A Comparative Study of the Predictive Accuracy of Four Iol Power Calculation Formulae in Phacoemulsification with Foldable Iol Surgery in Eye with Axial Length Less Than 22 Mm.

Abstract

Purpose: To compare the predictive accuracy of four intra-ocular lens power calculation formulae—Hoffer Q, Holladay 1, Haigis, and Hill-RBF—in cataractous eyes with axial length (AL) < 22 mm.

Methods: In this single-centre prospective study, 80 eyes of 80 patients underwent uneventful phacoemulsification with a single-piece hydrophilic acrylic IOL. Pre-operative biometry (IOLMaster 500) provided AL, keratometry and anterior chamber depth. IOL power was calculated with all four formulae, but implantation was based on Hoffer Q. Manifest refraction was recorded on postoperative day 7 and day 30. Absolute prediction error (AE = |predicted – achieved spherical-equivalent|) was the primary outcome. Differences were analysed with one-way ANOVA and Bonferroni post-hoc testing (α = 0.05).

Results: Mean age was 57.9 ± 5.8 years; 60% were women. Mean AL was 21.48 ± 0.40 mm and mean implanted IOL power was 24.76 ± 1.76 D. Spherical equivalent stabilized between day 7 and day 30 (p = 1.000). Mean ± SD AE differed significantly between formulae (p < 0.001): Hill-RBF 0.034 ± 0.217 D; Haigis 0.097 ± 0.403 D; Hoffer Q 0.231 ± 0.492 D; Holladay 1 0.403 ± 0.669 D. Hill-RBF out-performed Haigis (p = 0.031) and both surpassed Hoffer Q and Holladay 1 (all p < 0.001). Ninety-one percent of eyes calculated with Hill-RBF were within ±0.50 D of target versus 84% (Haigis), 65% (Hoffer Q) and 57% (Holladay 1).

Conclusions: Hill-RBF provided the most accurate refractive prediction in short eyes, followed by Haigis. Surgeons managing AL <22 mm should preferentially employ modern formulae that incorporate anterior segment metrics and artificial-intelligence modelling.