Effect of implantable collamer lens on biometric measurements and intraocular lens power calculation formulas

To investigate the effect of the implantable collamer lens (ICL) on biometric measurements and intraocular lens (IOL) power calculation formulas.

This prospective theoretical study included 92 eyes undergoing ICL surgery. Preoperative and 1-year postoperative biometric parameters, including axial length (AL), keratometry (K), anterior-chamber depth (ACD), lens thickness (LT), central corneal thickness (CCT) and corneal diameter (CD), were measured using IOLMaster 700. Nine formula types (10 variants) were evaluated: Barrett Universal II (BU II; BU II^a, IOLMaster 700-based; BU II^b, online calculator), Cooke K6, EVO, Hill-RBF, Hoffer QST, Holladay 1, Kane, Pearl-DGS and SRK/T. Primary outcomes were the difference in predicted spherical equivalent before and after ICL implantation (Δ-pSEQ) and the percentage of eyes within specified Δ-pSEQ ranges.

For all eyes (n = 92), significant post-ICL changes included increased AL, steep K (K2), LT, CCT (+0.02 mm, +0.08 D, +0.26 mm, +0.003 mm; p = 0.031, 0.021, <0.001, 0.008) and decreased ACD (-0.28 mm, p < 0.001). ACD segmentation errors occurred in 34.78% of eyes, associated with low vault (optimal cut-off, 0.491 mm; AUC = 0.921). Eyes were grouped by segmentation accuracy (correct, n = 60; incorrect, n = 32). All Δ-pSEQ values were within ±0.50 D; between-subgroup differences in Δ-pSEQ ranges were not significant (p > 0.05). The highest proportions within ±0.25 D were SRK/T (86% overall, 94% incorrect group) and BU II^a (85% correct group); EVO had the lowest (71%-72%).

ICL implantation alters biometric measurements and theoretically predicted Δ-pSEQ; although these changes are statistically significant, their clinical significance is negligible. While incorrect ACD segmentation is common, it is unlikely to substantially affect the theoretical calculation results.