A Comparison of Vitrectomy with and without Intravitreal Methotrexate for Advanced Proliferative Vitreoretinopathy

Abstract Objectives: To determine the anatomical visual outcomes and safety of intravitreal (IV) injection of self-compounded methotrexate (MTX) to treat proliferative vitreoretinopathy (PVR), since compounding pharmacies are limited or unavailable locally. Materials and Methods: This was a retrospective comparative, interventional study. Data was obtained from case records and surgical logs. IV MTX compounded in the operating room into aliquots was injected into the vitreous cavity containing silicone oil of eyes after vitrectomy for rhegmatogenous retinal detachment (RRD) complicated with advanced PVR-labelled with methotrexate PVR (WMPVR). Weekly IV MTX at 500 µg/0.1 mL was administered in a fixed dosing regimen in Nigerian PVR patients. The anatomical and visual outcome in WMPVR was compared with two groups matched for age and sex: PVR in which MTX was not used, labelled without MTX PVR (WOMPVR), and retinal detachment without PVR, labelled no PVR (NPVR). The primary outcome was the percentage of retina attachment after a minimum of 6 months follow-up, and the secondary outcome was the change in Snellen best corrected visual acuity conversion to LogMAR. Documented observed complications assessed the safety of IV MTX. Data analysis was done using Statistical Package for the Social Sciences version 22. Means were used for measuring continuous variables and percentages for discrete variables. The three study groups and the two PVR groups were compared. Kruskal–Wallis and Mann–Whitney U were used to calculate the relationship between variables. Results: A total of 90 RRD eyes were studied, 30 in each group. The number of detached quadrants, PVR grade, and macular involvement were similar in WMPVR and WOMPVR ( P > 0.05, 95% CI), but NPVR had fewer detached quadrants ( P < 0.05, 95% CI). Total IV MTX injections was 84, average 2.8 (range: 1–7). Anatomical reattachment was high, >93% in the three groups ( P = 0.81). Preoperative LogMAR vision in WMPVR and WOMPVR was 2.14/0.60 (20/2520) and 1.93/0.72 (20/1600) ( P = 0.13), and 1.30/0.83 (20/400) for NPVR ( P = 0.00). Postoperative LogMAR vision was 1.58/0.83 (20/800) for WMPVR, 1.48/0.95 (20/630) ( P = 0.92) for WOMPVR, and 0.79/0.91 (20/125) ( P = 0.002) for NPVR. Subgroup analysis of silicone oil removed eyes in all three groups showed similar anatomical outcomes ( P = 0.797); postoperative vision was superior in the WOMPVR and NPVR groups compared to WMPVR ( P = 0.001). Intraocular pressure in WMPVR increased from preoperative 5 mm Hg to postoperative 13 mm Hg. The three eyes that had keratopathy resolved with copious lubrication. There was no endophthalmitis. Conclusion: Preliminary findings demonstrate the safety of self-compounded IV MTX in Nigerians. There was no significant visual improvement despite favourable anatomical outcomes. In this research, a non-PVR group as a standard for comparison of the two PVR groups demonstrates the effectiveness of MTX for PVR by achieving equivalent anatomical outcomes with a group that had no PVR, which had a more favourable prognosis.