Indexing metabolic vulnerability stratifies risks of vascular complications of type 2 diabetes

Background: Vascular complications significantly harm long-term prognosis in type 2 diabetes mellitus (T2DM) patients yet current risk prediction tools remain insufficient. The Metabolic Vulnerability Index (MVX) is an NMR-derived metabolomic score capturing inflammation and nutrition–metabolic vulnerability (including signals from BCAAs, citrate, GlycA, and small HDL particles [S-HDLP]). We examined sex-specific associations of MVX with incident diabetic vascular complications beyond established risk factors. Methods: In the UK Biobank, we analyzed T2DM participants with NMR plasma profiling. For microvascular outcomes, 13,917 participants were followed for a median of 13.1 years; for macrovascular outcomes, 12,947 participants were followed for a median of 14.9 years. Sex-specific MVX was modeled continuously. Outcomes included composite microvascular complications and individual nephropathy (DKD), retinopathy (DR), neuropathy (DN), and macrovascular outcomes (non-fatal myocardial infarction [MI], non-fatal stroke and cardiovascular mortality). Sex-stratified multivariable Cox models were applied, and model performance was evaluated using Harrell’s C-statistic, NRI, and IDI. Results: During follow-up, 21.4% of participants developed microvascular complications. Addtionally, 15.5% developed macrovascular complications. In fully-adjusted models, higher MVX was associated with higher risk of composite microvascular complications in males (HR 1.32, 95% CI 1.15–1.51; [ P  < 0.001]) and females (HR 1.11, 95% CI 1.04–1.19; [ P  = 0.001]), and DKD in males (HR 1.24, 95% CI 1.15–1.33; [ P  < 0.001]) and females (HR 1.20, 95% CI 1.09–1.32; [ P  < 0.001]). MVX was associated with DN in males (HR 1.19, 95% CI 1.09–1.31; [ P  < 0.001]) but not with DR in either sex. For macrovascular outcomes, MVX remained associated with MI in males (HR 1.14, 95% CI 1.05–1.23; [ P  = 0.003]) and cardiovascular mortality in females (HR 1.16, 95% CI 1.00–1.35; [ P  = 0.047]). Adding MVX yielded improvement in discrimination and reclassification metrics. Conclusions: Higher MVX was independently associated with increased risks of diabetic vascular complications, with sex-specific patterns, and provided improved risk stratification beyond established factors.