Carbohydrate and fat oxidation in persons with lower limb amputation during walking with different speeds

Abstract

Background:

Studies suggest that the energy expenditure of healthy persons (control) during walking with the preferred walking speed in steady-state conditions is dominated by fat oxidation. Conversely, carbohydrate and fat oxidation during walking is little investigated in transfemoral amputees.

Objectives:

To investigate carbohydrate and fat oxidation, energy cost of walking, and percent utilization of maximal aerobic capacity $(% {\overset{\cdot}{V} O}_{2} \max)$ during walking.

Study design:

Eight transfemoral amputees and controls walked with their preferred walking speed and speeds 12.5% and 25% slower and faster than their preferred walking speed.

Methods:

Energy expenditure and fuel utilization were measured using a portable metabolic analyzer. Metabolic values are means ± standard deviation.

Results:

For transfemoral amputees (37.0 ± 10.9 years) and controls (39.0 ± 12.3 years), fat utilization at the preferred walking speed was 44.8% ± 7.2% and 45.0% ± 7.2% of the total energy expenditure, respectively. The preferred walking speed of the transfemoral amputees and controls was close to a metabolic cross-over speed, which is the speed where carbohydrate utilization increases steeply and fat utilization decreases. When walking fast, at 90 m min⁻¹ (preferred walking speed plus 25%), transfemoral amputees utilized 70.7% ± 5.6% of their ${\overset{\cdot}{V} O}_{2} \max$ , while the controls utilized 30.9% ± 4.5% (p < 0.001) at the matching speed (control preferred walking speed). At 90 m min⁻¹, carbohydrate utilization was 78% ± 4.7% and 55.2% ± 7.2% of the total energy expenditure for the transfemoral amputees and controls, respectively (p < 0.01). Compared to the control, energy cost of walking was higher for the transfemoral amputees at all speeds (all comparisons; p < 0.001).

Conclusion:

At the preferred walking speed, carbohydrate, not fat, dominates energy expenditure of both transfemoral amputees and controls. For the transfemoral amputees, consequences of fast walking are very high ${\overset{\cdot}{V} O}_{2} \max$ utilization and rate of carbohydrate oxidation.

Clinical relevance

Research on the relationships between physical effort and fuel partitioning during ambulation could provide important insights for exercise-rehabilitation programs for lower limb amputees (LLA). Regular endurance exercise will improve maximal aerobic capacity and enable LLA to walk faster and at the same time expend less energy and improve fat utilization.

Keywords

Metabolism fat utilization carbohydrate utilization

Background

Walking is the most common form of exercises and may for many people be the only break in an otherwise sedentary life.¹ Following a lower limb amputation and the resulting walking disability, persons often adopt a very sedentary lifestyle² which may, over time, further reduce aerobic power, physical fitness, and walking speed. During level walking, the rate of oxygen uptake relates to the walking speed³; hence, measurements of the rate of oxygen uptake $({\overset{\cdot}{V} O}_{2})$ during prosthetic walking are an important tool for assessing the energetic consequences of walking disabilities. Previous research has shown that there may exist an individual, optimal walking speed with regard to minimal energy expenditure,^3–5 and measurement of walking economy at this preferred walking speed (PWS) is frequently used as an indicator of overall gait performance of prosthetic walkers.^6,7 It is well known that the PWS of persons with a lower limb amputation is slower compared to healthy age-matched individuals,^8,9 and the walking economy (oxygen uptake per meter traveled) at the PWS is also substantially higher for persons with a lower limb amputation.¹⁰ The higher walking economy following lower limb loss is related both to the level⁶ and etiology of amputation,¹⁰ but it is argued that this could, in part, be caused by the fact that persons with lower limb amputation cannot reach their optimal (most economical) walking speed.¹¹ Consequently, this study aims to explore the impact of different walking speeds on the walking economy and energy expenditure of both healthy persons and persons with lower limb loss.

Moreover, there are very few studies on the relation between walking speed and fuel partitioning during prosthetic ambulation, but previous studies on healthy persons have demonstrated that fat is the preferred energy substrate when walking with the PWS.¹² The quantity of energy from carbohydrate (CHO) stores of the body is only 1% of that available in fat,¹³ and the rate of CHO oxidation increases with increasing physical effort.^14,15 Consequently, it is suggested that conservation of CHO energy reserves, rather than walking economy per se, governs the selection of a PWS, and that central nervous system (CNS) selects a PWS that is supported mainly by fat oxidation.¹² Except for one previous study, looking at fuel utilization at the PWS during treadmill walking,¹⁶ data on CHO and fat oxidation during prosthetic walking are virtually absent from the literature.

Thus, there is a need to investigate how overground walking with different speeds affect CHO and fat oxidation rates and walking economy of persons with a unilateral transfemoral amputation (TFA) and healthy, age-, and sex-matched individuals.

The main hypotheses are that persons with a TFA have a higher rate of CHO oxidation at similar relative speeds compared to healthy persons. In addition, we hypothesize that differences in walking speeds will have little effect on the walking economy, but that the physical effort, quantified as percent utilization of the maximal oxygen uptake $(% {\overset{\cdot}{V} O}_{2} \max)$ , will differ substantially between TFA and healthy persons.

Methods

Participants

Two groups of participants were recruited to this study. The participants of the TFA group were eight, non-smoking adults (50 % females) with unilateral TFA for other reasons than vascular diseases and no comorbidities. Causes of amputations were as follows: trauma (n = 1), cancer (n = 5), congenital (n = 1), and infection (n = 1). The TFA participants had in average (±standard deviation (SD)) used their prosthesis for 15.9 ± 13.9 (range: 3–39) years. Five persons had a microcontroller-assisted knee joint, while three persons used hydraulic-controlled knee joints, and all TFA participants used their prosthesis on a daily basis. The average weight of the prostheses was 3.80 ± 0.5 kg.

The participants in the control (CON) group were eight healthy, non-smoking adults (50% females) with no orthopedic problems and with similar weight, height, and age as the TFA. Exclusion criteria for both groups were use of medication that could affect heart rate or energy expenditure (i.e. beta-blockers and thyroid hormone replacements). Daily walking distance of both the TFA and CON was assessed by a self-report form, and inclusion criteria was that the participants were able to walk continuously for at least 500 m. Written informed consent was obtained from all subjects and the study was approved by the Regional Committees for Medical and Health Research Ethics in Norway.

Study design

The participants were instructed to avoid exercise and alcohol 24 h prior to testing and to abstain from coffee and tea on the day of testing. The TFA and CON reported to the laboratory in the morning, 2 h after eating a low-fat breakfast (bread, jam, sliced ham, juice, low-fat milk) and were subsequently instrumented for collection of expired air. During all walk trials, the ${\overset{\cdot}{V} O}_{2}$ consumption and VCO₂ production were measured breath-by-breath with a validated¹⁷ and portable metabolic analyzer (Metamax 3B; Cortex Biophysik, Leipzig, Germany). Heart rate was recorded beat-by-beat (Polar, Kempele, Finland) and interfaced with the metabolic analyzer. During trials, participants walked with their PWS and with speeds that were 25% and 12.5% lower and higher than their respective PWS. The sequence of walking speeds was determined for each individual by having an independent person randomly selecting between five closed envelopes, each containing a note specifying one of five specific walking speeds. Each walking trial (speed) lasted 7 min, and data reported on physiological measurements and metabolic calculations are average values over the last 2 min of each walking interval. Each walking trial was interspaced by rest intervals of 2 min where the participants sat quietly on a chair. The walking trials were performed around a 40-m oblong indoor course, and the walking speed was monitored by a 5-m-long optical gait analysis system (OptoGait; Microgate, Bolzano-Bozen, Italy). Prior to walking trials, the PWS was determined by having the participants walk a stretch of 10 m, with the speed measured by the OptoGait system during the last 5 m. This sequence was repeated twice and averaged. During trials, walking speed was measured twice for each 40-m round, and if necessary, verbal instructions, such as “walk a little slower/walk a little faster/keep the pace,” were given to the participants in order to adjust their speed. Furthermore, the maximal aerobic capacity $({\overset{\cdot}{V} O}_{2} \max)$ of the participants was determined according to the previous protocols¹⁸ on a separate occasion, 1–2 weeks earlier than the walking trials. In short, the participants walked on a treadmill (Woodway ELG 70; Woodway, Waukesha, WI, USA) with constant speed but with progressively increasing inclinations until volitional fatigue. The VO₂ measurements were considered maximal when the oxygen uptake did not increase >2 mL min⁻¹ kg⁻¹ (plateau in VO₂) despite increasing workload and with respiratory exchange ratio (RER) values >1.05.¹⁹

Gas exchange and energy expenditure measurements

The oxygen analyzer was calibrated for barometric pressure, and gas was calibrated with a reference gas mixture of 16% O₂ and 4% CO₂. The calibration was then verified with measurements of ambient air, according to the manufacturer’s instructions. In addition, a volume calibration was performed using a standardized 3-L syringe (Hans Rudolph, Shawnee, KS, USA).

VO₂, VCO₂, lung ventilation, heart rate, and RER values were continuously monitored during testing by telemetry in real time to verify steady-state conditions during walking trials. The RER is the ratio between the carbon dioxide production and the oxygen consumption, and all walking sessions were completed with RER values <1.0. CHO and fat oxidation was calculated by indirect calorimetry using standard methods.²⁰ Protein oxidation was assumed to be insignificant during these walking trials.²¹ The energy cost of walking (ECW), that is, the oxygen consumption per unit distance (mL kg⁻¹ m⁻¹), was calculated by dividing the participants ${\overset{\cdot}{V} O}_{2}$ consumption (mL kg⁻¹ min⁻¹) by their respective walking speed (m min⁻¹).

Statistics

Independent t tests were used to compare the TFA and CON for physical characteristics. A two-way mixed analysis of variance (ANOVA) was used to test if oxygen uptake ( ${\overset{\cdot}{V} O}_{2}$ , mL min⁻¹ kg⁻¹), $% {\overset{\cdot}{V} O}_{2} \max$ utilization, CHO and fat oxidation rates (cal kg⁻¹ min⁻¹), and walking economy (VO₂, mL kg⁻¹ m⁻¹) differed across walking speeds. Post hoc comparisons with Bonferroni corrections were conducted in case of a significant ANOVA. Specifically, the values at the PWS were compared to values at each of the other walking speeds. Data were tested for normality by the Shapiro–Wilk test. In those instances where the sphericity assumption was violated, Greenhouse–Geisser adjustments of the p values were reported. The criterion level for significance was set at p < 0.05. The effect size was evaluated with $η_{p}^{2}$ (partial eta-squared), where 0.01 < $η_{p}^{2}$ < 0.06 constitutes a small effect, 0.06 < $η_{p}^{2}$ < 0.14 constitutes a medium effect, and $η_{p}^{2} > 0.14$ constitutes a large effect.²² Pearson’s correlation was used to investigate the relationship between the pre-determined walking speeds and the actual measured walking speeds of the TFA and CON group. IBM SPSS Statistics for Windows, version 24.0 (IBM Corp., Armonk, NY, USA) was used for all statistical analyzes. Results are presented as mean ± SD or means and confidence intervals (CI).

Results

Physical characteristics of the participants

The mean ± SD age, height, weight, and body mass index of the TFA and CON were 37.0 ± 10.9 and 39.0 ± 12.3 years, 175.5 ± 4.6 and 170.0 ± 7.4 cm, 73.6 ± 10.4 and 72.7 ± 14.2 kg, 23.8 ± 2.7 and 25.2 ± 3.3 kg/m², respectively. The weight of the TFA is including their prosthesis. There were no statistical differences in physical characteristics between the two groups. The maximal aerobic capacities $({\overset{\cdot}{V} O}_{2} \max)$ of the TFA and CON were 30.6 ± 8.7 and 48.9 ± 14.4 mL min⁻¹ kg⁻¹ (p < 0.05), respectively. Mean daily, self-reported walking distance was 2187 ± 923 and 2688 ± 834 m for the TFA and CON (p = 0.243). All TFA participants reported that they were able to walk at least 500 m continuously.

Walking speed

All participants walked with their PWS and speeds 12.5% and 25% slower and faster than their respective PWS, thus all relative walking speeds were similar for the TFA and CON. In terms of actual walking speed in m min⁻¹, mean PWS of the TFA and CON were 73.2 ± 11.1 and 91.3 ± 8.8 m min⁻¹ (TFA vs CON; p < 0.001). The range of walking speeds from the slowest to the fastest walking speeds (PWS minus 25% to PWS plus 25%) was 54.8 ± 9.7 − 90.4 ± 13.2 and 69.0 ± 6.3 − 114.4 ± 10.9 m min⁻¹ for the TFA and CON, respectively (TFA vs CON; all comparisons, p < 0.001). Actual walking speeds were monitored by the OptoGait system (described in the “Methods” section), and there was a close correlation between OptoGait measurements and the pre-determined (calculated) walking speeds. For the TFA group, correlation coefficients for the measured and calculated walking speeds of PWS minus 25% and 12.5%, the PWS, and the PWS plus 12.5% and 25% were as follows: 0.997 (p < 0.001), 0.992 (p < 0.001), 0.994 (p < 0.001), 0.997 (p < 0.001), and 0.998 (p < 0.001), respectively. For the CON group, the correlation coefficients for the same speeds were as follows: 0.984 (p < 0.001), 0.989 (p < 0.001), 0.985 (p < 0.001), 0.993 (p < 0.001), and 0.997 (p < 0.001), respectively.

Oxygen uptake

The oxygen uptake (mL min⁻¹ kg⁻¹) following walking with different speeds is shown in Table 1. There was no significant interaction between group*time F(4, 56) = 0.572, p = 0.684, $η_{p}^{2} = 0.039$ , but there was a significant main effect of time F(4, 56) = 89.537, p < 0.001, $η_{p}^{2} = 0.865$ upon oxygen uptake. There was no significant main effect of group on oxygen uptake F(1, 14) = 1.027, p = 0.328, $η_{p}^{2} = 0.068$ ; hence, the oxygen uptake for the TFA and CON was similar at all speeds. Mean group difference (CI) for oxygen uptake was 1.403 (−1.566 to 4.371).

Table 1.

Oxygen uptake and percent utilization of ${\overset{\cdot}{V} O}_{2} \max$ at different walking speeds.

Relative speeds	Group	Oxygen uptake, mL min⁻¹ kg⁻¹		Oxygen uptake, $% {\overset{\cdot}{V} O}_{2} \max$
Relative speeds	Group	Means	95% CI	Means	95% CI
PWS minus 25%	CON	10.5***	9.10–11.8	22.8**	15.2–30.6
	TFA	12.2***	10.9–13.5	42.3***	34.7–49.9
PWS minus 12.5%	CON	11.7*	9.5–14.1	25.5	17.8–33.2
	TFA	13.6*	11.2–15.9	46.4*	38.8–54.1
PWS	CON	14.1	11.9–16.2	30.9	21.3–40.5
	TFA	15.9	13.8–18.1	54.5	44.9–64.1
PWS plus 12.5%	CON	16.7^#	13.8–19.7	36.7	25.9–47.6
	TFA	17.8	14.8–20.7	61.0^#	50.1–71.8
PWS plus 25%	CON	20.0^###	17.4–22.6	44.2^###	32.1–56.3
	TFA	20.6^###	17.9–23.2	70.7^###	58.2–82.7

PWS: preferred walking speed; CI: confidence interval; CON: control.

p < 0.05; **p < 0.01; ***p < 0.001; PWS compared to lower speeds.

p < 0.05, ^###p < 0.001; PWS compared to faster speeds.

Pairwise post hoc comparisons (for time) with Bonferroni corrections showed that for the TFA, the oxygen uptake at the PWS was significantly higher compared to PWS minus 25% (p < 0.001) and PWS minus 12.5% (p < 0.05) and lower compared to PWS plus 25% (p < 0.001). For the CON, the oxygen uptake at the PWS was significantly higher compared to PWS minus 25% (p < 0.001) and PWS minus 12.5% (p < 0.05) and lower compared to PWS plus 12.5% (p < 0.05) and PWS plus 25% (p < 0.001).

$% {\overset{\cdot}{V} O}_{2} \max$ utilization

There was no significant interaction between group*time F(4, 56) = 1.406, p = 0.244, $η_{p}^{2} = 0.091$ , but there was a significant main effect of time F(4, 56) = 75.747, p < 0.001, $η_{p}^{2} = 0.844$ and group F(1, 14) = 14.259, p < 0.01, $η_{p}^{2} = 0.505$ upon $% {\overset{\cdot}{V} O}_{2} \max$ . Mean group difference (CI) for $% {\overset{\cdot}{V} O}_{2} \max$ utilization was 22.936 (9.909–35.964) (Table 1).

Pairwise post hoc comparisons (for time) with Bonferroni corrections showed that for the TFA, the ${\overset{\cdot}{V} O}_{2} \max$ utilization at the PWS was significantly higher compared to PWS minus 25% (p < 0.001) and PWS minus 12.5% (p < 0.01) and lower compared to PWS plus 25% (p < 0.001). For the CON, the ${\overset{\cdot}{V} O}_{2} \max$ utilization at the PWS was significantly higher compared to PWS minus 25% (p < 0.01) and lower compared to PWS plus 25% (p < 0.001).

ECW

There was no significant interaction between group*time F(4, 56) = 1.107, p = 0.362, $η_{p}^{2} = 0.073$ , but there was a significant main effect of time F(4, 56) = 3.180, p < 0.05, $η_{p}^{2} = 0.185$ and group F(1, 14) = 29.873, p < 0.001, $η_{p}^{2} = 0.681$ upon the ECW (Figure 1).

Figure 1.

Energy cost of walking (ECW) of the TFA and CON during walking with similar relative speeds. Values are represented as mean ± SD. ***p < 0.001, TFA compared to CON.

Across the different walking speeds, the range of ECW values for the TFA and the CON were 0.213–0.226 and 0.146–0.174, respectively. Hence, since oxygen uptake and walking speed change more or less in parallel, the within-group ECW values show only small changes across the range of walking speeds. Pairwise post hoc comparisons with Bonferroni corrections showed that the ECW of the CON was significantly lower compared to the TFA at the all speeds (all comparisons, p < 0.001). For both the TFA and CON, the ECW at the PWS was similar to the ECW at the other walking speeds.

CHO and fat oxidation rates

For CHO, there was no significant interaction between group*time F(4, 56) = 0.576, p = 0.681, $η_{p}^{2} = 0.039$ , but there was a significant main effect of time upon CHO oxidation rates F(4, 56) = 41.225, p < 0.001, $η_{p}^{2} = 0.746$ (Figure 2). There was no significant main effect of group on CHO oxidation rates F(1, 14) = 0.477, p = 0.328, $η_{p}^{2} = 0.037$ .

Figure 2.

Fuel oxidation rates of the TFA and CON during walking on the floor with different speeds. Values are represented as mean ± SD. ***p < 0.001, CHO oxidation rates at the PWS of the TFA compared to other speeds. ^###p < 0.001, CHO oxidation rates at the PWS of the CON compared to other speeds.

At similar relative walking speeds, the mean difference in CHO oxidation rates between the TFA and CON was small and varied between 1 and 10 cal kg⁻¹ min⁻¹. For both the TFA and CON, the CHO oxidation rates at the PWS was significantly higher compared to PWS minus 25% (both, p < 0.05) and lower compared to oxidation rates at PWS plus 25% (p < 0.001).

For fat, there was no significant interaction between group*time F(4, 56) = 0.858, p = 0.495, $η_{p}^{2} = 0.085$ , no any significant main effect of time F(4, 56) = 1.304, p = 0.280, $η_{p}^{2} = 0.085$ or group F(1, 14) = 0.023, p = 0.881, $η_{p}^{2} = 0.002$ upon fat oxidation rates.

At similar relative walking speeds, the fat oxidation rates of the TFA and CON were quite similar, and the mean difference in fat oxidation rates was in the order of 2–7 cal kg⁻¹ min⁻¹. Fat oxidation rates are shown in Figure 2.

Discussion

This study compared walking economy, percent utilization of the ${\overset{\cdot}{V} O}_{2} \max$ , and fuel utilization of healthy individuals and persons with lower limb amputation across a wide range of walking speeds. Our data show that the ECW of both the TFA and CON is virtually similar over the range of walking speeds investigated (Figure 1). The explanation for this is that as the walking speed changes, the oxygen uptake changes in parallel; thus, the ECW remains stable for both groups. The TFA, however, have a significantly higher energy cost compared to the CON. As shown in Table 1, the oxygen uptakes of the TFA and CON are similar at similar relative walking speeds. Consequently, since the ECW is calculated as the oxygen uptake per meter traveled, the differences between lower limb amputees and healthy persons are chiefly the result of the much lower walking speed of the TFA. Based on this assumption, it is our opinion that the ECW does not provide much unique information about the effort of prosthetic ambulation that is useful in a clinical setting. We suggest that using calculations of percent utilization of the individual maximal aerobic capacity $(% {\overset{\cdot}{V} O}_{2} \max)$ is a better way of describing the physical effort of prosthetic gait than the ECW.

Our results also show that CHO and fat oxidation rates are similar for the TFA and CON when walking with similar relative speeds in the fed state. Thus, in the fed state, there is little support for the hypothesis that the CNS naturally selects a walking speed that requires little or no net CHO depletion.¹² Generally, this study shows that in the fed state, the PWS of the TFA is a metabolic cross-over speed,²³ meaning that when the TFA walking speed exceeds the PWS, the CHO oxidation rates increases more steeply and there is a concomitant reduction in the fat oxidation rates (Figure 2). A similar response is seen for the CON, and at the same relative speeds, the curves are virtually overlapping with the TFA curves. However, at the PWS, the TFA and CON walk at an absolute speed of 73 and 91 m min⁻¹; thus, the metabolic cross-over speed of the CON is at a higher absolute speed compared to the TFA. Furthermore, while the oxygen consumption at the PWS is similar (Table 1), the TFA utilized about 55 % of their ${\overset{\cdot}{V} O}_{2} \max$ , while the CON used about 31 % of their ${\overset{\cdot}{V} O}_{2} \max$ . Thus, the physical effort of walking at the PWS is substantially greater for the TFA compared to the CON. Thus, when healthy persons walk at a leisurely pace (i.e. ~90 m min⁻¹), persons with a lower limb amputation struggle to keep the same pace because in order to do so, they will need to increase their speed by about 25%. The consequences of increasing the speed to 90 m min⁻¹ (PWS plus 25%) are that the TFA utilized 71% of their ${\overset{\cdot}{V} O}_{2} \max$ (Table 1) and close to 80% of the total energy expenditure was provided by CHO oxidation (Figure 2). In contrast, at the same absolute speed, the CON utilized only 31% of their ${\overset{\cdot}{V} O}_{2} \max$ and about 55% of their energy expenditure was provided by CHO oxidation. At 90 m min⁻¹, the oxygen uptakes are similar for the TFA and CON, and the ECW for both groups is unchanged compared to the PWS. This underscores that the ECW is not a good indicator of the physical effort of prosthetic ambulation.

As can be observed in Figure 2, there is a great reliance on CHO oxidation during fast walking, and it is well established that endurance capacity and CHO availability are highly interrelated²⁴; hence, the size of the CHO stores in the body may be of special importance for persons with a lower limb amputation. Pertaining to this, persons with a TFA evidently have less lower extremity muscle mass than healthy persons and many are untrained and have low aerobic capacity.^16,25 The CHO (glycogen) stores in skeletal muscles is limited by the quantity of muscle mass and the intrinsic glycogen storage capacity of muscle tissue (80–150 mmol kg⁻¹ wet weight).²⁶ However, skeletal muscle of untrained individuals may store only 50% of the capacity of trained skeletal muscle.^27–29 Thus, it is plausible that an untrained muscle mass and corresponding low intramuscular CHO stores may considerably limit the endurance capacity and walking range of persons with lower limb amputation.

Furthermore, it is important to note that the rates of fat oxidation in the study of Willis et al.¹² are higher than the fat oxidation rates in this study. In the above study, fat accounted for about 65% of the total energy expenditure during walking with the PWS, and in a similar study of persons with post-stroke hemiparesis, fat supplied 58% of the total energy expenditure when walking with the PWS.³⁰ In this study, fat utilization was about 45% of the total energy expenditure for TFA and CON when walking with the PWS; hence, fat combustion was substantially lower than in the studies of Willis et al.¹² and Ganley et al.³⁰

The reason for these discrepancies may be related to the fact that the participants in the above studies were tested following an overnight fast, and it is expected that fat oxidation rates are quite high following prolonged fasting (>6 h).³¹ On the other hand, to investigate fuel oxidation during normal living, it is important to study the energy metabolism also in the non-fasted state, as in this study. The following example illustrates the impact of prior feeding on fuel partitioning. In Willis et al.’s¹² study, the total energy expenditure of healthy persons during treadmill walking with the treadmill PWS was 63 cal min⁻¹ kg⁻¹, and CHO and fat utilization accounted for 33% and 67% of the total energy expenditure, respectively. Oxygen uptake was 13.2 mL min⁻¹ kg⁻¹, and heart rate was 95 beats min⁻¹. In a previous study in our lab,³² using the same participants as in this study, we observed that during treadmill walking at the PWS, CON mean oxygen uptake, heart rate, and total energy expenditure were 13.4 mL min⁻¹ kg⁻¹, 92 beats min⁻¹, and 68 cal min⁻¹ kg⁻¹, that is, similar to Willis et al.’s¹² study. However, CHO and fat oxidation rates were 64% and 36% of the total energy expenditure, respectively, and consequently, the fuel partitioning in the fed state (this study) was reversed with regard to the fasted state.¹² This is, however, not unexpected as feeding suppresses fat oxidation during physical activity.³¹ Thus, to enable persons with a lower limb amputation to adopt a faster walking speed and at the same time to expend less energy, it is of importance that these persons perform aerobic endurance exercise. Regular endurance exercise will improve the maximal aerobic capacity $({\overset{\cdot}{V} O}_{2} \max)$ over time and switch substrate utilization toward fat oxidation.³³ This may result in a clinically relevant reduction in the relative oxygen uptake during prosthetic ambulation and a more sustainable substrate utilization. Consequently, this may possibly translate into functional improvements in walking speed, walking endurance, and a reduction in the perception of the physical effort. The interrelationship between improved physical work capacity and gait performance is little investigated in lower limb amputees, but Wezenberg et al.⁹ have developed a predictive quantitative model that shows that even small increases in aerobic capacity can result in substantial improvements in walking ability of older adults with a lower limb amputation.

Conclusion

In the fed state, the PWS of both the TFA and CON is close to a metabolic cross-over walking speed, above which CHO oxidation rates increases steeply. Thus, when the TFA walking speed exceeds the PWS, there is an increasing reliance on CHO oxidation and a concomitant reduction in the fat oxidation rates. The relative oxygen uptake (VO₂, mL min⁻¹ kg⁻¹) did not differ between groups across the range of walking speeds. Within each group, the ECW did not vary across walking speeds. In contrast, the $% {\overset{\cdot}{V} O}_{2} \max$ differed significantly across walking speeds for both groups. The $% {\overset{\cdot}{V} O}_{2} \max$ may be a better indicator of the physical effort of prosthetic ambulation than the ECW.

Possible limitations

The TFA in this study used different types of knee joints (hydraulic or microprocessor controlled) and one may speculate if this in any way could affect measures of ECW or oxygen uptake values. There is no indication in our data that differences in knee joint construction affect these measurements. Nonetheless, it may be prudent to conduct further studies on this matter with a larger number of participants to investigate the relation between walking speed and fuel selection during prosthetic walking.

Footnotes

Acknowledgements

The authors thank the participants for their patience and effort during the walking experiments.

Author contribution

All authors contributed equally in the preparation of this manuscript

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Passmore

Durnin

JV.

Human energy expenditure. Physiol Rev 1955; 5(4): 801–955.

Deans

Burns

McGarry

et al . Motivations and barriers to prosthesis users participation in physical activity, exercise and sport: a review of the literature. Prosthet Orthot Int 2012; 36(3): 260–269.

Ralston

HJ.

Energy speed relation and optimal speed during level walking. Int Z Angew Physiol 1958; 17: 2772–2783.

Zarrugh

Todd

Ralston

HJ.

Optimization of energy expenditure during level walking. Eur J Appl Physiol 1974; 33: 293–306.

Genin

Bastien

Franck

et al . Effect of speed on the energy cost of walking in unilateral traumatic lower limb amputees. Eur J Appl Physiol 2008; 103: 655–663.

Traballesi

Porcacchia

Averna

et al . Energy cost of walking measurements in subjects with lower limb amputations: a comparison study between floor and treadmill test. Gait Posture 2008; 27: 70–75.

Teixeira-Salmela

Nadeau

Milot

M-H

et al . Effects of cadence on energy generation and absorption at lower extremity joints during gait. Clin Biomech 2008; 23: 769–778.

Waters

Perry

Antonelli

et al . Energy cost of walking of amputees: the influence of level of amputation. J Bone Joint Surg Am 1976; 58: 42–46.

Wezenberg

Van der Woude

Faber

et al . Relation between aerobic capacity and walking ability in older adults with a lower-limb amputation. Arch Phys Med Rehabil 2013; 94: 1714–1720.

10.

Waters

Mulroy

The energy expenditure of normal and pathologic gait. Gait Posture 1999; 9(3): 207–231.

11.

Tesio

Roi

Möller

Pathological gaits: inefficiency is not a rule. Clin Biomech 1991; 6: 47–50.

12.

Willis

Ganley

Herman

RM.

Fuel oxidation during human walking. Metab Clin Exp 2005; 54: 793–799.

13.

Entin

Gest

Trancik

et al . Fuel oxidation in relation to walking speed: influence of gradient and external load. Eur J Appl Physiol 2010; 110(3): 515–521.

14.

Romijn

Coyle

Sidossis

et al . Regulation of endogenous fat and carbohydrate metabolism in relation to exercise intensity and duration. Am J Physiol 1993; 265(28): E380–E391.

15.

Van Loon

LJC

Greenhaff

Constantin-Teodosiu

et al . The effects of increasing exercise intensity on muscle fuel utilization in humans. J Physiol 2001; 536: 295–304.

16.

Starholm

Gjøvaag

Mengshoel

AM.

Energy expenditure of transfemoral amputees walking on a horizontal and tilted treadmill simulating different outdoor walking conditions. Prosthet Orthot Int 2010; 34(2): 184–194.

17.

Vogler

Rice

Gore

CJ.

Validity and reliability of the Cortex MetaMax3B portable metabolic system. J Sports Sci 2010; 28(7): 733–742.

18.

Gjøvaag

Starholm

Mirtaheri

et al . Assessment of aerobic capacity and walking economy of unilateral transfemoral amputees. Prosthet Orthot Int 2014; 38(2): 140–147.

19.

Aspenes

Lund

TIL

Skaug

E-A

et al . Peak oxygen uptake and cardiovascular risk factors in 4631 healthy women and men. Med Sci Sports Exerc 2011; 43(8): 1465–1475.

20.

Péronnet

Massicotte

Table of nonprotein respiratory quotient: an update. Can J Sport Sci 1991; 16(1): 23–29.

21.

Carraro

Stuart

Hartl

et al . Effect of exercise and recovery on muscle protein synthesis in human subjects. Am J Physiol 1990; 259: E470–E476.

22.

Cohen

Statistical power analysis for the behavioral sciences. 2nd ed. New York: Lawrence Erlbaum Associates, 1988.

23.

Brooks

Mercier

Balance of carbohydrate and lipid utilization during exercise: the “crossover” concept. J Appl Physiol 1994; 76(6): 2253–2261.

24.

Hawley

Burke

LM.

Carbohydrate availability and training adaptation: effects on cell metabolism. Exerc Sport Sci Rev 2010; 38(4): 152–160.

25.

Wezenberg

de Haan

Faber

et al . Peak oxygen consumption in older adults with a lower limb amputation. Arch Phys Med Rehabil 2012; 93: 1924–1929.

26.

Ivy

Katz

Cutler

et al . Muscle glycogen synthesis after exercise: effect of time of carbohydrate ingestion. J Appl Physiol 1988; 64(4): 1480–1485.

27.

Gollnick

Armstrong

Saubert

et al . Enzyme activity and fiber composition in skeletal muscle of untrained and trained men. J Appl Physiol 1972; 33(3): 312–319.

28.

Piehl

Adolfsson

Nazar

Glycogen storage and glycogen synthetase activity in trained and untrained muscle of man. Acta Physiol Scand 1974; 90(4): 779–788.

29.

Roedde

MacDougall

Sutton

et al . Supercom-pensation of muscle glycogen in trained and untrained subjects. Can J Appl Physiol 1986; 11(1): 42–46.

30.

Ganley

Herman

Willis

WT.

Muscle metabolism during overground walking in persons with poststroke hemiparesis. Top Stroke Rehabil 2008; 15(3): 218–226.

31.

Achten

Jeukendrup

AE.

Optimizing fat oxidation through exercise and diet. Nutrition 2004; 20: 716–727.

32.

Starholm

Mirtaheri

Kapetanovic

et al . Energy expenditure and walking economy of transfemoral amputees walking with different speeds on different walking surfaces. Prosthet Orthot Int 2016; 40(3): 335–342.

33.

Jeukendrup

Saris

Wagenmakers

AJ.

Fat metabolism during exercise: a review. Part I: fatty acid mobilization and muscle metabolism. Int J Sports Med 1998; 19(4): 231–244.

Carbohydrate and fat oxidation in persons with lower limb amputation during walking with different speeds

Abstract

Background:

Objectives:

Study design:

Methods:

Results:

Conclusion:

Clinical relevance

Keywords

Background

Methods

Participants

Study design

Gas exchange and energy expenditure measurements

Statistics

Results

Physical characteristics of the participants

Walking speed

Oxygen uptake

% V · O 2 max utilization

ECW

CHO and fat oxidation rates

Discussion

Conclusion

Possible limitations

Footnotes

Acknowledgements

Author contribution

Declaration of conflicting interests

Funding

References

$% {\overset{\cdot}{V} O}_{2} \max$ utilization