Pars Plana Vitrectomy for Retained Lens Fragments: Visual Outcomes and Complications in the IRIS Registry

Purpose: To evaluate visual acuity (VA) outcomes and postoperative complications after pars plana vitrectomy (PPV) for retained lens fragments (RLF) after cataract surgery. Methods: This retrospective cohort study used data from the IRIS (Intelligent Research in Sight) Registry from January 2013 to December 2022. Eyes that underwent PPV for RLF within 8 weeks of cataract surgery were identified using International Classification of Diseases, 9th and 10th editions, and Current Procedural Terminology codes. Eyes with preexisting ocular comorbidities were excluded. Multivariable regression identified predictors of good (VA ≥20/40) and poor (VA ≤20/200) vision at 3 months. Main outcomes included postoperative VA and complications within 3 months of cataract surgery. Results: A total of 5640 eyes from 5509 patients were analyzed. The mean logMAR VA improved from 0.63 ± 0.51 before cataract surgery to 0.46 ± 0.50 after PPV. PPV was performed the same day in 48.3% of eyes, within 1 week in 25.7%, and after 1 week in 26.0%. Common complications included ocular hypertension (17.1%), corneal edema (7.4%), vitreous hemorrhage (6.3%), and rhegmatogenous retinal detachment (RRD [3.5%]). Same-day PPV was associated with significantly lower rates of anterior uveitis, endophthalmitis, corneal edema, intraocular lens (IOL) displacement, and ocular hypertension (all P < .001). IOL implantation before PPV was an independent predictor of good vision (relative risk, 1.38; 95% CI, 1.21-1.56). Poor outcomes were associated with worse preoperative VA, Black race, older age, delayed PPV (>1 week), corneal edema, and RRD (all P < .05). Conclusions: Same-day PPV significantly reduces postoperative complications, and IOL implantation before PPV improves visual recovery. Delayed PPV, poor baseline vision, older age, and postoperative complications are predictive of worse outcomes. Early surgical intervention and individualized planning are critical for optimizing visual results in RLF management.