When walking at the given speed and slope, NW had higher exercise intensity than W. The V̇O2 (+ 15.8%), V̇CO2 (+ 17.0%), V̇E (+ 17.0%), RR (+ 18.2%), SBP (+ 7.7%), DBP (+ 6.9%), and HR (+ 8.4%) were significantly higher in NW than in W in all walking stages. The activities of the upper limb muscles (DEL, BB, and TB) were significantly higher in NW than in W. No significant differences were found between the two walking conditions in the lower extremity muscles, except the TA and GCM muscles during slow level walking.
In this study, we derived new energy estimation equations of NW and W from the measured V̇O2 and walking speed and slope: EW = 4.4 + 0.09 × speed (m·min−1) + 1.20 × speed × fractional grade; ENW = 6.1 + 0.09 × speed (m·min−1) + 1.19 × speed × fractional grade. The constant value in the final equation for W was 4.4 mL·kg−1·min−1, which is somewhat higher than the ACSM equation's known resting oxygen consumption of 3.5 mL·kg−1·min−1 . The coefficient for speed was 0.09, which is similar to the known value of 0.1 .
In the previous studies, the energy consumption of NW was presented as a percent difference compared to the consumption of W. The result varied depending on the walking speed and the slope. In our results, NW consumed more oxygen than walking, and the difference of V̇O2 between NW and W was rather constant than proportional. The difference between constants of the two formulae for NW and W was 1.7 mL·kg−1·min−1 in multiple linear regression analysis derived from speed and speed × grade.
The coefficient for walking speed × grade was 1.20 and smaller than ACSM’s coefficient. The ACSM regression equations developed to estimate oxygen uptake have known limitations that lead to overestimation of energy expenditure, particularly at higher work rates . Kokkinos et al. recently developed a new energy equation for walking from the Fitness Registry and the Importance of Exercise National Database [FRIEND] and suggested small cofficient value of 0.79 for walking speed × grade: E (mL·kg− 1·min− 1) = 3.5 + 0.17 × speed (m·min− 1) + 0.79 × speed × fractional grade [FRIEND equation] .
In our study, as the slope of the uphill increased (from 0–7%) at a constant speed (5 km·h− 1), the amount of oxygen consumption increased in both NW and W. However, the difference (ΔV̇O2) between NW and W was constant (1.7 to 2.3 mL·kg−1·min−1); thus, %V̇O2 decreased from the highest (17.4%) to the lowest (10.9%). Previous studies using a treadmill were reviewed to compare the uphill oxygen consumption patterns. Figard-Fabre et al.  measured oxygen consumption at 0% and 5% grade when walking at 4 km·h−1. Oxygen consumption was higher at NW; the %V̇O2 was 16% at 0% grade and reduced to to 12% at 5% grade. This value is consistent with the value for our Eq. (4 km·h− 1, 0%: 10.4 mL·kg−1·min−1 for W, 12.1 mL·kg−1·min−1 for NW, 16% for %V̇O2; 4 km·h− 1, 5%: 14.4 mL·kg− 1·min−1 for W, 16.1 mL·kg− 1·min−1 for NW, 12% for %V̇O2). Pellegrini et al.  measured oxygen consumption at 15% grade when walking at 4 km·h− 1, and %V̇O2 was 6.9%. This value is consistent with the value for our Eq. (4 km·h− 1, 15%: 22.4 mL·kg− 1·min− 1 for W, 24 mL·kg− 1·min− 1 for NW, 7% for %V̇O2).
On the other hand, the oxygen consumption results of studies conducted in an environment other than the treadmill (outdoor field study) showed a difference from the predicted value of our formula [4, 6]. Church et al.  reported that the average amounts of oxygen consumption at average self-selected walking speeds of 5.6 km·h− 1 (male participants) and 5.9 km·h− 1 (female participants) were overall 13.9 mL·kg− 1·min− 1 during W and 16.7 mL·kg− 1·min− 1 during NW. The estimated V̇O2 values using our equation were 12.8 and 13.3 mL·kg−1·min−1 for W and 14.5 and 15.0 mL·kg−1·min−1 for NW. The oxygen consumption values measured in the field study  were greater than our predicted values, particularly the V̇O2 during NW was higher than the predicted value. Regarding the field test, terrain characteristics would differ from a treadmill, potentially causing a difference in poling force or muscle activity. The difference in walking terrain leads to a difference in oxygen consumption during NW . According to a study by Schiffer et al. , the oxygen consumption during NW was significantly increased in a naturally grown soccer lawn than concrete.
The increase in exercise intensity and oxygen consumption by NW is due to the increased upper limb muscle activity. The upper extremity's increased activity was identified through the surface EMG signals. Walking with upper body exercise  or the use of hand weight  also increases the V̇O2. Oxygen consumption was significantly increased when walking with the weight in hand compared with walking at the same speed. When walking with 1.36 kg of weight held in each hand, it increased by about 3.5 mL O2·kg−1·min−1 compared with walking without weight, which was constant even at different walking speeds . The degree of increase in oxygen consumption increases further as the weight of the weight increases,, however, it is difficult to say that lifting heavy weights reflects the increase in oxygen consumption. According to Owens et al. , a significant increase in oxygen consumption was not observed using a 2.27 kg weight at walking speed (4.8 and 6.4 km·h−1), whereas a significant increase in oxygen consumption at running speed (8.0 and 9.6 km·h−1) was observed. Both the movement of the arm used for walking and the weight are potentially related to the increase in the amount of oxygen consumed. Regarding walking while exercising the upper limbs, the amount of oxygen consumption significantly increased compared with normal walking . The increase in oxygen consumption at NW seems to contribute to an increase in the activity of the upper limb muscles.
In our study, oxygen consumption increased significantly in NW compared with W, but the subjective difficulty, the RPE, did not show a significant difference between NW and W at level walking and rather lower RPE with NW during uphill walking. Previous studies showed higher RPE , lower RPE , or no difference  with pole walking compared with W. During uphill walking, decreased RPE with NW was reported in previous studies with pole walking [7, 19]. According to a study comparing the RPE of NW and W on downhill, uphill, and level walking, the RPE in NW on uphill walking was significantly decreased compared with W .
The participants of our study were all beginners with no Nordic pole experience and all were young males. The proficiency of NW methods can affect gait technique; however, even after a 4-week NW training three times a week, the difference in oxygen consumption between NW and W did not significantly change before and after the training . Moreover, the activation pattern of the surface electromyography of our results was similar to the NW-skilled participants'. In a result of surface electromyography analysis in participants who were experienced in NW, the activities of the BB, TB, and deltoid anterior in NW were significantly increased compared with those in W, which was significant in both level and uphill . Furthermore, the activities of the lower extremity muscles, such as the TA, gastrocnemius lateralis, and VL did not show a significant difference in NW and W, which was the same in level and uphill . These muscle activation patterns are similar to those found in our study. According to a previous study, the changes in the aspects of respiratory gas analysis did not differ by sex .
We proposed a prediction formula for oxygen consumption in NW; however, this is a formula derived only from a non-fast walking speed (3–5 km·h− 1) and a slope within 7% and may differ from the predicted values in a faster or higher slope. NW energy consumption could also be influenced by several factors other than walking speed and slope grade. Energy consumption during NW is known to be affected by the ground's condition . When a relatively short pole was used, energy consumption was increased in the uphill than in the case of a normal length pole . By contrast, pole weight does not appear to have a significant effect on energy consumption . As a result of comparing W and four different types of NW, muscle activity and metabolic response were different according to the type of NW, but all types of NW showed higher metabolic response and muscle activity than W .