Coronary artery disease is a stronger predictor of mortality than left ventricular ejection fraction among newly diagnosed patients with heart failure
European Heart Journal

Abstract
Heart failure (HF) prevalence remains high and prognosis is poor despite improvements in both medical treatment and implantation of primary prophylactic implantable cardioverter-defibrillator (ICD) and cardiac resynchronisation therapy (CRT) devices. At present, classification of HF and treatment algorithms are determined by the left ventricular ejection fraction (LVEF). However, in patients with LVEF <50% and newly diagnosed HF, evidence remains sparse as to whether LVEF per se or presence of CAD provide independent prognostic information on mortality.
Using the Western Denmark Heart Registry, we identified newly diagnosed patients with HF and LVEF 10–49% undergoing first-time coronary angiography (CAG) from 2003 to 2016 referred due to HF. Patients were stratified by LVEF (10–35% vs. 36–49%, according as to whether primary prophylactic ICD and CRT treatment should be considered) and presence of coronary artery disease (CAD). Maximum follow-up was 10 years. We estimated 10-year cumulative incidence of death and calculated hazard ratios (HR) adjusted for relevant comorbidities and risk factors.
Of 154,186 Western Danish residents undergoing CAG, 3,620 patients had HF and LVEF 10–49%. Among these, 2,780 (77%) patients had LVEF 10–35% and 840 (23%) patients had LVEF 36–49%. CAD was present in 1,592 (44%) patients.
There was a potential association in 10-year mortality when comparing patients with HF and LVEF 36–49% to those with LVEF 10–35% (37.3% vs. 42.1%, HR 1.15; 95% CI 0.99–1.34) (Figure 1A), with point estimates of mortality ranging from being 1% reduced to 24% increased. This result was not strongly influenced by the presence of CAD (HR 1.11; 95% CI 0.91–1.35) or absence of CAD (HR 1.24; 95% CI 0.97–1.57) (Figure 1B). There was no trend between LVEF categorized by 5-percentiles and mortality (p for trend = 0.24) (Table 1). In contrast, CAD was more strongly associated with increased 10-year mortality (55.0% vs. 31.5%, HR 1.43; 95% CI 1.25–1.64) irrespective of LVEF (Figure 1B).
Among newly diagnosed patients with HF and LVEF 10–49%, presence of CAD impacts mortality substantially more than LVEF per se. These results emphasize that assessment of CAD is pivotal for prognostication of newly diagnosed patients with HF and LVEF 10–49%.
Type of funding sources: None.
Figure 1
Table 1
Contributors

K Pryds
Author

R R Nielsen
Author

K K W Olesen
Author

M B Mortensen
Author

J C Nielsen
Author

M Maeng
Author
You may be interested in


