Biometry and Formula Accuracy With Intraocular Lenses Used for Cataract Surgery in Extreme Hyperopia

Purpose

To audit intraocular lens (IOL) power predictions for cataract surgery in extreme hyperopia and to compare the accuracy across different biometry formulae and IOL types.

Design

A retrospective analysis of 76 eyes from 56 patients undergoing cataract surgery with IOLs ranging in power from 30 to 35 diopters (D).

Methods

Axial lengths, corneal powers and anterior chamber depths were measured with ultrasound or optical methods, and the IOLMaster (Carl Zeiss Meditech, Inc, Dublin, California, USA) software was used to predict the refractive outcome for each IOL used. Differences between the predicted and actual postoperative refraction were then analyzed for each formula.

Results

In practice, 55% of patients were within ±1.0 D of the refraction predicted by their surgeon. In theory, the Haigis formula would have given the smallest mean refractive error (+0.51 ± 0.12 D), followed by the Hoffer Q (−0.70 ± 0.14 D), Holladay 1 (−1.11 ± 0.13 D), and SRK/T formulae (−1.45 ± 0.14 D). The Haigis formula overpredicted the lens power required, which would have generated a myopic result. The other formulae underpredicted the lens power required and would have generated a hyperopic result. There was a significant difference between lens designs: the Haigis was more accurate for open-loop, whereas the Hoffer Q was more accurate for plate-haptic lenses. The anterior chamber depth measurement could also be used to predict changes in intraocular pressure after surgery.

Conclusion

This represents the largest published series to date of biometry predictions for cataract surgery in extreme hyperopia and confirms the Haigis formula to be the most accurate. A consistent difference between open-loop and plate-haptic lenses suggests that haptic design may influence the effective lens position in very small eyes. We further propose a simple formula to optimize the Haigis and Hoffer Q formulae in patients with extreme hyperopia.