Eur J Appl Physiol DOI 10.1007/s00421-014-3071-y
ORIGINAL ARTICLE
Metabolic accommodation to running on a body weight‑supported treadmill David K. P. McNeill · Hendrik D. de Heer · Cody P. Williams · J. Richard Coast
Received: 10 July 2014 / Accepted: 27 November 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Body weight-supported treadmill training using positive air pressure has become increasingly popular, but little is known about the metabolic adaptations to these treadmills. This study aimed to evaluate the existence and length of a metabolic accommodation period to running on a lower body positive pressure (LBPP) treadmill. Methods A total of eight recreational runners (5 males and 3 females) ran 15 min trials (5 min at 50, 70, and 90 % body weight) on the AlterG Anti-gravity® P200 treadmill. No verbal instruction was given on how to run on the device. Their trial pace corresponded to 70–80 % of their velocity measured at V˙ O2max on a standard treadmill. Trials were continued until no significant metabolic change was observed. Two-way repeated measures analysis of variance was used to analyze changes in V˙ O2 across trials and levels of unloading. Results Participants completed 7 trials. Comparing trial 1 to the average of trials 5, 6, and 7, V˙ O2 decreased from 29.6 ± 3.8 to 23.6 ± 4.4 ml/kg/min at 50 % body weight (~20 % reduction), from 33.7 ± 4.5 to 29.2 ± 5.1 ml/kg/ min at 70 % body weight (~13 % reduction), and from 41.0 ± 7.7 to 36.6 ± 5.6 ml/kg/min at 90 % body weight (~11 % reduction). No significant reduction occurred after trial 4 at any level of support. Communicated by Peter Krustrup. D. K. P. McNeill · C. P. Williams · J. R. Coast (*) Department of Biological Sciences, Northern Arizona University, PO Box 5640, Flagstaff, AZ 86011‑5640, USA e-mail: [email protected] H. D. de Heer Department of Physical Therapy and Athletic Training, Northern Arizona University, Flagstaff, AZ, USA
Conclusions An accommodation effect of running on a treadmill with LBPP was observed and reached after 60 min of running (4 trials of 15 min). The accommodation effect was the largest at the greatest level of body weight support. These data suggest the importance of an accommodation period for reliable measures of metabolic cost to be made. Keywords Accommodation · Anti-gravity treadmill · AlterG® · Body weight support · Unloading · Metabolic test Abbreviations BW Body weight BWS Body weight support LBPP Lower body positive pressure ANOVA Analysis of variance
Introduction Running economy has been evaluated in a variety of situations and in many different populations over the years. Generally, the examination of running economy occurs on treadmills, which can pose a challenge to experienced runners’ regular, over-ground running economy when they are not experienced at running on a treadmill. Van Ingen Schenau (1980) described an apprehension to treadmill running upon initial exposure, which resulted in a shortened stride length and a higher stride frequency compared to above ground running. It is also possible that due to a smaller proprioceptive field posed by the narrow belt of the treadmill, and a constant visual field, mechanics are initially altered in such a way as to produce a more stable gait, but at the expense of running economy. Such
13
a notion is in agreement with Wall and Charteris (1982), who noted a lowering of the center of mass during the first tentative strides on the treadmill among novice treadmill walkers. However, these changes appear to occur acutely and have been shown by Schieb (1986) to stabilize after approximately 45 min of treadmill running. Research that has examined metabolic accommodation to treadmill running contends that 45–60 min of treadmill practice is sufficient for attaining stable measurements of V˙ O2 and running economy (Burleson et al. 1994; Morgan et al. 1991; Morgan and Craib 1992; Schieb 1986). In recent years, treadmill devices that provide body weight support (BWS) while running or walking have become increasingly popular in rehabilitation and performance settings. BWS treadmills are novel for most runners and may affect gait patterns. From a metabolic perspective, research using BWS treadmill devices has been concerned with quantifying the relationship between the degree of BWS provided, and the accompanying drop in metabolic cost. Harness systems have long been used to mechanically lift a runner off the treadmill surface. Farley and McMahon (1992) described a proportionality between the drop in the metabolic cost of running 3 m/s and the reduction in weight caused by a suspension harness. Teunissen et al. (2007) and Bijker (2003) suggested that the drop in metabolic cost was slightly less than proportional to the amount of BWS provided. More recent technologies used to examine the relationship between weight unloading and metabolic cost have used lower body positive pressure (LBPP) chambers surrounding a treadmill and attached at a person’s hips. Using an LBPP treadmill, Grabowski and Kram (2008) described a linear, but slightly less than proportional drop in metabolic cost with reduced body weight (BW) during running, which is in agreement with the findings of Teunissen et al. (2007). However, none of the studies to date have described incorporating a period of accommodation for participants in their study protocol as has been used in research using conventional treadmills. To our knowledge, no studies have evaluated the change in metabolic cost to see if there is an accommodation period with unweighting, as there is in conventional treadmill running. Therefore, in this study, we sought to examine how long such accommodation took on a LBPP treadmill. We hypothesized that the metabolic cost of running at a predetermined speed would stabilize after approximately 60 min of running on the device, as has been seen in regular treadmill running.
Methods The protocol was approved by the Institutional Review Board of Northern Arizona University. All participants
13
Eur J Appl Physiol
signed an informed consent form prior to initiation of the data collection. Details that might disclose the identity of the subjects under study have been omitted. A total of 8 participants (5 males and 3 females) were recruited from the local running and triathlon community. All participants were recreational runners or triathletes who had experience with treadmill running, but none had ever run on a treadmill providing BWS. All participants were capable of comfortably running 3.35 m/s (12.1 km/hr). Before the beginning of testing, participants filled out a questionnaire, detailing weekly running distance. The questionnaire also asked for a self-reported assessment of a regular easy run pace. This self-reported pace formed the basis for the prescription of an initial V˙ O2max test protocol, and the subsequent prescription of treadmill speed for the BWS running protocol. The V˙ O2max protocol, performed on a regular treadmill (Woodway USA Inc. Waukesha, WI), consisted of continuous, 3-min stages that progressed only in speed, and maintained a zero percent treadmill incline. This protocol better reflected the type of level running that the participants regularly partook in. Prior to performing the V˙ O2 max test, participants were weighed with their shoes off (Tanita BF679 W portable scale, Tanita Corporation, Arlington Heights, IL), and this weight was used for calculating relative V˙ O2. The initial speed for the maximal test was 0.67 and 0.54 m/s slower than each participant’s self-reported easy running pace, for male and female participants, respectively. Every 3 min, the treadmill speed was increased by 0.67 and 0.54 m/s for male and female participants, respectively, until volitional exhaustion. Each participant reached his or her V˙ O2max within no longer than 14 min. V˙ O2 was measured using a metabolic measurement system (TrueOne 2400, Parvo Medics, Sandy, UT), calibrated according to the manufacturer’s specifications. Where possible, each participant’s self-reported easy run pace was then used for the BWS running protocol if their measured V˙ O2 at that self-reported pace was between 70 and 80 % of their V˙ O2max. In six of the eight participants, this self-reported pace fell within this range of 70– 80 % of V˙ O2max. In two participants, the self-reported pace was associated with a V˙ O2 that was above 80 % of their V˙ O2max, and so their self-reported easy run pace was subsequently decreased for the BWS running protocol by less than 0.45 m/s to approximate an equivalent V˙ O2 in the appropriate sub-maximal range. This intensity was used to help ensure that the runners were able to maintain the pace for the duration of the subsequent tests without undue strain. Following V˙ O2max testing, participants completed seven 15-min running sessions on the LBPP treadmill (AlterG®, Fremont, CA), all run at the speed associated with 70– 80 % of V˙ O2max on the regular treadmill without BWS.
Eur J Appl Physiol
The sessions were performed with at least 2 days between them, such that the seven sessions were completed within a three-week span. The LBPP device incorporated the same type of treadmill (Woodway USA Inc. Waukesha, WI) as was used during the V˙ O2max test. At the beginning of each 15-min session, participants were first weighed using the same scales as were used prior to V˙ O2max testing with shoes off, and the specialized AlterG® shorts being worn. At the beginning of each test, participants stood on the AlterG® treadmill surface, and zipped the skirt of the specialized shorts into the aperture of the inflation chamber. The skirt of the shorts was positioned at the height of the anterior superior iliac spine (ASIS), while the height of the top of the frame of the positive pressure chamber, to which the skirt was attached (and therefore created the seal), was positioned in the frame rung that most closely corresponded to the height of the participant’s ASIS. Once “zipped in” the AlterG® pressure chamber was calibrated using the automatic computer interface. Additionally, the metabolic measurement system was calibrated before each individual session in the same manner as prior to the V˙ O2max test. Once both the AlterG® and metabolic measurement systems were calibrated, the session was started, with the computer interface covered and hidden from the participant, so that they did not know the speed or percentage unweighting applied. The participants were not given any specific instructions before or during each test on how to run with BWS. The only instruction given to each runner before the treadmill belt started was that running would initially feel a little strange. Each of the seven tests was identical for each participant. Speed was constant, having been set at the pre-determined pace derived from the self-reported comfortable pace (corresponding to 70–80 % of V˙ O2max) and the results of the V˙ O2max test (as described previously). During each 15-min session, percentage body weight was first set at 50 % body weight, then changed to 70 % body weight after 5 min, and then to 90 % body weight after 10 min. At 15 min, the test was terminated. V˙ O2 was continuously measured, and the data used for analysis were taken during the final minute of each 5-min increment at each of the three levels of BWS. A two-way repeated measures ANOVA (trial number × unloading level) was used to analyze the change in V˙ O2 across trials and levels of unloading.
Table 1 Participant demographics Variables
Total (n = 8)
Males (n = 5)
Females (n = 3)
Mean ± SD
Mean ± SD
Mean ± SD
Age (years) Weight (kg) V˙ O2max (ml/ kg/min) Height (cm) Runs/week
23.6 ± 5.4 67.3 ± 9.4 58.4 ± 7.1
24.4 ± 6.7 73.5 ± 4.3 61.2 ± 6.5
22.3 ± 2.9 57.0 ± 4.4 53.7 ± 6.2
176.4 ± 10.4 5.1 ± 1.8
183.2 ± 4.2 4.2 ± 1.6
165.0 ± 5.6 6.5 ± 1.3
Miles/week
36.3 ± 16.6
31.0 ± 15.2
45.0 ± 18.0
demographics describe experienced but not elite, trained male and female runners. The average V˙ O2 across trials is summarized in Table 2 and graphically in Fig. 1. There was a main effect of body weight (p
ORIGINAL ARTICLE
Metabolic accommodation to running on a body weight‑supported treadmill David K. P. McNeill · Hendrik D. de Heer · Cody P. Williams · J. Richard Coast
Received: 10 July 2014 / Accepted: 27 November 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Body weight-supported treadmill training using positive air pressure has become increasingly popular, but little is known about the metabolic adaptations to these treadmills. This study aimed to evaluate the existence and length of a metabolic accommodation period to running on a lower body positive pressure (LBPP) treadmill. Methods A total of eight recreational runners (5 males and 3 females) ran 15 min trials (5 min at 50, 70, and 90 % body weight) on the AlterG Anti-gravity® P200 treadmill. No verbal instruction was given on how to run on the device. Their trial pace corresponded to 70–80 % of their velocity measured at V˙ O2max on a standard treadmill. Trials were continued until no significant metabolic change was observed. Two-way repeated measures analysis of variance was used to analyze changes in V˙ O2 across trials and levels of unloading. Results Participants completed 7 trials. Comparing trial 1 to the average of trials 5, 6, and 7, V˙ O2 decreased from 29.6 ± 3.8 to 23.6 ± 4.4 ml/kg/min at 50 % body weight (~20 % reduction), from 33.7 ± 4.5 to 29.2 ± 5.1 ml/kg/ min at 70 % body weight (~13 % reduction), and from 41.0 ± 7.7 to 36.6 ± 5.6 ml/kg/min at 90 % body weight (~11 % reduction). No significant reduction occurred after trial 4 at any level of support. Communicated by Peter Krustrup. D. K. P. McNeill · C. P. Williams · J. R. Coast (*) Department of Biological Sciences, Northern Arizona University, PO Box 5640, Flagstaff, AZ 86011‑5640, USA e-mail: [email protected] H. D. de Heer Department of Physical Therapy and Athletic Training, Northern Arizona University, Flagstaff, AZ, USA
Conclusions An accommodation effect of running on a treadmill with LBPP was observed and reached after 60 min of running (4 trials of 15 min). The accommodation effect was the largest at the greatest level of body weight support. These data suggest the importance of an accommodation period for reliable measures of metabolic cost to be made. Keywords Accommodation · Anti-gravity treadmill · AlterG® · Body weight support · Unloading · Metabolic test Abbreviations BW Body weight BWS Body weight support LBPP Lower body positive pressure ANOVA Analysis of variance
Introduction Running economy has been evaluated in a variety of situations and in many different populations over the years. Generally, the examination of running economy occurs on treadmills, which can pose a challenge to experienced runners’ regular, over-ground running economy when they are not experienced at running on a treadmill. Van Ingen Schenau (1980) described an apprehension to treadmill running upon initial exposure, which resulted in a shortened stride length and a higher stride frequency compared to above ground running. It is also possible that due to a smaller proprioceptive field posed by the narrow belt of the treadmill, and a constant visual field, mechanics are initially altered in such a way as to produce a more stable gait, but at the expense of running economy. Such
13
a notion is in agreement with Wall and Charteris (1982), who noted a lowering of the center of mass during the first tentative strides on the treadmill among novice treadmill walkers. However, these changes appear to occur acutely and have been shown by Schieb (1986) to stabilize after approximately 45 min of treadmill running. Research that has examined metabolic accommodation to treadmill running contends that 45–60 min of treadmill practice is sufficient for attaining stable measurements of V˙ O2 and running economy (Burleson et al. 1994; Morgan et al. 1991; Morgan and Craib 1992; Schieb 1986). In recent years, treadmill devices that provide body weight support (BWS) while running or walking have become increasingly popular in rehabilitation and performance settings. BWS treadmills are novel for most runners and may affect gait patterns. From a metabolic perspective, research using BWS treadmill devices has been concerned with quantifying the relationship between the degree of BWS provided, and the accompanying drop in metabolic cost. Harness systems have long been used to mechanically lift a runner off the treadmill surface. Farley and McMahon (1992) described a proportionality between the drop in the metabolic cost of running 3 m/s and the reduction in weight caused by a suspension harness. Teunissen et al. (2007) and Bijker (2003) suggested that the drop in metabolic cost was slightly less than proportional to the amount of BWS provided. More recent technologies used to examine the relationship between weight unloading and metabolic cost have used lower body positive pressure (LBPP) chambers surrounding a treadmill and attached at a person’s hips. Using an LBPP treadmill, Grabowski and Kram (2008) described a linear, but slightly less than proportional drop in metabolic cost with reduced body weight (BW) during running, which is in agreement with the findings of Teunissen et al. (2007). However, none of the studies to date have described incorporating a period of accommodation for participants in their study protocol as has been used in research using conventional treadmills. To our knowledge, no studies have evaluated the change in metabolic cost to see if there is an accommodation period with unweighting, as there is in conventional treadmill running. Therefore, in this study, we sought to examine how long such accommodation took on a LBPP treadmill. We hypothesized that the metabolic cost of running at a predetermined speed would stabilize after approximately 60 min of running on the device, as has been seen in regular treadmill running.
Methods The protocol was approved by the Institutional Review Board of Northern Arizona University. All participants
13
Eur J Appl Physiol
signed an informed consent form prior to initiation of the data collection. Details that might disclose the identity of the subjects under study have been omitted. A total of 8 participants (5 males and 3 females) were recruited from the local running and triathlon community. All participants were recreational runners or triathletes who had experience with treadmill running, but none had ever run on a treadmill providing BWS. All participants were capable of comfortably running 3.35 m/s (12.1 km/hr). Before the beginning of testing, participants filled out a questionnaire, detailing weekly running distance. The questionnaire also asked for a self-reported assessment of a regular easy run pace. This self-reported pace formed the basis for the prescription of an initial V˙ O2max test protocol, and the subsequent prescription of treadmill speed for the BWS running protocol. The V˙ O2max protocol, performed on a regular treadmill (Woodway USA Inc. Waukesha, WI), consisted of continuous, 3-min stages that progressed only in speed, and maintained a zero percent treadmill incline. This protocol better reflected the type of level running that the participants regularly partook in. Prior to performing the V˙ O2 max test, participants were weighed with their shoes off (Tanita BF679 W portable scale, Tanita Corporation, Arlington Heights, IL), and this weight was used for calculating relative V˙ O2. The initial speed for the maximal test was 0.67 and 0.54 m/s slower than each participant’s self-reported easy running pace, for male and female participants, respectively. Every 3 min, the treadmill speed was increased by 0.67 and 0.54 m/s for male and female participants, respectively, until volitional exhaustion. Each participant reached his or her V˙ O2max within no longer than 14 min. V˙ O2 was measured using a metabolic measurement system (TrueOne 2400, Parvo Medics, Sandy, UT), calibrated according to the manufacturer’s specifications. Where possible, each participant’s self-reported easy run pace was then used for the BWS running protocol if their measured V˙ O2 at that self-reported pace was between 70 and 80 % of their V˙ O2max. In six of the eight participants, this self-reported pace fell within this range of 70– 80 % of V˙ O2max. In two participants, the self-reported pace was associated with a V˙ O2 that was above 80 % of their V˙ O2max, and so their self-reported easy run pace was subsequently decreased for the BWS running protocol by less than 0.45 m/s to approximate an equivalent V˙ O2 in the appropriate sub-maximal range. This intensity was used to help ensure that the runners were able to maintain the pace for the duration of the subsequent tests without undue strain. Following V˙ O2max testing, participants completed seven 15-min running sessions on the LBPP treadmill (AlterG®, Fremont, CA), all run at the speed associated with 70– 80 % of V˙ O2max on the regular treadmill without BWS.
Eur J Appl Physiol
The sessions were performed with at least 2 days between them, such that the seven sessions were completed within a three-week span. The LBPP device incorporated the same type of treadmill (Woodway USA Inc. Waukesha, WI) as was used during the V˙ O2max test. At the beginning of each 15-min session, participants were first weighed using the same scales as were used prior to V˙ O2max testing with shoes off, and the specialized AlterG® shorts being worn. At the beginning of each test, participants stood on the AlterG® treadmill surface, and zipped the skirt of the specialized shorts into the aperture of the inflation chamber. The skirt of the shorts was positioned at the height of the anterior superior iliac spine (ASIS), while the height of the top of the frame of the positive pressure chamber, to which the skirt was attached (and therefore created the seal), was positioned in the frame rung that most closely corresponded to the height of the participant’s ASIS. Once “zipped in” the AlterG® pressure chamber was calibrated using the automatic computer interface. Additionally, the metabolic measurement system was calibrated before each individual session in the same manner as prior to the V˙ O2max test. Once both the AlterG® and metabolic measurement systems were calibrated, the session was started, with the computer interface covered and hidden from the participant, so that they did not know the speed or percentage unweighting applied. The participants were not given any specific instructions before or during each test on how to run with BWS. The only instruction given to each runner before the treadmill belt started was that running would initially feel a little strange. Each of the seven tests was identical for each participant. Speed was constant, having been set at the pre-determined pace derived from the self-reported comfortable pace (corresponding to 70–80 % of V˙ O2max) and the results of the V˙ O2max test (as described previously). During each 15-min session, percentage body weight was first set at 50 % body weight, then changed to 70 % body weight after 5 min, and then to 90 % body weight after 10 min. At 15 min, the test was terminated. V˙ O2 was continuously measured, and the data used for analysis were taken during the final minute of each 5-min increment at each of the three levels of BWS. A two-way repeated measures ANOVA (trial number × unloading level) was used to analyze the change in V˙ O2 across trials and levels of unloading.
Table 1 Participant demographics Variables
Total (n = 8)
Males (n = 5)
Females (n = 3)
Mean ± SD
Mean ± SD
Mean ± SD
Age (years) Weight (kg) V˙ O2max (ml/ kg/min) Height (cm) Runs/week
23.6 ± 5.4 67.3 ± 9.4 58.4 ± 7.1
24.4 ± 6.7 73.5 ± 4.3 61.2 ± 6.5
22.3 ± 2.9 57.0 ± 4.4 53.7 ± 6.2
176.4 ± 10.4 5.1 ± 1.8
183.2 ± 4.2 4.2 ± 1.6
165.0 ± 5.6 6.5 ± 1.3
Miles/week
36.3 ± 16.6
31.0 ± 15.2
45.0 ± 18.0
demographics describe experienced but not elite, trained male and female runners. The average V˙ O2 across trials is summarized in Table 2 and graphically in Fig. 1. There was a main effect of body weight (p
