To comprehensively evaluate baseline characteristics of patients with central serous chorioretinopathy (CSC) and develop predictive risk scores to identify visual prognosis. This single-institute, retrospective cohort study included 144 eyes of 144 patients with CSC who underwent photodynamic therapy and achieved serous retinal detachment resolution. We developed and assessed the performance of several risk scores for best-corrected visual acuity (BCVA) outcomes six months post-treatment: i) BCVA improvement (≤-1.0 logMAR), and ii) BCVA deterioration (≥+ 1.0 logMAR). The BCVA improvement models used photoreceptor outer segment thickness, loss of photoreceptor outer segment, and neurosensory retinal thickness (NSRT), while the BCVA deterioration models included outer nuclear layer thickness and NSRT. The BCVA improvement models demonstrated a corrected area under the curve (AUC) of 0.786 (95% confidence interval [CI]: 0.699-0.864), with 80.4% sensitivity, and 71.2% specificity. The BCVA deterioration models achieved a corrected AUC of 0.864 (95% CI: 0.742-0.958), with 85.7% sensitivity, and 83.5% specificity. The predictive models for CSC exhibited favorable performance in predicting individual visual prognoses. A thinner outer nuclear layer may be associated with BCVA deterioration, whereas preservation of the photoreceptor outer segment may be correlated with BCVA improvement. WHAT IS KNOWN : Pre-treatment best-corrected visual acuity, thickness of each sensory retinal layer, time fromonset to treatment, and macular atrophy were each found to be associated with visualprognosis for patients with central serous chorioretinopathy (CSC). The current study comprehensively assessed potential prognostic factors and precisely identified individual likelihood of visual prognosis. The study found that different regions of the sensory retina were associated with either worsening or improving visual acuity. Accurately predicting visual outcomes after photodynamic therapy for CSC would help healthcare providers create personalized treatment plans and enable patients to make informed decisions about their treatment based on their expected visual results.
Read full abstract