Abstract
PURPOSE: To evaluate the relationship between choroidal thickness (CT) and anterior scleral thickness (AST) in patients with subclinical keratoconus (SKC) and established keratoconus (KC). METHODS: This single-center prospective case-control study included 97 eyes of 97 patients: 44 KC eyes, 14 SKC eyes, and 39 age- and axial length (AL)-matched healthy eyes. Using swept-source optical coherence tomography, the AST was manually measured in four directions and the CT was obtained automatically from the Early Treatment Diabetic Retinopathy Study (ETDRS) grid. Principal component analysis (PCA) was used to linearly reduce the dimensionality of nine CT inputs to one significant component, CT1. A multivariate model was created to evaluate the association between CT1, AST, and several ocular parameters in SKC and KC patients. Partial correlation was then performed to adjust the confounding factors and to examine the effect of AST on CT1. RESULTS: The PCA showed that CT1 accounts for 86.54% of the total variance in the nine original CTs of the ETDRS grid. The CT1 model was associated with age, AL, and AST in the superior meridian in SKC eyes, whereas in KC eyes, it was correlated with gender, age, AL, and AST in the inferior meridian (p < 0.001). The partial correlation between CT1 and AST in the superior zone was found to be significant, positive, and strong in SKC eyes (r = 0.79, p = 0.019), whereas a significant, positive, and moderate correlation between CT1 and AST at the inferior zone (r = 0.41, p = 0.017) was observed in KC eyes. CONCLUSIONS: Choroidal tissue was significantly correlated with the anterior sclera across the vertical meridian. This relationship was observed over the superior sclera in SKC eyes, whereas in established KC, it was over the inferior sclera. These results reveal new insights regarding the interactions between the anterior and posterior structures of the KC eyes and confirm the enigma of the pathophysiology of KC.