The association of retinal age gap with schizophrenia: a cross-sectional analysis

Schizophrenia, a chronic neuropsychiatric disorder increasingly recognized as a multisystemic disease, is associated with accelerated brain ageing. Using deep learning, we investigated the retina, as a window into the central nervous system, as a surrogate of biological ageing in individuals with schizophrenia. This cross-sectional study was nested within AlzEye, a retrospective cohort of individuals aged ≥40 years attending Moorfields Eye Hospital (2008-2018). Retinal age was predicted from color retinal photographs using a convolutional neural network. The difference between predicted retinal age and chronological age, termed the retinal age gap, was estimated in all individuals. Associations between schizophrenia and retinal age gap were assessed using adjusted linear mixed-effects models. From a cohort of 98,629, 214 individuals had schizophrenia. They were slightly younger than unaffected (61.6 +/- 12.1 vs 64.5 +/- 13.5 years) and more likely to have hypertension (82% vs 47%) and diabetes mellitus (72% vs 27%). Individuals with schizophrenia had a significantly greater retinal age gap (0.76 years, 95% CI: 0.03, 1.49, p=0.04) when adjusting for age, sex, ethnicity, and socioeconomic status. When adjusting for hypertension and diabetes mellitus, there was no significant difference in retinal age gap between groups (0.20, 95% CI: -0.53, 0.93). In this ethnically and socioeconomically diverse urban population, individuals with schizophrenia had an increased retinal age gap. This was attributable to the increased prevalence of hypertension and diabetes mellitus implicating medical comorbidity, which is modifiable, as the driver of accelerated central nervous system ageing.