Persistent subfoveal fluid in pneumatic retinopexy versus pars plana vitrectomy for rhegmatogenous retinal detachment: posthoc analysis of the PIVOT randomised trial

Purpose To assess the incidence of persistent subfoveal fluid (PSFF) in pneumatic retinopexy (PnR) versus pars plana vitrectomy (PPV) following rhegmatogenous retinal detachment (RRD) repair and to determine its association with functional outcomes. Methods Posthoc analysis of the PIVOT randomised trial. Eyes with gradable en face and cross-sectional spectral-domain optical coherence tomography (SD-OCT) scans at 1-2 months postoperatively were included. Primary outcome was the proportion of patients with PSFF following PnR versus PPV at 1-2 months postoperatively. Secondary outcomes included association of PSFF with Early Treatment Diabetic Retinopathy Study (ETDRS) letter score at 3, 6 and 12 months and metamorphopsia score (MCHARTs) at 12 months. Results Of 176 participants enrolled in PIVOT, 158 (89.8%) had gradable SD-OCT scans. Intergrader agreement was 0.870 (Cohen's kappa). The incidence of PSFF was 16% (13/81) following PnR and 10.4% (8/77) following PPV (p=0.298; OR=1.65, 95% CI 0.64 to 4.23). Median ETDRS score at 3 months postoperatively between eyes with and without PSFF was 71 (IQR=58-78) and 78 (IQR=70-84), respectively (difference=7 letters, p=0.037), with no significant difference at subsequent 6-month and 12-month visits. Median metamorphopsia scores in patients with versus without PSFF were: horizontal: 0.1 (IQR=0-0.3) vs 0 (IQR=0-0.2) (difference=0.1, p=0.228) and vertical: 0.25 (IQR=0-0.4) vs 0 (IQR=0-0.2) (difference=0.25, p=0.148), respectively. Conclusions There was no significant difference in the incidence of PSFF in eyes undergoing PnR versus PPV for RRD. The presence of PSFF at 1-2 months postoperatively was associated with worse ETDRS letter score at 3 months, but there was no difference at 1 year.