Heart failure with preserved and reduced ejection fraction : comorbidities, therapies and cause-specific outcomes

Abstract

Background: Heart failure (HF) is associated with worse prognosis and poor quality of life, and its rising global prevalence puts a growing strain on healthcare systems. While most previous studies have focused on cardiovascular (CV) comorbidities and CV death, especially in HF with reduced ejection fraction (HFrEF), less is known about the prognostic role of CV vs. non-cardiovascular (non-CV) comorbidities in HF with preserved ejection fraction (HFpEF). Although effective therapies exist for patients with HFrEF to improve survival, they remain underutilized in real-world healthcare settings.

Aims: The overall aim of this thesis was to evaluate the relative impacts of CV and non-CV comorbidities on specific causes of death and hospitalizations in HFpEF, and to investigate the adherence to guideline-recommended therapies in HFrEF considering the time of HFrEF diagnosis in a real-world setting. There were four specific aims: (1) Assess rates and predictors of CV and non-CV death and hospitalizations in a short-term follow-up study in patients with HFpEF (Study I). (2) Identify predictors for long-term all-cause death or HF hospitalization and all-cause death in patients with HFpEF (Study II). (3) Assess rates and predictors of CV and non-CV death in a long-term follow-up study in patients with HFpEF (Study III). (4) Describe the use of guideline-recommended therapies in relation to duration of HFrEF diagnosis in a large nationwide HF registry (Study IV).

Methods and Results: Study I-III included the Karolinska Rennes (KaRen) HFpEF cohort. Study I consisted of 539 patients with acute HF with data on short-term follow-up. Study II-III, consisted of 397 patients with data on long-term followup. Study IV was based on data from the Swedish Heart Failure (SwedeHF) Registry of 55 581 patients with HFrEF. Study I. Rates and predictors of CV and non-CV death and hospitalizations at short-term follow-up in HFpEF Over a median follow-up time of 2.0 (interquartile range 1.4, 3.2) years, the rates of CV vs. non-CV death (5.1 vs. 5.8 deaths per 100 patient-years) and CV vs. non-CV hospitalizations (33 vs. 27 per 100 patient-years) were similar. CV deaths were predominantly due to HF, while non-CV deaths were primarily linked to cancer. CV-related hospitalizations were mostly driven by HF and non-CV hospitalizations by lung disease. The severity of HF was the most important predictor of CV death, whereas history of anemia and prior stroke were predictors of non-CV death. Higher serum sodium levels were inversely associated with both outcomes. A range of CV and non-CV comorbidities were associated with hospitalizations, but they were difficult to assess as predictors. Study II. Predictors of all-cause death or HF hospitalization and all-cause death at long-term follow-up in HFpEF In the long-term follow-up (median [interquartile range]) 5.4 [2.1, 7.9] years) study, the rate of the primary endpoint (all-cause death or first HF hospitalization) was 227 events per 1000 patient-years and secondary outcome (all-cause death) 130 events per 1000 patients-years. Females had higher event-free survival. Several characteristics were associated with these outcomes with the strongest ones being tricuspid regurgitation peak velocity, diabetes mellitus, cancer, anemia, hyponatremia and male sex. Study III. Predictors of CV and non-CV death at long-term follow-up in HFpEF Nearly two-thirds of the 397 patients included died, half from CV and the other half from non-CV causes (62 vs. 58 deaths per 1000 patient-years, respectively). Significant predictors of CV death included coronary artery disease and tricuspid regurgitation peak velocity. Significant predictors of non-CV death included a history of stroke and kidney disease. Anemia and higher age were associated with both outcomes. Higher sodium concentrations and body mass index were inversely associated with non-CV death. Study IV. Guideline-recommended therapies in HFrEF in relation to the duration of HFrEF diagnosis Within 3 months, 3 to <6 months, 6 to 12 months and >12 months from diagnosis of HFrEF, 93%, 92%, 90% and 89% were on treatment with renin-angiotensin system inhibitors (RASI) or angiotensin receptor neprilysin inhibitors (ARNI), 9.8%, 17%, 19% and 22% on ARNI, 35%, 43%, 44% and 46% on mineralocorticoid receptor antagonist (MRA), 92%, 92%, 92% and 91% on beta-blockers, and 26%, 30%, 19% and 28% on sodium–glucose cotransporter 2 inhibitors, respectively. Additionally, 18% received cardiac resynchronization therapy or implantable cardioverter-defibrillator >12 months after diagnosis.

Conclusions: Patients with HFpEF have poor prognosis with similar rates of CV and non-CV death and hospitalizations. Several characteristics were associated with CV and non-CV outcomes at short- and long-term follow-up. In a comprehensive real-world registry of HFrEF patients, we found that the recent recommendation of immediate start of guideline-recommended HF therapies after a HFrEF diagnosis was adhered to regarding RASI and beta-blockers, but there was a substantial underuse of ARNI and MRA. These findings highlight the need to search for novel and tailored therapeutic approaches in HFpEF and close adherence to guideline-recommended therapies in HFrEF to improve outcomes.