Reference values for arm ergometry cardiopulmonary exercise testing (CPET) in healthy volunteers
•,.
...
Abstract
Introduction The performance of a cardiopulmonary exercise test (CPET) requires an individual to undertake a progressive, maximal exercise test to a symptom-limited end point. CPET is commonly performed using a treadmill or cycle ergometer (CE). Arm ergometry (AE) is an alternative exercise modality to CE; however, AE produces lower peak oxygen uptake (V̇O2) values as it involves smaller muscle groups and generates less cardiovascular stress. Current predicted equations for the interpretation of AE CPET are limited by small sample sizes, gender bias and limited age ranges.
Aims To develop predicted equations and reference ranges for AE exercise testing.
Design Incremental ramp protocol CPET, to a symptom-limited end point, via AE was performed in a group of 116 (62 F) healthy volunteers of median age 38 (IQR 29–48) years. Breath-by-breath gas analysis was performed using the Ultima CPX (Medical Graphics, UK) metabolic cart. Quantile regression analysis was used to develop regression equations for AE V̇O2, peak work rate (WR), anaerobic threshold, peak ventilation (VE), peak heart rate, oxygen pulse, V̇E/V̇CO2 slope and V̇O2/WR slope.
Results Reference equations including upper and/or lower limits, based on quantile regression, were generated and verified using a validation cohort.
Conclusions These findings represent the largest and most diverse set of predicted values and reference ranges for AE CPET parameters in healthy individuals to date. Implementation of these reference equations will allow AE to be more widely adopted, enabling the performance and interpretation of CPET in a wider population.
What is already known on this topic
Findings to date demonstrate that peak oxygen uptake (pV̇O2) is significantly lower when obtained using arm ergometry exercise. There are currently no recommended reference equations for use with arm ergometry. In contrast to published equations for cycle ergometry and treadmill, three of the current published regression equations for arm ergometry use achieved work rate to predict pV̇O2. In addition, many of the current studies were obtained from single-sex or patient populations.
What this study adds
Our study provides predicted values and reference ranges for pV̇O2, anaerobic threshold, peak work rate, peak ventilation (VE), peak heart rate, oxygen pulse, V̇E/VCO2 slope and the V̇O2/work rate slope for cardiopulmonary exercise using arm ergometry. It represents the largest data set for arm ergometry reference values to date.
How this study might affect research, practice or policy
Reference values developed by this study have the potential to significantly improve the interpretation of arm ergometry exercise tests and standardise interpretation. The data will enable the assessment of clinical populations against a normal range with the view to developing interpretation strategies for a range of clinical applications.
Introduction
Cardiopulmonary exercise testing (CPET) provides an integrated assessment of the response to exercise of the respiratory, cardiovascular, haematopoietic, neuropsychological and skeletal muscle systems. Standard, static tests of pulmonary and cardiac function, which are performed at rest, do not determine exercise capacity or causes of exercise limitation.1 However, stressing these systems using CPET can reveal pathologies that are not apparent at rest. Furthermore, exercise capacity is one of the most powerful predictors of cardiovascular and all-cause mortality.2–4 Consequently, CPET is increasingly used to provide a holistic assessment of exercise limitation and its causes and is widely used to provide clinically meaningful information regarding risks associated with surgical intervention.
CPET requires an individual to undertake a progressive, maximal exercise test to a symptom-limited end point. Tests are commonly performed using a treadmill (T) or cycle ergometer (CE). In normal individuals, the limiting symptoms will usually be physical fatigue or dyspnoea. Measurement of inspired and expired breath (breath-by-breath analysis) while exercising allows the calculation of various parameters which can be compared with expected normal values. Concurrent assessment of heart rate (HR) and rhythm, using an ECG, allows an integrative assessment of an array of parameters which evaluate the physiological response to exercise and causes for exercise intolerance. Clinical exercise testing is now routinely used for diagnosis of exercise limitation, surgical risk assessment or for monitoring of progression of disease.
However, there is a cohort of patients in whom the performance of CE or T testing is not possible because of medical conditions leading to impairment of the lower limbs or inability to perform weight-bearing exercise; commonly claudication, previous orthopaedic surgery, lower limb joint dysfunction/pain and certain neurodegenerative diseases. Consequently, the accepted method for determining surgical fitness or estimating prognosis is not available to these individuals, potentially preventing them from accessing essential surgery or therapy.
Arm ergometry (AE) is an alternative exercise modality; however, the data that have been obtained to date in healthy, normal populations for the purpose of generating a normal range of exercise indices are limited by the small size and heterogeneity of the study populations or are based on a fixed percentage of cycle ergometry predicted values.5–8 The largest study group to date included 60 (30M) healthy volunteers, aged 20–59 years.8 The output variables of work rate (WR), peak oxygen uptake (pV̇O2) and (HR) were derived using a cycle adapted for AE and a 10-watt step ramp protocol. Three further studies reporting reference values were derived using male-only populations, subjects with coronary artery disease and discontinuous step ramp protocols.5 6 There are currently no recommended reference equations for use with AE, with the wider literature predominately focusing on its relationship to CE instead.
Findings to date demonstrate that pV̇O2 is significantly lower when obtained using AE, and this can be explained by the physiology of the muscle groups used for each exercise.9 In addition, during submaximal phases of exercise, exercise parameters are higher for any given WR when exercising with the arms, suggesting decreased exercise efficiency.10 11 The fitness measures obtained from AE, therefore, cannot be assumed to correlate with the accepted methods of T and CE. Consequently, the clinical interpretation of maximal exercise parameters obtained from AE may, in the absence of direct comparison of these modalities and a robust normal reference range, be unreliable.
The aim of the present study was to assess the physiological responses of adult males and females (free from underlying cardiac or respiratory disease) to AE exercise testing and to use this data to generate a local set of regression equations for CPET parameters for use in the interpretation of AE exercise.
Methods
Participant recruitment
Participants were members of staff within our organisation, identified through advertisements including weekly electronic communications, intranet posts and all user emails. Participants were required to be of working age (18–69 years), free from any known cardiac or respiratory impairment and taking any prescribed medication likely to affect exercise performance.
Patient and public involvement
Patients and the public were not involved in the development of the research question, design or recruitment; no patient advisors were required. The results will be disseminated to the scientific community in academic writing.
Preparticipation screening
Prior to CPET, participants were assessed for health risks associated with exercise by completing the Physical Activity Readiness Questionnaire andYou (PAR-Q) questionnaire,12 consistent with the recommendations of The American College of Sports Medicine(ACSM) guidelines for cardiovascular disease risk stratification.5 After health screening, anthropometric measurements of height (cm), weight (kg) and calculation of body mass index (kg/m2) were made according to established procedures.13
Baseline physiological measurements of spirometry, ECG and blood pressure (BP) were performed to assess for any evidence of underlying respiratory or cardiac disease. Participants with normal baseline spirometry, ECG and BP were eligible to take part in the study.
Activity questionnaire
Following successful screening, participants were requested to complete the Recent Physical Activity Questionnaire (RPAQ) to quantify their level of physical activity.14
Arm ergometer exercise testing
AE exercise was performed using the Ergoselect 400+ (Ergoline, Germany) electronically braked arm ergometer (figure 1). Seat height was not adjustable; however, the height of the ergometric test unit was electrically adjustable and was adjusted to ensure that the participants’ arms were slightly bent at the elbow during furthest extension, as per knee flexion for CE.15
All participants undertook maximal CPET to volitional exhaustion. Incremental AE ramp protocols were calculated for each subject using the equations of Cooper and Storer.16
Arm work rate max=[(predicted V̇O2 max×0.70)+(BW×3.5)]/18.36
Predicted V̇O2 max was calculated using the Wasserman regression equations,7 where predicted V̇O2 max using the arms is estimated to be 70% that of the legs, hence the multiplication by 0.70. BW is the subjects’ body weight in kilograms.
All participants completed a rest phase of at least 2 min to allow familiarisation of the equipment and ensure accurate equipment function. This was followed by a 3 min unloaded phase.
Previously reported data suggest that arm crank speeds between 50 and 80 rpm have no influence on pV̇O217 but higher speeds can result in significantly higher values for HR and ventilation (VE),18 19 our centre employs a cadence of 60 rpm for CE, and so this was standardised across both modalities.
On completion of the unloaded exercise phase, the calculated exercise protocol commenced immediately. All participants were actively encouraged to exercise to a symptom-limited end point, for example, arm fatigue or breathlessness. The test was terminated by the Healthcare Scientist if, following a 60 s period of active encouragement, cadence fell by ≥5%. Measurements of gas exchange were made using the breath-by-breath analysis system Ultima CPX (Medical Graphics, UK) with an averaging method of five of seven breaths. HR was continually monitored using a 12-lead Mortara ECG system (Mortara, model X12+, Milwaukee, Wisconsin, USA). Maximal HR was defined as the highest value recorded during the test.20 Due to a lack of evidence to demonstrate maximal effort for AE CPET, the accepted CE criteria of a plateau in V̇O2, >90% predicted peak HR, peak VE>85% predicted and peak WR>85% were all accepted criteria for the achievement of maximal effort.21–23
Statistical analysis
The Kolmogorov-Smirnov test was used to assess data normality. Quantile regression analysis was performed to investigate the relationship between CPET variables and multiple independent variables. A standard bootstrap method with 5000 replications was completed to estimate the standard errors and construct CIs.
The regression equations were validated using a separate data set and differences were assessed by Wilcoxon signed-rank test. Spearman’s r was used to analyse the association between all studied parameters. The values p<0.05 were considered statistically significant. Statistical analysis was done using Stata V.14.2 (Stata).
Results
A total of 120 participants requested to take part in the study. One participant was excluded due to age (>70 years) and a further participant was excluded following initial screening measurements which demonstrated poorly controlled asthma. In total, 118 subjects were successfully recruited to the study. Of the 118 subjects, a further 2 (1.7%) were excluded after the performance of the first CPET. One due to dysfunctional breathing on exercise which prohibited accurate interpretation of the results, with the second demonstrating significant cardiac arrhythmias requiring a referral for a formal cardiological assessment.
Participants found the RPAQ difficult to accurately answer and demonstrated a tendency to over-report their recreational sporting activities. Consequently, data from the RPAQ questionnaire were excluded from further analysis. Despite participants not undertaking an assessment of their physical activity levels, it was confirmed that none of the participants undertook physical activity at a competitive level, consistent with that reported in previous studies.24 Many subjects recruited were recreational athletes, with cycling and running the most common pastimes reported.
Of the 116 participants recruited, 62 (53.45%) were female. Subjects met the age range criterion of 18–69 years; however, there were no subjects aged 18 years, only one aged 19 years and only one male and one female subject between the ages of 60 and 69 years. The age distribution of the study sample according to sex is summarised in figure 2. No participants reported a history of respiratory or cardiac disease.
Distribution of the study sample subdivided by decile.
No participants were active smokers; however, 14/116 (12%) had a history of cigarette/tobacco smoking, with 2/14 (14%) reporting exposure of >15 pack years. Both of these participants had normal spirometry (forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) ratio, FEV1 and FVC all within±1.645 SR), and therefore, no spirometric evidence of smoking-related lung disease.25 26 Relevant characteristics and distributions of study participants are shown in table 1.
Table 1
•
Subject characteristics
AE exercise
Results of AE CPET are summarised in table 2. All values are at peak exercise, except for anaerobic threshold. The median protocol was 7.50 watts/min with a total exercise duration (unloaded and incremental phase) of 14.27 min.
Table 2
•
Results of arm ergometry exercise testing
Individual subjects’ AE values were compared with their predicted values for cycle ergometry.7 For the group, median pV̇O2 was 80.4% of the predicted V̇O2 for cycle ergometry. Anaerobic threshold occurred at 37.3% of the cycle predicted V̇O2 and 46.5% of the achieved AE pV̇O2. Peak HR was 86.7% of the Tanaka peak predicted HR.20 Peak WR for AE exercise was 55.1% of that predicted for cycle ergometry,16 and peak VE reached 54% predicted (measured FEV1×35).
Reference values
Analysis of the AE data and the use of quantile regression analysis allowed the development of regression equations and normal ranges for the prediction of a normal cardiopulmonary exercise response to AE (table 3). A worked example of the regression equation can be found in the online supplemental appendix.
Table 3
•
Equations for reference values for arm ergometry cardiopulmonary exercise test parameters, based on the study sample (n=116)
Data are quantile regression analysis for pV̇O2, AT, WR, V̇E, HR, V̇E/V̇CO2 slope, V̇O2/WR slope and O2 pulse. For entering subject’s characteristics, sex (S) is coded as 0 for males and 1 for females, A=age in years, H=height in metres and W=weight in kilograms.
Validation
To determine the fit of the model for each exercise parameter, the predictive accuracy was measured in a separate group of healthy volunteers (n=16). The validation group was matched to the study group for age (p=0.299), height (p=0.100), weight (p=0.635) and BMI (p=0.070). Sex distribution was also comparable between the groups (Pearson’s χ2=0.000, p=1.000). The results of the Wilcoxon signed rank test and Spearman’s r and the statistical significance are summarised in table 4.
Table 4
•
Results of reference equation validation
The new reference values were also compared with the predicted equations of Kenney, Balady, Wasserman and Manfre (Healthy and Cardiac). Predicted values for each equation were calculated for study participants and then plotted against age (figure 3). There was a statistically significant difference between measured pV̇O2 and all currently available AE predicted values; p<0.001.
A comparison of the arm ergometry predicted V̇O2 equations according to age. Predicted values were calculated using the demographic data for the 116 study participants. V̇O2, peak oxygen uptake.
Discussion
We believe this to be the first study providing data for the development of regression equations for pV̇O2, anaerobic threshold, WR, peak VE, HR, oxygen pulse, V̇E/V̇CO2 and V̇O2/WR slopes obtained by the performance of AE CPET. Data were obtained from 116 normal healthy volunteers, which represents the largest sample size to date for AE reference values. In addition, validation of the regression values has demonstrated an ability to accurately predict CPET parameters in a separate cohort of subjects with a significant correlation between predicted and actual measurements of pV̇O2, anaerobic threshold, WR, peak VE and oxygen pulse. Despite a lack of correlation for HR and V̇E/V̇CO2 and V̇O2/WR slopes, there was no significant difference between the measured and predicted values.
Compared with other AE reference values, a strength of this study is the inclusion of a greater proportion of female participants at 53.4% and a wider age range of participants. As with the cycle ergometry predicted equations of Gläser et al we also included individuals with a BMI>30 kg/m2.27 It is recognised that obesity affects gas exchange; however, to exclude these individuals would ultimately limit the use of these equations in the current clinical population.28 It is felt that the statistical methods employed and the inclusion of weight in the regression equations allow an accurate determination of AE reference values across a wider range of individuals, including clinical cohorts.
In addition to the methodological and equipment disadvantages of the previous studies reporting reference values for AE, three of the developed regression equations used peak watts to determine predicted V̇O,2 which is in significant contrast to published equations for cycle ergometry. Systematic reviews of cycle ergometry and treadmill reference values illustrate that maximal oxygen uptake is dependent on age, sex and anthropometric properties and, therefore, it is these parameters that should be used in the development of reference values.29 30 Comparison of our data to that of previously available reference values demonstrates a significant difference with the current equations significantly underestimating pV̇O2. The equations which used wattage demonstrated the largest discrepancy in pV̇O2 with median differences between predicted equations and measured values varying from 236.75 to 1072.60 mL/min.
Our study methodology was designed to mirror the protocols used in cycle ergometry allowing easier adoption within clinical practice. The currently available reference equations all used step ramp protocols with WR increases of between 10 and 25 watts, with one study using a discontinuous protocol. These protocols were not individualised for each subject, with the same WR protocol used for both males and females and across the age ranges and were significantly greater than our median protocol of 7.5 watts. It is widely recognised that higher intensity exercise leads to increased fatigue and earlier termination of exercise, which may explain the underestimation of pV̇O2 by these studies. It has also been suggested that this methodology explains how reference values developed in a non-conditioned population8 significantly underestimated the measured pV̇O2 in a population with coronary artery disease.31
Our incremental ramp protocol was individualised per participant using the AE WR predicted equations of Copper and Storer.16 Our median total exercise test time of 14.27 min suggests that for some individuals, these equations may underestimate peak WR leading to an extended test duration. Further work needs to be undertaken to determine if the implementation of the predicted peak wattage determined by our cohort leads to a more optimal exercise test time.
The lack of widely available reference equations for AE led to the suggestion that 50%–70% of cycle ergometry V̇O2 reference values be used as an estimate of AE predicted values.21 32 Our data demonstrate that this is an underestimation and that pV̇O2 determined via AE more accurately aligns to 80% of CE predicted VO2 values. The arbitrary value of 70% was suggested based on studies of single sex and limited age range populations and, therefore, is unlikely to be representative of the wider population.33–35 This relatively broad predicted range (50%–70%) does not carry sufficient accuracy when utilising CPET data to make clinical decisions regarding surgical intervention or disease prognostication. Consequently, prior to the current study, interpretation of AE exercise test results carried significant difficulty which may be an important reason as to why this exercise modality has not been more widely adopted into routine clinical practice, despite its advantages in some patient groups.
The absolute physiological values (pV̇O2 and anaerobic threshold) measured during AE have been demonstrated to be lower than those obtained when using CE.9 36 This is likely attributed to the larger muscle mass of the legs37 38 and the mechanical differences of cycle and arm ergometers.11 The muscles used during AE CPET are predominantly the biceps, triceps, brachial and deltoid muscles. In contrast, during cycle ergometry, the primary muscles used are the quadriceps, hamstrings, gastrocnemius, soleus plantaris and gluteus maximus. Therefore, the muscles used when arm cycling are smaller and, compared with the legs, which are used for everyday activities such as walking, are generally less well conditioned. A smaller muscle needs to develop a greater percentage of its maximal tension, which leads to an increase in intracellular metabolites such as H+, lactate and inorganic phosphate (Pi) resulting in skeletal muscle fatigue earlier than would occur in a larger muscle group.37 38
In addition to muscle size, the muscles of the arms have a larger percentage of Type II fibres (fast twitch) which have a higher oxygen cost than slow twitch,39 resulting in acute fatigue during high-intensity dynamic exercise when there is an over-reliance on anaerobic metabolism resulting in earlier termination of exercise. A larger static exercise component during AE testing due to the requirement to grip the arm crank may also be a contributing factor.11 This is important as it has been shown that even light static exercise can induce a greater increase in HR and BP than dynamic exercise at the same V̇O2 level.40
Mean oxygen extraction in the muscle of the arms is closely related to the mean in vivo P50 value (oxygen tension when haemoglobin is 50% saturated with oxygen).41 For a given P50, the upper extremities extract less oxygen than the lower extremities, which is associated with lower oxygen conductance.41 Therefore, for a given oxygen demand, a greater oxygen delivery is needed for exercising arms than leg muscles, resulting in greater cardiovascular strain.42 This lower oxygen extraction in the arms is associated with several factors including a higher heterogeneity in blood flow distribution, shorter mean transit times, a smaller diffusing area and a larger diffusing distance in arm muscle when compared with leg muscles.41
This all leads to an earlier AT when exercising with the arms, and this needs to be considered when using AE CPET in a clinical population such as those requiring preoperative assessment. In our study cohort, 62 (53%) participants demonstrated an AT of <11 mL.kg.min with a normal range of between 9.50 and 13.30 mL.kg.min. This highlights the importance of using modality-specific recommendations when using CPET for preoperative risk stratification and prognosis. Further work is required to identify AE CPET risk thresholds prior to its wider adoption for preoperative assessments. The clinical application of other parameters such as the V̇E/V̇CO2 and V̇O2/WR slopes also needs to be further evaluated as they have been demonstrated in this cohort to be significantly greater than the values achieved when exercising with the legs. A higher V̇O2/WR slope can be explained by the lower mechanical efficiency when exercising with the arms, a greater contribution of accessory muscles and a higher cardiovascular strain. The higher V̇E/V̇CO2 slope could be influenced by the lower peak V̇CO2 generated when exercising with the arms. There is also the potential for compromised respiratory muscles when exercising with the arms due to the mechanical restraints of stabilising the torso and positioning the arms, which can influence tidal and end-expiratory lung volumes.43 44
The implementation of the predicted values and reference ranges developed in the current study for routine CPET parameters does have the potential, however, to improve the interpretation of AE exercise tests and standardise interpretation across sites in the UK. The data will enable the assessment of clinical populations against a normal range with the view to developing interpretation strategies for a range of clinical interventions.
Study limitations
Participants were deemed suitable for this study subject to successful completion of health screening and normal baseline measurements of spirometry, BP and ECG. It is, however, recognised that normal spirometry does not preclude a normal gas transfer (TLCO) in a symptomatic population.45 pV̇O2, V̇E/V̇CO2 slope and oxygen pulse all rely on effective oxygen diffusion and therefore the lack of TLCO data in our subject cohort may lead to bias by assuming homogeneity of lung function that does not exist. Excluding TLCO data from CPET regression equations may compromise their reliability; however, this has been demonstrated to be more of a risk in populations where potential pulmonary limitations are more likely than in our cohort.45 46
The performance of CPET relies on the concomitant measurements of ECG and BP. Both provide important clinical information on the cardiovascular response to exercise, but they also allow the identification of test termination criteria relating to patient safety. The most common site for BP measurement is the upper arm; however, AE exercise limits accurate measurement due to continual movement of the arm during testing. The lower leg is the most commonly chosen site for BP measurements when the arm is not possible; however, the large muscle bulk of the gastrocnemius, soleus and plantaris muscles leads to significantly increased participant discomfort during measurements.47 Previous studies have suggested BP measurements can be taken from the participants’ ankle using the posterior tibial artery due to ease of access, similar circumference to the arm muscles and lack of muscle bulk providing a more comfortable and tolerable mode of measurement.48
Unfortunately, the BP device used in our study repeatedly failed to obtain measurements when positioned on the ankle. Oscillometric BP measurements have previously been demonstrated to be a reliable alternative to conventional BP measurements made at the ankle.49 50 The poor reliability of the readings in the current study may, therefore, relate to the BP device itself which is not listed on the British and Irish Hypertension Societies list of recommended oscillometric devices for specialist use.51 As a result, accurate measurement of BP was not possible during AE testing. Due to the low risk of exercise-induced hypertension in the study population, this was not considered to pose a safety risk. However, prior to any further work in patient populations, it will be essential to identify equipment that will accurately measure ankle BP so that test termination criteria can be identified and acted on accordingly.
The reference values developed are from a study population aged 19–69 years of age; however, it should be recognised that only two subjects over 60 years were recruited. The applicability of the reference values to an older population is, therefore, not known. We recommend the results from individuals aged above 75 years should be interpreted with a degree of caution. The use of these predicted values and reference ranges is, however, still recommended as the alternative equations have additional significant limitations.
Our study methodology included the completion of the RPAQ to quantify the level of activity regularly undertaken by participants in an attempt to limit participation bias. Unfortunately, participants found the questionnaire difficult to accurately answer and demonstrated a tendency to over-report their sporting activities. Other studies have also suggested that these questionnaires have limited reliability and validity due to over-reporting of exercise rates and frequency, with the highest levels of physical activity having the greatest discrepancy with actual activity levels.52–54
In conclusion, the current study of 116 healthy male and female volunteers aged 19–69 years has provided data for the development of predicted values and reference ranges for V̇O2, anaerobic threshold, WR, peak VE, peak HR, oxygen pulse and the V̇E/V̇CO2 and V̇O2/WR slopes obtained using AE exercise. This is the largest data set for AE reference values to date and represents the widest age range for healthy individuals.
Contributors: JS designed the study and wrote the first draft. EP contributed to the data analysis, reviewed the first draft and contributed to subsequent revisions. JS takes full responsibility for the work and conduct of the study, had access to the data and controls the decision to publish. JS is the guarantor.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review: Not commissioned; externally peer reviewed.
Supplemental material: This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication:
Not applicable.
Ethics approval:
This study involves human participants and ethical approval for this study was granted by the Health Research Authority, Wales REC 7 (REC Reference 17/WA/0284; IRAS Project ID 226248). Participants gave informed consent to participate in the study before taking part.
Acknowledgements
This work is partially based on the authors' doctoral dissertation (Development of normal ranges for cardiopulmonary exercise testing performed using arm ergometry, Manchester Metropolitan University, 2021). We would like to thank all participants who agreed to take part in the study.
Balady GJ, Arena R, Sietsema K, et al. Clinician’s Guide to cardiopulmonary exercise testing in adults: a scientific statement from the American Heart Association. Circulation2010; 122:191–225.
Korpelainen R, Lämsä J, Kaikkonen KM, et al. Exercise capacity and mortality - a follow-up study of 3033 subjects referred to clinical exercise testing. Ann Med2016; 48:359–66.
Leeper NJ, Myers J, Zhou M, et al. Exercise capacity is the strongest predictor of mortality in patients with peripheral arterial disease. J Vasc Surg2013; 57:728–33.
Manfre MJ, Mallis GI, Varma A, et al. Oxygen Uptake, Electrocardiographic, and Symptomatic Response During Two Arm Cranking and Treadmill Testing in Healthy Older Adult Men and Patients with Coronary Artery Disease. J Cardiopulm Rehabil1990; 10:12–20.
Dekerle J, Dupont L, Caby I, et al. Ventilatory thresholds in arm and leg exercises with spontaneously chosen crank and pedal rates. Percept Mot Skills2002; 95:1035–46.
Pina IL, Balady GJ, Hanson P, et al. Guidelines for clinical exercise testing laboratories. A statement for healthcare professionals from the Committee on Exercise and Cardiac Rehabilitation, American Heart Association. Circulation1995; 91:912–21.
Bartsokas TW, Earnest CP, Heiselman DE, et al. MAXIMUM CARDIORESPIRATORY RESPONSES OF MEN AND WOMEN DURING ARM CRANK ERGOMETRY: EFFECT OF CRANKING SPEED: 519. Med Sci Sports Exerc1989; 21.
Price MJ, Collins L, Smith PM, et al. The effects of cadence and power output upon physiological and biomechanical responses to incremental arm-crank ergometry. Appl Physiol Nutr Metab2007; 32:686–92.
Radtke T, Crook S, Kaltsakas G, et al. ERS statement on standardisation of cardiopulmonary exercise testing in chronic lung diseases. Eur Respir Rev2019; 28.
Quanjer PH, Stanojevic S, Cole TJ, et al. Multi-ethnic reference values for spirometry for the 3-95-yr age range: the global lung function 2012 equations. Eur Respir J2012; 40:1324–43.
Gläser S, Ittermann T, Schäper C, et al. The Study of Health in Pomerania (SHIP) reference values for cardiopulmonary exercise testing. Pneumologie2013; 67:58–63.
Milani J, Fernhall B, Manfredi T, et al. Estimating oxygen consumption during treadmill and arm ergometry activity in males with coronary artery disease. J Cardiopulm Rehabil1996; 16:394–401.
Larsen RT, Christensen J, Tang LH, et al. A SYSTEMATIC REVIEW AND META-ANALYSIS COMPARING CARDIOPULMONARY EXERCISE TEST VALUES OBTAINED FROM THE ARM CYCLE AND THE LEG CYCLE RESPECTIVELY IN HEALTHY ADULTS. Int J Sports Phys Ther2016; 11:1006–39.
Franklin BA, Vander L, Wrisley D, et al. Aerobic Requirements of Arm Ergometry: Implications for Exercise Testing and Training. Phys Sportsmed1983; 11:81–90.
Sawka MN, Foley ME, Pimental NA, et al. Determination of maximal aerobic power during upper-body exercise. J Appl Physiol Respir Environ Exerc Physiol1983; 54:113–7.
Reybrouck T, Heigenhauser GF, Faulkner JA, et al. Limitations to maximum oxygen uptake in arms, leg, and combined arm-leg ergometry. J Appl Physiol1975; 38:774–9.
Muangkram Y, Honda M, Amano A, et al. Exploring the role of fatigue-related metabolite activity during high-intensity exercise using a simplified whole-body mathematical model. Inform Med Unlocked2020; 19:100355.
Ørtenblad N, Nielsen J, Boushel R, et al. The Muscle Fiber Profiles, Mitochondrial Content, and Enzyme Activities of the Exceptionally Well-Trained Arm and Leg Muscles of Elite Cross-Country Skiers. Front Physiol2018; 9.
Calbet JAL, Holmberg H-C, Rosdahl H, et al. Why do arms extract less oxygen than legs during exercise? Am J Physiol Regul Integr Comp Physiol2005; 289:R1448–58.
Calbet JAL, González‐Alonso J, Helge JW, et al. Central and peripheral hemodynamics in exercising humans: leg vs arm exercise. Scandinavian Med Sci Sports2015; 25:144–57.
Alison JA, Regnis JA, Donnelly PM, et al. End-expiratory lung volume during arm and leg exercise in normal subjects and patients with cystic fibrosis. Am J Respir Crit Care Med1998; 158:1450–8.
Tiller NB, Campbell IG, Romer LM, et al. Influence of Upper-Body Exercise on the Fatigability of Human Respiratory Muscles. Med Sci Sports Exerc2017; 49:1461–72.
Rasch-Halvorsen Ø, Hassel E, Brumpton BM, et al. The association between normal lung function and peak oxygen uptake in patients with exercise intolerance and coronary artery disease. PLoS One2020; 15.
Beckman JA, Higgins CO, Gerhard-Herman M, et al. Automated oscillometric determination of the ankle-brachial index provides accuracy necessary for office practice. Hypertension2006; 47:35–8.
Chongthawonsatid S, Dutsadeevettakul S. Validity and reliability of the ankle-brachial index by oscillometric blood pressure and automated ankle-brachial index. J Res Med Sci2017; 22.
Durante R, Ainsworth BE. The recall of physical activity: using a cognitive model of the question-answering process. Med Sci Sports Exerc1996; 28:1282–91.
