Immersible ergocycle prescription as a function of relative exercise intensity

Abstract

Purpose

The purpose of this study was to establish the relationship between various expressions of relative exercise intensity percentage of maximal oxygen uptake (%VO_2max), percentage of maximal heart rate (%HR_max), %VO₂ reserve (%VO₂R), and %HR reserve (%HRR)) in order to obtain the more appropriate method for exercise intensity prescription when using an immersible ergocycle (IE) and to propose a prediction equation to estimate oxygen consumption (VO₂) based on IE pedaling rate (rpm) for an individualized exercise training prescription.

Methods

Thirty-three healthy participants performed incremental exercise tests on IE and dryland ergocycle (DE) at equal external power output (P_ext). Exercise on IE began at 40 rpm and was increased by 10 rpm until exhaustion. Exercise on DE began with an initial load of 25 W and increased by 25 W/min until exhaustion. VO₂ was measured with a portable gas analyzer (COSMED K4b²) during both incremental tests. On IE and DE, %VO₂R, %HR_max, and %HRR at equal P_ext did not differ (p > 0.05).

Results

The %HRR vs. %VO₂R regression for both IE and DE did not differ from the identity line %VO₂R IE = 0.99 × HRR IE (%) + 0.01 (r² = 0.91, SEE = 11%); %VO₂R DE = 0.94 × HRR DE (%) + 0.01 (r² = 0.94, SEE = 8%). Similar mean values for %HR_max, %VO₂R, and %HRR at equal P_ext were observed on IE and DE. Predicted VO₂ obtained according to rpm on IE is represented by: VO₂ (L/min) = 0.000542 × rpm² − 0.026 × rpm + 0.739 (r = 0.91, SEE = 0.319 L/min).

Conclusion

The %HRR–%VO₂R relationship appears to be the most accurate for exercise training prescription on IE. This study offers new tools to better prescribe, control, and individualize exercise intensity on IE.

1. Introduction

Aerobic exercise training performed at an appropriate level of intensity has beneficial effects on health in the general population and improves aerobic capacity and exercise performance.¹ Prescription of exercise intensity using measured or estimated absolute values that may include either caloric expenditure (kcal/min) or absolute oxygen consumption (VO₂, in L/min) may result in misclassification of exercise intensity (e.g., moderate, vigorous) because they do not consider individual factors such as body mass, sex, and fitness level¹ or the environment in which the exercise is performed (i.e., water and land).2, 3

Individualized exercise training prescription is more appropriate using a relative measure of intensity and the following parameters can be used: maximal oxygen uptake (VO_2max), VO₂ reserve (VO₂R), maximal heart rate (HR_max), heart rate reserve (HRR), maximal metabolic equivalent of task (METs_max) and their relative expressions, %VO_max, %VO₂R, %HR_max, %HRR, and %METs_max.1, 4

Previous studies have shown conflicting results regarding the best approach to express %VO₂ (max or reserve) as a function of HR variables (max or reserve). Several studies have shown a better relationship between %HRR and %VO₂R in healthy adults using a treadmill or ergocycles,5, 6 among athletes⁷ and obese subjects.⁸ However, another study has demonstrated a better relationship between %VO_2max and %HRR.⁹ The American College of Sports Medicine (ACSM) has proposed a classification of relative and absolute exercise intensity for aerobic exercise where %VO₂R and %HRR remain interchangeable, but the ACSM emphasizes that the relationship among actual energy expenditure, HRR, VO₂R, %HR_max, and %VO_2max can vary considerably depending on exercise test protocol, exercise intensity, resting HR, fitness level, age, body composition, exercise mode (i.e., water and land) and other factors.¹

Lately, an increasing number of individuals are performing aerobic exercise training in an aquatic environment using various exercise modalities and devices. Water exercise allows participants to undergo hard workouts at intensities similar to dryland physical activities with a lower impact on joints and with different physiological responses.2, 10 Previous studies have concluded that the most accurate way to estimate exercise intensity in water is to use HR measurements and/or ratings of perceived exertion (RPE).3, 11 Giacomini et al.¹² studied the relationship between revolution per minute (rpm) and VO₂–HR responses on 4 different models of immersible ergocycle (IE). They showed that for a similar pedaling rate (70 rpm) the %VO_2max varied from 45% to 90% and the %HR_max varied from 60% to 90%, which could be explained by the difference between IE pedaling systems used in their study. Thus, various IE models may be responsible for producing different external power outputs (P_ext) for a similar rpm. Currently, the pedaling cadence (rpm) on various IE models is the only main parameter to increase or decrease exercise intensity (P_ext).13, 14, 15, 16

Previous studies have shown that immersion can reduce VO₂ and HR during deep water running, immersed treadmill running or immersible ergocycle pedaling at maximal15, 17, 18 and submaximal intensities (i.e., velocity or P_ext).10, 19 Consequently, the VO₂–HR relationship (in % of max or reserve) could be modified during exercise on IE and be different from that of dryland ergocycle (DE). Therefore, exercise prescription using the VO₂–HR relationship of DE could be less valid and accurate for IE exercise. The effects of immersion on the VO₂–HR relationship during IE has not been previously studied and compared with that of DE in healthy participants. Thus, the objectives of this work were: 1) to study the relationship between various expressions of relative exercise intensity (%VO_2max, % VO₂R, %HR_max, and %HRR) in order to obtain the more appropriate method for exercise intensity prescription when using an IE; and 2) to propose a prediction equation to estimate VO_2max based on IE pedaling rate (rpm) for individualized exercise training prescription.

2. Materials and methods

2.1. Experimental approach to the problem

All participants performed maximal incremental exercise tests in a random order on an IE (Hydrorider Aquabike professional; Hydrorider professional aquatic equipment^®, DIESSE S.R.L, Bologna, Italy) and a DE (Ergoline 800S; Ergoline GmbH, Bitz, Germany) and at similar P_ext in a laboratory with air temperature maintained at 21°C and in a swimming pool at a thermoneutral exercise water temperature of 30°C.20, 21 During incremental exercise tests, cardiopulmonary responses were measured with a portable gas analyzer (COSMED K4b²; COSMED, Rome, Italy). Gas analyzers were calibrated before each test using a standard certified commercial gas preparation (O₂: 16%; CO₂: 5%).²¹ HR was measured continuously using a HR monitor (T 61; Polar, Kempele, Finland).

2.2. Subjects

Thirty-three healthy young participants (age: 33 ± 10 years, 28 men and 5 women) were recruited at the Cardiovascular Prevention and Rehabilitation Centre of the Montreal Heart Institute. This study was approved by the Research Ethics Committee of the Montreal Heart Institute and all the subjects gave their written informed consent to participate in the study. Their baseline characteristics are presented in Table 1. Inclusion criteria were no apparent health problems and age 18 years and above. The exclusion criteria included: 1) any documented cardiovascular, pulmonary, musculo-skeletal, or metabolic diseases; and 2) inability to perform a maximal cardiopulmonary exercise test.

Table 1.

Subjects' physical characteristics and exercise testing parameters on IE and DE.

Parameters	Valuesa
Age (year)	33 ± 10
Sex (male/female)	28/5
Body mass (kg)	72 ± 9
Height (m)	1.74 ± 0.06
BMI (kg/m2)	23.7 ± 2.5
VO2max (L/min)
DE	3.46 ± 0.65
IE	2.48 ± 0.63**
VO2max (mL/min/kg)
DE	46.28 ± 9.18
IE	33.10 ± 9.07**
Resting HR
DE	75 ± 12
IE	73 ± 11
HRmax
DE	177 ± 14
IE	167 ± 12*
Maximal Pext (W)
DE	251 ± 55
IE	253 ± 58

Abbreviations: BMI = body mass index; DE = dry ergocycle; HR = heart rate; HR_max = maximal heart rate; IE = immersible ergocycle; P_ext = external power output; VO_2max = maximal oxygen uptake.

^aValues are presented as mean ± SD except sex as n.

^*p < 0.005, **p < 0.001, compared with DE.

2.3. Procedures

During data collection on both IE and DE, cardiopulmonary parameters were measured during a 3 min rest period, the exercise period, and a 5 min post-exercise recovery period. Data were averaged every 15 s for minute ventilation (VE, in L/min), body temperature, pressure, and saturation (BTPS), oxygen uptake (VO₂, in L/min), standard temperature and pressure dry (STPD), and carbon dioxide production (VCO₂, in L/min). Maximal exercise tests on IE and DE lasted until the attainment of 1 of the 2 primary maximal criteria: (1) a plateau of VO₂ (<150 mL) despite an increase in P_ext (rpm or W on IE and DE, respectively), and (2) respiratory exchange ratio >1.1; or 1 of the 3 secondary maximal criteria: (1) measured HR_max attaining 95% of age-predicted HR_max; (2) inability to maintain the required workload; and (3) subject exhaustion with cessation caused by fatigue or subjects and/or other clinical symptoms (dyspnea) and/or ECG abnormalities that required exercise cessation.15, 21

Following the 3 min rest period, the initial exercise load for incremental test on DE was 25 W and was increased by 25 W/min until exhaustion. The pedaling rate (rpm) was at a minimum cadence of 60 rpm; however, the participants were instructed to maintain a pedaling cadence of 80 rpm since previous studies have shown that in conditions simulating those seen during prolonged competitive cycling, higher cadences (i.e., 100 rpm vs. 80 rpm) are less efficient, resulting in greater energy expenditure and reduced peak power output (327 ± 27 W vs. 362 ± 38 W, respectively) during maximal performance.²²

The P_ext increases in water as a function of pedaling rate/velocity.16, 23, 24 Thus, in the large number of commercially available models of IE, the only method to either increase or decrease the intensity of exercise is by varying the rpm. On the IE, the subjects were immersed up to the xiphoid process level and the exercise protocol began at a pedaling rate of 40 rpm and was increased each minute by 10 rpm until 70 rpm. Afterwards, the rpm was increased by 5 rpm until the subject was unable to follow the pace or until exhaustion.15, 16 Pedaling rate (rpm) was controlled with the use of both a metronome (Matrix MR500; Metronome, Seoul, Korea) and a pedaling rpm meter (Echowell F2; CATEYE Co., Ltd., Taichung, Taiwan, China) to help the participant to maintain correct rpm. Following the exercise test, the participants recovered for 5 min while seated on the IE or DE. The posture of each subject on both cycle ergometers was adjusted for the correct height of the saddle by sitting the participant on the bicycle, according to previous studies.13, 16

The highest VO₂ and HR values reached during the exercise phase of each test were considered as the VO_2max and HR_max. The following values of HRR and VO₂R were calculated by subtracting, respectively, the value at rest from the maximal values.⁷ Each test on IE and DE was separated from each other by 1 week. For each subject, HR and VO₂ values were recorded at rest, were averaged during the last 15 s of each 1 min stage, and were expressed as percentages of their respective reserve (%VO₂R and %HRR) or maximum values (%VO_2max, %HR_max, data not shown).

$$\%{\text{HR}}_{\max }=\left(\text{HR}\text{of}\text{each}\text{stage}/{\text{HR}}_{\max }\right)\times 100\%$$ Eq. (1)

$$\begin{array}{c}\%\text{HRR}=\left(\text{HR}\text{of}\text{each}\text{stage}-{\text{HR}}_{\text{rest}}\right)/\\ \left({\text{HR}}_{\max }-{\text{HR}}_{\text{rest}}\right)\times 100\%\end{array}$$ Eq. (2)

$$\begin{array}{c}\%{\text{VO}}_{2\max }=\left({\text{VO}}_2\text{of}\text{each}\text{stage}/{\text{VO}}_{2\max }\right)\\ \times 100\%\end{array}$$ Eq. (3)

$$\begin{array}{c}\%{\text{VO}}_2\mathrm{R}=\left({\text{VO}}_2\text{of}\text{each}\text{stage}-{\text{VO}}_{2\text{rest}}\right)/\\ \left({\text{VO}}_{2\max }-{\text{VO}}_{2\text{rest}}\right)\times 100\%\end{array}$$ Eq. (4)

On the IE, the P_ext was produced by the pedaling rate that has been detailed elsewhere.13, 15, 17, 25 Briefly, the external forces during exercise on an IE are mainly caused by the mechanical components of the pedaling system (paddles, pedals, and rods) and by leg movement drag (calf, foot, and thigh) that is dependent on the surface area of the lower limbs and the pedaling rate (rpm).

The P_ext expressed in watt (W) was calculated by multiplying the total net force (F) overcoming the resistance of the system movement (pedaling system and legs) by the tangential velocity (m/s) of the pedal. Thus, the following general fluid equation was used to determine F mathematically:

$$F=\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.\rho A{v}^2{C}_d$$ Eq. (5)

where ρ is the density of water (at 30°C = 995.7 kg/m³), A is the projected frontal area (m²) in the direction of the movement for all segments involved (lower limbs, paddles, rods, and pedals), v is the velocity (m/s) ranging from 40 to 120 rpm, and C_d is the drag coefficient of shape for every element of the pedaling system and of the lower limbs.15, 16

2.4. Statistical analysis

Results are presented as mean ± SD. An ANOVA with repeated measures (condition × intensity) was performed to compare: (1) %VO₂R and %HRR during exercise on IE and DE for the same P_ext and (2) the VO₂ and HR responses during maximal incremental exercise test on DE or IE. Relationships between variables (%HRR and %VO₂R) obtained on IE and DE were performed using linear regression analysis. The level of equivalency was evaluated with analysis of the mean slopes and intercepts (i.e., slope = 1; intercept = 0) that was determined from linear regression equations. Statistical analysis was performed with Sigma Plot (Version 11.0; Sigma, San Jose, CA, USA), StatView (Version 5.0; SAS Institute Inc., Cary, NC, USA) and SPSS (Version 15.0; SPSS, Chicago, IL, USA). The Bland and Altman analysis was performed with Excel (Microsoft, Redmond, WA, USA).

3. Results

3.1. Absolute and %VO₂R

Fig. 1 illustrates the absolute and relative values of VO₂ obtained on IE in relationship to DE. The data points represent VO₂ measured during the incremental test at each stage (same P_ext) on IE and DE for each individual. As seen in Fig. 1A, the absolute VO₂ (L/min) obtained on IE was systematically lower and significantly correlated (r² = 0.81, p < 0.0001) to the VO₂ (L/min) on DE. The regression equation to predict VO₂ (L/min) on an IE from VO₂ (L/min) obtained on DE is: VO₂ IE (L/min) = 0.69 × VO₂ DE (L/min) + 130.09. Fig. 1B shows a significant correlation (r² = 0.89, p < 0.0001) of relative VO₂R (%) on IE as a function of relative VO₂R (%) on DE. The regression equation obtained is VO₂R IE (%) = 1.01 × VO₂R DE (%) + 0.02 and indicates that the slope is equal to 1 and that the intercept goes through 0, demonstrating that both forms of expression are equal.

Fig. 1.

Relationship of VO₂ measured on IE and DE. (A) VO₂ on IE relative to VO₂ on DE in absolute values of VO₂ (L/min); the filled black line represents the line of identity. (B) VO₂ on IE relative to VO₂ on DE in relative values of VO₂R (VO_2max − VO_2rest). All data points represent all participants. The dashed line in both graphs represents the line of the regression equation. DE = dryland ergocycle; IE = immersible ergocycle; VO₂ = oxygen uptake; VO_2max = maximal oxygen uptake; VO_2rest = oxygen uptake recorded at rest.

3.2. %HRR and %VO₂R

Table 2 presents %HRR and %VO₂R on IE and DE for the same P_ext. As well, Table 2 proposes a classification of RPE exercise intensity for both IE and DE. The average values of %HRR and %VO₂R were not significantly different for the same P_ext (p = 0.81 and p = 0.29, respectively) during exercise on IE and DE.

Table 2.

Relative intensity (%HRR and %VO₂R) for a same P_ext (W) on IE and DE and classification of exercise intensity on IE (mean ± SD).

rpm	Pext (W)	%HRR		%VO2R		Intensity
		IE	DE	IE	DE
50	50	21.2 ± 1.4	21.8 ± 1.9	21.3 ± 1.7	23.1 ± 1.1	Very light
60	75	35.3 ± 1.7	38.2 ± 1.7	38.5 ± 2.5	32.0 ± 1.3	Light
70	125	56.7 ± 2.1	57.6 ± 1.8	59.4 ± 3.3	50.9 ± 1.6	Moderate
80	200	85.3 ± 2.1	81.7 ± 1.9	85.2 ± 2.3	80.2 ± 2.5	Vigorous
90	300	98.4 ± 3.4	97.5 ± 3.9	96.7 ± 6.7	97.4 ± 4.7	Near-maximal

Note: very light < 30; light: 30–39; moderate: 40–59; vigorous: 60–89; near-maximal ≥ 90. Classification of exercise intensity adapted from ACSM.¹

Abbreviations: %HRR = percentage of heart rate reserve; %VO₂R = percentage of oxygen uptake reserve; DE = dryland ergocycle; IE = immersible ergocycle; P_ext = external power output; rpm = revolutions per minute.

Fig. 2 shows the relationships between %HRR and %VO₂R obtained for both IE and DE. As shown in Fig. 2A and 2B, %VO₂R was significantly correlated to %HRR for both IE and DE (r² = 0.91, p < 0.0001 and r² = 0.94, p < 0.0001, respectively), and the regression equations indicated that the 2 expressions of exercise intensity (%VO₂R and %HRR) were equal %VO₂R IE = 0.99 × HRR IE (%) + 0.01, SEE = 11%; %VO₂R DE = 0.94 × HRR DE (%) + 0.01, SEE = 8%, respectively. Fig. 2C shows the significant relationship (r² = 0.94, p < 0.0001) between %HRR IE and %HRR DE. The regression between both variables is %HRR IE = 0.97 × HRR DE (%) + 0.02. The equation slope and intercept are near equal to 1, respectively.

Fig. 2.

Relationship of %VO₂R with %HRR obtained with the IE and DE. (A) %VO₂R vs. %HRR obtained with IE; (B) %VO₂R vs. %HRR obtained with DE; (C) %HRR obtained on IE vs. %HRR obtained on DE; (D) level of agreement between %HRR obtained on IE and %HRR obtained on DE. All data points represent all participants. Dashed lines represent the regression equation in all graphs. %HRR = percentage of heart rate reserve; %VO₂R = percentage of oxygen uptake reserve; DE = dryland ergocycle; HRR = heart rate reserve; IE = immersible ergocycle; VO₂R = oxygen uptake reserve.

3.3. %HRR IE and %HRR DE level of agreement

Fig. 2D is a Bland and Altman plot illustrating the level of agreement (mean = −0.02) between the %HRR IE and %HRR DE difference. The regression line (medium hash) has a slope near equal to 0 (−0.08), indicating that the error in measure is nil and is constant throughout the range of 0–100%.

3.4. Estimated VO₂ prediction

Predicted VO₂ (L/min) obtained according to rpm on IE (data not shown) is represented by the following equation (r = 0.91, SEE = 0.319 L/min):

$$\begin{array}{c}{\text{VO}}_2\left(\mathrm{L}/\min \right)=0.000542\times {\text{rpm}}^2-0.026\times \text{rpm}+0.739\\ \left(r=0.91,\text{SEE}=0.319\mathrm{L}/\min \right)\end{array}$$ Eq. (6)

4. Discussion

The original findings of this study were that: 1) relative intensity was found to be similar for %VO₂R, %HR_max (data not shown) and %HRR at a similar P_ext on IE and DE; 2) on IE and DE, the %HRR vs. %VO₂R relationship was the closest to the identity line and the most accurate for exercise prescription in immersion. Linear regressions obtained on IE and DE to predict %VO₂R from %HRR, as shown in Fig. 2A and 2B, can be considered the most accurate for exercise training prescription for either exercise modality (IE and DE). To the best of our knowledge, this is the first study to compare the HR–VO₂ relationship (in % of reserve values) during incremental exercise on IE vs. DE at the same P_ext in healthy subjects.

We have used the method reported in previous studies using the same IE model to calculate the P_ext.13, 15, 16 This method provides a mathematical model for generalizability of calculation for IE P_ext with any IE type. The model takes into account rpm, IE pedaling system physical characteristics and lower limb size. Thus, from a performed incremental exercise test on IE, it is possible to obtain the relationship between rpm and P_ext to better prescribe relative to maximal exercise intensity on any IE. Currently, the differences between commercially available IE are in the pedaling system physical characteristics (the paddle and rod length varying between brands). The method proposed herein makes it possible to calculate P_ext.

In the current study, the predicted values to %HRR and %VO₂R at all levels of relative intensity agreed with the most recent exercise intensity scale of the ACSM.¹ In addition, the relationship between %VO₂R and %HRR (Fig. 2) is in agreement with the ACSM recommendations for healthy young participants despite the controversy raised by other investigators that have reported higher values at 85%VO_2max or VO₂R (i.e., 92%–93%HR_max).7, 8, 26

Other authors, however, who criticize the “traditional” concept to prescribe exercise intensity by means of a target % of HR_max, HRR, VO_2max, or VO₂R, have suggested that it might be more appropriate to consider, in addition, the metabolic demand of exercise by means of determining a lactate-threshold and to tailor exercise within target training zones of intensity.27, 28 Nonetheless, our study appears to offer a method for interchanging exercise prescription intensity for 2 different exercise devices (IE and DE) that is more accurate than the traditional %HR–%VO_2max relationship. Thus, if the following parameters, such as the absolute VO₂, HR and hemodynamic response (stroke volume, cardiac preload, cardiac output, and venous return) are affected during upright immersion exercise,14, 15, 19, 29, 30, 31 then, the rationale for using %VO₂R and %HRR for IE exercise prescription appears more appropriate. Therefore, as the theory of specificity suggests,32, 33 it is important to establish the value of VO_2max and HR_max directly in water to properly prescribe the intensity on IE.

We have previously reported that the relationship between P_ext (W) and rpm during incremental exercise on the IE is non-linear and could explain why VO₂ expressed as %VO_2max for intensities >60 rpm increases exponentially as a function of rpm.13, 16 This non-linear relationship, reported by us and others, reiterates the importance of using %HRR, as proposed herein, since as shown in Fig. 2A, the relationship between %VO₂R and %HRR is linear. This could have practical implications since small increases in rpm generate a more rapid increase of physiological responses. We have included a very very light category (Table 2) that corresponds to the lowest intensity on IE (≤40 rpm) and relates to the intensity recommended for warm-up.

There are some limitations in our study. This work is based on a sample of young healthy subjects; thus, our results apply only to a similar population and cannot be generalized to other groups, such as older subjects, subjects with cardiovascular risk factors or established cardiac disease.

Future studies in those populations would be necessary to see if similar results would be obtained.

Practically, however, the current study offers a new tool to better prescribe, control, and individualize exercise intensity on IE from the %HRR–%VO₂R relationship. It is possible to estimate these variables using the suggested method from IE pedaling cadencies (rpm)13, 16 for various water immersed bicycle models with a similar pedaling systems (i.e., Hydrorider^®, Archimedes^®, Poolbike^®) or by directly measuring cardiopulmonary and hemodynamic responses. However, for accurate prescription in different populations as quoted above, practitioners using any IE type will have to consider the following 4 elements when calculating the power output: (1) the pedaling rate; (2) the seat height adjustment; (3) the precise characteristics of the pedaling system (length and width of paddles, pedals, and rods); and (4) participant leg anthropometric characteristics.¹⁶

5. Conclusion

This study offers a new tool to better prescribe, control and individualize exercise intensity on IE. The %HRR–%VO₂R relationship appears to be the most accurate for exercise training prescription on IE. VO₂ (L/min) on IE can be obtained and predicted from the VO₂ measured on a DE. Similarly, VO₂ (L/min) obtained on IE can be predicted from IE pedaling cadencies (rpm) and is represented by: VO₂ (L/min) = 0.000542 × rpm² − 0.026 × rpm + 0.739 (r = 0.91, SEE = 0.319 L/min). Absolute cardiopulmonary responses (VO₂ and HR) during exercise on IE are different from that of DE, but relative intensity was found similar at a similar P_ext on both IE and DE. The classification of exercise intensity from rpm on IE for relative intensity (%HRR and %VO₂R) is in agreement with the 2011 ACSM exercise intensity scale.¹

Authors' contributions

MGarzon conceived of the study, carried out the experiments, participated in the analysis and drafted the manuscript; MGayda participated in the coordination and drafted the manuscript; AN helped to draft the manuscript; ASC conceived of the study, participated in the analysis, and drafted and revised the manuscript; MJ helped to draft the manuscript. All authors have read and approved the final version of the manuscript, and agree with the order of presentation of the authors.

Competing interests

That authors declare that they have no competing interests.