Seeking Ways to Address Non-Cardiovascular Death in Patients with Heart Failure

This article refers to ‘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’ by C. Settergren et al., published in this issue on pages 1150–1159. In the last few decades, several randomized controlled trials have demonstrated prognostic improvement with novel therapies, including both medications and devices, mainly reducing the incidence of cardiovascular (CV) deaths and/or heart failure (HF) rehospitalization which have been extensively used as clinical trial endpoints.1-3 However, morbidity and mortality in patients with HF remain high globally; therefore, further distinct approaches are warranted regarding the management of patients with HF. From the perspective of reducing mortality, it is of utmost importance to know the major cause of death in the HF population to develop a new therapeutic approach to improve mortality, although this is challenging considering that it requires a well-designed dataset, an elaborate definition of the cause of death, and an adequate number of patients followed up. In this context, the paper by Settergren et al.4 in this issue of the Journal added insightful data to the literature. Their study investigated the differences in cause-specific mortality among HF phenotypes—HF with reduced ejection fraction (HFrEF), HF with mildly reduced ejection fraction (HFmrEF), and HF with preserved ejection fraction (HFpEF)—using data from the nationwide Swedish Heart Failure (SwedeHF) Registry and the National Patient Register. In their analysis of more than 100 000 patients enrolled in SwedeHF from 2000 to 2021, the investigators found that most deaths were attributable to CV rather than non-CV causes across all HF phenotypes. However, HFpEF was associated with a significantly higher risk of non-CV death than HFrEF was, whereas HFrEF was associated with a greater risk of CV death than HFpEF was. Additionally, they observed that the relative risks of CV and non-CV deaths in patients with HFmrEF were more similar to those in patients with HFrEF than those in patients with HFpEF. Regarding specific causes of death, ischaemic heart disease (IHD) was the most common cause of CV death across all ejection fraction categories, whereas cancer and respiratory conditions were the most common causes of non-CV death across all ejection fraction categories. The incidence rate of CV death due to IHD was highest in the HFrEF group, whereas the incidence rates of CV death due to pulmonary vascular disease, stroke, valvular heart disease, and atrial fibrillation increased with increasing ejection fraction. The incidence rates of non-CV deaths owing to cancer, respiratory diseases, and infections increased with increasing ejection fraction. It should be acknowledged that this study focused on the cause of death, a concept that is different from the mode of death. The cause of death in this study was the most frequent cause leading to death which was extracted from the death certificate based on the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes, whereas the mode of death usually indicates the immediate cause of death and has been commonly used in recent studies, discussing CV or non-CV deaths, regarding patients with HF.5-8 For instance, atrial fibrillation as a cause of CV death or diabetes mellitus as a cause of non-CV death in this study is not included in the mode of death according to the HF Collaboratory and Academic Research Consortium.9 Despite this discrepancy in definition, the study results were almost consistent with those of previous studies, showing that non-CV death was as prevalent as the mode of death as CV death in patients with HFpEF. A participant-level, pooled analysis of data from the Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure (DAPA-HF) trial, and Dapagliflozin Evaluation to Improve the Lives of Patients With Preserved Ejection Fraction Heart Failure (DELIVER) trial suggested the higher proportionate contribution of non-CV death with rising ejection fraction, resulting in the comparable incidence in CV and non-CV deaths in patients with HFpEF.5 A Japanese prospective cohort study including 3717 patients with acute HF showed that 169 out of 392 (43%) were non-CV deaths in patients with HFpEF.7 In this study by Settergren et al., although total proportion of non-CV deaths out of all-cause deaths in patients with HFpEF was only 39%, the proportion increased over time from 2001 to 2021 in all of the patients with HFrEF, HFmrEF, and HFpEF, and the non-CV death to all-cause death ratio in patients registered around the year 2021 was almost 50% across the spectrum of HF phenotypes.4 This finding was spectacularly notable, and underlined the importance of developing management to tackle non-CV deaths in patients with HF regardless of ejection fraction (Figure 1). The study by Settergren et al.4 identified the specific cause of death in more than 100 000 patients across HF phenotypes. Regarding specific causes of CV death, IHD including myocardial infarction was the most common cause across all ejection fraction categories, with the second most common cause being HF death. This finding is inconsistent with those of recent randomized controlled trials. A pooled analysis of DAPA-HF and DELIVER showed that among CV deaths, the leading cause was sudden deaths (51%), followed by HF deaths (33%).5 Similarly, a post-hoc analysis of the Prospective Comparison of Angiotensin Receptor–Neprilysin Inhibitor With Angiotensin Receptor Blocker Global Outcomes in Heart Failure With Preserved Ejection Fraction (PARAGON-HF) trial demonstrated that among CV deaths, 37% were due to sudden death and 28% were due to HF deaths.6 This discrepancy between these randomized controlled trials and the study by Settergren et al. may arise from the definitions of cause of death based on ICD-10, which did not include sudden cardiac death as a code, so that sudden cardiac death might be coded as myocardial infarction or IHD in the national death register similar to an Australian cohort study.10 This is an important limitation of this study because preventing sudden cardiac death and IHD requires a completely different approach. Cancer was the most common cause of non-CV deaths across all ejection fraction categories, with the second most common cause being respiratory disease without pneumonia. These findings are consistent with previous studies,11, 12 and more studies are warranted to identify shared risk factors, the causal relationship between cancer and HF (i.e. cardio-oncology), and multidisciplinary strategies to improve non-CV mortality in patients with HF. In addition, we have also found that the increasing number of frailty domains (physical frailty, cognitive dysfunction, and social frailty) was independently associated with higher incidence of non-CV deaths but not with HF deaths or other CV deaths in 1181 older patients with HF, and this association was persistent regardless of ejection fraction.13 This finding is consistent with previous studies showing that advanced age, which is strongly associated with physical frailty, was associated with an increase in non-CV deaths rather than CV deaths in patients with HF,14, 15 and shed a light on multidomain frailty as a potential target to improve non-CV mortality in older patients with HF. Altogether, totality of these study results, including the study by Settergren et al., clearly indicate that intensification of guideline-directed medical therapy and advanced interventions alone may not improve survival in older patients with HF, especially those with multidomain frailty, and that recognizing the type of each patient's frailty may allow clinicians to provide a more personalized approach. In summary, the study by Settergren et al. added valuable information regarding cause-specific death and its transition over 20 years in more than 100 000 patients with HF across the spectrum of HF phenotypes, signifying the importance of non-CV deaths in the contemporary era, especially in patients with HFpEF. From both clinical and scientific perspectives, a multidisciplinary approach for improving non-CV mortality in patients with HF is warranted. Conflict of interest: Y.F. has nothing to disclose. Y.M. received an honorarium from Otsuka Pharmaceutical Co., Novartis Pharma K.K., Bayer Inc., and AstraZeneca and research grants from Pfizer Japan Inc., Otsuka Pharmaceutical Co., EN Otsuka Pharmaceutical Co., Ltd., and Nippon Boehringer Ingelheim Co., Ltd.