Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Oct 1.
Published in final edited form as: Circ Heart Fail. 2022 Sep 13;15(10):e009518. doi: 10.1161/CIRCHEARTFAILURE.122.009518

Sarcopenic Obesity is Associated with Reduced Cardiorespiratory Fitness Compared to Non-sarcopenic Obesity in Patients with Heart Failure with Reduced Ejection Fraction

Hayley E Billingsley 1,2, Marco Giuseppe Del Buono 2,3, Justin M Canada 2, Youngdeok Kim 1, Juan Ignacio Damonte 2,4, Cory R Trankle 2, Geza Halasz 5, Virginia Mihalick 2, Alessandra Vecchié 2,6, Roshanak R Markley 2, Dinesh Kadariya 2, Edoardo Bressi 2,7, Horacio Medina De Chazal 2,4, Juan Guido Chiabrando 2,4, James Mbualungu 2, Jeremy Turlington 2, Ross Arena 8,9, Benjamin W Van Tassell 2,10, Antonio Abbate 2, Salvatore Carbone 1,2
PMCID: PMC9588574  NIHMSID: NIHMS1825712  PMID: 36098058

Abstract

Background:

Sarcopenia impairs cardiorespiratory fitness (CRF) in patients with heart failure with reduced ejection fraction (HFrEF). Obesity has also been shown to impair CRF, however the effects of sarcopenia on CRF in patients with obesity and HFrEF are unknown. The aim of this analysis was to examine differences in CRF between patients with sarcopenic obesity (SO) and non-sarcopenic obesity (NSO) with HFrEF. We also assessed associations between skeletal muscle mass index (SMMI) and CRF.

Methods:

Forty patients with HFrEF and obesity underwent cardiopulmonary exercise testing to collect measures of CRF including peak oxygen consumption (VO2), circulatory power (CP), oxygen uptake efficiency slope (OUES), O2 pulse, and exercise time. Body composition was performed in all patients using bioelectrical impedance analysis to quantify fat mass index (FMI) and divide patients into SO and NSO based on SMMI cutoffs. Results presented as mean (standard deviation) or median [interquartile range] as appropriate.

Results:

Nearly half (43% [n=17]) of patients had SO. Patients with SO had a lower SMMI than those with NSO and no differences in FMI were observed between groups. Those with SO achieved a lower absolute peak VO2 (NSO, 1.62 ± 0.53 L•min−1 vs. SO, 1.27 ± 0.44 L•min−1, P=0.035), OUES (NSO, 1.92 ± 0.59 vs. SO, 1.54 ± 0.48, P=0.036), and exercise time (NSO, 549 ± 198 seconds vs. SO, 413 ± 140 seconds, P=0.021) compared to those with NSO. On multivariate analysis, SMMI remained a significant predictor of absolute peak VO2 when adjusted for age, sex, adiposity and HF severity.

Conclusions:

In patients with HFrEF and obesity, sarcopenia, defined as low SMMI, is associated with a clinically significant reduction in CRF, independent of adiposity.

Introduction

Reduced cardiorespiratory fitness (CRF), comprehensively assessed through several measurements obtained from cardiopulmonary exercise testing, is significantly compromised in patients with heart failure (HF).1 Reduced peak oxygen consumption (VO2), a primary measure of CRF, is a hallmark sign of HF.2 Low peak VO2 is associated with diminished quality of life and increased mortality risk,2, 3 with increases in peak VO2 linked to improvements in mortality in patients with HF with reduced ejection fraction (HFrEF).4 Peak VO2 is determined by arteriovenous oxygen difference and cardiac output, both of which are influenced by the quantity and quality of skeletal muscle.5–7 Patients with HF are at elevated risk of skeletal muscle abnormalities,8, 9 such as sarcopenia which is estimated to effect between 25–55% of this patient population.10

Sarcopenia, a reduction in skeletal muscle mass (SMM) with mostly preserved fat mass (FM), further impairs CRF and quality of life in patients with HF as well as reducing both exercise and functional capacity,11, 12 and thus, likely contributing to worsened clinical outcomes.13–15 Excess FM in conjunction with a reduction in SMM is referred to as sarcopenic obesity (SO), a condition with greater disability and functional impairment than sarcopenia alone.16–19 While excess FM impairs peak VO2 even further in patients with HF,20 the contribution of SO to CRF has not been explored in HFrEF, though 40–50% of patients present with obesity.21, 22 Moreover, the individual role of reduced SMM in CRF in HFrEF is not well understood.

The European Working Group for Sarcopenia in Older People (EWGSOP) acknowledges both SMM index (SMMI) estimated by bioelectrical impedance analysis (BIA), and appendicular lean mass index (ALMI) measured by dual energy x-ray absorptiometry (DEXA) as tools to quantify muscle mass in the assessment of sarcopenia.23 DEXA is presently the most widely utilized tool in the estimation of SMM and current EWGSOP cut-off values for low muscle mass were established with DEXA.23 In the clinical setting, however, BIA represents a faster, less-expensive and more portable tool for sarcopenia assessment. In patients with HFrEF, agreement between SMMI and ALMI measures has not yet been examined.

Therefore, the purposes of this study were to: 1) compare CRF and body composition parameters between patients with HFrEF with non-sarcopenic obesity (NSO) and SO 2) examine the association between SMM, body composition, and CRF and; 3) assess agreement between SMMI and ALMI in patients with HFrEF. We hypothesize that patients with SO have greater impairments in CRF compared to those with NSO, and that greater SMM is associated with greater CRF in patients with HFrEF and obesity.

Methods

This is a retrospective analysis of data from symptomatic (New York Heart Association Class II-III) patients with HFrEF (left ventricular ejection fraction [LVEF] ≤40%) and obesity (body mass index [BMI] ≥30 kg/m2) who were free of concomitant comorbidities that would interfere with the execution or interpretation of the cardiopulmonary exercise test (CPX) such as uncontrolled hypertension, orthostatic hypotension, tachy- or brady-arrhythmias, severe acute or chronic pulmonary disease, or neuromuscular disorders affecting respiration and provided a maximal effort (respiratory exchange ratio [RER] ≥ 1.0) on cardiopulmonary exercise testing. Institutional Review Board approval was obtained for this work and all subjects underwent informed consent. A limited database will be available upon request.

Body Composition

Body composition was assessed with single-frequency BIA (Quantum IV, RJL Systems). Fat mass as a percentage of body weight (FM%) was obtained and total FM in kilograms was indexed to height in meters squared to obtain FM index (FMI) (kg/m2). SMM in kilograms was also indexed to height in meters squared to obtain SMMI (kg/m2). Individuals with NSO were separated from individuals with SO with the use of SMMI cutoffs associated with increased risk of physical disability; values of ≤ 6.75 kg/m2 in women and ≤ 10.75 kg/m2 in men were considered indicative of sarcopenia.24 Lastly, phase angle, the correlation between resistance and reactance vectors measured by BIA, was also recorded. Phase angle is considered a prognostic nutritional marker and lower values are associated with an increased risk of mortality, reduced muscular strength, and lower CRF.25–27

DEXA was performed in a subset of the participants (n=15) to validate the BIA-measured body composition. We used DEXA to measure ALMI, or the lean mass of all four limbs in kilograms indexed to height in meters squared (kg/m2). SMM is the primary component of lean mass and thus lean mass is considered a surrogate for SMM.28

Cardiopulmonary Exercise Testing

Participants underwent maximal (RER ≥1.00), symptom-limited CPX with physician supervision. A conservative ramping protocol (increase of ≈0.6 Metabolic Equivalent of Task [METs] per minute)29 was performed on a metabolic cart interfaced with a treadmill (Ultima CardiO2, MGC Diagnostics). Peak VO2 was defined as the highest 10-second rolling average in the last 30 seconds of exercise1 and was expressed as an absolute value (L•min−1), relative to body weight (ml•kg−1•min−1) and relative to SMM (VO2SMM) (ml•kgSMM−1•min−1). Percent predicted relative peak VO2 (% predicted peak VO2) was calculated for each participant using the Wasserman-Hansen equations.30 Circulatory power (CP), a strong predictor of mortality,31 was calculated as the product of peak VO2 relative to body weight and peak systolic blood pressure (mmHg•ml•kg−1•min−1).31 Peak O2 pulse was calculated (ml/beat) and was also adjusted to total SMM (ml/beat/kgSMM).32 Exercise time was recorded in seconds. The oxygen uptake efficiency slope (OUES) was calculated via linear regression with the equation VO2 (L/min)= m (log10VE) +b, where m= OUES.33 The minute ventilation to carbon dioxide production (VE/VCO2) slope was also calculated with linear regression.33 Both OUES and VE/VCO2 were measured throughout the entire (100%) exercise period.33

Echocardiography

Doppler transthoracic echocardiography was performed at rest prior to exercise. Left ventricular ejection fraction (LVEF) was measured using the Simpson method. LV diastolic function was evaluated using trans-mitral diastolic flow tracings assessed with pulsed-wave Doppler from an apical four-chamber view with early (E)-wave and late (A)-wave velocity measurements, pulsed-wave tissue Doppler early diastolic mitral annular velocities (e’) averaged between the lateral and septal annulus, and calculation of the average E/e’ ratio according to the recommendations of the American Society of Echocardiography.34

Biomarkers

Venous blood samples were collected prior to exercise to measure hemoglobin, N-terminal pro-brain natriuretic peptide (NT-proBNP), a marker of myocardial wall stress, high-sensitivity C-reactive protein (hsCRP), a marker of systemic inflammation.

Quality of Life

The Kansas City Cardiomyopathy Questionnaire (KCCQ) was administered to participants in order to measure self-reported quality of life. Clinical summary score and overall summary score were calculated in a score of 0–100, with a score of 100 reflecting excellent health status and 0 reflecting very poor health status.35

Statistical Analysis

Normality of continuous variables was assessed via Shapiro-Wilk test and Q-Q plots. Descriptive statistics are reported as mean ± standard deviation (SD) or median [interquartile range] as appropriate for continuous variables and frequency (%) for categorical variables. Between-group differences (NSO vs. SO) were tested with independent sample t-test or Mann-Whitney U test for continuous variables and x2 test of independence for categorical variables. The magnitude of differences were quantified using an effect size calculated by Cohen’s d, Cramer’s V, or r as appropriate. Bivariate correlations of SMMI with CRF (i.e. absolute and relative peak VO2, OUES, exercise time, CP and O2 pulse) and other body composition parameters (i.e., phase angle and ALMI) were examined using Pearson or Spearman’s rank correlation. A series of linear regression models was constructed to further assess relative contributions of SMMI and FM% explaining variability of absolute peak VO2, the dependent variable. In model 1, we fitted simple linear regression analyses predicting absolute peak VO2 based on SMMI or FM% as a primary independent variable without covariate adjustment. In model 2, we constructed multiple linear regression extending model 1 by adjusting for study covariates including age and sex, important determinants of oxygen consumption at peak exercise36, and hsCRP and LVEF, indictors of HF severity.37 Model 3 was a fully adjusted multiple linear regression including both SMMI and FM% in the same model in addition to study covariates. For linear regression, variables hsCRP and LVEF were log-transformed due to non-normal distribution. There were three missing data points for hsCRP and LVEF variables that are assumed to be missing at random. Multiple imputation was used to create 5 complete datasets using a Markov Chain Monte Carlo method, with auxiliary variables included such as waist circumference, resting heart rate, and resting blood pressures, to improve the accuracy of the imputed values. The results for model 2 and 3 were obtained from the pooled analysis of the five multiple imputed datasets. Assumptions of linear regression including normality of residuals and multicollinearity were verified for all models. Lastly, Bland-Altman plot was created to test agreement between ALMI and SMMI. 95% limits of agreement were calculated as mean difference ±1.96 × SD of the differences and compared against the predefined maximum allowed difference, that was calculated based on the inherent imprecision of both methods using coefficient of variations: 0±1.96×CVSMMI2+CVALMI2×Mean of SMMI and ALMI.38 Acceptable agreement of ALMI and SMMI was considered as the 95% limits of agreement falling within the maximum allowed difference calculated. Simple linear regression was also performed to assess proportional bias, whether mean difference was related to the magnitude of the measurements. All analyses were performed in SPSS (v27.0). Statistical significance was considered as a P value of <0.05.

Results

Participant Characteristics

Baseline characteristics of participants (N=40) are displayed in Table 1. Majority of the participants were male (73%), Black (68%), and with a median LVEF of 33 [24, 36] %.

Table 1.

Participant Characteristics

All (N=40) NSO (n=23) SO (n=17) ES P Value
Age 57 (10) 55 (11) 60 (8) 0.464 0.155
Male (%) 29 (73%) 16 (70%) 13 (76%) 0.076 0.629
Black (%) 27 (68%) 18 (78%) 9 (53%) 0.267 0.171
ACEi/ARB/ARNI (%) 33 (83%) 19 (83%) 14 (82%) 0.003 0.983
Beta Blocker (%) 36 (90%) 20 (87%) 16 (94%) 0.118 0.455
Aldosterone Antagonists (%) 14 (35%) 12 (52%) 2 (12%) 0.419 0.008
Hydralazine (%) 11 (28%) 7 (30%) 4 (24%) 0.076 0.629
Nitrates (%) 12 (30%) 6 (26%) 6 (35%) 0.099 0.530
SGLT2i (%) 4 (10%) 4 (17%) 0 (0%) 0.287 0.070
Loop Diuretics (%) 36 (90%) 20 (87%) 16 (94%) 0.118 0.455
Weight (kg) 100.5 (13.1) 103.7 (13.5) 96.2 (11.5) 0.589 0.073
BMI (kg/m 2 ) 34.0 (31.9, 36.8) 34.9 (32.9, 37.4) 31.9 (30.9, 36.0) 0.400 0.032
LVEF (%) 33 (24, 36) 32 (24, 36) 34 (29, 36) 0.082 0.605
E/e’ 15.8 (6.6) 16.5 (6.0) 14.8 (7.4) 0.247 0.456
HsCRP (mg/L) 3.2 (1.7, 6.5) 3.0 (1.6, 5.5) 3.7 (1.8, 8.7) 0.105 0.506
NT-proBNP (pg/mL) 601 (231,1411) 548 (231,926) 1047 (179, 2804) 0.171 0.280
Hemoglobin (g/dL) 13.9 (1.7) 14.1 (1.7) 13.7 (1.8) 0.214 0.514
KCCQ Overall Summary Score 56.8 (22.3) 60.8 (22.3) 52.0 (22.1) 0.393 0.285
KCCQ Clinical Summary Score 58.8 (23.9) 65.3 (22.8) 50.9 (23.6) 0.623 0.095
SMMI (kg/m 2 ) 9.9 (2.0) 10.7 (1.7) 8.8 (1.8) 1.140 0.001
FMI (kg/m 2 ) 11.8 (3.2) 12.1 (3.6) 11.4 (2.4) 0.217 0.502
FM (%) 34.5 (7.8) 34.6 (8.6) 34.5 (7.0) 0.017 0.959
SSM/FM Ratio 0.85 (0.62, 1.12) 0.87 (0.65, 1.17) 0.84 (0.58, 0.98) 0.175 0.268
Peak VO2 (ml•kg−1•min−1) 14.4 (4.3) 15.5 (4.6) 13.0 (3.6) 0.588 0.074
Peak VO2SMM (ml•kgSMM−1•min−1) 50.3 (11.7) 50.3 (11.6) 50.3 (12.3) 0.001 0.997
% Predicted Peak VO2 (ml•kg−1•min−1) 52 (11) 55 (12) 49 (10) 0.607 0.065
Circulatory Power (mmHg•ml•kg −1 •min −1 ) 2269 (1512, 3212) 3008 (1892, 3477) 1847 (1321, 3068) 0.283 0.073
VE/VCO2 slope 35.1 (7.9) 35.0 (8.9) 35.3 (6.5) 0.033 0.919
Peak O2 Pulse (ml/beat) 11.9 (3.7) 12.9 (3.7) 10.6 (3.3) 0.641 0.052
Adjusted Peak O2 Pulse (ml/beat/kgSMM) 0.41 (0.10) 0.40 (0.08) 0.43 (0.12) 0.242 0.482
Peak SBP (mmHg) 164 (32) 168 (29) 158 (36) 0.310 0.339
Peak HR (beats/min) 123 (19) 125 (16) 121 (22) 0.188 0.561
Peak RER 1.10 (1.02, 1.17) 1.10 (1.02, 1.15) 1.15 (1.04, 1.25) 0.210 0.183
Open in a new tab

Data are presented as frequency (%), mean (standard deviation) or median (interquartile range). ES is obtained from Cramer’s d, Cramer’s v or r. P-values are differences between those with sarcopenic obesity (SO) and NSO. P-values are obtained via independent T-test, Mann-Whitney U or Chi-squared. There were missing values for KCCQ for nine participants (NSO [n=17], SO [n=14]), echocardiography in two participants (NSO [n=22], SO (n=16]), missing NT-proBNP for two participants (NSO [n=22], SO [n=16]) and one participant had missing hsCRP (NSO [n=22], SO [n=17]). Abbreviations: ES, effect sizes; CI, confidence intervals; ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; ARNI, angiotensin receptor-neprilysin inhibitor; SGLT2i, sodium-glucose transport protein 2 inhibitor; BMI, body mass index; LVEF, left ventricular ejection fraction; hsCRP, high-sensitivity C-reactive protein; NT-proBNP, N-terminal pro-brain natriuretic peptide; KCCQ, Kansas City Cardiomyopathy Questionnaire; SMMI, skeletal muscle mass index; FM, fat mass; VO2, oxygen consumption; VO2SMM, oxygen consumption indexed to skeletal muscle mass; VE/VCO2, minute ventilation to carbon dioxide production slope; kgsmm, kilograms of skeletal muscle mass; SBP, systolic blood pressure; HR, heart rate; RER, respiratory exchange ratio.

Nearly half (43%) (n=17) of the 40 patients were classified as having SO. There were no significant differences in age, race, weight, LVEF, hemoglobin, KCCQ clinical summary score, KCCQ overall summary score, NT-proBNP, and hsCRP between those with SO and NSO (Table 1). Patients with SO had a significantly lower BMI than those with NSO (NSO, 34.9 [32.9, 37.4] kg/m2 vs. SO, 31.9 [30.9, 36.0] kg/m2, ES= 0.400, P= 0.032). With the exception that patients with SO were less likely to be taking aldosterone antagonists (NSO, 12 [52%] vs. SO, 2 [12%], ES= 0.419, P= 0.008), there were no differences in medications between the two groups.

On bivariate analysis, SMMI was found to be significantly positively associated with KCCQ clinical summary score (r=0.393, P=0.029), but not KCCQ overall summary score (r= 0.297, P= 0.105), NT-proBNP (r= −0.253, P=0.126), hsCRP (r= −0.279, P=0.085) nor LVEF (r=−0.181, P=0.276).

Body Composition

Body composition analysis by BIA suggested there were no statistically significant differences in adiposity between those with NSO and SO as neither FMI nor FM% differed between the groups (Table 1). There was also no difference in SMM/FM ratio between those with NSO and SO (Table 1). Individuals with SO had a lower SMMI compared to those with NSO (NSO, 10.7 ± 1.7 vs. SO, 8.8 ± 1.8, ES = 1.14, P=0.001). Phase angle was also significantly lower in individuals with SO as compared to individuals with NSO (NSO, 6.4 ± 1.0 vs. SO, 5.0 ± 0.9, ES= 1.55, P<0.001) (Figure 1A), and on univariate analysis SMMI displayed a significant positive association with greater phase angle (r=0.790, P<0.001) (Figure 1B).

Figure 1. Sarcopenic Obesity, SMMI and Phase Angle.

Figure 1.

Open in a new tab

Participants with sarcopenic obesity had a lower phase angle, a prognostic nutritional marker, than those with non-sarcopenic obesity (Panel A). Moreover, skeletal muscle mass index (SMMI) demonstrated a positive relationship with phase angle (Panel B), suggesting that increasing SMMI may result in a greater phase angle for individuals with HFrEF and obesity.

Cardiopulmonary Exercise Testing

In general, individuals with SO had less favorable CRF on CPX than those with NSO (Table 1). When compared to patients with NSO, those with SO achieved a lower absolute peak VO2 (NSO, 1.62 ± 0.53 L•min−1 vs. SO, 1.27 ± 0.44 L•min−1, ES= 0.700, P=0.035), OUES (NSO, 1.92 ± 0.59 vs. SO, 1.54 ± 0.48, ES= 0.695, P=0.036), and exercise time (NSO, 549 ± 198 seconds vs. SO, 413 ± 140 seconds, ES= 0.771, P=0.021) (Figure 2A–C). There was also a trend towards reduced relative peak VO2 (NSO, 15.5 ± 4.6 mL•kg−1•min−1 vs. SO, 13.0 ± 3.6 mL•kg−1•min−1, ES= 0.558, P=0.074), % predicted peak VO2 (NSO, 55 ± 12% vs. SO, 49 ± 10%, ES= 0.607, P=0.065), CP (NSO, 3008 [1892–3477] mmHg•ml•kg−1•min−1 vs. SO, 1847 [1321–3068] mmHg•ml•kg−1•min−1, ES= 0.283, P=0.073), and peak O2 pulse (NSO, 12.9 ± 3.7 mL/beat vs. SO, 10.6 ± 3.3 mL/beat, ES= 0.641, P=0.052) in patients with SO. Peak VO2SMM, VE/VCO2 slope, peak RER, adjusted O2 pulse, and systolic blood pressure (SBP) did not differ between groups.

Figure 2. Sarcopenic Obesity and Cardiorespiratory Fitness.

Figure 2.

Open in a new tab

Participants with HFrEF who had concomitant sarcopenic obesity demonstrated lower cardiorespiratory fitness than their peers with non-sarcopenic obesity. Absolute peak oxygen consumption (peak VO2) (Panel A), oxygen uptake efficiency slope (OUES) (Panel B), and exercise time (Panel C) were all reduced in patients with sarcopenic obesity versus non-sarcopenic obesity.

On bivariate analysis, SMMI was found to correlate with greater absolute peak VO2, relative peak VO2, CP, OUES, O2 pulse, and exercise time (Figure 3A–F). There was no association of SMMI with % predicted peak VO2 (r= −0.037, P=0.821).

Figure 3. SMMI and Cardiorespiratory Fitness.

Figure 3.

Open in a new tab

Skeletal muscle mass index (SMMI) demonstrated positive associations with measures of cardiorespiratory fitness in patients with HFrEF and obesity. Absolute peak oxygen consumption (VO2) (Panel A), relative peak VO2 (Panel B), circulatory power (Panel C), oxygen uptake efficiency slope (Panel D), oxygen pulse (Panel E), and exercise time (Panel F) were all associated with SMMI, suggesting that increasing skeletal muscle may favorably augment cardiorespiratory fitness in patients with HFrEF and obesity.

On simple linear regression, both SMMI and FM% were significantly associated with absolute peak VO2 (Model 1, Table 2).Upon multiple linear regression, after adjustment for age, sex, hsCRP and LVEF, SMMI remained associated with absolute VO2 (b=0.096, P=0.038, semi-partial R2=5.47%) but FM% did not (b=0.004, P=0.776, semi-partial R2=0.13%) (Model 2, Table 2). When SMMI and FM% were added to the same model and adjusted for the covariates in Model 2, SMMI again remained associated with absolute VO2 (b=0.110, P=0.024, semi-partial R2=6.45%) while FM% did not (b=0.011, P=0.340, semi-partial R2=1.14%) (Model 3, Table 2).

Table 2.

Linear Regression Analyses Predicting Peak Oxygen Consumption (L•min−1) (n = 40)

Model 1* Model 2** Model 3***
b P r2 (%) b P Semi-partial r2 (%) b P Semi-partial r2 (%)
SMMI (kg/m 2 ) 0.685 <0.001 46.92% 0.096 0.038 5.47% 0.110 0.024 6.45%
FM (%) −0.414 0.008 17.14% 0.004 0.776 0.13% 0.011 0.340 1.14%
Open in a new tab

*Simple linear regression for SMMI and FM (%) separately** multiple linear regression adjusted for study covariates including age, sex, high-sensitivity C-reactive protein (hsCRP), and left ventricular ejection fraction (LVEF) for SMMI and FM (%) separately***SMMI and FM (%) are entered in the same model adjusted for study covariates. b = unstandardized regression coefficient. The results for model 2 and 3 are from the pooled analysis of five multiple imputed datasets for the missing values (missing n = 3).

Abbreviations: SMMI, skeletal muscle mass index; FM, fat mass

Validation of SMMI in HFrEF

A subset of patients (n=15) underwent both DEXA and BIA. Mean ALMI in this subset was 9.9 ± 1.3 kg/m2 and median SMMI was 10.5 ± 1.4 kg/m2. On the Bland-Altman plot, the 95% limits of agreement (−1.395, 2.430) fell within the maximum allowed difference (−3.805, 3.805), demonstrating acceptable agreement (Figure 4A). The slope of the regression line (b = 0.113; SE = 0.208; P=0.597) was not statistically significant, therefore proportionate bias was not evident. The two measures were also significantly associated (r = 0.793; Figure 4B).

Figure 4. Validating SMMI against ALMI.

Figure 4.

Open in a new tab

The 95% limits of agreement fell within maximally allowed differences on Bland-Altman analysis, demonstrating acceptable agreement between skeletal muscle mass index (SMMI) and appendicular lean mass index (ALMI) (Panel A), suggesting that bioelectrical impedance analysis (BIA) may be used in place of dual energy x-ray absorptiometry (DEXA) in clinical settings to estimate SMM in patients with HFrEF and obesity. SMMI estimated by BIA was also positively associated with ALMI measured by DEXA in patients with HFrEF and obesity (Panel B).

Discussion

The data presented herein show that in patients with HFrEF, SO is associated with a more severe reduction in CRF as compared with NSO. Adiposity was not statistically different between the two groups and when the association between SMMI and absolute peak VO2 was adjusted for FM%, it remained significant, demonstrating the independent role of SMMI in mediating CRF. We also demonstrated the association between SMMI and absolute peak VO2 remained significant when adjusted for age, sex, and biomarkers of HF severity. Absolute peak VO2, OUES, and exercise time were reduced in patients with SO versus NSO. Relative peak VO2 trended lower in patients with SO but importantly, peak VO2SMM did not differ between groups. Greater absolute peak VO2 without a difference in peak VO2SMM suggests that a greater total amount of SMM may be driving the observed differences in CRF between the two groups. This is supported by a higher SMMI and BMI in individuals with NSO (NSO, 34.9 [32.9, 37.4] kg/m2 vs SO, 31.9 [30.9, 36.0] kg/m2, P=0.032) without differences in FM% or FMI between the two groups. A reduced amount of SMM without differences in FM% or FMI may suggest a lower SMM quality, which is recognized as an even more vital determinant of adverse events than SMM in isolation.23, 39 Low muscle quality, particularly greater intermuscular fat and intermuscular fat to SMM ratio are associated with reduced peak VO2, possible mechanisms include reduced O2 delivery, mitochondrial mass, and SM oxidative metabolism.7 This is further supported by a significantly lower phase angle, a nutrition status marker and measure of electrical function of cell membranes, in individuals with SO.27 Greater phase angle was associated with increasing SMMI in our sample and is positively correlated with muscular strength and quality.27 Further characterization of muscular strength and quality should be performed with the use of functional testing such as handgrip strength, as well as advanced imaging techniques such as magnetic resonance imaging, to better elucidate the relationship between SO and CRF in individuals with HFrEF.8, 28

Although we speculate that greater SMMI leads to increased total O2 extraction and utilization contributing to more favorable CRF, modulation of cardiac output must be considered. Patients with HFrEF are characterized by reduced stroke volume and cardiac output in response to exercise,40 resulting in lower peak VO2. Increased SMM necessitates increases in blood flow, leading to greater total and central blood volume which in turn produces higher stroke volume and therefore cardiac output.5 We observed a trend towards lower CP and peak O2 pulse in those with SO, as well as the positive association of these measures with SMMI.5 Peak O2 pulse adjusted for SMM, however, did not differ between groups, suggesting that reduced total SMM may contribute to decreased stroke volume. Peak O2 pulse has been offered as a surrogate for stroke volume,41 however it should be noted that in accordance with the Fick equation, stroke volume= (VO2/HR)/arteriovenous oxygen difference, and as our CPET was noninvasive, we cannot account for the arteriovenous oxygen difference. CP is a noninvasive estimate of cardiac power and a powerful predictor of outcomes in patients with HF.31, 42 Importantly, CP is calculated as a product of peak systolic blood pressure (SBP) and relative peak VO2, and SBP did not differ between groups. This suggests that lower peak VO2 alone, not reduced blood pressure at peak exercise, contributed to lower CP in patients with SO. Further study is required, however, to determine the role SMM plays in modulating key components of peak VO2, cardiac output, and arteriovenous oxygen difference, in individuals with HFrEF and SO. This includes examining differences in CRF between individuals with SO and non-obese sarcopenia, as individuals with sarcopenia alone demonstrate lower SMM than those with SO,43 suggesting the possibility of further impairment in some measures of CRF.

While magnetic resonance imaging can accurately assess SMM and quantity and DEXA is commonly considered the surrogate for estimating SMM,23, 28 BIA represents an inexpensive, quick, and portable alternative for body composition assessment. The strong agreement and association between ALMI (DEXA) and SMMI (BIA) measures demonstrated here emphasize that BIA could be widely employed for clinical identification of lean mass abnormalities in the HF population. Moreover, BIA is able to identify the edema index, a surrogate of extracellular volume status particularly relevant in HF.44 Our findings here justify greater validation of BIA for body composition analysis in the HF population.

This analysis was exploratory and cross-sectional in nature and therefore we cannot assume that SMMI played a casual role in modulating CRF. Our sample size was also relatively small, and it is possible that an increase in type I error occurred due to multiple testing burden. Therefore, It will be important to replicate these results in a larger cohort of patients. Further study is also required to determine whether: 1) accounting for muscular strength and quality in defining SO modulates differences in CRF in individuals with HFrEF; and 2) increasing SMMI results in improvements in CRF in individuals with obesity and HFrEF. Although our sample was mostly Black and male, it is worth noting that Black men have the highest risk of being hospitalized for HFrEF, making exploration of risk factors in this population of critical importance.45 Of note, Black patients are underrepresented in heart failure clinical trials and there is a concerning trend towards decreasing representation.46, 47 Additionally, this post-hoc analysis focused on changes in SMMI, and data was not available to explore differences in cardiac muscle mass and quality in this group of patients with HFrEF. Lastly, although the EWGSOP now confirms the presence of sarcopenia through first identifying low muscular strength,23 a measure lacking in our sample, it should be noted that there is currently no standardized criteria for identifying sarcopenic obesity.8

In conclusion, in patients with obesity and HFrEF, the presence of low SMMI, or sarcopenic obesity, is associated with less favorable CRF. Randomized control trials are required to confirm that increasing SMMI results in increased CRF independent of changes in FM, and whether increases in CRF result from improvement in muscle quality, cardiac output, or both.

What is new?

This manuscript demonstrates in a well characterized cohort of patients with heart failure with reduced ejection fraction and sarcopenic obesity present with reduced cardiorespiratory fitness compared to their counterparts without sarcopenic obesity. Moreover, the results of this work shows that skeletal muscle mass index is an independent predictor for reduced cardiorespiratory fitness independent of the amount of body fat and other commonly used prognostics factors in this population.

What are the clinical implications?

The results of this manuscript highlights the importance of measuring body composition in clinical practice in patients with heart failure with reduced ejection fraction to identify those patients with reduced skeletal muscle mass in presence of excess adiposity (i.e., sarcopenic obesity), which are likely to present with a reduced cardiorespiratory fitness. Moreover, it highlights that therapies aimed at improving body composition, particularly those that can increase skeletal muscle mass, could potentially increase cardiorespiratory fitness, even in absence of changes in body fat, in patients with heart failure with reduced ejection fraction.

Sources of Funding:

This work was supported by an NIH/NHLBI Grant (R61HL139943) (Dr. Van Tassell and Dr. Abbate), Career Development Award 19CDA34660318 from the American Heart Association (Dr. Carbone), and Clinical and Translational Science Awards Program UL1TR002649 from the National Institutes of Health to Virginia Commonwealth University.

Footnotes

Disclosures: The authors have no disclosures in relation to the content of this manuscript.

References

  • 1.Guazzi M, Adams V, Conraads V, Halle M, Mezzani A, Vanhees L, Arena R, Fletcher GF, Forman DE, Kitzman DW, Lavie CJ and Myers J. Clinical Recommendations for Cardiopulmonary Exercise Testing Data Assessment in Specific Patient Populations. Circulation. 2012;126:2261–2274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Del Buono MG, Arena R, Borlaug BA, Carbone S, Canada JM, Kirkman DL, Garten R, Rodriguez-Miguelez P, Guazzi M, Lavie CJ and Abbate A. Exercise Intolerance in Patients With Heart Failure: JACC State-of-the-Art Review. J Am Coll Cardiol. 2019;73:2209–2225. [DOI] [PubMed] [Google Scholar]
  • 3.Carbone S, Kim Y, Kachur S, Billingsley H, Kenyon J, De SA, Milani RV and Lavie CJ. Peak Oxygen Consumption Achieved at the End of Cardiac Rehabilitation Predicts Long-Term Survival in Patients with Coronary Heart Disease. Eur Heart J Qual Care Clin Outcomes. 2021;8:361–367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Swank AM, Horton J, Fleg JL, Fonarow GC, Keteyian S, Goldberg L, Wolfel G, Handberg EM, Bensimhon D, Illiou MC, Vest M, Ewald G, Blackburn G, Leifer E, Cooper L and Kraus WE. Modest increase in peak VO2 is related to better clinical outcomes in chronic heart failure patients: results from heart failure and a controlled trial to investigate outcomes of exercise training. Circ Heart Fail. 2012;5:579–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Elagizi A, Kachur S, Lavie CJ, Carbone S, Pandey A, Ortega FB and Milani RV. An Overview and Update on Obesity and the Obesity Paradox in Cardiovascular Diseases. Progress in Cardiovascular Diseases. 2018;61:142–150. [DOI] [PubMed] [Google Scholar]
  • 6.Zamani P, Proto EA, Mazurek JA, Prenner SB, Margulies KB, Townsend RR, Kelly DP, Arany Z, Poole DC, Wagner PD and Chirinos JA. Peripheral Determinants of Oxygen Utilization in Heart Failure With Preserved Ejection Fraction: Central Role of Adiposity. JACC BTS. 2020;5:211–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Haykowsky MJ, Kouba EJ, Brubaker PH, Nicklas BJ, Eggebeen J and Kitzman DW. Skeletal Muscle Composition and Its Relation to Exercise Intolerance in Older Patients With Heart Failure and Preserved Ejection Fraction. The American Journal of Cardiology. 2014;113:1211–1216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Carbone S, Billingsley HE, Rodriguez-Miguelez P, Kirkman DL, Garten R, Franco RL, Lee D-C and Lavie CJ. Lean Mass Abnormalities in Heart Failure: The Role of Sarcopenia, Sarcopenic Obesity, and Cachexia. Curr Probl Cardiol. 2019;45:100417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Anker SD, Ponikowski P Fau - Varney S, Varney S Fau - Chua TP, Chua Tp Fau - Clark AL, Clark Al Fau - Webb-Peploe KM, Webb-Peploe Km Fau - Harrington D, Harrington D Fau - Kox WJ, Kox Wj Fau - Poole-Wilson PA, Poole-Wilson Pa Fau - Coats AJ and Coats AJ. Wasting as independent risk factor for mortality in chronic heart failure. Lancet. 1997;349:1050–3. [DOI] [PubMed] [Google Scholar]
  • 10.Zhang Y, Zhang J, Ni W, Yuan X, Zhang H, Li P, Xu J and Zhao Z. Sarcopenia in heart failure: a systematic review and meta-analysis. ESC Heart Fail. 2021;8:1007–1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Emami A, Saitoh M, Valentova M, Sandek A, Evertz R, Ebner N, Loncar G, Springer J, Doehner W, Lainscak M, Hasenfuß G, Anker SD and von Haehling S. Comparison of sarcopenia and cachexia in men with chronic heart failure: results from the Studies Investigating Co-morbidities Aggravating Heart Failure (SICA-HF). Eur J Heart Fail. 2018;20:1580–1587. [DOI] [PubMed] [Google Scholar]
  • 12.Fülster S, Tacke M, Sandek A, Ebner N, Tschöpe C, Doehner W, Anker SD and von Haehling S. Muscle wasting in patients with chronic heart failure: results from the studies investigating co-morbidities aggravating heart failure (SICA-HF). European Heart Journal. 2013;34:512–519. [DOI] [PubMed] [Google Scholar]
  • 13.Lavie CJ, Carbone S and Neeland IJ. Prevention and Treatment of Heart Failure: We Want to Pump You Up. JACC Cardiovascular imaging. 2021;14:216–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Selvaraj S, Kim J, Ansari Bilal A, Zhao L, Cvijic Mary E, Fronheiser M, Mohan-Rao Vanjarapu J, Kumar Anupam A, Suri A, Yenigalla S, Satija V, Ans Armghan H, Narvaez-Guerra O, Herrera-Enriquez K, Obeid Mary J, Lee Jonathan J, Jehangir Q, Seiffert Dietmar A, Car Bruce D, Gordon David A and Chirinos Julio A. Body Composition, Natriuretic Peptides, and Adverse Outcomes in Heart Failure With Preserved and Reduced Ejection Fraction. JACC: Cardiovascular Imaging. 2021;14:203–215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.von Haehling S, Garfias Macedo T, Valentova M, Anker MS, Ebner N, Bekfani T, Haarmann H, Schefold JC, Lainscak M, Cleland JGF, Doehner W, Hasenfuss G and Anker SD. Muscle wasting as an independent predictor of survival in patients with chronic heart failure. Journal of Cachexia, Sarcopenia and Muscle. 2020;11:1242–1249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Baumgartner RN, Wayne SJ, Waters DL, Janssen I, Gallagher D and Morley JE. Sarcopenic obesity predicts instrumental activities of daily living disability in the elderly. Obes Res. 2004;12:1995–2004. [DOI] [PubMed] [Google Scholar]
  • 17.Huo YR, Suriyaarachchi P, Gomez F, Curcio CL, Boersma D, Gunawardene P, Demontiero O and Duque G. Phenotype of sarcopenic obesity in older individuals with a history of falling. Archives of Gerontology and Geriatrics. 2016;65:255–259. [DOI] [PubMed] [Google Scholar]
  • 18.Öztürk ZA, Türkbeyler İH, Abiyev A, Kul S, Edizer B, Yakaryılmaz FD and Soylu G. Health-related quality of life and fall risk associated with age-related body composition changes; sarcopenia, obesity and sarcopenic obesity. Internal Medicine Journal. 2018;48:973–981. [DOI] [PubMed] [Google Scholar]
  • 19.Sanada K, Chen R, Willcox B, Ohara T, Wen A, Takenaka C and Masaki K. Association of sarcopenic obesity predicted by anthropometric measurements and 24-y all-cause mortality in elderly men: The Kuakini Honolulu Heart Program. Nutrition. 2018;46:97–102. [DOI] [PubMed] [Google Scholar]
  • 20.Carbone S, Canada JM, Buckley LF, Trankle CR, Dixon DL, Buzzetti R, Arena R, Van Tassell BW and Abbate A. Obesity Contributes to Exercise Intolerance in Heart Failure With Preserved Ejection Fraction. J Am Coll Cardiol. 2016;68:2487–2488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Horwich TB, Broderick S, Chen L, McCullough PA, Strzelczyk T, Kitzman DW, Fletcher G, Safford RE, Ewald G, Fine LJ, Ellis SJ and Fonarow GC. Relation Among Body Mass Index, Exercise Training, and Outcomes in Chronic Systolic Heart Failure. The American Journal of Cardiology. 2011;108:1754–1759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kenchaiah S, Pocock SJ, Wang D, Finn PV, Zornoff LAM, Skali H, Pfeffer MA, Yusuf S, Swedberg K, Michelson EL, Granger CB, McMurray JJV and Solomon SD. Body Mass Index and Prognosis in Patients With Chronic Heart Failure. Circulation. 2007;116:627–636. [DOI] [PubMed] [Google Scholar]
  • 23.Cruz-Jentoft AJ, Bahat G, Bauer J, Boirie Y, Bruyère O, Cederholm T, Cooper C, Landi F, Rolland Y, Sayer AA, Schneider SM, Sieber CC, Topinkova E, Vandewoude M, Visser M and Zamboni M. Sarcopenia: revised European consensus on definition and diagnosis. Age and Ageing. 2019;48:16–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Janssen I, Baumgartner RN, Ross R, Rosenberg IH and Roubenoff R. Skeletal Muscle Cutpoints Associated with Elevated Physical Disability Risk in Older Men and Women. American Journal of Epidemiology. 2004;159:413–421. [DOI] [PubMed] [Google Scholar]
  • 25.Alves FD, Souza GC, Clausell N and Biolo A. Prognostic role of phase angle in hospitalized patients with acute decompensated heart failure. Clinical Nutrition. 2016;35:1530–1534. [DOI] [PubMed] [Google Scholar]
  • 26.Garlini LM, Alves FD, Ceretta LB, Perry IS, Souza GC and Clausell NO. Phase angle and mortality: a systematic review. European Journal of Clinical Nutrition. 2019;73:495–508. [DOI] [PubMed] [Google Scholar]
  • 27.Custódio Martins P, de Lima TR, Silva AM and Santos Silva DA. Association of phase angle with muscle strength and aerobic fitness in different populations: A systematic review. Nutrition. 2022;93:111489. [DOI] [PubMed] [Google Scholar]
  • 28.Sheean P, Gonzalez MC, Prado CM, McKeever L, Hall AM and Braunschweig CA. American Society for Parenteral and Enteral Nutrition Clinical Guidelines: The Validity of Body Composition Assessment in Clinical Populations. JPEN J Parenter Enteral Nutr. 2020;44:12–43. [DOI] [PubMed] [Google Scholar]
  • 29.Arena R, Humphrey R, Peberdy MA and Madigan M. Predicting peak oxygen consumption during a conservative ramping protocol: implications for the heart failure population. J Cardiopulm Rehabil. 2003;23:183–9. [DOI] [PubMed] [Google Scholar]
  • 30.Hansen JE, Sue DY and Wasserman K. Predicted values for clinical exercise testing. Am Rev Respir Dis. 1984;129:S49–55. [DOI] [PubMed] [Google Scholar]
  • 31.Lala A, Shah KB, Lanfear DE, Thibodeau JT, Palardy M, Ambardekar AV, McNamara DM, Taddei-Peters WC, Baldwin JT, Jeffries N, Khalatbari S, Spino C, Richards B, Mann DL, Stewart GC, Aaronson KD and Mancini DM. Predictive Value of Cardiopulmonary Exercise Testing Parameters in Ambulatory Advanced Heart Failure. JACC Heart Fail. 2021;9:226–236. [DOI] [PubMed] [Google Scholar]
  • 32.Lavie CJ, Milani RV and Mehra MR. Peak exercise oxygen pulse and prognosis in chronic heart failure. The American Journal of Cardiology. 2004;93:588–593. [DOI] [PubMed] [Google Scholar]
  • 33.Arena R, Myers J, Hsu L, Peberdy MA, Pinkstaff S, Bensimhon D, Chase P, Vicenzi M and Guazzi M. The Minute Ventilation/Carbon Dioxide Production Slope is Prognostically Superior to the Oxygen Uptake Efficiency Slope. Journal of Cardiac Failure. 2007;13:462–469. [DOI] [PubMed] [Google Scholar]
  • 34.Nagueh SF, Smiseth OA, Appleton CP, Byrd BF 3rd, Dokainish H, Edvardsen T, Flachskampf FA, Gillebert TC, Klein AL, Lancellotti P, Marino P, Oh JK, Popescu BA and Waggoner AD. Recommendations for the Evaluation of Left Ventricular Diastolic Function by Echocardiography: An Update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2016;29:277–314. [DOI] [PubMed] [Google Scholar]
  • 35.Spertus JA, Jones PG, Sandhu AT and Arnold SV. Interpreting the Kansas City Cardiomyopathy Questionnaire in Clinical Trials and Clinical Care: JACC State-of-the-Art Review. Journal of the American College of Cardiology. 2020;76:2379–2390. [DOI] [PubMed] [Google Scholar]
  • 36.Peterman JE, Arena R, Myers J, Marzolini S, Ades PA, Savage PD, Lavie CJ and Kaminsky LA. Reference Standards for Cardiorespiratory Fitness by Cardiovascular Disease Category and Testing Modality: Data From FRIEND. Journal of the American Heart Association. 2021;10:e022336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Abbate A, Toldo S, Marchetti C, Kron J, Van Tassell BW and Dinarello CA. Interleukin-1 and the Inflammasome as Therapeutic Targets in Cardiovascular Disease. Circulation Research. 2020;126:1260–1280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Jensen AL and Kjelgaard-Hansen M. Method comparison in the clinical laboratory. Vet Clin Pathol. 2006;35:276–86. [DOI] [PubMed] [Google Scholar]
  • 39.Carbone S, Kirkman DL, Garten RS, Rodriguez-Miguelez P, Artero EG, Lee D-c CJ. Muscular Strength and Cardiovascular Disease: An Updated State-of-the-Art Narrative Review. Journal of Cardiopulmonary Rehabilitation and Prevention. 2020;40:302–309. [DOI] [PubMed] [Google Scholar]
  • 40.Haykowsky MJ, Tomczak CR, Scott JM, Paterson DI and Kitzman DW. Determinants of exercise intolerance in patients with heart failure and reduced or preserved ejection fraction. J Appl Physiol. 2015;119:739–744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Oliveira RB, Myers J, Araújo CGS, Arena R, Mandic S, Bensimhon D, Abella J, Chase P, Guazzi M, Brubaker P, Moore B, Kitzman D and Peberdy MA. Does peak oxygen pulse complement peak oxygen uptake in risk stratifying patients with heart failure? Am J Cardiol. 2009;104:554–558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Cohen-Solal A, Tabet JY, Logeart D, Bourgoin P, Tokmakova M and Dahan M. A non-invasively determined surrogate of cardiac power (‘circulatory power’) at peak exercise is a powerful prognostic factor in chronic heart failure. European Heart Journal. 2002;23:806–814. [DOI] [PubMed] [Google Scholar]
  • 43.Fonseca G, Dos Santos MR, de Souza FR, Takayama L, Rodrigues Pereira RM, Negrão CE and Alves MNN. Discriminating sarcopenia in overweight/obese male patients with heart failure: the influence of body mass index. ESC Heart Fail. 2020;7:84–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Marawan A, Thomas GK, Canada JM, Billingsley HE, Dixon DL, Van Tassell BW, Kadariya D, Markley R, Rotelli B, Shah KB, Kang L, Celi FS, Abbate A and Carbone S. Edema Index Predicts Cardiorespiratory Fitness in Patients With Heart Failure With Reduced Ejection Fraction and Type 2 Diabetes Mellitus. Journal of the American Heart Association. 2021;10:e018631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Virani SS, Alonso A, Aparicio HJ, Benjamin EJ, Bittencourt MS, Callaway CW, Carson AP, Chamberlain AM, Cheng S, Delling FN, Elkind MSV, Evenson KR, Ferguson JF, Gupta DK, Khan SS, Kissela BM, Knutson KL, Lee CD, Lewis TT, Liu J, Loop MS, Lutsey PL, Ma J, Mackey J, Martin SS, Matchar DB, Mussolino ME, Navaneethan SD, Perak AM, Roth GA, Samad Z, Satou GM, Schroeder EB, Shah SH, Shay CM, Stokes A, VanWagner LB, Wang N-Y and Tsao CW. Heart Disease and Stroke Statistics—2021 Update. Circulation. 2021;143:e254–e743. [DOI] [PubMed] [Google Scholar]
  • 46.Sullivan LT, Randolph T, Merrill P, Jackson LR, Egwim C, Starks MA and Thomas KL. Representation of black patients in randomized clinical trials of heart failure with reduced ejection fraction. American Heart Journal. 2018;197:43–52. [DOI] [PubMed] [Google Scholar]
  • 47.Azam TU and Colvin MM. Representation of Black patients in heart failure clinical trials. Current Opinion in Cardiology. 2021;36:329–334. [DOI] [PubMed] [Google Scholar]

RESOURCES