Outcomes of two-stage phacoemulsification and pars plana vitrectomy for proliferative diabetic retinopathy with coexisting cataract: a comparative study.

OBJECTIVE: This study aims to compare the outcomes of a two-stage surgical approach in treating proliferative diabetic retinopathy (PDR) coexisting with cataract. METHODS: This retrospective study enrolled 97 eyes. Thirty eyes were assigned to sequential surgery: initial intravitreal pharmacotherapy administered concurrently with phacoemulsification and intraocular lens (IOL) implantation, followed by pars plana vitrectomy (PPV) one week postoperatively. The remaining 67 eyes underwent combined surgery, which entailed initial intravitreal pharmacotherapy followed one week later by simultaneous PPV, phacoemulsification and IOL implantation. The primary outcomes measured included early postoperative anterior chamber (AC) inflammation (degree of inflammatory cells and fibrinous exudate), posterior synechiae of the iris, best-corrected visual acuity (BCVA), intraocular pressure (IOP), and other intraoperative and postoperative complications. RESULTS: Post-vitrectomy day 1, in the sequential group, postoperative findings included AC cells ≥2+ in five eyes, fibrinous exudation in two eyes, and posterior synechiae of the iris in one eye. In the combined group, 28 eyes exhibited inflammatory cells≥2+ postoperatively, nine eyes presented with fibrinous exudation, and seven eyes developed posterior synechiae of the iris. The difference in the incidence of AC cell grade ≥2+ between the two groups was statistically significant (P < 0.05). Preoperative BCVA (LogMAR) in the sequential group was 1.88 ± 1.13, improving to 0.73 ± 0.77 at the final follow-up, showing statistically significant improvement (P < 0.05). In the combined group, preoperative BCVA (LogMAR) was 1.71 ± 1.08, improving to 0.74 ± 0.91 at the final follow-up, also showing significant improvement (P < 0.05). There was no statistically significant difference in postoperative BCVA (LogMAR) between the two groups (P>0.05). Binary logistic regression analysis identified sequential surgery as a significant protective factor against AC inflammation (OR 0.187, 95% CI 0.056-0.621, P = 0.006), indicating a substantially reduced risk of postoperative inflammatory reaction in the staged intervention cohort. CONCLUSION: Both sequential and combined surgical approaches are viable for the treatment of diabetic retinopathy with concomitant cataract. These findings indicate that sequential surgery helps reduce early postoperative inflammatory reactions; however, no significant difference in long-term efficacy has been observed between the two surgical methods. Clinical decision-making should be personalized, taking into account the patient's specific condition and the surgeon's expertise.