Introduction
An exercise stress echocardiogram includes both a maximal exercise test and echocardiographic imaging. The exercise stress echocardiogram has many clinical uses, such as detecting ischaemia, assessing outflow gradient during exercise and/or identifying intraventricular gradient with exercise to evaluate medical management and stratify risk for adverse cardiac events. Reference Cotrim, Café, João and Cardiol1 Historically, exercise stress echocardiograms were completed by having a patient complete a maximal exercise test on a treadmill before quickly lying down in a supine position for immediate post-exercise echocardiography imaging. Reference Dillon, Dausin and Claessen2 The semi-supine recumbent ergometer is a useful tool for exercise stress echocardiography as it allows for optimal echocardiographic imaging during progressive stages of exercise, including maximal exercise. Reference Ciliberti, McLeod and Cairello3 The semi-supine recumbent ergometer is a particularly useful alternative to treadmill-based exercise stress echocardiograms in children, who demonstrate a precipitous reduction in heart rate upon lying supine for echocardiogram images, Reference Ciliberti, McLeod and Cairello3,Reference Pahl, Duffy and Chaudhry4 which reduces diagnostic yield. Reference Ciliberti, McLeod and Cairello3
Given the utility of the semi-supine recumbent ergometer for stress echocardiography, there must be clarity regarding appropriate interpretation of cardiopulmonary exercise parameters in a semi-supine recumbent ergometer exercise test as body position can impact cardiopulmonary exercise parameters. Reference Dillon, Dausin and Claessen2,Reference Wan, Liu, Olson, Chen, Lu and Jiang5–Reference May, Punn, Olson, Kazmucha, Liu and Chin13 Although exercising on a semi-supine recumbent ergometer uses a cycling motion, the semi-supine body position rather than a standard upright body position may nevertheless alter cardiopulmonary exercise parameters. To date, there is no established normative data for cardiopulmonary exercise parameters in the semi-supine position. This absence of normative data creates ambiguity as to what may be a pathological exercise value versus normal exercise value on the semi-supine recumbent ergometer as there is conflicting data regarding the effect of a semi-supine body position on cardiopulmonary exercise parameters. Some studies show lower peak oxygen consumption on the semi-supine recumbent ergometer versus the treadmill or upright cycle ergometer, Reference Dillon, Dausin and Claessen2,Reference Forbregd, Aloyseus, Berg and Greve6–Reference Le, Zhou and Tao10 while others show no difference in VO2 peak between semi-supine recumbent ergometer and treadmill. Reference Quinn, Smith, Vroman, Kertzer and Olney11,Reference Fernie, Patel, Jacobsen and Rausch12 Still other studies demonstrate lower VO2 peak on a supine ergometer, Reference May, Punn, Olson, Kazmucha, Liu and Chin13 but it is unclear if or to what degree this may differ from a semi-supine recumbent ergometer. Many of these studies involve small sample sizes (sample size 6 to 23) with predominantly adult populations, which limits interpretation in a paediatric population.
Stress echocardiograms are routinely performed in children/adolescents with a history of Kawasaki disease and anomalous origin of the right coronary artery, among other conditions. Kawasaki disease without coronary artery aneurysm is a non-progressive condition that does not require exercise restriction. Reference Thompson, Myerburg and Levine14 Cardiopulmonary exercise testing with echocardiogram is standard of care following Kawasaki disease as a screen for myocardial ischaemia. Reference Gravel, Curnier, Dallaire, Fournier, Portman and Dahdah15,Reference Pellikka, Arruda-Olson and Chaudhry16 Exercise capacity in children with a history of Kawasaki disease without coronary artery aneurysm(s) is not significantly different from children without a history of Kawasaki disease. Reference Tuan, Chung and Chen17 Similarly, anomalous origin of the right coronary artery is another non-progressive cardiac condition. Cardiopulmonary exercise testing is a routine test in this population to screen for myocardial ischaemia that may indicate the need for corrective surgery. Reference Gentile, Castiglione and De Caterina18 Exercise capacity is not reduced in children/adolescents with anomalous origin of the right coronary artery even after surgical repair and/or recommended exercise restriction. Reference Elias, Meza, McCrindle, Brothers, Paridon and Cohen19,Reference Brothers, McBride and Marino20 Thus, cardiopulmonary exercise tests in children with a history of Kawasaki disease without coronary artery aneurysm or anomalous origin of the right coronary artery without history of intervention are a reasonable surrogate for cardiopulmonary exercise parameters in children with structurally normal hearts.
The goal of this study was to identify differences in cardiopulmonary exercise parameters between the semi-supine recumbent ergometer and the cycle ergometer and treadmill in children/adolescents. This information provides a foundation for possible appropriate normative data in the absence of semi-supine recumbent ergometer-specific normative data.
Materials and methods
The Colorado Multiple Institutional Review Board approved the protocol with a waiver of consent in view of the retrospective nature of the study. All procedures performed were part of routine clinical care. Only maximal tests were included for analysis. Since peak respiratory exchange ratio is lower in young children/adolescents, Reference Fernie, Silveira, Jacobsen and Rausch21 maximal tests were defined per Takken et al as respiratory exchange ratio ≥ 1.00 and subjective assessment by the exercise physiologist. Reference Takken, Bongers, van Brussel, Haapala and Hulzebos22 Tests with submaximal effort and/or poor test compliance were excluded from analysis.
As a retrospective study, all cardiopulmonary exercise tests included for analysis were limited to those performed per routine clinical care. Cardiopulmonary exercise tests are routinely performed on the treadmill and cycle ergometer in children/adolescents with normal cardiac structure and function, but children/adolescents with normal cardiac structure and function do not perform cardiopulmonary exercise tests on the semi-supine recumbent ergometer per routine clinical care. Therefore, only semi-supine recumbent ergometer tests performed by children/adolescents with a history of uncomplicated Kawasaki disease without coronary artery aneurysm or anomalous origin of the right coronary artery with no prior intervention were included for analysis.
There were three arms of analysis in this study (see Figure 1):
Enrolment/analysis schema. KD = Kawasaki disease; AORCA = Anomalous Origin of the Right Coronary Artery; SSRE = Semi-Supine Recumbent Ergometer; TM = Treadmill; CE = Cycle Ergometer; BMI = Body Mass Index.
-
1. Maximal cardiopulmonary exercise tests completed on the semi-supine recumbent ergometer by children/adolescents with a history of uncomplicated Kawasaki disease or unrepaired anomalous origin of the right coronary artery were compared against tests completed by children/adolescents with a history of uncomplicated Kawasaki disease or unrepaired anomalous origin of the right coronary artery on the treadmill or cycle ergometer.
-
2. Maximal cardiopulmonary exercise tests completed on the semi-supine recumbent ergometer by children/adolescents with a history of uncomplicated Kawasaki disease or unrepaired anomalous origin of the right coronary artery were compared against matched tests completed by children/adolescents who were referred for cardiopulmonary exercise test for family history of cardiac disease and/or symptoms with exercise and had normal cardiac structure and function confirmed by cardiac exam, electrocardiogram, and/or echocardiogram prior to the cardiopulmonary exercise test. Children/adolescents with structural abnormalities or channelopathies such as long QT syndrome or catecholaminergic polymorphic ventricular tachycardia were excluded from the cohort of matched controls. Tests performed on the semi-supine recumbent ergometer were matched to tests completed on the treadmill by exercise device, sex, age, height, weight, body mass index, and body surface area. Tests completed March–August 2021 were only matched with tests from that era, as exercise parameters have been found to be altered during that period due to effects of the COVID-19 pandemic. Reference Fernie, Pettit, Schaffer and Rausch24
-
3. Same as Arm 2, except tests performed on the semi-supine recumbent ergometer were matched to tests completed on the cycle ergometer.
Semi-supine recumbent ergometer tests were completed on an Ergoline Ergoselect 1200 (Davis Medical, Vista, California, USA, see Figure 2) with a targeted peak workload of approximately 2.5 watts/kg, with four 3-minute stages and a pedal speed of 60 revolutions per minute as described by Cifra et al. Reference Cifra, Dragulescu, Border and Mertens23 Our laboratory’s practice is to increase workload as needed within the final minute of exercise to reach/exceed maximal exercise criteria.
Ergoline 1200 set up for patient testing.
Cardiopulmonary exercise tests completed on the upright electronically braked cycle ergometer (Lode Corival CPET, Groningen, Netherlands) used a ramp protocol of 5 to 30 watts/minute for a targeted peak workload of approximately 3 watts/kg and a targeted exercise duration of 8 to 12 minutes. Patients were instructed to maintain a pedal speed of 60 to 70 revolutions per minute. Treadmill tests were performed using the Bruce protocol. The treadmill was used for children/adolescents who were not tall enough for the cycle ergometer or who reported symptoms primarily while running.
The cardiopulmonary exercise test systems (MGC Diagnostics Ultima CPX, Saint Paul, MN, USA) were maintained per manufacturer guidelines and calibrated according to manufacturer specifications. Patients completed baseline pulmonary function testing. Patients then underwent continuous metabolic monitoring, including oxygen consumption, carbon dioxide production, minute ventilation, and respiratory exchange ratio. Heart rate was measured by electrocardiogram. Cardiopulmonary exercise parameters were reported as 10-second averages. Peak values were defined as the highest 10-second average value for each variable during exercise. Ventilatory anaerobic threshold was identified using the V-slope method and was independently verified by an exercise physiologist and a cardiologist. The ventilatory equivalent of carbon dioxide slope was determined as the slope of minute ventilation versus carbon dioxide production from rest to ventilatory anaerobic threshold. Oxygen saturation was measured with a forehead probe and pulse oximeter (Nellcor PM1000N, Medtronic, Minneapolis, MD) to minimise disruption of signal while exercising. Oxygen saturation was validated by confirmation of good plethysmography signal and matching heart rate by electrocardiogram and the pulse oximeter. Blood pressures were measured using an appropriately sized cuff. A SunTech Tango M2 system (Morrisville, NC) was used to auto-inflate the blood pressure cuff, and blood pressure was auscultated by the supervising exercise physiologist. Resting blood pressure was taken prior to starting exercise. Peak exercise blood pressure was measured within the last 45 seconds of exercise. Peak workload on the cycle ergometer and semi-supine recumbent ergometer was identified as the peak watts achieved.
Statistical analysis
All analysis was performed in R version 4.4.0 (R Core Team 2024). Control and case groups were compared using two-sample T-tests for numeric variables and Fisher’s exact tests for categorical variables. Summaries are presented as N (%) for categorical variables and mean (SD) and median [range] for continuous variables. P-values for exercise outcomes were adjusted for multiple testing using the Benjamini–Hochberg method.
Results
In total, 115 cardiopulmonary exercise tests completed on the semi-supine recumbent ergometer by children/adolescents with a history of Kawasaki disease or anomalous origin of the right coronary artery were identified (see Figure 1). Seventeen of these tests were excluded for submaximal effort and 55 were excluded for Kawasaki disease with history of coronary artery dilation and/or coronary artery aneurysm or anomalous origin of the right coronary artery with history of surgical repair, leaving 43 maximal semi-supine recumbent ergometer cardiopulmonary exercise tests eligible for matching.
Arm 1: Semi-supine recumbent ergometer versus cycle ergometer/treadmill in children/adolescents with Kawasaki disease or anomalous origin of the right coronary artery
In Arm 1 of analysis, 43 cardiopulmonary exercise tests were matched against cardiopulmonary exercise tests completed on the cycle ergometer or treadmill by children/adolescents with a history of uncomplicated Kawasaki disease or unrepaired anomalous origin of the right coronary artery (n = 5). Age was nearly statistically significantly different between the semi-supine recumbent ergometer and cycle ergometer/treadmill groups (14.2 ± 2.51 versus 9.60 ± 3.78 years, p = 0.0525, see Table 1A).
Demographics
* Significant at p < 0.05.
Arm 2: Semi-supine recumbent ergometer in children/adolescents with Kawasaki disease or anomalous origin of the right coronary artery matched against treadmill in children/adolescents with normal cardiac structure/function
There were 20 cardiopulmonary exercise tests completed on the semi-supine recumbent ergometer that were matched to cardiopulmonary exercise tests completed on the treadmill by control patients (see Arm 2 of analysis, Figure 1). Demographics (age, sex, race/ethnicity, height, weight, body surface area, and body mass index) were identical between the semi-supine recumbent ergometer and treadmill groups (see Table 1B). There was no difference in baseline spirometry (p = 0.865) or the use of attention-deficit/hyperactivity disorder or blood pressure medications (p = 1) that may potentially alter heart rate or blood pressure response to exercise between the semi-supine recumbent ergometer and treadmill groups (see Table 1B). Ventilatory anaerobic threshold (20.6 ± 5.57 mL/kg/min vs 30.3 ± 8.24 mL/kg/min, p = 0.00193), peak oxygen consumption (37.0 ± 6.60 mL/kg/min vs 49.3 ± 8.99 mL/kg/min, p < 0.001, 2.04 ± 0.443 L/min vs 2.73 ± 0.558 L/min, p = 0.00182), and minute ventilation (81.9 L/min ± 18.8 vs 104 ± 22.8, p = 0.0335) were lower in the semi-supine recumbent ergometer group than the treadmill group (Table 2). Peak exercise oxygen saturation was lower on the treadmill compared to the semi-supine recumbent ergometer (94.4% ± 1.95 vs 91.7 ± 2.80, p = 0.0201; Table 2). There was no difference in baseline heart rate (69.6 ± 12.5 vs 80.4 ± 17.4 bpm, p = 0.523) or peak HR (193 ± 10.4 bpm vs 185 ± 12.1 bpm, p = 0.787) between the treadmill and semi-supine recumbent ergometer groups (Table 2). See Table 2 for all comparative exercise values between the semi-supine recumbent ergometer and treadmill groups.
Cardiopulmonary exercise parameters. Values are mean (SD)
Arm 3: Semi-supine recumbent ergometer in children/adolescents with Kawasaki disease or anomalous origin of the right coronary artery matched against cycle ergometer in children/adolescents with normal cardiac structure/function
There were 34 cardiopulmonary exercise tests completed on the semi-supine recumbent ergometer that were matched to cardiopulmonary exercise tests completed on the cycle ergometer by control patients. The only demographic difference between the semi-supine recumbent ergometer and cycle ergometer group was a higher percentage of Caucasian ethnicity in the cycle ergometer group (73.5% versus 38.2%, p = 0.0138, see Table 1C). There was no difference in baseline spirometry (p = 0.846) or the use of attention-deficit/hyperactivity disorder or blood pressure medications (p = 1) between the semi-supine recumbent ergometer and cycle ergometer groups (see Table 1C). The ventilatory equivalent of carbon dioxide slope was higher (28.7 ± 3.10 vs 26.3 ± 2.82, p = 0.038) on the cycle ergometer versus semi-supine recumbent ergometer. Peak cardiopulmonary exercise parameters (watts, peak oxygen consumption, heart rate, blood pressure, oxygen saturation, ventilation, respiratory exchange ratio, respiratory rate, and ventilatory anaerobic threshold) were not statistically different between semi-supine recumbent ergometer and cycle ergometer (Table 2).
Discussion
This study identifies limited differences in cardiopulmonary exercise parameters between the semi-supine recumbent ergometer and cycle ergometer and treadmill in a matched case-control design.
Arm 1 of analysis evaluated tests performed on the semi-supine recumbent ergometer versus cycle ergometer or treadmill only within children/adolescents with Kawasaki disease or anomalous origin of the right coronary artery. A very small sample size (n = 5) of tests completed on the cycle ergometer and treadmill and trends towards differences in age (14.2 ± 2.51 vs 9.60 ± 3.78, p = 0.0525) and height (164 ± 13.4 cm vs 139 ± 23.8, p = 0.0737) limited the validity of comparisons between groups as age and height are significant contributors to normative cardiopulmonary exercise test values. Reference Fernie, Pettit, Schaffer and Rausch24,Reference Burstein, McBride and Min25 Therefore, we consider the matching of cardiopulmonary exercise tests completed in children/adolescents with history of uncomplicated Kawasaki disease or anomalous origin of the right coronary artery to cardiopulmonary exercise tests completed in children/adolescents with normal cardiac structure and function with no history of acquired heart disease to be the best retrospective surrogate for identifying differences in cardiopulmonary exercise parameters between exercise devices.
Analysis of Arm 2 revealed that tests completed on the treadmill demonstrated higher peak oxygen consumption (2.73 ± 0.558 L/min versus 20.4 ± 0.443, p = 0.00182 and 49.9 ± 8.66 mL/kg/min versus 37.0 ± 6.60, p < 0.001) and higher ventilatory anaerobic threshold (30.9 ± 7.66 mL/kg/min versus 20.6 ± 5.57, p < 0.001) compared to tests completed the semi-supine recumbent ergometer. This indicates that both peak oxygen consumption and submaximal exercise capacity (ventilatory anaerobic threshold) are reduced on the semi-supine recumbent ergometer compared to the treadmill. There is limited data directly comparing semi-supine cardiopulmonary exercise parameters to those on the treadmill, but one study of 23 adults with stable heart failure with preserved ejection fraction reported a 23% difference in peak oxygen consumption between the semi-supine recumbent ergometer and treadmill. Reference Le, Zhou and Tao10 Similarly, the mean peak oxygen consumption on the treadmill in our dataset was 34.8% higher than on the semi-supine recumbent ergometer. This has important implications for interpreting peak oxygen consumption values on the semi-supine recumbent ergometer, as comparing semi-supine recumbent ergometer values against normative values on the treadmill could falsely report pathological per cent-predicted peak oxygen consumption when in fact the difference is due to testing device.
Minute ventilation was higher on the treadmill compared to the semi-supine recumbent ergometer (104 ± 22.8 L/min versus 81.9 ± 18.8, p = 0.027) but peak respiratory rate was not statistically different between treadmill and semi-supine recumbent ergometer (54.2 ± 8.11 breaths/min versus 55.4 ± 17.7, p = 1). Since minute ventilation is a factor of tidal volume and respiratory rate, this means that peak tidal volume was higher on the treadmill than on the semi-supine recumbent ergometer. The mechanism of reduced tidal volume on the semi-supine recumbent ergometer is unclear. Potential underlying mechanisms of reduced tidal volume include differences in pulmonary blood flow and pulmonary perfusion in the semi-recumbent position, restricted chest wall expansion due to the hard back of the semi-supine recumbent ergometer thereby restricting peak tidal volume, or restricted diaphragm flexion due to the semi-bent leg position of the semi-supine recumbent ergometer. Research generally finds an increase in forced vital capacity when standing versus sitting versus supine,Reference Katz, Arish, Rokach, Zaltzman and Marcus 26 but it is unclear how the body position of the semi-supine recumbent ergometer may impact tidal volume during exercise. Ventilatory efficiency as measured by ventilatory equivalent of carbon dioxide slope was not different between semi-supine recumbent ergometer and treadmill (26.1 ± 2.53 versus 27.3 ± 2.61, p = 1; Table 2), so it is possible that the semi-supine recumbent ergometer restricts chest wall expansion and peak exercise tidal volume. Peak oxygen saturation may be lower in the treadmill group than the semi-supine recumbent ergometer group due to one outlier (91.1% ± 3.47, range 82 to 95 versus 94.4% ± 1.95, range 90 to 98; Table 2).
Analysis of Arm 3 identified that nearly all cardiopulmonary exercise parameters were identical between the cycle ergometer and semi-supine recumbent ergometer, with the only statistically significant difference being ventilatory equivalent of carbon dioxide slope (26.3 ± 2.82 semi-supine recumbent ergometer versus 28.7 ± 3.10 cycle ergometer, p = 0.038; Table 2). This differs from Wehrle et al who reported lower peak oxygen consumption, peak watts, and ventilatory anaerobic threshold on the semi-supine recumbent ergometer versus cycle ergometer, while ventilatory equivalent of carbon dioxide slope was not statistically significant between semi-supine recumbent ergometer and cycle ergometer. Reference Wehrle, Waibel, Gollhofer and Roecker9 In the current study, minute ventilation was not significantly different between semi-supine recumbent ergometer and cycle ergometer, so the difference in ventilatory equivalent of carbon dioxide slope between devices was due to increased carbon dioxide production on the semi-supine recumbent ergometer compared to the cycle ergometer. Anecdotally, patients tend to report more localised muscle fatigue on the semi-supine recumbent ergometer than the cycle ergometer, so the higher carbon dioxide production could reflect localised muscle fatigue. Interestingly, Wehrle et al reported lower carbon dioxide production peak on the semi-supine recumbent ergometer versus the upright cycle ergometer. Reference Wehrle, Waibel, Gollhofer and Roecker9 The subjects in Wehrle’s study were adults, so it is possible that age contributes to differences in cardiopulmonary exercise parameters between devices. Given that increases in ventilatory equivalent of carbon dioxide slope are closely associated with increased risk of adverse cardiac outcomes in some congenital heart conditions such as Tetralogy of Fallot and after single ventricle (Fontan) repair, it is imperative that normative ventilatory equivalent of carbon dioxide slope values on the semi-supine recumbent ergometer be established for specific clinical populations for valid interpretation. Minute ventilation and respiratory rate were not significantly different between semi-supine recumbent ergometer and cycle ergometer (minute ventilation 77.2 ± 19.6 versus 83.2 ± 23.7, p = 1; respiratory rate 54.5 ± 15.6 versus 54.0 ± 9.02, p = 1). This suggests that tidal volume was not significantly different between semi-supine recumbent ergometer and cycle ergometer, which could be due to both devices restricting full hip extension.
Limitations/future directions
While all efforts to only include semi-supine recumbent ergometer tests in children/adolescents with presumptively normal exercise capacity were made, the history of Kawasaki disease or anomalous origin of the right coronary artery in the semi-supine recumbent ergometer group may nevertheless impact cardiopulmonary exercise parameters and introduce bias that reflects clinical history rather than differences due to exercise device. This is a limitation of the retrospective design of this study in which all tests were completed per standard of care rather than prospective enrolment. Future prospective studies may focus on enrolling patients with no cardiac history and conducting exercise tests on patients on multiple devices for improved control between exercise devices.
Conclusions
Ventilatory anaerobic threshold, peak oxygen consumption, and minute ventilation were lower in the semi-supine recumbent ergometer versus treadmill group. The uniformity in nearly all cardiopulmonary exercise parameters between the semi-supine recumbent ergometer and cycle ergometer suggests that normative data for the cycle ergometer are a reasonable surrogate for normative data on the semi-supine recumbent ergometer until semi-supine recumbent ergometer-specific normative data are developed. However, the difference in ventilatory equivalent of carbon dioxide slope between the semi-supine recumbent ergometer and cycle ergometer suggests altered ventilatory efficiency that should be explored, particularly for valid interpretation of ventilatory equivalent of carbon dioxide slope in clinical populations where increases in ventilatory equivalent of carbon dioxide slope are associated with increased risk for adverse cardiac events.
Acknowledgements
The authors would like to thank the exercise physiologists and cardiac sonographers who conducted the exercise tests.
Financial support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Competing interests
None.
Ethical standard
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008, and have been approved by the institutional committees (Colorado Multiple Institutional Review Board).
