Author + information
- Received April 26, 2019
- Revision received May 28, 2019
- Accepted May 29, 2019
- Published online September 11, 2019.
- Yevgeniy Khariton, MDa,∗ (, )
- Gregg C. Fonarow, MDb,
- Suzanne V. Arnold, MD, MHAa,
- Ann Hellkamp, MSc,
- Michael E. Nassif, MD, MSd,
- Puza P. Sharma, MBBSe,
- Javed Butler, MD, MPH, MBAf,
- Laine Thomas, PhDc,
- Carol I. Duffy, DOe,
- Adam D. DeVore, MD, MHSg,
- Nancy M. Albert, PhDh,
- J. Herbert Patterson, PharmDi,
- Fredonia B. Williams, EdDj,
- Kevin McCague, MAe and
- John A. Spertus, MD, MPHa
- aDepartments of Cardiology and Cardiovascular Outcomes Research, Saint Luke’s Mid America Heart Institute/University of Missouri-Kansas City, Kansas City, Missouri
- bDepartment of Cardiology, Ahmanson-UCLA Cardiomyopathy Center, Ronald Reagan University of California Los Angeles Medical Center, Los Angeles, California
- cDuke Clinical Research Institute, Durham, North Carolina
- dDepartment of Cardiology, Washington University School of Medicine in Saint Louis, Saint Louis, Missouri
- eNovartis Pharmaceuticals Corp, East Hanover, New Jersey
- fDepartment of Medicine, University of Mississippi Medical Center, Jackson, Mississippi
- gDivision of Cardiology, Department of Medicine, and the Duke Clinical Research Institute, Duke University School of Medicine, Durham, North Carolina
- hCleveland Clinic, Cleveland, Ohio
- iEshelman School of Pharmacy, University of North Carolina, Chapel Hill, North Carolina
- jMended Hearts, Huntsville, Alabama
- ↵∗Address for correspondence:
Dr. Yevgeniy Khariton, Saint Luke’s Mid America Heart Institute, 4401 Wornall Road, Kansas City, Missouri 64111.
Objectives This study sought to describe the short-term health status benefits of angiotensin-neprilysin inhibitor (ARNI) therapy in patients with heart failure and reduced ejection fraction (HFrEF).
Background Although therapy with sacubitril/valsartan, a neprilysin inhibitor, improved patients’ health status (compared with enalapril) at 8 months in the PARADIGM-HF (Prospective Comparison of ARNI with ACE inhibitor to Determine Impact on Global Mortality and Morbidity in Heart Failure) study, the early impact of ARNI on patients’ symptoms, functions, and quality of life is unknown.
Methods Health status was assessed by using the 12-item Kansas City Cardiomyopathy Questionnaire (KCCQ) in 3,918 outpatients with HFrEF and left ventricular ejection fraction ≤40% across 140 U.S. centers in the CHAMP-HF (Change the Management of Patients with Heart Failure) registry. ARNI therapy was initiated in 508 patients who were matched 1:2 to 1,016 patients who were not initiated on ARNI (no-ARNI), using a nonparsimonious time-dependent propensity score (6 sociodemographic factors, 23 clinical characteristics), prior KCCQ overall summary (KCCQ-OS) score and angiotensin-converting enzyme inhibitor/angiotensin receptor blocker status.
Results Multivariate linear regression demonstrated a greater mean improvement in KCCQ-OS in patients initiated on ARNI therapy (5.3 ± 19 vs. 2.5 ± 17.4, respectively; p < 0.001) over a median (interquartile range [IQR]) of 57 (32, 104) days. The proportions of ARNI versus no-ARNI groups with ≥10-point (large) and ≥20-point (very large) improvements in KCCQ-OS were 32.7% versus 26.9%, respectively, and 20.5% versus 12.1%, respectively, consistent with numbers needed to treat of 18 and 12, respectively.
Conclusions In routine clinical care, ARNI therapy was associated with early improvements in health status, with 20% experiencing a very large health status benefit compared with 12% who were not started on ARNI therapy. These findings support the use of ARNI to improve patients’ symptoms, functions, and quality of life.
A primary treatment goal in patients with heart failure and reduced ejection fraction (HFrEF) is to optimize their health status (i.e., symptoms, functions, and quality of life) (1). Health status is not only important from patients’ and providers’ perspectives (2) but also a strong and independent predictor of cardiovascular morbidity and mortality (3–6). Accordingly, regulatory agencies are increasingly recognizing the significance of systematically quantifying patients’ health status by using patient-reported outcomes, to assess the health status benefits of novel therapies (7–9). However, despite this era of rapidly expanding treatments, few pharmacological interventions in HFrEF have been shown to improve patients’ quality of life and reduce their symptom burden (1,10).
The PARADIGM-HF (Prospective Comparison of ARNI with ACE inhibitor to Determine Impact on Global Mortality and Morbidity in Heart Failure) trial compared sacubitril/valsartan therapy (angiotensin receptor-neprilysin inhibitor [ARNI]) with that of enalapril and demonstrated improved survival and lower hospitalization rates with ARNI. The study also showed significantly less deterioration in patients’ health status with ARNI from baseline to 8 months (11). These data led the 2016 European and North American guideline authors to recommend ARNI for patients with HFrEF or as a replacement for angiotensin converting-enzyme (ACE) inhibitor or angiotensin receptor blocker (ARB) in patients with HFrEF (12,13). However, a limitation of the PARADIGM-HF trial was that patients’ health status was not assessed before the run-in phase, precluding an assessment of the early health status benefits of ARNI. Moreover, the effectiveness of ARNI in patients’ health status in routine clinical practice is unknown.
To address these gaps in knowledge, this study used data from the CHAMP-HF (Change the Management of Patients with Heart Failure) registry (14) to examine the association between ARNI therapy and patient-reported health status. The CHAMP-HF study was a prospective, multicenter, observational registry of outpatients with HFrEF that captured serial health status outcomes by using the 12-item Kansas City Cardiomyopathy Questionnaire (KCCQ) (15), making it an ideal data source with which to describe the health status benefits of ARNI in routine clinical practice.
CHAMP-HF is a multicenter, observational registry designed to capture the care and outcomes of patients with HFrEF across heterogeneous outpatient practices in the United States (14). Eligibility criteria included a diagnosis of chronic HFrEF with left ventricular ejection fraction ≤40% and use of ≥1 oral pharmacotherapy management of heart failure. Minimum required exposure to ARNI was defined as uninterrupted ARNI therapy at any dose, either at the time of or between the most recent KCCQ assessment and the next closest assessment occurring at least 2 weeks after the initiation of ARNI treatment. Patients with limited life expectancy who were considered for advanced mechanical support (e.g., left ventricular assist device, heart transplantation) and who required hemodialysis were excluded. Study coordinators at each practice site were responsible for consecutively identifying and enrolling patients during the course of a routine outpatient HFrEF visit. For this analysis, other exclusion criteria included: 1) use of ARNI before enrollment; 2) documented contraindication or intolerance to ARNI; and 3) completion of <2 KCCQ assessments (1 before or at the initiation of ARNI and 1 at least 2 weeks after initiation) (Figure 1). The CHAMP-HF study was funded by Novartis. All participating sites received institutional review board approval, and informed consent was signed by each participant before enrollment.
Data collection and primary analysis outcome
At the time patients were enrolled, study coordinators interviewed them to collect their baseline sociodemographic and health status information and performed chart abstraction to establish a comprehensive clinical history and medication profile. On all subsequent visits (1, 3, 6, and 12 months), patient-reported data were collected either during in-person or telephone interviews. The study did not dictate or recommend any changes in therapy nor mandatory laboratory measurements.
The primary outcome for this analysis was changes in scores of the KCCQ, a well-validated disease-specific patient-reported outcome instrument that measured patients’ health status over the preceding 2 weeks that preserves the psychometric properties of the KCCQ 23-question version used in the PARADIGM-HF trial (15). The KCCQ overall summary score (KCCQ-OS) consists of 4 equally-weighted domains: physical limitation (KCCQ-PL), symptom frequency (KCCQ-SF), quality of life (KCCQ-QoL), and social limitation (KCCQ-SL), which were secondary outcomes of the study. All domains and the KCCQ-OS scores range from 0 to 100, where higher scores indicate better health status (15). Prior and extensive work has defined the clinical significance of both group mean and individual patient-level changes. Patient-level changes of <5, 5 to <10, 10 to <20, and ≥20 points represent worse to small, moderate, large, and very large improvements, respectively (3,6,16,17).
Study cohorts and defining ARNI use
Participants were allocated to receive ARNI therapy or not (no-ARNI treatment group), contingent on whether they began ARNI at any time after enrollment. Because ARNI might have been a preferred initial treatment, or patients might have switched from an ACE inhibitor/ARB to an ARNI, patients were matched directly on their pre-ARNI ACE inhibitor/ARB status (use of an ACE inhibitor/ARB within the preceding 2 weeks) (Online Figure 1). Due to the time-dependent nature of this analysis, if a no-ARNI patient completed ≥2 KCCQ assessments after initial propensity matching and before ARNI initiation, that patient was then potentially eligible to be included twice in the analysis, initially as a no-ARNI patient using the pre-ARNI KCCQ data and then as an ARNI patient using the post-ARNI KCCQ data.
Time-dependent propensity score matching
Because patients might have been prescribed ARNI at different times throughout the registry, matching was used to ensure that no-ARNI patients were identified at the same time during follow-up as those newly prescribed ARNI. All eligible participants were propensity matched at an ARNI-to-no-ARNI ratio of 1:2 based upon on a time-dependent propensity score, their most recent KCCQ-OS and ACE inhibitor/ARB status. Propensity scores were calculated using Cox proportional hazards models, where “time to ARNI” (days from enrollment to ARNI initiation) was the dependent variable and all patient-level predictors (except sociodemographic factors) were allowed to vary over time. Variables in the propensity score included 6 sociodemographic (age, sex, race, Hispanic ethnicity, household income, and employment status) and 23 clinical characteristics including medical history (atrial fibrillation, ventricular tachycardia/fibrillation, cardiac resynchronization therapy, chronic obstructive pulmonary disease, coronary artery disease, diabetes mellitus, depression/anxiety, essential hypertension, ischemic cardiomyopathy, current smoker, prior HF hospitalization, chronic kidney disease); medication use (beta-blocker, mineralocorticoid antagonist, loop diuretic agent, hydralazine, digoxin, ivabradine); physiological measurements (body mass index, systolic blood pressure, heart rate, left ventricular ejection fraction); and their most recent KCCQ scores. Due to nonmandatory reporting of laboratory data, these characteristics were not included in the model due to high missing rates. Left ventricular ejection fraction, health status data, and medications were updated throughout the analysis, whereas physiological data (e.g., vital signs) were updated every 6 months (the first value within each 6-month period was used for the matching within that period). Continuous variables were assessed for linearity of their relationship with the primary outcome of the propensity analysis by using restricted cubic splines. For categorical variables, multilevel variables (e.g., household income) were divided into binary categories (white vs. nonwhite, income <$25,000 vs. ≥$25,000, and full-time vs. part-time employment vs. not working) to simplify interpretation. The quality of propensity matching was evaluated by using standardized differences: the absolute differences in means (or proportions) divided by the average SD. A standardized difference of <0.10 (10%) reflected good covariate balance between groups (18).
Propensity models and matching were conducted separately for patients who were versus those who were not taking ACE inhibitors/ARBs, and then the cohorts were combined. To be included in the ACE inhibitor/ARB cohort, patients were required to have been treated with ACE inhibitor/ARB within 2 weeks before enrollment (otherwise, they were assigned to the no- ACE inhibitor/ARB cohort). These steps ensured comparability of newly prescribed ARNI patients and switchers from ACE inhibitor/ARB to ARNI. Medians (IQR) of time (days) from pre-match to post-match KCCQ assessments were calculated. The primary analysis included all patients (ARNI vs. no-ARNI), with comparisons of newly prescribed (ARNI vs. no-ACE inhibitor/ARB) and switching (ARNI vs. ACE inhibitor/ARB) reported as secondary analyses.
Most patient-level variables had ≤5% missing data, with the exception of household income (21%) and body mass index (8%). For covariates in the propensity model missing values were imputed using single imputation with full conditional specification. KCCQ scores were not imputed.
Linear regression and responder analysis
Changes in KCCQ scores were calculated as the difference between post-match and pre-match KCCQ scores. After comparing the propensity-matched cohorts, we compared the mean differences in the matched cohorts and further compared the association of ARNI initiation with changes in health status using 5 linear regression models of change: KCCQ-OS as the primary outcome and changes in KCCQ domain scores as secondary outcomes. The regression models used generalized estimating equations to adjust for all covariates and clustering of patients within practice sites.
Because mean KCCQ scores represented a population average effect, the distribution of changes was also described so that the proportion of patients with clinically important changes in health status could be appreciated. For each KCCQ score, where there was a significant effect in the main model, the proportion of patients (n = %) across categories of KCCQ improvement (worse to small [<5 points], moderate [≥5 to <10 points], large [10 to <20 points], and very large [≥20 points] changes) were calculated (15). When statistically significant, the number needed to treat (NNT) was calculated by first finding the absolute risk reduction as [ARR = PA − PN], where PA and PN represent the proportion of ARNI and no-ARNI patients, respectively, with a given level of health status improvement and 95% confidence intervals (CIs) were calculated as ARR ± 1.96 × SE, where SE is Wald SE. The reciprocals of these estimates were used to calculate the NNTs.
All estimates were reported using 95% CI and a p value of ≤0.05 was considered statistically significant. All analyses were performed using SAS version 14.3 software (SAS Institute, Cary, North Carolina).
At the time of this analysis, 3,918 patients across 140 sites were enrolled in the CHAMP-HF study between 2015 and 2017, after excluding those patients who had been prescribed ARNI before enrollment (n = 608); or who were missing demographic, medical history, or medication data (n = 198); or who had documented contraindications or intolerance to ARNI (n = 119); and those who had missing KCCQ data before or after ARNI initiation (n = 183) (Figure 1). Of these, 580 were newly prescribed ARNI and 508 (88%) were successfully matched with 1,016 patients who had not begun ARNI therapy at the same point during follow-up (Table 1, Online Figure 2). For those patients who began treatment with ARNI, 267 (53%) had been taking ACE inhibitor/ARB therapy and 241 (47%) had not, within 2 weeks of ARNI initiation. The matched groups were well balanced in sociodemographic and clinical characteristics, with the exception of chronic kidney disease (standardized difference: 11.7%), beta-blocker (14.1%), mineralocorticoid antagonists (MRA) (15.3%), and loop diuretic treatment (10.2%). Similar comparability between groups was observed within the ACE inhibitor/ARB (267 ARNI and 534 ACE inhibitor/ARB) and no-ACE inhibitor/ARB (241 ARNI and 482 no-ACE inhibitor/ARB) cohorts (Online Tables 1 and 2, Online Figures 3 and 4).
Characteristics of the analytic cohort are presented in Online Table 3. Of the total sample, 42.0% were 40 to 64 years of age and 44.4% were 65 to 80 years of age; and 29.2% were women, and 74.2% were white. Cardiac and non-cardiac comorbidities were common, with 64.6% having coronary artery disease, 27.9% having chronic obstructive lung disease/asthma, 33.7% having depression/anxiety, 42.7% having diabetes, and 38.4% having atrial fibrillation/flutter. Most patients (62.0%) had no HF hospitalizations in the year before enrollment, reflecting a relatively stable HF cohort. Mean systolic blood pressure and left ventricular ejection fraction in the cohort were 120 ± 18 mm Hg and 28 ± 8%, respectively. Use of HF therapies in the matched cohorts was high, including beta-blockers in 94.3%, MRAs in 41.4%, and loop diuretic agents in 69.5%. A minority were prescribed digoxin (14.6%) and ivabradine (1.6%). The average KCCQ-OS score at enrollment in the CHAMP-HF trial was 63.6 ± 23.7, corresponding to New York Heart Association functional class II.
Association between use of ARNI and changes in health status
Improvements in KCCQ score from the last pre-match to the first post-match health status assessment were observed over a median (IQR) of 57 (32, 104) days. Overall, ARNI patients experienced an average 5.3 ± 18.6-point improvement in the KCCQ-OS compared with 2.5 ± 17.4 points for their no-ARNI counterparts (differences in mean changes of 3.2 [95% CI: 1.5 to 4.9); adjusted group-level differences with regression modeling of 2.9 points (95% CI: 1.14 to 4.6; p < 0.001) (Table 2). This was largely driven by patients’ improvements in the KCCQ-PL (4.8 ± 24.8 points vs. 2.0 ± 22.2 points, respectively) and KCCQ-QoL (6.4 ± 23.9 points vs. 2.7 ± 24.1 points, respectively) domains. Similar findings were observed in both the de novo ARNI initiates (mean difference of 2.9 points [95% CI: 0.3 to 5.5; p = 0.028]) and in those who switched between ACE inhibitor/ARB and ARNI therapy (mean difference of 2.7 points [95% CI: 0.4 to 5.0; p = 0.024]) (Online Tables 4a, 4b, 5a, and 5b).
The proportion of patients experiencing at least moderate, large, and very large health status improvements according to changes in KCCQ-OS scores were calculated (Table 3). Overall, 43.7% of ARNI patients (vs. 39.8% of the no-ARNI patients), 32.7% of ARNI patients (vs. 26.9% of no-ARNI patients), and 20.5% of ARNI patients (vs. 12.1% of no-ARNI patients) experienced at least a moderate (≥5-point increase), large (≥10-point increase), and very large (≥20-point increase) health status benefit, respectively (Online Figures 5 and 6). Based on these results, it was found that for every 18 patients (95% CI: 10 to 111) and 12 patients (95% CI: 9 to 24) started on ARNI therapy, 1 more patient would be expected to have a large and a very large health status benefit, respectively, compared with the patient who had not been started on an ARNI therapy (Tables 3 and 4, Central Illustration). Moreover, fewer patients were likely to experience minimal to no improvement or deterioration in their health status (absolute difference of 3.9% between groups), translating to a NNT of 26 to prevent such an outcome.
Consistent findings were observed among ARNI patients who switched from ACE inhibitor/ARB therapy and among patients not previously taking an ACE inhibitor/ARB therapy (Online Figures 7 and 8). The proportions experiencing very large clinical improvements in the ACE inhibitor/ARB cohort were 19.1% versus 10.5%, respectively, suggesting that switching to an ARNI resulted in an NNT of 12 (95% CI: 8 to 31) for a patient to experience a very large improvement in their health status. Among patients not previously taking an ACE inhibitor/ARB, initiation of the ARNI therapy was associated with a 22% (vs. 13.9%, respectively) chance of a very large improvement, corresponding to an NNT of 13 (95% CI: 8 to 50) (Table 4).
Although a principal goal of HF management is to alleviate patient symptoms and improve health status, few pharmacological therapies have been shown to reliably reduce symptoms, improve functions, and enhance quality of life. This study examined the early health status benefits of instituting an ARNI in a multicenter observational registry of outpatients with chronic HFrEF. Independent of prior treatment with ACE inhibitor/ARB, it was found that patients prescribed ARNI experienced early and robust improvements in disease-specific health status as measured by the KCCQ. These findings represent the first real-world evidence describing the potential health status benefits of ARNI in patients with HFrEF.
These findings extend the clinical trial data from the PARADIGM-HF trial (11,19,20), which described significant reductions in cardiovascular mortality and HFrEF hospitalization and greater preservation of health status over 8 months with ARNI. A notable limitation of PARADIGM-HF arose from the use of a run-in phase during which all patients underwent dosages of enalapril followed by dose escalation of ARNI before baseline KCCQ assessment. As such, early improvements in health status associated with the use of ARNI could not be calculated. To address this gap, the CHAMP-HF registry offers a unique perspective by capturing patients’ true baseline health status before ARNI initiation, thereby providing a more accurate assessment of early health status changes over time after ARNI initiation. Finding a clinically significant mean increase in KCCQ-OS scores of 5.3 points in the patients treated with ARNI suggests that the early benefits of treatment may have been missed in the PARADIGM-HF study but that, given the minimal further decrease in scores over time in the PARADIGM-HF, these benefits were likely sustained over time. Confirmation of these sustained benefits from the CHAMP-HF study should be explored as more follow-up in this registry accrues.
In comparing mean differences between groups, it can often be difficult to interpret the clinical significance of changes. Accordingly, the distribution of changes against well-established thresholds of clinically important changes in the KCCQ were also examined (15,17). By comparing the proportions of patients according to their magnitude of changes, a “responder” analysis could also be conducted and an estimate of the number of patients who would need to be treated for one to have a clinically important change in their health status. The distribution of changes in KCCQ scores for those started on ARNI was shifted, such that a substantial proportion of patients experienced very large clinical improvements in their health status. It was found that an NNT of 12, meaning for every 12 patients treated with ARNI, compared with similar patients not treated, 1 patient would experience a very large improvement in their health status, regardless of whether ARNI was used as a substitute for ACE inhibitor/ARB treatment or as a de novo therapy. This change of >20 points is comparable to the mean health status benefits after transcatheter aortic valve replacement or insertion of a left ventricular assist device (21,22). Understanding whether or not there are particular patient profiles associated with such large health status benefits from ARNI is an important area for future investigation.
First, the CHAMP-HF study was an observational registry and thereby susceptible to bias as a result of unmeasured confounding. In particular, the role of placebo effect due to open-label ARNI use cannot be excluded. Second, patients’ laboratory values were unable to be matched due to missing laboratory data. Third, although a primary aim of the CHAMP-HF study was to recruit participating sites with diverse treatment backgrounds (e.g., physician specialty, patient population), the present findings may not be generalizable throughout the United States or to other countries. In particular, although rates of medical therapy for heart failure were (at least) similar to those observed in prior HFrEF registries and HFrEF clinical trials (23–25), rates of MRA, digoxin, and diuretic agents were lower than those in the PARADIGM-HF registry (25,26). Finally, these results apply only to outpatients with HFrEF, and the benefits of ARNI for other outcomes and in other heart failure populations require further investigation.
In an observational registry of outpatients with HFrEF across the United States, this study found an association between ARNI initiation and early improvements in patient-reported health status. These improvements were largely driven by a substantially larger proportion of patients treated with ARNI who experienced a very large health status benefit shortly after ARNI initiation. These data supplement the benefits of ARNI therapy in reducing mortality and hospitalizations and underscore the need for future research to better identify patients who are most likely to benefit from ARNI.
COMPETENCY IN MEDICAL KNOWLEDGE: Initiation of ARNI is associated with early, clinically meaningful health status (symptoms, functions, and quality of life) benefits in outpatients with heart failure and reduced ejection fraction.
TRANSLATIONAL OUTLOOK: This work extends prior insights into the clinical benefits of ARNI initiation and demonstrates its early health status benefits in a real-world HFrEF population.
The CHAMP-HF (Change the Management of Patients with Heart Failure) and the present study were funded by the Novartis Pharmaceuticals Corp. Drs. Khariton and Nassif are supported by U.S. National Heart, Lung, and Blood Institutes award T32HL110837. The content is solely the responsibility of the authors and does not necessarily represent the official views of the U.S. National Institutes of Health (NIH). Dr. Khariton had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Dr. Sharma was an employee of Novartis at the time of the development and review of this paper. Dr. Spertus has received research support from National Institute of Health (NIH), American College of Cardiology Foundation, Bayer, Novartis, and Abbott Vascular; and he serves on a Scientific Advisory Board for United Healthcare; and is a consultant for Novartis, V-Wave, AstraZeneca, Janssen, Corvia, and Bayer; and holds patent rights to the Kansas City Cardiomyopathy Questionnaire; and he holds equity in Health Outcomes Sciences. Dr. Thomas has received research support from Novartis Pharmaceuticals Corp. Dr. Fonarow has received research support from NIH; and is a consultant for Abbott, Amgen, Bayer, Janssen, Medtronic, and Novartis. Dr. DeVore has received research support from the American Heart Association, Amgen, NIH, and Novartis; and is a consultant for Novartis. Dr. Butler has received research support from NIH and European Union; and is a consultant for Amgen, Bayer, Boehringer Ingelheim, Cardiocell, CVRx, Gilead, Janssen, Medtronic, Merck, Novartis, Relypsa, and ZS Pharma. Dr. Albert is a consultant for Novartis, AstraZeneca, and Boston Scientific; and has received honoraria from Novartis. Dr. J. Herbert Patterson reports consulting and research support for Novartis. Drs. Duffy and McCague are employees of Novartis. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- angiotensin-neprilysin inhibitor
- heart failure with reduced ejection fraction
- Kansas City Cardiomyopathy Questionnaire
- Received April 26, 2019.
- Revision received May 28, 2019.
- Accepted May 29, 2019.
- Rumsfeld J.S.,
- Alexander K.P.,
- Goff D.C. Jr..,
- et al.
- Pokharel Y.,
- Khariton Y.,
- Tang Y.,
- et al.
- Kosiborod M.,
- Soto G.E.,
- Jones P.G.,
- et al.
- Soto G.E.,
- Jones P.,
- Weintraub W.S.,
- Krumholz H.M.,
- Spertus J.A.
- Committee on Quality of Health Care in America Institute of Medicine
- ↵Kansas City Cardiomyopathy Questionnaire (KCCQ) | FDA Voice. Blogs.fda.gov. Available at:. https://blogs.fda.gov/fdavoice/index.php/tag/kansas-city-cardiomyopathy-questionnaire-kccq/. 762018.
- National Institutes of Health
- Ponikowski P.,
- Voors A.A.,
- Anker S.D.,
- et al.
- Yancy C.W.,
- Jessup M.,
- Bozkurt B.,
- et al.
- DeVore A.D.,
- Thomas L.,
- Albert N.M.,
- et al.
- Spertus J.A.,
- Jones P.G.
- Dreyer R.P.,
- Jones P.G.,
- Kutty S.,
- Spertus J.A.
- Lewis E.F.,
- Claggett B.L.,
- McMurray J.J.V.,
- et al.
- Chandra A.,
- Lewis E.F.,
- Claggett B.L.,
- et al.
- Osnabrugge R.L.,
- Arnold S.V.,
- Reynolds M.R.,
- et al.
- Rogers J.G.,
- Aaronson K.D.,
- Boyle A.J.,
- et al.
- Patel D.B.,
- Shah R.M.,
- Bhatt D.L.,
- et al.
- Fonarow G.C.,
- Albert N.M.,
- Curtis A.B.,
- et al.