Pars plana vitrectomy, glycemic control, and systemic adverse events in patients with diabetes mellitus: a retrospective cohort study

This is the authors' abstract. We don't add key points for this paper.

Background The relationship between hemoglobin A1c (HbA1c), pars plana vitrectomy (PPV), and systemic health risks remains incompletely understood. We aimed to determine whether undergoing PPV is associated with increased risk of systemic adverse events in diabetic patients (Analysis 1) and whether higher HbA1c is associated with increased risk of systemic events among PPV patients (Analysis 2). Methods We conducted a retrospective cohort study using the TriNetX US Collaborative Network. In Analysis 1, we stratified adults with diabetes mellitus who underwent PPV by preoperative HbA1c (<8.0 or ≥8.0) recorded within six months before PPV and propensity score matched them to diabetic patients who attended an outpatient eye care visit and did not undergo PPV. We estimated 30- and 60-day hazard ratios (HRs) for mortality, stroke or transient ischemic attack, major cardiovascular event, hospitalization, and a composite outcome. In Analysis 2, we analyzed all diabetic PPV patients as a single cohort using Cox proportional hazards models to estimate the association between HbA1c, modeled continuously, and each systemic outcome. Results In Analysis 1, the PPV and outpatient visit cohorts each included 5,029 patients with HbA1c ≥ 8.0 and 10,518 patients with HbA1c < 8.0. After matching, no statistically significant increase in any selected 30- or 60-day outcome across HbA1c strata was detected among patients selected and cleared to undergo PPV. For the composite outcome, 30-day HRs were 1.00 (95% CI 0.65, 1.55) among patients with HbA1c ≥ 8.0 and 0.76 (95% CI 0.56, 1.02) for patients with HbA1c < 8.0. In Analysis 2, among 15,622 patients, HbA1c was not associated with a statistically significant increase in risk of any systemic outcome; HRs ranged from 0.96 to 1.07 at 30 days and from 0.99 to 1.02 at 60 days. Conclusions In this study, after matching, no statistically significant increase in selected short-term systemic outcomes was detected among diabetic patients selected to undergo PPV, including those with elevated HbA1c. Similarly, higher preoperative HbA1c was not associated with a statistically significant increase in risk of the selected systemic outcomes following PPV. Preoperative HbA1c alone may have limited utility for identifying patients at increased short-term risk following PPV, while comprehensive clinical assessment remains necessary.