Home / Resources / Articles / Higher HbA1c Levels Provide Better Outcomes in Heart Failure with Diabetes

Higher HbA1c Levels Provide Better Outcomes in Heart Failure with Diabetes

Apr 19, 2012

Patients with advanced heart failure who were also diabetic had better two-year survival if their baseline HbA1c levels were more than 7.3%….

Dr. Tamara Horwich, lead investigator (University of California, Los Angeles) stated that, “Among the nondiabetic patients with heart failure, HbA1C levels did not predict survival outcomes.” “We’re finding that in heart failure [plus diabetes], higher HbA1C levels are associated not with worse outcomes, but with better outcomes.” This suggests that for patients with both diseases, “the focus should not be on lowering the HbA1C levels to as low as possible,” she said, adding that “aiming for a midrange of 7.2% to 8.2% may be very reasonable.”


Dr. David Aguilar(Baylor College of Medicine, Houston, TX), who was not involved with this research added that, This “adds to a small but growing body of literature demonstrating a complex relationship between levels of glycemic control and survival in patients with advanced, established heart failure.” However, because it was an observational study, residual confounding factors may be contributing to adverse outcomes.

The implications for clinical practice are that “in someone with advanced heart failure who may be having difficulties with hypoglycemia or other adverse effects of diabetic medications, less stringent glycemic control (HbA1C <8%) may be acceptable," he said. On the other hand, "if patients are tolerating the medications without difficulties, current glycemic guidelines should continue to be followed as we await further information from prospective clinical trials."

In patients with established heart failure, the relationship between HbA1C levels and survival is not clear, the group writes. TheCandesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program found that as HbA1C levels rose, cardiovascular risk increased. However, in a study of close to 6000 older veterans with heart failure, Aguilar and colleagues reported a U-shaped relationship between HbA1C and survival: optimal survival was seen among patients with “modest” glycemic control — an HbA1C of 7.1% to 7.8%.

Horwich and colleagues performed a retrospective analysis on data from 845 patients with advanced heart failure who had been referred to a cardiomyopathy center for heart transplant or other evaluation from 1999 to 2010. The patients had a mean age of 55 years, and 28% were women.

The team subdivided the cohort into two subgroups, 358 patients with diabetes and 487 patients without diabetes, who were then stratified into quartiles according to baseline HbA1C levels.

The primary end points were death or urgent need for heart transplantation and all-cause mortality at two years. At two years, 181 patients had undergone urgent heart transplantation and 180 patients had died.

Among the patients with diabetes, two-year event-free survival was highest in the two highest quartiles of HbA1C.

Patients with Diabetes: Two-Year Survival Free from Death/Urgent Heart Transplantation


HbA1C (%)

2-y survival (%)


In the subgroup of patients without diabetes, patients with HbA1C levels of 5.7% to 6.0% had better survival, but this was not statistically significant.

Patients without Diabetes: Two-Year Survival Free from Death/Urgent Heart Transplantation


HbA1c (%)

2-y survival (%)


In the total cohort, after adjustment for age, gender, body-mass index, and LVEF, for each unit increase in HbA1C, patients had an 8% decreased risk of death or need for urgent transplantation. In the subgroup of patients with diabetes, for each unit increase in HbA1C, patients had a 15% decreased risk of death or need for urgent transplantation. Low HbA1c levels may reflect an inflammatory state or malnutrition, or some antidiabetic therapies may be causing the adverse cardiac outcomes, Horwich speculated.

Many, large-scale, phase 3 trials are currently testing the safety and efficacy of diabetic therapies in high-risk patients, including those with heart failure, a group that has often been excluded from glycemia clinical trials, said Aguilar. “Hopefully, we will have more prospective data on the best treatment strategies in this patient population over the next several years. . . . For now, I believe that practicing clinicians should follow published guidelines, which state that target glycemic goals should be individualized.”

Tomova GS, Nimbal V, Horwich TB. Relation between hemoglobin A1c and outcomes in heart failure patients with and without diabetes mellitus. Am J Cardiol 2012; DOI:10.1016/j.amjcard.2012.02.022. Available at: http://www.ajconline.org

Careful analysis of similar studies, showed that those with the lowest A1c’s were treated with high doses of potent anti-glycemic medications (instead of low carbohydrate diets), gained weight, had wide blood sugar swings, and had frequent episodes of severe hypoglycemia. If this turns out to be the case here, then it is likely the treatments that caused the outcomes and not merely the A1c values.

Richard K. Bernstein, MD, FACE, FACN, FCCWS