Low stress hyperglycemia ratio predicts worse prognosis in diabetic acute heart failure patients.

INTRODUCTION: Acute blood glucose but not glycated hemoglobin (HbA1c) predicts poor outcome in acute heart failure (HF). The stress hyperglycemia ratio (SHR) has been proposed as a prognostic predictor in various clinical settings. OBJECTIVES: We assessed the prognostic implications of the SHR in acute HF patients with and without diabetes. METHODS: We performed a retrospective analysis of an acute HF registry conducted between 2009 and 2010. Estimated average glucose (eAG) was calculated as (28.7×HbA1c)-46.7 and SHR as acute blood glucose divided by eAG. The primary endpoint was all-cause mortality. Follow-up was three months. Patients were grouped by SHR tertiles (≤0.88, 0.89-1.16, and >1.16). Cox regression analysis was used to test the association of SHR (cut-off 0.88) with all-cause mortality. Analysis was stratified according to the presence of diabetes. Multivariate models were built accounting for acute blood glucose and for eAG (models 1 and 2, respectively). RESULTS: We studied 599 patients, mean age 76±12 years, of whom 62.1% had reduced ejection fraction and 50.9% had diabetes. Median acute blood glucose, eAG and SHR were 136 (107-182) mg/dl, 131 (117-151) mg/dl, and 1.02 (0.20-3.34), respectively. During follow-up 102 (17.0%) died. In patients with diabetes, those in the lowest SHR tertile had a hazard ratio (HR) of 2.24 (95% CI: 1.05-5.22) (model 1) and 2.34 (1.25-4.38) (model 2). In patients without diabetes, the HR of three-month death in the lowest SHR tertile was 0.71 (95% CI: 0.36-1.39) and 1.02 (0.58-1.81). Significant interaction was observed between diabetes and SHR. CONCLUSIONS: In HF patients with diabetes, a SHR ≤0.88 was associated with a more than twofold higher three-month mortality risk. No such association was found in non-diabetic patients. The presence of diabetes influences the association of the SHR with mortality.