Long‑Term Visual Outcomes of Cataract Surgery in Pediatric Patients

Purpose To report long-term visual and refractive outcomes following pediatric cataract surgery over a 22-year period at a tertiary referral center and to explore associations with clinical and surgical factors after accounting for within-child clustering. Design Retrospective cohort Subjects Consecutive patients ≤18 years who underwent primary cataract extraction (January 2000-December 2021). Methods Extracted data included demographics, laterality, IOL status, etiology, cataract morphology, primary posterior capsulotomy/anterior vitrectomy status, recorded complications/events, preoperative biometry/cycloplegic refraction, follow-up duration, and final recorded visual and refractive outcomes. Final logMAR acuity was recalculated from recorded Snellen fractions. Final BCVA was analyzed using a mixed-effects linear model with a random intercept for child and adjustment for age at surgery, laterality, IOL status, follow-up duration, etiology, and primary posterior capsulotomy/anterior vitrectomy; BCVA ≤0.30 logMAR was analyzed with a GEE logistic model using the same covariates and robust cluster standard errors. Better-seeing-eye, traumatic-cataract-exclusion, and multiple-imputation sensitivity analyses were performed. Results Forty-four children (61 eyes) met inclusion criteria (mean age at surgery per operated eye 4.45 ± 4.58 years; range 0.1-15.6). Seventeen children (38.6%) contributed both eyes. Primary IOL implantation was performed in 42 eyes (68.9%), and primary posterior capsulotomy with anterior vitrectomy was performed in 35 eyes (57.4%). Mean follow-up was 53.3 ± 37.2 months; median 48 months (range 2-115). Quantitative final BCVA was available for 47/61 eyes (77.0%); among measured eyes, mean final BCVA was 0.41 ± 0.37 logMAR and 24/47 (51.1%) achieved ≤0.30 logMAR. Final spherical equivalent was available for 45/61 eyes and averaged 1.88 ± 5.60 D overall, 0.11 D in 30 pseudophakic eyes, and 5.43 D in 15 aphakic eyes. In the adjusted mixed-effects model, older age at surgery was directionally associated with better BCVA (beta -0.045, 95% CI -0.092 to 0.002; p=0.060), while the adjusted GEE showed higher odds of BCVA ≤0.30 logMAR with older age (OR 1.37 per year, 95% CI 1.05 to 1.78; p=0.019). PPC/anterior vitrectomy was associated with higher logMAR in the continuous model (beta 0.162, 95% CI 0.001 to 0.323; p=0.049) but not with the binary BCVA endpoint (OR 0.24, 95% CI 0.04 to 1.42; p=0.1). Conclusions Many children achieved useful long-term vision after pediatric cataract surgery, and pseudophakic eyes were closer to emmetropia at last follow-up. After accounting for within-child clustering and the available covariates, the age-outcome association was attenuated and primary pseudophakia was not independently associated with better final BCVA. These findings should therefore be interpreted as exploratory and susceptible to residual confounding by phenotype, severity, and surgical era.