Predicting early postoperative corneal edema after phacoemulsification: A retrospective observational study.

This study aimed to develop and internally validate a multivariable prediction model for early postoperative corneal edema following phacoemulsification by integrating patient-related and intraoperative parameters. This retrospective cohort study included 912 consecutive eyes from 912 patients who underwent standard phacoemulsification at a tertiary ophthalmic center in Southwest China between January 2021 and January 2023. Early postoperative corneal edema was defined as either a ≥5% increase in central corneal thickness from baseline on postoperative day 1 or the presence of clinically significant stromal haze on slit-lamp examination. Demographic, systemic, preoperative, and intraoperative variables were evaluated using univariate logistic regression, followed by multivariable logistic regression for model development. Model performance was assessed in terms of discrimination and calibration. Discrimination was quantified using the area under the receiver operating characteristic curve, whereas calibration was evaluated using calibration plots, calibration intercept and slope, and the Hosmer-Lemeshow goodness-of-fit test. A nomogram was constructed based on the final multivariable model to estimate individualized risk. Early postoperative corneal edema occurred in 138 of 912 eyes (15.1%). In the multivariable model, older age, diabetes mellitus, lower preoperative endothelial cell density, and greater cataract and intraoperative complexity, reflected by higher lens nucleus hardness, longer operative duration, higher cumulative dissipated energy, and greater irrigation volume, were associated with early postoperative corneal edema. Intraoperative anterior chamber instability and capsular rupture or vitreous loss were also associated with an increased risk. The final model achieved an area under the receiver operating characteristic curve of 0.892 (95% confidence interval: 0.865-0.919) and demonstrated acceptable calibration in internal validation. Both systemic and intraoperative factors were associated with early postoperative corneal edema after phacoemulsification. The internally validated prediction model demonstrated good discrimination and calibration and may support preliminary perioperative risk estimation in similar clinical settings. However, external validation is required before the nomogram can be recommended for routine clinical implementation.