Association Between Blood Urea Nitrogen/Albumin and the Incidence as Well as Progression of Type 2 Diabetes

Background: An increased risk of multiple secondary diseases has been observed in individuals with diabetes, which contributes to the growing economic burden. Few studies have established the connection of blood urea nitrogen/albumin (BAR) with diabetes, and its link to subsequent diabetic complications and mortality remains unclear. We aimed to explore the association of BAR with the onset of type 2 diabetes mellitus (T2DM) and its dynamic progression. Methods: A total of 20,039 prediabetic participants aged 40–70, without diabetes or related events at baseline, were selected from the UK Biobank. We used a multistate regression model to assess the relationships between BAR and the trajectories of T2DM progression—from pre-T2DM to T2DM, complications, and ultimately mortality. Akaike information criterion (AIC), area under the curve (AUC), and C-statistic analyses were performed to compare the diagnostic performance of BAR with that of HbA1c for assessing T2DM progressions risk. Results: Over a mean 13-year follow-up, 5621 incident T2DM cases were identified, and among them, 1295 developed diabetes macrovascular complications, 574 developed diabetes microvascular complications, and 1264 died. BAR was significantly associated with the increased risk of T2DM (HR: 1.05, 95% CI: 1.02, 1.09), subsequent macrovascular complications (HR: 1.18, 95% CI: 1.12, 1.25), microvascular complications (HR 1.18, 95% CI: 1.08, 1.28), and further death (HR 1.18, 95% CI: 1.00, 1.39). The ability of BAR (AUC: 0.938, C-statistic: 0.739, p < 0.001) to predict diabetes progression was inferior to that of hemoglobin A1c (HbA1c) (AUC: 0.949, C-statistic: 0.809, p < 0.001). Conclusions: Although BAR is significantly positively correlated with the risk of progression at different stages of diabetes, its predictive performance is inferior to that of HbA1c and comparable to that of BUN and albumin. BAR does not demonstrate sufficient clinical significance in predicting diabetes progression, highlighting the need for further investigation.