To evaluate the accuracy of intraocular lens (IOL) power calculations according to the chosen formulas and anterior chamber depths in eyes with short axial lengths.

A retrospective analysis was performed on 57 eyes of 50 patients (axial length < 22.0 mm) and 42 eyes of 36 patients (22.0 mm ≤ axial length < 25.0 mm) who underwent cataract surgery. IOL power was calculated with the SRK II, SRK/T, Binkhorst, Holladay I, and Hoffer Q formulas. The differences between the predicted refraction and the actual refraction were compared and analyzed. The errors according to the anterior chamber depth were also evaluated.

The SRK II formula showed a lower predictive accuracy, and the other formulas showed similar accuracies in eyes with short axial lengths. The Holladay 1 and Hoffer Q formulas showed good predictive accuracies in eyes with short axial lengths. Hyperopic shift tended to occur with all formulas in eyes with short axial lengths. When using SRK II and SRK/T formulas, a correlation between axial length and hyperopic shift was observed, with shorter axial length patients shifting to more hyperopic conditions. In eyes with short axial lengths, larger hyperopic shifts tended to occur in eyes with relatively deeper anterior chambers.

In eyes with short axial lengths, preoperative predicted IOL power showed good accuracies with Holladay 1 and Hoffer Q formulas. Preoperative anterior chamber depth and axial length had a strong influence on the accuracies of predicted IOL power.