Diabetes and all-cause mortality in a national telemonitored heart failure cohort: role of comorbidities, heart failure etiology burden, and short-term weight variability

Abstract

Abstract Background Diabetes mellitus is common in patients with heart failure (HF) and is associated with worse outcomes. However, its prognostic role in real-world telemonitored HF populations remains insufficiently characterized. We evaluated the association between diabetes and all-cause mortality in a national telemonitored HF cohort while considering HF severity, comorbidities, HF etiology burden, and short-term weight variability (WV). Methods We conducted a retrospective cohort study of adults enrolled in the French Satelia ® Cardio telemonitoring program between 2018 and 2024 (NCT07561866). Baseline clinical characteristics, HF etiologies, comorbidities, and repeated weight measurements were analyzed. Short-term WV was defined as the mean absolute 10-day weight change per patient and categorized according to the cohort median. Survival was assessed using Kaplan–Meier analyses and multivariable Cox proportional hazards models. Results Among 16,556 patients (mean age 74.9 ± 12.9 years; 64% male), 4,723 (28.5%) had diabetes. Median follow-up was 13 months in patients with diabetes and 11 months in those without diabetes; 2,266 patients died (13.7%). Patients with diabetes more frequently had ischemic and hypertensive HF etiologies, obesity and greater HF etiology burden. In multivariable models without NYHA class, diabetes was independently associated with higher mortality, with a significant diabetes-by-age interaction suggesting a stronger association among patients aged < 75 years. After adjustment for NYHA class in the smaller subgroup, this association was attenuated and no longer statistically significant. Male sex, clinician-reported renal failure, combined chronic obstructive pulmonary disease and sleep apnea syndrome, and higher NYHA class were associated with higher mortality, whereas obesity was associated with lower mortality. Individual HF etiologies were not consistently associated with mortality across adjusted models, whereas increasing HF etiology burden remained associated with higher mortality. High short-term WV was associated with poorer survival. Conclusions In this large national telemonitored HF cohort, the association between diabetes and mortality was attenuated after accounting for HF severity assessed by NYHA class. Mortality was more consistently associated with HF severity, comorbidities and cumulative HF etiology burden than with individual HF etiologies. Short-term WV emerged as an exploratory telemonitoring-derived marker associated with poorer survival. These findings support a multidimensional approach to risk assessment in patients with diabetes.