Algorithm for determining equivalent A-constants and surgeon factors.

Purpose: To describe statistical algorithms for determining surgeon factors and corresponding A-constants and compare them to empirical data.

Methods: The Holladay and SRK/T equations are rearranged to develop a series of equations expressing the surgeon factor as a function of the A-constant or, alternatively, the A-constant as a function of the surgeon factor. These expressions are statistically manipulated using keratometric and axial length distributions to determined clinically equivalent A-constant-surgeon factor pairs. Predictions made by this algorithm are fit to a linear equation that accurately relates surgeon factors to equivalent A-constants and are compared to a set of corresponding A-constants and surgeons factors used by the Food and Drug Administration for labeling purposes. Algorithm performance is assessed by calculating the difference between the Holladay and SRK/T intraocular lens power equations, establishing equivalence criteria. These calculations are performed using clinically equivalent A-constant-surgeon factor pairs with an axial length and average corneal curvature corresponding to the population mean.

Results: The difference calculation is less than or equal to 0.02 diopter (D) for A-constants ranging between 110.0 and 120.0. A comparison of the algorithm with empirically derived corresponding A-constant-surgeon factor pairs shows that the two methods are identical for A-constants ranging between 117.0 and 119.0; however, differences in equivalent surgeon factors predicted by these methods increase with decreasing A-constants. The magnitude of this difference is 0.44 mm at an A-constant of 110.0, resulting in a difference of 0.45 D in equivalence criteria.

Conclusions: These data demonstrate that the statistical algorithm provides an improvement in A-constant-surgeon factor equivalent for A-constants less than 117.0. The structure of this algorithm can easily be adapted to interrelate other pairs of personal constants and serves as a theoretical method to standardize personal constants.