Correlation analysis between the deep-superficial flow-density ratio and treatment response of macular edema secondary to branch retinal vein occlusion / 中华眼底病杂志

ABSTRACT

Objective:To observe the correlation analysis between the deep-superficial flow-density ratio (DSFR) and treatment response of macular edema secondary to branch retinal vein occlusion (BRVO).Methods:Forty-eight patients (48 eyes) with macular edema secondary to BRVO from December 2018 to December 2019 in the Department of Ophthalmology of Beijing Hospital were enrolled in this study. There were 29 males (29 eyes) and 19 females (19 eyes), with the mean age of 58.77±10.88 years. All eyes were treated with intravitreal injection of ranibizuma once a month for 3 months, and then treated as needed. According to the central retinal thickness (CRT) 12 months after treatment, the patients were divided into good response group (CRT≤250 μm) and refractory group (CRT> 250 μm). The flow density in the superficial capillary plexus (SCP) and deep capillary plexus (DCP) of all subjects was measured by optical coherence tomography angiography. The flow density of DCP and SCP measured at 3 follow-up times was selected and DSFR was calculated. The DSFR was recorded by the Study for the Treatment of Diabetic Retinopathy (ETDRS) -grid and Nine-grid. The flow density of DCP, SCP and DSFR were compared between the two groups by paired t test. At 3 months post-treatment, the efficacy of DSFR in ME treatment response was evaluated according to area under curve (AUC) of receiver operating characteristic. Univariate and multivariate binary logistic regression were used to analyze the factors affecting the response to ME treatment. Results:At 12 months after treatment, there were 27 eyes in good response group and 21 eyes in refractory group. There was no statistical significance in the flow density of DCP ( t=1.804, 1.064, 0.660) and SCP ( t=0.581, 0.641, 0.167) and DSFR ( t=0.393、-0.553、0.474) in all area of response group and refractory group using ETDRS-GRID recording method ( P>0.05). The SCP, DCP and DSFR of the most severe non-perfusion area were (27.10±5.70) %, (28.33±8.95) %, 1.35±0.54 and (27.54±6.70) %, (29.11±0.42) %, 1.01±0.40 in the response group and refractory group, respectively. There was no significant difference in the flow density of DCP and SCP between the two groups ( t=-0.237,-0.340; P>0.05). The difference of DSFR between two groups was statistically significant ( t=2.288, P=0.024). Univariate and multivariate binary logistic regression analysis showed that DSFR in the most severe non-perfusion area was associated with ME response (odds ratio=0.212, 0.085; P=0.027, 0.024). The AUC was used to evaluate the efficacy of DSFR in ME treatment response, the results showed that the AUC was 0.800, P=0.001, Youden index was 1.348, sensitivity was 67.7%, and specificity was 86.7%. Conclusions:DSFR reduction is more common in BRVO secondary to ME patients. DSFR correlates with ME treatment response.