Related Experiment Videos
Influence of the Corneal Posterior-to-Anterior Radius Ratio on IOL Power Prediction Using Standard and Total
Zhongxiu Yang1, Fu-Man Yang2, Zi-Qi Meng1
1National Clinical Research Center for Ocular Diseases, Eye Hospital, Wenzhou Medical University, Wenzhou, 325027, Zhejiang Province, China; State Key Laboratory of Eye Health, Eye Hospital, Wenzhou Medical University, Wenzhou, 325027, Zhejiang Province, China; Eye Hospital of Wenzhou Medical University at Hangzhou, Hangzhou, 310000, Zhejiang Province, China.
Purpose:
To compare the predictive accuracy of standard keratometry (K) and total keratometry (TK) for intraocular lens (IOL) power calculation across eyes stratified by the posterior-to-anterior corneal radius ratio (PAR).
Design:
Retrospective evaluation of IOL power calculation accuracy.
Methods:
A total of 1,314 eyes of 1,314 patients who underwent uneventful phacoemulsification with monofocal IOL implantation were enrolled. Preoperative biometry was performed with the IOLMaster 700 (Carl Zeiss Meditec). Eyes were divided into three subgroups according to the PAR: low (<25th percentile), moderate (25th-75th percentiles), and high (>75th percentile). Refractive prediction errors (PEs) were calculated with the Barrett Universal II (BUII) and Kane formulas, using both K and TK values. Main outcome measures were PE, median absolute error (MedAE), and the proportions of eyes within predefined absolute prediction error thresholds.
Results:
PAR was independently associated with the discrepancy between K- and TK-based prediction errors in both the BUII and Kane formulas (both P<0.001). Compared with conventional K, TK produced a consistent myopic shift in eyes with low PAR (-0.10 D for both formulas; both P<0.001) and a hyperopic shift in eyes with high PAR (+0.14 D for both formulas; both P<0.001), whereas only minimal changes were observed in eyes with moderate PAR. Despite these systematic directional shifts, MedAE remained comparable between K- and TK-based calculations across all PAR subgroups. In the high PAR subgroup, incorporation of TK significantly reduced the proportion of eyes within ±0.50 D of target refraction for the BUII formula (79.9% vs. 75.4%, P=0.041), whereas no significant difference was observed for the Kane formula.
Conclusions:
PAR is an independent factor associated with the discrepancy between K- and TK-based refractive predictions. Rather than uniformly improving refractive accuracy, TK primarily induces systematic, PAR-dependent directional refractive shifts. These findings suggest that PAR may serve as a clinically useful biomarker for selecting between K- and TK-based calculations instead of supporting the routine use of TK in all eyes.
