Comorbidities, quality of life and cause-specific outcomes in heart failure across the ejection fraction spectrum

Background

Heart failure (HF) is frequently accompanied by multiple comorbidities that correlate with poor outcomes and reduced health-related quality of life (HRQoL). However, the burden and influence of comorbidities may vary in HF with preserved (HFpEF), mildly reduced (HFmrEF), and reduced (HFrEF) ejection fraction (EF). Increased knowledge about the links between specific comorbidities and the different HF phenotypes and outcomes, along with a better understanding of the causes of death in HF patients, may contribute to improved phenotyping, prognostication, clinical management, and improved design of clinical trials in HF. Furthermore, there is a need for a better understanding of the characteristics associated with HRQoL and its prognostic implications in HF, especially across the EF categories.

Aims

1) To perform a comprehensive comparison of concomitant Type 2 diabetes mellitus (T2DM), atrial fibrillation (AF), and/or chronic kidney disease (CKD) in HFrEF, HFmrEF, and HFpEF regarding prevalence, clinical correlates, predictors and prognosis.

2) To assess the proportions and incidence rates (IRs) of cause-specific death within cardiovascular (CV) and non-CV death in HF in the three different EF categories.

3) To examine patients' characteristics and outcomes associated with HRQOL measured by EuroQoL 5-Dimensional Visual Analogue Scale (EQ-5D-vas) in HFrEF, HFmrEF, and HFpEF in hospitalized and outpatient settings.

4) To assess the characteristics of patients and their clinical outcomes related to self-reported anxiety/depression across different EF categories in hospitalized and outpatient settings.

Methods

In all four studies, data from the SwedeHF, linked to other national registries were used. The main statistical methods included logistic regression, cumulative incidence, Cox regression, and Fine-Gray regression models.

Comorbidities and cause-specific outcomes in different EF categories

Out of 42,583 patients included 2000-2012, 24% had T2DM, 51% CKD, 56% AF, and 8% all three comorbidities. Patients with HFpEF had a higher prevalence of CKD and AF, HFmrEF had an intermediate prevalence of AF, and the prevalence of T2DM was similar across the EF spectrum. Cardiovascular events were highest in HFrEF, and non-CV events were highest in HFpEF. Type 2 diabetes increased CV and non-CV events similarly but less so in HFpEF. Chronic kidney disease increased CV-events more than non-CV events, less so in HFpEF. Atrial fibrillation increased CV events substantially more than non-CV events, particularly in HFpEF and HFmrEF.

Cause-specific death in HF across the EF spectrum

Among 100,584 patients included 2000-2021, most deaths across all EF categories were attributed to CV causes. Within five years, HFpEF showed a higher adjusted risk of non- CV death, and a lower adjusted risk of CV death compared to HFrEF. Regardless of the EF category, the leading causes of death were ischemic heart disease (IHD) and cancer. The IR of CV death due to IHD was highest in HFrEF, while IRs of CV death due to pulmonary vascular disease, stroke, valvular heart disease, and AF increased with increasing EF. As EF increases, the IRs of non-CV deaths caused by cancer, respiratory diseases, and infections also rise.

HRQoL in HFrEF, HFmrEF and HFpEF

We studied 40,809 patients 2000-2021. Patients were categorized into four strata of HRQoL levels based on the EQ-5D-vas. Twenty-nine percent fell into the "best" category, 41% into the "good", 25% into the "bad", and 5% into the "worst", evenly distributed across all EF categories. Regardless of EF categories, a higher New York Heart Association (NYHA) class was strongly associated with lower EQ-5D-vas scores, as were liver disease, chronic obstructive pulmonary disease, smoking, higher body mass index, a history of stroke, the use of diuretics, and living alone. Conversely, higher income and being male were inversely related to lower EQ-5D-vas categories. Patients in the "worst" EQ-5D-vas category as compared with the "best" had the highest risk of all-cause death.

Anxiety/depression in HF across the EF spectrum

Out of 57,251 patients 2008-2023, 58% reported no anxiety/depression, 38% moderate, and 4% severe anxiety/depression, showing a comparable distribution across EF categories. Anxiety/depression was closely linked to HRQoL components, most strongly symptoms of fatigue, out of breath, and incapability to maintain main activities. When not accounting for HRQoL components, significant associations included higher NYHA class, liver disease, smoking, female sex, younger age, and inpatient status. Anxiety/depression levels correlated with increased risks of hospitalization and death within 12 months of follow-up. After adjusting for other HRQoL components, the association with mortality was no longer present.

Conclusions

Patients with HFpEF is characterized by more comorbidities and non-CV events compared to HFmrEF and HFrEF, but the impact of T2DM and CKD on events is lower in this group. Cardiovascular events occur most frequently in HFrEF. Across all EF categories, CV deaths were more prevalent than non-CV deaths, although the five-year risk of non-CV death increased with higher EF. Ischemic heart disease and cancer were the leading causes of CV and non-CV deaths, respectively, regardless of EF category. Most patients fell into the top two EQ-5D-vas categories. In all EF categories, a higher NYHA class exhibited the strongest relationship with lower EQ- 5D-vas levels. Patients with the worst EQ-5D-vas category faced the highest risk of mortality. Self-reported anxiety/depression was prevalent in all EF categories, closely linked to HRQoL, and associated with higher risks of HF- and all-cause hospitalization and mortality. However, the effect on mortality decreased once adjustments for HRQoL were made.

List of scientific papers

I. Comorbidities and cause-specific outcomes in heart failure across the ejection fraction spectrum: A blueprint for clinical trial design. Savarese G, Settergren C, Schrage B, Thorvaldsen T, Löfman I, Sartipy U, Mellbin L, Meyers A, Farsani SF, Brueckmann M, Brodovicz KG, Vedin O, Asselbergs FW, Dahlström U, Cosentino F, Lund LH. Int J Cardiol. 2020 Aug 15;313:76-8. https://doi.org/10.1016/j.ijcard.2020.04.068

II. Cause-specific death in heart failure across the ejection fraction spectrum: A comprehensive assessment of over 100 000 patients in the Swedish Heart Failure Registry. Settergren C, Benson L, Shahim A, Dahlström U, Thorvaldsen T, Savarese G, Lund LH, Shahim B. Eur J Heart Fail 2024 May;26(5): 1150-1159. https://doi.org/10.1002/ejhf.3230

III. Health-related quality of life across heart failure categories: associations with clinical characteristics and outcomes. Settergren C, Benson L, Dahlström U, Thorvaldsen T, Savarese G, Lund LH, Shahim B. ESC Heart Failure. 2024. https://doi.org/10.1002/ehf2.15206

IV. Anxiety/depression in Heart Failure Across the Ejection Fraction Spectrum: Analyses from the SwedeHF. Settergren C, Benson L, Thorvaldsen T, Dahlström U, Savarese G, Lund LH, Shahim B. [Submitted]