1104Random plasma glucose predicts long-term mortality in patients with heart failure without previously known diabetes - insights from the Swedish heart failure registry (SwedeHF)

Abstract

Background: Dysglycaemia often coexists with heart failure and adversely impacts long-term prognosis. A random, elevated plasma glucose (RPG) during hospitalisation for an acute coronary syndrome or stroke often reflects undetected dysglycaemia and is associated with a poor prognosis. Purpose: To investigate the association between RPG and long-term prognosis in real world patients with heart failure with and without a reported diabetes (DM) diagnosis. Methods: Patients with and without previously known DM included in the Swedish national heart failure registry (SwedeHF) between 2003-2011 and with a reported RPG (n=10098) were followed for all cause mortality until December 31, 2014 (median 4.4 (IQR 2-6) years). Mortality was analysed by DM diagnosis or by glycaemic levels as follows: RPG <6.1 mmol/L; 6.1-6.9 mmol/L; ≥7.0 mmol/L (the latter = DM cut off level for fasting glucose according to WHO). Hazard ratios (HR) were calculated in a Cox regression model adjusting for age, gender, renal function and heart failure care (hospitalisation or out patient clinic-visit). Results: The number of patients without reported DM was 7223 of whom 22% (n=1574) had an RPG ≥7.0 mmol/L, 25% (n=1771) 6.1-6.9 mmol/L and 54% (n=3878) <6.1 mmol/L. DM was reported in 28% (n=2875). The mean age was lowest in patients with known DM (76 years) while it increased with higher RPG category; 77 (<6.1 mmol/L); 79 (6.1-6.9 mmol/L) and 80 (≥7.0 mmol/L) years respectively (p<0.0001). Left ventricular ejection fraction did not differ across the groups, (≈22% with LVEF ≥50%; p=0.81). Age adjusted survival by RPG category compared with known DM is depicted in Figure 1. Mortality increased by increasing RPG in patients without known DM and was highest in those with known DM (log-rank p=0.0001). Known DM was associated with an increased risk of mortality vs. the lowest RPG category (<6.1 mmol/L; adjusted HR 95% CI 1.51:1.42-1.60). The highest RPG category was associated with increased mortality even among those without known DM (adjusted HR 1.17, CI 1.08-1.25 comparing ≥7.0 vs. <6.1 mmol/L). There was no increased mortality risk comparing the slightly elevated vs. lowest RPG category (6.1-6.9 vs. <6.1 mmol/L; adjusted HR 1.04:0.96-1.11). Conclusions: In patients with heart failure without previously reported DM, increased levels of RPG were associated with greater risk of long-term mortality compared with lower RPG levels. DM was as expected associated with the highest mortality risk. These findings highlight the importance of searching for previously undetected dysglycemia and DM in heart failure populations. (Figure Presented) .