Secondary Logo

Journal Logo

V̇o2 Requirements of Boxing Exercises

Arseneau, Eric; Mekary, Saïd; Léger, Luc A

Journal of Strength and Conditioning Research: February 2011 - Volume 25 - Issue 2 - p 348-359
doi: 10.1519/JSC.0b013e3181ef64cb
Original Research
Free

Arseneau, E, Mekary, S, and Léger, LA. V̇o2 requirements of boxing exercises. J Strength Cond Res 25(2): 348-359, 2011-The purpose of this study was to quantify the physiological requirements of various boxing exercises such as sparring, pad work, and punching bag. Because it was not possible to measure the oxygen uptake (V̇o2) of “true” sparring with a collecting gas valve in the face, we developed and validated a method to measure V̇o2 of “true” sparring based on “postexercise” measurements. Nine experienced male amateur boxers (Mean ± SD: age = 22.0 ± 3.5 years, height = 176.0 ± 8.0 cm, weight = 71.4 ± 10.9 kg, number of fights = 13.0 ± 9.5) of regional and provincial level volunteered to participate in 3 testing sessions: (a) maximal treadmill test in the LAB, (b) standardized boxing training in the GYM, and (c) standardized boxing exercises in the LAB. Measures of V̇o2, heart rate (HR), blood lactate concentration [LA], rated perceived exertion level, and punching frequencies were collected. V̇o2 values of 43.4 ± 5.9, 41.1 ± 5.1, 24.7 ± 6.1, 30.4 ± 5.8, and 38.3 ± 6.5 ml·kg−1·min−1 were obtained, which represent 69.7 ± 8.0, 66.1 ± 8.0, 39.8 ± 10.4, 48.8 ± 8.5, and 61.7 ± 10.3%V̇o2peak for sparring, pad work, and punching bag at 60, 120, and 180 b·min−1, respectively. Except for lower V̇o2 values for punching the bag at 60 and 120 b·min−1 (p < 0.05), there was no V̇o2 difference between exercises. Similar pattern was obtained for %HRmax with respective values of 85.5 ± 5.9, 83.6 ± 6.3, 67.5 ± 3.5, 74.8 ± 5.9, and 83.0 ± 6.0. Finally, sparring %HRmax and [LA] were slightly higher in the GYM (91.7 ± 4.3 and 9.4 ± 2.2 mmol·L−1) vs. LAB (85.5 ± 5.9 and 6.1 ± 2.3 mmol·L−1). Thus, in this study simulated LAB sparring and pad work required similar V̇o2 (43-41 ml·kg−1·min−1, respectively), which corresponds to ∼70%V̇o2peak. These results underline the importance of a minimum of aerobic fitness for boxers and draw some guidelines for the intensity of training.

Department of Kinesiology, University of Montreal, Montreal, Canada

Address correspondence to Luc Léger, luc.leger@umontreal.ca.

Back to Top | Article Outline

Introduction

Boxing, also known as the “sweet science,” has ironically very little scientific evidence concerning its physiological requirements (14). Therefore, boxers mostly rely on empirical approaches. For a majority of boxers, training methods are established by trial and errors after numerous practices with their boxing team. Of course, this lack of tangible guidelines makes it difficult to determine if the training methods used are optimal for the physiological requirements of competitions. Consequently, the margin of progress and chances of success at the highest level become limited. As a result, it can be discouraging and possibly lead to premature withdrawal or amplify the risk of injuries or failures. The few studies that were found on the physiological requirements of boxing used various methodologies and often reported different results. Measures such as heart rate (HR) and lactate [LA] are easy to obtain and have been reported several times (1,7-9,12,13,17,18,20,24,25). In general, HR values around 170-180 b·min−1 are reported for sparring (7-9,12,18,24,25) and ∼167 b·min−1 for pad work (18), whereas a wider range between 145 and 195 b·min−1 is observed for the punching bag (1,13,17,18,20,24,25). For [LA], values around 9-12 mmol·L−1 are reported after 3 rounds of sparring (9,12) or punching bag (17). On the other hand, oxygen uptake (V̇o2) during “true” sparring or other boxing exercises is difficult to measure and reported values are controversial (see below). Reinvestigating the V̇o2 cost of boxing exercises is thus the focus of this study.

Concerning sparring, Seliger (24,25) reported higher V̇o2 values of 38.9 ± 5.9 ml·kg−1·min−1 (Mean ± SD) than the ones obtained for the punching bag using the same subjects and similar methodology. In their compendiums, Durnin (11) reported values of at least 21.8 and 23.1 ml·kg−1·min−1 or higher for women and men, respectively, whereas Ainsworth et al. (2,3) reported values of 31.5 and 42.0 ml·kg−1·min−1 for sparring and “in ring general” [sic], respectively. In a review article, Ostyn and S'Jongers (21) reported a V̇o2 range of 26.3-42.0 ml·kg−1·min−1. More recently using the treadmill HR/V̇o2 regression approach, Chatterjee et al. (7-9) reported higher mean values from 40.3 ± 7.0 to 46.6 ± 6.6 ml·kg−1·min−1 with peak values between 45.9 ± 6.0 and 50.5 ± 7.9 ml·kg−1·min−1 from round 1 to round 3 (2-minute rounds) for women sparring.

Another typical boxing exercise is “pad work,” which is punching on a partner's pads or mitts. In their study, Morita et al. (18) reported peak values of 48.2 ± 3.8 ml·kg−1·min−1. More recently for similar exercise with Muay Thai Boxing, Crisafulli et al. (10) reported values of 42.5 ± 2.2 ml·kg−1·min−1 (average for 3 3-minute rounds) for pad work involving fists, elbows, knees, and feet to strike and block.

For the punching bag (heavy bag), Seliger (24,25) reported values of 27.0 ± 4.2 ml·kg−1·min−1 using meteorological balloons to measure V̇o2. About 15 years later and using Douglas bags, Morita et al. (18) reported very high V̇o2peak values of 52.5 ± 7.1 ml·kg−1·min−1, which was superior to the V̇o2max of their subjects. In their compendiums (no methodological details), Ainsworth et al. (2,3) and Moy et al. (19) reported lower values of 21.0 and 18.9 ml·kg−1·min−1, respectively. Adams et al. (1) reported V̇o2 costs of 22.9 ± 10.0 ml·kg−1·min−1 for punching at 120 b·min−1, whereas young adult male and female boxers (n = 5 + 5) were continuously connected to a metabolic system without any foot work. Two years later, O'Driscoll et al. (20) did a similar study and reported higher V̇o2 costs of 31.5 ± 6.9 ml·kg−1·min−1 for a higher synchronized pace of 134 b·min−1 for “punches and lower body movements” [sic], whereas young adult male and female boxers (n = 7 + 9) were also continuously connected to a metabolic system. Recently, Kravitz et al. (13) established the V̇o2 cost of punching at 6 different frequencies (60-120 b·min−1) with fitness boxing subjects hitting a SLAMMAN (Fitness Quest, Canton, OH, USA), which is a type of dummy standing on the floor instead of a regular hanging punching bag, and they reported almost similar V̇o2 values of ∼27.0-30.0 ml·kg−1·min−1 regardless of the punching frequency.

As we can see, the energy cost of boxing exercises varies a lot depending on the type of subjects, the methodology used to estimate V̇o2 cost and according to the type of boxing exercises. For sparring, previous studies used hindering equipment (collecting gas valve in the face and portable V̇o2 systems) or estimated V̇o2 from HR/V̇o2 regression, which is not a very accurate procedure (18). In an attempt to solve these problems and to enable proper comparisons, we reinvestigated the energy cost of key boxing exercises such as sparring, pad work and punching bag with a new method to measure V̇o2 without hindering the boxers movement. Furthermore, because these measurements were obtained in the LAB, we wanted to know whether the sparring intensity in the LAB and in the GYM was similar. This was done by comparing the HR, [LA], rated perceived exertion (RPE), and punching frequency of sparring in the GYM and in the LAB. A final objective was to study the effect of punching frequency on the V̇o2 cost of the punching bag. To our knowledge, this study is unique because it is the only one that measures V̇o2 of “true” sparring without any motion hindrance. It is also the sole study investigating the effect of punching frequency on the energy cost of punching the bag with experienced boxers. Knowing accurate V̇o2 requirements is important to properly adjust the training load of boxers.

Back to Top | Article Outline

Methods

Experimental Approach to the Problem

To fulfill our objectives, subjects participated in 3 testing sessions: (a) maximal treadmill test in the LAB, (b) standardized boxing training in the GYM, (c) and standardized boxing exercises in the LAB. Three 2-minute rounds of various boxing exercises with 1-minute rest in between were used as the independent variables. In the GYM, these exercises were sparring and free punching bag, and in the LAB, they were sparring, pad work, and cadence controlled punching bag with alternating straight punches only, without any foot work. The timing of dependent variables (V̇o2, HR, [LA], RPE, and video recording of punching frequency) is illustrated in Figure 1 and detailed thereafter. A minimum of 1 day was required between 2 sessions, and all sessions were conducted within a 3-month span.

Figure 1

Figure 1

Back to Top | Article Outline

Subjects

Nine experienced male amateur boxers (age = 22.0 ± 3.5 years, height = 176.0 ± 8.0 cm, weight = 71.4 ± 10.9 kg, number of fights = 13.0 ± 9.5) of regional and provincial level volunteered to participate in this study (Table 1). All owned a competition passport licensed by the Canadian Amateur Boxing Association and provided written consent. The study protocol was approved by the Ethics Committee of University of Montreal.

Table 1

Table 1

Back to Top | Article Outline

Procedures

Session 1. Maximal Treadmill Test in the LAB

Before the treadmill test, the height and weight were measured without shoes. For the treadmill (Q65, model 645, Quinton Instruments Co., Bothell, WA, USA) protocol, subjects warmed up for 3 minutes at 8 km·h−1 and 0% slope. Thereafter, the speed was increased by 1 km·h−1 every 2 minutes till exhaustion (15). The objective was to determine the V̇o2peak, HRmax, [LA]max, and RPEmax of boxers. The temperature and relative humidity were around 21° C and 35%.

Back to Top | Article Outline

Session 2. Standardized Boxing Training in the GYM

The second session took place in the respective boxing GYM of each boxer, and the goal was to simulate a typical boxing training session including 3 sequences of typical exercises: shadow boxing (2 rounds), sparring (3 rounds on approximately 4.9 by 4.9 m [approximately 16 × 16 ft] competition ring) and freely hitting a punching bag (3 rounds) weighing around 34.0 kg (75.0 lb) depending on the gym. A 10-minute rest was given between each exercise, and this used to put on a head gear, a jock strap, a pair of gloves, and a mouth piece for the sparring. Rounds were of a 2-minute duration with 1-minute standing rest in between. Subjects wore hand wraps for the 3 exercises and 0.45-kg (16-oz) gloves for sparring and punching the bag. Each boxer wore his customary boxing gear and was informed of the procedure beforehand. Subjects were randomly paired and instructed to perform as they usually would during these exercises. The RPE was measured after the third round of sparring and punching bag exercise, and [LA] was recorded at the end of the third round of sparring only. For the whole GYM session, HR was continuously monitored. Sparring and punching bag rounds were videotaped to count the number and frequency of punches thrown. Temperature and relative humidity were not measured but subjectively were relatively constant and similar to LAB conditions.

Back to Top | Article Outline

Session 3. Standardized Boxing Exercises in the LAB

During the third testing session, simulated boxing exercises such as shadow boxing (2 rounds), sparring (3 rounds), pad work (3 rounds), and punching bag (3 rounds) were performed in the LAB. Boxing gear, round duration, rest period between exercises and rounds were the same as in session 2. Exercises were done on a 4.3 by 4.3 m (14 × 14 ft) free space without ropes. Sparring was a simulation of the sparring done in the GYM. For the pad work (new exercise), subjects had to hit the partner's pads in a predetermined routine of 4 different combinations that were always repeated in the same order till the end of the rounds. For the right-handed boxers, the combinations consisted of (a) jab, jab, straight right; (b) jab, straight right, left hook, straight right; (c) jab, straight right, jab, straight right; and (d) right uppercut, left uppercut, straight right, and left hook. For the sole left-handed boxer, left and right punches were reversed. Approximately 30 seconds of the first round was needed by the subjects to get familiar with the combinations and afterward they would hold a steady rate. For the punching bag rounds, subjects had to hit a 25-kg (55-lb) bag held motionless by an assistant in a normalized way. They were asked to stand at arm length from the punching bag with their feet stationary at the width of their shoulders while hitting the bag alternating with jabs and straight rights from chin to bag at the beat of a metronome. Rounds 1-3 were done at 60, 120, and 180 b·min−1, respectively.

As for the GYM, the whole session was videotaped, and the HR data were continuously collected. Temperature and relative humidity were around 21° C and 37%. The RPE was obtained at the end of sparring, pad work, and after each punching bag round. In addition, V̇o2 was measured at the end of the third round of sparring and pad work and continuously during each round of punching bag. The [LA] was measured at the end of the sparring and pad work only. For sparring and pad work, boxers were connected to the mouth piece of our collecting gas system immediately after the third round. The system was located about 2 m away from the boxing ring. The whole procedure took about 3-5 seconds, including nose-clip fixation. Once connected to the metabolic system, the subjects were asked to exert themselves at the same intensity as before, by continuing foot work and punching on the gloves or pads of his partner to minimize the recovery process for 60-75 seconds to give time to the metabolic system to reach an equilibrium value. Only equilibrium values were retained as the V̇o2 cost of the activity. The calibration of the metabolic system was completed about 5-10 minutes before the first round of sparring.

That approach to measure postexercise V̇o2 was validated with another group of 9 subjects using similar “rounds” of treadmill exercise, to enable V̇o2 measures during the exercise and postexercise phases. V̇o2 obtained during treadmill exercise was used as our gold standard to assess the accuracy of V̇o2 measured after reconnecting the subjects at the end of a “round” of running (Annex 1). In summary, V̇o2 values obtained at 3 different intensities on the treadmill between 30 and 75 seconds after the reconnection of the subjects were similar to the values obtained during steady state exercise (2-way analysis of variance [ANOVA] for repeated measures and Tukey a posteriori tests (n.s., p > 0.05) and regression analysis: r = 0.96, SEE = 1.6 ml·kg−1·min−1).

Figure 1 describes the kinetics of dependent variables measured for the 3 sessions and also indicates potential direct comparisons of boxing exercises for each variable.

Back to Top | Article Outline

Measuring Apparatus and Tests

For V̇o2 measures, expired gases passed through an open circuit and were analyzed in real time by an automated and computerized system (Moxus Modular Metabolic System, AEI Technologies, Naperville, IL, USA) (5). Results were averaged every 30 seconds for the maximal treadmill test and every 15 seconds for the simulated boxing exercises in the LAB. The HR was measured throughout every session with a Polar S810 HR monitor. Data were recorded every 5 seconds and later transferred on a computer to be analyzed (Polar Electro, Kempele, Finland). The HR measured with the Polar S810 is very reliable (16). The RPE was recorded after every exercise using the Borg scale (6) with scores ranging from 6 to 20 (very, very light to very, very hard). Finally, as a witness of the anaerobic solicitation, [LA] was measured from a fingertip blood sample taken 4 minutes after the third round of the exercise. The blood sample was analyzed with the Lactate Pro (ARKRAY, Inc., Kyoto, Japan), which has been reported valid (22,23).

Back to Top | Article Outline

Statistical Analyses

Four 2-way (round and boxing exercise) ANOVA for repeated measures with Tukey a posteriori tests were used to describe and compare %HRmax and punching frequency of each round of sparring and punching bag in the GYM (Figure 2) and sparring and pad work in the LAB (Figure 3). Six 1-way ANOVAs were used to compare average or peak V̇o2, %V̇o2peak, %HR, [LA], RPE, and punching frequency values of the 3 rounds combined for all possible boxing exercises whether they were done in the GYM or in the LAB (Table 2 and Figure 4). First degree Pearson correlation and SEE between V̇o2 cost of sparring as the dependent variable and body weight and V̇o2peak as the independent variables were also computed to see a possible effect of these 2 variables on V̇o2 cost of sparring. Unless otherwise stated, all reported differences are significant at the p ≤ 0.05 level. Analyses were performed with or without missing data being replaced, giving exactly the same results. Thus, only results with missing data being replaced are reported (n = 9). Mean ± SD format is used in the text.

Table 2

Table 2

Figure 2

Figure 2

Figure 3

Figure 3

Figure 4

Figure 4

Back to Top | Article Outline

Results

GYM Vs. LAB

Sparring was the sole exercise being studied both in the GYM and LAB in the same conditions (Table 2), and it was found that %HRmax and [LA] were higher (p < 0.05) in the natural GYM environment (91.7 ± 4.3 and 85.5 ± 5.9 and 9.4 ± 2.2 and 6.1 ± 2.3 mmol·L−1), whereas RPE and punching frequency were similar. Note that these %HRmax and punching frequency values represent the average of 3 rounds, whereas RPE and [LA] were only taken at the end of last round. V̇o2, %HRmax and RPE obtained for multistage treadmill test, sparring, pad work, and punching the bag at different frequencies (60, 120, and 180 b·min−1) in the LAB are also illustrated in Figure 4 and %HR and RPE for sparring and free punching bag in the GYM.

Back to Top | Article Outline

Round Effects

In the GYM, only HR and punching frequency were recorded from round to round. %HRmax increased (p < 0.05) from round to round for sparring and punching bag (Figure 2). For the punching frequency, however, it increased (p < 0.05) only for the punching bag (Figure 2). For each round, %HRmax was higher for sparring (vs. punching bag, p < 0.05), whereas punching frequency was higher on the punching bag (vs. sparring, p < 0.05).

In the LAB, %HRmax increased only from round 1 to round 2 for sparring and pad work (p < 0.05), whereas the punching frequency increased from round 1 to round 2 for pad work only (p < 0.05) (Figure 3). No pad work exercise was performed in the GYM to allow a similar comparison. In the LAB, punching the bag was done at increasing imposed frequency from round 1 to round 3 and %HRmax, and punching frequency obviously increased (p < 0.05). %HRmax was higher for sparring (vs. pad work, p < 0.05) in round 1 only, whereas punching frequency was higher for each round for pad work (p < 0.05).

Back to Top | Article Outline

Exercise Effects

Because GYM and LAB results were different (p < 0.05) for %HRmax and [LA] and because different exercises were done in each location, exercise effects are reported separately in each location (Table 2 and Figure 4).

In the GYM, concerning %HRmax and punching frequency (Table 2), %HRmax was higher for sparring than for punching bag (91.7 ± 4.3 and 86.9 ± 5.7, p < 0.05), but it was the opposite for punching frequency (35.7 ± 9.9 and 70.6 ± 22.6 b·min−1, p < 0.05). The same was true for each round (Figure 2). In the LAB (Table 2), %HRmax was similar for sparring and pad work (85.5 ± 5.9 and 83.6 ± 6.3), except for round 1 (Sparring > Pad, p < 0.05, Figure 3), whereas punching frequency was higher for pad work than for sparring (61.4 ± 7.9 and 34.9 ± 7.1 b·min−1, p < 0.05). No free punching bag exercise was performed in the LAB to allow similar comparisons.

In the GYM, concerning RPE and [LA], they were measured only after the third round and [LA] was measured after sparring only. Thus, only RPE exercise effects could be assessed, and no difference was found between sparring and punching bag (12.9 ± 1.8 and 12.6 ± 1.9). In the LAB, RPE was similar for sparring and pad work (12.2 ± 1.3 and 13.6 ± 2.1), but [LA] was almost higher for sparring than for pad work (6.1 ± 2.3 and 4.0 ± 1.7 mmol L−1, respectively, p < 0.05).

Most maximal values of the treadmill test (V̇o2peak, HRmax, and RPEmax) were higher (p < 0.05) than values observed while doing sparring, pad work, or punching bag in the GYM (no %V̇o2peak values in that condition) or in the LAB (Figure 4). V̇o2 values of LAB sparring, pad work, and punching bag at 180 b·min−1 were similar (43.4 ± 5.9, 41.1 ± 5.1, and 38.3 ± 6.5 ml·kg−1·min−1, respectively, p > 0.05). However, V̇o2 increased with the punching frequency (24.7 ± 6.1, 30.4 ± 5.8, and 38.3 ± 6.5 at 60, 120, and 180 b·min−1, respectively, p < 0.05). Significant increases with punching frequency (p < 0.05) were also observed for RPE and %HRmax (Figure 4). For %HRmax, sparring, pad work, and punching bag at 180 b·min−1 in the LAB and punching bag in the GYM were all similar, whereas sparring in the GYM was higher (p < 0.05) and punching bag at 60 and 120 b·min−1, was lower (p < 0.05) (Figure 4). For RPE, only punching the bag at 60 b·min−1 was lower than other boxing exercises whether they were done in the LAB or in the GYM (Figure 4).

Our regression analyses indicated that the V̇o2 cost of sparring (or its intensity) was not significantly (p > 0.05) related to body weight (r = 0.53) nor to V̇o2peak (r = 0.51). However the V̇o2 cost of sparring tended to be inversely proportional to the body weight of boxers (Figure 5, upper graph), but proportional to their V̇o2peak (Figure 5, lower graph), a trend that became significant when we duplicated the same results (same dispersion and twice the numbers of subjects).

Figure 5

Figure 5

Back to Top | Article Outline

Discussion

Before discussing our results per se, it is worthwhile to discuss some methodological aspects of this study. The main purpose of this study was to quantify V̇o2 requirements of various boxing exercises such as sparring, pad work, and punching bag. Because V̇o2 could not be measured in the natural GYM environment, we simulated boxing exercises in the LAB and developed a method to measure V̇o2 for those boxing exercises by connecting subjects to the metabolic system immediately after 3 2-minute rounds of sparring and pad work while asking them to maintain the same level of exertion. However, for the punching bag, subjects were connected to the metabolic system during the entire exercise. As indicated previously (Annex 1), this method was proven valid without systematic error when normalized and controlled exercises were performed on the treadmill. Because the subjects subjectively attempted to continue exercising at the same level of exertion, it is possible that exercise intensity may have been slightly lower or even higher. That is one limitation of our study, but we do not expect a large error from that point of view.

From another point of view, we can ask ourselves if simulated boxing exercises in the LAB realistically corresponds to natural GYM boxing exercises. Based on the higher sparring values of HR and [LA] observed in the GYM as compared to the LAB (Table 2), we may expect a slight underestimation of V̇o2 values observed for sparring in the LAB. The fact that lower HRs were observed in the LAB for punching the bag at 60 and 120 b·min−1 (67.5 ± 3.5 and 74.8 ± 5.9%HRmax, respectively, Figure 4) compared to free punching bag in the GYM at a frequency of 70.6 ± 22.6 b·min−1 (86.9 ± 5.7%HRmax, Figure 4 and Table 2), also indicates that power of the punches or the foot work rather than the punching frequency may explain the higher intensity of the GYM boxing exercises compared to the LAB exercises. The same conclusions could be reached from RPE measures. Because we do not have any LAB-GYM comparison data for the pad work exercise, it is difficult to say if the observed energy cost is also underestimated for this activity, but there is no reason to believe so.

We did our measurements using 2-minute rounds of boxing exercises, but actual ruling sets the round duration to 3-minute for “elite amateur” and “professional” male boxers. The 2-minute round rule is still valid for other boxers. In any event and at the same pace of exercise, the energy cost should not be affected by the length of the round because metabolic steady state is reached in about 2 minutes (4) as seen by continuous V̇o2 recording. On the other hand, it is not excluded that boxers could decrease their metabolic rate because of fatigue, but this would need additional study to demonstrate it. From another perspective, our measurements only reflect the energy cost at the end of the last or third 2-minute round. In other words, the exercises may not be steady state tasks and the values observed may not represent the average cost for those exercises, but some kind of a peak or an end value of the third round of exercise assuming that V̇o2 increases from round to round and from the first to second minutes of each round.

Now that we better understand the methodological limitations, it seems easier to discuss the energy cost of boxing exercises. From our LAB measures, we can say that at least 43.4 ± 5.9, 41.1 ± 5.1 ml·kg−1·min−1 are required for sparring and pad work, respectively (Figure 4). These values correspond to 69.7 ± 8.0 and 66.1 ± 8.0% of treadmill V̇o2peak. Three months separated the V̇o2peak test from V̇o2 measurements during the boxing exercises. A possibility exists that these values changed during that time lapse. However, for already well trained subjects such as our boxers (62.2 ± 4.1 ml·kg−1·min−1), V̇o2peak improvement is usually <2 ml·kg−1·min−1 over a 3-month period even with very intense training (4). In summary, with V̇o2 requirements just >40 ml·kg−1·min−1, boxing may not be a typical aerobic sport such as middle or long distance running with known V̇o2max values in the 70.0-85.0 ml·kg−1·min−1 range, but a minimal of aerobic fitness may help to maintain first round pace till the end of the fight.

Our sparring average V̇o2 value is higher (see Introduction) than the ones reported by Seliger (24,25), Durnin (11), Ainsworth et al. (2,3), and Ostyn and S'Jongers (21) possibly because of the limitations of their methodological equipment (24,25) or computation (2,3,11,21). However, using treadmill HR/V̇o2 regression, Chatterjee et al. (7-9) reported similar V̇o2 values for women sparring with an increase of 40.3 ± 7.0 to 46.6 ± 6.6 ml·kg−1·min−1 from round 1 to round 3. That is also expected from the increased %HRmax observed from round 1 to round 3 by our own subjects while sparring in the GYM. It is difficult to say which values represent the true values. Morita et al. (18) found that treadmill HR/V̇o2 regression yielded biased V̇o2 estimates as compared to boxing HR/V̇o2 regression or simply V̇o2 measured during shadow, pad work, heavy bag, and punching ball exercises. The lower values reported by Seliger (24,25) may be because of hindered movement while carrying meteorological balloons during sparring and punching bag. Furthermore, with a collecting gas valve in the face, it is difficult to study “true” sparring as we did in our study. The studies that involved “true” sparring, were estimating V̇o2 from HR/V̇o2 regression, a procedure that is not very accurate particularly if the regression was obtained on the treadmill (18). On the other hand, measurements of HR and [LA] were easily obtained during “true” sparring, and values of ∼180 b·min−1 (7-9,12,18) and 9-10 mmol·L−1 (9,12) were reported, which is similar to our “true” sparring HR and [LA] values and thus confirms the intensity of sparring in the GYM. However, lower HR values of ∼170 b·min1 were reported by Seliger (24,25), but their boxers wore a mouth piece connected with tubing to meteorological balloons on their back. Thus, we feel that our values represent a good estimate of the average cost of “true” sparring and pad work. In any event, the average values may not be representative for all boxers. With an average around 40 ml·kg−1·min−1 and an SD around 5 ml·kg−1·min−1 for sparring and pad work, it means that based on a normal distribution, around 32% of boxers are either <35 or >45 ml·kg−1·min−1 when doing those activities. Thus, it is important to exercise caution before generalizing average values to all individuals. Some of these differences may be because of the weight or the fitness level of the boxers. Typically, compared to light boxers, heavy boxers tend to move slowly in the ring. In our study, there was a tendency for a lower V̇o2 cost in heavier boxers (Figure 5), a tendency that might be significant with a larger number of subjects. Similarly, the boxers with better V̇o2peak tend to have a higher V̇o2 cost of sparring or to invest themselves more during sparring (Figure 5). Although logical and interesting, this needs to be confirmed with a larger sample.

Now let us discuss the energy cost of punching the bag. Our LAB measurements at frequencies of 60, 120, and 180 b·min−1, yielded V̇o2 costs of 24.7 ± 6.1, 30.4 ± 5.8, and 38.3 ± 6.5 ml·kg−1·min−1, respectively. For free punching bag in the GYM at 70.6 ± 22.6 b·min−1, using LAB V̇o2 data at 60 b·min−1 grossly underestimates the metabolic cost because the punching intensity was probably much larger. However, because %HRmax and RPE are similar to the ones observed during punching the bag at 180 b·min−1 in the LAB, the V̇o2 cost of free punching bag exercise in the GYM is probably closer to the LAB cost at 180 b·min−1 (38.3 ± 6.5 ml·kg−1·min−1, Figure 4). This value is a bit lower than the values we obtained for sparring and pad work, which is consistent with the results of Seliger (24,25).

Thus, in addition to foot work, higher power of the punches in the GYM may explain these higher RPE and HR values. In their study, Kravitz et al. (13) also measured the V̇o2 cost of punching at different frequencies. Keeping in mind that they have used recreational fitness boxers hitting a SLAMMAN instead of a regular suspended punching bag, they have reported values around 27.0-30.0 ml·kg−1·min−1 for punching frequencies between 60 and 120 b·min−1. This is almost identical to our values but probably underestimates the real cost of punching in the GYM with elite boxers for the reasons mentioned above. Furthermore, O'Driscoll et al. (20) reported values of 31.5 ± 6.9 ml·kg−1·min−1 for a punching frequency of 134 b·min−1, which is in between our values at 120 and 180 b·min−1, and Adams et al. (1) reported values of 22.9 ± 10.0 ml·kg−1·min−1, which is lower than our values obtained for the same punching frequency of 120 b·min−1. However, other studies (2,3,19,24,25) reported values between 18.9 and 27.0 ml·kg−1·min−1 for free training of boxing on the punching bag, which would be lower than our estimate of 38.3 ± 6.5 ml·kg−1·min−1 (see above).

These discrepancies in the punching bag cost may be because of different methodologies such as the V̇o2 measures, the punching power, or frequency and the integration or not of footwork. However, Morita et al. (18) reported values of 52.5 ± 7.1 ml·kg−1·min−1, which this time is much larger than our 38.3 ± 6.5 ml·kg−1·min−1 estimate for free punching on the bag. We do not have any explanation for this, but it is interesting to note that their values were slightly higher than the V̇o2max of their subjects. Thus, our 38.3 ± 6.5 ml·kg−1·min−1 estimate for 3 rounds of free punching bag corresponds to 61.7 ± 10.3%V̇o2peak and is a much lower value than ∼100%V̇o2peak. On the other hand, it is quite feasible to reach V̇o2max on the punching bag if a boxer punches as fast and as hard as possible and gets exhausted at the end of a single round of punching bag. Interestingly, one study reports that between 2 boxers with similar HR response, the one with the largest punching power has the lowest punching frequency (17). The energy cost of punching is thus related to many conditions, and it is very difficult to attribute a single value for this particular exercise.

Back to Top | Article Outline

Practical Applications

Even though boxing requires a combination of technical, tactical, mental, and physical skills, this study indicates that aerobic fitness is certainly one of the important physical qualities to consider as seen by a V̇o2peak of 62.2 ± 4.1 ml·kg−1·min−1 and a relative intensity of ∼70%V̇o2peak for the most demanding boxing exercises such as sparring and pad work. This also gives an indication for the minimal level of aerobic training stimuli required for the boxers because the intensity could be higher in real competition. However, aerobic fitness is probably more important as the duration or the number of rounds increases. From another point of view, punching frequencies of around 35 and 60 b·min−1 in sparring and pad work, respectively, were observed in this study and gives an idea of the frequency that should be used when training those abilities. To be more specific, our study is the only one to measure the V̇o2 cost of “true” sparring and those values indicate (a) the importance of V̇o2max in the training program of the boxer and (b) the minimal intensity at which training loads should be set for aerobic training. This study also reports punching frequency data for sparring, pad work and free training on the punching bag that could be used as training guides for boxers.

Back to Top | Article Outline

Acknowledgments

The authors would like to thank Mr. Daniel Trépanier, Technical Coordinator at Boxing Canada, who facilitated our work and also all the boxers that participated in this study and the subjects that participated in the validation process (Annex 1). The authors have no undisclosed professional relationships with companies or manufacturers that would benefit from the results of this study. The results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association.

Back to Top | Article Outline

References

1. Adams, KJ, Allen, NB, Schumm, JE, and Swank, AM. Oxygen cost of boxing exercise utilizing a heavy bag [Abstract]. Med Sci Sports Exerc 29: S187, 1997.
2. Ainsworth, BE, Haskell, WL, Leon, AS, Jacobs, DR, Montoye, HJ, Sallis, JF, and Paffenbarger, RS. Compendium of physical activities: Classification of energy costs of human physical activities. Med Sci Sports Exerc 25: 71-80, 1993.
3. Ainsworth, BE, Haskell, WL, Whitt, MC, Irwin, ML, Swartz, AM, Strath, SJ, O'Brien, WL, Bassett, DR, Schmitz, KH, Emplaincourt, PO, Jacobs, DR, and Leon, AS. Compendium of physical activities: An update of activity codes and MET intensities. Med Sci Sports Exerc 32: S498-S504, 2000.
4. Astrand, PO and Rodahl, K. Textbook of Work Physiology (Figures 9.3 and 12.13).New York, NY: McGraw-Hill, 1977.
5. Babineau, C, Léger, L, Long, A, and Bosquet, L. Variability of maximal oxygen consumption measurements in various metabolic systems. J Strength Cond Res 13: 318-324, 1999.
6. Borg, GA. Perceived exertion: A note on history and methods. Med Sci Sports 5: 90-93, 1973.
7. Chatterjee, P. Energy expenditure of women boxers in India. Nigerian J Health Biomed Sci 5: 17-20, 2006.
8. Chatterjee, P, Banerjee, AK, and Majumdar, P. Energy expenditure in women boxing. Kathmandu Univ Med J (KUMJ) 4: 319-323, 2006.
9. Chatterjee, P, Banerjee, AK, Majumdar, P, and Chatterjee, P. Oxygen consumption, heart rate and blood lactate response during sparring on Indian women boxers. Int J Appl Sports Sci 17: 9-16, 2005.
10. Crisafulli, A, Vitelli, S, Cappai, I, Milia, R, Tocco, F, Melis, F, and Concu, A. Physiological responses and energy cost during a simulation of a Muay Thai boxing match. Appl Physiol Nutr Metab 34: 143-150, 2009.
11. Durnin, JV. The energy cost of exercise. Proc Nutr Soc 44: 273-282, 1985.
12. Ghosh, AK, Goswami, A, and Ahuja, A. Heart rate & blood lactate response in amateur competitive boxing. Indian J Med Res 102: 179-183, 1995.
13. Kravitz, L, Greene, L, Burkett, Z, and Wongsathikun, J. Cardiovascular response to punching tempo. J Strength Cond Res 17: 104-108, 2003.
14. Lane, AM.Introduction to the special issue on combat sport (Combat Sports Special Issue). J Sports Sci Med 5: i-iii, 2006.
15. Léger, L and Boucher, R. An indirect continuous running multistage field test: The Université de Montréal Track Test. Can J Appl Sport Sci 5: 77-84, 1980.
16. Léger, L and Thivierge, M. Heart rate monitors: Validity, stability, and functionality. Physician Sportsmed 16: 143-151, 1988.
17. Martos, E and Jako, P. Exercise physiology in boxing. Hungarian Rev Sports Med 39: 197-220, 1998.
18. Morita, Y, Seiiti, N, and Takashi, I. Energy expenditure during boxing training (In Japanese). Nippon Sport Sci Univ J 13: 59-66, 1984.
19. Moy, K, Scragg, R, McLean, G, and Carr, H. Metabolic equivalent (MET) intensities of culturally-specific physical activities performed by New Zealanders. N Z Med J 119: U2000, 2006.
20. O'Driscoll, E, Steele, J, Perez, H, Yreys, S, Snowkroft, N, and Locasio, F. The metabolic cost of two trials of boxing exercise utilizing a heavy bag [Abstract]. Med Sci Sports Exerc 31: S158, 1999.
21. Ostyn, M, and S'Jongers, JJ. Cout métabolique des sports de combat (In French). Méd Sport 54: 3, 1980.
22. Pyne, DB, Boston, T, Martin, DT, and Logan, A. Evaluation of the Lactate Pro blood lactate analyser. Eur J Appl Physiol 82: 112-116, 2000.
23. Saunders, AC, Feldman, HA, Correia, CE, and Weinstein, DA. Clinical evaluation of a portable lactate meter in type I glycogen storage disease. J Inherit Metab Dis 28: 695-701, 2005.
24. Seliger, V. Energeticky metabolismus u vybranych telesnych cviceni (In Czech). Praha, Czechoslovakia: Karlova Universita, 1967.
25. Seliger, V. Energy metabolism in selected physical exercises. Int Z Angew Physiol 25: 104-120, 1968.
Back to Top | Article Outline

Appendix 1. Validation of Postexercise Measurements to Estimate Exercise V̇o2

Back to Top | Article Outline

Introduction

In some exercises such as sparring, while the boxer may receive punches to the head and upper body, it is not possible to wear a collecting gas valve in the face or a portable V̇o2 system on the chest to collect expired gasses. V̇o2 could be estimated from individual HR/V̇o2 regression, but those estimations are not accurate, particularly if the regression is obtained from another type of exercise (e.g., Treadmill vs. Sparring) (4). Because none of these 2 traditional methods permit proper measurements of true sparring V̇o2, we developed a new method to do so, that is, a method that measures V̇o2 of true sparring with actual punches to the face and chest. With this method, the subjects are connected to the metabolic system right at the end of an exercise bout done at a relatively constant intensity that lasts at least 2 minutes to reach steady V̇o2 (e.g., third 2-minute round of sparring) and are instructed to keep moving their legs and arms to maintain their metabolic activity at the same level as it was before the connection. When the metabolic system reaches equilibrium (≥30 seconds to wash the system), V̇o2 is recorded as the exercise steady state or average value. The general purpose of this study was to validate this approach to measure V̇o2 in field conditions.

Back to Top | Article Outline

Methods

Subjects

Nine kinesiology students (age = 27.0 ± 13.4 years, height = 174.0 ± 7.9 cm, and weight = 76.2 ± 16.8 kg) volunteered to participate in this validation process.

Back to Top | Article Outline

Factors to Consider

When the method is used in the field, gases are collected only at the end of a task done at a relatively constant intensity, for example, after the third 2-minute round of sparring. To validate these measures, a criterion measure was required, which was V̇o2 during the exercise itself. Thus, in the validation process, V̇o2 was measured both during the exercise and after the exercise. Furthermore, for the postexercise measures in field conditions, we asked the subjects to continue to move at the same intensity immediately after being connected to the metabolic system to avoid the recovery process. In the validation study, treadmill exercise was used to be able to measure V̇o2 both during the exercise itself and after the exercise.

One more aspect needs to be considered. When used in field conditions, the metabolic system may often be disconnected from the subjects for 10-16-minute periods before reconnecting the subjects for the postexercise measurements. Therefore, during the validation process, the exercise steady state criterion measures were taken ∼15-30 minutes before the postexercise measures. This happened many times after the initial calibration to validate the method at different intensities. We thus developed an exercise pattern that takes into consideration all these aspects including the calibration that was made up to 85 minutes before numerous connections and disconnections to the metabolic system.

The validation study was thus designed to answer the following questions: (a) How are V̇o2 values affected by many 10-16 minute disconnections from the metabolic system after initial calibration? (b) Can the gas analyzing system stay calibrated for up to 85 minutes? (c) How long does it take to wash the metabolic system from zero input and to reach V̇o2 equilibrium? (d) Are 3 2-minute rounds of running exercise with 1-minute rest in between enough for the subjects to reach a steady state V̇o2? (e) How does exercise intensity affect the accuracy of the postexercise measures? (f) Are the postexercise values significantly different from the criterion values?

Back to Top | Article Outline

Validation Protocol

Each subject ran on the treadmill at 3 different intensities for at least 2 minutes continuously to reach metabolic steady state. Intensities were chosen to cover subjects capacity and a wide range of energy costs. Each intensity was done twice: first time to obtain criterion V̇o2 values and second time to see if the postexercise V̇o2 values were the same as the criterion values. For the postexercise measures of the validation process, the subjects briefly stopped running to connect themselves to the gas analyzing system and hopped back on the treadmill to continue running at the same speed to avoid recovery process for the collection periods. To summarize, the subjects (a) ran on the treadmill while their criterion exercise V̇o2 were measured, (b) took a 8-10-minute rest, (c) got back on the treadmill for 4-8 minutes without being connected to the system to mimic various field conditions, (d) stopped running for 10-15 seconds to put a nose clip and to connect themselves to the metabolic system, and (e) got back on the treadmill for 1.75-2 minutes to collect the postexercise measures. The protocol is illustrated in Figure 1. The black line is the predicted V̇o2 costs of treadmill running using Léger's equation (2,3):

The predicted cost is thus proportional to the speed pattern of the protocol and illustrates the actual exercise workload and also enables a comparison of the V̇o2 values obtained with our system and the values reported in the literature. Each subject had a different set of speeds. The individual kinetic of V̇o2 measured during and after exercise at 3 different speeds for 1 subject is also illustrated in Figure 1 (gray line) and enables comparison between exercise and postexercise measures and between literature predicted costs for 3 different intensities. Similar curves were obtained for all the subjects.

Back to Top | Article Outline

Statistical Analyses

“Postexercise V̇o2 measures,” “criterion exercise V̇o2 measures,” and “literature predicted V̇o2 values” (equation 1) were compared for each running intensity using a 2-way ANOVA for repeated measures and Tukey a posteriori tests. A linear regression and a scatterplot were also done to compare “postexercise V̇o2 measures” and “criterion exercise V̇o2 measures.” Unless otherwise stated, all reported differences are significant at the p ≤ 0.05 level.

Back to Top | Article Outline

Results

Figure 1 is a typical representation of V̇o2 predictions and V̇o2 measurements during and after exercise at 3 different intensities during the exercise protocol for 1 subject. Similar patterns were obtained for all the subjects. Figure 2 summarizes the results of the whole group and illustrates similar values between criterion exercise, postexercise, and literature predicted V̇o2 values. This is observed for the whole range of investigated metabolic levels. Not only were the values obtained immediately after reconnecting the subjects to the metabolic system similar to the criterion exercise values, but these 2 values were also similar to reported values in the literature indicating no important systematic differences between these V̇o2 values. Furthermore, random variation between V̇o2 measured during running or immediately after reconnecting the subjects to the metabolic system were very small as seen by the high correlation and the small SEE (r = 0.96, SEE = 1.6 ml·kg−1·min−1, Figure 3) between these 2 variables.

Figure 1

Figure 1

Figure 2

Figure 2

Figure 3

Figure 3

A closer look at the individual curves with extended abscissas (Figure 4) illustrates that the system equilibrium is reached between 30 and 75 seconds (second to fifth 15-second sample) after reconnection. Similar results were systematically obtained for all the subjects, indicating that only a short delay or a short time-sampling collection is required to obtain proper V̇o2 values after reconnecting the subject.

Figure 4

Figure 4

Back to Top | Article Outline

Discussion

Our metabolic system yields similar exercise and postexercise V̇o2 values for the same workload after being disconnected many times for 10-16 minute periods. Thus, it is not a problem to measure V̇o2 many times after different tasks of various intensities between 8 and 13 km·h−1 or ∼25 to 45 ml·kg−1·min−1 while the metabolic system is disconnected in between and calibrated up to 85 minutes before. Furthermore, we can see that the metabolic system adjusts itself within 30-45 seconds after reconnecting the subjects. Finally, we can see that 3 2-minute periods of exercise are enough to achieve metabolic steady state and that attained V̇o2 values truly reflect the energy cost of the activity being measured, which is in conformity with the classical work of Astrand and Rodahl (1). Of course, we used well controlled treadmill running instead of subjective arm and leg exercise during the postexercise period of the validation process. Although we cannot be totally sure, the error should not be much different when the level of activity is subjectively maintained after reconnecting the subject to the metabolic system when the method is used with other types of activities where it is not possible to wear a respiratory face mask (e.g., sparring). Therefore, it is difficult to say if the limitations of the present approach are smaller or larger than the limitations of approaches used in previous sparring studies. However, the results we obtained with this new approach certainly bring new insight in the V̇o2 cost of true sparring. Thus, the results of this validation process confirm the potential of this new approach in field conditions for our metabolic system at least.

These results were obtained with the same metabolic system used with the boxers in the experimental study. With other models or brands of metabolic system, however, we recommend to check as we did the stability of the V̇o2 measures for many disconnections and connections after the initial calibration and to check the time required by the metabolic system to reach equilibrium values after reconnecting the subjects from a zero input signal. Furthermore, before measuring the energy cost of a new activity with this approach, it is also recommended to check if the metabolic system used yields treadmill or cycling V̇o2 values that are similar to accepted literature values.

Back to Top | Article Outline

Conclusion and Practical Applications

That method was developed to determine the V̇o2 cost of 2-minute steady-state rounds of sparring and pad work in boxing but could also be useful in other field situations (a) where the use of a portable V̇o2 system seriously hinders the motion pattern or (b) where such a portable system is not available. It is however important to make sure that the subjects are relatively in a steady state before connecting the subjects to the metabolic system and that the subjects continue to exercise at the same level during the measurement phase.

Back to Top | Article Outline

References

1. Astrand, PO and Rodahl, K. Textbook of Work Physiology (Figure 9.3, p. 296).New York, NY: McGraw-Hill, 1977.
    2. Hall, C, Figueroa, A, Fernhall, B, and Kanaley, JA. Energy expenditure of walking and running: Comparison with prediction equations. Med Sci Sports Exerc 36: 2128-2134, 2004.
    3. Léger, L and Mercier, D. Gross energy cost of horizontal treadmill and track running. Sports Med 1: 270-277, 1984.
    4. Morita, Y, Seiiti, N, and Takashi, I. Energy expenditure during boxing training (in Japanese). Nippon Sport Sci Univ J 13: 59-66, 1984.
      Keywords:

      energy cost; physiological demand; heart rate; sparring; pad work; punching bag

      Copyright © 2011 by the National Strength & Conditioning Association.