Secondary Logo

Journal Logo

Effects of Hiking Pole Inertia on Energy and Muscular Costs During Uphill Walking

FOISSAC, MATTHIEU J.1,2; BERTHOLLET, ROMAIN1; SEUX, JULIEN1; BELLI, ALAIN1; MILLET, GUILLAUME Y.1

Medicine & Science in Sports & Exercise: June 2008 - Volume 40 - Issue 6 - p 1117-1125
doi: 10.1249/MSS.0b013e318167228a
APPLIED SCIENCES: Biodynamics
Free

Introduction/Purpose: The purpose of the present study was to investigate the effects of using hiking poles with different inertia on oxygen cost (V˙O2) and muscular activity.

Methods: Eleven subjects walked at 3 km·h−1 on a treadmill inclined at 20% grade. Three mass (240, 300, and 360 g), load distribution, and walking frequency (preferred, −20% and +20%) conditions were tested. Each subject also walked without poles and carried a 360-g mass. V˙O2 and average EMG (aEMG) of nine muscles from lower (soleus, gastrocnemius lateralis, vastus lateralis, biceps femoris, gluteus maximus) and upper (latissimus dorsi, biceps brachii, triceps brachii, and anterior deltoid) limbs were recorded.

Results: Using poles significantly reduced lower limb muscle aEMG values (P < 0.001) by about 15% and increased upper limb muscle aEMG values (P < 0.001) by about 95%. Hand-masses of 360 g did not result in an increased V˙O2, and the only modification in terms of muscular activation was greater biceps brachii activity (+55%, P = 0.006). Biceps brachii and anterior deltoid activity were also influenced by pole mass and load distribution (P < 0.01). Walking at high frequency increased both aEMG and V˙O2, whereas walking at low frequency redistributed the muscular work from the thigh muscles to calf and upper limb muscles although this did not lead to an increased V˙O2 compared with that at preferred frequency. No interaction between mass and frequency was found for aEMG or V˙O2.

Conclusion: Using poles and changing frequency have important effects on muscle recruitment, whereas the effects of mass were limited when considering poles available on the market.

1PPEH Laboratory, Jean Monnet University, Saint-Etienne; and 2Decathlon Research Center, Biomechanics Laboratory, Villeneuve d'Ascq, FRANCE

Address for correspondence: Matthieu Foissac, MSc, 4, bd de Mons. BP299, 59665 Villeneuve d'Ascq, France; E-mail: matfoissac@yahoo.fr.

Submitted for publication August 2007.

Accepted for publication January 2008.

For the last decade, the use of walking poles during hiking, Nordic walking, or trail running has become extremely popular. During level or downhill walking, an appropriate use of poles increases stability and significantly decreases mechanical stress on the lower limbs, particularly on the ankle and knee joints (1,23,26).

The effect of hiking poles on energy cost is disputed and conflicting results have been obtained for different grades or walking speeds. The growing use of poles in mountain running competitions tends to show that they contribute to alleviating fatigue, particularly during uphill walking. Jacobson et al. (9) demonstrated that the steeper the gradient, the better the perceived exertion with poles compared to no-poles. The more effective forces exerted by the arms at high grades were considered to be the main explanation. However, even though the perceived exertion was shown to be reduced, no significant decrease in metabolic response was found during loaded uphill walking with poles at 2.4 km·h−1 for +10% to +25% grades. Knight and Caldwell (10) showed a significant decrease in the myoelectric activity of some lower limb muscles and a higher activation of the triceps brachii (TB) during pole walking compared to normal walking. However, no EMG evaluation of the specific muscles involved in pole use was carried out for this study.

During level ground walking at 6 to 7 km·h−1, increases of 12% to 23% in energy consumption have been reported when using poles weighing between 340 and 450 g (19,20). Not only the mass of the poles but also the frequency of the movement and its amplitude (i.e., hands swinging more or less from the back to the front while walking) were hypothesized as being responsible for this increase. Indeed, hand-carried masses as low as 0.45 kg per hand associated with an exaggerated movement of the arms increased oxygen consumption compared to the no-load condition (7). Nevertheless, without any instruction relative to arm movement, no significant increase in V˙O2 could be found (18) for hand-masses from 0.45 to 2.27 kg. The mass of hiking poles has decreased substantially in recent years usually weighing from 200 to 360 g-less than all the masses previously investigated on upper limbs. For instance, the mean (±SD) mass of 28 poles randomly chosen in several mountaineering stores in France was about 300 (±30) g.

For the same mass, the increase in oxygen consumption during walking is higher when the mass is carried with the hands rather than on the feet (6,24). Previous research on running has shown significant metabolic and mechanical adaptations with loads of 0.50 kg attached to each foot, but failed to identify differences for lighter masses of 0.18, 0.25, and 0.335 kg (3,15). Oxygen consumption increased with 0.5 kg masses by 2.72 mL·kg−1·min−1 (15).

The results cited above were obtained in very heterogeneous conditions. However, two main conclusions can be drawn. First, the amount of mass carried has a direct influence on the energy cost of walking, although this is low when masses are below a few hundred grams. Second, an elevated speed and its associated high frequency seem to be determinant factors for the negative effect of poles on the energy cost of walking. It can therefore be hypothesized that there is an interaction between walking frequency and mass carried on energy cost. To the best of our knowledge, this assumption has never been examined.

Finally, for lower limbs, it has been shown that proximo-distal distribution of mass influences energy cost. A load with a center of gravity close to the hips requires less energy and fewer mechanical adaptations than one close to the feet during both running and walking (2,15,21). No similar information can be found relative to upper limbs. In addition, the poles being held far from the shoulder, it can reasonably be hypothesized that the effects of their inertia on shoulder muscle activity would be amplified.

Therefore, the first objective of the present study was to investigate the physiological effects of using hiking poles, independently from their mass. The second objective was then to measure the effects of pole inertia on energy cost and EMG activity. This second objective was tested through changing the walking frequency, pole mass, and center of gravity.

Back to Top | Article Outline

METHODS

Subjects.

Eleven male volunteers (mean ± SD, age 24.0 ± 4.6 yr; mass, 74.1 ± 11.4 kg; height, 177.5 ± 8.0 cm) participated in this experiment. Written informed consent was obtained from the subjects and the Ethics Committee of Rhône-Alpes Loire (CCPPRB) gave its approval for the experimentation.

Back to Top | Article Outline

Procedures.

One or two 1-h practice sessions of walking with poles on the treadmill were done before the main experiment. During these familiarization sessions, subjects were instructed to use the alternate stride technique, i.e., with an antiphase movement of the arms relative to the legs. The volunteers were not selected as subjects if they were not able to use the poles correctly. For each subject, the poles were adjusted to 70% of body height (23). This pole height roughly corresponded to the method with the elbow at 90° while the pole is held in a vertical position and in contact with the ground (9,10).

The subjects walked at 3 km·h−1 on a treadmill inclined at 20% in line with previous studies (9,14). The study consisted of 14 bouts of 4 min separated by 2 min recovery periods. First, the subjects walked normally without poles (condition i: NPpre) in order to determine their preferred walking frequency. During this condition, the mean walking frequency was measured by counting the number of steps over about 1 min (e.g., 87 steps in 58.4 s = 1.49 Hz). Then, 12 randomized bouts of walking with different pole masses and step frequencies were performed. For these 12 bouts, an electronic metronome was used to impose the walking frequency. Before each trial, it was adjusted to the preferred +20 or −20% frequencies, and subjects were asked to strictly follow this cadence. After analysis, mean walking frequencies were checked and found to be −19.0 ± 2.4% and +19.6 ± 3.4% for NF−20 and NF+20, respectively. The 12 conditions included one with the subjects carrying poles but not using them (condition ii: HW) and 11 with the subjects using the poles normally (conditions iii to v) at preferred walking frequency with heavy, medium, and light poles (HP, MP, LP, respectively), (conditions vi and vii) with HP poles but with the center of mass lowered or elevated (HPLCoG and HPHCoG, respectively; Fig. 1) and (conditions viii to xiii) with HP, MP, and LP but at a stride frequency 20% higher or 20% lower than the preferred frequency (HPNF+20, MPNF+20, LPNF+20, HPNF−20, MPNF−20, LPNF−20, respectively). Finally, a second bout with the subjects walking normally without poles (condition xiv: NPpost) was performed. The protocol is summarized in Figure 2.

FIGURE 1

FIGURE 1

FIGURE 2-Ex

FIGURE 2-Ex

The three mass conditions roughly corresponded to the minimum, mean, and maximum masses observed among the products on the market, i.e., 240 g for LP, 300 g for MP, and 360 g for HP. The two conditions using heavy poles with a different mass distribution (HPLCoG and HPHCoG) were fixed at 52% and 72% of pole size, respectively, to cover the range of CoG distributions observed on existing hiking poles (see Fig. 1). The mass distribution was measured as the distance between the tip and its center of gravity, determined by putting it in equilibrium on a sharp base, the pole always being adjusted to 1.20 m in length.

Back to Top | Article Outline

Gas exchange measurements.

Mean values for oxygen uptake (V˙O2), carbon dioxide output (V˙CO2), and RER were measured during the last minute of each 4-min period. The subjects breathed through a two-way nonrebreathing valve (Hans Rudolph, series 2700) connected to a three-way stopcock for the collection of gases (100-L bag). Expired air was collected for about 30-45 s. The volume of the expired gas was measured in a Tissot spirometer. Fractions of expired gases were determined with a paramagnetic O2 analyzer (Servomex, cell 1155B) and infrared CO2 analyzer (Normocap Datex). The analyzers were calibrated with mixed gases, the composition of which was determined using Scholander's method (22). For clarity, because the RER were very similar among conditions, it was decided to express the energy cost of walking in mL O2·kg−1 rather than in J·kg−1.

Back to Top | Article Outline

EMG recording and analysis.

Surface EMG of nine muscles was measured for about 30s immediately before the gas exchange measurements so that the subjects were not perturbed by the mouthpiece during the EMG recording. Among these nine muscles, five were located on the lower limbs: soleus (Sol), gastrocnemius lateralis (GL), vastus lateralis (VL), biceps femoris (BF), and gluteus maximus (GM). Four were located on the upper part of the body: latissimus dorsi (LD), biceps brachii (BB), TB, and anterior deltoid (AD). These muscles were chosen based on previous studies using EMG during walking with (10) or without (5,21,25) poles.

The EMG signals were recorded by using Ag/AgCl preamplified bipolar surface electrodes (electrode ECG universelle, Control Graphic Medical, Brie Comte Robert, France). The recording electrodes were fixed lengthwise over the muscle belly with an interelectrode distance of 25 mm. The single reference electrode was placed on the patella. Low impedance (< 5 kΩ) was obtained by abrading the skin with emery paper and cleaning with alcohol. EMG activity was recorded online (EISA 16-2, Freiburg, Germany and DAQCard-6062E acquisition card, National Instruments, Austin, TX, USA) with a sampling frequency of 2000 Hz and amplified (gain = 1000-2500). The signal was treated using the Imago software developed under Labview (National Instruments).

The EMG signals from the nine muscles were analyzed over 20 consecutive strides. The raw EMG signals were rectified and integrated (iEMG). Then, the average EMG was determined for each considered cycle (aEMGcycle) by dividing iEMG by the cycle duration. The mean aEMG value over the 20 cycles was retained. Based on the net onset of BB and TB activity, two phases were visually determined for each cycle, namely the pushing and recovery phases (see Fig. 3). The average EMG was also considered for these two phases (aEMGpush and aEMGrec). aEMGtot was calculated as the sum of aEMGcycle of each of the nine muscles. aEMGupper and aEMGlower were calculated as the sums of aEMGcycle of the 4 upper limb muscles and the 5 lower limb muscles, respectively.

FIGURE 3

FIGURE 3

Back to Top | Article Outline

Statistical analysis.

A one-way repeated measures ANOVA with the pole and the no-pole conditions as dependant variable was first used to compare the conditions NPpre, NPpost, HW, and HP. A 3 (poles masses) × 3 (walking frequencies) repeated-measures ANOVA was used to test for main effects of pole mass and walking frequency and the interaction between mass and frequency. When an effect was detected, Tukey post hoc test was used for multiple pairwise comparisons. For all analyses, a P value of 0.05 was accepted as the level of statistical significance. The effect sizes (d) were calculated for each comparison using standardized mean differences (8). When d< 0.50, the effect was considered small; when 0.50 < d < 0.80, it was considered medium; and when d > 0.80, it was considered large.

Back to Top | Article Outline

RESULTS

Effect of using poles.

The ANOVA revealed no difference in V˙O2 between NPpre and NPpost. Consequently, the two conditions without poles were averaged (NP) in order to compare this single value with HW and HP. No significant difference in V˙O2 was found among these three conditions. A significant main effect was detected for aEMGtot, with HP being higher than NP (P = 0.003, d = 0.46), whereas no significant difference was observed between holding weights and the HP or the NP conditions. All upper limb muscles except BB exhibiteda significant (P < 0.001, d = 1.56 for TB, P = 0.005, d = 1.01 for AD and P < 0.001, d = 1.49 for LD) increase in aEMG for HP compared to both NP and HW (Fig. 4, upper panel). BB activation was significantly greater for HW and HP compared to NP (respectively, P = 0.006, d = 1.24 and P = 0.012, d = 1.01).

FIGURE 4-aE

FIGURE 4-aE

Regarding the aEMGlower, ANOVA revealed a significant main effect (P < 0.001) and the post hoc test showed that aEMG for HP was 14.8% lower (P < 0.001, d = 0.77) than NP and 13.8% lower (P < 0.001, d = 0.70) than HW, again with no significant differences between the two conditions without poles. As seen in Figure 4 (lower panel), this difference is due to reduced activation of VL (−14 ± 5%), GL (−26 ± 8%), and Sol (−22 ± 6%) when using the poles. No difference was observed between NP and HW for any lower limb muscles. Neither BF nor Glut activity were influenced by the masses or the use of poles.

Back to Top | Article Outline

Effect of frequency and inertia.

ANOVA revealed a significant influence (P = 0.009) of walking frequency on V˙O2, independently from mass. High-frequency conditions (NF+20) exhibited a significant average increase in V˙O2 compared to both the low (NF−20) and preferred frequency (NF) conditions (respectively P = 0.025, d = 0.68 and P = 0.013, d = 0.86). No significant difference in V˙O2 was found between NF−20 and NF (Fig. 5).

FIGURE 5

FIGURE 5

aEMGupper and aEMGtot were higher for NF−20 than for other frequencies (P = 0.045, d = 0.44 for NF and P = 0.018, d = 0.49 for NF+20; Fig. 6). This was mainly due to an increase in aEMG of TB and AD at low frequency (Table 1). In lower limbs, GL also exhibited a greater activation at low frequency than at high frequency, but the opposite result was found for Glut and VL (Table 1). No significant difference between frequencies was found for BF and Sol.

FIGURE 6

FIGURE 6

TABLE 1

TABLE 1

There was a general trend towards a reduced aEMGpush of TB and AD and lower aEMGrec of AD at high frequency (see Table 2). For lower limb muscles, the only significant difference was a higher aEMGpush of Glut for NF+20 compared to NF−20 (P = 0.025, d = 0.76).

TABLE 2

TABLE 2

The mass or mass distribution of the poles did not significantly affect V˙O2, aEMGtot or aEMGlower. In addition, ANOVA did not show any cross-interaction of mass-frequency on V˙O2, aEMGtot and aEMGlower. More surprisingly, the same results were found on global upper limb muscle activation. Looking more specifically at each muscle, it was found that only two upper limb muscles were influenced by mass or mass distribution when considering aEMGcycle. BB activation was significantly higher with MP and HP than with LP (P = 0.003, d = 0.23), and AD activation was significantly higher for LCoG than either MCoG (P = 0.005, d = 0.51) or HCoG (P = 0.015, d = 0.42). Moreover, when focusing on the recovery phase, TB and BB activation was significantly higher with heavy poles (see Table 3) and exhibited a slightly higher aEMGrec with HP and MP than with LP (P = 0.037 for TB, P = 0.01 and 0.007 for BB, d between 0.08 and 0.19).

TABLE 3

TABLE 3

Back to Top | Article Outline

DISCUSSION

The present study provides additional evidence to support the assumption that the use of hiking poles during uphill walking does not lead to an increased energy cost and redistributes muscular activity from lower to upper limbs. It also gives a precise description of muscular activity patterns during pole walking. More importantly, three main results were obtained: (i) hand-carrying loads or using poles of up to 360 g required more muscular activity of upper limb muscles but did not result in a significantly higher energy consumption, (ii) walking with poles at high frequency increased energy cost compared to preferred frequency, whereas low frequency did not affect the global energy cost despite changes in muscle recruitment, and (iii) no interaction between pole mass and walking frequency was observed on either EMG or V˙O2.

Back to Top | Article Outline

Influence of using poles.

Hiking with poles or carrying loads of up to 360 g in the hands does not significantly change energy costs during uphill walking (mean values around 0.54 mL O2·kg−1·m−1). However, for rapid level walking and exaggerated arm movements with poles having a similar mass, opposite results have been found in the literature because the metabolic response increased by between 12% and 23% (19,20). Besides the exaggerated arm movements in these two studies, another possible explanation for these increases lies in the fact that vertical forces exerted by the poles on the ground are probably much greater in the vertical than in the horizontal direction, as shown by Komi (11) in cross-country skiing. Thus, it can be assumed that lower propulsive forces can be exerted with hiking poles on level ground (i.e., horizontal forces) than in uphill walking where the vertical component plays a more important role. Cross-country skiers, for whom propulsive arm forces are essential, exhibit low cycle frequencies (0.7 to 1.1 Hz), longer poles, and trunk flexion to produce a more effective force at the end of the propulsive phase (17). In spite of these characteristics, horizontal forces were shown to be three times lower than vertical ones on level ground in cross-country skiing (11).

EMG profiles of the lower limb muscles for the NP condition are consistent with those previously observed (10,12,13). In addition, aEMG values displayed in the present study provide additional evidence of the efficient redistribution of muscular activity from lower to upper limbs. When the subjects walked with poles, there was a significant lowering in aEMGlower by 15% through increased activation of the upper limb muscles. The reduced aEMGlower concerned only the muscles implicated in knee extension and plantarflexion and activated during the stance phase. Indeed, the aEMG of VL, GL, and Sol decreased respectively by 14%, 26%, and 22% when using poles, which was similar to the results of Knight and Caldwell (10), who showed decreases of 12% to 16% for the same muscles.

Neither Glut nor BF were influenced by the use of the poles. This can partly be explained by the fact that one important role of hamstring muscles and Glut was shown to be to decelerate the leg prior to heel contact (12,25,27). Hamstring muscles actually counteract the quadriceps action during knee movements in order to protect the anterior cruciate ligament (4,12). This deceleration is dependant on velocity, which in the present study was identical with or without poles. There is also more Glut activation for steep grades of more than 15% than on level ground after mid-stance and until heel-off, in order to elevate the body's center of mass, as described by Tokuhiro et al. (25). Again, the elevation was similar with or without poles. Yet hamstring muscles act synergistically with the quadriceps during the stance phase. Because VL activity was reduced with poles, one might expect that BF would exhibit the same tendency, as found by Knight and Caldwell (10). This was not the case in the present study. A complementary explanation could be that another role of these two muscles is to counterbalance gravitational forces during weight acceptation (27). Unlike Knight and Caldwell's study, no backpack was used, so the gravitational forces found were slightly lower, which could explain the lower modifications of EMG activity with poles observed for these muscles. Hiking poles therefore seem to significantly relieve the muscles implicated in propulsion but have a lesser influence on the muscles involved in leg swing or to face gravitational forces.

There was significantly more activation of all four upper limb muscles with poles than without, and the increases amounted to roughly +50% for BB, +70% for AD, +100% for LD, and +150% for TB. This result is in agreement with and completes the data of Knight and Caldwell (10) who noticed a similar increase in EMG for TB with poles when uphill walking, but with a 9% grade at a speed corresponding to 55-65% of their theoretical maximal heart rate, an intensity comparable to that of the present study.

For BB activation, the difference between HW and NP was similar to the difference between HP and NP (about +50%) which suggests that the mass of the poles was solely responsible for the higher activation of BB. For AD, the increase of 70% for HP when compared with either NP or HW indicates that this muscle was only affected by the use of poles and not their mass. In addition, when comparing HP and HW, LD and TB exhibited a much higher activation when the poles were used, which is merely due to a stronger contraction during the pushing phase of the cycle (Figs. 3 and 7) where elbow and shoulder extensions are performed.

FIGURE 7

FIGURE 7

Overall, these results suggest that walking with poles using the alternate stride technique, by relieving knee extensor muscles, could alleviate knee joint forces and, during long walks, lessen fatigue of the lower limb muscles. This should also lead to increased stability and a lower risk of injury. However, complementary studies need to be performed to confirm this hypothesis.

Back to Top | Article Outline

Effect of inertia and frequency.

In the range of masses studied in the present study, the mass of the poles was shown to have no direct influence on V˙O2 during uphill walking. Similarly, the position of the center of gravity of the pole had no significant influence on global V˙O2. Other studies have reported opposing results regarding the mass but using much heavier loads, e.g., 1.36 kg for Graves et al. (6). For lower limbs, no author has shown any significant influence of masses of less than 500 g (3,15) while running. It could therefore be assumed that equipment weighing less than 500 g does not lead to significant increases in energy cost.

Regarding the EMG activity, BB was the only muscle affected by the additional mass of the poles (+9% between both MP and HP compared to LP). This could indicate that a 360-g pole induces fatigue in this muscle more rapidly than a 240-g one, even though it is probably not a limiting factor during hiking. No upper limb muscles other than BB are influenced by pole mass.

Similarly, the position of the center of gravity of the pole has no major influence on muscle activation. The only difference was found for AD which showed higher activation during the recovery phase for low load distribution than with either medium or high load distributions. This also seems to indicate that a distal mass could be more strenuous for some shoulder flexion muscles. This result is in accordance with those of Royer et al. (2) who studied walking with masses at different proximo-distal positions on lower limbs but with much higher loads (2 kg).

The present study shows that uphill walking with poles at high frequency is less efficient than uphill walking with poles at preferred or low frequency. At the beginning of the cycle, with the pole in an upright position, propulsive forces were shown to be lower (16). Therefore, it can be hypothesized that for high frequency, as cycle time becomes shorter, propulsive forces exerted on the poles, mainly at the end of the cycle, are lower. On the contrary, a longer cycle time during low frequency allows the walker to exert a longer and more effective force with the arms. This was illustrated by the higher activation of TB during the pushing phase and of AD during the whole cycle for NF−20. This therefore allowed a significant reduction of VL and Glut activation at NF−20 compared to NF+20. Yet, it resulted in a 15% higher activation of GL. These opposing results did not lead to a modification of global metabolic cost, probably because of the relatively lower mass of AD, TB, and GL compared to either VL or Glut. Nevertheless, local muscular fatigue could arise from this situation. Further studies on the propulsive forces exerted by the upper limbs during pole walking at different gradients should be performed to characterize the kinetic contributions of hiking poles.

Back to Top | Article Outline

CONCLUSIONS

The actual mass of hiking poles does not influence energy cost even though differences of mass of about 100 g or of 20 cm in load distribution on hiking poles can be detected through EMG analysis of the BB and AD. Within this mass or COM distribution range, the low alterations observed suggest that poles could contain additional concepts (adjustments, shock absorbers, additional features, etc.) without significantly increasing neuromuscular fatigue. This result was reinforced by the fact that no interaction between mass and frequency was found for any parameter.

At preferred frequency, the use of poles reduced the activation of lower limb muscles by 15% and increased upper limb muscle activation by 95%. High frequency was physiologically inefficient, whereas walking at low frequency redistributed the muscular work from thigh muscles to calf and upper limbs muscles, but this did not lead to an increased V˙O2. Thus, during uphill walking, repeatedly changing the frequency from preferred to low frequency could be a useful strategy to lessen fatigue. Reducing the frequency of arm movements separately from legs (i.e., one pole planted every two steps) could also provide physiological and biomechanical benefits. Complementary studies on walking techniques would be necessary to investigate this hypothesis.

Back to Top | Article Outline

REFERENCES

1. Bohne M, Abendroth-Smith J. Effects of hiking downhill using trekking poles while carrying external loads. Med Sci Sports Exerc. 2007;39(1):177-883.
2. 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.
3. Cavanagh PR, Kram R. Stride length in distance running: velocity, body dimensions, and added mass effects. Med Sci Sports Exerc. 1989;21(4):467-79.
4. Ciccotti MG, Kerlan RK, Perry J, Pink M. An electromyographic analysis of the knee during functional activities. I The normal profile. Am J Sports Med. 1994;22:645-50.
5. Ericson MO, Nisell R, Ekholm J. Quantified electromyography of lower-limb muscles during level walking. Scand J Rehabil Med. 1986;18:159-63.
6. Graves JE, Martin AD, Miltenberger LA, Pollock ML. Physiological responses to walking with hand weights, wrist weights, and ankle weights. Med Sci Sports Exerc. 1988;20(3):265-71.
7. Graves JE, Pollock ML, Montain SJ, Jackson AS, O'Keefe JM. The effect of hand-held weights on the physiological responses to walking exercise. Med Sci Sports Exerc. 1987;19(3):260-5.
8. Howell DC. In: Fundamental Statistics for the Behavioral Sciences, Belmont, CA: Wadsworth; 2007. pp. 402-6.
9. Jacobson BH, Wright T, Dugan B. Load carriage energy expenditure with and without hiking poles during inclined walking. Int J Sports Med. 2000;21:356-9.
10. Knight CA, Caldwell GE. Muscular and metabolic costs of uphill backpacking: are hiking poles beneficial? Med Sci Sports Exerc. 2000;32(12):2093-101.
11. Komi PV. Force measurements during cross-country skiing. Int J Sports Biomech. 1987;8:370-881.
12. Lange GW, Hintermeister RA, Schlegel T, Dillman CJ, Steadman JR. Electromyographic and kinematic analysis of graded treadmill walking and the implications for knee rehabilitation. J Orthop Sports Phys Ther. 1996;23:294-301.
13. Leroux A, Fung J, Barbeau H. Adaptation of the walking pattern to uphill walking in normal and spinal-cord injured subjects. Exp Brain Res. 1999;126:359-68.
14. Lloyd R, Cooke CB. The oxygen consumption associated with unloaded walking and load carriage using two different backpack designs. Eur J Appl Physiol. 2000;81:486-92.
15. Martin PE. Mechanical and physiological responses to lower extremity loading during running. Med Sci Sports Exerc. 1985;17(4):427-33.
16. Millet GY, Hoffman MD, Candau RB, Buckwalter JB, Clifford PS. Cycle rate variations in roller ski skating: effects on oxygen uptake and poling forces. Int J Sports Med. 1998;19:521-5.
17. Millet GY, Hoffman MD, Candau RB, Clifford PS. Poling forces during roller skiing: effects of grade. Med Sci Sports Exerc. 1998;30(11):1637-44.
18. Owens SG, al-Ahmed A, Moffatt RJ. Physiological effects of walking and running with hand-held weights. J Sports Med Phys Fitness. 1989;29:384-7.
19. Porcari JP, Hendrickson TL, Walter PR, Terry L, Walsko G. The physiological responses to walking with and without Power Poles on treadmill exercise. Res Q Exerc Sport. 1997;68:161-6.
20. Rodgers CD, VanHeest JL, Schachter CL. Energy expenditure during submaximal walking with Exerstriders. Med Sci Sports Exerc. 1995;27(4):607-11.
21. Royer TD, Martin PE. Manipulations of leg mass and moment of inertia: effects on energy cost of walking. Med Sci Sports Exerc. 2005;37(4):649-56.
22. Scholander PF. Analyzer for the accurate estimation of respiratory gases in one-half cubic centimeter samples. J Biol Chem. 1947;167:235-50.
23. Schwameder H, Roithner R, Muller E, Niessen W, Raschner C. Knee joint forces during downhill walking with hiking poles. J Sports Sci. 1999;17:969-78.
24. Soule RG, Goldman RF. Energy cost of loads carried on the head, hands, or feet. J Appl Physiol. 1969;27:687-90.
25. Tokuhiro A, Nagashima H, Takechi H. Electromyographic kinesiology of lower extremity muscles during slope walking. Arch Phys Med Rehabil. 1985;66:610-13.
26. Willson J, Torry MR, Decker MJ, Kernozek T, Steadman JR. Effects of walking poles on lower extremity gait mechanics. Med Sci Sports Exerc. 2001;33(1):142-7.
27. Winter DA, Yack HJ. EMG profiles during normal human walking: stride-to-stride and inter-subject variability. Electroencephalogr Clin Neurophysiol. 1987;67:402-11.
Keywords:

OXYGEN COST; EMG; MASS; LOAD DISTRIBUTION; WALKING FREQUENCY

©2008The American College of Sports Medicine