New kids on the CPET: age-appropriate outdoor cardiopulmonary exercise testing in preschoolers

Lack of physical activity and the often-related problem of obesity are well-known risk factors for cardiovascular diseases (Boyer et al. 2015; Werneck et al. 2019). Among the health-related physical fitness components in children and adolescents, we know cardiorespiratory fitness (CRF) to be the most important marker (Ortega et al. 2007; Ruiz et al. 2009). Higher levels of CRF during early childhood are favorably associated with multiple health indicators (Carson et al. 2017). They especially result in lower cardiovascular disease risk factors during later life (Nordström, Högström, and Nordström 2014) and reduce the risk of becoming overweight across puberty (Ortega et al. 2007).

Cardiopulmonary exercise testing (CPET) is a commonly performed, non-invasive method to evaluate cardiac symptoms and assess functional capacity in children (Paridon et al. 2006) and can be considered as safe (Ghosh et al. 2015). In very young children (4–6 years) though, CPET is highly challenging. Bicycle testing is usually not feasible in this age group due to, cognitive, anatomic and technical limitations (average height for a 6-year-old is about 118 cm, cutoff for bicycle testing according to the German medical products law is 120 cm) (Neuhauser 2013). Therefore, exercise testing can only be performed on a treadmill (TM). Requirements for a representative treadmill test are good motor skills and a high motivation, both difficult to achieve in this age group. Most studies investigating CPET have, therefore, been limited to older children (> 8 years of age) (Ghosh et al. 2015; Duff et al. 2017; Leger et al. 1988; Akkerman et al. 2010; Armstrong and Welsman 2001; Armstrong et al. 1991; van der Steeg and Takken 2021; Blanchard et al. 2018; Bongers et al. 2016).

Since children have relatively underdeveloped knee extensors, treadmill testing is generally preferred over cycle ergometry (CE) in young children (Bar-Or 2004). The most common TM protocol is the Bruce protocol which was originally designed for diagnosing coronary artery disease in middle-aged men (Bruce et al. 1963). It has since become the most common TM protocol, even in children from the age of 4 years on (Wessel et al. 2001). In this protocol, speed and inclination are increased every 3 min (Bruce et al. 1963). However, increasing the incline during a treadmill test represents a high demand for those relatively underdeveloped knee extensors and may lead to premature local leg fatigue, causing the children to stop running before achieving maximal performance (Duff et al. 2017). In addition, the rather large increases between the different steps and a step-duration of 3 min each are difficult, as boredom and demotivation step in (Duff et al. 2017). Several attempts have been made, to modify the Bruce protocol for a better suitability for children (van der Cammen-van Zijp et al. 2009; Dubowy et al. 2008; Eiberg et al. 2005; Tuan et al. 2019, 2018).

Furthermore, field-based tests for the indirect estimation of \(\dot{V}O\)_2peak have been proposed, and inter alia, derived as a consequence of lacking equipment, trained personnel or even financial aspects (Mora-Gonzalez et al. 2017; Leger et al. 1988). One approach is the adapted 20 m Shuttle-run test for preschool children, the so-called 20mSRT-PREFIT (Mora-Gonzalez et al. 2017). However, in this approach, a validation of data is not possible without true cardiopulmonary exercise data (Tomkinson et al. 2019).

The development of mobile cardiopulmonary exercise testing instead offers a whole new opportunity for accurate measurements of \(\dot{V}{O}_{2peak}\) in children (Schoffl et al. 2020a). In a recently published study examining cardiopulmonary field testing using a mobile CPET device, young school children were successfully tested by running in an outdoor park. The children were able to increase the speed according to their own capabilities, allowing for age- and fitness-adapted testing and keeping the test duration comparable (Schoffl et al. 2020a). One limitation of this first study was the fact that it compared the field test with indoor CE testing. Therefore, in this study CE testing was used for the indoor protocol. As maximal exertion during biking and running cannot directly be compared, no conclusion could be drawn as to the benefits of outdoor vs. treadmill testing. Still, this was the first study to investigate field testing in children (7–10 years of age).

In this study, we want to implement the upper mentioned field test (Schoffl et al. 2020a) in preschool children (4–6 years) and compare it to the ramped modified Bruce protocol on the treadmill which was developed by Dubowy et al. (2008) and is actually recommended by the German Association for Pediatric Cardiology (DGPK) for TM testing in children. Thus, the comparison between the indoor and outdoor tests was more realistic as walking and running were used in both the tests.

The study was approved by the Ethics Committee of the University of Erlangen-Nuremberg (159_19B).

Evaluation of the cardiorespiratory fitness of healthy children and those who suffer from a chronic cardiopulmonary disease is essential for determining cardiovascular risks but also for potential disease management. Due to anatomic limitations, CPET with very young children can only be performed on the TM. But this is often problematic, as the required motor skills may not yet be sufficiently developed in this age group (Hebestreit 2004).

Our study population consisted of a total number of 22 children aged 4–6 years, which is comparable to other studies (Schoffl et al. 2020a). Average height (114 cm) was below the minimum height required for bicycle testing (Hebestreit 2004). Both indoor and outdoor testing represented no health risk to the children. The preschoolers in this study showed a clear preference for the outdoor test. They were more reluctant to run on the treadmill, presenting coordinative difficulties, culminating in one indoor-dropout due to fear of the TM. This observation has previously been described (van der Cammen-van Zijp et al. 2009) but was not evident during the outdoor testing, where the children were highly motivated to run.

Hebestreit et al. (2004) reported children to prematurely quit during the first few seconds of a new step, likewise implying an increment of workload (Hebestreit 2004). The preschoolers in this study showed comparable behavior on the TM. The reason for this phenomenon is probably linked to the motor-skill-related problems occurring on the TM as they were especially noticeable during low speeds and at the start of a new step. Another reason for the premature termination of tests at the beginning of a new step could be the increasing inclination of the TM leading to leg fatigue due to underdeveloped knee extensors in this age group (van der Cammen-van Zijp et al. 2009). Furthermore, uphill running requires higher energy expenditure than running on a flat ground (Cavagna et al. 1964). To address this specific problem when testing children, efforts have been made to develop new TM protocols for children without inclination of the treadmill (Duff et al. 2017). Duff et al. 2017 used a protocol (BCCH) where the incline stayed at a constant 1% starting at a speed of 2.0 mph, increasing by 0.5 mph every minute until volitional fatigue. In comparison, the Bruce protocol simultaneously increases the speed and grade of the treadmill every 3 min. As a result, the BCCH-protocol requires children to run at an earlier stage and at faster speeds. Furthermore, it addresses the problem of local leg fatigue, which is often a complaint in the Bruce protocol. Endpoint data such as \(\dot{V}\)E, \(\dot{V}\)O₂, \(\dot{V}\)CO₂, RER and HR were comparable between both protocols. Nevertheless, the BCCH-protocol requires more exercise time than the Bruce protocol and was only tested in children aged 10–18 years, which is one of the reasons why we did not use it for our preschoolers. In our clinical experience, long exercise time leads to boredom and premature quitting of the test, especially in younger children. On top, the German Association of Pediatric Cardiologists (DGPK) recommends the TM protocol by Dubowy, which is the one we used indoors, and it is, therefore, probably the most common TM protocol used for testing children in Germany.

Both protocols performed by Eiberg and LeMura, as well as the Bruce protocol, used an inclination on the TM, which we did not choose due to the problem of local leg fatigue while running uphill and the upper mentioned fact of the problems encountered using protocol by Dubowy.

We commonly observed that the children were able to chatter a few seconds after TM test termination, whereas after the outdoor tests, none of the children was able to hold up to a normal conversation due to exertion. Studies about the talk test as a marker of exercise training intensity suggest there is a strong correlation between a decrease in spoken words and training above the ventilatory threshold (Reed and Pipe 2014). We presume that running on the TM did not lead to the same level of cardiorespiratory exertion as running outdoors.

When working with children, motivation and fun are crucial factors. Hence, it is important to note that all the children enjoyed the outdoor test more than running on a treadmill. This observation has previously been described in a study comparing outdoor testing with indoor cycle testing (Schoffl et al. 2020a).

Essentially, the children were able to reach significantly higher RER values running outdoors than during the TM test (RER outdoor 1.10 vs. indoor 0.97). This parameter is representative of the maximum exertion achieved by the test person. A study on 5–6-year-old children has defined maximum effort as RER values > 1.0 (LeMura et al. 2001). Average values of 1.05–1.10 have been reported for testing 6–7-year-old-children on a treadmill (Eiberg et al. 2005). In a TM-study on preschoolers using a ramp protocol instead of a stepwise increase of the workload, RER values as high as 1.03–1.18 were achieved (Tuan et al. 2019). Our results for the outdoor test are comparable, showing maximum cardiorespiratory exertion. However, the mean RER values achieved on the TM using the modified Bruce protocol were below the recommended value of 1.0 (LeMura et al. 2001), suggesting that cardiorespiratory exertion was not achieved in our cohort. Therefore, this protocol may not be suitable for testing preschoolers. The same outdoor protocol has previously been used in 7–10-year-old children (Schoffl et al. 2020a). The RER values achieved in this study were even higher (RER 1.19 in boys and 1.25 in girls), reflecting on the fact, that testing older children is more feasible (Schoffl et al. 2020a).

In our study, most of the CPET values showed a significant difference between the two tests, with higher values achieved during the outdoor test. The peak oxygen consumption \(\dot{V}{O}_{2}peak\), which is a surrogate parameter of cardiopulmonary fitness, was much higher outdoors than on the TM (51.5 ml/min/kg vs. 40.6 ml/min/kg). This, we think, is due to a better cardiopulmonary exertion, reflected by the higher RER (Schoffl et al. 2020a). Other studies determining \(\dot{V}{O}_{2}peak\) in 5–7-year-old children on the TM and on a stationary bike, recorded comparable results with values ranging from 44.8–48.5 for \(\dot{V}{O}_{2}peak\) (Eiberg et al. 2005; LeMura et al. 2001). The same outdoor test performed by older children (7–10-year-olds) also provided comparable results with mean \(\dot{V}{O}_{2}peak\) ranging from 50.0 m/min/kg in girls to 52.8 ml/kg/min in boys (Schoffl et al. 2020a). The outdoor test, therefore, seems to provide reasonable results even in children as young as 4 years of age.

There has been a variety of studies in children, reporting about higher maximum HR when tested on the TM than on the CE (Armstrong et al. 1991; Boileau et al. 1977; Cumming 1985), including one examining preschoolers (LeMura et al. 2001). A possible explanation for this discrepancy is the limiting influence of underdeveloped knee extensor muscle in young children, inducing local muscle fatigue and premature termination of the CPET (van der Cammen-van Zijp et al. 2009). In our study, the HR was significantly higher outdoors compared to the indoor TM test (187/min vs. 177/min). Again, this could be due to the same phenomenon, since an increase in inclination will also lead to local muscle fatigue of the underdeveloped knee extensors and the TM test demanded going uphill in the protocol, whereas the outdoor protocol did not. The same outdoor test used in this study elicited mean heart rates of 192.5 in boys and 199.3 in girls, who were 7–10 years old (Schoffl et al. 2020a). The most reasonable explanation for this discrepancy is the difficulty to reach maximum exertion in children so young. Other TM-studies reported maximum heart rates ranging from 194 ± 11/min in 6-year-olds (van der Cammen-van Zijp et al. 2010), up to an average of nearly 200/min in 5–6-year-old children (LeMura et al. 2001). However, a study with a comparable age group (5.58 years in girls and 5.86 years in boys) recorded comparable values to our study (178–184/min) (Tuan et al. 2019). These considerations underline the fact that better cardiopulmonary exertion was reached during the outdoor test compared to the indoor TM test, which was also reflected by the significantly higher values of \(\dot{V}E\)_max, O₂pulse, and maximum breath rate (BR).

The oxygen uptake efficiency slope (OUES), an age- and sex-dependent parameter of submaximal cardiopulmonary effort (Bongers et al. 2016), provides a valid measure of cardiopulmonary fitness in children who cannot reach maximal exertion (Bongers et al. 2011). As OUES and \(\dot{V}{O}_{2peak}\) strongly correlate (Bongers et al. 2016), the OUES can be used as a surrogate parameter for \(\dot{V}{O}_{2peak}\), especially in study populations, where maximal CPET results are difficult to achieve due to, e.g., motivational issues (Ten Harkel and Takken 2019). Furthermore, \(\dot{V}{O}_{2peak}\) and OUES show excellent interest reproducibility and reliable results among the same study population (Baba et al. 1999). This is consistent with our study results as the preschoolers show comparable results for OUES both during indoor TM and outdoor testing. Despite the comparable results for OUES, the children were able to achieve significantly higher maximal CPET values during the outdoor testing than on the TM. This characterizes the OUES as a test-independent, reproducible, submaximal parameter.

Average peak velocity was significantly higher during the outdoor tests (8.3 km/h vs. 5.4 km/h), which is a consequence of running without inclination outdoors but with increasing steepness on the TM. Compared to the mean peak velocity obtained in a comparable study on 7–10-year-olds (10.11 km/h), the preschoolers were significantly slower (Schoffl et al. 2020a). The mean speeds recorded at each step could be used for developing an age-appropriate TM test with age-appropriate speeds (first step: 3.16 km/h, second step: 4.29 km/h, third step: 5.65 km/h, fourth step: 7.22 km/h).

The fact that TM protocols working for adults might not fit for children has led to the development of more age-appropriate test protocols for children. Several authors have altered the Bruce protocol, by shortening the steps, decreasing the increments of speed and inclination (van der Cammen-van Zijp et al. 2009; Dubowy et al. 2008) and starting with a higher speed (Eiberg et al. 2005). This has led to a reduction in mean test time to values as low as 7.28 min (Eiberg et al. 2005). With these average test durations in mind, we used a test protocol with a defined maximum duration time of 8 min. Mean outdoor test duration in our study was 6.57 min., a time that was significantly shorter than the average test duration of 11.20 min. using the common TM protocol. Even though, this test duration seems comparably short, better cardiopulmonary exertion was achieved and ventilatory thresholds were easily discernible. We believe that keeping exercise times short will prevent boredom and consequently premature termination of CPET and that comparable exercise times are more important than comparable test protocols for a better comparability of CPET data (Schoffl et al. 2020a). As these children grow older, their relative speed may, therefore, increase for each step, however, the test duration will remain comparable.

The study limitations in our study are the same as in a previous study using mobile CPET equipment in the park (Schoffl et al. 2020a). Namely, the need for special equipment (mobile CPET and specialized ECG), the restrictions due to bad weather conditions and the need for trained test personnel for ensuring the proper test protocol as well as the safety of the child.

Concluding, common TM problems such as fear of TM testing, local leg fatigue, boredom and subsequent premature test ending were evaded by letting the children use natural movement patterns in a natural habitat, namely running on flat ground in an outdoor park (Duff et al. 2017). In our opinion, proper cardiorespiratory exertion does not depend on the applied protocol but on its feasibility in the corresponding study population. All the children in this study preferred the outdoor testing over running on the TM, and consistently showed significantly higher values of cardiorespiratory exertion. It is our aim to standardize exercise times, using individual protocols adapted to the capacity of each child, its stage of disease and chronological age and physiological development instead of standardizing protocols with individually lasting exercise times, thus allowing for a better long-term comparability.