ICL Sizing and Vault Prediction: A Systematic Review of Mathematical Models and Calculation Formulas.

PURPOSE: To systematically identify and analyse published formulas and nomograms for implantable collamer lens (ICL) sizing and vault prediction, evaluating their input parameters and reported performance. METHODS: A systematic review was conducted including studies that developed nomograms for ICL sizing or vault prediction. PubMed and Scopus databases were searched, and the references of included studies were screened. The search period ranged from January 2010 to July 2025. The review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Risk of bias was assessed using the Prediction model Risk of Bias Assessment Tool (PROBAST). RESULTS: A total of 675 records were identified, of which 38 studies were included. Twenty studies developed AS-OCT-based nomograms (11,177 eyes from 7,085 patients), and 18 used UBM (21,767 eyes from 15,393 patients). No significant differences were found between AS-OCT and UBM nomograms in predictive performance for R2 (p = .169, 95% CI: -0.155 to 0.372) or MAE (p = .159, 95% CI: -152.36 to 203.80 µm). This finding should be interpreted cautiously given substantial heterogeneity in populations, devices, and validation strategies across studies, and does not imply true clinical equivalence between the two modalities. CONCLUSION: Multiple nomograms and machine learning models are available for ICL sizing and vault prediction; but no single approach performed universally better across populations, devices, and clinical settings. Because measurement conditions, imaging platforms, institutional protocols, and surgeon-specific factors can all influence model performance, any chosen formula should be validated locally before routine use, and combining multiple validated nomograms may support safer clinical decision-making.