Portable metabolic carts are a popular tool to assess aerobic capacity and affirm many cardiorespiratory conditions. They may also measure strength training performance. Given their popularity and increased usage to assess strength training performance; their data accuracy and consistency are important to determine.
OBJECTIVE:
Measure Cosmed K4 b2 portable metabolic cart data repeatability from consecutive seated calf press workouts.
METHODS:
Fifteen women and twelve men did two workouts that began with a stationary cycling warm-up followed by calf presses. Gases were measured before the calf press portion of workouts to establish baseline VO and VCO values, as well as continually throughout and after the calf press protocol. Subjects were detached from the cart once gas values returned to baseline after workouts concluded. In addition to VO and VCO, repeatability was quantified for: breaths per minute, tidal volume, ventilation, O uptake relative to body mass, expired O and CO fractions, percent fat and carbohydrate utilization, METS and total energy cost. Mean and peak values per variable were analyzed. Repeatability was assessed separately for male and female data, as well as with values pooled, by the following: intraclass correlation coefficients, eta squared, limits of agreement, coefficient of variation and smallest real difference percent.
RESULTS:
Per variable, repeatability values across workouts were low. Female intraclass correlation coefficient mean values were more repeatable for variables related to gas measurements, yet male data were generally more repeatable for those related to substrate usage.
CONCLUSIONS:
Results for some repeatability indices were influenced by measurement magnitude. Peak values were predictably less repeatable than those for mean values. Most smallest real differences percent scores are so high they were rendered irrelevant or meaningless to determine true differences among paired values. Results suggest low data repeatability that are likely appropriate and realistic for the exercise protocol, hardware and intensity examined.
Portable metabolic carts are a popular tool to assess aerobic capacity and help affirm many cardiorespiratory conditions. Advancements in miniaturization within the field of computer science enabled creation of this novel technology. Miniaturization offers this technology to settings where standard carts, due to size and operational constraints, are impractical. Such scenarios include their use aboard manned space flights [1]. While portable metabolic carts, like their more conventional predecessors, quantify numerous variables, they perhaps best measure VO and VCO concentrations within the body under steady-state conditions [2, 3, 4, 5, 6, 7]. One such device is the Cosmed K4 b2 portable metabolic cart (Cosmed Corporation; Rome, Italy). It is a wearable unit that uses telemetry and breath-by-breath gas sampling to assess cardiorespiratory function for field-based performances. It is also equipped with a mixing chamber and gas sensors that allow its use in both clinical and laboratory settings.
Research evaluated the Cosmed K4 b2 portable metabolic cart’s use during aerobic exercise and reported positive outcomes that helped validate the device [2, 3, 4, 5, 6, 7]. Cosmed K4 b2 portable metabolic carts were even used by disuse populations, sometimes as they performed aerobic exercise [8, 9]. However, unlike aerobic activity, strength training is a non-steady state exercise modality that sees rapid perturbations in VO and VCOconcentrations within the body that are difficult to accurately quantify. Yet metabolic carts are increasingly used to measure strength training performance; they include quantify energy expenditures from resistive exercise workouts, whereby estimated caloric costs were computed from net O uptakes [10, 11]. While strength training relies primarily on anaerobic glycolysis for ATP resynthesis, it may also see considerable contributions from oxidative phosphorylation and other aerobic pathways [12, 13]. Gender-based differences in substrate utilization may exist, as some suggest women rely more on oxidative metabolism for strength training workouts [12, 13]. Thus, continued inquiry on the nature of substrate data obtained from strength training workouts by metabolic carts is warranted.
Aside from potential disparities in metabolic demands and substrate usage for each type of exercise, there are distinct differences in the manner aerobic and strength training workouts are performed. Aerobic exercise entails continuous activity done at a pace that delivers O to cells commensurate to its demand, yet strength training entails brief bursts of intense physical effort separated by rest periods. Under non-steady state conditions, the ability of Cosmed K4 b2 portable metabolic carts to offer accurate and consistent data for the variables they provide is unknown. Given the use of metabolic cart variables to assess strength training performance, the accuracy and consistency of Cosmed K4 b2 variables are important to assess. Variables include direct gas measurements, and those which estimate substrate usage and energy cost. Cosmed K4 b2 portable metabolic cart data accuracy and consistency, whereby the same measurements are systematically collected over time and compared, relate to the repeatability of values they obtain.
Repeatability quantifies measurement variations from a single instrument or person under the same conditions [14]. Data repeatability results differ considerably among scientific disciplines and are perhaps lowest for human responses to non-steady state exercise. Lower data repeatability is likely due to an inability to consistently exert comparable amounts of effort over successive workouts. While non-steady state exercise may evoke strength, power and muscle mass changes, less data exists on its cardiorespiratory responses. Given the paucity of cardiorespiratory data from 1): non-steady state exercise and 2): portable metabolic carts, inquiry is warranted. Our purpose quantifies data repeatability for cardiorespiratory dependent variables from a Cosmed K4 b2 portable metabolic cart across two identical strength training workouts. Given advancements in metabolic cart miniaturization, and in addition to assessments of gender-based substrate utilization differences and work-energy cost relationships, we hypothesize this hardware will yield repeatable data from consecutive strength training workouts.
Methods
Subjects and their first laboratory visits
This study was approved by a university-based institutional review board (IRB # 14–56) for human data collection. College-age participants (12 men, 15 women) gave informed written consent, and then filled out a medical questionnaire that stated they were free of conditions that could impair their workout performance. They were healthy non-smokers who each made three laboratory visits close to the same times of day to limit the impact of circadian influences on our results. First visits entailed collection of anthropometric data in addition to familiarization to our study procedures. For their final two visits, participants did resistive exercise workouts on a flywheel ergometer; per subject, the time between workouts averaged 17.8 2.6 days. Data were collected at within a thermoneutral (21–22C, 40–45% humidity) laboratory at sea level.
At their first visits, subject’s height, mass, and body fat percentage values were recorded as they stood upright while barefoot. Heights were measured with a stadiometer (Detecto Corporation; Webb City, Missouri). Mass and body fat percentage were recorded with a calibrated bioimpedance scale (Model BF-350, Tanita Corporation; Tokyo, Japan). First visits continued as they did a stationary cycling warm-up (Model 812E, Ergotest; Stockholm, Sweden) against a workload of 75–90 watts for five minutes. To conclude their first visits, they performed seated calf presses (ankle plantar flexion with full knee extension) on the ergometer (YoYo Technologies; Stockholm, Sweden) at a submaximal level of effort to become accustomed to its operation. A pair of flywheels imparted resistance that was overcome by torque exerted against the ergometer’s footplate. Per participant, first visits lasted approximately 20–25 minutes.
Second and third visits
Second and third visits entailed identical workouts, which allowed cardiorespiratory data to be compared and measured for repeatability. A Cosmed K4 b2 portable metabolic cart was calibrated, and used for data collection, in accordance with the manufacturer’s guidelines. Subjects were instructed to arrive at our laboratory well rested at the start of workouts, which began with a stationary cycling warm-up identical to that of their first visit. When the warm-up concluded, they next sat on the ergometer for ten continuous minutes. With the portable cart beside the ergometer, the former device sampled respiratory gases as participants sat quietly. They breathed normally into a respiratory tube that connected them to the metabolic cart’s gas sensors as they wore a nose clip, which permitted collection of baseline (pre-exercise) VO and VCO values per gas, averaged over 15-second intervals, and displayed in real time.
After ten minutes and still tethered to the cart and wearing the nose clip, subjects did a 4-set 15-repetition calf press protocol on the ergometer, which is a prototype of a device now aboard the International Space Station [15]. With subject’s knees fully extended and ankles dorsiflexed 10–15, they aligned the balls of their feet against the lower edge of the ergometer’s footplate to begin sets. They were encouraged to exert maximal effort and rested 120 seconds between sets. Resistance was commensurate to flywheel radii as rotation rates were recorded at 10 Hz with software (Labview 8.1; Austin, TX) [16]. Concentric, eccentric, and total work were calculated from the following: whereby rotation inertia and angular velocity. Our ergometer’s instrumentation methods and data validity were deemed acceptable [16, 17, 18].
Subject seated on the flywheel ergometer as they performed calf presses.
Figure 1 shows subject’s position on the ergometer. When the protocol concluded, they remained tethered to the cart and wore the nose clip until VO and VCO values returned to baseline. In addition to VO and VCO, the following variables were recorded to assess repeatability: breaths per minute (FREQ), tidal volume (TV), ventilation (VE), O uptake relative to body mass, expired O and CO fractions (FeO, FeCO), percent fat and carbohydrate utilization (FAT%, CHO%), METS and total energy expenditure (EE). Both mean and peak values per dependent variable were used for analysis. Exceptions were CHO% and EE; the former was only analyzed for mean, and the latter for peak, values. Repeatability was assessed separately for men and women, as well as with data pooled. Second and third visits each lasted 45 minutes.
Calf press work results (mean sem) in Joules
Subjects
Concentric work
Eccentric work
Total work
Men
9514
505
8593
439
18107
938
Women
7502
440
6807
403
14309
841
Total
8359
356
7568
319
15927
672
Descriptive data (mean sem) for Cosmed K4 b2 mean values
Variable
Men
Women
Total
FREQ (breaths min)
21.7
0.3
22.7
0.6
22.3
0.4
TV (liters)
1.2
0.03
0.89
0.03
1.03
0.03
VE (liters min)
25.7
0.7
19.9
0.8
22.5
0.7
VO (ml min)
710.3
23.5
484.4
16.7
584.8
20.7
VCO (ml min)
805
28.1
548.5
22.3
662.5
24.7
VO/KG (liters kg)
8.9
0.22
7.5
0.20
8.1
0.18
FeO (%)
17.5
0.06
17.8
0.07
17.7
0.05
FeCO (%)
3.8
0.05
3.4
0.06
3.6
0.05
FAT% (%)
9.2
1.5
8.1
1.6
8.7
1.06
CHO% (%)
90.8
1.4
91.9
1.5
91.3
1.05
METS (ml O kg min)
2.52
0.06
2.16
0.06
2.33
0.05
Descriptive data (mean sem) for Cosmed K4 b2 peak values
Variable
Men
Women
Total
FREQ (breaths min)
36.9
0.9
38.0
1.9
37.5
1.1
TV (liters)
2.23
0.07
1.66
0.08
1.91
0.07
VE (liters min)
46.8
1.9
35.6
1.75
40.6
1.5
VO (ml min)
1621.8
68.7
1004.9
43.8
1279.1
57.2
VCO (ml min)
1571.6
54.6
1010.8
42.5
1260.0
50.9
VO/KG (liters kg)
20.4
0.81
15.4
0.50
17.6
0.56
FeO (%)
18.5
0.08
18.8
0.07
18.6
0.05
FeCO (%)
5.3
0.08
4.8
0.49
5.0
0.27
FAT% (%)
62.0
4.55
58.3
4.47
59.9
3.18
METS (ml O kg min)
5.83
0.23
4.40
0.14
5.04
0.16
EE (kilocalories)
73.9
2.3
47.5
2.0
59.3
2.3
Statistics
Z-scores were used to identify outliers, with values more than 1.96 removed from further analyses. Data were then examined for compliance to ANOVA assumptions (normality, independence, equal variances). To assess if women relied more on oxidative phosphorylation than men, FAT% data from both genders were compared with t-tests for independent samples. The merits of concentric, eccentric, and total work to predict the METS and EE variance were measured by Pearson correlations. Mean and peak cardiorespiratory dependent variable values obtained across workouts were compared with the following repeatability statistics: intraclass correlation coefficient (ICC), eta squared (), limits of agreement (LOA), coefficient of variation (CV%) and smallest real difference percent (SRD%). ICC analyses were done with a two-way mixed single measures model. ICC and were derived from ANOVA correlations for unordered pairs; F and p values per computation were also recorded. LOA spanned 1.96 standard deviations from the mean and equaled the range for 95% of the population. CV% was computed as a mean/standard deviation ratio, and then multiplied by 100 to express values as a percentage. A measure of within-subjects variance, SRD% is the least perturbation needed to infer an actual dissimilarity among paired data and was tabulated as the standard error of the estimate/1.414 ratio, and then multiplied by 2.77 [19]. That value was then divided by the mean per dependent variable, and then multiplied by 100 to derive each SRD% score.
Results
Each subject completed three visits. Height values for men and women were 179 2 and 168 2 cm respectively. Male and female masses were 79.9 4.1 and 65.1 2.6 respectively. Body fat percentages were 13.6 1.9 and 24.7 1.8 for men and women respectively. Table 1 shows work values obtained from exercise done on the ergometer. Those values were averaged across the second and third laboratory visits and parsed by contractile mode and gender. Raw respiratory variable descriptive data, collected from those same visits by the Cosmed K4 b2 metabolic cart, are provided as workout mean (Table 2) and peak (Table 3) values respectively.
Our data met all ANOVA assumptions. Z-score analysis identified less than 0.1% of our data as outliers. The data eliminated relate to challenges in collecting cardiorespiratory data during non-steady state activity. Given that acknowledgement, our inter-gender FAT% results yielded non-significant differences due to a high degree of data variability. Yet our Pearson results yielded far different outcomes, as separate analyses revealed mean METS values correlated significantly with concentric ( 0.60), eccentric ( 0.57) and total ( 0.59) work performed from calf press workouts. In similar fashion, EE correlated significantly with concentric ( 0.50), eccentric ( 0.51) and total ( 0.51) work. Our Pearson correlations support the positive relationships between resistive exercise volume and metabolic cost seen previously [13, 20, 21].
Mean data repeatability results
Variable
ICC
ANOVA
ANOVA
LOA (95% CI)
CV%
SRD%
FREQ
Men
0.32
0.64
2.0
0.13
21.8 (18.8–24.7)
7
13.7
Women
0.91
0.95
22.3
0.001
22.7 (16.2–29.2)
14.6
28.7
Total
0.83
0.91
10.8
0.001
22.3 (17.0–27.6)
12.1
23.7
TV
Men
0.65
0.81
4.7
0.001
1.2 (0.9–1.5)
13.7
26.8
Women
0.75
0.87
6.9
0.001
0.9 (0.5–1.3)
21
41.2
Total
0.83
0.91
10.8
0.001
1.1 (0.8–1.4)
22.8
44.7
VE
Men
0.82
0.90
10.0
0.001
25.7 (18.8–32.6)
13.6
13.6
Women
0.78
0.88
8.1
0.001
19.9 (11.4–28.4)
21.7
42.8
Total
0.86
0.93
13.3
0.001
22.5 (12.9–32.1)
21.7
42.7
VO
Men
0.65
0.81
4.8
0.005
7.1 (4.9–9.4)
16.2
31.7
Women
0.58
0.78
3.8
0.007
4.8 (3.0–6.6)
18.9
37.1
Total
0.83
0.91
10.5
0.001
5.9 (2.9–8.8)
26
51.0
VCO
Men
0.71
0.84
5.8
0.003
8.1 (5.4–10.8)
17.1
33.5
Women
0.75
0.87
7.1
0.001
5.5 (3.1–7.9)
22.3
43.6
Total
0.86
0.93
13.4
0.001
6.7 (3.1–10.2)
27.4
53.7
VO/KG
Men
0.35
0.65
2.1
0.115
8.9 (6.8–11)
12.1
23.7
Women
0.31
0.64
1.9
0.117
7.5 (5.3–9.6)
14.7
28.8
Total
0.53
0.76
3.2
0.002
8.2 (5.6–10.7)
16.1
31.5
FeO
Men
0.23
0.60
1.6
0.21
17.5 (16.9 -18.1)
1.7
3.4
Women
0.66
0.82
4.9
0.003
17.8 (17.1–18.5)
2.1
4.0
Total
0.61
0.80
4.1
0.003
17.7 (16.9–18.5)
2.2
4.3
FeCO
Men
0.17
0.56
1.4
0.28
3.8 (3.3–4.3)
7
13.8
Women
0.86
0.92
13.1
0.001
3.4 (2.8–4.0)
9.6
18.9
Total
0.71
0.85
6.0
0.001
3.6 (2.9–4.3)
10
19.5
FAT%
Men
0.61
0.79
4.2
0.008
15 (5–24.3)
78.8
154.3
Women
0.24
0.60
1.6
0.19
8 (8–24.6)
100.2
182.9
Total
0.38
0.68
2.2
0.02
9 (6.6–24.4)
88.8
167.0
CHO%
Men
0.61
0.79
4.2
0.008
91 (76.1–105.1)
8
15.6
Women
0.23
0.60
1.6
0.19
91.9 (75.8–108)
8.9
17.4
Total
0.38
0.68
2.2
0.02
91.3 (76–106.6)
5.4
16.5
METS
Men
0.45
0.71
2.6
0.05
2.5 (1.9–3.1)
0.6
25.2
Women
0.32
0.64
2.0
0.11
2.2 (1.5–2.8)
0.7
31.0
Total
0.51
0.74
3.1
0.002
2.4 (1.6–3.0)
0.7
31.6
Table 4 shows repeatability results for the current mean values. With exercise research that implies ICC values greater than 0.75–0.80 exhibit excellent repeatability [22, 23, 24], pooled Table 4 results show comparable values only for TV, VE, VO and VCO. Inspection of Table 4 ICC values by gender shows female data were generally more repeatable for variables directly related to gas measurements, yet male data were more repeatable for substrate utilization. Other results (LOA, CV%, SRD%) have low acceptability and may in part to be impacted by measurement magnitude for individual dependent variables. This is especially true FAT% and CHO%, which had a potential range of scores from 0 to 100. With SRD% values 20% considered acceptable, few Table 4 values reached that threshold. Table 5 shows repeatability results for the current peak values. Predictably, they were less repeatable than Table 4 mean values. Table 5 ICC results show little proof of excellent data repeatability; the lone exceptions were FeCO values from our women and pooled samples. In contrast, FAT% results were the least repeatable. Table 5 SRD% values exceed those for Table 4; with FeO a notable exception.
Peak data repeatability results
Variable
ICC
ANOVA
ANOVA
LOA (95% CI)
CV%
SRD%
FREQ
Men
0.55
0.76
3.5
0.02
36.9 (28.1–45.7)
12.3
24.2
Women
0.63
0.80
4.4
0.004
38 (18–59)
26.9
52.8
Total
0.61
0.80
4.1
0.001
37.5 (21.5–53.5)
21.8
42.6
TV
Men
0.55
0.76
3.5
0.02
2.2 (1.5–2.9)
15.6
30.5
Women
0.49
0.73
2.9
0.02
1.7 (0.8–2.5)
25.5
50.0
Total
0.67
0.83
5.1
0.001
2.0 (1.0–2.9)
25.2
49.3
VE
Men
0.51
0.74
3.1
0.03
47 (29–64.6)
19.5
26.7
Women
0.66
0.82
4.9
0.002
36.7 (16.9–54.5)
26.9
42.8
Total
0.70
0.84
5.7
0.001
40.6 (19.3–61.9)
26.8
42.7
VO
Men
0.68
0.83
5.3
0.003
16.2 (9.6–22.8)
20.8
40.7
Women
0.56
0.77
3.6
0.009
10 (5.4–14.8)
23.9
46.7
Total
0.83
0.91
10.7
0.001
12.8 (4.6–21.0)
32.8
64.3
VCO
Men
0.72
0.85
6.2
0.002
15.7 (10.5–21)
17
33.3
Women
0.71
0.85
5.9
0.001
10 (5.4–14.8)
23.9
45.1
Total
0.87
0.93
14.7
0.001
12.3 (5.3–19.3)
29.7
58.1
VO/KG
Men
0.54
0.55
3.3
0.03
20.4 (12.7 – 28)
19.4
38.0
Women
0.19
0.58
1.5
0.23
15.4 (10–20.7)
17.7
34.6
Total
0.62
0.80
4.2
0.001
17.6 (9.5–25.7)
23.4
45.9
FeO
Men
0.56
0.76
3.5
0.02
18.5 (17.8–19.2)
2
3.9
Women
0.54
0.76
3.4
0.013
18.7 (18 – 19.5)
2.1
4.0
Total
0.61
0.80
4.1
0.001
18.6 (17.9 – 19.4)
2.2
4.2
FeCO
Men
0.31
0.63
0.9
0.14
5.3 (4.5–6.1)
7.4
14.5
Women
0.90
0.95
18.9
0.001
4.8 (0.5–10.0)
55.9
21.4
Total
0.86
0.93
12.9
0.001
5.0 (1.1–8.9)
40.2
26.8
FAT%
Men
0.24
0.60
1.6
0.21
62 (18.3–105.7)
36
64.9
Women
0.42
0.69
2.4
0.06
58 (10.3–105.7)
42
73.6
Total
0.32
0.65
1.9
0.005
59.9 (14–105.8)
39
71.5
METS
Men
0.54
0.75
3.3
0.03
5.8 (3.6–8.0)
19.4
38.0
Women
0.19
0.58
1.5
0.23
4.4 (2.9–5.9)
17.7
34.6
Total
0.62
0.80
4.2
0.001
5.1 (2.7–7.4)
23.5
46.0
EE
Men
0.40
0.68
2.4
0.08
47.5 (26.3–68.7)
22.7
29.5
Women
0.61
0.79
4.1
0.005
74.4 (52.1–96.7)
15
44.4
Total
0.80
0.90
4.2
0.001
59.3 (25.8–92.8)
28.8
56.5
Discussion
Repeatability quantifies measurement variations from a single instrument or person under the same conditions [14]. Data from exercise done under non-steady state conditions are less likely to be repeatable, which stem from an inability to consistently exert comparable efforts over successive bouts of high-intensity activity [25, 26]. An examination of data from elite rugby players produced ICC values with moderate to high reliability for tackle indices that are comparable to some of the Table 4 results [26]. A recent study examined intra-rater reliability and repeatability of heart rate and percent O saturation values from a pulse oximeter [25]. On a flywheel ergometer, subjects (15 men, 17 women) underwent a similar exercise protocol as the current study’s [25]. Results exhibited acceptable intra-rater reliability and repeatability for heart rate, but not percent O saturation that was in part attributed to the non-steady state protocol [25].
While repeatability may be quantified with various statistics, SRD% was implied as superior to ICC since it could potentially identify responsiveness to experimental treatments [19]. SRD% scores are derived from standard error of estimate values. SRD’s merits to assess reproducibility were apparent from a test-retest tool to detect changes to subjects, which was deemed useful for both sample size estimation and data interpretation since it identified minimally important variations [27]. SRD values were obtained from studies that also employed non-gravity-dependent hardware; values ranged from 12 to 850 for workout data obtained from the non-gravity-dependent high-speed device [28], and 4 to 94 for the flywheel ergometer model used in the current study [16]. However, most current SRD% scores are so high they are irrelevant or meaningless to help determine true differences among paired values [19]. An overall assessment of all current statistics shows a low, yet realistic level of data repeatability for Cosmed’s K4 b2 portable metabolic cart given the exercise protocol, hardware and intensity examined.
Unlike current data repeatability results, there were significant correlations between the current work volume and dependent variables related to metabolic cost, which concur with prior results [13, 20, 21]. Like the current protocol, prior studies entailed both concentric and eccentric actions [10, 11, 13]. Total-body workouts, done with standard resistive exercise equipment at two different intensities, saw non-significant inter-gender differences in absolute energy cost measured by a Cosmed K4 b2 portable cart [13]. Yet relative energy cost was significantly higher in women at each intensity. It was implied women rely more on aerobic metabolism as they strength train [13]. Energy costs for leg press workouts, done with standard resistive exercise equipment with different amounts of concentric and eccentric actions, were examined in men and showed eccentric actions accounted for only 14% of the net energy cost [11].
Yet a similar leg press investigation done on the same exercise hardware used in the current study saw eccentric actions did not raise energy costs beyond those produced by concentric activity [10, 20]. With an additional 3600 Joules of eccentric work done at no extra caloric cost, the lower energy expenditure from flywheel workouts was attributed to greater elastic energy usage for this hardware [10, 20]. Finally, from perhaps the only prior study on the metabolic cost of seated calf press workouts done on a flywheel ergometer, subjects did two separate 3-set 10-repetition exercise bouts [21]. Net caloric cost was 52.8 3.3 kcals and yielded an inter-workout ICC value of 0.88. Stepwise multivariate regression revealed body surface area, height, and mass explained 97, 99 and 100% respectively of net caloric cost variance [21]. Current results support significant correlations between work volume and metabolic cost seen previously [13, 20, 21].
Tables 4 and 5 offer little support for the hypothesis that Cosmed K4 b2 portable metabolic carts yield repeatable data from consecutive and identical strength training workouts. Yet partial support for our hypothesis is evident from prior steady-state studies [5, 7, 29]. Cosmed K4 b2 data was assessed from VO and VCO measurement comparisons from a protocol that entailed walking 400 meters, followed by a ten-minute rest and then a second 400-meter walk [29]. Cosmed K4 b2 values were compared to a Medgraphics D-series gas exchange unit. ICC values between the systems ranged from 0.94–0.98 for VO and VCO measurements. It was concluded the Cosmed K4 b2 produced acceptable VO and VCO values during steady-state submaximal exercise [29]. With data from maximal and submaximal cycle ergometry workouts, measurement accuracy for the Cosmed K4 b2 was compared to a CPX metabolic system [3]. Both protocols yielded non-significant VO inter-system differences. It was concluded, like the aforementioned study [29], the Cosmed K4 b2 accurately measured VO from steady-state activity [3].
Other steady-state studies also saw low data variability from the Cosmed K4 b2 portable metabolic cart [5, 7]. Male endurance athletes ( 19) did three running tests with data collected by the Cosmed K4 b2 system [7]. Two tests were run on a treadmill, while a third entailed outdoor running. Results yielded non-significant amounts of systematic error for most dependent variables examined [7]. Another study assessed the Cosmed K4 b2 portable metabolic system’s accuracy, with data from ten men who did identical cycle ergometry workouts on consecutive days as part of a counter-balanced design [5]. The Cosmed system quantified respiratory data for one workout, while Douglas Bags were used for the other. Despite significantly higher VO values measured by the Cosmed system, the magnitude of the difference was small and only reached significance due to very small standard errors [5]. It was concluded the Cosmed K4 b2 unit had acceptable accuracy for VO values over a wide range of steady-state efforts [5].
Yet like many current dependent variables, other studies that either did, or did not, entail steady-state activity saw greater variability that infers less data repeatability from Cosmed K4 b2 units [6, 30, 31]. Resting energy costs measured by a Cosmed K4 b2 system were compared to those from a Deltatrac II metabolic unit [31]. Subjects lay supine for 30 minutes as their data were collected over the final 20 minutes [31]. Resting energy costs for Cosmed K4 b2 and Deltatrac II units were 1846 543 and 1579 495 kcals day respectively. The Cosmed K4 b2 was not deemed accurate or reliable to assess resting energy costs [31]. At several submaximal intensities, a Cosmed K4 b2 unit was assessed for FeO, FeCO, and VE measurement accuracy [6]. Subjects ran five minutes each at 8, 11 and 14 km hr at 1% grades. Cosmed values were compared to those obtained from a Morgan ventilation monitor. FeO, FeCO, and VE results varied differently among the systems and the Cosmed routinely yielded lower values, yet paired data produced similar correlations. It was implied consistent error existed between systems that limited Cosmed’s accuracy and repeatability, which was deemed physiologically significant [6].
Finally, VE, VO, VCO, FeO and FeCO were measured to assess test-retest data repeatability from a Cosmed K4 b2 unit over time [30]. Cosmed values were also compared to those from a laboratory-based metabolic cart [30]. To assess these objectives, 12 trained men performed multiple runs on a treadmill. The runs lasted ten, three and one minute each; they were separated by 10-minute rest periods. ICC values ranged from 0.40–0.88; with comparatively lower VE, VO and VCO values from the one-minute run, yet that same run elicited higher ICC FeO and FeCO values. It was implied Cosmed K4 b2 units yielded accurate VE, VO, VCO, FeO and FeCO values from the ten- and three-minute runs. Yet one-minute run results, as well as when all data are compared to the same variables from a laboratory-based metabolic cart, were less accurate [30]. Like the current study, reasons for less accurate and repeatable data from the Duffield and Pinnington investigations are likely due to the exercise protocols, hardware and intensity employed [6, 30]. Thus, current repeatability results concur with outcomes from other investigations that entailed non-steady state exercise [6, 30].
Conclusion
Data repeatability was assessed for Cosmed K4 b2 portable metabolic cart data collected before, during and after identical strength training workouts. Data variability for FAT% elicited non-significant inter-gender substrate utilization differences. Significant correlations were noted between work volume and metabolic cost. Yet current results imply low data repeatability as evidenced by many high SRD% values, which are likely realistic given the exercise protocol, hardware and intensity examined. With these constraints, perhaps only shorter times between workouts would have improved data repeatability. Current data repeatability results agree with study outcomes that also entailed non-steady state exercise [6, 30].
Author contributions
All authors contributed equally.
Ethical considerations
This study was approved by a university-based institutional review board (IRB # 14–56) for human data collection. All subjects gave informed written consent prior to participation in the study.
Funding
This project received no financial assistance.
Footnotes
Acknowledgments
We thank our subjects for their participation.
Conflict of interest
The authors declare they have no conflicts of interest. Given his role as an Editorial Board Member, John F Caruso had no involvement nor access to information regarding the peer review of this article.
References
1.
KoscielniakAA. Development and testing of a metabolic workload measurement system for space suits. Master’s Thesis. University of Maryland, Department of Aerospace Engineering, 2007.
2.
De LorenzoASorgeRPBertiniI, et al. Validity and reliability of a new portable telemetric calorimeter designed to measure oxygen consumption and carbon dioxide production. Diabetes Nutr Metab.2001; 14(5): 268-76.
3.
HausswirthCBigardAXLe ChevalierJM. The COSMED K4 telemetry system as an accurate device for oxygen uptake measurements during exercise. Int J Sports Med.1997; 18(6): 449-53.
4.
MaioloCMelchiorriGIacopinoL, et al. Physical activity energy expenditure measured using a portable telemetric device in comparison with a mass spectrometer. Br J Sports Med.2003; 37(5): 445-7.
5.
McLaughlinJEKingGAHowleyET, et al. Validation of the Cosmed K4 b2 portable metabolic system. Int J Sports Med.2001; 22(4): 280-4.
6.
PinningtonHCWongPTayJ, et al. The level of accuracy and agreement in measures of FEO2, FECO2 and VE between the Cosmed K4 b2 portable, respiratory gas system and a metabolic cart. J Sci Med Sport.2001; 4(3): 324-35.
7.
RossRDuhishyALAGonzalez-HaroC. Validation of the Cosmed K4 b2 portable metabolic system during running outdoors. J Strength Cond Res.2020; 34(1): 124-33.
8.
BoydRFatoneSRoddaJ, et al. High- or low-technology measurements of energy expenditure in clinical gait analysis? Dev Med Child Neurol.1999; 41: 676-82.
9.
KervioGCarreFVilleNS. Reliability and intensity of the six-minute walk test in healthy elderly subjects. Med Sci Sports Exerc.2003; 35(1): 169-74.
10.
CarusoKFHernandezDASaitoK, et al. Inclusion of eccentric actions on net caloric cost resulting from isoinertial resistance exercise. J Strength Cond Res.2003; 17: 549-55.
11.
DudleyGATeschPAHarrisRT, et al. Influence of eccentric actions on the metabolic cost of resistance exercise. Aviat Space Environ Med.1991; 62: 678-82.
12.
DavisonSWRumplerWVBarbosaAG, et al. Metabolic responses to high-speed exercise measured via whole room calorimetry. Med Sportiva.2014; 18: 185-91.
13.
MorganBWoodruffSJTiidusPM. Aerobic energy expenditure during recreational weight training in females and males. J Sport Sci Med.2003; 2: 117-22.
14.
https://www.labmate-online.com/news/news-and-views/5/breaking-news/what-is-the-difference-between-repeatability-and-reproducibility/30638. Accessed September 22, 2021.
15.
ParmarPPerryRCesarzG, et al. Physiological effects of spaceflight/unloading and the mitigating effects of flywheel-based resistive exercise. Grav Space Res.2016; 4(1): 64-77.
16.
CarusoJFWilliamsJAHariP, et al. Data reproducibility from the instrumentation of an inertial resistance exercise device suggested for use during space travel. Isok Exerc Sci.2006; 14: 371-82.
17.
AlknerBTeschP. Efficacy of a gravity-independent resistance exercise device as a countermeasure to muscle atrophy during 29-day bed rest. Acta Physiol Scand.2004; 181: 345-57.
18.
CarusoJFHernandezDAPorterA, et al. Integrated electromyography and performance outcomes to inertial resistance exercise. J Strength Cond Res.2006; 20: 151-6.
19.
BeckermanHRoebroeckMELankhorstGJBecherJGBezemerPDVerbeekALM. Smallest real difference, a link between reproducibility and responsiveness. Qual Life Res.2001; 10: 571-8.
20.
CarusoJFHamillJLYamauchiM, et al. Net caloric cost and electromyographic responses to seated leg presses performed against inertial resistance. Isok Exerc Sci.2006; 14: 15-23.
21.
CarusoJFHerronJCHernandezDA, et al. Metabolic responses to the seated calf press exercise performed against inertial resistance. Aviat Space Environ Med.2005; 76: 1019-23.
22.
FlansbjerUBHombackAMDownhamD, et al. Reliability of gait performance tests in men and women with hemiparesis after stroke. J Rehabil Med.2005; 37: 75-82.
23.
KovelskiJEIngersollCDKnightKL, et al. Reliability of the BTE dynatrack isotonic dynamometer. Isok Exerc Sci.1996; 14: 53-62.
24.
KovelskiJEHeitmanRJGurichekLR, et al. Reliability and effects of leg dominance on lower extremity isokinetic force and work using the closed chain rider system. J Sports Rehabil.1997; 6: 319-26.
25.
CarusoJFSheffletteACS. Intra-rater reliability and repeatability of pulse oximetry values obtained before, between sets, and after resistive exercise. Isok Exerc Sci.2021; doi: 10.3233/IES-200237.
26.
VachonABerrymanNMujikaI, et al. Reliability of a repeated high-intensity effort test for elite rugby union players. Sports.2020; 8(5): 72.
27.
HérbertRSpiegehalterDJBrayneC. Setting the minimally metrically detectable change on disability rating scales. Arch Phys Med Rehabil.1997; 78: 1305-8.
28.
CarusoJFHariPCodayMA, et al. Performance evaluation of a high-speed inertial exercise trainer. J Strength Cond Res.2008; 22: 1760-8.
29.
ShrackJFerrucciL. Validation of the Cosmed K4 b2 portable metabolic system in measuring steady-state walking energy expenditure. Med Sci Sports Exerc Suppl.2008; 40(5): pS104.
30.
DuffieldRDawsonBPinnongtonHC, et al. Accuracy and reliability of a Cosmed K4 b2 portable gas analysis system. J Sci Med Sport.2004; 7: 11-22.
31.
LittlewoodRAWhiteMSBellKL, et al. Comparison of the Cosmed K4 b2 and the Deltatrac IITM metabolic cart in measuring resting energy expenditure in adults. Clin Nutr.2002; 21(6): 491-7.
