Associations of body mass index on worsening of heart failure and mortality in patients with heart failure and reduced left ventricular ejection fraction: a 10-year follow-up study (a NorthStar substudy)

Figures

• 

  Fig. 1. Cumulative incidence curves and adjusted hazard ratios of cardiovascular outcomes in groups by body mass index (reference: body mass index = 24–27). BMI, body mass index; aHR, adjusted Hazard ratio; CI, confidence intervals. Hazard ratios with 95% confidence intervals were adjusted for key prognostic factors: sex, age (< 70, > 70), eGFR (< 60, > 60), NT-proBNP level (< 1000, > 1000), NYHA class (1–2, 3), diabetes (yes, no), atrial fibrillation (yes, no), HF etiology (ischemic, non-ischemic), LVEF (< 30, > 30), use of MRAs, beta-blockers, and RAAS inhibitors

• 

  Fig. 2. Interaction analyses for the primary outcome based on interactions between prognostic factors and body mass index groups > 27 vs. 24–27 (reference). BMI, body mass index; eGFR, estimated glomerular filtration rate; NT-proBNP, amino-terminal-pro-brain-natriuretic-peptide; LVEF, left ventricular ejection fraction; NYHA class, New York Heart Association; HF, heart failure; MRA, mineralocorticoid receptor antagonists; RAAS inhibitors, Renin–angiotensin–aldosterone system inhibitors

• 

  Fig. 3. Interaction analyses for the primary outcome based on interactions between prognostic factors and body mass index groups < 24 vs. 24–27 (reference). BMI, body mass index; eGFR, estimated glomerular filtration rate; NT-proBNP, amino-terminal-pro-brain-natriuretic-peptide; LVEF, left ventricular ejection fraction; NYHA class, New York Heart Association; HF, heart failure; MRA, mineralocorticoid receptor antagonists; RAAS inhibitors, Renin–angiotensin–aldosterone system inhibitors

• 

  Fig. 4. Associations between body mass index, treated as a continuous variable, and the primary and secondary outcomes using restricted cubic splines in Cox proportional-hazards models. Hazard ratios with 95% confidence intervals were reported, using a body mass index of 27 kg/m2 as the reference. Models were adjusted for key prognostic factors: sex, age (< 70, > 70), eGFR (< 60, > 60), NT-proBNP level (< 1000, > 1000), NYHA class (1–2, 3), diabetes (yes, no), atrial fibrillation (yes, no), HF etiology (ischemic, non-ischemic), LVEF (< 30, > 30), use of MRAs, beta-blockers, and RAAS inhibitors

• 

  Fig. 5. Associations between body mass index treated as a continuous variable, and the primary and secondary outcomes stratified by heart failure etiology using restricted cubic splines in adjusted Cox proportional-hazards models. Hazard ratios with 95% confidence intervals were reported, using a body mass index of 27 kg/m2 in patients with ischemic HF as the reference. Models were stratified by HF etiology (ischemic, non-ischemic) and adjusted for key prognostic factors: sex, age (< 70, > 70), eGFR (< 60, > 60), NT-proBNP level (< 1000, > 1000), NYHA class (1–2, 3), diabetes (yes, no), atrial fibrillation (yes, no), LVEF (< 30, > 30), use of MRAs, beta-blockers, and RAAS inhibitors