The Real World of <i>de novo</i> Heart Failure: The Next Frontier for Heart Failure Clinical Trials?

Abstract

This article refers to ‘Readmission and death in patients admitted with new-onset versus worsening of chronic heart failure: insights from a nationwide cohort’ by J.H. Butt et al., published in this issue on pages 1777–1785. Across the world, as many as 64 million people carry a diagnosis of heart failure (HF) and HF remains one of the most common reasons for hospitalization.1 Inherently, every decompensated HF patient presenting to the hospital must fall into one of two categories: de novo/recently diagnosed HF or worsening chronic HF. While this paradigm has always existed, only in the past few years have the implications of these two different types of patients been systematically studied. Notably, data from the ASCEND-HF (Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure) trial showed that a diagnosis of HF for ≤1 month before hospitalization was independently associated with roughly a twofold lower risk of death at 180 days compared to patients hospitalized with worsening of chronic HF.2 While the heterogeneity of patients hospitalized for HF has long been recognized and been a topic of constant research, these data from ASCEND-HF were among the first to establish HF duration and the simple schema of ‘new’ vs. ‘old’ HF as clinically relevant. While the data from ASCEND-HF were informative, ASCEND-HF included a selected population in the context of a clinical trial. Trial databases have the advantage of more granular and comprehensive data collection, but may be less generalizable to populations seen in routine clinical practice. In terms of real-world data, reports for rates of de novo vs. worsening chronic HF have varied widely (i.e. de novo rates of 13% in North America to 53% in South Korea), perhaps reflecting different demographic characteristics and/or differences in definitions or data capture.3-5 Most recently, using a single worldwide protocol, the global REPORT-HF (International Registry to Assess Medical Practice with Longitudinal Observation for Treatment of HF) registry of patients hospitalized for HF reported the overall contemporary proportion of de novo HF to be 43%, with rates ranging from 20% in North America to 79% in Southeast Asia.6 REPORT-HF investigators also found de novo HF patients to have comparatively better outcomes, with worsening chronic HF independently associated with a 43% higher risk of 1-year mortality.7 Nonetheless, it is important to recognize that REPORT-HF did require patient consent and had low in-hospital mortality, features that suggest possible selection bias and potentially challenge the ‘real-world’ nature of the data. Thus, large-scale data from contemporary unselected populations are needed to best inform the true real-world prevalence and outcomes of patients hospitalized with de novo HF. In this context, the article by Butt et al.8 in this issue of the Journal is a welcomed addition to the literature. To our knowledge, the authors provide the first study to examine the relationship between long-term outcomes and pre-existing HF duration in a large unselected nationwide and contemporary cohort of hospitalized HF patients. The study included 17 176 patients at first hospital admission for HF in the period 2013–2015 using data from Danish nationwide registries. New-onset HF was defined as either no history of HF or a history of HF of ≤30 days, and worsening of chronic HF was defined as a history of HF of >30 days. The duration of HF was determined using inpatient and outpatient diagnosis codes up to 10 years prior to index admission. In total, 8860 (52%) patients were admitted with new-onset HF and 8316 (48%) with worsening chronic HF. After adjustment for potential confounders, worsening of chronic HF was independently associated with a substantially higher rate of all-cause mortality and HF readmission, compared with new-onset HF. The results remained consistent across subgroups and in a sensitivity analysis using a stricter definition of de novo HF restricted to only patients diagnosed at time of index admission (i.e., excluding 30-day window). Butt et al. should be congratulated for an elegant and timely analysis that broadens the scope of real-world data surrounding de novo vs. worsening chronic HF (Figure 1). These data are consistent with existing literature establishing de novo or recently diagnosed HF as an inherently lower risk condition within the broader pool of patients hospitalized for HF.2, 4 A hypothesized mechanism for this consistent finding is that de novo patients inherently involve a subset of patients that experience cardiac recovery.2 The possibility of cardiac recovery is supported across multiple analyses by the baseline characteristics of de novo patients, who consistently tend to have a higher proportion of non-ischaemic cardiomyopathy and typically younger age with less comorbidities.2, 8 Moreover, some de novo patients may have conditions often associated with transient left ventricular dysfunction such as stress (i.e. ‘Takotsubo’) cardiomyopathy and acute myocarditis. Likewise, the possibility of tachycardia-induced cardiomyopathy for a proportion of de novo patients is further supported in the current analysis, whereby comorbid atrial fibrillation was associated with better prognosis in de novo HF, suggesting that rhythm or rate control may lead to significant improvement or resolution of the cardiomyopathy.8 In contrast, atrial fibrillation in the setting of chronic established HF was associated with worse prognosis. Multiple therapies, including angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, angiotensin receptor–neprilysin inhibitors, beta-blockers, mineralocorticoid receptor antagonists, sodium–glucose co-transporter 2 inhibitors and most recently vericiguat, a soluble guanylate cyclase stimulator, have been shown to improve outcomes for patients with HF with reduced ejection fraction (HFrEF).9-11 Moreover, omecamtiv mecarbil, a novel cardiac myosin activator has also been shown to improve cardiac function, and is being investigated in the the GALACTIC-HF (Global Approach to Lowering Adverse Cardiac Outcomes Through Improving Contractility in Heart Failure) trial.12 For drug approval, it is reasonable that the initial landmark trial with a new therapy demonstrates incremental efficacy beyond already indicated background medications to justify its value. However, despite the sheer number of effective HFrEF drugs available, use of these medications in clinical practice remains low and has not meaningfully improved over time.13, 14 Likewise, a recent real-world study of de novo HF in the U.S. found that patients generally receive little guideline-directed medical therapy (GDMT) by the time they have their first worsening HF event, suggesting a blurred line between true worsening HF (i.e. breakthrough worsening despite true maximally tolerated therapy) and undertreated HF.15 In that study, the first worsening HF event took place on average ∼ 18 months after initial diagnosis, which should have provided ample time to initiate and up-titrate GDMT to maximally tolerated or target doses.15 The above described paradox between many effective drugs and overall poor utilization in clinical practice highlights the critical need for improved implementation. In this context, research and clinical trials in de novo HF populations may open unique doors for implementation science (Figure 1). This may include sequencing trials to identify which combinations of drugs have the best adherence and tolerability, and/or lead to best rates of cardiac recovery and clinical outcomes. While guidelines and totality of evidence generally support maximally tolerated doses of all drugs for all patients with HFrEF, determining optimal sequencing may bridge the divide between clinical trial evidence and real-world use.13, 14 Sequencing trials (or even ‘withdrawal of therapy’ trials) may have logistical and ethical challenges among patients with chronic established HFrEF already on background therapy. In contrast, in the current era of ‘HFrEF polypharmacy’ and cost considerations, de novo HF patients inherently starting on zero to little evidence-based medical therapy may be ideal for answering questions regarding ‘the next best drug to initiate/titrate’. De novo HF is also a key time for implementation research to improve the overall velocity and completeness of GDMT initiation/titration, and has been termed the ‘other vulnerable phase’ given the importance of optimizing GDMT quickly before HF chronicity develops.16 Improved use of GDMT after an initial HFrEF diagnosis could have major implications for implantable cardioverter-defibrillators and other device therapies, as appropriate upfront initiation and escalation of GDMT would improve rates of left ventricular recovery and potentially obviate the need for device implantation in some instances. From a feasibility perspective, available data suggest the prevalence of de novo HF is likely high enough to support dedicated trials. Indeed, HF hospitalizations are exceedingly common and Butt et al. found that most patients hospitalized for HF in Denmark were de novo or recently diagnosed (52%), roughly aligned with the 43% seen in REPORT-HF. However, there remain key unanswered questions for de novo HF patients in the real-world setting, all of which will be important for understanding feasibility and design of any future de novo HF trials. For example, one key unknown is the proportion of de novo HF patients that are actually hospitalized – i.e. where is de novo HF diagnosed? where would we enrol patients in a trial of de novo HF? While clinical experience may suggest the majority of de novo HF patients present to the hospital, a study from an administrative database suggested that nearly 50% of incident HF may be diagnosed as an outpatient.17 Further research must validate this result, assess for differences across countries and global regions, and further characterize the patient journey of de novo HF patients in the real world. In summary, the analyses by Butt et al.8 provides important real-world evidence validating de novo HF as a large proportion of patients hospitalized for HF, and confirming this subset as lower risk than patients hospitalized with worsening chronic HF. On one hand, the markedly higher event rates among worsening chronic HF support the continued focus on better therapies and implementation strategies for this exceptionally high-risk subset. However, on the other hand, despite the lower clinical risk, de novo HF may represent an underappreciated opportunity for clinical research and is well-positioned to answer questions regarding medication sequencing and implementation science. As the clinical trial arena for chronic HFrEF pharmacotherapy becomes increasingly crowded and complex, innovative studies of de novo HF have the potential for far-reaching impact that could influence care for the overall HFrEF condition. Conflict of interest: M.S.K. reports no relevant disclosures. J.B. has received research support from the NIH, PCORI and the European Union; and serves as a consultant for Amgen, Array, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, CVRx, G3 Pharmaceutical, Innolife, Janssen, Luitpold, Medtronic, Merck, Novartis, Relypsa, Stealth Peptides, SC Pharma, Vifor, and ZS Pharma. S.J.G. has received a Heart Failure Society of America/Emergency Medicine Foundation Acute Heart Failure Young Investigator Award funded by Novartis; has received research support from Amgen, Bristol-Myers Squibb and Novartis; has served on advisory boards for Amgen and Cytokinetics; and serves as a consultant for Amgen and Merck.

Topics & Concepts