CHOOSING IOL POWER IN PATIENTS WITH PREVIOUS RADIAL KERATOTOMIES: A FORMIDABLE CHALLENGE

Eyes previously submitted to radial keratotomies (RK) present an unusual challenge when selecting IOL power before cataract surgery. Post-RK eyes have unusually flat keratometric readings, irregular topography and are associated with small optical zones (less than 3 mm). ZEISS IOLMaster 700 swept source OCT biometer features total and posterior keratometry and may help provide better refractive outcomes in these altered corneas. With this study we intend to compare the accuracy of several IOL calculation formulae with the inclusion of posterior/ total keratometry measurements in refractive outcomes after cataract surgery in patients previously submitted to RK.

Refractive Surgery Unit of the Ophthalmology Department of Unidade Local de Saúde de Santo António (ULSSA), a tertiary hospital in Oporto, Portugal.

Retrospective analysis of patients with previous RK surgery submitted to cataract surgery with IOL implant in the bag at a tertiary hospital between January 2021 and July 2024. Refraction was performed by the same surgeon at least 3 months after surgery. According to the number of RK-cuts we defined 2 subgroups: group 1 (8 cuts) and group 2 (12 cuts). Studied formulae included: Cooke K6 with TK, Hoffer QST with TK, Kane with TK, Pearl DGS’s regular eyes module with TK, Pearl DGS’s RK module, Barret TrueK RK module with predicted and measured PK, EVO with PK and Holladay I’s double K method (as in the ASCRS calculator). Prediction error (PE) was calculated as the difference between the subjective refraction spherical equivalent (SE) and the targeted SE of each formula. Absolute error (AE) is the absolute value of the PE.

We included 13 eyes of 9 patients. All eyes but one were implanted with Clareon® CNA0T0 IOLs, the other was implanted with Acrysof® SN60AT IOL. 6 eyes showed 8 KR-cuts and 7 eyes showed 12 KR-cuts. Mean axial length was 27.55±1.97 mm, mean anterior keratometry was 36.25±4.29 D and mean total keratometry was 36.33±4.27 D. Mean AE was lowest for Barret TrueK No History RK with measured PKs (0.48±0.58 D), followed by Barret TrueK No History RK with predicted PKs (0.49±0.59 D) and Pearl DGS RK (0.69±0.44 D). Hoffer QST TK (0.93±1.00 D), Kane TK (0.89±0.50 D), Holladay 1 DoubleK method (0.84±0.67 D), EVO PK (0.76±0.48 D), CookeK6 TK (0.73±0.45 D) performed worse, with higher MAEs. For Barret TrueK No History with measured PKs, 77% eyes had a PE inferior to 0.5 D. No statistical differences were assessed comparing eyes with 8 or 12 KR-cuts.

Even when using total keratometry measurements, other modern formulae appear to be inferior to Barret TrueK No History RK with measured or predicted PKs. Accordingly, this formula ought to be preferred when choosing IOL power in cataract patients previously submitted to RK.