Frailty and risk of microvascular complications in patients with type 2 diabetes: a population-based cohort study

Abstract

Abstract Background Cross-sectional studies found that frailty was associated with prevalent diabetic microvascular complications (DMC). Longitudinal evidence in this regard is inconclusive and insufficient. We aimed to prospectively evaluate the association of pre-frailty and frailty with DMC in patients with type 2 diabetes (T2D). Methods We included 18,062 adults (mean age 59.4 ± 7.2 years, 37.4% female) with T2D at baseline in the UK Biobank. Frailty was defined using the frailty phenotype according to five components (weight loss, exhaustion, low physical activity, slow gait speed, and low grip strength). DMC, defined as diabetic nephropathy, diabetic neuropathy, or diabetic retinopathy, was identified using hospital inpatient records and death registries. Cox proportional hazard regression models considering competing risks were used to evaluate the associations of frailty phenotype with overall DMC events and subtypes. Results Among all participants, 6101 (33.8%) were classified as non-frail, 10,073 (55.8%) were classified as pre-frail, and 1888 (10.4%) were classified as frail. During a median follow-up of 12.0 years, 3678 DMC cases were documented, including 2213 diabetic nephropathy, 1520 diabetic retinopathy, and 673 diabetic neuropathy events. In the multivariable-adjusted model, compared with participants with non-frail, both pre-frailty and frailty were significantly associated with increased risk of overall DMC (HR 1.10, 95% CI: [1.02, 1.18] for pre-frailty and HR 1.52 [95% CI: 1.36, 1.69] for frailty). Similar results were observed in the subtypes of DMC. For each one-point increase in frailty phenotype score, the risk of overall DMC, diabetic nephropathy, diabetic retinopathy, and diabetic neuropathy event increased by 13%, 16%, 10%, and 20%, respectively. Conclusions Both pre-frailty and frailty were associated with an increased risk of DMC in patients with T2D. These findings have important implications for integrating early assessment and surveillance of frailty in diabetes and may favor the identification of at-risk patients.