The global burden of obesity is increasing at an alarming rate worldwide. Approximately 2 billion people are overweight, and one third of them are obese; in the United States alone, over 200 million adults (72%) are overweight or obese, with the anticipated rate even higher in 2030.1,2 Obesity may adversely affect the cardiorespiratory fitness (CRF) of patients with cardiac disease and thus limit their ability to engage in and benefit from cardiac rehabilitation (CR) programs.3,4 Our objective was to test the effect of obesity on the improvement in CRF (i.e., change in metabolic equivalents, MET).
METHODS
We utilized data from an established retrospective cohort of 312 consecutive patients who underwent CR at the University of California, San Diego, between January 1, 2018 and March 15, 2019 (IRB #190538).5,6 Patients completed either standard (~36 hr) or intensive CR (~72 hr). Eleven participants lacked data on the change in MET (primary outcome), thus the final cohort consisted of 301 patients.
Charts were manually reviewed to collect baseline characteristics including demographics, body mass index, reason for CR, type of CR (standard/intensive), duration of CR (d), CR adherence (% completion of all sessions), smoking and alcohol status, and select co-morbidities that could affect exercise (asthma, chronic obstructive pulmonary disease, and obstructive sleep apnea [OSA]).
Cardiorespiratory fitness was assessed by MET using a treadmill exercise test before and after CR. This testing was performed according to the standard Bruce protocol, which estimates maximum MET using a standard formula and the performance of the person on a treadmill as the workload was increased.7 The changes in MET (primary outcome) were calculated by subtracting the MET documented on the first day of CR from the MET documented on the last day of CR.
For univariable comparisons of continuous and categorical data by obesity status, we used Wilcoxon rank-sum and Fisher’s exact tests, respectively. Multivariable linear regression was used to adjust for imbalances in baseline covariates. Moreover, structural equation models were employed for exploratory mediation analyses, using linear probability models for binary mediator candidates.8 P < .05 was used to determine statistical significance (R, version 3.6.1).
RESULTS
Among 301 participants (29% women), 75 (25%) subjects were obese. Compared to participants who were not obese, those who were obese had a higher body mass index, were slightly younger, and were more likely to have a history of stable angina, stable congestive heart failure, and OSA. Otherwise, both groups had similar characteristics (Table 1).
Table 1.
General Characteristics and Univariable Results
| Characteristic | Overall N = 301a | Nonobese N = 226a | Obese N = 75a | P Valueb |
|---|---|---|---|---|
| Age, yr | 67 (59, 73) | 67 (60, 73) | 64 (58, 71) | .10 |
| Sex, female | 86 (29) | 64 (28) | 22 (29) | >.9 |
| Body mass index, kg/m2 | 26.9 (24.3, 29.9) | 25.4 (23.5, 27.5) | 33.2 (31.8, 37.3) | |
| Intensive program | 100 (33) | 76 (34) | 24 (32) | .9 |
| Race | .6 | |||
| Asian | 26 (9) | 21 (9) | 5 (7) | |
| Black | 16 (5) | 10 (4) | 6 (8) | |
| Other | 38 (13) | 28 (12) | 10 (13) | |
| White | 221 (73) | 167 (74) | 54 (72) | |
| Hispanic | 27 (9) | 18 (8) | 9 (12) | .3 |
| Reason for cardiac rehabilitation | ||||
| Stable angina pectoris | 35 (12) | 19 (8) | 16 (21) | .006 |
| CABG | 44 (15) | 37 (16) | 7 (9) | .2 |
| PCI | 103 (34) | 83 (37) | 20 (27) | .12 |
| Valvular surgery | 37 (12) | 31 (14) | 6 (8.0) | .2 |
| Heart transplant/LVAD | 12 (4) | 11 (5) | 1 (1) | .3 |
| Acute MI (past 12 mo) | 41 (14) | 28 (12) | 13 (17) | .3 |
| Stable HFrEF | 54 (18) | 31 (14) | 23 (31) | .002 |
| PAD | 6 (2) | 5 (2) | 1 (1) | >.9 |
| CR-duration, d | 91 (58, 126) | 90 (58, 121) | 95 (59, 143) | .3 |
| CR adherence, % sessions | 100 (50, 100) | 100 (56, 100) | 89 (50, 100) | .3 |
| Smoking status | .7 | |||
| Current | 5 (2) | 3 (1) | 2 (3) | |
| Former | 115 (38) | 88 (39) | 27 (36) | |
| Never/Unclear | 181 (60) | 135 (60) | 46 (61) | |
| Alcohol status | .2 | |||
| Current | 131 (44) | 105 (46) | 26 (35) | |
| Former | 23 (8) | 17 (8) | 6 (8) | |
| Never/Unclear | 147 (49) | 104 (46) | 43 (57) | |
| Asthma | 34 (11) | 26 (12) | 8 (11) | >.9 |
| COPD | 37 (12) | 30 (13) | 7 (9) | .4 |
| OSA | 100 (33) | 61 (27) | 39 (52) | <.001 |
| Untreated OSA | 71 (24) | 47 (21) | 24 (32) | .06 |
| Weight, kg | ||||
| Baseline | 81 (71, 92) | 76 (67, 85) | 103 (92, 113) | <.001 |
| Follow-up (n = 254) | 80 (70, 92) | 75 (67, 84) | 100 (91, 114) | <.001 |
| Change (n = 254) | −0.6 (−2.5, 0.3)c | −0.5 (−2.1, 0.4)c | −1.4 (−5.0, 0.0)c | .02 |
| MET | ||||
| Baseline | 3.70 (3.00, 4.80) | 3.80 (3.10, 4.97) | 3.30 (2.80, 4.20) | .006 |
| Follow-up | 5.70 (4.30, 7.30) | 5.85 (4.60, 7.83) | 5.00 (3.40, 6.40) | .001 |
| Change | 2.00 (0.80, 2.90)c | 2.15 (1.00, 3.17)c | 1.30 (0.50, 2.50)c | .01 |
Abbreviations: CABG, coronary artery bypass graft; COPD, chronic obstructive pulmonary disease; CR, cardiac rehabilitation; HFrEF, heart failure with reduced ejection fraction; LVAD, left ventricular assist device; MET, metabolic equivalents; MI, myocardial infarction; OSA, obstructive sleep apnea; PAD, peripheral arterial disease; PCI, percutaneous coronary intervention.
Data are presented as median (IQR); n (%).
Wilcoxon rank-sum test; Fisher’s exact test.
Indicate P-values < .05 based on Wilcoxon signed rank tests to assess changes within groups.
Compared to patients who are nonobese, patients with obesity had significantly lower CRF at the start (P = .006) and at the end of CR (P = .001; Table 1). Thus, while both groups experienced a significant improvement in CRF during CR (Pwithin-group changes < .001), there was a significantly smaller improvement in patients who were obese vs patients who were nonobese (median [IQR] changes in MET 1.3 [0.5, 2.5] vs 2.2 [1.0, 3.2]; Pacross-groups = .01). This difference in MET changes remained significant when adjusting for imbalances in baseline characteristics (i.e., age, stable angina, congestive heart failure, and OSA; beta = −.51; 95% CI, −.99 to −.03, P = .04).
Patients with and without obesity experienced a significant reduction in body weight during CR (Pwithin-group changes < .001), but the median weight loss was significantly greater in patients who were obese vs patients who were nonobese (−1.4 [−5.0, 0] vs −0.5 [−2.1, 0.4] kg; Pacross-groups = .02; Table 1). We explored weight changes and OSA diagnosis as potential mediators, but analyses did not reach significance for either variable (Pindirect effect ≥ .19).
DISCUSSION
Our study provides important new insights based on data that demonstrate the potential impact of increasingly prevalent obesity on CR outcomes. Patients who were obese had lower CRF at baseline and at the end of CR, and they experienced less improvement in CRF with CR than patients who were nonobese. On the other hand, patients who were obese experienced a greater reduction in weight than nonobese individuals. This suggests that CR can play an important role in weight optimization for patients who were obese, which is expected to provide additional (cardiac) health benefits beyond what is attained through the improvement in CRF. Further interventional studies testing approaches to help patients who are obese achieve greater improvements in CRF via CR are needed. While weight changes and OSA status were not significant mediators in our formal mediation analyses, we speculate that these are nonetheless likely important targets for future interventions. For example, weight loss may in part reflect a reduction in muscle mass, which may antagonize efforts to improve CRF. Interventions aiming to facilitate the loss of body fat while minimizing the loss of muscle mass (e.g., special diets and exercise regimens) may augment the benefits that patients who are obese reap from CR. Similarly, poor sleep likely affects adversely the other two “pillars of health” diet and exercise.9 Thus, interventions to treat OSA (which can severely disrupt sleep, cause intermittent hypoxemia, and is highly prevalent in patients with obesity) may help improve CR outcomes for patients who are obese as well,5 especially in those with comorbid pulmonary hypertension.10
Apart from the fact that several CR programs do not have a standardized behavioral weight loss component, which has been shown to be widely effective,11 the impact of CR on patients who are obese with cardiovascular disease may also be limited by some patients with established coronary artery disease are not being referred to CR, representing a missed opportunity. For example, a survey of 427,267 US Medicare beneficiaries who survived ≥30 d after discharge in 199710 showed that only 13.9% of patients with acute myocardial infarctions and 31.0% of patients with coronary artery bypass surgery participated in CR. Expanding the inclusion criteria of CR (a class 1 recommendation for patients with cardiovascular disease) coupled with a greater focus on weight loss as part of CR programs may be a promising strategy for highly impactful secondary prevention.
CONCLUSION
Patients who are obese may experience slightly less improvements in their CRF from CR than patients who are nonobese, but their median improvement of 1.3 MET exceeds the minimal clinically important difference of 1 MET, suggesting tangible exercise benefits in addition to benefits from weight loss. The reason for the lower improvement may be lower CRF but warrants further investigation to help develop interventions that augment the CR benefits for patients who are obese.
Acknowledgments
Dr Schmickl is supported by the American Heart Association (AHA; CDA#940501), the National Institutes of Health (NIH; K23HL161336), and the American Academy of Sleep Medicine Foundation (AASMF; #277-JF-22). The content is solely the responsibility of the authors and does not necessarily represent the official views of the AHA, NIH, or AASMF. Drs Malhotra and. Taub are funded by the NIH.
Dr Schmickl also reports income from consulting for Verily, outside of the submitted work. ResMed provided a philanthropic donation to UCSD. Dr Malhotra reports income from Livanova, Eli Lilly, Jazz, and Zoll related to medical education.
REFERENCES
- 1.Ward Z, Bleich S, Cradock A, et al. Projected U.S. state-level prevalence of adult obesity and severe obesity. N Engl J Med. 2019;381 (25):2440–2450. [DOI] [PubMed] [Google Scholar]
- 2.Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. Trends in obesity among adults in the United States, 2005 to 2014. JAMA. 2016;315(21):2284–2291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.McAuley PA, Keteyian SJ, Brawner CA, et al. Exercise capacity and the obesity paradox in heart failure: The FIT (Henry Ford Exercise Testing) Project. Mayo Clin Proc. 2018;93 (6):701–708. [DOI] [PubMed] [Google Scholar]
- 4.Powell-Wiley TM, Poirier P, Burke LE, et al. American Heart Association Council on Lifestyle and Cardiometabolic Health; Council on Cardiovascular and Stroke Nursing; Council on Clinical Cardiology; Council on Epidemiology and Prevention; and Stroke Council. Obesity and cardiovascular disease: a scientific statement from the American Heart Association. Circulation. 2021;143(21):984–1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sonners C, Schmickl CN, Raphelson J, et al. The impact of obstructive sleep apnea on exercise capacity in a cardiac rehabilitation program. Sleep Breath. 2023;27(4):1269–1277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sykes AV, Sonners C, Schmickl CN, et al. The impact of underlying obstructive sleep apnea treatment on exercise capacity in patients with pulmonary hypertension undergoing a cardiac rehabilitation program. J Cardiopulm Rehabil Prev. 2023;43(3):186–191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Harb S, Bhat P, Cremer P, et al. Prognostic value of functional capacity in different exercise protocols. J Amer Heart Assoc. 2020;9(13):e015986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rijnhart JJM, Valente MJ, Smyth HL, MacKinnon DP. Statistical mediation analysis for models with a binary mediator and a binary outcome: the differences between causal and traditional mediation analysis. Prev Sci. 2023;24(3):408–418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kurnool S, McCowen KC, Bernstein NA, Malhotra A. Sleep apnea, obesity, and diabetes—an intertwined trio. Curr Diab Rep. 2023;23(7):165–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Keteyian SJ, Jackson SL, Chang A, et al. Tracking cardiac rehabilitation utilization in Medicare beneficiaries: 2017 UPDATE. J Cardiopulm Rehabil Prev. 2022;42(4):235–245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Brinkley TE, Hsu FC, Bowman BM, Addison T, Kitzman DW, Houston DK. Targeting obesity to optimize weight loss in cardiac rehabilitation: a pilot study. J Cardiopulm Rehabil Prev. 2023;43(1):39–48. [DOI] [PMC free article] [PubMed] [Google Scholar]