Authors: Harrish Nithianandan, Avner Hostovsky, Rajeev Muni, Bernard Hurley, Peter J. Kertes

Author Disclosure Block: H. Nithianandan: None. A. Hostovsky: None. R. Muni: None. B. Hurley: None. P.J. Kertes: None.

Abstract Body:

Purpose: Epiretinal membranes (ERMs) are a common cause of visual decline following the repair of rhegmatogenous retinal detachments (RRDs). The purpose of this study was to determine the risk factors associated with the formation of an ERM following RRD repair.
Study Design: This was a REB-approved multicentre retrospective case series.
Methods: Patients who underwent primary RRD repair at St. Michael’s Hospital (R.M), the University of Ottawa Eye Institute (B.H) or the Sunnybrook Health Sciences Centre (P.K) from January 1, 2016 to December 31, 2016 were included. Patients who underwent either pneumatic retinopexy (PR) or pars plana vitrectomy (PPV) were included. Patients were excluded from the analysis if they had a prior history of any retinal disease or retinal surgery. All data was collected via electronic medical records. Variables of interest included patient sex, age, systemic comorbidities, lens status, macular status at presentation, visual acuity at baseline, RD characteristics and postoperative retinal findings. Odds ratios (OR) and their 95% confidence intervals (CI) were computed using multivariate logistic regression analyses to determine which characteristics were significantly associated with postoperative ERM development. P<0.05 was considered statistically significant.
Results: This study included 214 eyes of 214 patients, whose mean±SD age was 58.0±13.2 years and mean preoperative logMAR visual acuity at presentation was 1.1±0.99 (Snellen: 20/250). The patients presented with macula-off detachment in 52% of cases and had 1.7±1.1 retinal breaks on average. PR was employed in 159 eyes (74%) and PPV in 55 eyes (26%). There were no differences in rates of macula-off detachment (PR: 51% vs. PPV: 54%, p=0.85), number of breaks (p=0.27), or RD size (p=0.50) when comparing eyes treated by PR vs. PPV. The mean length of follow-up across all eyes was 21.6±8.1 months. Postoperative ERM was identified in 73 eyes, of which 19 underwent peeling. The only preoperative characteristic that was associated with ERM formation was the presence of macula-off RD (OR=2.2, 95%CI: 1.3-3.9, p=0.035). RD size (p=0.96) and the number of retinal breaks (p=0.59) were not significant risk factors for ERM formation. Patients were more likely to develop an ERM when treated by PPV vs. PR (OR=2.7, 95% CI 1.3-5.9, p=0.009).
Conclusions: Our results indicate that PPV for primary RRD repair in patients with no prior retinal disease history is associated with greater risk of ERM formation. Larger prospective studies are needed to discern whether RRD repair modality is truly a modifiable risk factor for postoperative ERM formation.