AUTHOR=Li Jing , Xu Qian , Yuan Xiao , Hu Wen 

TITLE=Red blood cell count and its inverse association with diabetic retinopathy: Exploratory development of a risk assessment model in a retrospective cohort

JOURNAL=Frontiers in Endocrinology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2025.1571192

DOI=10.3389/fendo.2025.1571192

ISSN=1664-2392

ABSTRACT=ObjectiveThe purpose of this exploratory study was to investigate the association between red blood cell count (RBC) and diabetic retinopathy (DR) and to develop a preliminary risk assessment framework.MethodsA total of 413 individuals diagnosed with type 2 diabetes mellitus (T2DM) at Suqian First Hospital’s Endocrinology Department were included in this study. These participants were divided into training and validation groups in a 7:3 ratio, consisting of 289 and 124 patients respectively. In the training cohort, potential predictive variables were determined through both univariate and multivariate analyses utilizing forward-backward stepwise selection. Only variables with p < 0.05 were included in the nomogram, which encompassed demographic information, clinical laboratory results, and diabetes-associated complications. The performance of the model was evaluated in both groups using receiver operating characteristic (ROC) curve analysis, the Hosmer-Lemeshow test for calibration, and decision curve analysis (DCA) to determine clinical utility.ResultsOut of 20 clinical variables examined, five were chosen to develop the nomogram: RBC, serum creatinine (SCR), diabetes duration, diabetic peripheral neuropathy (DPN), and diabetic kidney disease (DKD). The ROC analysis revealed that the area under the curve (AUC) for the training cohort was 0.765 (95% CI 0.709-0.821) and for the validation cohort was 0.707 (95% CI 0.616-0.798). Results from the Hosmer-Lemeshow test were p = 0.233 and p = 0.579, indicating a good fit. The nomogram demonstrated excellent predictive accuracy and provides a quantitative tool for assessing the risk of DR in individuals with T2DM.ConclusionOur findings suggest an inverse association between RBC levels and DR risk. The exploratory model incorporating RBC provides an initial framework for evaluating DR risk in patients with T2DM. Further validation in prospective cohorts is needed to refine this framework before considering clinical applications.