Model Predicts Risk of Clinically Significant Intraocular Lens Tilt, Decentration

Risk models incorporating preoperative crystalline lens tilt and decentration as well as axial length show good performance and may be useful tools for predicting clinically significant intraocular lens (IOL) tilt and decentration in patients with age-related cataract undergoing phacoemulsification combined with IOL implantation, according to research published in Journal of Cataract & Refractive Surgery.

Researchers conducted a single-center, prospective cohort study to explore the risk factors and construct models to predict the risks of IOL tilt and decentration after cataract surgery. 

The team measured the tilt and decentration of patients’ crystalline lens and IOL before and 3 months after surgery, used logistic regression analyses to determine the risk factors of clinically significant IOL tilt and decentration, and constructed prediction models based on the results of the multivariate logistic regression analysis.

A total of 207 eyes of 207 patients (mean age, 67.23±8.05 years; 55% women and 45% men) were included in the study. At 3 months after cataract surgery, 11.59% and 7.73% of eyes had clinically significant IOL tilt and decentration, respectively. 

With multivariate logistic regression analysis, the investigators demonstrated preoperative crystalline lens tilt was a risk factor for clinically significant IOL tilt (OR, 3.519; P <.001) and preoperative crystalline decentration (OR, 410.22; P =.001) and axial length were risk factors for clinically significant IOL decentration (OR, 2.155; P =.019). 

Their risk models showed good calibration and discrimination for prediction of clinically significant IOL tilt (area under the receiver-operating characteristic curve [AUC], 0.833; cut-off value, 6.5; sensitivity, 62.5%; specificity, 93.48%) and decentration (AUC, 0.757; cut-off value, 0.08; sensitivity, 81.25%; specificity, 73.82%).

“It is necessary to measure crystalline lens tilt and decentration before cataract surgery for patients applying multifocal and toric IOLs,” according to the researchers. “[O]ur findings suggested that our models may be useful risk-prediction tools for postoperative IOL tilt and decentration.”

Limitations of the study included limitation to only a single type of IOL and the relatively short follow-up period.