Purpose: To assess the accuracy of different corneal powers for intraocular (IOL) power calculation in combined Descemet membrane endothelial keratoplasty (DMEK) and cataract surgery and investigate whether preoperative parameters correlate to the prediction error (PE).
Methods: This prospective case series involved patients with Fuchs endothelial dystrophy receiving combined DMEK and cataract surgery. Preoperatively, patients underwent optical biometry and anterior segment OCT (AS-OCT). AS-OCT measurements were repeated 6 months postoperatively, when final refraction was assessed. The PE was calculated using the preoperative average keratometry (Kave) measured by the optical biometer and User Group for Laser Interference Biometry (ULIB) constants. It was also calculated, after constant optimization, using the preoperative Kave from both devices and the total corneal power (TCP) measured by AS-OCT, as well as the postoperative Kave and TCP measured by AS-OCT.
Results: ULIB constants resulted in the highest hyperopic PE (P < .0001). Constant optimization improved the results, because the PE was zeroed out and the absolute PEs decreased. No significant difference was found among the Barrett Universal II, Emmetropia Verifying Optical 2.0, Haigis, Hoffer Q, Holladay 1, Kane, and SRK/T formulas. Further improvement was achieved with the postoperative Kave and TCP, although the accuracy remained moderate. The PE based on preoperative corneal measurements was correlated to the amount of corneal flattening; the latter could be predicted by multiple linear regression accounting for anterior and posterior corneal radii (P = .0002) and was correlated to the preoperative anterior/posterior ratio.
Conclusions: Constant optimization is beneficial for combined DMEK and phacoemulsification. Predicting postoperative corneal flattening may improve the results of IOL power accuracy. [J Refract Surg. 2022;38(7):435-442.].