Author + information
- W.H. Wilson Tang, MD∗ ( and )
- Justin L. Grodin, MD
- Department of Cardiovascular Medicine, Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio
- ↵∗Reprint requests and correspondence:
Dr. W.H. Wilson Tang, Cleveland Clinic, 9500 Euclid Avenue, Desk J3-4, Cleveland, Ohio 44195.
Worsening heart failure (WHF) has been proposed over the years as a clinically relevant endpoint in assessing therapies for acute decompensated heart failure (ADHF) (1). Although varied across different studies, WHF was often defined as “either failure to improve (persistent symptoms and signs of HF during treatment) or worsening symptoms and signs of HF, pulmonary edema, or cardiogenic shock after initial stabilization and treatment of at least 24 h any of which requires rescue therapy” (1). This short-term endpoint (often within 7 days of admission) parallels early reinfarction after myocardial infarction as an indicator of initial treatment failure (2). As such, WHF has been used as a component of a primary or secondary endpoint in many clinical trials testing both novel medications (3–7) and new methods for established therapies for ADHF (8–10) as summarized in Table 1. In this issue of JACC: Heart Failure, Davison et al. (11) added another analysis to this growing body of literature on WHF during treatment for ADHF. They retrospectively analyzed a pooled contemporary ADHF cohort to identify baseline predictors of WHF and its associated outcomes. As with the previous studies (1,12–15), respiratory status, increased neurohormonal activation, and other familiar (and some less familiar such as “Western-like country”) baseline markers of clinical severity were found to be significant, but modest, predictors in a multivariable model for incident WHF (C-statistic: 0.67, 95% confidence interval: 0.65 to 0.70). After adjusting for baseline risk factors, laboratory values with their changes, and physical examination changes, WHF was associated with increased length of stay, cardiovascular death, or heart failure/renal failure rehospitalization by 60 days, and death by 180 days (p < 0.0001 for all). When stratified by treatment type—additional intravenous loop diuretic drugs or inotropes or mechanical therapies—WHF remained associated with death by 180 days. As with other studies, the investigators concluded that WHF is an important clinical event during the clinical course of ADHF that is not fully explained by baseline risk factors and can track with adverse outcomes after ADHF.
Any clinician that has cared for patients with ADHF will no doubt recall their daunting encounters with patients not improving after initial treatment and will relate to the often-haunted-by-downward-spiraling nature of their clinical courses as portrayed in these studies. So far, the relentless effort to establish WHF as a potential therapeutic target or clinical endpoint should be highly commended. Yet, as tempting as it is to believe the appealing nomenclature, overcoming the following challenges is necessary, in part due to the fundamental assumptions being made in the conception of WHF.
The first challenge is to understand exactly what WHF is. Currently, WHF is arbitrarily defined by the clinician’s therapeutic decisions rather than an underlying biologic process. Though easily ascertained and readily documented, this assumes that clinicians are capable of providing accurate, reproducible, and timely prescription of advanced therapies on the basis of their subjective assessment. The reality is that defining WHF can be unclear and quite inconsistent given the wide array of local and regional practice patterns as well as diverse range of clinical expertise and subjective bedside assessments. Although attempts have been made to standardize this definition (13) to include both worsening signs and symptoms and requiring additional therapy, it is still a situational outcome, and like incident heart failure (16), WHF may occur differently depending on how patients present and how clinicians respond.
The second challenge lies with the lack of understanding of just who these patients might be. A corollary and often overlooked question is whether or not the clinical course leading to WHF is directly related to the ADHF admission or the intended treatment. Calling it “worsening” itself implies a clear progression of disease course despite standard pharmacological therapy (mainly in the form of diuretic therapy to relieve congestion) shortly after hospital admission for ADHF. Yet, in most cases, we simply do not have adequate knowledge or understanding of why WHF occurs and whether the deterioration is in any way related to the treatment. Unlike reinfarction after myocardial infarction, there is no uniform pathophysiological process(es) leading to WHF. It is also important to remember that incident WHF is associated with worse presenting symptoms, more respiratory compromise (e.g., lower oxygen saturation and higher respiration rate), worse renal function, and more neurohormonal activation (higher natriuretic peptide and blood urea nitrogen levels) (1,12–15). Furthermore, patients with WHF have more hemodynamic perturbations, as exemplified by lower mean arterial pressure, higher pulmonary capillary wedge pressure, and lower cardiac power output (14). As these features are commonly observed in advanced stages of HF, this begs the question regarding whether the clinical manifestations of WHF are indeed “new” or “recurrent” (akin to that of reinfarction), or whether these are simply persistent exacerbations and even pre-existed before admission. In other words, it is questionable whether the clinical deterioration depicted by WHF may have evolved with or without ADHF admission. None of the published studies can adequately dissect these possibilities.
The third challenge is the lack of appreciation of how WHF evolved in those experiencing WHF. Whereas WHF may be influenced by underlying disease severity, it may also be due to inappropriate treatment (either overzealous vs. insufficient) or a lack of appropriate or adequate treatment response. These may be important confounders that may in part help to explain why adverse outcomes occur beyond physiologic measures or traditional risk factors. Furthermore, if a specific intervention is to be tested to counter such clinical deterioration that leads to WHF, it has to be capable of targeting the very mediators that promote such adverse processes. Among such heterogeneity of ADHF phenotypes, we simply do not have a good sense whether a broad endpoint such as WHF can be altered by a single therapeutic intervention.
Theoretically it makes good sense to consider WHF as a potential therapeutic target and to pursue strategies to prevent such devastating clinical situations. It is a telling story when an intervention can prevent WHF. Unfortunately, the current concept of WHF defies simple definition and has no easily quantitative attributes. Indeed, the 3 challenges described herein remained largely unanswered despite the wealth of supportive studies. It is unlikely that the concept of WHF will be any better clarified by this or any future contributions in the form of post-hoc analyses or observational studies. We therefore need to step back and critically ask ourselves how much we truly understand the underlying biology of ADHF and the factors that precipitate WHF. These are necessary insights if we seek to develop better heart failure therapies and design successful clinical trials to demonstrate their effectiveness.
↵∗ Editorials published in JACC: Heart Failure reflect the views of the authors and do not necessarily represent the views of JACC: Heart Failure or the American College of Cardiology.
Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Packer M.,
- Colucci W.,
- Fisher L.,
- et al.
- Davison B.A.,
- Metra M.,
- Cotter G.,
- et al.,
- on behalf of the PROTECT and RELAX-AHF Executive Committees
- DeVore A.D.,
- Hammill B.G.,
- Sharma P.P.,
- et al.
- Cotter G.,
- Metra M.,
- Weatherley B.D.,
- et al.
- Zannad F.,
- Stough W.G.,
- Pitt B.,
- et al.