[Logistic regression analysis of risk factors for subretinal fluid after rhegmatogenous retinal detachment repair]

Objective: To identify factors associated with persistent subretinal fluid (SRF) after scleral buckling and to evaluate surgical outcomes of patients with primary rhegmatogenous retinal detachment (RRD). Methods: Cross-sectional study and cohort study. This study included 104 patients (104 eyes) who underwent scleral buckling for repair of RRD between January 2016 and June 2017. Several statistically significant risk factors associated with SRF were screened out with univariate analysis. Then independent risk factors were determined with multivariate stepwise logistic regression analysis. Examinations were taken preoperatively, at 1, 3, 6, and 12 months postoperatively, and thereafter every six months. Patients were divided into two groups depending on the presence or absence of persistent SRF on optical coherence tomography at 1 month. Results: Persistent SRF occurred in 69.2% (72/104) of patients. In multivariate analysis, younger age (OR=32.860, 95%CI=3.636-296.986, P=0.002), high myopia (OR=7.229, 95%CI=1.217-42.94, P=0.03) and macula-involving retinal detachment (OR=357.397, 95%CI=29.761-4292.0, P=0.000) were associated with persistent SRF. Best corrected visual acuity in patients with SRF at 1 month (0.71±0.18 vs. 0.58±0.11; t=4.047, P<0.01) and 3 months (0.55±0.15 vs. 0.43±0.12; t=3.914, P<0.01) was worse than that in patients with absence of SRF. Best corrected visual acuity in patients with SRF was 0.44±0.16 at 6 months, 0.37±0.12 at 12 months and 0.36±0.10 at the last follow-up, with no significant difference from patients without SRF at the three time points (0.39±0.13, 0.38±0.12 and 0.35±0.09; t=1.643, -0.202, 0.464; P>0.05). Conclusions: RRD patients with younger age, high myopia and macular involvement were more likely to develop persistent SRF after scleral buckling. The presence of persistent SRF slowed visual recovery but did not influence the final visual outcomes. (Chin J Ophthalmol, 2021, 57: 426-432).