AGREEMENT BETWEEN REAL VAULT IN IMPLANTABLE COLLAMER LENS (ICL) AND PREDICTED VAULT USING ANTERIOR SEGMENT OPTICAL COHERENCE TOMOGRAPHY (AS-OCT) BASED FORMULAS.

Selecting the most appropriate ICL size is of high importance as it minimizes the chance of postoperative complications. The vault, distance from the crystalline lens to the ICL posterior surface, is a postoperative parameter associated with the ICL size. Several studies aimed to predict this distance preoperatively, by using AS-OCT parameters, and claim to explain approximately 40% of the vault variability. However these formulas’ performance has been rarely tested. This study aimed to characterise the prediction ability of two vault formulas (ST-formula and KS-formula) by comparing the measured vault against the predicted one.    

Ophthalmology Clinic Vista Sánchez Trancón, Badajoz, Spain

Retrospective case series analysed 209 eyes (1eye-1patient) implanted with spherical or toric myopic ICL (EVO-V4c, STAAR Surgical, USA). Preoperative parameters, namely anterior chamber width, anterior chamber depth, crystalline lens rise and central corneal thickness were measured using the AS-OCT (Visante, Zeiss). At three-months postoperatively the vault was measured using the same device. The two formulas tested were the ST-formula (Sánchez-Trancón et al. 2020, JCRS) and the KS-formula (Igarashi et al. 2019, JCRS). Agreement between measured and predicted vaults was assessed through mean differences (MD) and the limits of agreement (LoA) calculated as the 95% confidence interval of the differences.  

The postoperative ICL mean vault was 629.5±322.7 µm. The ST-formula predicted a mean vault of 689.7±194.5 µm and the KS–formula 732.22±216.6 µm. The MD and LoA were, MD:-57.2 (LoA:-563.8; 449.5) µm and MD:-102.2 (LoA:-675.3; 469.9) µm for the ST-formula and KS-formula, respectively. Approximately 27% and 29% of the eyes had measured and predicted vault differences within 100 µm, rising to 65% and 57% within 250 µm, for the ST-formula and KS-formula, respectively. When divided by vault type, 62.9% (ST-formula) and 58.2% (KS-formula) of the vault predictions fell into the same vault group as the measured vault.    

The vault prediction formulas tend to show a small overestimation the vault, with the ST-formula showing a smaller difference to the measured vault compared to the KS-formula. Still the LoA indicate that differences can extend up to 670 µm. Approximately 60% of the eyes had predictions differing less than 250 µm from the measured vault. Both formulas failed to predict extremely low vaults (≤100 µm) but predicted 73% (ST-formula) and 78% (KS-formula) of extremely high vaults. Although vault prediction formulas do not show high precision in determining the exact vault, they are useful in detecting preoperatively high vaults.