Sociodemographic predictors of acute corneal hydrops in patients with unstable keratoconus

Abstract

Objective To determine whether sociodemographic factors affect the outcomes of patients with unstable keratoconus. Design Retrospective case control study. Subjects All patients diagnosed with unstable keratoconus at Mass Eye and Ear between January 2016 and October 2022. Those who experienced acute corneal hydrops during the study period were considered cases, and those who did not were considered controls. Methods Potential subjects were identified by billing code and verified with review of clinical documentation. Charts were reviewed for pertinent sociodemographic, medical, and ocular risk factors. Main outcome measures Clinical diagnosis of acute corneal hydrops was the primary outcome. Receipt of corneal crosslinking (CXL) was a secondary outcome. Results Of 762 patients diagnosed with unstable keratoconus, we identified 128 episodes of hydrops in 113 patients (14.9%). Compared to controls, cases were a significantly higher proportion Black (27.6% vs 14.6%, P  < 0.0001), unemployed (22.4% vs 12.5%, P  < 0.01), and/or covered by public health insurance plans (50.4% vs 27.7%, P  < 0.0001). Body mass index (BMI) was significantly higher in cases (34.1 ± 10.0) than controls (29.2 ± 8.1, mean±SD, P  < 0.0001). Developmental delay (17.7% vs 3.4%, P  < 0.0001), atopy (48.9% vs 30.4%, P  = 0.0001), sleep apnea (9.7% vs 1.2%, P  < 0.0001), and eye rubbing (85.5% vs 67.7%, P  < 0.01) were all more common in the hydrops group. In a stepwise multivariable logistic regression analysis, Black race [2.10 (1.02–4.27), P  = 0.041], public health insurance coverage [2.06 (1.12–3.80), P  = 0.020], BMI [1.05 (1.02–1.08), P  = 0.003], and developmental delay [16.42 (5.62–56.30), P  < 0.001] were each associated with a higher odds of incident hydrops [OR (95%CI)]. Similar trends were found to influence the receipt of CXL, with male sex [0.34 (0.13–0.77), P  = 0.015], Black race [0.39 (0.16–0.97), P  = 0.038], and public health insurance [0.37 (0.16–0.82), P  = 0.014] negatively influencing the odds of receiving CXL. Conclusions This study establishes several sociodemographic (Black race, unemployment, and publicly insured) and medical (high BMI and developmental delay) factors as independent risk factors influencing the incidence of hydrops and/or the receipt of CXL in a large group of patients with unstable keratoconus. Special attention is warranted when caring for patients within these subpopulations with unstable keratoconus, and further study into the mechanisms of these associations is warranted.