Predictors of Outcomes in Diastolic Heart Failure

Study Questions:

What are the characteristics and outcomes in patients with diastolic heart failure (HF) with and without diabetes?


The study authors examined clinical and echocardiographic characteristics and outcomes in the I-Preserve (Irbesartan in Heart Failure With Preserved Ejection Fraction) trial, according to history of diabetes. They used Cox regression models to estimate hazard ratios (HR) for cardiovascular outcomes adjusted for known predictors, including age, gender, natriuretic peptides, and comorbidity.


The overall study cohort was comprised of 4,128 patients, of whom 27% (n = 1,137) had diabetes. Echocardiographic data were available in 745 patients. Of the 745 patients in the echocardiographic substudy, 25% (n = 187) had diabetes. Diabetics compared to those without were more likely to have a history of myocardial infarction (MI) (28% vs. 22%), higher body mass index (31 kg/m2 vs. 29 kg/m2), worse Minnesota Living With Heart Failure score (48 vs. 40), higher median N-terminal pro–B-type natriuretic peptide (NT-proBNP) concentration (403 vs. 320 pg/ml; all p < 0.01), more signs of congestion, but no significant difference in LVEF. The investigators found that patients with diabetes had a greater LV mass (173 ± 48 vs. 161 ± 48 g, p = 0.004), and left atrial area (24 ± 6 vs. 23 ± 6 cm2, p = 0.003), than patients without diabetes. Doppler E wave velocity (86 ± 32 vs. 76 ± 27 cm/sec, p < 0.0001) and the ratio of E/e' (11.7 vs. 10.4, p = 0.010) were higher in patients with diabetes. Over a median follow-up of 4.1 years, cardiovascular death or HF hospitalization occurred in 34% of patients with diabetes vs. 22% (n = 253) of those without diabetes (adjusted HR, 1.75; 95% confidence interval [CI], 1.49-2.05), and 28% vs. 19% of patients with and without diabetes died (adjusted HR, 1.59; 95% CI, 1.33-1.91).


The authors concluded that in diastolic HF patients, patients with diabetes have more signs of congestion, worse quality of life, higher NT-proBNP levels, a poorer prognosis, and display greater structural and functional echocardiographic abnormalities. They opined that further studies are needed to determine the mediators of the adverse impact of diabetes on outcomes in HF with preserved EF, and whether they are modifiable.


The hunt for lifesaving therapies for diastolic HF has been elusive. This is an important study because it suggests that diabetes itself, independent of underlying renal function, has an adverse impact on outcomes in diastolic HF. One question that comes to mind: “Is diabetes a marker of noncompliance?” As the authors point out, further studies are needed to determine whether preventing or improving diabetes results in better outcomes in patients with diastolic HF.

Clinical Topics: Anticoagulation Management, Heart Failure and Cardiomyopathies, Noninvasive Imaging, Prevention, Acute Heart Failure, Chronic Heart Failure, Heart Failure and Cardiac Biomarkers, Echocardiography/Ultrasound

Keywords: Comorbidity, Diabetes Mellitus, Diagnostic Imaging, Echocardiography, Heart Failure, Heart Failure, Diastolic, Myocardial Infarction, Natriuretic Peptide, Brain, Peptide Fragments, Primary Prevention, Quality of Life, Stroke Volume

< Back to Listings