Health-Related Quality of Life Outcomes in PARADIGM-HF

Health-related quality of life (HRQL) is a key target of therapy in the management of patients with chronic heart failure (HF).¹ The HRQL of HF patients is more impaired than age-matched patients without chronic illnesses and those with other comorbidities,² and HRQL perceptions are predictive of future risk for morbidity and mortality.^3,4 For many patients with HF, in addition to prolonging life, improving HRQL is important.⁵ Given this focus on HRQL by both patients and clinicians, it is relevant to determine the impact of novel interventions on these important patient-reported outcomes.

Angiotensin-converting enzyme (ACE) inhibitors have been the standard of care for management of patients with HF with reduced ejection fraction because of improved survival and reduced hospitalizations for HF.⁶ ACE inhibitors and angiotensin receptor blockers have variable long-term effect on HRQL.^2,7,8 Neprilysin is a neutral endopeptidase that facilitates breakdown of natriuretic and other vasoactive peptides. Inhibition of this enzyme provides higher endogenous levels of vasoactive peptides, including natriuretic peptides, which may promote several changes that could improve HRQL.⁹ Sacubitril/valsartan combines the neprilysin inhibitor, sacubitril, with the angiotensin receptor blocker, valsartan, to minimize the risk of serious angioedema previously seen with drugs that act to inhibit both neprilysin and ACE. The PARADIGM-HF trial (Prospective Comparison of ARNI [Angiotensin Receptor–Neprilysin Inhibitor] With ACEI [Angiotensin-Converting–Enzyme Inhibitor] to Determine Impact on Global Mortality and Morbidity in Heart Failure) demonstrated that sacubitril/valsartan was superior to enalapril with regards to reduction in cardiovascular death, HF hospitalizations, and all-cause death.¹⁰ Sacubitril/valsartan also prevented the clinical progression of HF in surviving patients and risk of sudden cardiac death.^11,12 This article provides a comprehensive analysis of HRQL in PARADIGM-HF, a key prespecified secondary outcome measure.⁹ The principal objective of this analysis was to determine whether sacubitril/valsartan was superior to enalapril on HRQL changes at 8 months. Secondary objective was to provide an assessment of long-term HRQL changes beyond 8 months.

The details of PARADIGM-HF have been previously published.^9,10 Briefly, HF patients ≥18 years of age with New York Heart Association (NYHA) class II, III, or IV functional capacity, left ventricular ejection fraction (LVEF) ≤40%, and either a plasma BNP (B-type natriuretic peptide) >150 pg/mL or NT-proBNP (N-terminal pro-B-type natriuretic peptide) >600 pg/mL or a hospitalization for HF within past 12 months were eligible to be enrolled. Patients already taking ACE inhibitors or angiotensin receptor blockers were eligible if they were taking a daily dose equivalent to enalapril 10 mg and were on stable dose of β-blocker for a minimum of 4 weeks. Key exclusion criteria included symptomatic hypotension, systolic blood pressure <100 mm Hg at screening, estimated glomerular filtration rate <30 mL/(min 1.73 m²), history of angioedema, or potassium >5.2 mmol/L. Eligible patients were entered (in a single blinded fashion) into a run-in phase where they took enalapril 10 mg twice daily for 2 weeks followed by sacubitril/valsartan 100 mg twice daily initially followed by 200 mg twice daily for a 4- to 6-week period. Patients without significant intolerances to either drug were randomized in 1:1 ratio to either enalapril 10 mg twice daily or sacubitril/valsartan 200 mg twice daily in a double-blinded fashion. The primary outcome was a composite of cardiovascular death or hospitalization for HF. The study was approved by an institutional review committee and informed consent was obtained.

Among the 8399 patients enrolled in PARADIGM-HF, 7623 (91%) in 38 countries completed KCCQ scores at randomization. The baseline characteristics of the patients who completed the KCCQ were different than the 776 patients who did not complete KCCQ (Table 1). The characteristics of the patients who did not complete KCCQ include younger age, lower body mass index, less comorbid illnesses, higher percentage of women, and higher proportion from Asia Pacific region. Among patients who completed KCCQ, the patient characteristics in the 2 treatment groups were similar (Table 2). Compared with enalapril, the mean baseline KCCQ-CS scores (76.6±19 versus 75.3±19; P=0.004) and KCCQ-OS scores (73.5±20 versus 72.3±19; P=0.007) were higher in sacubitril/valsartan groups. The overall completion rate decreased to 90% at 8 months, 87% at 12 months, 61% at 24 months, and 29% at 36 months (Figure 1; Table I in the Data Supplement). The distribution of KCCQ scores are skewed toward the right (Figure 2).

At 8 months, the difference in change scores favored the patients in sacubitril/valsartan group in both KCCQ-CS score (+0.64 versus −0.29; P=0.008) and KCCQ-OS score (+1.13 versus −0.14; P<0.001) in comparison to those in enalapril group (Table 3). The patients in the sacubitril/valsartan group also noted improvements in all domains of KCCQ, with an exception of symptom stability with a 2.9-point decline (Table 3); notably, the enalapril group had a 4.3-point decline in that domain. In contrast, patients in the enalapril arm noted declines in most domains. There were consistent improvements in KCCQ-CS and KCCQ-OS scores in the sacubitril/valsartan group (in comparison to enalapril) for almost every subsequent visit point through 36 months (Table 4). At 8 months, the proportion of patients with clinically meaningful improvements in KCCQ-OS scores was significantly greater for sacubitril/valsartan group than for enalapril group (35% versus 33%), and the proportion with deterioration was less for sacubitril/valsartan (27% versus 31%; Table IIa and IIb in the Data Supplement).

In a multivariable model, several clinical factors were independently associated with deteriorations in both KCCQ-OS and KCCQ-CS scores, including higher body mass index, higher NT-proBNP, NYHA functional class III/IV, female sex, and history of myocardial infarction, atrial fibrillation, and diabetes mellitus (Table 5). Conversely, patients enrolled in Latin America and Asia noted improvements in these scores. Randomization to sacubitril/valsartan remained an independent predictor of improvements in KCCQ-OS and KCCQ-CS scores after adjustment for these factors. With an exception of older age, all factors were associated with changes in KCCQ scores with and without NYHA in the model.

A total of 305 (4.4%) patients were hospitalized for HF between baseline and 8 months. The patients without a HF hospitalization had modest improvements in KCCQ scores in both sacubitril/valsartan and enalapril (Table 6). In contrast, patients with HF hospitalization had significant decreases in KCCQ scores at 8 months. In comparison to enalapril group, patients randomized to sacubitril/valsartan had a smaller decline in KCCQ-OS scores (5.66 point difference; P=0.003) and KCCQ-CS scores (5.14 difference; P=0.005). There was no difference in mean time between HF hospitalization and 8-month KCCQ assessment in sacubitril/valsartan (−4.0+2.3 months) and enalapril (−4.3+2.6 months) groups (P=0.27).

When imputing 0 for death during follow-up, there were significant differences favoring sacubitril/valsartan for KCCQ-CS, KCCQ-OS, and all 8 domains, with least square mean differences ranging from 1.41 to 2.56 points (Table III in the Data Supplement).

In a large, well-treated population with HF with reduced ejection fraction, sacubitril/valsartan was superior to enalapril in improving disease-specific HRQL. The principal HRQL end points of KCCQ-CS and KCCQ-OS were improved by 4 months, and these improvements persisted throughout 36 months. There were statistically significant differences favoring sacubitril/valsartan across all 8 KCCQ domains and extending to 36 months, adding further support for the principal findings. There was a significant proportion of patients receiving sacubitril/valsartan who noted clinically meaningful changes of HRQL, despite all patients receiving the drug during the run-in phase. Patients who were hospitalized for HF within the first 8 months noted significant deteriorations in KCCQ scores; however, randomization to sacubitril/valsartan attenuated this decline in comparison to enalapril.

The KCCQ scores at baseline are much higher than those of contemporary cohorts with HF evaluating novel interventions, including ivabradine,⁴ iron replacement,¹⁷ coronary artery bypass grafting,¹⁶ cardiac rehabilitation,¹⁸ sertraline use for treatment of depression,¹⁹ and spironolactone.²⁰ This higher score likely reflects the unique features of the HRQL component of PARADIGM-HF, in that the patients did not complete instruments until the time of randomization, which occurred after the run-in phase with both enalapril and sacubitril/valsartan up to 10 weeks. It has been well established that HRQL perceptions can improve quickly in a clinical trial setting independent of any treatment effect because of potential extra attention, more intense management, and potential optimism for improved outcomes.^21,22 This higher baseline KCCQ score and prior exposure to sacubitril/valsartan possibly impacted the magnitude of improvement over time. Despite this limitation, randomization to sacubitril/valsartan still was consistently associated with improved HRQL. Despite a comparison with an active treatment rather than placebo, sacubitril/valsartan had similar magnitude of improvements of the KCCQ change scores as were seen with cardiac resynchronization therapy in the MADIT-CRT trial (Multicenter Automatic Defibrillator Implantation Trial–Cardiac Resynchronization Therapy).²³

Hospitalizations for HF are associated with increased risk for mortality.^24,25 Several studies have demonstrated worse HRQL after patients are discharged from the hospital. Sacubitril/valsartan appeared to attenuate the marked decline in HRQL in comparison to enalapril. It is unclear if the decline in HRQL among these patients with a HF hospitalization prompted the decision for admission or was a consequence of the progression of symptomatic HF. The greater magnitude of decline as a function of the time between the hospitalization and KCCQ assessment suggests that decisions that incorporate HRQL should be reassessed as patients stabilize in the outpatient setting.

There are several characteristics that were associated with impaired HRQL at baseline, including older age, female sex, higher body mass index, and comorbid illnesses. These variables are complex and may reflect the impact of symptom burden, patient perception, and disease severity on their overall sense of well-being.²⁶ Several studies have identified many of these factors in HF populations.^20,27,28 Some of these variables are modifiable and should be studied further with targeted interventions as we aim to improve HRQL. In the present study, there were multiple factors that consistently were associated with decreases in KCCQ scores by 8 months, including higher body mass index, higher NT-proBNP, older age, history of atrial fibrillation, diabetes mellitus, and higher NYHA class. This study expands on the study by Allen et al²⁷ that identified diabetes mellitus, older age, and arrhythmias as predictors of future unfavorable HRQL. Comorbidities affect self-care management and result in worse HRQL.²⁶ Exploration of these factors may identify a population who may be targeted for specific therapies directed specifically for preservation/improvements of HRQL and discussions about goals of care/advanced directives. Nevertheless, randomization to sacubitril/valsartan was an independent predictor of improved perceptions after adjusting for all other variables.

There are several limitations that should be highlighted. First, the KCCQ was initially measured after the run-in phase, which potentially resulted in a higher baseline score. Second, although the overall effect size might seem small, there are limited data assessing the clinical meaningfulness of change scores in patients who start with relatively good perceptions of their HRQL. Given the lack of validity of this instrument for some countries, we have limited experience of the impact of therapy in some parts of Asia. There may be >1 language in a given country, and thus, we cannot cross-validate responses by language. Nevertheless, KCCQ is the most well-validated instrument in HF, and the consistency of improvements of KCCQ scores across all domains suggest that patients can enjoy an improved survival and HRQL in comparison to ACE inhibitors, the long-standing standard of care.

In patients with HF with reduced ejection fraction who are well managed, HRQL specific to HF is significantly improved with sacubitril/valsartan compared with enalapril.