Exercise capacity is an important index to assess exercise stress in cardiac rehabilitation patients. Exercise capacity is generally expressed in units of metabolic equivalents (METs) calculated during cardiopulmonary exercise testing (CPX) using a ventilatory expired gas analysis system. While guidelines of academic association recommend conducting CPX to calculate METs [1,2], CPX induces heavy exercise stress and requires the use of equipment such as a cycle ergometer and treadmill, which is not always feasible in frail elderly patients or those with unstable circulation dynamics. The present study explored the possibility of predicting METs during exercise from heart rate (HR) of patients with cardiovascular disease.
A total of 251 patients aged ≥18 years underwent CPX at the cardiac rehabilitation center of our hospital on an outpatient basis between April 2012 and August 2017. Of these, 243 patients with ischemic heart disease, valvular disease, and/or heart failure were included in this study. Patients with atrial fibrillation, implanted cardiac pacemakers or ventricular assist devices, and a history of orthopedic conditions or cerebrovascular diseases were excluded. Background factors including age, sex, body mass index (BMI), diagnosis, left ventricular ejection fraction (LVEF), brain natriuretic peptide (BNP), and medication use (e.g., β-blockers, angiotensin converting enzyme inhibitors, angiotensin II receptor blockers (ACEIs/ARBs), and diuretics) were extracted from clinical records as parameters. CPX was conducted using a cycle ergometer (Strength Ergo 8; Mitsubishi Electric Engineering Co., Ltd., Tokyo, Japan) with a ventilatory expired gas analysis system (Cpex-1; Inter Reha, Tokyo, Japan). Exercise intensity was increased by 10 or 20 W/min. HRrest was defined as the lowest HR recorded while seated on the cycle ergometer for 3 min, and HRs measured every minute after the start of CPX were used as HRabsolute. METs corresponding to HRrest and HRabsolute were examined.
The results of CPX from 243 patients yielded 2048 data sets for analysis. The mean age was 58.7 ± 12.6 years, 74.5% were men, and the median LVEF and BNP were 57.6 ± 15.5% and 75.5 (interquartile range, 36.3–152.6) pg/mL, respectively; 66.8% of patients used β-blockers (Table 1). Relationships between HR parameters and METs are shown in Fig. 1. Significant positive correlations were observed between METs during exercise and HRabsolute (r = 0.740), HRnet (HRabsolute/HRrest) (r = 0.799), and HRindex (HRabsolute − HRrest) (r = 0.764) (p < 0.001, respectively). A simple linear regression analysis was performed with independent variables of each HR parameter, yielding the following prediction formulae: METs = 0.05 × HRabsolute − 1.0 (adjusted R2 = 0.584) for HRabsolute; METs = 0.05 × HRnet + 2.1 (adjusted R2 = 0.638) for HRnet; and METs = 3.2 × HRindex − 0.7 (adjusted R2 = 0.548) for HRindex (p < 0.001, respectively). Subgroup analyses were performed for age (young vs. elderly), sex (men vs. women), BMI (<25 kg/m2 vs. ≥25 kg/m2), exercise type (10 W vs. 20 W), and the use of β-blockers. Although the regression equations were different for HRabsolute and HRindex in each subgroup analysis, the regression equation for HRnet had the minimum variance for the slope and intercept, suggesting that METs could be predicted using 0.05 × HRnet + 2.
Table 1.
Clinical characteristics.
| n = 243 | (Minimum–maximal) | |
|---|---|---|
| Age (years) | 58.7 ± 12.6 | (26–79) |
| Male (%) | 74.5 | |
| BMI (kg/m2) | 23.8 ± 3.90 | (16.4–38.9) |
| LVEF (%) | 57.6 ± 15.5 | (10.3–89.6) |
| Diagnosis (%) | ||
| Ischemic heart disease | 48.9 | |
| Cardiac surgery | 27.5 | |
| Heart failure | 23.6 | |
| BNP (pg/ml)a | 75.5 (36.3–152.6) | (5.4–2154.2) |
| HR rest (beats/min) | 69.0 ± 11.1 | (44–101) |
| Peak HR (beats/min) | 130.1 ± 21.80 | (73–202) |
| Peak VO2 (ml/kg/min) | 20.3 ± 5.4 | (8.3–43.3) |
| Medications (%) | ||
| β-Blockers | 66.8 | |
| ACEIs/ARBs | 68.1 | |
| Diuretics | 35.7 |
Mean ± SD; BMI = body mass index; BNP = brain natriuretic peptid; HR = heart rate; ACEI = angiotensin converting enzyme inhibitor; ARB = angiotensin receptor blocker.
Median (interquartile range).
Fig. 1.
Relationships between HR parameters and METs.
Upper left: Relationship between METs and HRabsolute. Lower left: Relationship between METs and HRnet. Upper right: Relationship between METs and HRindex.
The results of the present study showed that HR parameters could be used to predict METs during exercise. HRnet was least influenced by background factors and β-blocker use, suggesting that METs during exercise can be roughly predicted using a formula of 0.05 × HRnet + 2. This formula could be useful for predicting METs during exercise from HR alone, without using specialized equipment in clinical settings.
There are some limitations in this study. First, HRrest may have affected the prediction formula. Future studies are needed to investigate whether the prediction formula calculated separately by HRrest affects R2 values. Second, since the present study used a cycle ergometer for CPX, the results of our analysis should be validated with data obtained using a treadmill. Third, when applying the prediction formula, if HRnet is 0, estimated METs would be 2; therefore, METs <2 cannot be predicted. However, activities corresponding to METs <2 are considered resting physical activities (e.g., reading in a lying or seated position) [3], and thus, no potential issues resulting from this are likely to limit the application of this formula.
The present study was the first to explore the possibility of predicting METs from HRrest and HRabsolute in patients with cardiovascular disease. METs during exercise could be predicted by 0.05 × HRnet + 2, regardless of age, sex, BMI, or the use of β-blockers that affect HR. This method could be useful for estimating METs in cardiac rehabilitation patients, as it requires no specialized equipment and enables instant feedback to patients.
Conflict of interest
The authors have no conflicts of interest to report.
Acknowledgments of grant support
This study was supported by the Japan Society for the Promotion of Science (JSPS), KAKENHI Grant Number JP18K17715, Japan. Furthermore, the authors thank Ms. Keiko Oka for her editorial assistance.
References
- 1.Fletcher G.F., Ades P.A., Kligfield P., Arena R., Balady G.J., Bittner V.A. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation. 2013;128:873–934. doi: 10.1161/CIR.0b013e31829b5b44. [DOI] [PubMed] [Google Scholar]
- 2.Steg P.G., James S.K., Atar D., Badano L.P., Blomstrom-Lundqvist C., Borger M.A. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur. Heart J. 2012;33:2569–2619. doi: 10.1093/eurheartj/ehs215. [DOI] [PubMed] [Google Scholar]
- 3.Ainsworth B.E., Haskell W.L., Herrmann S.D., Meckes N., Bassett D.R., Jr., Tudor-Locke C. 2011 Compendium of Physical Activities: a second update of codes and MET values. Med. Sci. Sports Exerc. 2011;43:1575–1581. doi: 10.1249/MSS.0b013e31821ece12. [DOI] [PubMed] [Google Scholar]