Ana Piñar-Gutiérrez, José María Madruga-Rubio, Noelia Gros-Herguido, Sandra Amuedo, Gema López-Gallardo, Carmen Amelia Ruiz-Trillo, Fernando López Herrero, María Loreto Saez Ortega, Ana Muñoz Morales, Virginia Bellido, Alfonso Soto-Moreno
In this real-world screening-based cohort of adults with T1D, mild NPDR demonstrated a highly dynamic course, with more than half of patients experiencing regression over 5 years. Overall glycemic control, reflected by HbA1c, was the primary determinant of regression. CGM-derived metrics did not provide additional predictive value beyond HbA1c in this setting. These findings highlight the potential reversibility of early diabetic retinopathy under contemporary care and underscore the importance of intensive long-term glycemic management.
BACKGROUND: Contemporary data on the natural history of mild non-proliferative diabetic retinopathy (NPDR) in adults with type 1 diabetes managed under intensive therapy and continuous glucose monitoring (CGM) are limited. We aimed to characterize the long-term evolution of mild NPDR in this modern care setting and to explore clinical and glycemic factors associated with regression or progression, including CGM-derived metrics.
METHODS: We conducted a retrospective cohort study of adults with type 1 diabetes and mild NPDR identified by digital fundus photography within a screening program between 2018 and 2020. Participants underwent follow-up retinal evaluation after approximately 5 years. Retinopathy outcomes were classified as regression (absence of retinopathy), stability (persistent mild NPDR), or progression (moderate/severe NPDR and/or diabetic macular edema [DME]). Clinical, biochemical, and glycemic variables, including HbA1c and CGM-derived metrics, were collected. Multivariable logistic regression identified factors independently associated with regression of mild NPDR.
RESULTS: Among 113 participants (median age 39 years; median diabetes duration 21 years), regression occurred in 57.5%, 31.0% remained stable, and 11.5% progressed. Lower HbA1c at follow-up was observed in participants with regression compared with those without. Several CGM-derived metrics, including time in range and glucose management indicator, were associated with regression in univariate analyses; however, these associations were not independent of HbA1c due to strong collinearity. In multivariable analysis, lower HbA1c was the only independent predictor of regression.
CONCLUSIONS: In this real-world screening-based cohort of adults with T1D, mild NPDR demonstrated a highly dynamic course, with more than half of patients experiencing regression over 5 years. Overall glycemic control, reflected by HbA1c, was the primary determinant of regression. CGM-derived metrics did not provide additional predictive value beyond HbA1c in this setting. These findings highlight the potential reversibility of early diabetic retinopathy under contemporary care and underscore the importance of intensive long-term glycemic management.