Home / Resources / Articles / Elevated Glucose Predicts Mortality

Elevated Glucose Predicts Mortality

Jan 25, 2013
Elevated blood glucose levels at admission for acute heart failure significantly predicted 30-day mortality in a large observational study, independently of whether there was coronary artery disease, impaired ventricular or renal function, or even a prior diagnosis of diabetes….

Although the findings don’t show whether derangements in glucose metabolism are a cause or result of the heart failure, adding blood glucose levels significantly improved 30-day risk stratification for the patients compared with a clinically based prediction model.

Lead Investigator Dr. Alexandre Mebazaa (Université Paris Diderot and INSERM, Paris, France), stated that, "Because blood glucose is widely measured, easily interpreted, and inexpensive to measure, its use for risk assessment in acute heart failure is worthy of consideration.


Further, they write, their findings "suggest that stress-induced impaired glucose tolerance and/or occult diabetes among patients with acute HF is common and deserves further study, especially with respect to efforts at follow-up care and longer-term management." Their preferred interpretation, Mebazaa et al make clear, is that elevated glucose in this setting likely represents a modifiable treatment target, not just a prognostic marker.

"One can envision numerous reasons why hyperglycemia might lead to an adverse outcome in patients with acute HF," they write. "In our cohort, the early and substantial risk associated with elevated glucose concentrations argues most compellingly for a direct myocardial effect, related to either reduced pump function or arrhythmia promotion. Considerably more data are needed to answer whether hyperglycemia is a marker or mediator, but our compelling data suggest the latter rather than the former."

Mebazaa et al looked at patient-level data from 6212 patients with acute heart failure in cohorts primarily from Western and Eastern Europe but also the US and Japan. Their median admission blood glucose level was 135mg/dL.(7.5 mmol/L); diabetes in the analysis was defined as patient self-reported diabetes, ongoing antidiabetic therapy, or a diabetes diagnosis in the patient’s medical record.

Elevated glucose was defined as >126mg/dL.(7 mmol/L) in nondiabetic patients and >180mg/dL.( 10 mmol/L) in patients with diabetes. Defined that way, elevated blood glucose was seen in 42% and 50% of nondiabetics and diabetics, respectively.

Overall 30-day mortality was 10%; the 618 who died had a mean blood glucose concentration of 160mg/dL.(8.9 mmol/L), compared with 133mg/dL.(7.4 mmol/L) for those alive at 30 days (p<0.0001).

Elevated blood glucose significantly predicted 30-day mortality, with a hazard ratio of 2.19 (95% CI 1.69–2.83; p<0.001) after adjustment for age, sex, history of heart failure or coronary artery disease, history of diabetes, systolic and diastolic blood pressure, heart rate, serum sodium, and renal function. (Other significant predictors included hyponatremia, poor renal function, older age, and increasing systolic blood pressure.) The elevated mortality risk remained significant across all analyzed patient subgroups.

Hazard Ratio (HR) for 30-Day Mortality in Selected Subgroups

HR (95% CI)
Preserved systolic function
5.41 (2.44–12.0)
Impaired systolic function
2.37 (1.57–3.59)
2.18 (1.67–2.83)
1.65 (1.23–2.22)
No CAD history
2.27 (1.58–3.25)
History of CAD
2.1 (1.45–3.06)
eGFR >60 mL/min/1.73 m2
2.36 (1.4–3.98)
eGFR <60 mL/min/1.73 m2
2.12 (1.58–2.86)
p<0.0001 for all hazard ratios

Increasing blood glucose as a continuous variable was also a significant predictor of 30-day mortality, with the adjusted risk going up 9% for every 18mg/dL.(1-mmol/L) increment (p<0.0001).

If blood glucose elevations actually promote acute decompensation in heart failure and aren’t just a signal, a next step would be an intervention trial.

Mebazaa A, Gayat E, Lassus J, et al. Association between elevated blood glucose and outcome in acute heart failure. Results from an international observational cohort. J Am Coll Cardiol 2013; DOI:10.1016/j.jacc.2012.11.054. Available at: http://content.onlinejacc.org.