Validation of a brief step-test protocol for estimation of peak oxygen uptake

Transcription

1 Original scientific paper Validation of a brief step-test protocol for estimation of peak oxygen uptake Frank Beutner 1,2, *, Romy Ubrich 1, *, Silke Zachariae 1,3, Christoph Engel 1,3, Marcus Sandri 2, Andrej Teren 1,2 and Stephan Gielen 1,4 EU RO PEAN SOCIETY OF CARDIOLOGY European Journal of Preventive Cardiology () 1 1! The European Society of Cardiology 214 Reprints and permissions: sagepub.co.uk/journalspermissions.nav DOI: / ejpc.sagepub.com Abstract Background: Physical exercise capacity has been shown to predict cardiovascular disease incidence and is increasingly measured in epidemiological studies. However, direct measurement of peak oxygen uptake is too time consuming in large-scale studies. We therefore investigated whether a brief 3-minute step-test protocol can be used to estimate peak oxygen uptake in these settings. Design and methods: A group of 97 subjects performed the YMCA step test and a maximal treadmill test with continuous measurement of oxygen uptake. Correlation and linear regression analyses were used to identify VO 2 peak predictors obtained from the step test and to develop models for VO 2 peak estimation. Results: The YMCA model, including the 1-minute heart beat count, predicted VO 2 peak with R ¼.83. A novel simplified model based on the heart rate at 45 s of recovery performed comparable (R ¼.83). However, models based on heart rate measures were only valid in subjects who completed the test according to protocol, but not in subjects who terminated prematurely. For the applicability in subjects with low exercise capacity, a new model including gas exchange analysis enabled prediction of VO 2 peak (R ¼.89). All models were validated in an independent sample (r ¼.86.91). Exercise time of the step test was less than one-hird of standard ergospirometry (treadmill test: s, step test: 18 s, p <.1). Conclusion: In large-scale epidemiological studies with limited time slots for exercise testing and significant proportions of subjects with low exercise capacity a modified version of the YMCA step test may be used to predict VO 2 peak. Keywords step test, maximum oxygen uptake, VO 2 max, VO 2 peak, exercise testing Received 7 November 213; accepted 7 April 214 Introduction Physical fitness has been identified as a predictor of mortality and cardiovascular events in long-term follow-up studies both among cardiovascular patients and healthy subjects. 1 7 However, standard maximal exercise testing using direct measurement of peak oxygen uptake (VO 2 peak) is time consuming and may therefore not be feasible in certain settings (e.g. largescale epidemiological studies, primary health care, etc.). 8 In consequence, there is a clear need for a brief cardiovascular fitness test, which can be performed as part of a multifaceted health assessment battery. Single-stage, fixed-time step tests, such as the commonly used Young Men s Christian Association (YMCA) step test, could be used to predict VO 2 peak. 9 The step test was originally developed by Kasch in 1961, using the heart beat count (HBC) after the first minute of recovery to predict maximum oxygen uptake. 1,11 1 LIFE Leipzig Research Center for Civilization Diseases, Medical Faculty, University Leipzig, Leipzig, Germany 2 Heart Center University Leipzig, Department of Internal Medicine/ Cardiology, Leipzig, Germany 3 Institute for Medical Informatics, Statistics and Epidemiology, University Leipzig, Leipzig, Germany 4 Department of Medicine III, Martin Luther University Halle-Wittenberg, University Hospital Halle, Halle/Saale, Germany *Frank Beutner and Romy Ubrich contributed equally to this paper. Corresponding author: Stephan Gielen, Department of Internal Medicine III, University Hospital, Martin Luther University of Halle/Wittenberg, Ernst-Grube-Strasse 4, 612 Halle, Germany. stephan.gielen@uk-halle.de

2 2 European Journal of Preventive Cardiology () The National YMCA recognized the test as a simple and time-effective method for categorization of individual fitness levels in the target population of young healthy men and adopted Kasch s step test as part of the National YMCA Physical Fitness Test Battery. Robust correlations have been repeatedly observed between VO 2 peak and HBCs obtained from modified step test protocols However, alternative cardiorespiratory measures during a step test have never been studied for estimation of individual VO 2 peak. Thus, the aim of the present study was: 1. to analyse the individual physiological response during the YMCA step test using continuous heart rate monitoring and gas exchange measurement; 2. to compare VO 2 peak prediction by a model using the standard 1-minute post-exercise HBC with new models using multivariate regression based on additional cardiorespiratory measurements during the step test; and 3. to reflect whether the YMCA step test is suited for VO 2 peak prediction in population-based studies. Methods Study sample Subjects were recruited as participants of the LIFE Project (Leipzig Research Center for Civilization Diseases), a combined large-scale population- and disease-based epidemiological cohort study conducted in the city of Leipzig, Germany. LIFE aims to study the interaction between genetic, environmental and individual lifestyle factors on the risk of various civilization diseases. In this feasibility study a total of 111 subjects (55 men, 56 women aged years) completed a maximum treadmill test in addition to the selected exercise test battery including step test, hand grip test and physical activity questionnaires. Subjects with ECG abnormalities, severe pulmonary or cardiovascular diseases, orthopaedic disabilities or taking beta blocker medication were excluded from the active exercise studies. The Ethics Committee at the University of Leipzig approved the study protocol. All participants gave their written informed consent prior to study inclusion. Study protocol Testing was performed between 7. a.m. and 3. p.m. Exercise capacity was assessed by (1) the standardized YMCA step test with continuous heart rate monitoring by a single channel ECG and a breath-by-breath analysis of oxygen uptake and (2) a maximal cardiopulmonary treadmill exercise test with determination of VO 2 peak (at respiratory exchange rates > 1.1). A minimum period of 2 hours of rest was given between the two exercise tests. Participants were introduced to the programme and familiarized with the test protocols before starting the test battery. Step test We adopted basic characteristics of the YMCA step test including a 12-inch bench height, a duration of 3 minutes and the metronome-based step frequency (96 beats per minute) of 24 step cycles per minute. 9 Instead of counting post-exercise 1-minute HBC manually, we continuously recorded the heart rate using a POLAR single channel ECG chest belt device (RS8CX, POLAR Electro, Germany) starting at 2 minutes prior to stepping and finishing 5 minutes after workload termination. The POLAR system together with a wireless electrode chest belt allows detection of R-R intervals with a resolution of 1 ms. 18 POLAR was set to store heart rate in 5 s intervals continuously during all test phases. The test started with a 2-minute resting phase while subjects were seated on a chair in a temperature- and humidity-controlled quiet room. Subjects had to step 72 times in 3 minutes on the prepared box (step upup-down-down). Stepping frequency was indicated by a metronome. Finally, subjects were seated again during a 5-minute recovery phase. Participants were categorized in sets of subjects who finished the step test conforming to protocol, i.e. kept up stepping for 3 minutes and managed to do at least 7 steps (completers) and subjects who terminated the test prematurely or did not achieve 7 steps (noncompleters). The 1-minute heart beat count (1 min-hbc) as defined by the original YMCA step test was approximated calculating the mean of twelve consecutive POLAR heart rate records in 5 s-intervals, starting 5 s after workload termination. Similarly, we calculated continuous HBCs in 5 s intervals from 15 s to 2 minutes of the post-exercise. In order to evaluate the exercise intensity of the YMCA step-test protocol, the oxygen uptake was measured using a breath-by-breath gas exchange analysis system (MasterScreen CPX, CareFusion, Hoechberg, Germany) during the step test. Standard treadmill cardiopulmonary exercise testing The peak oxygen uptake (VO 2 peak) was measured using the same breath-by-breath analysis system during an exhaustion-limited treadmill exercise test (MasterScreen CPX, CareFusion, Hoechberg,

3 Beutner et al. 3 Germany). According to the ACC/AHA guidelines for exercise testing, the treadmill protocol was customized individually to achieve maximal capacity within 6 12 minutes. 19 Self-reported exercise capacity by the Veterans Physical Activity Questionnaire (VSAQ) was used for protocol alignment. 2 The Bruce protocol was used for individuals having VSAQ metabolic equivalent (MET) 9 and the modified Bruce protocol was used for individuals having VSAQ MET 6. Individuals having intermediate VSAQ MET 7/8 were assigned to Bruce protocol when step-test criteria were fulfilled; otherwise they were assigned to the modified protocol. The protocol began with a 1-minute warm-up, during which participants walked comfortably at selfpaced speed and at % grade. The workload was then increased according to the Bruce or modified Bruce protocol, respectively. 21 Oxygen uptake was considered maximal if any two of the following criteria were met: (a) respiratory exchange ratio 1.1; (b) a heart rate at or near agepredicted peak (22-age 1 beats/min); and (c) when further increase in workload did not result in a significant rise in oxygen uptake (<15 mlkg 1 min 1 ). These criteria were met in 97 of 111 subjects. Fourteen subjects who did not achieve these criteria were excluded from the analyses. Age- and gender-adjusted predicted VO 2 peak was determined using the treadmill equation of Wassermann et al. 22 Statistical analysis Statistical analyses were done with IBM SPSS 2.. Measures are given as mean standard deviation. Means were compared using Student s t-tests. Pearson s product-moment correlations and linear regression models were used to validate the standard YMCA step-test measure as well as to develop and validate new models for VO 2 peak prediction. The study sample was randomly split into a learning set and an independent validation set (3:1). The analysis was performed in three steps:. Identification of VO 2 peak predictors (learning set): we correlated VO 2 peak with heart rate (HR) based parameters, which were derived from continuous heart rate measurement during the step test (including heart rate, HBCs, heart rate reserve, heart rate recovery) and respiratory parameters, which were derived from continuous breath-bybreath gas exchange (GE) analysis (respiratory exchange ratios (RERs), oxygen uptake).. Development of VO 2 peak-predictive models (learning set): we integrated the parameters showing the highest correlation with VO 2 peak into a linear regression model (including age and gender) and compared them with the model including the standard YMCA 1 min-hbc.. Model verification and validation: the HBC-based YMCA model and the novel HR- and GE-based models were validated in an independent subset of the investigated sample (validation set). P-values <.5 were considered statistically significant. Results Baseline characteristics and results of exercise testing are presented for completers and non-completers of the step test, the learning and validation set, respectively (Table 1). Observed peak oxygen uptake (treadmill test) ranged from 21 to 66 mlkg 1 min 1 (6 19 metabolic equivalents METs). Fifteen per cent of participants were unable to complete the step test according to protocol due to physical exhaustion. Age > years, female gender, BMI >29 kgm 2, VSAQ score <9 and VO 2 peak <29 mlkg 1 min 1 were significant predictors for failing to complete the step test. Sixtyseven per cent of participants reached 8% of their predicted maximum heart rate during the step test. Analyses of continuous heart rate monitoring The time course of changes in heart rate during the YMCA step test are shown in (Figure 1a). The applicability of heart rate measures to predict VO 2 peak was restricted to subjects who completed the step test, while it failed in non-completers (Figure 1b). Continuous correlation analysis determined heart rate at 45 s of recovery (45 s-phr) as the time point having the highest correlation with VO 2 peak (Figure 1c). Courses of the post-exercise heart beat counts are shown in Figure 1d. Among these the standard measure of the YMCA step test the 1-minute heart beat count (1 min-hbc) was best correlated with VO 2 peak (Figure 1e and f). 45 s- phr and 1 min-hbc were correlated with r ¼.992. Further investigated HR-associated parameters (e.g. heart rate reserve, heart rate recovery) were also correlated with VO 2 peak, but inferior compared to 45 s-phr and 1 min-hbc (data not shown). Value of gas exchange analysis during the YMCA step test To evaluate the exercise intensity of the YMCA step-test protocol we analysed the oxygen uptake at workload termination in 82 completers (47 men). The mean

4 4 European Journal of Preventive Cardiology () Table 1. Baseline characteristics and results of exercise testing. Total Completer Non-completer Learning Set Validation Set n Age (years) b Male 51 (53%) 47 (57%) 4 (27%) a 4 (55%) 11 (46%) BMI (kg m 2 ) b VSAQ b Treadmill test Modified Bruce protocol 19 (2%) 8 (1%) 11 (73%) b 16 (22%) 3 (13%) VO 2 peak (ml kg 1 min 1 ) b %predicted VO Peak RER b Duration (s) Step test Non-completer 19 (17%) (%) 15 (1%) 11 (15%) 4 (17%) VO 2 step (ml kg 1 min 1 ) b Peak RER b Duration (s) b Notes: Statistical significance was tested for completer vs. non-completer and learning set vs. validation set. Significance level: a p <.5; b p <.1. oxygen uptake at workload termination (VO 2 step) was mlkg 1 min 1 in men and mlkg 1 min 1 in women (p ¼.74) equivalent to MET and MET, respectively. Completers achieved significantly higher maximum oxygen uptake during the step test (VO 2 step) compared to non-completers ( vs mlkg 1 min 1, p <.1, Figure 2a). Respiratory exchange ratio at workload termination indicated maximal exercise exhaustion in non-completers (RER 1.2.1), while completers remained submaximal (RER.99.12, p <.1). VO 2 step was suitable to predict VO 2 peak in non-completers and RER was predictive in completers. Development and validation of the HBC-based prediction of VO 2 peak (YMCA model) A linear regression model including age, gender and 1 min-hbc (YMCA model) predicted VO 2 peak with R ¼.828 in completers of the learning set. Transferring the YMCA model to the validation set, the predicted VO 2 peak showed a good correlation with the observed VO 2 peak (r ¼.862, p <.1, Figure 3a). Standard error of estimate (SEE) was 5.5 mlkg 1 min 1. Development and validation of a novel model for VO 2 peak prediction based on heart rate parameters Age, gender and 45 s-phr were selected as predictors to estimate VO 2 peak in completers of the learning set (R ¼.833). In the validation set, the predicted VO 2 peak correlated well with the observed VO 2 peak (r ¼.868, p <.1, Figure 3b) with a SEE of 5.3 mlkg 1 min 1. Development and validation of a novel model for VO 2 peak prediction based on step-test gas exchange measurements Age, gender and RER at work termination were selected as predictors to estimate VO 2 peak in completers (R ¼.833). VO 2 step was the only sufficient predictor in non-completers (R ¼.885). By combining both models, prediction of VO 2 peak improved compared to the YMCA model (R ¼.89). In the validation set, the predicted VO 2 peak correlated with observed VO 2 peak (r ¼.96, p <.1, Figure 3c) with a SEE of 4.7 mlkg 1 min 1. Table 2 shows details of the parameters included in the described models. Test duration The treadmill test took a mean of s with a range of s. The exercise time of the original 3-minute step test was significantly shorter 18 s for completers and s for non-completers due to preterm exhaustion with a global range of 1 18 s (p <.1 compared to treadmill test).

5 Beutner et al. 5 Heart rate Heart beat count (a) 1 (d) 25 Course Beats per minute (b) /TT Heart beat count (e) /TT Correlation with VO 2 peak Parameter with best Correlation Correlation (abs,value of R) (c) 7 Peak oxygen uptake sec-pHR C-R=.625 NC-R=.52 Correlation (abs,value of R) Peak oxygen uptake /TT (f) 7 1min-HBC C-R=.582 NC-R= s recovery heart rate 1min recovery heart beat count Non-completer Completer (VO 2 peak < 4ml kg 1 min 1 ) All completer Completer (VO 2 peak 4ml kg 1 min 1 ) Figure 1. Results from continuous heart rate analysis. Courses of continuous monitored heart rate (a) during the step test and recovery heart beat count (d) are shown for non-completers and completers; the completers were separated in subjects with performance lower or higher than 4 mlkg -1 min -1. Correlation with peak oxygen uptake (VO 2 peak in ml kg 1 min 1 ) is shown continuously for heart rates along the whole step test (b) and for heart beat counts during the recovery phase (e). Time points of highest correlation with VO 2 peak were selected for heart rate at 45 s of the recovery phase (45 s-phr) (c) and for the HBC at s of the recovery phase (1 min-hbc) (f).

6 6 European Journal of Preventive Cardiology () Oxygen uptake RER (a) (d) 1.6 Course ml kg 1 min 1 (b) /TT RER (e) /TT Correlation with VO 2 peak Correlation (abs,value of R) Correlation (abs,value of R) (c) (f) 7 Step test Oxygen Uptake 7 RER test termination C-R=.58 C-R=.734 Parameter with best Correlation Peak oxygen uptake 5 4 NC-R=.92 Peak oxygen uptake 5 4 NC-R= Step test oxygen uptake RER step test Non-completer Completer (VO 2 peak < 4ml kg 1 min 1 ) All completer Completer (VO 2 peak 4ml kg 1 min 1 ) Figure 2. Results from continuous gas exchange analysis. Courses of continuous recorded oxygen uptake (a) and respiratory exchange ratio (RER) (d) during the step test are shown for non-completers and completers; the completers were separated in subjects with performance lower or higher than 4 mlkg 1 min 1. Correlation of VO 2 peak with oxygen uptake (b) and RER (e) is continuously shown along the whole step test. Time points of highest correlation with VO 2 peak were selected for the oxygen uptake at workload termination (VO 2 step) (c) in non-completers and for the RER at workload termination for completers (f).

7 Beutner et al. 7 (a) 7 Observed peak oxygen uptake Learning set (LS) Validation set (VS) YMCA model LS-R=.828 VS-R=.862 (b) 7 Observed peak oxygen uptake 5 4 HR based model LS-R=.833 VS-R=.868 (c) Predicted peak oxygen uptake Predicted peak oxygen uptake Predicted peak oxygen uptake Observed peak oxygen uptake 5 4 GE based model LS-R=.89 VS-R=.86 Figure 3. Correlations between model-predicted peak oxygen uptake and observed peak oxygen uptake. Correlations are shown for the YMCA model based on the original one-minute heart beat count (a) and novel heart rate-based (b) and gas exchange-based (c) models that were developed in this study. Learning set is represented by empty dots, validation set by black dots. Accuracy of predictive models is given as standard error of the estimates (SEE). Table 2. Development of VO 2 peak-predictive models using linear regression analysis (learning set). R P-value Regression coefficient Lower 95% CI Upper 95% CI YMCA model Completer.828 Constant < Age < Gender min-hbc < HR-based model Completer.833 Constant < Age < Gender s-phr < Non-completer.9 Constant.68 Age.842 Gender.997 Continuous HR-analysis NS Step count.113 GE-based model Completer.847 Constant < Age < Gender < s-rer < Non-completer.885 Constant VO 2 step < Combined.89

8 8 European Journal of Preventive Cardiology () Discussion The following key messages emerge from this feasibility study using the YMCA step test combined with continuous heart rate monitoring and oxygen uptake analysis as a surrogate for standard VO 2 peak measurement in a population-based study sample:. In subjects who are able to exercise equal or higher to 9 MET, the original YMCA step-test HBC is well correlated with exercise capacity and adequately predicts VO 2 peak using an age- and gender-adjusted model. The majority of subjects with <9 MET exercise capacity is not able to complete the step test. In those subjects the original YMCA step-test HBC fails to predict maximal exercise capacity.. An alternative model developed in the current study based on another heart rate parameter (i.e. heart rate at 45 s of the recovery phase) performed comparably to the YMCA model. As with the YMCA model it was limited to subjects with exercise capacity of 9 MET.. A novel model developed in the current study based on gas exchange measurement parameters broadens the applicability of the step test to subjects with lower exercise performance (<9 MET).. The step-test protocol is time saving in large-scale trials with testing time being two-thirds quicker than a standard maximum treadmill test. The original YMCA step test was used as a submaximal stress test targeted at healthy young men until now. However, the majority of participants in this population-based sample reached >8% of their predicted maximum heart rate, which demonstrates that the exercise level achieved in the step test exceeded submaximal exercise levels although all participants had adequate or even superior performance relating to their age and gender. A total of 15% of the whole sample and 33% of subjects older than sixty years did not complete the test. However, the interpretation of the original YMCA step test requires finishing the test according to the protocol including step time and step pace. Continuous heart rate monitoring in the current study did not reveal any heart-rate-based parameter that enabled prediction of VO 2 peak in non-completers. Therefore, we allowed the use of gas exchange parameters for the development of a new model predicting VO 2 peak in all subjects who take the step test irrespective of its completion. Good prediction of VO 2 peak in step-test completers Heart rate measures were determining factors of VO 2 peak estimation only in subjects who completed the step test according to protocol. The standard measure of the YMCA step test, the post-exercise 1-minute heart beat count, was significant in predicting VO 2 peak and resulted in r and SEE values comparable to other studies using modified step-test protocols. 15 The impact of standard HBC to estimate VO 2 peak was validated in an independent sample of subjects fulfilling the test criteria (r ¼.87). The optimal time period for the recovery HBC has been questioned in previous studies since Watkins and McArdle reported greatest reliability at 2 s of recovery, while Santo found no difference between the 15 s and s HBC. 15,23,24 In our study, continuous heart rate analysis identified the recovery HBC at s as a good predictor for VO 2 peak. We also identified heart rate at 45 s of the recovery phase as a further predictor of VO 2 peak. As heart rate at 45 s of recovery and 1 min-hbc were highly correlated (r ¼.99), the heart rate at 45 s recovery may be a surrogate for the HBC. Additional value of oxygen uptake measurements in step-test non-completers Gas exchange measurement during the step test abolishes the ease of the intended simple step test. However, the current study revealed that heart rate recovery and other heart rate measures of the step test did not predict VO 2 peak in subjects who did not complete the test, while gas exchange analysis provided valuable information in these subjects. Peak oxygen uptake during the step test was strongly correlated with peak oxygen uptake during the maximum treadmill test in subjects who did not complete the step test (r ¼.89). On the other hand, in completers there was only a weak correlation between peak oxygen uptake during the step test and VO 2 peak. The weakness of the latter correlation can be explained by a ceiling effect in VO 2 step, based on the fact that the step test represents only a submaximal workload to subjects with moderate-to-high cardiorespiratory fitness, whereas it applies a maximal workload to non-completers. In our model, including gas exchange analysis, we avoided the ceiling effect in the subgroup of step-test completers by using the respiratory exchange ratio for calculation, which was highly correlated with VO 2 peak in subjects who completed the step test conforming to protocol. The gas-exchange based model does not notably improve prediction of VO 2 peak compared to heartrate-based models in subjects who complete the step test conforming to protocol. Therefore, the effort of additional gas exchange measurement is only required in subjects with expected low performance and a possible premature end of the test. To assess the expected exercise capacity, the Veterans Specific Activity Questionnaire (VSAQ) might be a suitable instrument.

9 Beutner et al. 9 Safety aspects of the step test Cardiopulmonary exercise testing and outpatient exercise training are overall safe procedures, even if there is a high risk of cardiovascular diseases in the investigated cohort However, the majority of subjects reached >8% of their predicted maximum heart rate in the current study despite their adequate or superior ageand gender-related performance. It has to be assumed that the YMCA step test does not represent a strict submaximal exercise test in a cohort including female and elderly subjects. All models developed in the present study were validated in an independent sample of the study, which has never been previously done for any step test. The required precision of VO 2 peak prediction depends on the intended purpose of use. VO 2 peak estimates from our models would not be appropriate for precise measurements of maximal exercise capacity in high-performance sports, but may be an adequate and objective solution for cardiorespiratory fitness testing in primary health care and population-based studies. Limitations In population-based studies including a significant proportion of low and intermediate performers the original YMCA step test requires attention and safety precautions comparable to other maximum exercise tests. The heart rate based models for VO 2 peak estimation are not appropriate for subjects with beta blocker intake or atrial fibrillation. The impact in those subjects has to be investigated, particularly with regard to the wide use of beta blockers in older subjects. The practical applicability of the step test is limited in orthopaedic conditions (such as knee or hip arthritis). Patients with such conditions were excluded in the present study. Conclusion The YMCA single stage step test is appropriate for estimating individual peak oxygen uptake in settings with limited time resources. The YMCA step-test protocol could be ideal for primary health care settings and population-based epidemiological studies to obtain an objective measurement of exercise capacity and physical fitness in less than 5 minutes. We were able to confirm that the original YMCA step-test model accurately predicts peak oxygen uptake in a fit and healthy population (9 MET). In the current study we extracted another heart rate parameter that equally predicts peak oxygen uptake, but is easier and more precise to obtain. We propose a novel model utilizing gas exchange parameters to permit the application of the step test and the prediction of peak oxygen uptake in subjects with lower cardiorespiratory fitness. Acknowledgements The work is part of Romy Ubrich s MD thesis. The authors thank Antonia Christen and Kay Olischer for their technical assistance. Funding This publication is supported by LIFE Leipzig Research Center for Civilization Diseases, University Leipzig. LIFE is funded by the European Union, by the European Regional Development Fund (ERDF) and by the Free State of Saxony within the framework of the excellence initiative. Conflict of interest The authors declare that they have no conflict of interest. References 1. Myers J, Prakash M, Froelicher V, et al. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med 22; 346: Ekelund LG, Haskell WL, Johnson JL, et al. Physical fitness as a predictor of cardiovascular mortality in asymptomatic North American men. The Lipid Research Clinics Mortality Follow-up Study. N Engl J Med 1988; 319: Blair SN, Kohl 3rd HW, Paffenbarger RS, Jr., et al. Physical fitness and all-cause mortality. A prospective study of healthy men and women. JAMA 1989; 262: Lakka TA, Venalainen JM, Rauramaa R, et al. Relation of leisure-time physical activity and cardiorespiratory fitness to the risk of acute myocardial infarction. N Engl J Med 1994; 3: Nocon M, Hiemann T, Muller-Riemenschneider F, et al. Association of physical activity with all-cause and cardiovascular mortality: a systematic review and meta-analysis. Eur J Cardiovasc Prev Rehabil 28; 15: Mandic S, Myers J, Oliveira RB, et al. Characterizing differences in mortality at the low end of the fitness spectrum in individuals with cardiovascular disease. Eur J Cardiovasc Prev Rehabil 21; 17: Kokkinos P, Myers J, Doumas M, et al. Heart rate recovery, exercise capacity, and mortality risk in male veterans. Eur J Prev Cardiol 212; 19: Vanhees L, Lefevre J, Philippaerts R, et al. How to assess physical activity? How to assess physical fitness? Eur J Cardiovasc Prev Rehabil 25; 12: Golding LA. YMCA fitness testing and assessment manual. Champaign, IL: Human Kinetics, Kasch FW. A comparison of exercise tolerance of postrheumatic and normal boys. J Assoc Phys Ment Rehabil 1961; 15: Kasch FW, Phillips WH, Ross WD, et al. A comparison of maximal oxygen uptake by treadmill and step-test procedures. J Appl Physiol 1966; 21:

10 1 European Journal of Preventive Cardiology () 12. Siconolfi SF, Garber CE, Lasater TM, et al. A simple, valid step test for estimating maximal oxygen uptake in epidemiologic studies. Am J Epidemiol 1985; 121: Francis K and Culpepper M. Height-adjusted, rate-specific, single-stage step test for predicting maximal oxygen consumption. South Med J 1989; 82: Francis K and Brasher J. A height-adjusted step test for predicting maximal oxygen consumption in males. J Sports Med Phys Fitness 1992; 32: Santo AS and Golding LA. Predicting maximum oxygen uptake from a modified 3-minute step test. Res Q Exerc Sport 23; 74: Chatterjee S, Chatterjee P, Mukherjee PS, et al. Validity of Queen s College step test for use with young Indian men. Br J Sports Med 24; 38: Chatterjee S, Chatterjee P and Bandyopadhyay A. Validity of Queen s College Step Test for estimation of maximum oxygen uptake in female students. Indian J Med Res 25; 121: Gamelin FX, Berthoin S and Bosquet L. Validity of the polar S81 heart rate monitor to measure R-R intervals at rest. Med Sci Sports Exerc 26; 38: Gibbons RJ, Balady GJ, Beasley JW, et al. ACC/AHA guidelines for exercise testing: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Exercise Testing). Circulation 1997; 96: Myers J, Bader D, Madhavan R, et al. Validation of a specific activity questionnaire to estimate exercise tolerance in patients referred for exercise testing. Am Heart J 21; 142: Chaitman BR. Exercise stress testing. In: Libby P, Bonow RO, Mann DL, et al. (eds) Braunwald s heart disease: A textbook of cardiovascular medicine, 8th ed. Amsterdam: Elsevier, 28, pp Guazzi M, Adams V, Conraads V, et al. EACPR/AHA Joint Scientific Statement. Clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Eur Heart J 212; 33: Watkins J. Step tests of cardiorespiratory fitness suitable for mass testing. Br J Sports Med 1984; 18: McArdle WD, Katch FI, Pechar GS, et al. Reliability and interrelationships between maximal oxygen intake, physical work capacity and step-test scores in college women. Med Sci Sports 1972; 4: Skalski J, Allison TG and Miller TD. The safety of cardiopulmonary exercise testing in a population with highrisk cardiovascular diseases. Circulation 212; 126: Dhoble A, Sarano ME, Kopecky SL, et al. Safety of symptom-limited cardiopulmonary exercise testing in patients with aortic stenosis. Am J Med 212; 125: Keteyian SJ, Isaac D, Thadani U, et al. Safety of symptom-limited cardiopulmonary exercise testing in patients with chronic heart failure due to severe left ventricular systolic dysfunction. Am Heart J 29; 158: S72 S Soga Y, Yokoi H, Ando K, et al. Safety of early exercise training after elective coronary stenting in patients with stable coronary artery disease. Eur J Cardiovasc Prev Rehabil 21; 17: van der Meer S, Zwerink M, van Brussel M, et al. Effect of outpatient exercise training programmes in patients with chronic heart failure: a systematic review. Eur J Cardiovasc Prev Rehabil. Epub ahead of print 18 July 212. DOI: /