Combined sleep behaviors, mediating biomarkers, and incident vascular complications among individuals with type 2 diabetes: A prospective study in <scp>UK</scp> Biobank

ABSTRACT Aims To prospectively examine how combined sleep behaviors affect vascular complications in type 2 diabetes, whether these associations are modified by disease severity, as well as the mediating roles of clinical biomarkers. Methods A total of 12,456 participants with type 2 diabetes from the UK Biobank were included. A sleep score was defined by: sleep duration, chronotype, insomnia, snoring, and daytime sleepiness. Cox regression models were used to examine the association between healthy sleep score and risk of vascular complications. Results Over 13.5 years, 1,878 macrovascular and 3,066 microvascular complications occurred. Compared with a healthy sleep score of 0–1, the multivariable‐adjusted HRs (95% CIs) of the score 4–5 were 0.75 (0.63, 0.89) for macrovascular complications [coronary artery disease 0.72 (0.57, 0.91), stroke 0.92 (0.66, 1.28), peripheral artery disease 0.79 (0.56, 1.10)]; 0.86 (0.75, 0.99) for microvascular complications [neuropathy 0.62 (0.47, 0.83), nephropathy 1.01 (0.84, 1.21), retinopathy 0.84 (0.68, 1.03)]. The aggregated diabetes severity score ( P interaction = 0.037) significantly modified the association with microvascular complications. Total cholesterol, LDL‐C, triglycerides, apolipoprotein B, and CRP collectively explained 11.34% (4.24%, 37.70%) of macrovascular associations [including 11.14% (2.85%, 43.03%) for coronary artery disease], while total bilirubin explained 3.57% (0.43%, 20.63%) of microvascular associations, collectively with triglycerides, GGT, and CRP mediating 9.33% (3.95%, 27.58%) for diabetic neuropathy. Conclusions A healthy sleep pattern is associated with lower risks of vascular complications in type 2 diabetes, independent of traditional risk factors, and the association with microvascular complications is modified by diabetes severity. Lipid profiles and inflammatory factors partly mediate the favorable association.