Journal Logo

APPLIED SCIENCES

Energy Cost and Mechanical Work of Walking during Load Carriage in Soldiers

GRENIER, JORDANE G.1,2; PEYROT, NICOLAS3; CASTELLS, JOSIANE1,4; OULLION, ROGER4; MESSONNIER, LAURENT5; MORIN, JEAN-BENOIT1

Author Information
Medicine & Science in Sports & Exercise: June 2012 - Volume 44 - Issue 6 - p 1131-1140
doi: 10.1249/MSS.0b013e3182456057

Abstract

For occupational (e.g., handling, school), recreational (e.g., hiking), or pathological (e.g., obesity) reasons, the human body may be led to carry external or “additional” load during locomotion. This restricting and sometimes unnatural situation may have consequences on individuals’ musculoskeletal and metabolic systems (25). Therefore, studies have long been conducted to evaluate and understand the physiological and biomechanical effects of load carriage (33), taking into account load position (e.g., proximal vs distal placement, see Browning et al. [11]) and the various modes of carrying (e.g., manual lifting, backpack carriage, or other modes; see Datta and Ramanathan [16] and Legg [28]). Furthermore, studies have investigated specific populations, such as African women (31), Nepalese porters (3), or obese patients (10,34), and specific activities such as trekking (21) or military exercise (26). As a whole, studies revealed that load carrying capabilities could substantially vary with gender, familiarization, training, load distribution, body characteristics and composition, total mass carried, or use of ergonomic systems (23,29).

In the military context (missions or training), soldiers carry their own equipments, protections, and supplies during locomotion. The total mass carried often exceeds 30%–40% of their body mass (BM) and can even reach 70% of BM (5,27), which represents high- to extreme-load carrying situations. Moreover, soldiers are often required to carry their pack over long durations and therefore develop high abilities and endurance for load carriage. Finally, military loads (i.e., equipments) are numerous and conveniently organized around soldiers’ body to satisfy operational exigencies, such as (i) quickly accede to battle necessities (munitions, weapons, and electronics located onto the front of the trunk), (ii) benefit from a large protection against enemies and environments (helmet, joint protections, and heavy boots located in distal regions of the body), or (iii) bring supplies and supplementary equipments (located in an additional backpack), which represents a distinct and complex load distribution compared to backpack carriage only, for instance.

Considering these characteristics, we thought interesting to investigate the specific effects of military load carriage on soldiers’ walking energetics and mechanics.

Concerning the energetics, the physiological effects of military load carriage have been studied for almost a century (26), and many studies focused on the metabolic responses during walking. These studies tended toward the conclusion that absolute energy expenditure increases linearly as a function of the military load carried (35,36) up to 70% of soldiers’ lean BM (5). Interestingly, this linear load–energy expenditure relationship also existed at the most commonly used military walking speeds (32). Complementary studies have also shown that both optimal load–speed combination (39) and loads distribution on soldiers’ body and within their backpack could reduce the physical stress imposed to soldiers before battle (26,28,40).

Overall, the nonmilitary literature showed comparable results. However, to study walking energetics and load carriage with more precision, Griffin et al. (22) proposed to differentiate the metabolic energy used by the muscles to walk and carry the load from the residual non–locomotor-related metabolism. To this aim, they calculated a net (locomotor) metabolic rate by subtracting the standing (nonlocomotor) metabolic rate from the gross (total) metabolic rate of walking. Thanks to this gross and net distinction and when normalizing parameters to their unloaded walking values (i.e., ratio of metabolic rates during loaded to unloaded walking vs ratio of total mass to BM), these authors (22) showed that (i) the net metabolic rate increased greater than proportionally to the load carried at any given speed and (ii) the gross metabolic rate increased less than proportionally to the load carried at walking speeds below 1.5 m·s−1 and in direct proportion or greater than proportionally at 1.5 m·s−1 or higher speeds, respectively. These results, obtained for additional loads up to 30% of BM carried symmetrically around the waist, also showed that the gross metabolic parameters underestimates the specific energy used by the muscles to carry the load, especially at slow walking speeds (22).

In light of these results, a first part of the present study focused on the effects of military load carriage on the metabolic cost of walking. To differentiate the locomotor and nonlocomotor metabolism during military-type walking, the gross and net calculation approach was used.

Concerning the mechanics of walking during military load carriage, few studies have investigated this issue, and when they did, they mainly focused on spatiotemporal and kinematic parameters (1,30). Nevertheless, these military-related studies indicated that loaded walking presented fewer deviations from normal walking in the following conditions: (i) with carriage on the back only, when heavy items are positioned high vertically in the backpack and close to the back of the carrier (26,40); (ii) with carriage on the whole body, when the center of gravity of the load is close to the body center of mass (COM) in the three directions of space (25,26,28), and when loads are equilibrated around the trunk to promote anteroposterior and lateral stability (24,28,29); (iii) in both conditions, when a large part of the weight is redistributed toward the strong muscles of the legs via a hip belt, for example (25,26,28); unfortunately, exigencies of military operations do not always allow to respect these ways of optimal load carriage (26). In addition, Birrell et al. (7,8) investigated military walking kinetics through changes in ground reaction forces (GRF) while carrying various configurations of an actual military equipment of 8–40 kg. They showed that vertical and anteroposterior GRF increased linearly in proportion to the amount of additional loads carried (8). However, they did not report other kinetic results or associated metabolic data, and we are not aware of any study that has reported kinetic data such as mechanical work, inverted pendulum recovery of mechanical energy, or locomotor efficiency during walking with military load carriage.

Mechanical work represents the muscular actions performed to lift and accelerate the COM in the surrounding space (external work), transfer the COM from one leg to the other leg (internal work done during the double contact phase), make the body rigid, and swing the limbs relatively to the COM (internal work) (for details, see Bastien et al. [2], Cavagna et al. [14], and Donelan et al. [18]). As a consequence, the net metabolic rate varies according to the mechanical work done by the muscles during walking. However, studies comparing metabolic and mechanical energies during walking showed that an important part of the total mechanical work is done by passive mechanisms of exchange between kinetic and gravitational potential energy (13). Concerning the energy related to COM displacement, this mechanism of exchange has been described as an inverted pendulum, the participation of which can be quantified through the parameter of mechanical recovery (15,37).

In line with these data, a second part of the present study centered on the mechanical work parameters of walking during military load carriage.

The aim of the present study was therefore to describe the effects of military intermediate- and high-load carriage (∼22 kg corresponding to ∼27% of subjects’ BM and ∼38 kg corresponding to ∼46% of subjects’ BM, respectively) on the energy cost and mechanical work of walking in experienced infantrymen. To this aim, an unloaded walking condition was used as reference, and results were discussed with respect to those reported in backpack and experimental waist pack carriage conditions to better interpret the specific effects of military load carriage. We hypothesized that (i) military intermediate- and high-load carriage would affect the energy cost of walking and the associated mechanical work and that (ii) the changes observed would be greater than those reported while carrying loads symmetrically around the waist and close to the COM (22), which appears in the literature as the less metabolically disturbing method of load carriage on the whole body referential (25,26,28).

METHODS

Subjects

Ten males (mean ± SD; age = 38.9 ± 8.9 yr, height = 177 ± 5 cm, leg length = 90.9 ± 3.6 cm, BM = 82.9 ± 9.3 kg, BMI = 26.7 ± 2.2 kg·m−2, body fat percentage = 19.4% ± 3.1%, HRmax = 190 ± 17 bpm, V˙O2max = 3.53 ± 0.36 L·min−1) volunteered to participate in this study. They were all involved in regular physical activities (5.45 ± 2.75 h·wk−1), not presenting recent muscular, joint, or bone conditions or receiving any medication that could interfere with their walking pattern or influence their energetic metabolism. Subjects were recently retired infantrymen (mainly from the French Foreign Legion) with a career of 14.1 ± 8.3 yr and had an extensive high-load carriage experience. The main self-reported career-related injuries were as follows: (i) repeated ankle and knee sprains suffered during training or actual road marches and battles (n = 6), (ii) tibia/fibula and wrist fractures caused by falls during obstacle courses tests and military simulations (n = 3), and (iii) lumbar and cervical pains mainly caused by repeated load carriage efforts (n = 5). Written informed consent was obtained from the subjects, and the study was conducted according to the Declaration of Helsinki. The protocol was approved by the local ethics committee (Comité de Protection des Personnes, Sud-Est 1, France) and registered at http://clinicaltrial.gov (reference no. NCT01127191).

Protocol overview

Subjects were included 1 wk before the beginning of the specific research protocol. Inclusion sessions consisted of (i) a complete medical examination with anthropometric data collection and fat mass estimation (20), (ii) a complete familiarization with the different devices used in the experimentation, especially the instrumented treadmill, and (iii) a standardized incremental maximal aerobic test.

The protocol consisted of two identical laboratory sessions separated by 2–4 wk. Variables computed during each of these two sessions were averaged before statistical analyses, to increase the representativeness of the values obtained. No significant difference between sessions was shown by t-tests and by mean variation tests performed for each experimental condition.

During each laboratory session, expired gases were first collected during 10 min of unloaded standing, according to the method proposed by Griffin et al. (22). Then, subjects performed three 3-min level walking trials at 4 km·h−1 on an instrumented treadmill, during which walking energetics and mechanics were assessed (see details below). The 4-km·h−1 (1.11 m·s−1) speed was chosen for its economical character in normal adult locomotion (4,14) and its consistency with the average walking speed currently used during military missions or experimentations (32). Three conditions were tested: a sportswear condition taken as reference (SP, mass ≤ 1 kg) and two configurations of a military equipment, namely, battle equipment (BT, 22.4 ± 1.1 kg) and road march equipment (RM, 37.9 ± 1.4 kg). Rifle carriage was voluntarily excluded to possibly compare our results with nonmilitary studies because Birrell and Haslam (7) showed that walking kinetics were altered by a limitation of the arm swing during rifle carriage. These three conditions were assigned in a randomized and counterbalanced order, and trials were separated by 5–10 min during which subjects rested and changed their equipment.

Equipment characteristics

The new French infantry combat system (FELIN, Sagem, France) was used in this study in two different configurations. Its two notable characteristics are as follows: (i) from a biomechanical viewpoint, a distribution of the loads balanced around the body; and (ii) from a military viewpoint, it will soon be used in actual military missions. Thus, the data reported here are representative of a typical infantry system of the coming years. Equipments characteristics are detailed in Table 1 and illustrated in a supplemental figure (Figure showing pictures of a typical subject in the three conditions of equipment tested in the study, Supplemental Digital Content 1, https://links.lww.com/MSS/A145). The military configurations tested were designed to meet the common necessities of a 24-h patrol and recognizance mission (which may include battles and road marches). Each participant was familiarized with all of the equipments tested because they took part in several operational tests and simulations with the device (such as military obstacle course test, marches, or simulated missions) before the experiment.

T1-21
TABLE 1:
Main characteristics of the equipments used in the study.

Walking energetics

Energetic data were obtained by indirect calorimetry. Expired gases were collected during the last 30 s of each 3-min walking trial to allow the stabilization of the aerobic metabolism before gas collection. Subjects breathed through a two-way nonrebreathing valve (series 2700; Hans Rudolph, Kansas City, MO) connected to a three-way stopcock that stemmed into a 100-L Douglas bag. The volume of the expired gas was measured by means of a Tissot spirometer (Gymrol, Roche-la-Molière, France), and fractions of expired gases were determined with a paramagnetic O2 analyzer (cell 1155B; Servomex, Crowborough, England) and an infrared CO2 analyzer (Normocap Datex, Helsinki, Finland). These analyzers were calibrated with mixed gases whose composition was determined using Scholander’s method (38).

Oxygen consumption (L·min−1) and carbon dioxide production (L·min−1) were first calculated. Unloaded standing metabolic rate and gross metabolic rate of walking (W) were then determined from the steady state and using the following standard equation proposed by Brockway (9) because the RER was lower than 1.0:

where

(mL·s−1) represent mean V˙CO2, respectively.

In the present study, we subtracted the metabolic rate measured during unloaded standing from all walking values to calculate the net metabolic rate (W), as proposed by Griffin et al. (22). Indeed, these authors showed that metabolic rate while standing did not significantly change with loads of up to 50% of BM. Gross and net metabolic rates (W) were divided by walking speed (m·s−1) to obtain the gross and net energy costs of walking (CW (J·m−1); see di Prampero [17]). Gross and net CW were also divided by the total mass in motion (TM, kg) to obtain mass-relative gross and net CW (CW,TM, J·kg−1·m−1).

Walking mechanics

Walking mechanics was analyzed using an instrumented three-dimensional force treadmill (ADAL; HEF Tecmachine, Andrézieux-Bouthéon, France) consisting of two identical left–right frames and belts separated by a 7-mm gap, one for the left foot GRF and one for the right foot GRF (for details and validation, see Belli et al. [6]). Parameters were recorded over 20 s, 1.5 min after the beginning of each trial to ensure the stabilization of the gait pattern and avoid disturbances from the metabolic measurements. All data were sampled at 200 Hz and low-pass filtered at 30 Hz.

Mechanical analyses were performed over five consecutive strides, a stride being defined as the period between two consecutive right heel strikes and identified from vertical GRF signals. Mechanical parameters were computed for each stride and then averaged to describe a mean typical stride with standard deviation.

Spatiotemporal parameters of walking were calculated from vertical GRF signal. Stride duration (s) was delimited by two consecutive right heel strikes and stance duration (s) by right heel strike to consecutive right toe-off. A duty factor (%) was calculated as the ratio of stance duration to stride duration. Double-support duration (%) was expressed relatively to stride duration and corresponded to the phases during which GRF were measured simultaneously on the two frames of the treadmill. Single-support duration (%) was also expressed relatively to stride duration and corresponded to the phases during which GRF were measured on only one frame of the treadmill. Step frequency (Hz) and step length (m) were computed as follows:

Accelerations of the COM in the vertical (V), anteroposterior (A-P), and mediolateral (M-L) directions were computed from the corresponding GRF components, according to the law of dynamics. Velocities of the COM in the V, A-P, and M-L directions were then computed using a trapezoidal integration of the corresponding acceleration components, as proposed by Cavagna (12) and used for instance by Schepens et al. (37) and Peyrot et al. (34). The total instantaneous kinetic (Ek, J) and potential (Ep, J) energies of the COM were calculated as follows:

where TM is expressed in kilograms; v (m·s−1) is the resultant COM velocity determined from its V, A-P, and M-L components; g (m·s−2) is the gravitational constant; and h (m) is the vertical position of the COM relative to heel strike, calculated by integration of the vertical velocity. The total mechanical energy of the COM (Etot, J) was computed as the sum of the Ek and Ep curves over each stride. External mechanical work (Wext, J·m−1) was calculated as the sum of the positive increments in Etot (J; for each stride) (14) divided by stride length (m). Wext was also normalized by TM to obtain mass-relative Wext (Wext,TM, J·kg−1·m−1).

The inverted pendulum recovery of mechanical energy of the COM was calculated according to Schepens et al. (37), as follows:

where recovery is expressed in percentage, Wext in joules, and Wk and Wp (both in joules) are the sum of the positive increments in Ek and Ep, respectively.

During the double-contact phase of walking, positive work (Wint,dc, J) is done by the back leg pushing forward, whereas negative work is done by the front leg pushing backward. In this study, Wint,dc was calculated from the forces exerted by each lower limb on the ground measured separately, as proposed and detailed by Bastien et al. (2). Wint,dc was related to the double-contact lengths (i.e., walking speed multiplied by double-contact duration) to be expressed in joules per meter (J·m−1) before analysis (2). Wint,dc was also divided by TM to obtain mass-relative Wint,dc (Wint,dc,TM, J·kg−1·m−1).

Finally, locomotor efficiency was calculated as the ratio of Wext and Wint,dc to net CW, as follows:

where efficiency is expressed in percentage and Wext, Wint,dc, and net CW are expressed in joules per meter (J·m−1).

Statistical analyses

Normal distribution of the data was checked by the Shapiro–Wilk normality test and variance homogeneity between samples was tested by the F Snedecor test. All variables were normally distributed, and variances were homogeneous. Then, a series of one-way ANOVA with repeated measures was performed to test whether mechanical and metabolic parameters changed with equipment (load) carriage. When warranted, Newman–Keuls multiple-comparison post hoc tests were used to identify differences among conditions (SP, BT, and RM). The statistical significance was accepted at P < 0.05.

RESULTS

TM (subject’s BM + equipment) were 83.9 ± 9.3 kg in SP, 105 ± 10 kg in BT, and 121 ± 11 kg in RM. These TM corresponded to 101% ± 1%, 127% ± 2%, and 146% ± 4% of subjects’ BM in SP, BT, and RM conditions, respectively.

Walking Energetics

Gross CW was significantly altered by military load carriage (P < 0.0001). On average, gross CW was 22.3% ± 16.3% higher in BT versus SP, 12.3% ± 12.1% higher in RM versus BT, and 37.1% ± 22.9% higher in RM versus SP. Similarly, net CW was significantly modified by equipment carriage (P < 0.0001). Net CW was 42.5% ± 29.2% higher in BT versus SP, 20.8% ± 21.1% higher in RM versus BT, and 70.8% ± 43.0% higher in RM versus SP. On the other hand, when expressed per kilogram of total mass in motion, gross and net CW,TM were not significantly different between load conditions (Table 2 and Table showing data of oxygen consumption during walking in the three conditions of equipment tested in the study, Supplemental Digital Content 2, https://links.lww.com/MSS/A146).

T2-21
TABLE 2:
Metabolic costs of walking in SP, BT, or RM conditions.

Walking Mechanics

Spatiotemporal parameters

As shown in Table 3, military equipments carriage significantly altered the spatiotemporal parameters of subjects’ walking pattern (all P < 0.01), yet in a different manner between BT and RM. Indeed, all the spatiotemporal parameters of walking were significantly different in RM compared to BT or SP. Contrastingly, only stance duration, duty factor, and single- and double-support durations significantly changed during walking in BT compared to SP.

T3-21
TABLE 3:
Spatiotemporal parameters of walking in SP, BT, or RM conditions.

Kinetic parameters

Military load carriage induced changes in absolute and mass-relative mechanical works of walking (all P < 0.01; Table 4). On average, Wext was 44.6% ± 14.7% higher in BT versus SP, 12.5% ± 11.0% higher in RM versus BT, and 62.5% ± 21.0% higher in RM versus SP. Wext,TM was 15.4% ± 11.4% higher in BT versus SP, 13.1 ± 13.6% higher in RM versus SP and not significantly different between RM and BT. Furthermore, Wint,dc increased by 38.1% ± 10.9% from SP to BT, 32.5% ± 19.2% from BT to RM, and 82.5% ± 26.5% from SP to RM. Wint,dc,TM was 10.3% ± 9.2% higher in BT versus SP, 15.5% ± 15.1% higher in RM versus BT, and 26.8% ± 14.5% higher in RM versus SP. Finally, military load carriage did not significantly modify the inverted pendulum property of mechanical energy of the COM transfer (recovery) or the locomotor efficiency of walking (efficiency).

T4-21
TABLE 4:
Kinetic parameters of walking in SP, BT, or RM conditions.

DISCUSSION

The main results of the present study are that, compared to SP that represents “natural” (i.e., unloaded) walking, military BT and RM equipments carriage induced (i) significant alterations in the spatiotemporal walking pattern (all P < 0.01); (ii) significantly higher gross and net CW, Wext, and Wint,dc (all P < 0.0001); (iii) significantly higher Wext,TM and Wint,dc,TM (P < 0.01 and P < 0.0001, respectively) but did not induce change in locomotor efficiency or in the inverted pendulum mechanism of recovery.

Walking Energetics

First of all, the present data of walking energetics are consistent with those reported in the nonmilitary literature. For instance, Bastien et al. (4) showed mass-relative gross CW value of ∼3.6 J·kg−1·m−1 during backpack carriage at the speed of 1.1 m·s−1, for loads up to 75% of subjects’ BM. They also found mass-relative net CW of ∼2.1 and ∼2.4 J·kg−1·m−1 with loads of 30% and 45% of subjects’ BM, respectively (4). In the present study, gross CW,TM value was ∼3.0 J·kg−1·m−1, and although not statistically different between BT and RM conditions (∼27% and ∼46% of subjects’ BM, respectively), net CW,TM was ∼1.9 J·kg−1·m−1. Our slightly lower values might result from the subjects’ high abilities to perform load carriage efforts, the different modes of carriage, the different methods of metabolic measurement (Douglas bags vs portable gas analysis system, see Duffield et al. [19]) or a combination of these possibilities.

In the present study, absolute gross and net CW increased significantly with the carriage of military intermediate (BT, ∼22 kg) and high-load (RM, ∼38 kg) equipments (P < 0.0001; Table 2). Furthermore, when CW in BT and RM were expressed as a percent of CW in SP and plotted in function of the ratio of TM to BM (as suggested by Griffin et al. [22]), we observed the two following phenomena. First, net CW significantly increased by ∼43% from SP to BT (P < 0.001), whereas TM increased by ∼27% only. Further, from SP to RM conditions, net CW significantly increased by ∼71% (P < 0.001), whereas TM increased by ∼46% only. Therefore, net CW increased greater than proportionally to the military load carried (Fig. 1A), and as a result, net CW,TM values increased from SP to BT and from BT to RM conditions (Table 2). In other words, the metabolic cost specifically required (i.e., used by the muscles) to carry one unit of equipment mass was higher than the metabolic cost required to carry one unit of BM. Contrary to previous military studies that did not differentiate this specific locomotor (net) energy expenditure from the total energy expenditure (5,32,35), these results show how important are the muscular demands to carry the military loads. These results also indicate how the walking effort performed while approaching enemies may substantially involve soldiers’ metabolic resources before the (possible) subsequent battle phase, which is often risky and crucial for the success of a mission. Second, gross CW increased slightly less than proportionally to the military load carried (∼22% vs ∼27% in BT and ∼37% vs ∼46% in RM; Fig. 1A). Consequently, gross CW,TM values decreased from SP to BT and from BT to RM (Table 2). In other words, the total (gross) metabolic cost for carrying one unit of BM was higher than the total metabolic cost for carrying one unit of equipment mass during military walking. This percent gross CW increasing trend from SP to BT and RM (Fig. 1A) can be explained by the fact that the nonlocomotor (standing) metabolic rate, which is considered and computed as constant across load conditions (i.e., from SP to BT and to RM), represents a large part of the gross metabolic rate at moderate walking speeds (22). Therefore, at the moderate 4-km·h−1 speed used here, the specific metabolic effects of military load carriage (net) were appreciably underestimated by the gross CW, as in numerous military-related studies (5,32,35). However, the gross values remained important because they characterized the total metabolic demand of military intermediate- and high-load carriage.

F1-21
FIGURE 1:
Energy costs (gross and net CW; A), mechanical works (W ext, W int,dc, and W total = W ext + W int,dc; B) and total mass (TM) in motion on the treadmill during walking. The conditions presented are: military battle (BT) and road march (RM) with values expressed as a percentage of the unloaded walking (SP) condition values (mean ± SD). TM is also expressed as a percentage of BM. Bold lines indicate direct proportionality (line of identity).

It should be noted that the variability of subjects’ metabolic adaptations to military load carriage did not allow us to statistically establish the decrease in gross CW,TM and the increase in net CW,TM while carrying military BT and RM equipments [Table 2 and also the statistical tendency of increase in net O2,TM in the supplementary table (Table showing data of oxygen consumption during walking in the three conditions of equipment tested in the study, Supplemental Digital Content 2, https://links.lww.com/MSS/A146)]. That being said, the evolution of gross and net CW values in the function of TM to BM values (Fig. 1A) was very close to that reported by Griffin et al. (22) at the speed of 1 m·s−1 while carrying loads about the waist. Therefore, contrary to what was hypothesized, the apparently complex distribution of the military equipments around soldiers’ body (i.e., loads, up to ∼46% of BM) did not seem to induce greater metabolic demands compared to load carriage around the COM. These unexpected results also indicate that the approach aiming at organizing and counterbalancing loads around soldiers’ body, while respecting military battle needs, is energetically efficient and beneficial.

Walking Mechanics

Spatiotemporal parameters

Walking pattern significantly changed with military equipments carriage (all P < 0.01). In BT, although stride duration and the associated step length and frequency did not significantly change compared to SP, stance duration increased by 2.4% ± 1.1% (P < 0.01) and duty factor increased by 2.2% ± 0.6% (P < 0.001) because of the 10.1% ± 2.6% increase in double-support duration (P < 0.001). These results show that adaptations of soldiers to military intermediate and trunk-equilibrated load carriage (i.e., BT, ∼27% of BM) occurred within the stride. In RM condition (∼46% of subjects’ BM), all the spatiotemporal parameters were modified compared to SP and BT (all P < 0.05). The two major adaptations observed in RM versus SP were (i) an increase in stride duration (by 2.5% ± 2.3%, P < 0.01) resulting from an increase in step length (by 1.8% ± 2.1%, P < 0.01) and (ii) an associated redistribution of the gait phases within the walking stride as shown by the 3.1% ± 0.7% increase in duty factor (P < 0.001) caused by the 13.8% ± 2.5% higher double-support duration (P < 0.001). These results indicate that a progressiveness in the adaptations occurred when subjects carried additional loads that shifted from “null” (SP) to military intermediate-equilibrated (BT) and then to high-load-backward (RM) carriage. The main changes consisted in alterations of the gait phases within the stride (i.e., from SP to BT) and, when the constraint represented by the additional load increased (i.e., from BT to RM), adaptations of the entire stride while at the same time changing the gait phases within.

Kinetic parameters

Absolute Wext and Wint,dc significantly increased with military intermediate (BT) and high-load (RM) equipments carriage (P < 0.0001). Thus, adding masses to soldiers’ BM induced a significant increase in the muscular mechanical actions performed to lift/accelerate and transfer the COM (center of total mass, COTM) from one foot to the other during the double contact phase. Moreover, this increase was so large that Wext,TM and Wint,dc,TM were significantly higher in BT and RM conditions compared to SP (P < 0.01; Table 4). In other words, the Wext and Wint,dc done per unit of additional mass were higher than those done per unit of BM. This suggests that military equipment carriage had more than a “passive” effect on the mechanical works and shows how important were these muscular works during military-type walking. Further, Recovery did not significantly change in any of the tested conditions, showing that the inverted pendulum mechanism of mechanical energy exchange was not disturbed by the carriage of military loads. Thus, it seems that the subjects’ capabilities to save mechanical energy of the COTM by exchanging kinetic (i.e., changes in COTM velocity) and gravitational potential (i.e., changes in COTM vertical position) energies were not altered by military load carriage. Finally, locomotor efficiency did not significantly change during military load carriage, suggesting that the mechanical muscular actions remained at the same level of metabolic effectiveness to produce COTM displacements (Wext) and stance-to-stance transfers (Wint,dc) in the different walking conditions (SP, BT, and RM).

To further analyze these kinetic data, the mechanical work values computed in BT and RM were also expressed as a percent of the SP condition values and plotted as a function of the TM-to-BM ratio (Fig. 1B). By doing so, we observed that the effect of military load carriage on Wext was more moderate when soldiers were already carrying a part of their equipments (i.e., from BT to RM) than when shifting from “null” to intermediate load carriage (i.e., from SP to BT, see the indicative slopes in Fig. 1B). Conversely, Wint,dc increased more from BT to RM than from SP to BT. Therefore, the higher the total military load carried, the higher the relative muscular work needed to transfer the COTM from one foot to the other compared to accelerating/lifting it, even if Wext represented, in absolute value, the major part of the total mechanical work computed in this study.

Furthermore, the increases in total mechanical work (i.e., Wext + Wint,dc) from unloaded (SP) to military intermediate-loaded (BT) and high-loaded (RM) walking (∼42% from SP to BT and ∼69% from SP to RM; Fig. 1B) were extremely close to those of net CW in the same conditions (∼43% from SP to BT and ∼71% from SP to RM; Fig. 1A). Consequently, the net metabolic cost of walking increased in nearly direct proportion to the total mechanical work (as confirmed by the constant locomotor efficiency; Table 4), showing that the quasitotality of the metabolic adaptations to military load carriage was associated with the work done on the COTM (vs making the body rigid and swinging the limbs relative to the COM, i.e., internal work).

Once again, these results were in accordance with the findings of Griffin et al. (22) for moderate speeds (0.5–1.5 m·s−1) while carrying loads about the waist. Therefore, this contradicts our initial hypothesis that the complex distribution of military equipments would induce larger effects than carrying loads very close and totally equilibrated around the COM, as done by these authors. This observation also shows that when the major part of the load is located close to subjects’ COM and counterbalanced between the front and the back of the trunk, the load effects are interestingly reduced, even if some equipments (i.e., loads) remain positioned at distal locations (e.g., helmet, boots, supple joint protections).

This study showed that mass-relative external mechanical work (Wext,TM, J·kg−1·m−1) was higher during walking with external military load carriage (BT and RM) than during unloaded walking (SP) at the speed of 4 km·h−1. When comparing the present kinetic data with those reported in other research contexts like obesity, it is interesting to note that the external mechanical work relative to TM in motion (i.e., subjects’ BM) did not significantly change between normal-weight and obese subjects at similar walking speeds (34). These results indicate that external military load carriage induces greater changes in walking kinetics (when expressed relative to TM) than internal–additional load carriage, as obesity could be interpreted from a mechanical standpoint. We suggest that this effect can be attributed to the immediate effect of external military load carriage compared to the continuous and chronic aspect of internal–additional load faced by obese individuals while walking (10). Taken together, these results show that even if infantrymen develop high abilities to load carriage over their career, it remains insufficient to make them develop an ability to reduce the “negative” mechanical effects of load carriage, as obese individuals may experience.

Finally, when considering both spatiotemporal and kinetic parameters, we can assume that the higher Wint,dc observed here was related to the increase in double-support duration during walking with external load carriage. Indeed, when the TM in motion increases, the transfer of the COTM from one foot to the other requires more work (Table 4) and thus higher GRF. This phenomenon may affect the subject’s gait stability and walking balance. Consequently, an increase in double-support duration may induce a decrease in the extent of this effect through the actions of both feet during their simultaneous contact against the ground, which allows a better control of the COTM displacements. Interestingly, the significant (P < 0.0001) correlation between Wint,dc,TM (i.e., the Wint,dc independently of the load) and the double-support duration partly supported this hypothesis (Fig. 2; r = 0.472).

F2-21
FIGURE 2:
Correlation between double-support duration (in percentage of stride duration) and total mass-relative internal work done during double contact (W int,dc,TM). Pooled data of all subjects (n = 10), conditions (n = 3), and measurement sessions (n = 2).

CONCLUSIONS

To our knowledge, this study is the first to describe energy cost, mechanical work, and the associated locomotor efficiency during walking with military intermediate- and high-load carriage in experienced infantrymen. Compared to unloaded walking, the mechanical work and energy cost parameters of walking increased significantly during military load carriage. However, contrary to what was expected, these mechanical and metabolic effects observed during military equipments carriage (which represents a complex load distribution form) appeared not greater than those reported in studies in which the loads were carried around the waist. This result suggests that the ergonomic approach aiming at organizing and counterbalancing the loads around soldiers’ body, while respecting military battle needs, is effective and beneficial. However, even if load distribution has positive effects, each unit of mass carried by soldiers has a considerable effect on their mechanical and metabolic behaviors (and also indirectly on the risk of injuries). As a consequence, military equipments definition by armies and development by manufacturers should consider the loads carried ever more carefully to place the soldiers in the best and most secure conditions for campaigns. Further, future investigations should seek whether the immediate effects of load carriage reported here are also observed, or differ, when soldiers perform military missions in the field and experience fatigue, which may represents a factor of injuries during military load carriage.

No funding was received for this study from National Institutes of Health, Wellcome Trust, Howard Hughes Medical Institute, or others.

The authors thank Dr. Pierre Samozino for his stimulating discussion of the data and assistance in statistical analyses.

The authors have no conflict of interest.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

REFERENCES

1. Attwells RL, Birrell SA, Hooper RH, Mansfield NJ. Influence of carrying heavy loads on soldiers’ posture, movements and gait. Ergonomics. 2006; 49 (14): 1527–37.
2. Bastien GJ, Heglund NC, Schepens B. The double contact phase in walking children. J Exp Biol. 2003; 206 (Pt 17): 2967–78.
3. Bastien GJ, Schepens B, Willems PA, Heglund NC. Energetics of load carrying in Nepalese porters. Science. 2005; 308 (5729): 1755.
4. Bastien GJ, Willems PA, Schepens B, Heglund NC. Effect of load and speed on the energetic cost of human walking. Eur J Appl Physiol. 2005; 94 (1–2): 76–83.
5. Beekley MD, Alt J, Buckley CM, Duffey M, Crowder TA. Effects of heavy load carriage during constant-speed, simulated road marching. Mil Med. 2007; 172 (6): 592–5.
6. Belli A, Bui P, Berger A, Geyssant A, Lacour JR. A treadmill ergometer for three-dimensional ground reaction forces measurement during walking. J Biomech. 2001; 34 (1): 105–12.
7. Birrell SA, Haslam RA. The influence of rifle carriage on the kinetics of human gait. Ergonomics. 2008; 51 (6): 816–26.
8. Birrell SA, Hooper RH, Haslam RA. The effect of military load carriage on ground reaction forces. Gait Posture. 2007; 26 (4): 611–4.
9. Brockway JM. Derivation of formulae used to calculate energy expenditure in man. Hum Nutr Clin Nutr. 1987; 41 (6): 463–71.
10. Browning RC, Kram R. Pound for pound: working out how obesity influences the energetics of walking. J Appl Physiol. 2009; 106 (6): 1755–6.
11. Browning RC, Modica JR, Kram R, Goswami A. The effects of adding mass to the legs on the energetics and biomechanics of walking. Med Sci Sports Exerc. 2007; 39 (3): 515–25.
12. Cavagna GA. Force platforms as ergometers. J Appl Physiol. 1975; 39 (1): 174–9.
13. Cavagna GA, Kaneko M. Mechanical work and efficiency in level walking and running. J Physiol. 1977; 268 (2): 467–81.
14. Cavagna GA, Saibene FP, Margaria R. External work in walking. J Appl Physiol. 1963; 18: 1–9.
15. Cavagna GA, Thys H, Zamboni A. The sources of external work in level walking and running. J Physiol. 1976; 262 (3): 639–57.
16. Datta SR, Ramanathan NL. Ergonomic comparison of seven modes of carrying loads on the horizontal plane. Ergonomics. 1971; 14 (2): 269–78.
17. di Prampero PE. The energy cost of human locomotion on land and in water. Int J Sports Med. 1986; 7 (2): 55–72.
18. Donelan JM, Kram R, Kuo AD. Mechanical work for step-to-step transitions is a major determinant of the metabolic cost of human walking. J Exp Biol. 2002; 205 (Pt 23): 3717–27.
19. Duffield R, Dawson B, Pinnington HC, Wong P. Accuracy and reliability of a Cosmed K4b2 portable gas analysis system. J Sci Med Sport. 2004; 7 (1): 11–22.
20. Durnin JV, Womersley J. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J Nutr. 1974; 32 (1): 77–97.
21. Foissac MJ, Millet GY, Geyssant A, Freychat P, Belli A. Characterization of the mechanical properties of backpacks and their influence on the energetics of walking. J Biomech. 2009; 42 (2): 125–30.
22. Griffin TM, Roberts TJ, Kram R. Metabolic cost of generating muscular force in human walking: insights from load-carrying and speed experiments. J Appl Physiol. 2003; 95 (1): 172–83.
23. Haisman MF. Determinants of load carrying ability. Appl Ergon. 1988; 19 (2): 111–21.
24. Harman EA, Frykman PN, Knapik JJ, Han KH. Backpack vs. front–back: differential effects of load on walking posture. Med Sci Sports Exerc. 1994; 26: S140.
25. Knapik J, Harman E, Reynolds K. Load carriage using packs: a review of physiological, biomechanical and medical aspects. Appl Ergon. 1996; 27 (3): 207–16.
26. Knapik JJ, Reynolds KL, Harman E. Soldier load carriage: historical, physiological, biomechanical, and medical aspects. Mil Med. 2004; 169 (1): 45–56.
27. Koerhuis CL, Veenstra BJ, van Dijk JJ, Delleman NJ. Predicting marching capacity while carrying extremely heavy loads. Mil Med. 2009; 174 (12): 1300–7.
28. Legg SJ. Comparison of different methods of load carriage. Ergonomics. 1985; 28 (1): 197–212.
29. Lloyd R, Cooke CB. The oxygen consumption associated with unloaded walking and load carriage using two different backpack designs. Eur J Appl Physiol. 2000; 81 (6): 486–92.
30. Majumdar D, Pal MS, Majumdar D. Effects of military load carriage on kinematics of gait. Ergonomics. 2010; 53 (6): 782–91.
31. Maloiy GM, Heglund NC, Prager LM, Cavagna GA, Taylor CR. Energetic cost of carrying loads: have African women discovered an economic way? Nature. 1986; 319 (6055): 668–9.
32. Pal MS, Majumdar D, Bhattacharyya M, Kumar R, Majumdar D. Optimum load for carriage by soldiers at two walking speeds on level ground. Int J Ind Ergonom. 2009; 39 (1): 68–72.
33. Pandolf KB, Givoni B, Goldman RF. Predicting energy expenditure with loads while standing or walking very slowly. J Appl Physiol. 1977; 43 (4): 577–81.
34. Peyrot N, Thivel D, Isacco L, Morin JB, Duche P, Belli A. Do mechanical gait parameters explain the higher metabolic cost of walking in obese adolescents? J Appl Physiol. 2009; 106 (6): 1763–70.
35. Quesada PM, Mengelkoch LJ, Hale RC, Simon SR. Biomechanical and metabolic effects of varying backpack loading on simulated marching. Ergonomics. 2000; 43 (3): 293–309.
36. Ricciardi R, Deuster PA, Talbot LA. Metabolic demands of body armor on physical performance in simulated conditions. Mil Med. 2008; 173 (9): 817–24.
37. Schepens B, Bastien GJ, Heglund NC, Willems PA. Mechanical work and muscular efficiency in walking children. J Exp Biol. 2004; 207 (Pt 4): 587–96.
38. Scholander PF. Analyzer for accurate estimation of respiratory gases in one-half cubic centimeter samples. J Biol Chem. 1947; 167 (1): 235–50.
39. Scott PA, Christie CJ. “Optimal” speed–load combinations for military manoeuvres. Int J Ind Ergonom. 2004; 33 (1): 63–8.
40. Stuempfle KJ, Drury DG, Wilson AL. Effect of load position on physiological and perceptual responses during load carriage with an internal frame backpack. Ergonomics. 2004; 47 (7): 784–9.
Keywords:

INSTRUMENTED TREADMILL; OXYGEN CONSUMPTION; LOCOMOTION; LAND WARFARE; MILITARY EQUIPMENT

Supplemental Digital Content

©2012The American College of Sports Medicine