Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Dorairaj Prabhakaran

General Cardiology

View

Dr. Pugazhendhi Vijayaraman

EP · Geisinger

View

Dr. Robert Schaller

EP · HUP

View

Dr. Trey Woods

General Cardiology · Emory Healthcare

View

Dr. Brahmajee Nallamothu

Interventional Cardiology · UMHS

View

Dr. Felix Yang

EP · NYU Langone

View

Dr. Jennifer Rymer

Interventional & Peripheral Cards · Duke University

View

Dr. Ami Bhatt

Adult Congenital Heart Disease · MGH

View

Dr. Gregory Marcus

EP · UCSF

View

Dr. Connor O'Brien

Cardiac Critical Care · UCSF Medical Center

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Life's Essential 8 Cardiovascular Health Score and Cardiorespiratory Fitness in the Community.

    May 3, 2024DOI: 10.1161/jaha.123.032944Journal ArticleAHA

    Written by: Sandhiya RavichandranPriya GajjarMaura E WalkerBrenton PrescottConnie W TsaoMawra JhaPrashant RaoPatricia MillerMartin G LarsonRamachandran S VasanRavi V ShahVanessa XanthakisGregory D LewisMatthew Nayor

    Find these authors on Roon

    PR
    Dr. Prashant Rao
    Sports Cardiology · BIDMC

    Methods

    Anonymized data and materials from the Framingham Heart Study (FHS) have been made publicly available through the National Institutes of Health Database of Genotypes and Phenotypes, and procedures for obtaining FHS data can be accessed at www.framinghamheartstudy.org.The FHS is a prospective epidemiological cohort study of community‐dwelling adults. The FHS began with the original cohort in 1948 and now includes their children and children's spouses (Generation 2) and their grandchildren and grandchildren's spouses (Generation 3).30 Generation 3 (N=4095) was initially enrolled in 2002 to 2005 and has attended study visits alongside the separate multiethnic Omni‐2 (N=410) and New Offspring Spouse Cohorts (N=103), as described.30, 31 These 3 cohorts attended their first study visit in 2002 to 2005 and their second study visit in 2008 to 2011. At the third study visit (2016–2019), 3117 of 3521 study visit attendees (89%) consented to undergo maximum‐effort CPET.32 Reasons for nonparticipation in CPET included home visit (rather than attending at the study center; N=35), medical conditions precluding maximum‐effort exercise (N=185), patient preference (n=124), or technical issues (n=61).33 From individuals who completed CPET, we excluded those with missing gas exchange measures (N=12), submaximum volitional effort (peak respiratory exchange ratio <1.05; N=139), and missing data at visit 3 for defining LE8 components (N=1128: N=674 for accelerometry, N=394 for dietary assessment, N=36 for nicotine exposure, N=22 for lipid values, N=1 for hemoglobin A1c, and N=1 for hypertension treatment status), resulting in a sample size of 1838 participants. For analyses including LE8 assessment at the prior study visit (visit 2), we excluded an additional 450 participants who did not attend visit 2 or with missing data for ≥1 LE8 components at visit 2. Boston University Medical Campus and Massachusetts General Hospital Institutional Review Boards approved all study protocols, and all participants provided written informed consent.Participants were encouraged to fast overnight before the study visit, and N=1796 (98%) fasted for at least 8 hours before CPET assessment. Maximum‐effort CPET was performed on a cycle ergometer (Lode, Netherlands) with breath‐by‐breath gas exchange measures (MedGraphics, St. Paul, MN). The exercise protocol consisted of 3 minutes of unloaded (“freewheel”) exercise followed by incremental ramp exercise using 1 of 2 incremental ramp protocols (15 and 25 watts/min; assigned by study staff on the basis of estimated predicted peak watts and exercise time).28 Exercise was terminated early if there were concerning ECG findings, symptomatic relative hypotension, severe hypertension, symptomatic hypoxia, or neurological compromise as determined by the supervising physician.33 Peak V̇O2 was assessed as the highest 30‐second median during the final minute of exercise. The majority of participants did not reach maximum achievable V̇O2 (as evidenced by a plateauing of the V̇O2 uptake curve), and maximum effort was determined on the basis of a peak respiratory exchange ratio ≥1.05 as previously described.33 Predicted V̇O2 was calculated using the Wasserman–Hansen equations.34 Ventilatory efficiency, as assessed by the VE/VCO2 ratio, was measured as the lowest 30s median value achieved during exercise (the “nadir”).32 Heart rate was monitored continuously during exercise with wireless ECG equipment (Mortara, Milwaukee, WI). Blood pressure was measured every 2 minutes during exercise via manual sphygmomanometer. The systolic blood pressure (SBP) response to exercise was assessed using the “ΔSBP/W slope,” which was calculated as ([peak SBP − rest SBP]/peak workload)*sex‐specific 1‐SD change in workload, as reported previously.35The LE8 score was calculated according to the definition outlined by Lloyd‐Jones et al.11 Scores of 0 to 100 were assigned for categories of each of the 8 LE8 score components (4 lifestyle behaviors and 4 health factors), and the total LE8 score was calculated as the average of the 8 individual component scores (summarized in Table 1).Life's Essential 8 Cardiovascular Health Score Definitions and DistributionsSleep healthBlood lipidsBlood glucose*We defined diabetes as either currently taking diabetes medications or having a history of diabetes. We defined no diabetes as not currently taking diabetes medication and having no history of diabetes. For 60 points, we used “No diabetes and HbA1c 5.7–7” rather than “No diabetes and HbA1c 5.7–6.4” to avoid having individuals who did not fit in any category. BMI indicates body mass index; BPs, blood pressures; DASH, Dietary Approaches to Stop Hypertension; HbA1c, hemoglobin A1c; HDL, high‐density lipoprotein; LE8, Life's Essential 8; MVPA, moderate‐to‐vigorous physical activity; and NDS, nicotine‐delivery system.Dietary intake was assessed using the Harvard semiquantitative food frequency questionnaire, which accounts for the intake of 126 dietary products ranging from no intake to ≥6 servings/day, and a Dietary Approaches to Stop Hypertension score was calculated for each participant, as described.36 Physical activity was assessed using both objective accelerometry and questionnaire‐based data in this study. There is no “gold standard” approach for assessment of physical activity in a community‐based cohort, as each method has advantages and limitations.37, 38, 39, 40 Accelerometer data provide accurate estimates of the frequency, duration, and intensity of movement but do not capture all types of exercise (eg, stationary cycling), is reflective of activity only during the time it was worn, and is not available for all individuals. On the other hand, questionnaire‐based methods are widely applicable and can accurately rank individuals into high/low categories but are limited by the potential for recall bias and low validity for assessing less structured (or incidental) activity. For the present investigation, accelerometry‐based physical activity was used as our primary assessment, and minutes spent performing moderate‐to‐vigorous physical activity per day was assessed as average number of minutes with >1535 counts/min (as described previously28) across the 8 days of wear time. In sensitivity analyses, we also assessed physical activity via questionnaires using 2 different methods. First, we calculated a physical activity index for each participant by adding the products of hours spent at each level of activity per day by a weighting factor based on the oxygen consumption required for each activity, as previously reported.41 As specific thresholds of physical activity index are not reported, we elected to categorize physical activity by septiles of the physical activity index to maintain the 7‐level LE8 scoring system. Second, we calculated minutes of moderate‐to‐vigorous physical activity per week using responses to questions about individual physical activities (detailed description in Data S1). Smoking history, vaping or other inhaled nicotine delivery system, and exposure to indoor secondhand smoke were assessed by questionnaire. Sleep health was defined using self‐reported hours of sleep per night.Height and weight were measured at each study visit and BMI was calculated by dividing weight in kilograms by height in meters squared. Lipid levels were measured from fasting blood samples at each study visit (as described42), and use of lipid‐lowering medications was based on a medication review by study staff. Blood glucose categories were assessed on the basis of serum hemoglobin A1c (assessed at each study visit using Tin‐quant Hemoglobin A1c Gen.3 kit on a Roche Cobas 501 Analyzer [Roche Diagnostics, Indianapolis, IN]) and review of medications for diabetes. Resting blood pressure was assessed at each study visit using standardized protocols. Blood pressure was measured on seated participants using a mercury column sphygmomanometer, and the average of 2 measurements was recorded for each participant.We summarized clinical characteristics and evaluated the distribution of the LE8 score and its components. Using linear models, we evaluated cross‐sectional associations of the overall LE8 score (exposure/independent variable) with CPET exercise response measures (dependent variables: peak V̇O2 [absolute, in mL/min; relative, in mL/kg per minute], percent predicted peak V̇O2, VE/VCO2 nadir, resting heart rate, and ΔSBP/W slope). Peak V̇O2 was log‐transformed for analysis, and the dependent variables were mean‐centered and standardized to 1‐unit variance to facilitate comparison of effect sizes. Multivariable models were adjusted for age and sex. We then evaluated for effect modification by age, sex, and prevalent CVD on the association of LE8 score and log(peak V̇O2) using multiplicative interaction terms. We tested for nonlinearity and threshold effects using generalized additive models.Next, we sought to evaluate the associations of each of the component scores with peak V̇O2. As the individual score components mostly did not display normal distributions, we elected to model each component as a categorical variable. We created 2‐level (optimal versus nonoptimal) and 3‐level (poor, suboptimal, and optimal) categories for each of the 8 LE8 components (Table S1). We then evaluated separate linear models relating each LE8 component (exposures, entered as “factor variables”) to log(peak V̇O2) with adjustment for age and sex. Next, all LE8 components were entered into the same multivariable model with additional adjustment for age and sex. No evidence of collinearity among LE8 components was observed.Finally, we related changes in the LE8 score over time (from study visit 2 to visit 3) with peak V̇O2. The change in LE8 score was modeled continuously (visit 3 score – visit 2 score) and in categories by dichotomizing the LE8 score at each exam cycle: (1) category 1: below median LE8 score at visit 2 and 3; (2) category 2: LE8 score higher than median at visit 2 and median or below at visit 3; (3) category 3: LE8 score median or below at visit 2 and higher than median at visit 3; (4) category 4: higher than median at both visits. The continuous LE8 score change and categories were related to log(peak V̇O2), which was measured only at visit 3, in linear models adjusted for age, sex, and visit 2 LE8 score.

    Results

    Characteristics of the FHS participants are shown in Table 2. At the baseline (third) study visit, participants had a mean age of 54±9 years, the majority (55%) were women, and mean BMI was 27.8±5.2 kg/m2, with the majority (67%) being overweight or obese. Men displayed higher levels of standard risk factors and lower LE8 scores compared with women. Mean peak V̇O2 was 21.1±6.0 mL/kg per minute in women and 26.6±6.9 mL/kg per minute in men, which was 97±20% of the predicted value. The total LE8 score was normally distributed, and the mean value was 76±12 at visit 3 (Figure S1). Average scores for individual components of LE8 ranged from 43±31 for a healthy diet to 92±19 for blood glucose (Table S2). The proportion of individuals reaching the optimal amount of moderate‐to‐vigorous physical activity per week was higher when moderate‐to‐vigorous physical activity was assessed via questionnaire rather than by accelerometer (Table S3).Characteristics of the Study SampleValues represent the mean ± for continuous variables and n (%) for categorical variables. DASH indicates Dietary Approaches to Stop Hypertension; LDL, low‐density lipoprotein; LE8, Life's Essential 8; MVPA, moderate‐to‐vigorous physical activity; SBP, systolic blood pressure; VE/VCO2, ventilatory efficiency; and V̇O2, oxygen uptake.Household income and educational status were available on 1692 participants; LE8 score at visit 2 was available on 1388 participants.In linear models adjusted for age and sex, a higher LE8 score was associated with higher peak CRF (relative and absolute peak V̇O2 and % predicted peak V̇O2). The LE8 score, age, and sex accounted for 52% of the variance explained (model R 2) in the peak V̇O2 (in mL/kg per minute), compared with 31% for age and sex alone. A higher LE8 score was also associated with better ventilatory efficiency (lower VE/VCO2 nadir), lower resting heart rate, and favorable SBP exercise response (lower ΔSBP/workload slope; Table 3). The association of LE8 score and peak V̇O2 demonstrated linearity without evidence of threshold effects (Figure 2). A 5‐point higher LE8 score (reflecting a clinically meaningful improvement in 1–2 of the LE8 components) was associated with a 6.0% greater peak V̇O2 (equivalent to 1.4 mL/kg per minute at the sample mean). We did not observe evidence of effect modification by age, sex, or CVD status on the association of LE8 score and peak V̇O2 (Table S4). In sensitivity analyses, we observed consistent associations when the physical activity metric was assessed via questionnaire instead of accelerometry (Table S5) and with additional adjustment for education and household income (Table S6).Association of LE8 Score With CPET MeasuresRegression coefficients represent the change in standardized CPET measures for a 1‐unit higher LE8 score with adjustment for age and sex.CPET, cardiopulmonary exercise testing; LE8, Life's Essential 8; SBP, systolic blood pressure; VE/VCO2, ventilatory efficiency; V̇O2, oxygen uptake.Panel (A) shows a scatterplot of the total LE8 score vs log relative peak V̇O2stratified by sex overlaid with fitted curves from a general additive model with a penalized regression spline applied to the total LE8 score with 95% confidence bounds. The general additive model was adjusted for age and sex. Panel (B) displays the mean peak V̇O2 for age‐specific percentiles of the LE8 score calculated in the study sample. Percentile values were 25th (65 for men, 69 for women); 50th (73 for men, 78 for women); 75th (79 for men, 85 for women); 100th (99 for men, 100 for women). Peak V̇O2 values and SE were calculated using linear models with log(peak V̇O2) as the dependent variable and total LE8 scores as the independent variable with adjustment for age and sex and age held at the sample mean of 54 y. FHS indicates Framingham Heart Study; LE8, Life's Essential 8; and V̇O2, oxygen uptake.Next, we sought to understand the relative contribution of each of the LE8 components to the score's relation with peak V̇O2. In age‐ and sex‐adjusted models, we observed stepwise higher peak V̇O2 values for suboptimal and optimal component scores in comparison with the poor CVH category (Figure 3A; Table S7). With mutual adjustment for all other LE8 components, we observed modest attenuation of effect estimates (Figure 3B). Suboptimal and optimal values of physical activity, nicotine exposure, BMI, and blood glucose remained associated with higher peak V̇O2, whereas the associations of diet, blood lipids, and sleep health with peak V̇O2 were no longer significant. The findings were largely similar in sensitivity analyses categorizing physical activity levels using questionnaire data (Table S8), with some attenuation of the effect estimates when the physical activity index was used to assess physical activity. In a secondary analysis, we categorized each of the LE8 components into 2 groups (optimal versus nonoptimal). When compared with nonoptimal scores, the optimal score for each component variable was associated with higher peak V̇O2 in age‐ and sex‐adjusted models, and the associations were modestly attenuated after multivariable adjustment, with diet and sleep no longer being statistically significant (Table S9).Panel (A) displays the estimated regression coefficients (circle) and standard error (bars) for the association of suboptimal and optimal values of each LE8 component (compared with poor LE8 values as a referent) with standardized log(peak V̇O2) in models adjusted for age and sex. Panel (B) shows the estimated regression coefficients from a multivariable model including all LE8 score components, age, and sex. BMI indicates body mass index; LE8, Life's Essential 8; and V̇O2, oxygen uptake.Finally, we evaluated relations of changes in LE8 score between study visits 2 and 3 (mean interval, 8±1 years) with peak V̇O2. The mean LE8 score was higher in visit 2 compared with visit 3 (Table 2; Table S2), consistent with decreasing LE8 scores with aging.10 Accounting for the earlier study visit, a 5‐unit increase in LE8 score between visits 2 and 3 was associated with a 3.7% greater peak V̇O2 at visit 3 (Table S10). We then divided individuals into 4 categories on the basis of median LE8 scores at the 2 study visits. Individuals with high LE8 scores at both study visits had the highest peak V̇O2, and individuals with low LE8 scores at both study visits had the lowest (Figure 4). Intriguingly, individuals with high LE8 scores at study visit 2 but low scores at visit 3 had similar peak V̇O2 to individuals who improved from low LE8 scores at visit 2 to high scores at visit 3.Estimated marginal means of peak V̇O2 (and 95% CIs) are plotted by category of high (median or higher) or low (lower than median) LE8 score at study visits 2 and 3. Each marginal mean value was fitted individually adjusting for age and sex. The coefficient was then exponentiated for plotting purposes. LE8 score categories were defined as participants having low LE8 scores at both exams (N=532), low LE8 at exam 3/high LE8 at exam 2 (N=152), high LE8 at exam 3/low LE8 at exam 2 (N=153), and high LE8 at both exams (N=551). LE8 indicates Life's Essential 8; and V̇O2, oxygen uptake.

    Discussion

    We evaluated the relations of the LE8 CVH score with CRF in a large community‐based sample, with several key findings. First, a higher LE8 score was associated with greater peak V̇O2 (explaining ≈20% of its variance beyond age and sex), and complementary fitness measures such as better ventilatory efficiency, lower resting heart rate, and favorable SBP response to exercise. Second, in models including all LE8 components, the highest effect estimates in relation to peak V̇O2 were observed for the physical activity, BMI, blood glucose, and nicotine exposure components, while the associations of diet, blood lipids, and sleep health were notably attenuated. Third, individuals with a low LE8 score at visit 2 but a high score at visit 3 (“improvers”) had similar peak V̇O2 values to individuals with high scores at visit 2 but low scores at visit 3, suggesting that the cumulative exposure to LE8 behaviors may be relevant to current fitness levels. Taken together, these findings delineate the relations of CVH assessed by LE8, with CRF demonstrating consistent findings across age and sex and suggesting that maintaining and improving LE8 scores over time may be important for preserving CRF.Few prior studies have investigated the relation of a composite CVH score with the gold‐standard measures of CRF. In one noteworthy report, Ross et al observed that individuals with greater estimated CRF (based on treadmill exercise time) had a 45% higher mean CVH score as assessed by Life's Simple 7 metrics.43 On the other hand, several of the LE8 components have individually been shown to correspond to CRF levels, but most prior studies have relied on estimated CRF based on exercise time, focused on mostly young adult populations or specific disease states, and have used varying assessments of lifestyle factors.26, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54 Here, we demonstrate relations of the LE8 components (mostly scored on the basis of evidence‐based thresholds) with the gold‐standard assessment of CRF. The strongest associations were observed for physical activity, BMI, nicotine exposure, and blood glucose, suggesting that maintenance of physical activity, metabolic health (optimal BMI, euglycemia), and avoiding nicotine exposure may be the most important LE8 components for preserving CRF. Indeed, the associations of physical activity with CRF were observed using objective accelerometry or questionnaire‐based physical activity data, extending the applicability of this finding across different physical activity assessment methods. On the other hand, healthy diet, sleep health, and blood lipids may be less relevant for CRF. In fact, while impaired CRF is associated with earlier development of dyslipidemia,55, 56 the converse is not well studied. Despite the importance of non–high‐density lipoprotein cholesterol for atherosclerotic CVD pathogenesis, subclinical atherosclerotic CVD (eg, coronary artery calcium) and CRF appear to provide orthogonal prognostic information, consistent with a nonlinear relation of subclinical atherosclerotic disease and CRF.57, 58Not only was the LE8 score related to CRF (peak V̇O2), but it was also associated with lower (better) ventilatory efficiency (VE/VCO2 nadir), favorable SBP response to exercise (lower ΔSBP/W slope), and better autonomic function as assessed by lower resting heart rate. Ventilatory efficiency is determined primarily by central cardiac and pulmonary vascular features, and we may therefore have expected it not to be directly associated with CVH metrics.32, 59 However, these findings are in line with our previous work linking VE/VCO2 to several standard CVD risk factors in the community and likely reflect subclinical cardiac, pulmonary, and pulmonary vascular dysfunction. We have recently shown that the ΔSBP/W slope is an integrative measure of SBP response to exercise that is directly related to higher arterial stiffness and lower cardiac and peripheral exercise performance.35 The association of the LE8 score with lower ΔSBP/W slope indicates that one mechanism by which greater CVH may lead to better CRF is through improved vascular function with exercise. Similarly, the association of the LE8 with resting HR is consistent with prior work linking lower resting HR with higher fitness and suggests the potential for autonomic function to play a role.60, 61Intriguingly, we observed that individuals with higher LE8 scores at visit 3 or visit 2 and lower LE8 scores at the other visit had similar peak V̇O2 values whether the high score was contemporaneous with CPET or not. These findings are consistent with numerous studies that have determined that the cumulative burden of risk factors, and not single time point measures, are relevant for CVH,62, 63, 64 and extend these findings to the study of CRF. Another possibility is that CRF may change in response to, or in parallel with, general CVH over time. Accordingly, our findings suggest that individuals who maintain higher LE8 scores over time are more likely to exhibit higher CRF.The present study provides a comprehensive analysis of the relations of the LE8 score and its components with CRF in a large, contemporary, community‐based sample. Nevertheless, our study has several limitations. The analyses of LE8 score and CRF measures are cross‐sectional and may be subject to reverse causation (ie, individuals with greater CRF may potentially make healthier lifestyle choices). While inclusion of longitudinal changes in LE8 scores partially addresses the potential for reverse causation, separate study designs with prospective assessment of longitudinal measures of CRF after an LE8‐based intervention would be necessary to definitively establish causality. While most of our assessments were matched well with LE8 definitions, our physical activity metric was defined using accelerometry data rather than questionnaire‐based assessments. In addition, our primary intention was to investigate the association of a contemporary CVH metric and its components with CRF and not to evaluate threshold values for each specific component in relation to CRF or to evaluate all potential lifestyle factors that might contribute to CRF. For example, the nicotine exposure metric for the LE8 score incorporates only nicotine exposure via smoking or vaping and does not include noninhaled nicotine delivery systems, which might have differing health consequences.65 Future studies may be warranted to further refine the LE8 score definitions themselves. While CRF assessment via cycle ergometer may result in slightly lower peak V̇O2 values versus other exercise methods (eg, treadmill), peak V̇O2 assessments made using different techniques are well correlated and, therefore, should result in similar associations.66 Our study sample was composed of predominantly White, middle‐aged individuals, and generalizability to other populations is therefore unknown.In conclusion, the LE8 CVH score was associated with peak V̇O2 and complementary fitness measures in a large sample of community‐dwelling individuals across categories of age, sex, and CVD status. Of the LE8 components, physical activity, BMI, nicotine exposure, and blood glucose displayed the highest effect estimates in relation to peak V̇O2 after mutual adjustment for other LE8 metrics underscoring their relevance to CRF. Our longitudinal analyses suggest that maintaining favorable CVH or improving one's CVH has a positive effect on peak V̇O2. Taken together, these study findings indicate that behavioral interventions to optimize LE8 metrics may lead to higher CRF, an important component of physiological health and resilience. Whether targeting LE8 metrics specifically may lead to improvement in CRF would require a different design with serial CRF assessment, which was not available for this study. Future studies are needed to evaluate whether interventions to target improvements in LE8 may improve or prevent poor CRF and its downstream consequences on CVD risk.

    Sources of Funding

    The Framingham Heart Study is supported by the National Heart, Lung, and Blood Institute (contracts N01‐HC‐25195, HHSN268201500001I, and 75N92019D00031). M.E.W. is supported in part by the American Heart Association (grant number 20CDA35310237), the Doris Duke Charitable Foundation (grant number 2021261), and the National Center for Advancing Translational Sciences at the National Institutes of Health through the Clinical and Translational Science Institute at Boston University (grant number 1UL1TR001430). R.S.V. was supported in part by the Evans Medical Foundation and the Jay and Louis Coffman Endowment from the Department of Medicine, Boston University School of Medicine. G.D.L. is supported by the National Institutes of Health (grant number R01‐HL131029) and the American Heart Association (grant number 15GPSGC24800006). M.N. is supported by the National Institutes of Health (grant numbers K23‐HL138260, R01‐HL156975, R01‐HL131029) and by a Career Investment Award from the Department of Medicine, Boston University School of Medicine.

    Disclosures

    In the past 12 months, Dr Shah has served as a consultant for Myokardia, Best Doctors, Amgen, and Cytokinetics. Dr Shah is a coinventor on a patent for ex‐RNAs signatures of cardiac remodeling. The spouse of Dr Shah works for UpToDate (Wolters Kluwer). Dr Lewis has research funding from Amgen, Cytokinetics, Applied Therapeutics, AstraZeneca, and Sonivie in relation to projects and clinical trials investigating exercise capacity that are distinct from this work. He has served as a scientific advisor for Pfizer, Merck, Boehringer‐Ingelheim, Novartis, American Regent, Relypsa, Cyclerion, Cytokinetics, and Amgen and receives royalties from UpToDate for scientific content authorship related to exercise physiology. Dr Nayor has received speaking honoraria from Cytokinetics. The remaining authors have no disclosures to report.

    Supporting information

    Data S1Data S2
    PRDr. Prashant Rao

    Dr. Prashant Rao

    Cardiology · Sports & Performance Cardiology
    Beth Israel Deaconess Medical Center
    prashant