For average eyes, formula differences are modest. At the extremes of axial length, selection has a measurable effect on outcomes.
For the average cataract patient, the formula printed on the biometry report matters less than the quality of the measurements, the optimization of lens constants, and the surgeon's target refraction. But at the extremes of axial length, formula selection has a measurable effect on outcomes — and understanding where that threshold lies is clinically useful.
In eyes with axial length between approximately 22mm and 25mm, differences between modern optimized formulas are generally modest. Published comparisons consistently show that well-optimized versions of SRK/T, Hoffer Q, Holladay 1, and Barrett Universal II perform within a narrow range of each other in this population. That said, small systematic differences may still matter in premium refractive cataract surgery, where a 0.15–0.20D bias can shift a patient from their target refraction. Melles et al. (2018, Ophthalmology), in a study of over 13,000 eyes, found Barrett Universal II among the top-performing formulas across the full axial length range, including in eyes of average biometry.
Legacy vergence formulas become less reliable in long eyes, partly because effective lens position assumptions — derived from regression analyses of average populations — become less accurate at the extremes of biometric geometry. SRK/T in particular tends to underestimate required IOL power in long eyes, increasing the risk of postoperative hyperopia.
Melles et al. found Barrett Universal II among the strongest performers in longer eyes. Cooke and Cooke (2016, JCRS), comparing nine formulas across a large dataset, similarly found newer formulas outperforming legacy options in eyes outside the average range. In highly myopic eyes with axial lengths beyond 28–30mm, axial length correction methods (such as the Wang-Koch adjustment) and careful attention to biometry quality are additional important variables that no formula alone can compensate for.
For axial lengths below 22mm, Hoffer Q has historically been among the stronger performers. Its personalized anterior chamber depth model gives it an advantage in hyperopic eyes, and it has been validated specifically in this population by Eom et al. (2014, JCRS), who found it outperformed SRK/T and Holladay 1 in eyes under 22mm. Modern formulas including Barrett Universal II and Kane also perform well in short eyes and are reasonable defaults when a single formula approach is preferred across all axial lengths.
Rather than fixed thresholds — which depend on biometer, lens model, constant optimization, keratometry quality, and ACD — a more defensible approach is:
IOLDx Clinical displays comparative formula outputs across SRK/T, Hoffer Q, Holladay 1, and Haigis simultaneously, allowing surgeons to assess spread and identify outliers before finalizing power selection.
IOLDx Clinical displays SRK/T, Hoffer Q, Holladay 1, and Haigis side by side for every lens.
Open IOL Planner →