TOPIC: To examine the potential role of systemic neurodegeneration, this study aimed to provide an overview of the available evidence on the relationship between diabetic retinopathy (DR) and systemic neurodegeneration.
CLINICAL RELEVANCE: The association between DR and systemic neurodegeneration is inconsistent in the literature. A summary estimate on the measures of association is important to establish whether DR may be used as a risk marker of systemic neurodegeneration.
METHODS: We searched the literature databases PubMed/MEDLINE, EMBASE, and Cochrane Library on October 3, 2020, for all observational studies on humans evaluating the association between DR and systemic neurodegenerative diseases. Two authors conducted the literature search, study selection, and data extraction in an independent fashion. Studies were reviewed qualitatively in text and quantitatively in meta-analyses. Heterogeneity was evaluated with Cochran's Q and I 2, and Funnel plot was used to investigate for skewed results and possible publication bias.
RESULTS: We identified 27 eligible studies with a total of 1 398 041 patients with diabetes. Diagnosis of DR was made using fundus photography or examination (n = 20), health registries (n = 4), was self-reported (n = 1), or was not disclosed in the remaining studies. Neurodegenerative conditions studied were cognitive impairment (n = 23), Alzheimer's disease (n = 3), and Parkinson's disease (n = 1). In cross-sectional and longitudinal studies, respectively, presence of any DR was associated with present (odds ratio [OR], 1.57; 95% confidence interval [CI], 1.02-2.43, P = 0.043) and incident (OR, 2.36; 95% CI, 1.50-3.71, P = 0.00021) systemic neurodegeneration, but severity of DR was not associated with differences in systemic neurodegeneration (OR, 0.98; 95% CI, 0.45-2.15, P = 0.96).
CONCLUSIONS: In this systematic review, DR appears to be a marker of systemic neurodegeneration. Further studies are warranted to better elucidate the clinical practical implications of this relationship.