Hemodynamic phenotypes defined by arterial stiffness index and pulse pressure with risk of diabetic microvascular complications in type 2 diabetes
Objective To identify hemodynamic phenotypes based on arterial stiffness index (ASI) and pulse pressure (PP) and examine their associations with incident diabetic microvascular complications (DMC) in individuals with type 2 diabetes (T2D). Methods A total of 9,163 participants with T2D free of DMC at baseline were included from the UK Biobank. K-means clustering based on ASI and PP was used to identify hemodynamic phenotypes. Associations with incident DMC were evaluated using cause-specific Cox and Fine–Gray models. Restricted cubic spline analyses assessed nonlinear associations. Secondary analyses examined diabetic kidney disease (DKD), diabetic retinopathy (DR), and diabetic neuropathy (DN). Results During a median follow-up of 12.9 years, 2,349 participants developed DMC. Three phenotypes were identified. Compared with the low ASI–low PP phenotype, the high PP phenotype was associated with a higher risk of DMC in both cause-specific Cox models (hazard ratio [HR] 1.16, 95% CI 1.05–1.28) and Fine–Gray models (subdistribution hazard ratio [SHR] 1.16, 95% CI 1.05–1.29), whereas no significant association was observed for the high ASI phenotype. ASI and PP showed L-shaped and J-shaped associations with DMC, respectively. ASI was primarily associated with DKD, whereas PP was associated with both DKD and DR. Conclusions Hemodynamic phenotypes defined by ASI and PP were associated with differential risks of DMC in T2D. Elevated PP, rather than elevated ASI, was consistently associated with increased microvascular risk. Distinct associations of ASI and PP suggest heterogeneous hemodynamic pathways underlying DMC.