Correlation between redefined optical coherence tomography parameters and best-corrected visual acuity in non-resolving central serous chorioretinopathy treated with half-dose photodynamic therapy

PURPOSE: To assess parameters on optical coherence tomography (OCT), and their correlation with best-corrected visual acuity (BCVA) in patients with non-resolving central serous chorioretinopathy (CSC). METHODS: For 25 non-resolving CSC patients treated with photodynamic therapy (PDT), the thickness of retinal layers was assessed on the foveal spectral-domain (SD) OCT scan. Evaluated OCT parameters included the central retinal thickness (CRT), defined as the internal limiting membrane (ILM) to ellipsoid zone (EZ) distance, and the second band thickness (SBT), defined as the EZ to hyperreflective subretinal accumulation distance. Integrity of the external limiting membrane (ELM) and the EZ bands was also determined. These parameters, along with BCVA and CRT measured automatically by SD-OCT device software were obtained before PDT, after PDT, and at final visit. After Bonferroni correction, a p-value <0.007 was considered statistically significant. RESULTS: Twenty-five patients could be included at last visit before PDT and first visit after PDT. At final visit, 24 patients could be included, since 1 patients was lost to follow-up. Mean CRT was 112 μm at last visit before PDT, 118 μm at first visit after PDT (p = 0.030), and 127 μm at final visit (p<0.001compared to baseline). Mean SBT was 74 μm, 26 μm (p<0.001 compared to baseline), and 21 μm (p<0.001 compared to baseline), respectively. Mean BCVA in Early Treatment of Diabetic Retinopathy Study letters was 79 at baseline, 85 at first visit after PDT (p = 0.005 compared to baseline), and 87 at final visit (p = 0.001 compared to baseline). BCVA had an estimated correlation of β = 0.103 (p = 0.114) with CRT, β = -0.051 (p = 0.014) with SBT, β = 0.615 (p = 0.600) with the integrity of the ELM, and β = 4.917 with the integrity of the EZ (p = 0.001). CONCLUSIONS: In non-resolving CSC patients treated with half-dose PDT, the CRT increased at final visit in comparison to the last visit before PDT. The continuity of the EZ on SD-OCT was positively correlated with BCVA. We propose that the distance between ILM and EZ should be used as a reliable CRT measurement in non-resolving CSC patients treated with half-dose PDT.