Methods for Intraocular Lens Power Calculation in Cataract Surgery after Radial Keratotomy

To compare methods of calculating the required intraocular lens (IOL) power for patients undergoing cataract surgery after radial keratotomy (RK), including the 2016 update of the True K formula.

Retrospective case series.

A total of 52 eyes of 34 patients who had sequential RK and cataract surgery performed in the same institution by 1 of 2 surgeons.

Seven IOL calculation formulae were evaluated: True K [History], True K [Partial History], True K [No History], Double-K Holladay 1 (DK-Holladay-IOLM), Potvin-Hill, Haigis, and Haigis with a -0.50 diopter (D) offset. Biometry was obtained with the IOLMaster 500 (Carl Zeiss Meditec AG, Jena, Germany) and Pentacam (OCULUS Inc, Arlington, WA) devices. Subjective refraction was performed at 4 to 6 weeks postoperatively. The achieved spherical equivalent outcome was compared with the target outcome to calculate the absolute error for each eye with each formula.

Median absolute error (MedAE) and mean absolute error (MAE), and percentage of patients within ±0.50 D, ±0.75 D, and ±1.00 D of refractive target. Mean error (ME) was also calculated to demonstrate whether a formula tended toward more myopic or hyperopic outcomes.

Best results were achieved with the True K [History]. The MedAE was higher (0.382 vs. 0.275) with the True K [Partial History], but a similar percentage of patients (75.0%-76.6%) achieved within ±0.50 D of target. Of the methods that do not require refractive history, the True K [No History] and unadjusted Haigis were most accurate (69.2% within ±0.50 D of target), with the True K [No History] returning the lowest MedAE but also more of a tendency toward hyperopia (ME +0.269 vs. -0.006 for Haigis). The DK-Holladay-IOLM and Potvin-Hill methods were the least accurate.

Knowledge of the refractive history significantly improves the accuracy of IOL calculations in patients undergoing cataract surgery after previous RK. The post-RK refraction appears to be the most important parameter, with inclusion of the pre-RK refraction offering a further slight improvement in MedAE. When no refractive history is available, the True K [No History] and Haigis formulae both perform well, with the added advantage of not requiring data from separate biometric devices.