The Effects of Using Geared Wheels on Energy Expenditure During Manual Wheelchair Propulsion in Adults with Spinal Cord Injury

Abstract

Objectives: To quantify the effects of geared wheelchair wheels on energy expenditure during manual wheelchair propulsion in individuals with spinal cord injury (SCI). Methods: Eleven adult manual wheelchair users with SCI propelled their personal manual wheelchairs, which were equipped with a pair of geared wheels, on a passive wheelchair ergometer in low-gear and standard-gear conditions for six minutes. The energy cost of transport, distance traveled, rate of oxygen consumption (SCI MET), rate of perceived exertion, heart rate, and stroke cycle frequency were measured and compared across the gear conditions. Results: The distance traveled and SCI MET were significantly lower (p = 0.003) and cost of transport was significantly higher under the low-gear condition compared with the standard-gear condition. Gear condition exerted a moderate effect on the level of exertion; however, the decrease in the rate of perceived exertion under the low-gear condition was not statistically significant. Gear condition did not significantly affect heart rate and stroke cycle frequency. Conclusions: Geared manual wheelchair propulsion was significantly more energy-demanding, but less intense (easier) under the low-gear condition than the standard-gear condition. Using geared wheels may be beneficial for manual wheelchair users to independently accomplish strenuous propulsion tasks during typical activities of daily living, such as propulsion on carpeted floor. However, the small sample size and inclusion of only male participants limit the generalizability of these findings, and future studies with larger and more diverse cohorts are warranted.

1. Introduction

Approximately 2.3% (5.5 million) of adults in the United States used a wheelchair for their daily mobility in 2014 [1]. Spinal cord injury (SCI) is one of the leading conditions associated with wheelchair use, and manual wheelchairs are the most common alternative mode of mobility chosen by this population [2,3].

Manual wheelchair propulsion is an efficient form of locomotion for people with SCI; however, when compared with normal walking, wheelchair propulsion is relatively inefficient [2,4,5]. The inefficiency is mainly attributed to the low muscle mass of the upper limbs, which are not specialized for ambulatory activities, and the biomechanical disadvantages of using hand-rims for propulsion [6,7]. The health conditions of people with SCI might also contribute to additional propulsion inefficiencies. Energy expenditure and efficiency factors are necessary for the evaluation of function and participation of manual wheelchair users and the assessment of their physical fitness [8]. Therefore, the effects of wheelchair type on energy expenditure and efficiency are essential information that influences wheelchair prescription for individuals with SCI.

Geared manual wheelchairs are a promising alternative propulsion mechanism that may reduce the biomechanical demands of the upper extremities while maximizing function. Similar to a multi-speed bicycle, geared wheels are designed with a low gear, which makes propulsion easier. Geared wheels are relatively new; thus, there is still limited scientific evidence surrounding geared manual wheelchair mobility. Furthermore, there are no guidelines available for the prescription of or transition to geared manual wheelchairs. Studies on people with SCI and non-wheelchair users have shown that using geared manual wheelchairs decreases the upper-extremity biomechanical demands during propulsion and may be beneficial for strenuous tasks such as ramp ascent and propulsion on carpet [9,10,11,12,13]. Finley and colleagues also demonstrated the potential of geared manual wheelchair use in decreasing shoulder pain in manual wheelchair users [14]. However, the effects of geared wheels on energy expenditure and propulsion efficiency in manual wheelchair users have not yet been evaluated.

The aim of this study was to evaluate the effects of geared wheels on the rate of oxygen consumption, perceived exertion, and propulsion efficiency during manual wheelchair propulsion in individuals with SCI. Our previous study on hand-rim biomechanics during geared manual wheelchair propulsion indicated a significant increase in normalized stroke cycle frequency and a significant reduction in propulsion speed and hand-rim propulsive torque under the low-gear condition compared with the standard-gear condition [10]. Building on these results, we hypothesized that using geared wheels under the low-gear condition would increase the energy cost of propulsion and significantly decrease the intensity of wheelchair propulsion compared with the standard-gear condition. To test the first hypothesis, distance traveled and energy cost of transport, and to test the second hypothesis, rate of oxygen consumption for SCI individuals (SCI MET), average heart rate, and rate of perceived exertion were contrasted under low-gear and standard-gear conditions.

2. Methods

This study was approved by the Department of Veterans Affairs (DVA, Milwaukee, WI, USA) and the University of Wisconsin–Milwaukee (UWM) Institutional Review Boards (IRBs). Prior to research participation, all participants submitted written informed consent.

2.1. Participants

The inclusion criteria for participants were to be aged between 18 and 70 years old, use a manual wheelchair as the primary mode of mobility, have a minimum of six months of experience as a manual wheelchair user, and have the ability to perform independent transfers. If the wheelchair users were found to have health complications that would inhibit their ability to participate in a two-hour data collection session, such as pressure ulcers, extensive comorbidities, and severe pain, they were excluded. Participants were examined by an SCI specialist (MV) in the SCI Unit at the Clement J. Zablocki VA Medical Center (Milwaukee, WI, USA) to confirm eligibility. The participants’ characteristics are listed in

Table 1. Participant’s characteristics.

Participant	Age (Years)	Weight (kg)	Height (cm)	Arm Dominance	SCI Level	Years as Wheelchair User
1	53	87	178	Left	T4, ASIA A	27
2	42	85	188	Right	T10, ASIA C	21
3	55	98	185	Right	T5, ASIA A	31
4	36	80	175	Right	L2, ASIA C	12
5	68	73	170	Right	T10, ASIA A	1.5
6	57	81	180	Right	T11, ASIA C	0.6
7	50	66	180	Right	T6, ASIA C	9.5
8	24	71	180	Right	T5, ASIA A	2
9	51	112	188	Left	T12, ASIA C	30
10	29	93	188	Right	T1, ASIA A	10
11	54	136	193	Right	T12, ASIA A	36
Mean (SD)	47.2 (13.1)	89 (20)	182.3 (6.8)	--------------	------------	16.4 (13.1)

T#: Thoraic spinal injury level, L#: Lumbar spinal injury level; ASIA A: Complete spinal cord injury; ASIA C: Incomplete spinal cord injury.

.

2.2. Experimental Protocol and Instrumentation

Participants were instructed to refrain from caffeine and energy beverages (for six hours) and vigorous exercise (for 12 h) prior to wheelchair propulsion testing at the UWM Mobility Lab. After the collection of anthropometric and demographic information, participants transferred from their wheelchair to an adjustable-height medical exam table. The geared wheels (IntelliWheels, Inc., Champaign, IL, USA) were then mounted on their personal wheelchairs. The participants then returned to their wheelchairs. The tire pressure was adjusted to approximately 700 kPa. After a 15 min acclimation period to propulsion with the geared wheels on passive rollers (McLAIN, Traverse City, MI, USA), the participants were instrumented with a portable metabolic system (COSMED K4b², Rome, Italy), heart rate monitor (T34, Polar Electro Inc., Lake Success, NY, USA), and retroreflective markers (

Table 2. Locations of retroreflective markers for calculation of the distance traveled and stroke cycle frequency, as well as characterization of propulsion pattern.

Marker	Location
M3	Dorsal aspect of the hand on the third metacarpal joint, dominant side
Wrist	Dorsal aspect of the wrist midway between the radial and ulnar styloid processes, dominant side
Wheel	Center of wheel hub, non-dominant side
Off-Center	On the wheel, at a distance of 15 cm from the center of the wheel hub, non-dominant side

). The test procedures were reviewed with the participants and they were given additional time to acclimate to wearing the COSMED mask and data acquisition unit (Figure 1).

The geared wheels supported both standard-gear (gear ratio of 1:1) and low-gear (gear ratio of 1.5:1) propulsion conditions. The test protocol included six-minute trials at the participants’ self-selected speed (they were instructed to propel at their “normal comfortable” pace) under both the standard- and low-gear conditions, in random order. A mandatory ten-minute rest period separated test conditions. If needed, the participants could rest for a longer time. Measurements of energetics (breath-by-breath measures of O₂ and CO₂ to estimate oxygen uptake (mL/min) and energy expenditure (kcal/min)) and heart rate were conducted during wheelchair propulsion. Hand-rim kinematics and spatiotemporal parameters were also collected using a 15-camera motion capture system (Vicon Motion Systems, Oxford, UK; 120 Hz) and a cycling speedometer (Bell Dashboard 100; Bell Sports, Inc., Scotts Valley, CA, USA).

All tests were conducted in the morning; the relative humidity was 40–60% and air temperature ranged from 20–22 °C. The K4b² system used for this study has been reported as a valid and reliable measure of oxygen uptake and has been effectively used to measure the energy costs of individuals with SCI [15,16]. The system analyzer (K4b² device) was calibrated before each test and verified with reference gases and room air according to the manufacturer’s guidelines.

At the end of each task, the rate of perceived exertion was measured using the Borg 6–20 scale [17]. Borg 6–20 is a subjective rating scale that has been used as a valid method for rating perceived exertion and measuring exercise intensity in people with SCI and manual wheelchair users [18,19].

2.3. Data Processing

The outcome measures for each participant and test condition included distance traveled, energy cost of transport, and rate of perceived exertion during the 6 min push test, as well as SCI MET, average heart rate, and stroke cycle frequency during the steady-state phase of wheelchair propulsion. The distance traveled (m) was calculated using the Off-Center marker (Table 2) kinematic data (sagittal plane). The distance traveled for each minute of propulsion was the product of the number of wheel cycles and the wheel circumference (m). For one participant, the distance traveled was calculated using the cycle speedometer mounted on the front roller of the passive wheelchair ergometer. This method was considered acceptable because, for the other ten participants, the distance from the speedometer differed by less than 3% from the distance calculated using kinematic data. The K4b² software was used to establish summary estimates of energy expenditure, including oxygen uptake (VO₂), CO₂ production (VCO₂), and rate of energy expenditure. The rate of energy expenditure (EE, kcal/min) was calculated based on the measured VO₂ (L/min) and VCO₂ (L/min), calculated breath by breath (Equation (1), [20]). The rate of energy expenditure data were then used to calculate the total energy expenditure (kcal) during the 6 min push test on the passive wheelchair ergometer.

EE = 3.781 × VO₂ + 1.237 × VCO₂

(1)

To calculate SCI MET, oxygen consumption was normalized by the participant’s weight (VO₂/kg, mL/min/kg). Steady-state values were reviewed during minutes 2 through 5 of the 6 min trial (coefficient of variation < 10%); these normalized steady-state values were averaged across 30 s periods. As per Collins et al., one metabolic equivalent (METs) for SCI individuals is 2.7 mL/kg/min (for able-bodied adults, one MET is 3.5 mL/kg/min), [16]. Therefore, the normalized steady-state values of oxygen consumption were divided by 2.7 to determine the SCI MET during propulsion for both gear conditions. The average heart rate, serving as a physiological indicator of activity intensity, was also calculated for the period of steady state. Additionally, the effect of gear ratio on the average self-selected stroke cycle frequency was explored in this study, as it is an important biomechanical indicator of propulsion strategy. The sagittal plane kinematics of the wrist marker were used to calculate the stroke cycle frequency. The average stroke cycle frequency (expressed in pushes per minute) during the steady-state phase was then computed for each task to control for the effects of cadence and propulsion strategy on energy expenditure and cost of transport.

2.4. Statistical Analysis

The data from wheelchair propulsion testing were analyzed using gear condition as the independent variable and the energy cost of transport, distance traveled, SCI MET, heart rate, rate of perceived exertion, and stroke cycle frequency as the main dependent variables.

To test the research hypotheses, the dependent variables were compared across gear conditions using separate Wilcoxon Signed-Rank tests. Test statistics (Z), significance (p), and effect size (r) are reported for each metric. The effect size was calculated as the ratio of the Z-value to the square root of the number of observations (number of participants times two). Statistical analyses were performed using SPSS 25 (IBM, Armonk, NY, USA) and the level of significance was reduced from 0.05 to 0.0083, using a Bonferroni correction for six dependent variables.

3. Results

All participants performed the wheelchair propulsion tasks. The heart rate data during wheelchair propulsion from two participants were not analyzed due to technical issues that occurred during testing. The box plot graphs combined with the individual measurements for all dependent variables are depicted in Figure 2.

The results of the 6 min push test on the passive wheelchair ergometer indicated that the distance traveled (p = 0.003) and SCI MET (p = 0.006) were significantly lower under the low-gear condition compared with the standard-gear condition (

Table 3. Mean and standard deviation (SD), as well as median and first and third quartiles (Q1, Q3), energy cost of transport (CT), distance traveled (Distance), SCI MET, average heart rate (HR), stroke cycle frequency (Frequency), and rate of perceived exertion (RPE) for the standard gear and low gear conditions.

	Standard Gear		Geared		Statistical Results
	Mean (SD)	Median (Q1, Q3)	Mean (SD)	Median (Q1, Q3)	N	Z	p	r
CT (cal/m)	85.9 (30.1)	71.4 (67.2, 100.1)	112.5 (37.8)	99.9 (84.4, 133.1)	11	2.93	0.003 *	0.62
Distance (m)	341.1 (144.6)	314.3 (247.2, 413.9)	224.1 (91.8)	212.1 (162.2, 266.9)	11	2.93	0.003 *	0.62
SCI MET	4.3 (1.2)	4.2 (3.5. 4.8)	3.8 (1.0)	3.9 (3.0, 4.0)	11	2.75	0.006 *	0.59
HR (bpm)	102 (16)	105.3 (89.5, 116.5)	100 (19)	93.3 (86.0, 114.5)	9	0.71	0.47	0.17
RPE	12.4 (2.8)	12.0 (11.0, 13.0)	10.9 (1.9)	11.0 (9.5, 12.5)	11	1.72	0.085	0.37
Frequency (push/min)	55 (9)	54.3 (49.3, 60.5)	54 (10)	52.7 (45.8, 62.8)	11	0.76	0.45	0.16

N = number of participants; Z = test statistic; p = significance level; r = effect size; *: p < 0.01.

). The energy cost of transport was significantly higher (p = 0.003) under the low-gear condition. Gear condition was moderately effective on the rate of perceived exertion (Table 3, Figure 2). However, the decrease in perceived exertion under the low-gear condition in comparison with the standard condition was not statistically significant (p = 0.085). Gear condition did not affect heart rate (p = 0.47; Table 3). Gear condition did not alter the stroke cycle frequency during wheelchair propulsion on passive rollers (p = 0.45; Table 3).

4. Discussion

We characterized the effects of using geared wheelchair wheels on the energy cost and intensity of wheelchair propulsion in manual wheelchair users with paraplegic SCI. To the authors’ knowledge, this study is the first time that the effects of using geared manual wheelchairs on energy expenditure have been investigated with experienced manual wheelchair users. The results of the study support the main hypotheses.

Consistent with our first hypothesis, the energy cost of propulsion increased significantly under the low-gear condition in comparison with the standard-gear condition (Table 3, Figure 2). This indicated that using the geared wheels under the low-gear condition was significantly more energy-demanding for propelling a given distance in comparison with the standard-gear condition. As expected, distance traveled during the six-minute propulsion on the passive rollers was significantly lower compared with the standard gear condition, whereas the stroke cycle frequency was relatively similar across gear conditions (Table 3, Figure 2). This suggests that a higher number of stroke cycles are required to propel the same distance under the low-gear condition relative to the standard condition, which could explain the increased energy cost.

Using the geared wheels under the low-gear condition significantly decreased the SCI MET in comparison with the standard-gear condition (Table 3, Figure 2). This could be interpreted as a significant decrease in the intensity of the wheelchair propulsion task. The decrease in the rate of perceived exertion is consistent with this interpretation, although it was not statistically significant. The significant decrease in SCI MET and the decrease in the rate of perceived exertion support the second hypothesis that using the geared wheels under the low-gear condition decreased the physiological and perceived intensity of wheelchair propulsion.

The findings of this study regarding the increase in the cost of propulsion and decrease in the intensity of task during manual wheelchair propulsion under the low-gear condition suggest that geared manual wheelchair use might be beneficial for individuals with SCI to perform mobility tasks that cannot be performed easily using standrad manual wheelchairs. For example, to ascend a steep slope, it might be much easier for wheelchair users to propel under the low-gear mode. They might still consume more energy and take a much longer time, but the benefit is it is much easier on the upper limbs (perceived exertion) so that the effort is sustainable.

The lack of statistically significant differences in average heart rate between gear conditions may be attributed to the similar stroke cycle frequency that the participants used for wheelchair propulsion under both conditions. Participants were instructed to propel their wheelchair at their normal comfortable speed. The similar stroke cycle frequency under both wheel conditions suggests that they performed both tasks at their comfortable stroke cycle frequency (optimum frequency). Previous studies have also reported optimal energy cost and efficiency at stroke cycle frequencies close to the self-selected frequency [21,22]. Our observation of a fixed cycle stroke frequency across conditions is also consistent with the findings of Salm et al. [23], who reported a fixed cycle frequency for both single-task and dual-task propulsion in individuals with SCI. They suggested that this may indicate an overarching control function, which could also explain why participants in the present study maintained their preferred stroke cycle frequency, despite changes in wheel condition. Additionally, if the task was overground, the subject might feel the speed through the surrounding environment change and distance traveled. However, when their physical location was fixed on the passive rollers, they might not be sensitive to the speed. This could result in a fixed stroke cycle frequency, instead of speed.

The similar stroke cycle frequency under both wheel conditions might indicate that wheelchair propulsion at the self-selected pace for only 6 min was such an easy task so that when the gear ratio increased from the low gear to standard gear, they did not feel the challenge and kept the same stroke frequency. Therefore, when the workload is increased to a more intensive level (e.g., fast-paced propulsion, ramp ascension, etc.), the task would challenge the subject’s physical limit. Then, different gear ratios would introduce a more significant difference; thereby significantly reducing perceived exertion, maybe even reducing the cost of propulsion. Thus, experiments in future studies should include data collection during overground propulsion tasks with high intensity levels, such as fast-paced propulsion and ramp ascension.

The energy expenditure results for the standard gear condition demonstrated a mean (SD) intensity of 4.33 (1.24) SCI METS for 11 adult male manual wheelchair users with SCI; this means that the intensity level was within the same range reported for wheeling on carpet/grass (3.35–6.22 SCI METs) [16,24]. The SCI METs for both conditions indicate that propulsion on passive rollers can be classified as a moderate-intensity (3.0–6.0 METs) physical activity for manual wheelchair users with SCI [24]. This is consistent with the results of a recent study by Rice and colleagues; they reported that the intensity of physical activity performed during manual wheelchair propulsion reaches moderate intensity at a push rate close to optimum frequency (55 pushes/min) [25]. Therefore, habitual manual wheelchair propulsion may help to improve the physical activity level in individuals with SCI who are among the least physically active in society [26]. Strong evidence suggests that increasing the physical activity level from low to moderate could sharply decrease the risks associated with an inactive lifestyle [27]. Additionally, previous studies indicated that using geared manual wheelchairs might decrease upper-extremity biomechanical demands during propulsion [9,10,11,12,13], which are the main risk factors for high prevalence of pain and upper-extremity secondary injuries in this population. These results, combined with the findings on the effect of geared manual wheelchair use on the energy cost of propulsion and task intensity, could have important implications for rehabilitation, fatigue management, and exercise prescription. They suggest the potential of using geared manual wheelchairs in promoting the level of physical activity while decreasing the risk of secondary upper-extremity injuries. However, further studies with a larger sample size are warranted to investigate the effects of using geared wheels on daily fatigue in manual wheelchair users with SCI and whether it could promote greater levels and more sustained physical activity for them during the day.

The small sample size was the main limitation that might limit the interpretation of the findings of this study. As our sample only included male manual wheelchair users with different levels of paraplegic SCI, we are unable to generalize the findings to female wheelchair users or individuals with other disabilities. Additionally, the effects of the level of injury on the observed results was not investigated. All the tasks in this study were performed on a passive wheelchair ergometer, which differs from over-ground propulsion. To minimize this potential impact, the rear roller of the passive ergometer was connected to a flywheel to provide momentum during the recovery phase. This research was a cross-sectional study and participants had limited time to acclimate to the geared wheels. Further research with a longitudinal design and with a larger sample size of wheelchair users is warranted. Finally, as we used a passive wheelchair ergometer and we did not measure the hand-rim kinetics, we were unable to calculate the gross mechanical efficiency; this can be another suggestion for further research.

5. Conclusions

The findings of this study demonstrated that geared manual wheelchair propulsion aa t self-selected frequency had a significantly higher energy cost of propulsion and was significantly less intense (easier) under the low-gear condition than the standard-gear condition and accompanied by a reduced perception of effort. These results suggest that using geared wheels may be beneficial for manual wheelchair users with SCI to independently accomplish strenuous propulsion tasks during typical activities of daily living. These results may be utilized to develop practical and accessible exercise recommendations based on manual wheelchair propulsion for individuals with SCI to maintain and promote their health. This work also has the potential to impact clinical decision making for wheelchair prescription and usage for clinicians and manual wheelchair users.