Post-Pars Plana Vitrectomy Endophthalmitis in Patients with and without Systemic Immunosuppression

To determine whether systemic immunosuppression is associated with endophthalmitis after pars plana vitrectomy (PPV), and whether risk varies by drug class and time window.

Retrospective propensity score matched cohort study using TriNetX, a federated network of 72 US healthcare organizations. Exposed adults undergoing PPV used corticosteroids, classic nonbiologic immunosuppressants, targeted small molecule immunomodulators, or biologics for at least 3 months before surgery; controls had none in the prior year. Matching controlled for demographics, comorbidities, and surgical indication. The outcome was endophthalmitis within 365 days, assessed over early (1-42 days), 1-90 day, and later (43-365 day) windows. Cox modeling was secondary.

After matching, 25,997 patients per cohort were analyzed. No class was associated with endophthalmitis during the early window. During the later window, risk was higher for all immunosuppressants (RR 1.74; 95% CI 1.33 to 2.28; P < .0001), corticosteroids (RR 1.81; 95% CI 1.38 to 2.37; P < .0001), and biologics (RR 2.00; 95% CI 1.00 to 4.03; P = .045). In Cox regression, corticosteroids (HR 1.19; 95% CI 1.01 to 1.39; P = .037) and biologics (HR 1.51; 95% CI 1.08 to 2.11; P = .016) remained independent risk factors.

Systemic corticosteroid use prior to PPV was associated with late-onset endophthalmitis, with excess risk appearing between 43 and 365 days rather than in the early period. This timing is atypical for surgically acquired infection, and some late events may reflect endogenous or inflammatory disease. Immunosuppressed patients may warrant a longer window of suspicion.