopters were used for each of the independently adjusted variables. The main outcomes were the calculated IOL power (CIOL), and expected manifest refraction required to achieve emmetropia (ExMRx). The absolute error and standard deviations were calculated for each independently adjusted variable. ANOVA tests were performed for all variables. Hyperopic changes in BS & TMRx were compared to myopic changes using ANOVA, Tukey HSD and Bonferonni tests.
Results :
There was no statistically significant difference in effect on CIOL or ExMRx between all variables on ANOVA testing (p = 1.00). However, hyperopic changes in TMRx resulted in greater CIOL changes than myopic changes for each tested variable and were statistically different on ANOVA testing [F(5,24) = 4.436, p = 0.005]. Post hoc analysis using Tukey HSD and Bonferroni tests indicated that the difference between VTX -3 and VTX +3 variables were statistically significant (p = 0.001), while other variables were not statistically different.
Conclusions :
Hyperopic refractive states appear to be subject to greater predicted effect from input variables than myopic states, but this was only significant for the variable of vertex distance. Other IOL input variables of base sphere, white-to-white distance and lens thickness did not have a significantly different effect on CIOL or ExMRx compared to one another. Personnel performing lens calculations should be made aware of potential sources of human error during lens calculations and be especially careful with data entry for high hyperopic eyes.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.