American Diabetes Association
Sacre.Supplement.R1_clean.pdf (442.13 kB)

Incidence of hospitalization for heart failure relative to major atherosclerotic events in type 2 diabetes: A meta-analysis of cardiovascular outcomes trials

Download (442.13 kB)
posted on 2020-09-21, 19:03 authored by Julian W. Sacre, Dianna J. Magliano, Jonathan E. Shaw
Background. Emerging evidence points to heart failure being a common first presentation of cardiovascular (CV) disease in type 2 diabetes.

Purpose. To determine whether hospitalization for heart failure (HHF) occurs more or less frequently than major adverse CV events (MACE) in people with type 2 diabetes.

Data Sources. Placebo arms of CV outcomes trials in type 2 diabetes.

Study Selection. Dipeptidyl peptidase-4 inhibitor, glucagon-like peptide-1 receptor agonist, and sodium-glucose co-transporter-2 inhibitor CV outcomes trials (n=16).

Data Extraction. We meta-analyzed incidence rates of HHF, myocardial infarction (MI), stroke, and the composite outcomes of ‘CV death or HHF’ and MACE (CV death, nonfatal MI, or nonfatal stroke).

Data Synthesis. In two trials enriched with CKD patients, HHF was more common than both MI and stroke. Among the remaining 14 trials, HHF was less frequent than MI in 13 studies (93%), with this difference being significant in 8 studies (57%); however, HHF surpassed stroke in all but one study (93%; significant in seven studies [50%]). Heterogeneity among trials was moderate/high (I2>50%), and partly explained by HHF/MI correlating with age and previous MI history (p<0.05). In seven trials that reported events stratified by presence/absence of pre-existing CV disease, ratios of HHF/MI and HHF/stroke were similar between groups.

Limitations. Enrichment of trial populations with those at high risk of CV events limits generalizability.

Conclusions. Although less frequent than MI, HHF is a common event in type 2 diabetes – both in those with and without prior CV disease.


This work was supported by the National Health and Medical Research Council of Australia (NHMRC; APP1107361 to DJM and APP1173952 to JES) and the Victorian Government’s Operational Infrastructure Support Program. The paper includes previously unpublished data from the REWIND trial, which was sponsored by Eli Lilly. This sponsor had no role in the design of this meta-analysis, nor in preparation, review, or approval of the enclosed manuscript.