Objective: We aimed to investigate whether adjustment for frailty influences the relationship of blood pressure with mortality in elderly patients with type 2 diabetes mellitus (T2DM).
Methods: Patients aged 60 years and older (n = 858) were selected from a prospective observational cohort study of primary care patients with T2DM. Frailty was defined as a score less than 80 on the subscale 'physical functioning' of the RAND-36 questionnaire. After median follow-up for 14 years, multivariate Cox regression analyses were performed to evaluate the association between blood pressure and (cardiovascular) mortality. Analyses were performed in strata according to the frailty level ('physical functioning' score <80 and ≥80) and were repeated for patients older than 75 years.
Results: Frailty was highly prevalent in our study population; 629 out of 858 patients (73%) fulfilled the criterion. For patients aged at least 60 years, the hazard ratios (95% confidence interval) of a 10 mmHg increase in SBP and DBP for cardiovascular mortality in nonfrail patients were 1.38 (1.15-1.68%) and 1.60 (1.07-2.37%), respectively. No relationship was observed for frail patients. For the oldest frail elderly, the hazard ratios of SBP and DBP for all-cause mortality were 0.92 (0.87-0.98%) and 0.83 (0.73-0.93%), respectively. For the oldest nonfrail elderly, a positive relationship between SBP and all-cause mortality was observed.
Conclusions: Frailty modifies the relationship between blood pressure and mortality in elderly patients with T2DM. Higher blood pressure was related to increased cardiovascular mortality in nonfrail patients, even in the oldest elderly, and to lower all-cause mortality in frail patients.