Subjects

The study population was 9 male NW instructors (mean age 36.8±11.9 years, height 1.78±0.10 m, body weight 75.8±5.8 kg, body–mass index [BMI weight in kg divided by height in meters squared] 24.2±1.8 kg/m²) licensed by the ANWI (Associazione Nordic Walking Italia) and with at least 2 years of experience in NW (mean 2.89±1.00 years). The general health status was normal; none had any health condition that could affect exercise capacity.

Ethics Statement

The study was approved by the Ethical Committee of Verona University. All participants were informed verbally and in written form about the nature and procedures of the study before they gave their written consent to participate. They were also informed about their right to withdraw from the study at any time without the need to provide explanation.

Experimental Procedure and Protocols

Tests were performed on a motorized treadmill with a belt surface 2.5 m wide and 3.5 m long (RL3500E, Rodby, Sweden). Subjects used NW poles (Exel, Nordic Walker, Espoo, Finland) equipped with special carbide tips to ensure appropriate grip with the treadmill belt surface. As recommended by the INWA, correct pole length was determined by multiplying the subject’s height in cm by 0.68 rounded down to the nearest 5 cm within a tolerance of 2.5 cm. The subjects performed 5-min tests on the treadmill under the following combination of conditions: conventional walking and NW on level (0%) and uphill walking at a 15% incline. These inclines were chosen because, according to the Nordic walking instructors, both are common conditions encountered during NW. Walking speed was maintained at 4 km/h in each slope and gait condition because it is the usual pace that can be comfortably sustained on either incline. During NW, all subjects used the diagonal technique, which is the most common NW technique worldwide and is characterized by contralateral leg and arm coordination. All tests were presented in random order.

Measurements

Gas exchange and ventilatory parameters were collected breath-by-breath by means of a portable metabolic system (Cosmed K4b2, Rome, Italy) with the main sample unit attached to the chest and the battery pack on the back (800 g). The subjects wore a facemask (70 mL dead space) that directed the respiratory gases via the sampling tube to the analyzer. A low-resistance bidirectional turbine incorporated in the mask measured the breath volume. Before each test, the turbine was calibrated with a 3-L volume syringe (Cosmed, Rome, Italy); the gas analyzer was calibrated with ambient air (20.93% O₂ and 0.03% CO₂) and a known concentration of gasses (16.00±0.04% O₂ and 5.00±0.01% CO₂) (Air Liquide Italia S.p.A., Milan, Italy).

A portable EMG system (Myomonitor, Delsys Inc., Boston, MA, USA) recorded the surface EMG activity from seven upper body (lumbar erector spinae [ES], trapezius [TR], latissimus dorsi [LD], deltoideus anterior [DA], biceps brachii [BB], caput lateralis of triceps brachii [TB], and rectus abdominis [RA]) and five lower limb (gluteus medius [GM], vastus lateralis [VL], biceps femoris [BF], gastrocnemius lateralis [GL] and tibialis anterior [TA]) muscles of the arm dominant side of the body. Handedness was determined on the basis of self-reported hand use. We chose the lateral head of triceps because we wanted to isolate elbow extensor action. Parallel-bar EMG electrodes (DE-2.3, 5-mm single differential surface EMG sensor with two 1-mm Ag contacts 10 mm apart) were positioned longitudinally on the belly of each muscle with respect to the underlying muscle fibres in accordance with standard recommendations [19,20] to minimize cross-talk and geometrical artifacts [21]. For the erector spinae muscle, the electrode was attached at the height of the third lumbar vertebra on the dominant side, 2 cm lateral to the spinous processes [20]. A reference electrode was placed on the posterior aspect of the proximal epiphysis of the right radius. To minimize impedance, the skin was shaved, slightly abraded, degreased, and disinfected with alcohol before attaching the electrodes. The electrode wires were kept close to the skin with elastic nets to avoid artifacts due to movement.

The EMG signals were sampled at a frequency of 1000 Hz, hardware amplified (gain 1000 V/V ± 1%), band-pass filtered (20–450 Hz; 20 dB/oct) to remove noise, converted A/D and transmitted wirelessly (D-link WUA-1340 Wireless G USB adapter) to a computer for real-time data display and storage (EMGworks Acquisition Software, Delsys Inc.).

A foot switch (DC-F01, Delsys Inc.) was used to detect foot contact with the ground. The switch consists of a circular pressure sensitive resistive membrane (radius 1.27 cm) that was attached with double-layer tape to the skin on the rear part of the foot plantar surface under the heel. The signal was acquired with the same device used to sample and record the EMG signals.

Pole force was measured during NW with a lightweight single-axial load cell (Deltatech, Sogliano al Rubicone, Italy) mounted inside the poles under the handgrip. Analog signals from the force transducer were sampled at 100 Hz by means of a data acquisition board (NI DAQ-PAD-6016, 16 bit; National Instruments, Austin, TX, USA). Force transducers were dynamically calibrated a few minutes before each test using a load cell as a reference (546QD; DSEurope, Milan, Italy), as described previously [22]. Data collection for EMG, foot switches, and pole force was triggered by a digital signal in order to ensure synchronization between the force signals, the EMG signal, and the foot switch.

Data Processing

The average value of oxygen consumption (VO₂) was calculated over the last 30 s of each condition. The root mean square amplitudes (RMS) of the EMG signal were calculated for each muscle and for each complete gait cycle occurring within the 30-s acquisition period. Gait cycle was defined as the interval between two consecutive heel strikes of the foot on the investigated body side. Instants of heel strike were recognized when the signal of the foot sensor rose and passed a threshold value of 10% of the average signal measured during the acquisition period. For visualization purposes, the EMG signals were processed by an RMS moving window (window size 250 ms; stepwise value for value); each walking cycle was then standardized in time to 100 of the whole cycle. For each subject and muscle, the RMS values were normalized using the mean dynamic method [23] and expressed as percentage of the mean RMS calculated from the EMG signal obtained during conventional walking on level incline. Ensemble averages for each subject were calculated from 20 consecutive cycles and then used to calculate the ensemble average for the group. The magnitude of differences in RMS values between the gaits at a given incline were quantified and are reported as the percentage differences between NW and conventional walking expressed with respect to the conventional walking values. Similarly, the magnitude of the differences in RMS between the inclines for a given gait are reported as the percentage differences between the values obtained at inclines of 15% and 0% expressed with respect to the values at 0% incline.

For NW, the poling cycle was defined as beginning at pole ground contact and ending at the subsequent ground contact of the same pole. Pole contact and pole take-off were identified in the force data as the first point above and the first point below a force threshold of 4 N, respectively. The cycle time for poling action was calculated as the time between two subsequent pole contacts and the duration of the pole thrust action; poling time (PT) was calculated as the time between pole ground contact and pole take-off. The average poling force (PF) over the entire cycle was calculated by dividing the integral of the force–time curves by the duration of poling cycle time. Data were processed using Matlab 7.0 (MathWorks Inc., Natick, MA, USA) and Excel 2003 (Microsoft Corporation, Redmond, WA, USA).

Statistical Analysis

The Shapiro–Wilk test for normality showed that the data were normally distributed. For each condition, the values for each subject are expressed as the mean of all the cycles during the 30-s period of data acquisition. Two-way repeated measures multivariate analysis (MANOVA) was applied to evaluate the influence of gait and incline on the parameters of interest. Assumption of sphericity was checked using Mauchly's test. A Holm–Bonferroni test to correct for alpha inflation was applied post hoc to confirm differences between the gaits at each degree of incline and between the degrees of incline at each gait. Statistical analyses were carried out using SPSS 15.0 Software for Windows (IBM SPSS Inc., Chicago, IL, USA). Statistical significance was set at p < .05.

Metabolic and EMG Differences in Nordic Walking vs Conventional Walking

Oxygen uptake was greater during NW than conventional walking at the same speed, with an estimated extra energy cost of about 23% during level walking at 4 km/h and of 7% during uphill walking. The metabolic data are in line with previous studies investigating energy expenditure [1,2,4] during NW on flat terrain. An extra energy cost of 68% was found in subjects using a vigorous NW technique [6], but no differences were noted when the subjects walked with hiking poles [25] or walked at a brisk speed (2 m/s) [7]. Therefore, the amount of increase in energy expenditure when walking with poles is not a universal value and it is more likely explained by the technique the subjects adopted and by proper pole use.

EMG muscle activity of the upper body muscles during NW was highest in the triceps brachii muscle, with an average 16-fold increase in muscle activation during NW as compared to conventional walking, and a 40-fold increase at the peak of activation during poling action, followed by the latissimus dorsii muscle, with an average 4-fold increase during NW as compared to conventional walking. Both the triceps and the latissimus dorsii muscles are involved in the poling phase during elbow extension and shoulder back extension, respectively. Because muscle activation is increased in the second part of the gait cycle, corresponding to the poling phase, it can be hypothesized that the increased activity of these two muscles concurs during poling force exertion.

A significant increase in biceps brachii and deltoideus anterior muscle activation during NW was also seen, with an average 2- to 3-fold increase throughout the entire gait cycle. As the biceps brachii serves to flex the elbow and the deltoideus anterior to medially rotate and stabilize the shoulder, we can hypothesize that they are not only involved in pole recovery during the arm swing phase but also work in coactivation with the triceps brachii and latissimus dorsii muscles during the poling phase so as to stabilise the elbow and shoulder joints. The increased activity of the rectus abdominis muscles throughout the entire gait cycle during NW may be imputable to a greater need for trunk stabilization. The higher oxygen consumption measured during NW as compared to conventional walking can be attributed to increased muscle activation primarily for thrust action in the poling phase and secondarily for joint stabilisation at pole impact and during the poling phase. A minor role in increasing energy expenditure may be attributed to muscle activation for pole recovery.

Previous studies on muscle activation in NW have produced contradictory results. Our data are in line with those reported by Shim et al. [12] who found similar increases in activation of the triceps brachii (+600%) and latissimus dorsii muscles (+59%), which are involved in backward arm extension. Two other studies [4,5] reported a 3-fold higher increase in triceps brachii muscle activation during NW as compared to conventional walking. Discordant results have been reported for biceps brachii muscle activation: Shim et al. (10) found a very high increase in biceps brachii muscle activation (+632%), whereas Schiffer et al. [7] found no difference in biceps brachii muscle activation between conventional walking and NW. In their study, Schiffer et al. demonstrated, however, that loading the pole increases activation of the biceps brachii muscle, confirming its role in pole recovery during the arm swing phase. The same conclusion on the role of the biceps brachii muscle was drawn from the observation that simply carrying poles but not using them causes a similar increase in biceps brachii muscle activation as when using them (about +50%). The authors suggested that recovery of the mass of the poles was solely responsible for the higher activation of the biceps brachii muscle [11].

Smaller increases in muscle activation were also found for walking with hiking poles (9,16). The increase in muscle activation was roughly 50% for the biceps brachii, 70% for the deltoideus anterior, 100% for the latissimus dorsii, and 150% for the triceps brachii muscles during uphill walking with poles [11] and 300% for the triceps brachii muscle when using a walking pole while carrying a heavy backpack [26]. The lower response reported in the last-mentioned study may have been due to a less intense use of walking poles and to smaller range of movement.

NW was originally developed as summer training exercise for cross-country skiers because it resembles the locomotion pattern of the diagonal stride technique. To date, no direct comparison of muscle activation during these two types of locomotion has been carried out in this context. Studies investigating upper body muscle activity demonstrated considerable trunk and arm muscle activation in double poling cross-country skiing [13,14], with activation exceeding the level reached during maximal voluntary contraction at high-velocity exercise [13]. Comparing muscle-activation patterns in cross-country skiing and NW is difficult owing to the differences in speed of locomotion and the particular joint movement patterns of double poling. What can be hypothesised is that muscular activity is lower in NW than in cross-country skiing.

The only difference we noted in lower limb muscle activation when the subjects walked with poles was a slight but significant reduction in activation of the gastrocnemius lateralis muscle, which is activated during the push-off phase to propel the body forward. This may be because the force exerted through the poles relieves part of the propulsion achieved with the leg. Shim et al. [12] found no differences in lower body muscle activation when walking poles were used. In contrast, another study [5] reported significantly lower activation of the vastus lateralis and gastrocnemius muscles at all walking speeds investigated and of the biceps femoris muscle at a very high walking speed. However, because the participants in the study by Shim et al. were not experienced in NW, their poling action was probably not such that it could have an effect on lower limb action. On the basis of our and previously published results, we can say that proper pole technique during NW can provide forward propulsion and reduce gastrocnemius muscle action during push off [27].

Different from NW, the relative contribution of the poles to gain forward propulsion in cross-country skiing can range from 20% of total work up to being nearly the only source of propulsion depending on the technique and the conditions considered [22,28]. The amount of force exerted through the pole we noted for NW is comparable to that reported for low-intensity diagonal stride cross-country skiing, and it is 3 times lower than for double poling [14,29]. The larger difference in poling action between NW and cross-country skiing can be ascribed to the shorter poling time in cross-country skiing, resulting in higher power production [14,29].

Nordic Walking Adaptation to Uphill Walking

Additional power from muscle action is needed for uphill walking. We found that uphill walking increased activation of all the leg muscles investigated. This adaptation was similar during both conventional walking and NW and comparable to that previously reported for upslope walking [16,17,18]. The activity of the tibialis anterior muscle, which primarily serves to lift the foot during leg swing during uphill walking, was increased by about 10% between the level and the uphill walking condition. Our observation that the tibialis anterior muscle plays a minor role in sustaining the increased incline is shared by other studies that found small [18] or no [16] changes with grade in ankle flexor activity. A greater adaptation of muscular activity was seen for the gastrocnemius, vastus lateralis, and biceps femoris muscles, with an estimated increase of between 60% and 80% during uphill walking. Based on previous studies investigating the effects of slope on leg muscle activation during conventional walking [16,18], we can hypothesise that the leg extensor muscles in particular are activated to meet the demand required by increased slope. The lack of a significant gait x slope interaction suggests, however, that walking with poles does not increase the work required by the leg muscles.

A marked increase in activation of the biceps brachii, trapezius and erector spinae muscles was noted in the uphill walking condition. The 50% increase in activation of the biceps brachii muscle during uphill NW can be explained by the need to lift the pole before ground contact to a higher point than during level NW. No increase in muscle activity in force exertion or in force exerted through the pole was found during uphill NW. This suggests that pole use played no role in exploiting the extra mechanical work required to overcome gravity for uphill locomotion. This observation is corroborated by the fact that the differences in oxygen consumption between NW and conventional walking were smaller during uphill walking than level walking and that the differences between walking with and without the poles at 15% incline became even smaller. Importantly, the increase in metabolism during NW at 0% and 15% incline was roughly the same in absolute values, suggesting that the contribution from the upper body during NW remains roughly the same during level and uphill walking. Moreover, It has been demonstrated that walking with trekking poles does not increase the energy cost and that muscular activity is redistributed from the lower to the upper limbs [11].

Interestingly, we found a gait x slope interaction for the erector spinae muscle, with greater increases in activation during conventional walking than NW in the uphill condition. The primary function of the erector spinae muscle is to control trunk flexion during gait [30]. However, the epaxial muscles are thought to serve different and somehow conflicting functions in mammalian terrestrial locomotion [31]. Besides mobilising the trunk, and so contribute to propulsion through the production of mechanical work, they also dynamically stabilize the trunk by counteracting movements that are passively induced by external forces or actively produced by antagonist muscles. In uphill trotting dogs, for example, elevated epaxial muscle activity works to stabilise the pelvis against the increased moment imposed by the hindlimb retractor muscles and assists in the production of lateral bending of the trunk [31].

Similar muscular adaptations to meeting increased work demand are seen in stair climbing in humans. When muscle activation patterns during level walking and stair ambulation were compared, the erector spinae muscle was found to help the climbing limb swing over the next step, thus elevating the pelvis on this side [32]. Two different mechanisms may underlie the decreased activity of the erector spinae muscle with pole use. First, adaptation to uphill walking involves increasing trunk tilt [33], which increases activation of the erector spinae muscle [34]. A study on uphill backpacking [26] demonstrated that peak trunk flexion and extension velocities were reduced when hiking poles were used. The authors reported that pole use had a significant effect on reducing trunk movement in the sagittal plane, with less need to actively control trunk flexion by engaging the erector spinae muscle. Second, the activity of the erector spinae muscle during uphill walking with poles may be decreased due to the propulsive work done by the upper body, which reduces the work of the hip retractor muscles.