Question  What are the effects of sacubitril/valsartan on hospitalizations of any cause in patients with heart failure (HF)?

Findings  In this post hoc pooled analysis of the PARADIGM-HF and PARAGON-HF randomized clinical trials including 13?194 patients, sacubitril/valsartan significantly reduced the risk of all-cause hospitalization, primarily driven by lower rates of cardiac and pulmonary hospitalization. These benefits were most apparent in patients with a left ventricular ejection fraction (LVEF) below normal.

Meaning  Sacubitril/valsartan significantly reduced hospitalization for any reason in patients with HF, with benefits most apparent in patients with an LVEF below normal.

Importance  Sacubitril/valsartan is indicated to reduce the risk of cardiovascular death and heart failure (HF) hospitalizations in patients with chronic HF. However, many of these patients are older and have multiple comorbidities that increase the risk of hospitalization for causes other than HF.

Objective  To assess the effects of sacubitril/valsartan on hospitalizations of any cause across the spectrum of left ventricular ejection fraction (LVEF).

Design, Setting, and Participants  This post hoc, participant-level, pooled analysis of the PARADIGM-HF (in patients with an LVEF ≤40%) and PARAGON-HF (in patients with an LVEF ≥45%) randomized clinical trials was conducted from February 5, 2024, to April 5, 2024. Participants with chronic HF, New York Heart Association classes II through IV symptoms, and elevated natriuretic peptides were randomized to treatment with either sacubitril/valsartan or a renin-angiotensin system inhibitor (RASi)—enalapril in the PARADIGM-HF trial or valsartan in the PARAGON-HF trial.

Intervention  Sacubitril/valsartan vs RASi (enalapril or valsartan).

Main Outcomes and Measures  The effects of sacubitril/valsartan on time to first investigator-reported all-cause and cause-specific hospitalizations were examined using Cox proportional hazards models, stratified by geographic region and trial. Effect modification by LVEF as a continuous function was examined.

Results  Among 13?194 participants in the PARADIGM-HF and PARAGON-HF trials, mean (SD) patient age was 67 (11) years, 8883 patients (67.3%) were male, and mean (SD) LVEF was 40% (15%). Sacubitril/valsartan significantly reduced the risk of all-cause hospitalization (ACH) compared with RASi over a median (IQR) follow-up period of 2.5 (1.8-3.1) years (hazard ratio [HR], 0.92; 95% CI, 0.88-0.97; P?=?.002). The incidence rate of first ACH was 25 (95% CI, 24-26) per 100 patient-years in the sacubitril/valsartan arm and 27 (95% CI, 26-28) per 100 patient-years in the RASi arm. The absolute risk reduction (ARR) was 2.1 per 100 patient-years, corresponding to a number needed to treat (NNT) of 48 patient-years of treatment exposure to prevent 1 ACH. Reductions in overall hospitalizations seemed primarily driven by lower rates of cardiac and pulmonary hospitalizations with sacubitril/valsartan. Patients in the 2 treatment arms had similar rates of composite noncardiac hospitalizations. Treatment heterogeneity on ACH by LVEF was observed (P for interaction?=?.03), with benefits most apparent in patients with an LVEF less than 60% (HR, 0.91; 95% CI, 0.86-0.96), but not in patients with an LVEF of 60% or more (HR, 0.97; 95% CI, 0.86-1.09).

Conclusions and Relevance  In this post hoc pooled analysis of 13?194 patients with chronic HF in the PARADIGM-HF and PARAGON-HF randomized clinical trials, sacubitril/valsartan significantly reduced hospitalization for any reason, with benefits most apparent in patients with an LVEF below normal. This reduction appeared to be principally driven by lower rates of cardiac and pulmonary hospitalizations.

Trial Registrations  ClinicalTrials.gov Identifiers: NCT01035255 (PARADIGM-HF) and NCT01920711 (PARAGON-HF)

There has been increasing interest in understanding the effects of disease-specific medical therapies on hospitalizations of any cause, given the significant economic and population health implications. Sacubitril/valsartan is indicated to reduce the risk of cardiovascular death or hospitalization for heart failure (HF) in patients with chronic HF in many countries worldwide, with particular benefits noted among patients with a left ventricular ejection fraction (LVEF) below normal, which can be defined as 60% or above.^1-4 However, patients with HF are often older and carry high comorbid disease burden, increasing their risk of hospitalization for reasons unrelated to HF. The effects of sacubitril/valsartan on all-cause hospitalizations (ACH) have not been comprehensively described, and it is unclear whether the benefits of sacubitril/valsartan in lowering HF hospitalizations may be offset by an increase in noncardiovascular hospitalizations. In this post hoc participant-level pooled analysis of the PARADIGM-HF and PARAGON-HF randomized clinical trials, the effects of sacubitril/valsartan on ACH and the main causes of hospitalization were assessed across the LVEF spectrum.

PARADIGM-HF and PARAGON-HF were global, multicenter, randomized clinical trials testing sacubitril/valsartan against a renin-angiotensin system inhibitor (RASi)—enalapril and valsartan, respectively—in patients with HF and an LVEF of 40% or less in PARADIGM-HF or 45% or more in PARAGON-HF.^1,2 The characteristics of both trials are described in eMethods 1 in Supplement 1.

We identified predictors of ACH using Cox proportional hazards regression analyses using a forward stepwise selection approach. Treatment effects on time to first all-cause and cause-specific hospitalizations were analyzed using Cox proportional hazards models, stratified according to geographic region and trial.³ Causes of hospitalization were derived from adverse event reports, as described in eMethods 2 in Supplement 1. To address the competing risk of death, a composite end point of first ACH or all-cause death was also assessed. The number needed to treat (NNT) to prevent 1 hospitalization was determined by calculating the inverse of the absolute risk reduction (ARR) between treatment arms and was expressed as person-years of exposure to sacubitril/valsartan or RASi. Potential treatment effect modification of sacubitril/valsartan in each trial population and by LVEF as a continuous function was assessed using negative binomial regression, with LVEF modeled using restricted cubic splines (model-selected knots at 23%, 35%, and 62%). Poisson regression was used to estimate associations between continuous LVEF and incidence rates of cause-specific hospitalizations, with potential nonlinearity addressed by using polynomial terms. Statistical analyses were conducted using STATA version 18 (StataCorp). Two-sided P?<?.05 was considered statistically significant.

Among 13?194 participants in the PARADIGM-HF and PARAGON-HF randomized clinical trials, mean (SD) patient age was 67 (11) years, 8883 participants (67.3%) were male, and mean (SD) LVEF was 40% (15%). Over a median (IQR) follow-up period of 2.5 (1.8-3.1) years, among 13?194 participants with available LVEF, 6145 first ACHs occurred (2995 with sacubitril/valsartan and 3150 with RASi). Patients who were hospitalized were significantly older and had a higher comorbid disease burden (eTable in Supplement 1). Rates of ACH were modestly higher in patients with an LVEF above 40% (29 [95% CI, 28-30] per 100 patient-years) compared with those with LVEF of 40% or lower (25 [95% CI, 24-26] per 100 patient-years). Independent predictors of first ACH included older age, male sex, geographic region, higher heart rate, higher comorbidity burden, higher LVEF, and worse baseline health status (eFigure 1 in Supplement 1).

Sacubitril/valsartan significantly reduced the risk of ACH compared with RASi (hazard ratio [HR], 0.92; 95% CI, 0.88-0.97; P?=?.002) (Figure 1). The incidence rate of first ACH was 25 (95% CI, 24-26) per 100 patient-years in the sacubitril/valsartan arm and 27 (95% CI, 26-28) per 100 patient-years in the RASi arm. The ARR was 2.1 per 100 patient-years, corresponding to an NNT of 48 patient-years of treatment exposure to prevent 1 hospitalization. We observed heterogeneity by LVEF (P for interaction?=?.03), with benefits most apparent in those with an LVEF less than 60% (HR, 0.91; 95% CI, 0.86-0.96), but not in those patients with an LVEF of 60% or higher (HR, 0.97; 95% CI, 0.86-1.09) (eFigure 2 in Supplement 1). We did not find that sex modified this association between LVEF and treatment response on ACH (3-way P for interaction?=?.57) (eFigure 3 in Supplement 1).

Of the 6145 instances of ACH, 5783 (94.1%) had identifiable specific causes. Reductions in overall hospitalizations with sacubitril/valsartan seemed to be primarily driven by lower rates of cardiac and pulmonary hospitalizations (Figure 2). The risk for composite noncardiac hospitalizations was similar in the 2 treatment arms, despite a higher rate of hospitalizations for injuries, poisoning, or procedural complications in the sacubitril/valsartan arm.

There was a significant decline in the proportion of cardiac-related hospitalizations as LVEF increased (Figure 3). Conversely, there was a significant increase in the proportion of noncardiac admissions with higher LVEF that appeared driven by higher proportions of pulmonary-related hospitalizations and hospitalizations categorized as “other.”

Sacubitril/valsartan reduced the risk of the composite of ACH or all-cause mortality (HR, 0.92; 95% CI, 0.87-0.96; P?<?.001), with an ARR of 2.5 per 100 patient-years and an NNT of 40 patient-years (eFigure 4 in Supplement 1). We observed similar treatment heterogeneity by LVEF as a continuous measure, with most pronounced benefits in patients an LVEF below normal (P for interaction?=?.02).

In this post hoc pooled analysis of 13?194 patients with chronic HF from the PARADIGM-HF and PARAGON-HF randomized clinical trials, sacubitril/valsartan significantly reduced the risk of hospitalization for any reason compared with RASi; these benefits were most apparent among patients with an LVEF below normal. Despite a high burden of noncardiac hospitalization, sacubitril/valsartan significantly reduced the risk of first ACH by 8%, which translated to an NNT of 48 patient-years of treatment to prevent 1 hospitalization. These reductions appeared principally driven by reductions in cardiopulmonary hospitalizations.

The benefits on ACH appear to mirror the treatment effects on HF-specific hospitalization. No clear benefits on ACH were observed in patients with normal LVEF (≥60%). The relative proportion of hospitalizations related to cardiovascular causes was lower in the normal LVEF range⁵; modulation of a single disease-specific pathway may be challenging to modify ACH in this heterogeneous patient population. Certain adverse effects, such as hypotension, may be particularly pronounced in those with a normal LVEF, which might heighten risks of noncardiovascular hospitalizations. While sodium-glucose cotransporter-2 inhibitors have been shown to modify ACH across the LVEF spectrum,^6-9 many other pharmacological or nonpharmacological interventions have not clearly demonstrated improvements in ACH.

The mechanisms driving lower pulmonary hospitalizations with sacubitril/valsartan may be multifactorial. Sacubitril/valsartan may stabilize cardiac status and thus attenuate triggers for exacerbation of underlying pulmonary conditions, such as chronic obstructive pulmonary disease. Additionally, patients with HF have been shown to be at higher risk for pulmonary infection due to the accumulation of alveolar fluid, which may impair bacterial clearance and disrupt local defense against infection. Conversely, pulmonary infection may alter membrane permeability and thus predispose to pulmonary congestion.¹⁰ Some investigator-reported pulmonary hospitalizations may have been misclassified and could have been primarily related to worsening HF. We found, reassuringly, that the treatment benefits of sacubitril/valsartan on cardiopulmonary hospitalizations were not offset by relative increases in noncardiac hospitalizations. There was a slight imbalance, with higher rates of hospitalizations for injuries, poisoning, or procedural complications in the sacubitril/valsartan arm, which hypothetically might be related to treatment-related risks of hypotension and resultant falls or injuries. Furthermore, although sacubitril/valsartan may be associated with early changes in kidney function,¹¹ no significant between-arm difference was found regarding hospitalizations for kidney-related causes.

In patients with HF, advanced age and a high burden of comorbidities contribute to high rates of noncardiovascular hospitalizations.^12,13 Indeed, noncardiac comorbidities are closely associated with hospitalizations and death, higher costs of care, and extended length of stay.¹⁴ Interestingly, risk of ACH was subject to regional variation, with the highest rates observed in North America, which may be driven by greater health care access and use and greater comorbid disease burden.¹⁵ Reducing the global burden of hospitalizations, including those not related to cardiovascular issues, in patients with HF remains a paramount objective for various stakeholders, including patients, caregivers, clinicians, and health systems.¹⁶

This analysis has several limitations. This was a post hoc analysis, and therefore findings should be considered as hypothesis-generating. Causes for hospitalizations other than HF were not centrally adjudicated, which might have contributed to misclassification and imprecision, and not all hospitalizations had a clear identifiable cause designated by the site investigator. Analyses were not adjusted for multiple comparisons. Participants with LVEFs between 40% and 45% were not well represented in this trial program. Finally, because both trials excluded patients with advanced noncardiovascular illness or significantly limited life expectancy, it remains uncertain if similar results would be observed in less selected patients in clinical practice.

In this pooled post hoc analysis of 13?194 patients with chronic HF in the PARADIGM-HF and PARAGON-HF randomized clinical trials, sacubitril/valsartan reduced hospitalization for any reason, with benefits most apparent among patients with an LVEF below normal. This reduction appeared to be primarily driven by lower rates of cardiac and pulmonary hospitalizations, which was not offset by an increase in noncardiovascular hospitalizations.

Accepted for Publication: June 28, 2024.

Published Online: August 30, 2024. doi:10.1001/jamacardio.2024.2566

Corresponding Author: Muthiah Vaduganathan, MD, MPH, Division of Cardiovascular Medicine, Brigham and Women’s Hospital, Harvard Medical School, 75 Francis St, Boston, MA 02115 (mvaduganathan@bwh.harvard.edu).

Author Contributions: Dr Vaduganathan had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Lu, Packer, Lam, Zile, Lefkowitz, Solomon, Vaduganathan.

Acquisition, analysis, or interpretation of data: Lu, Claggett, Packer, Lam, Swedberg, Rouleau, Zile, Lefkowitz, Desai, Jhund, McMurray, Vaduganathan.

Drafting of the manuscript: Lu.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Lu, Claggett, Jhund.

Obtained funding: Packer, Lefkowitz.

Administrative, technical, or material support: Packer.

Supervision: Packer, Lam, Solomon, Vaduganathan.

Conflict of Interest Disclosures: Dr Lu reported grants from the Gottfried und Julia Bangerter-Rhyner Foundation, the SICPA Foundation, and the Société Académique Vaudoise during the conduct of the study. Dr Claggett reported consulting fees from Alnylam Pharmaceuticals, Cardurion Pharmaceuticals, Corvia Medical, Cytokinetics, Intellia Therapeutics, Rocket Pharmaceuticals, and CVRx outside the submitted work. Dr Packer reported consulting fees from 89bio, AbbVie, Actavis, ARMGO Pharma, AstraZeneca, Ardelyx, Alnylam Pharmaceuticals, Attralus, Biopeutics, Caladrius, Cytokinetics, Casana, Lilly, Moderna, Novartis, Pharmacosmos, Regeneron, Relypsa, Salamandra, and Boehringer Ingelheim outside the submitted work. Dr Lam reported research support from a Clinician Scientist award from the National Medical Research Council of Singapore, Novo Nordisk, and Roche Diagnostics; serving as a consultant or on the advisory board, steering committee, or executive committee for Alleviant Medical, Allysta Pharmaceuticals, AnaCardio, Applied Therapeutics, AstraZeneca, Bayer, Biopeutics, Boehringer Ingelheim, Boston Scientific, Bristol Myers Squibb, CardioRenal, CPC Clinical Research, Eli Lilly, Impulse Dynamics, Intellia Therapeutics, Ionis Pharmaceutical, Janssen Research & Development, Medscape/WebMD, Merck, Novartis, Novo Nordisk, Prosciento, the Quidel Corporation, Radcliffe Group, Recardio, ReCor Medical, Roche Diagnostics, Sanofi, and Siemens Healthcare Diagnostics outside the submitted work; a pending patent for PCT/SG2016/050217 for diagnosis and prognosis of chronic heart failure; a patent for an automated clinical workflow that recognizes and analyzes 2-dimensional and Doppler echo images for cardiac measurements and the diagnosis, prediction, and prognosis of heart disease; and serving as cofounder and nonexecutive director of Us2.ai. Dr Rouleau reported personal fees from Novartis, BMS, Bayer, AstraZeneca, Intellia, and Alexion during the conduct of the study. Dr Zile reported personal fees from Novartis for serving as a consultant on the steering committee during the conduct of the study. Dr Lefkowitz reported employment by and personal fees from Novartis Pharmaceuticals outside the submitted work. Dr Desai reported personal fees from Novartis during the conduct of the study; research grants to Brigham and Women’s Hospital from Novartis, Abbott, Alnylam, AstraZeneca, Intellia Therapeutics, Bayer, and Pfizer; and personal consulting fees from Abbott, Alnylam, AstraZeneca, Avidity Biopharma, Axon Therapeutics, Bayer, Biofourmis, Boston Scientific, Medpace, Medtronic, Merck, New Amsterdam, Parexel, Porter Health, Regeneron, River2Renal, Roche, scPharmaceuticals, Veristat, Verily, and Zydus outside the submitted work. Dr Jhund reported his employer (University of Glasgow) received remuneration from Novartis during the conduct of the study and from AstraZeneca, Novo Nordisk, and Bayer outside the submitted work; grants from AstraZeneca, Boehringer Ingelheim, Analog Devices, and Roche Diagnostics outside the submitted work; consulting fees from Boehringer Ingelheim; speaking fees from Intas Pharma and Sun Pharmaceuticals; and serving as the director of GCTP Ltd. Dr McMurray reported payments to his employer (Glasgow University) through a consultancy by Novartis for time spent as an executive committee member and then co–principal investigator of the ATMOSPHERE, PARADIGM-HF and PARAGON-HF, PARADISE-MI, PERSPECTIVE, and PARACHUTE-HF trials (with sacubitril/valsartan) and for travel and accomodations for meetings and presentations related to these trials, aliskiren, and sacubitril/valsartan; payments to his employer from Amgen for time spent as a steering committee member for the ATOMIC-HF, COSMIC-HF, and GALACTIC-HF trials and for travel and accomodations for meetings and other activities related to these trials; payments to his employer from Bayer for time spent as co–principal investigator of the FINEARTS trial with finerenone; payments to his employer from Cardurion for participation in a company advisory board about drug development and design of clinical trials; payment to his employer from Cytokinetics and GSK for time spent as steering committee member for the GALACTIC-HF trial and for travel and accomodations for meetings and other activities related to this trial; payments to his employer from KBP Biosciences for time spent as steering committee member for the ASCEND-D and ASCEND-ND trials using daprodustat and for travel and accomodations for meetings related to these trials; personal lecture fees from Abbott, Alkem Metabolics, AstraZeneca, Blue Ocean Scientific Solutions, Boehringer Ingelheim, Canadian Medical and Surgical Knowledge, Emcure Pharmaceuticals, Eris Lifesciences, the European Academy of CME, Hikma Pharmaceuticals, Imagica Health, Intas Pharmaceuticals, JB Chemicals & Pharmaceuticals, Lupin Pharmaceuticals, Medscape/Heart.Org, ProAdWise Communications, Radcliffe Cardiology, Sun Pharmaceuticals, The Corpus, Translation Research Group, and the Translational Medicine Academy; personal consultancy fees from Alynylam Pharmaceuticals, Bayer, BMS, Ionis Pharmaceuticals, Novartis, Regeneron Pharmaceuticals, and the River 2 Renal Corporation; serving as the director and on the data safety board of Global Clinical Trial Partners; payment to his employer from AstraZeneca for time spent as principal/co–principal investigator of the DAPA-HF, DELIVER, DETERMINE, and DAPA-Resist trials with dapagliflozin in heart failure; and time spent as a steering committee member for the DAPA-CKD trial with dapagliflozin in chronic kidney disease outside the submitted work. Dr Solomon reported grants from Novartis during the conduct of the study; grants from Alexion, Alnylam, Applied Therapeutics, AstraZeneca, Bellerophon, Bayer, BMS, Boston Scientific, Cytokinetics, Edgewise, Eidos/BridgeBio, Gossamer, GSK, Ionis, Lilly, the National Institutes of Health/National Heart, Lung, and Blood Institute, Novartis, Novo Nordisk, Respicardia, Sanofi Pasteur, Tenaya, Theracos, and US2.AI outside the submitted work; and personal fees from Abbott, Action, Akros, Alexion, Alnylam, Amgen, Arena, AstraZeneca, Bayer, BMS, Cardior, Cardurion, Corvia, Cytokinetics, GSK, Lilly, Novartis, Roche, Theracos, Quantum Genomics, Tenaya, Sanofi-Pasteur, Dinaqor, Tremeau, CellProThera, Moderna, American Regent, Sarepta, Lexicon, Anacardio, Akros, and Valo outside the submitted work. Dr Vaduganathan has received research grant support, served on advisory boards, or had speaker engagements with American Regent, Amgen, AstraZeneca, Bayer AG, Baxter Healthcare, BMS, Boehringer Ingelheim, Chiesi, Cytokinetics, Fresenius Medical Care, Idorsia Pharmaceuticals, Lexicon Pharmaceuticals, Merck, Milestone Pharmaceuticals, Novartis, Novo Nordisk, Pharmacosmos, Relypsa, Roche Diagnostics, Sanofi, and Tricog Health, and participates on clinical trial committees for studies sponsored by AstraZeneca, Galmed, Novartis, Bayer AG, Occlutech, and Impulse Dynamics outside the submitted work. No other disclosures were reported.

Funding/Support: The PARADIGM-HF and PARAGON-HF randomized clinical trials were funded by Novartis Pharmaceuticals.

Role of the Funder/Sponsor: Novartis Pharmaceuticals had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentation: This paper was presented at the European Society of Cardiology Congress 2024; August 30, 2024; London, England.

Data Sharing Statement: See Supplement 2.