Manual wheelchair biomechanics while overcoming various environmental barriers: A systematic review.

During manual wheelchair (MWC) locomotion, the user's upper limbs are subject to heavy stresses and fatigue because the upper body is permanently engaged to propel the MWC. These stresses and fatigue vary according to the environmental barriers encountered outdoors along a given path. This study aimed at conducting a systematic review of the literature assessing the biomechanics of MWC users crossing various situations, which represent physical environmental barriers. Through a systematic search on PubMed, 34 articles were selected and classified according to the investigated environmental barriers: slope; cross-slope; curb; and ground type. For each barrier, biomechanical parameters were divided into four categories: spatiotemporal parameters; kinematics; kinetics; and muscle activity. All results from the different studies were gathered, including numerical data, and assessed with respect to the methodology used in each study. This review sheds light on the fact that certain situations (cross-slopes and curbs) or parameters (kinematics) have scarcely been studied, and that a wider set of situations should be studied. Five recommendations were made at the end of this review process to standardize the procedure when reporting materials, methods, and results for the study of biomechanics of any environmental barrier encountered in MWC locomotion: (i) effectively reporting barriers' lengths, grades, or heights; (ii) striving for standardization or a report of the approach conditions of the barrier, such as velocity, especially on curbs; (iii) reporting the configuration of the used MWC, and if it was fitted to the subject's morphology; (iv) reporting rotation sequences for the expression of moments and kinematics, and when used, the definition of the musculoskeletal model; lastly (v) when possible, reporting measurement uncertainties and model reconstruction errors.

1. Introduction

In 2019, it was estimated that 75 million people in the world require a manual wheelchair (MWC) [1]. MWC users daily face physical environmental barriers such as slopes, cross-slopes, curbs, and uneven terrain that affect their access to buildings and urban areas. Yet, accessibility for people with disabilities is crucial for their social and professional integration [2-4]. Standards and regulations have been established to impose some architectural rules to make public buildings and squares accessible to everyone. However, the regulations are mainly based on the aspects of required space and maximum slope inclination [5]. Despite the improvement of the overall accessibility of public areas, these regulations remain unsatisfactory for a large proportion of MWC users [5-7].

The limitations imposed by environmental barriers in MWC locomotion can be described using the International Classification of Functioning, Disability, and Health (ICF) [8]. The ICF is a framework for describing “dynamic interactions between a person’s health condition, environmental factors, and personal factors” [8]. The ICF can therefore be used to identify key elements that need to be addressed in rehabilitation [9, 10], to guide the classification of assistive technology [9], or even to determine the relationship linking wheelchair skills and capabilities with participation frequency and mobility [11]. From that, previous studies have, in particular, revealed the need for better training in overcoming environmental barriers [10]. In addition, the ICF framework could be used by clinicians to adapt MWC training programs according to their patients’ capabilities and life projects [12]. To that end, it appears necessary to be able to associate a barrier’s difficulty with the user’s capabilities. This could be achieved by the quantification and comparison of the physical demands associated with the various environmental barriers encountered.

Biomechanical analysis of locomotion is a reference method to investigate physical demands associated with MWC locomotion. Such biomechanical analysis classically includes the quantification of joint motion and intersegmental loads (forces and torques). Thus, several studies have investigated the physical demands of MWC propulsion when crossing various environmental barriers from a biomechanical point of view [13-17]. Illustrations of environmental barriers that were recreated in a laboratory to that end can be found Fig 1. However, in general, only one type of barrier was investigated in each study, and it appears that no study investigated more than two types of obstacles, hindering the comparison of results between barriers. Moreover, studies seem to use a variety of experimental protocols and investigated different biomechanical parameters. For these reasons, researchers may encounter difficulties when looking for concise data on the influence of environmental barriers on a biomechanical evaluation of MWC locomotion. To address this gap, the purpose of this study was to identify and synthesize data and experimental methods from the literature on the biomechanics of MWC propulsion for various and frequent environmental barriers that are encountered daily by MWC users.

Fig 1

Reproduction of environmental barriers in a laboratory.

Picture A: reproduction of a slope. Picture B: reproduction of a cross-slope. Picture C: reproduction of a curb.

2. Methods

The present study conducted a systematic review to identify and analyze existing studies that reported biomechanical parameters of MWC propulsion while overcoming environmental barriers. Because handrim propulsion is the most frequent system of manual propulsion adopted by MWC users due to its higher compliance with the constraints of activities of daily living indoors [18-21], the review focuses on the biomechanics of manual handrim propulsion.

2.1 Systematic literature review

To answer the question: “What are the biomechanics involved to overcome specific environmental barriers?”, a systematic search was performed based on the methodology of Harris et al. [22] and Moher et al. [23] to identify relevant articles published until May 2021 within the Pubmed and Scopus database.

The request, launched on May 3, 2021, focused on biomechanical parameters and especially on spatio-temporal parameters, kinematics, kinetics, and muscle activations during MWC propulsion to overcome environmental obstacles, as well as on the experimental methods used to obtain the aforementioned parameters. More precisely, the request was:

(bioengineering OR biomechanic* OR kinematic* OR velocity OR velocities OR (joint angle*) OR kinetic* OR force* OR torque* OR moment* OR (motion capture) OR electromyography) AND wheelchair AND (propulsion OR slope OR kerb OR curb OR ground OR floor OR rolling resistance OR activities OR activity OR ambulation OR locomotion OR situation)

The keywords used for this search were determined after reviewing the results of a preliminary search, which had identified the four most studied in the literature: slope, cross-slope, curb, and ground type.

2.2 Article selection

Articles were selected following the flow diagram (Fig 2) recommended by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [23]. After eliminating duplicates, all titles were screened for inclusion by three of the authors. The inclusion criteria were: original study or systematic review; study written in English; and features experimental results on slopes, cross-slopes, curbs, and ground types during MWC locomotion. Exclusion criteria were articles about electric wheelchairs, power-assisted wheelchairs, sports wheelchairs, other propulsion systems than manual handrim, and hemiplegia-pattern propulsion. All other abstracts and articles were screened by the same authors. The articles on subject-based studies dealing with an environmental barrier were selected and then sorted according to the barrier type: slope, cross-slope, curb, and ground type.

Fig 2

PRISMA 2020 flow diagram for new systematic reviews which included searches of databases and registers only.

For the analysis, biomechanical parameters were divided into four a priori defined categories: spatio-temporal parameters (push time, recovery time, cycle time, speed, etc); kinematics (joint angles); kinetics (handrim forces and torques, rate of rise, fraction of effective force, net joint moments, mechanical work and power, etc); and muscle activity. A more detailed definition of these biomechanical parameters can be found in S1 Appendix.

3. Results

The first search resulted in a total of 1429 references, and 1093 articles remained after removing duplicates. The screening through the title filter resulted in 266 references. After reading the abstracts, 59 articles were selected, and finally, 34 papers were included in this review after the full text read. This selection process is summarized in Fig 2.

The 34 retrieved articles included populations between 7 and 128 participants (Total: 756, Mean [M]: 22, standard-deviation [SD]: 25). Cohorts included able-bodied (AB) subjects and MWC users (MWU), among whom spinal cord injured (SCI) subjects, subjects with lower limb amputation, cerebral palsy, neuropathy, or Friedreich’s Ataxia. AB and SCI subjects were studied in 10 and 22 studies, respectively (Table 1).

Table 1

Synthesis of all studies.

Reference	Ground types	Slope	Cross-Slope	Curb	Able-bodied	MWC Users	Video camera	opto-electronic motion capture	IMUs1	optical encoder	Instrumented wheel	Spatio- temporal parameters	Kinematics	Handrim Kinetics	Body kinetics	EMG2
Bertocci et al., 2019 [7]		x			7		x				x	x		x
Chow et al., 2009 [24]		x				9	x					x				x
Cowan et al., 2008 [25]	x	x				128					x	x		x
Cowan et al., 2009 [26]	x					52					x	x		x
Dysterheft et al., 2015 [27]	x					10					x	x		x
Gagnon et al., 2014 [28]		x				18		x			x	x		x
Gagnon et al., 2015 [29]		x				18		x			x	x	x		x	x
Holloway et al., 2015 [30]		x	x			7			x		x			x	x	x
Hurd et al., 2008 [15]	x		x			12					x	x		x
Hurd et al., 2008 [31]	x					14					x	x		x
Hurd et al., 2009 [32]	x	x				13					x			x
Kim et al., 2014 [33]		x			30							x				x
Koontz et al., 2005 [34]	x					11					x	x		x
Koontz et al., 2009 [35]	x					28		x			x	x		x
Kulig et al., 1998 [36]		x				17		x			x	x			x
Lalumiere et al., 2013 [37]				x		15		x			x		x		x	x
Levy et al., 2004 [38]	x	x				11										x
Martin-Lemoyne et al., 2020 [39]	x					13						x				x
Morrow et al., 2010 [40]		x				12		x			x				x
Morrow et al., 2011 [41]		x				12		x					x
Mulroy et al., 2005 [42]		x				13		x			x	x			x
Newsam et al., 1996 [43]	x	x				70				x	x	x
Oliveira et al., 2019 [44]	x	x				7			x		x	x	x	x
Qi et al., 2013 [45]		x			15						x	x		x	x	x
Requejo et al., 2008 [46]		x				20		x			x	x				x
Richter et al 2007 [47]			x			25		x			x	x		x
Slavens et al., 2019 [48]		x			14			x				x	x			x
Soltau et al., 2015 [49]		x				80		x			x	x	x	x
Symonds et al., 2016 [50]		x	x		6	7			x		x	x	x			x
van der Woude et al., 1989 [51]		x			6	6	x					x		x
van drongelen et al., 2005 [52]		x		x	5	12					x	x		x	x
van drongelen et al., 2013 [13]		x			12			x			x	x		x
Veeger et al., 1998 [53]		x			5	4		x					x
Wieczorek et al., 2020 [54]		x			8					x		x				x

1IMU: Inertial Measurement Unit;

2EMG: Electromyography.

Experimental design, acquisition methods, and measurement tools were also found to differ between studies. The MWC was propelled overground, on a treadmill, or over a stationary ergometer. Kinematics were recorded either with motion capture systems, inertial measurement units, video cameras or optical encoders. Kinetics were systematically recorded with instrumented wheels.

An overview of the retrieved studies is provided in Table 1. A subsection dedicated to each investigated environmental barrier (slope, cross-slope, curb, ground type) summarizes the experimental methods used in these studies (also reported in Tables 2–5) as well as the obtained biomechanical results. A compilation of the detailed numerical results of the studies is appended as supplementary material (S2 Table).

Table 2

Study review for slope investigation.

Reference	Population		Experimental condition	Speed	slope grade (°) and length (m)	Kinematics	Kinetics	Muscle activity	Model
Slavens et al., 2019 [48]	14 (7F, 7M)	AB1	overground	self selected	0° (10 m) and 4.8° (2.5 m)	opto-electronic (15 cameras, 120 Hz)		EMG2 (3 muscles)	Schnorenberg et al., 2014 [55]
Bertocci et al., 2019 [7]	7 (2F, 5M)	AB	overground	self selected	3.5, 9.8, 15° (1.22 m)	1 video camera (30 Hz)	instrumented wheel (dominant side)
Holloway et al., 2015 [30]	7 (7M)	SCI3	overground	self selected	0, 3.7, 6.8° (lengths not reported)	IMU4 (50 Hz)	instrumented wheel (side not reported)	EMG (3 muscles)	‘Dynamic Arms 2013’ (Holzbaur et al., 2005 [56])
Gagnon et al., 2015 [29] Gagnon et al., 2014 [28]	18 (1F, 17M)	SCI	motorized treadmill	self selected (but identical for all slopes)	0, 2.7, 3.6, 4.8, 7.1° (length: N/A)	opto-electronic (4 cameras, 30 Hz)	instrumented wheels (both sides)	EMG (4 muscles)	ISB Recommendation (Wu et al., 2005 [57]) adapted for shoulder sequence (Senk et Chèze 2006 [58])
Qi et al., 2013 [45]	15 (7F, 8M)	AB	overground	self selected	4° (4.1 m)		instrumented wheel (side not reported)	EMG (7 muscles)
van drongelen et al., 2013 [13]	12 (12M)	AB	motorized treadmill	imposed (1.1 m/s)	0.6, 1.4, 2.3° (length: N/A)	opto-electronic (6 cameras, 100 Hz)	instrumented wheel (left side)		only measurement of the hand marker
Chow et al., 2009 [24]	10 (10M)	5 SCI, 5 with various disabilities	overground	self selected (normal and fast speed)	0, 2, 4, 6, 8, 10, 12° (7.3 m)	1 video camera (60 Hz)		EMG (6 muscles)	2D analysis
Oliveira et al., 2019 [44]	8 (1F, 7M)	4 SCI,3 Cerebral palsy, 1 Friedrich’s Ataxia	overground	self selected	0° (10m) and slope with non-constant grade (max grade: 5°, total length: 4.8m)	IMU (11 sensors, 60 Hz)	instrumented wheel (right side)		Xsens MVN Biomech model
Morrow et al., 2010 [40]	12 (1F, 11M)	11 SCI, 1 spina bifida	overground	not reported	0° and 4.6° (length: 10 m)	opto-electronic (10 cameras, 240 Hz)	instrumented wheels (both sides)		ISB recommendations (Wu et al., 2005 [57])
van drongelen et al., 2005 [52]	17	12 SCI, 5 AB	motorized treadmill	imposed (0.56 m/s)	0° and 1.7° (length: N/A)	opto-electronic (3 cameras, 100Hz)	instrumented wheel (right side)		Delft Shoulder and Elbow Model
Veeger et al., 1998 [53]	9	4 SCI, 5 AB	motorized treadmill	imposed (0.83, 1.11, 1.39 m/s)	0.6, 1.1, 1.7° (length: N/A)	opto-electronic (60Hz)		EMG (1 muscle group)
van der Woude et al., 1989 [51]	12 (12M)	6 MWU5, 6 AB	motorized treadmill	imposed (0.55, 0.83, 1.11, 1.39 m/s)	1, 2° (length: N/A)	1 video camera (54 Hz)
Wieczorek et al., 2020 [54]	8	AB	overground	self-selected	4.6° (4m)	incremental encoder (500 steps)		EMG (4 muscles)
Symonds et al., 2016 [50]	13 (1F, 12M)	7 SCI, 6 AB	overground	self-selected	0, 3.7, 6.8° (8.4, 7.2, 1.5m)	IMU (50Hz)	instrumented wheel (left side)	EMG (3 muscles)
Hurd et al., 2009 [32]	13 (1F, 12M)	SCI	overground	self-selected	3° (30m)		instrumented wheels (both sides)
Kim et al., 2014 [33]	30 (19F, 11M)	AB	overground	self-selected	4.1, 4.8, 5.7, 7.1, 9.4° (0.9, 1.2, 1.5, 1.8, 2.1, 2.4, 3, 3.6, 4.2, 2.7, 3.6, 4.5, 5.4, 6.3m)
Kulig et al., 1998 [36]	17 (17M)	SCI	stationnary ergometer	self-selected	0, 4,6° (length: N/A)	opto-electronic (50Hz)	instrumented wheel (right side)		4 rigid bodies linked by 3 degrees of freedom joints
Levy et al., 2004 [38]	11 (3F 8M)	MWU	overground	self-selected	0, 5° (100m, 9m)			EMG (8 muscles)
Morrow et al., 2011 [41]	12 (1F 11M)	11 SCI, 1 spina bifida	overground	self-selected	0° and 4.6° (10 m)	opto-electronic (10 cameras, 240 Hz)			ISB recommendations (Wu et al., 2005 [57])
Requejo et al., 2008 [46]	20 (20M)	12 Tetra, 8 Para	stationnary ergometer	self-selected	0, 2.3, 4.6° (length: N/A)	opto-electronic (6 cameras, 50 Hz)		EMG (4 muscles)
Cowan et al., 2008 [25]	128 (102 M, 26 F)	SCI (various levels)	overground	self-selected	0, max 5°		instrumented wheel (side not reported)
Mulroy et al., 2005 [42]	13 (13 M)	SCI	stationnary ergometer	self-selectect	0, 4.6° (length: N/A)	optoelectronic moetio capture (6 cameras, 50 Hz)	instrumented wheel (right side)		Inverse dynamics: Kulig et al., 1998 [36]
Newsam et al., 1996 [43]	70 (70M)	SCI	stationnary ergometer	self-selected	0, 2.3, 4.56° (length: N/A)	incremental encoder
Soltau et al., 2015 [49]	80 (74 M, 6 F)	MWU (paraplegic)	stationnary ergometer	self-selected	0, 4.6° (length: N/A)	opto-electronic motion capture	instrumented wheels (both sides)		ISB recommendation (Wu et al., 2005 [57])

1AB: Able-bodied;

2EMG: Electromyography;

3SCI: Spinal cord-injured;

4IMU: Inertial measurement unit;

5MWU: Manual wheelchair user.

Table 5

Study review for ground type investigation.

Study	Population		Test ground types	Length (m)	Kinematics	Kinetics	Muscle activity
Oliveira et al., 2019 [44]	8 (7M 1F)	4 SCI1, 3 Cerebral palsy, 1 Friedrich’s Ataxia	tile; polyfoam mat	10; 2.2m	IMU2 (11 sensors, 60Hz)	instrumented wheel (right side, 240Hz)
Koontz et al., 2009 [35]	29 (28M 1F)	25 SCI, 3 lower-limb amputees, 1 Neural palsy	linoleum; carpet	1.2; 1.5m	opto-electronic (6 cameras, 60Hz)	instrumented wheels (both sides, 240Hz)
Cowan et al., 2009 [26]	53 (20M 33F)	MWU3	tile; low-pile carpet; high-pile carpet	12; 7.3; 7.3m		instrumented wheels (both sides, 240Hz)
Hurd et al., 2008 [15]	12 (11M 1F)	SCI	smooth concrete; aggregate concrete; carpet; tile	N/A; N/A; 10; 10m		instrumented wheels (both sides, 240Hz)
Hurd et al., 2008 [31]	14 (12M 2F)	SCI	aggregate concrete; smooth concrete; carpet; tile	30; 30; 10; 10m		instrumented wheels (both sides, 240Hz)
Koontz et al., 2005 [34]	11 (10M 1F)	10 SCI, 1 multiple sclerosis, 1 transfemoral amputee	high-pile carpet; low-pile carpet; concrete; pavers; grass; tile; wood	7.6; 18.3; 15.2; 15.2; 6.1; 15.2; 15.2m		instrumented wheel (right side, 240Hz)
Hurd et al., 2009 [32]	13 (11M 2F)	SCI	smooth concrete; aggregate concrete	30m		instrumented wheels (both sides, 240Hz)
Levy et al., 2004 [38]	11 (8M 3F)	MWU	linoleum; carpet	100; 21m			EMG4 (8 muscles)
Cowan et al., 2008 [25]	128 (102 M, 26 F) hard-tile: 123 low-pile: 94	SCI (various levels)	hard tile; low-pile carpet	10; 10m		instrumented wheel (one side, not reported)
Dysterheft et al., 2015 [27]	10 (7 M 3F)	Teenage MWU	tile; carpet; concrete	15; 15; 15m		instrumented wheel (both sides, analyzed only at the right side, 240 Hz)
Martin-Lemoyne et al., 2020 [39]	13 (9M, 4 F)	SCI	tiled abrasive floor; padded carpet fllor	10; 10m			Surface EMG (4 muscles, dominant arm)
Newsam et al., 1996 [43]	70 (70M)	SCI	tile; carpet	15; 12m	optical encoder	Force transducers

1SCI: Spinal cord-injured;

2IMU: Inertial measurement unit;

3MWU: Manual wheelchair user;

4EMG: Electromyography.

3.1 Slope

3.1.1 Methods on slopes

Twenty-five articles investigated MWC propulsion on a slope, all during slope ascent (). The number of participants ranged between 7 and 128 (M: 23, SD: 29) and the studied populations were mostly MWU (SCI or other motor disabilities).

Experimental design differed across studies, both in terms of the propulsion experimental environment (overground, treadmill, or on a stationary ergometer) and the slope (grade mostly between 2° and 5° but could reach up to 15°) (Table 2). Similarly, the acquisition methods and measurement tools were not consistent between studies. Kinematics recording was most often based on opto-electronic motion capture systems, but also on systems based on inertial measurement units, simple 2D cameras, or optical encoders. Kinetics were always measured through instrumented wheels (six-component dynamometers), generally mounted on one side only. One study investigated the kinetics of both wheels using only one instrumented wheel mounted separately on the right and left sides in different trials [49]. Two of the ten studies which used only one instrumented wheel reported having mounted a matching "dummy" wheel on the opposite side to ensure inertial symmetry [7, 13].

Outcome measurements included spatio-temporal parameters (e.g. MWC mean velocity, cycle frequency, push and recovery phases durations, etc.), kinematics (glenohumeral, elbow, neck, and trunk angles), handrim kinetics (tangential, radial, and total forces; fraction of effective force, mechanical work and power), joint kinetics (shoulder net joint moments and glenohumeral joint contact force), and muscle activity (percentage of maximal voluntary isometric contraction).

3.1.2 Results on slopes

3.1.2.1 Spatio-temporal parameters. Under uncontrolled conditions (i.e. overground), MWC speed was found to decrease with increasing slope. Contradictory results were obtained on cycle frequency: MWU tended to increase their cycle frequency with slope on a long ramp [24], whereas AB decreased cycle frequency with slope on a short ramp [7, 48]. Moreover, when the MWC speed was constant across the different slope inclinations (speed imposed by the treadmill belt), cycle frequency tended to increase with increasing slope in SCI subjects [28, 29], but was not affected with AB subjects [13]. Push phase duration at the reference level (i.e. grade = 0°) was similar in all studies that reported this information [24, 28, 29, 36, 45, 48–50]. When the speed was imposed (i.e. on a motorized treadmill), the push phase duration was not modified [13, 28, 29] by slope. On the opposite, in overground and stationary ergometer studies, where speed was self-selected, push phase duration increased with the grade [7, 24, 30, 36, 45, 48, 49]. All studies reported a decrease in recovery phase duration with the increase of slope inclination. Seven studies [13, 25, 28, 29, 49–51] reported data on contact angles. Four of these studies used treadmills but investigated different populations, namely AB and MWU, and highlighted significant differences between those populations in contact angle even on a zero grade slope [13, 28, 29, 51]: contact angle was higher on the same slope when experimenting on AB subjects, and seemed to remain constant with different grades of slope in AB subjects [13], whilst contact angle tended to decrease with increasing slope in MWU [28, 29]

3.1.2.2 Joint kinematics. Important differences can be noted between studies in all degrees of freedom (DoF) of the glenohumeral joint. In particular, the evolution of the glenohumeral flexion-extension range of motion (RoM) with the grade differed with either an increase [29], no observed change [44, 48, 50], or even a decrease for AB users in one study [50]. On the contrary, results on trunk inclination are in agreement between studies with an increase of trunk flexion-extension RoM with the grade [29, 44, 50]. An increase of the neck extension with the grade, consistent with the increase of the trunk extension to keep the gaze orientation, was also observed [44]. Wrist flexion-extension and radio-ulnar deviation RoM also tended to increase [53], as well as elbow flexion-extension and pronation-supination RoM [49]. Finally, one study reported maximal scapular angles (down-up, antero-posterior, and internal-external rotations), showing a decrease in maximal downward and anterior rotations, and an increase in internal rotation with increasing slope [41].

3.1.2.3 Handrim and joint kinetics. Results on handrim kinetics show noticeable differences between studies when compared to similar or close grades. However, evolution with grade was consistent between studies with an increase of both mean and peak total force, as well as of its tangential and radial components. The handrim mechanical work and power also increased with the grade. Results on the fraction of effective force were however less clear with a mean value that tended to slightly decrease [28], be maintained [13], or increase [45, 49]. Few and disparate outcome data were provided on joint kinetics during slope ascent. An increase of the mean and peak glenohumeral net joint moment and the peak elbow net joint moment with slope was however reported [29, 36, 41, 42, 45]. Two studies reported data on glenohumeral joint contact forces, which require the assessment of muscle forces through a musculoskeletal model, and found a significant increase of the three components of this force with the slope grade [30, 36].

3.1.2.4 Muscle activity. Most studies reported peak EMG value [29, 30, 33, 38, 45, 48, 50, 60], but five studies reported mean EMG activity during propulsion [24, 29, 46, 50, 61]. Although most studies reported normalized muscle activity using maximum voluntary contraction testing, one article reported un-normalized EMG activity as voltage measured by the sensor [38]. The muscles investigated in the studies were often different, although most studies measured the muscle activity of the anterior deltoid and pectoralis major [24, 29, 30, 38, 46, 48, 50]. On equivalent slopes, the different studies gave different values of normalized muscle activity for these two muscles. However, it was observed that muscle activity of all of the studied muscles was found to consistently increase with the grade. Some studies reported muscle activity during locomotion higher than the one observed during maximum voluntary contraction testing for some subjects [46, 61].

3.2 Cross-slope

3.2.1 Methods on cross-slopes

Four articles studied cross-slope propulsion [14, 15, 30, 50] (). Seven to twenty-five (M: 14, SD: 8) MWU—mainly SCI subjects—took part in these experiments. Trials were performed overground [15, 30, 50], or on a treadmill [14], always at self-selected speeds. Cross-slope inclination ranged between 1.4 and 6°. Cross-slope length was only reported in one study (7.2 m) [50].

Kinematics recording was based on an opto-electronic motion capture system or on an inertial measurement unit-based system. Kinetics were systematically measured through a six-component instrumented wheel. The downhill side was systematically measured [14, 15, 30, 50], with only one study reporting using a dummy wheel [14], and only one study equipping both wheels [15]. EMG activity of the downhill side was recorded in two studies and focused on three muscles: the pectoralis major, the anterior deltoid, and the infraspinatus [30, 50].

Outcome data were spatio-temporal parameters (MWC speed, cycle frequency, push and recovery phase duration, contact angle), handrim kinetics (tangential and total handrim forces, fraction of effective force, propelling torque, mechanical work, and mechanical power), shoulder joint kinetics (glenohumeral joint contact force) and muscle activity (peak and/or mean of the percentage of maximal voluntary isometric contraction). One study compared the kinetics at the dominant and non-dominant hand sides, while the MWC’s right wheel was downside, without investigating the effect of the side of the dominant hand (two participants left-handed) [15].

3.2.2 Results on cross-slopes

3.2.2.1 Spatio-temporal parameters. The only study reporting data across different grades of cross-slopes showed a decrease of the speed, an increase of the cycle frequency (i.e. decrease of the cycle duration), an increase of the push phase duration, and a decrease of the recovery phase duration with increasing slope [14]. Contact angles on the downhill side did not appear to be affected by the grade of the cross-slope.

3.2.2.2 Joint kinematics. The only study investigating body kinematics during cross-slope propulsion found an increase in downhill glenohumeral flexion/extension and internal/external rotation RoM compared to level-ground propulsion [50]. On the contrary, downhill glenohumeral abduction/adduction RoM decreased on the cross-slope and trunk flexion/extension RoM tended to increase only in SCI subjects (and not in AB subjects).

3.2.2.3 Joint and handrim kinetics. Peak and mean total forces were shown to increase with increasing grade of the cross-slope [14] or compared to level-ground [14, 30]. The propelling torque on the downhill wheel as well as the mechanical power of this torque were also increased with the grade of the cross-slope. The downhill glenohumeral joint contact force, assessed through a musculoskeletal model, was increased by the cross-slope with respect to level ground in every direction (posterior, superior, medial, and total) [30].

3.2.2.4 Muscle activity. Finally, results on downhill side muscle activity based on EMG data showed an increase of mean muscle activity for all investigated muscles during propulsion on a cross-slope compared to level ground for AB and SCI populations [50]; with an increase of peak muscle activity for the anterior deltoid and pectoralis majors, and a decrease of peak activity for the infraspinatus muscle in SCI participants [30].

3.3 Curb

3.3.1 Methods on curbs

Two studies investigated curb ascent with a MWC [16, 52], involving five and fifteen SCI participants (). Curb height ranged from four to twelve centimeters and curbs were negotiated overground with momentum. Initial instantaneous MWC speed at the beginning of the curb ascent was not reported in any publication.

Kinematics measurements were performed through an opto-electronic motion capture system in both articles but with a small number of cameras for both (less than four). Handrim kinetics were measured using a six-component instrumented wheel, either on one [52] or on both sides [16]. It was not reported whether a dummy wheel was used to equilibrate the MWC when only one instrumented wheel was mounted. EMG data were recorded in one study and focused on four muscles: biceps, triceps, pectoralis major, and anterior deltoid muscles. Outcome data were trunk inclination and upper-limb joint angles (shoulder, elbow, and wrist joints), upper-limb net joint moments (shoulder, elbow, and wrist joints), and muscle activity.

3.3.2 Results on curbs

3.3.2.1 Joint kinematics. Reported results on kinematics [16] showed an increase in the RoM of the shoulder and elbow joints with increasing curb height. In general, this increase was related to an increase of the maximal angle value or a decrease of the minimal value of the angle only. The shoulder internal-external rotation RoM was noticeably increased in both the internal and external rotation ranges. Changes in the wrist RoM remained limited in spite of a slight increase of the peak flexion angle. Finally, the trunk inclination was also modified by the curb height with an increase of the RoM and a noticeable increase of the trunk flexion.

3.3.2.2 Joint and handrim kinetics. Regarding results on net joint moments, both studies found consistent results for peak total shoulder and elbow moments at high curb level (i.e. 10 and 12 cm). Furthermore, peak and mean net shoulder moments were increased for all three moment components, but more especially for the flexion and internal rotation moments. At the elbow, there was also an increase in the total net joint moment, lower than that of the shoulder. The flexion component was the most affected. At the wrist, the increase with curb height was also more limited than at the shoulder and the elbow. The extension and radial deviation components were the most affected. Comparison between joints showed that the higher the initial moment value (i.e. at a curb height of four centimeters), the higher the increase. It can also be noticed that extremely high variability (i.e. standard deviation) was found in upper-limb joint kinetics.

3.3.2.3 Muscle activity. Finally, regarding muscle activity, all four muscles were found to increase their activity with curb height. The biceps brachii and the anterior deltoid muscles appeared to be the most involved between the four studied muscles. Very high variability was also found on these outcome variables.

3.4 Ground type

3.4.1 Methods on ground types

Twelve studies investigated the influence of various ground types on MWC propulsion [15, 25–27, 31, 32, 34, 35, 38, 39, 43, 44] (). The experiments were conducted on MWU populations ranging from eight to 128 participants (M: 31, SD: 36), among which most were SCI participants. Indoor ground types were mostly studied and one study investigated grass and pavers [34].

Kinematics were recorded using an opto-electronic motion capture system [35] or inertial measurement units [44]. Kinetics were recorded using instrumented wheels mounted on both sides of the MWC [15, 26, 27, 31, 32, 35] or one side only [25, 34, 44]. It was not reported if a dummy wheel was also mounted when only one instrumented wheel was used. Muscle activity was recorded using EMG [38, 39].

Outcome parameters included the spatio-temporal parameters of propulsion (speed, stroke frequency, push phase duration, contact angle), handrim kinetics (tangential, radial, and total handrim forces, fraction of effective force, propelling torque, mechanical work, and power), and EMG data expressed in percentage of maximal voluntary contraction for normalization purposes, or directly as measured in voltage.

3.4.2 Results on ground types

3.4.2.1 Spatio-temporal parameters. Results showed that self-selected speed was the highest on smooth concrete, tile, and paved grounds, whereas it was the lowest on high-pile carpet, polyfoam mat, grass, and wood grounds [26, 27, 31, 34, 44]. Stroke frequency was the highest on concrete, grass, and paving. High-pile carpets seemed to induce a decrease in speed compared to low-pile carpets [26, 34], and so did aggregate concrete compared to smooth concrete [31]. In one of two studies, a decrease of stroke frequency was also reported between high-pile and low-pile carpets [26], while in general, similar stroke frequencies were reported for carpet and tile [25, 27, 31, 34, 43].

3.4.2.2 Joint kinematics. Regarding the kinematics of upper limbs, results indicated an increase in the RoM of the shoulder, elbow, neck, and trunk during locomotion on a polyfoam mat compared to locomotion on tiles [44].

3.4.2.3 Joint and handrim kinetics. The reported results on handrim kinetics showed that propulsion on smooth concrete, tile, and linoleum resulted in the lowest values in peak and mean handrim forces, propelling torque, as well as output work and power [15, 31, 34, 35]. Propulsion on low-pile carpet also presented low values in handrim forces, propelling torque, and output work and power [15, 27, 31, 34]. High-pile carpet, aggregate concrete, polyfoam mat, pavers, and grass were the most constraining ground types with high values in peak, mean, and rate of rise handrim forces, propelling torque, and output work and power, with grass propulsion having the highest of these values [15, 31, 34, 44]. Fraction of effective force was reported in two articles only, and showed propulsion asymmetry between the subjects’ dominant and non-dominant sides and presented a high variance among subjects; it was the lowest on smooth concrete, and the highest on grass, as well as generally high on ground types that present higher values in handrim forces and propelling torque [15, 34].

3.4.2.4 Muscle activity. Lastly, regarding muscle activity, an increase of the mean activity was found for the anterior deltoid and the triceps brachii from abrasive tile to padded carpet [39], while similar to decreased voltage values were found from linoleum to carpet for these muscles in [38]. Muscle work was also found to double for the anterior deltoid from tile to padded carpet [39].

4. Discussion

4.1 Investigated environmental barriers

Four different barrier types representing obstacles encountered daily by MWC users were considered and investigated in the literature: slopes; cross-slopes; curbs; and ground types. Among these four barrier types, the slope has been studied the most, always during the ascent, while cross-slopes and curbs (ascent only) were scarcely studied. Yet, the study of curbs and cross-slopes appears particularly relevant since they require specific propulsion strategies. It should be noted that differences in the biomechanics of the uphill and downhill sides during cross-slopes were not investigated.

Out of the thirty-four retrieved studies, nine investigated multiple barriers at once—albeit not more than two [15, 25, 30, 32, 38, 43, 44, 52]. The scarcity of studies on cross-slopes and curbs diminishes the strength of the conclusions drawn by these studies. Indeed, a larger number of studies may have demonstrated contradictory results, as is the case for the retrieved studies on slopes (due to different experimental setups, processing, or populations). The discrepancy of focus between slopes/ground types and curbs/cross-slopes cannot possibly be explained by the lack of cross-slopes or curbs encountered during MWC locomotion in urban areas, since the uneven ground usually encountered may present such environmental barriers, albeit of low grades [2]. Similarly, descending slopes and curbs, or technically challenging situations such as crossing a door threshold with or without a ramp [6] deserve to be studied. For some of these environmental situations, a task analysis could also be considered by separating start-up, propulsion, braking, and turning.

Future studies should therefore be conducted on several different environmental barriers simultaneously, with a special focus on the reproduction of the environments and tasks that are encountered daily by MWUs. Indeed, measuring spatio-temporal parameters, kinematics, kinetics, and muscle activity using the same methods for all barriers would allow the identification of a set of parameters reflecting the difficulty of any environmental barrier encountered in daily MWC locomotion. Furthermore, to allow for comparison of results between studies, the experimental methods and protocols must be clearly defined and explained. Indeed, the speed of the MWC when approaching a curb strongly influences curb negotiation. Similarly, muscle fatigue may impact how the different barriers are approached, and especially curbs and cross-slopes. Consequently, future research should focus on the standardization of protocols and experimental methods regarding MWC locomotion.

4.2 Experimental design

4.2.1 Studied populations

Significant variations were observed in the recruited populations, composed mainly of SCI and AB subjects (twenty-two and ten articles, respectively), but also of lower-limb amputees or subjects affected by cerebral palsy, neuropathy, or Friedreich’s Ataxia. Although the level of experience in MWC locomotion has been shown to significantly affect user biomechanics [62], the MWC locomotion skills of the AB subjects were not specified and therefore this may have influenced the results obtained on each environmental barrier. Even when discarding AB subjects, the included MWC users were characterized by various physical conditions, anthropometries, and abilities. While such differences can lead to different propulsion strategies over the same locomotion conditions, it is interesting to have this variety represented in the studied cohorts, to have a representative population of real-world MWC users.

4.2.2 Reproduction of environmental barriers

The difference in the number of studies investigating each barrier may not only be due to a heterogeneous distribution of interest amongst researchers, but also due to practical reasons regarding the methods available to study each barrier. Indeed, researchers can use inclined motorized treadmills or stationary ergometers to simulate slopes and potentially cross-slopes, whereas experiments with curbs and ground types all need to be conducted overground.

Propulsion strategies implemented on a motorized treadmill or a stationary ergometer replicating a slope or cross-slope may differ from those typically used overground. When motorized treadmills were used, the subjects were sometimes secured using safety belts, which were reported to have some looseness in order to limit their influence on the subject’s propulsion [14, 28, 29]. Yet, even when secured, the subject may unconsciously fear to fail to sustain the speed of the treadmill and therefore fall, leading to safer propulsion strategies than those that they would have adopted overground. When a stationary ergometer is used, slope simulation is achieved by adding a rolling resistance equivalent to the work needed to ascend the desired slope, sometimes coupled with an incline of the MWC [42, 46]. Yet, the stationary ergometer fails to reproduce the increased risk of wheelchair tipping during slope ascension, as well as the risk of backtracking when an insufficient moment of propulsion is applied to the handrim by the user.

It should also be noted that when using a treadmill, propulsion biomechanics may be impacted by the surface of the treadmill belt which differs from everyday overground surfaces, leading to different strategies over a similar slope. This remark is also valid for different surfaces during overground propulsion on slope and cross-slope.

4.2.3 MWC configuration

MWC configuration is one of the main determining factors when optimizing locomotion for a given user, as it affects propulsion biomechanics as well as other locomotion factors, such as stability [63]. MWC stability, for example, is strongly affected by environmental barriers such as curbs or slopes [62, 64, 65]. Yet, most of the reviewed studied did not report the configuration of the investigated MWC, and those that did provided only a brief description of the MWC dimensions. The issue lies in the lack of consensus on methodology to characterize and report MWC characteristics/configuration, leading to a major bias limiting the comparison across studies and subjects.

4.3 Joint kinematics and kinetics estimation

Upper-limb kinematics and subject kinetics were reported for ascending slope propulsion as well as, to a lesser extent, for curb and cross-slope, but not for ground types. Yet, when reported, methodological differences in kinetic and kinematic acquisition (opto-electronic motion capture system, system based on inertial measurement units) and in data processing (musculoskeletal model used for computation of joint angles and moments [66, 67], point and basis of expression of net joint moments [68, 69]) hinder rigorous comparisons of studies on the same barrier, and prevent the formulation of a reliable evidence-based synthesis of the propulsion biomechanics for each barrier. Lastly, poor data acquisition accuracy may lead to improper conclusions, especially for kinematics and kinetics quantities [70, 71]. This observation may explain some of the contradictory results reported in the studies such as those involving slopes.

When investigating handrim kinetics, all studies used instrumented wheels, but most of them only mounted such wheels on one side of the MWC, whereas mounting them on both sides would enable the comparison of kinetics on each side of the MWC user and the evaluation of possible asymmetries in propulsion strategies. Moreover, only four studies reported the use of a dummy wheel to balance the MWC equipped with one instrumented wheel, which is crucial to ensure natural propulsion strategies. During level-ground propulsion over concrete, which is a situation expected to stress the user symmetrically, a relative difference of 20% between dominant and non-dominant sides of the user was found [15]. The only study that investigated cross-slope locomotion using instrumented wheels on both sides of the subjects’ MWC also reported results indicating an asymmetry in handrim forces, propelling torques, mechanical works, and powers when comparing dominant and non-dominant sides of the user [15]. However, they did not report which side was uphill or downhill, which is the most interesting paradigm for interpretation of the results on cross-slopes.

Reported studies also tended to use different musculoskeletal models, yet the definition of joint coordinate systems linked to musculoskeletal models influences both kinematic and kinetics results [72]. Although there is consensus on upper-extremity joint coordinate system definition for kinematics since 2005 [57], the ISB has made recommendation on the reporting of kinetics only as of [73]. Only two studies [30, 36] reported joint contact forces estimations. The reason could be that such a parameter requires a deeper dive into musculoskeletal modeling and simulation because it requires, as a prerequisite, to assess muscle forces [74]. Furthermore, the definition of such a model influences the accuracy with which joint contact forces are estimated [75]. Further studies should take better advantage of musculoskeletal models specifically developed and tailored to study MWC locomotion, and the sharing of these models would favor the standardization of the results.

It should be noted that none of the studies presented in this review reported the uncertainties in the determination of the parameters of interest, while the different choices of models or measurement devices might have resulted in significant uncertainties. For instance, multibody kinematics optimization was found to generally carry reconstruction residual errors on markers ranging from four to forty millimeters, and between three and ten degrees of error against true bone kinematics for shoulder rotations [76]. Moreover, the measurement uncertainty of kinetic measurement devices given by manufacturers has to be applied and propagated with those kinematic uncertainties to rigorously compare results on body kinetics. Therefore, future studies should provide recommendations on how to assess and propagate modeling and measurement uncertainties in order to allow a more rigorous comparison of results across different studies.

4.4 Muscle activity estimation

Fourteen studies reported results acquired using EMG, ten of which focused on slope propulsion. All studies but one normalized EMG data acquired during locomotion by EMG data of maximum voluntary contraction, hence reported muscle activity highly depends on the physical capacity of each participant. It is therefore difficult to give an estimate of activity for a specific muscle and barrier, as these results are highly dependent on both the subject’s physiology and propulsion strategy. Moreover, maximum voluntary contraction normalization is subjected to uncertainty under the risk of incorrectly testing for maximum voluntary contraction. In particular, when normalization is done improperly, there may be trials where recorded muscle activity is higher than its maximal value, characterized by results above 100% of maximum voluntary contraction. For instance, Requejo et al. reported mean muscle activity higher than 100% for eight subjects [46], but it might also be the case for some subjects in other studies in which the mean muscle activity was averaged over all the participants. One study reported un-normalized EMG data, which is therefore presented in Volts [38], preventing the comparison of muscle activity with other studies.

5. Conclusion

This review highlighted discrepancies in focus given to each environmental situation in the literature. Slope ascent and ground types were studied much more than cross-slope or curb ascent. Furthermore, the review evidences a lack of consensus on the parameters of interest to report and on the methods used to conduct experiments. These variations and lack of consensus make it impossible to cross-reference studies to compare situations. Nevertheless, for each environmental barrier, this review provides an unprecedented overview of its current biomechanical assessment through the report of numerical values of all biomechanical parameters retrieved from the relevant literature (in tables provided in supplementary material).

At the end of this review process, we recommend a more systematic approach when reporting materials, methods, and results for the reflection of the difficulty of any environmental barrier encountered in MWC locomotion: (i) effectively reporting barriers’ lengths, grades, or heights; (ii) striving for standardization or a report of the approach conditions of the barrier, such as velocity, especially on curbs; (iii) reporting the configuration of the used MWC, and if it was fitted to the subject’s morphology; (iv) reporting rotation sequences for the expression of moments and kinematics, and when used, the definition of the musculoskeletal model; (v) when possible, reporting measurement uncertainties and model reconstruction errors.

Biomechanical parameters definition.

(DOCX)

Click here for additional data file.

PRISMA checklist.

(DOCX)

Click here for additional data file.

Study and results review for slopes, cross-slopes, curbs, and ground types.

(XLSX)

Click here for additional data file.

26 Dec 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Lucas van der Woude

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

Apologies for the delayed return of your manuscript and the accompanying reviews. The reviewers appreciate the manuscript and the work it has entailed. They are positive, yet have a number of major suggestions to improve the manuscript and its readability. Other than that they have provided extensive smaller remarks.

Major points are the overlap between the results text section and the table info. A more complementary text to the central role of the tables is suggested. Use SI-system units throughout the manuscript. What is the role for Power output (W), external or user-related power production in your analyses? I miss this info in the tables. Add definitions The conclusion should be short and concise, not repeating or adding to the discussion. Did the authors conduct a quality assessment of the different papers, which is often typical for the systematic review? Be consistent in terminology, abbreviations between the different tables and sections.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This review gives a very good and extensive overview on the existing literature of manual wheelchair propulsion while overcoming different barriers in daily life of wheelchair users. The results are presented in detail in the tables which is very convenient for the interested reader to extract the information of interest. However, the presentation of the results in the text is sometimes lengthy and a repetition of what is presented in the table. This makes the review difficult to read. In my view, the review would benefit from shortening the text in the results section and referring to the tables instead. Detailed comments to the manuscript are written below.

Introduction:

Line 55: delete (MWU)

Line 55-57: This sentence is not clear. What is the link between ICF and physical demands associated with barriers? Please rephrase.

Line 67-68: Sentence is not complete.

Line 69: replace "in the same study" with "in each study".

Line 78-80: these two sentences are not well placed, they should be placed earlier in the introduction not just at the end where it's out of context.

Line 80: to be specific, add "manual" to handrim propulsion

Methods:

Line 83: replace "consisted of" with "consisted in". There are other small language flaws in the manuscript, please let it check by a native English speaker.

Line 95-98: did you also try to include "inclination" as an addition to "slope"? This might have resulted in more studies.

Line 118: S1 Appendix: The appendix does not includ all parameters mentioned here, and not all parameters from the appendix are mentioned here. Please make it consistent.

Regarding the S1 Appendix:

- Contact angle: you define it as "angle distance travelled by the non-dominant hand on the handrim during the push phase". Why is it limited to the non-dominant hand?

- Rate of rise: your definition is not correct, not detailed. Is it the mean resultant force divided by the contact time of the whole cycle?

Besides that, in literature there are different definitions:

• taking the derivative of FR with respect to time and then determining the maximum value during the first third of the stroke

• (first) peak of the resultant force, divided by the time to reach the peak.

Therefore, write it more precisely in your definition.

- Fraction of effective force: There are also different definitions used, either FEF = (Mwheel·r−1)·Ftot−1 or Ftan2/Fres2

Therefore, be more specific in your definition

Results

Line 133: include a reference to table 1.

Table 2: van Drongelen 2005: information on kinematics is missing. I did not check the whole table on missing information, but just spotted this one. So please check again whether everything is included here.

Table 3: - what does * mean?

- what does sEMG mean?

- Since it is a new table, explain the abbreviations below. Same accounts for the following tables.

Table 5: why is in this table height and weight of the participants included, but not in the others? Keep it consistant over the tables.

Line 156 and further: As already indicated in the general comments, all the information you give in here is already displayed in the table. The first part of the methods, where you summarize the range of participant number, the studied population and the range of slopes measured, is ok (although also not really needed since it is already indicated in the table). The second part on how the measurements were performed is obsolete, since you list again what is mentioned in the table (i.e. 19 articles measured kinematics, 15 article measured kinetics,..). Please shorten this part a lot or even delete it and refer to the table. The same accounts for cross-slope, curb and ground types.

Line 157: it should be referenced to table 2 instead of table 1.

Line 187 and further: In general, I like the overview on the results in the appendix, it gives a good overview. If possible it would be good to take the tables out of the appendix and place it in the manuscript Then you could also shorten the text of the results part.

Line 187: Please refer to appendix Table S2, also further on in the results section.

Line 188: replace "speed decreases with the grade of the slope" with "speed decreases with increasing slope"

Line 193: replace "increase with the grade" with "increase with increasing slope"

Line 215: replace "increase in internal rotation with the slope" with "increase in internal rotation during propulsion on a slope"

Line 241: should be referred to table 3 instead of 2.

Line 250: before you always called it instrumented wheel instead of handrim dynanometer. Please use one of the labels conistently throughout the manuscript

Line 270: "on speed, cycle frequency and duration of push and recovery phase" can be deleted. In general, write more concisely and omit the information that is not really necessary or is a repetition. This makes it easier and more convenient for the reader.

Line 274-275: give reference to the study and Appendix

Line 296: should be referred to table 4 instead of 3.

Line 296: "real" MWC sounds odd, please delete.

Line 313: add reference to appendix.

Line 344: should be referred to table 5 instead of 4.

Line 365: this is another example of sentences that are not needed, a reference to the appendix is more meaningful.

Discussion

Line 408: "investigated in literature" instead of "from literature"

Line 409-414: This information belongs to the results, there you already listed how many studies investigated which barrier. It does not have to be mentioned again, especially not with numbers of studies. You might want to say in one sentence what has been studied most.

Line 431: safety belts were not used in all of the studies conducted on treadmills

Line 443: Number of studies reporting kinematics and kinetics is for the results section, not for the discussion.

Line 467: include reference to the study.

References

Reference 1: what kind of a reference is this? if it is a website, please indicate the url.

Reference 11: Title is written twice in the reference

Figure 1: indicate in the figure why the records (n=1098) they are excluded

S2 Table: Check the table, units are not always indicated. For example in Slope: MWC speed, recovery phase duration, contact angle, sometime it's indicated after the numbers instead of in the the title line (i.e. body kinetics), or both in the title line and after numbers (Peak un-normalized EMG). Please make it consistent.

Reviewer #2: Comments to editor:

The authors presented a nicely written systematic review on environmental barriers and manual wheelchair biomechanics. The introduction, rationale and methods of this study are presented in a clear way. The results still have quite some redundant information, which is already present in the tables, it is recommended to shorten parts of the results to improve the readability. The discussion section raises very interesting points, only the structure needs some finetuning. The conclusion should be shorter and to the point. A recommendation of Acceptance with Major Revision has been given.

General comments:

A nicely written systematic review, which presents the work in a very detailed and extensive way. In more detail:

Introduction: the introduction is written well, only the ending could use some finetuning. Methods: Methods section is well presented and only lacks a clear inclusion/exclusion overview.

Results: The results were separated for the four areas (slopes, cross-slopes, curbs and ground type), as well as methods and results.

Regarding the methods on slopes, cross-slopes, curbs and ground type: It is unnecessary to present all results on methods this extensively. Most information can also be found in the presented tables, which already cover a lot of information. Removing parts in these sections might help with the readability of the whole paper.

On the Results on slopes, cross-slopes, curbs and ground type part: Adding subheadings for the subcategories: spatio-temporal, kinematics, kinetics and muscle activity are recommended.

Discussion: In the discussion section a lot of important information is discussed. It feels like a lot of separate interesting discussion points are raised but the structure is a bit lacking. The transitions from one paragraph to another, as well as the general structure should use some finetuning.

Conclusion: In this section a lot of information is discussed, which can either be removed or combined with some of the discussion points. Try to get the conclusion short and to the point.

Smaller general points: The consistency in the use of references, sometimes all references are listed extensively, but there are cases in which some the references are missing, examples are Lines 380-382 and Lines 426-442.

Maybe a figure with the 4 environmental barriers might be a good addition to have a clear overview and explanation of the 4 barriers (cross-slope might be unclear to some naïve readers).

Specific comments:

Abstract:

Page 2, Lines 33-35: Make the conclusion at the end a bit stronger, something similar to: “It is recommended to standardize the procedure, studying a wide set of…”

Introduction:

Page 3, Lines 45-46: The sentence seems confusing, maybe something similar to: “Despite the improvement of overall accessibility of public areas…”

Page 3, Lines 46: Recommended to start a new paragraph when ICF is introduced

Page 3, Lines 61-62: Last sentence of this paragraph feels a bit redundant

Page 4, Line 75: Remove the word “therefore” in this sentence

Page 4, Lines 78-80: This paragraph is unnecessary, the end of the previous paragraph is a more proper ending of the introduction

Methods:

Page 4, Line 83: Change the sentence a bit: “The present study conducted a systematic review to identify…”

Page 5, Lines 105-106: Remove “Appended to this paper…” and add “(S1 table)” add the end of “…(PRISMA) Checklist [24]”

Page 5, Lines 106-110: Separate into inclusion/exclusion criteria, inclusion: biomechanical data, experimental work, English languages, exclusion: electric and power assisted wheelchairs etc.

Page 5, Line 111: Remove the word “three” from the sentence

Page 5, Lines 114-116: Very nice and detailed description of the parameters

Page 6, Line 117: Change towards: “A more detailed definition of the biomechanical parameters can be found in S1 Appendix”

Page 6, Line 118-119: Sentence feels a bit redundant

Results:

Page 6, Lines 129-132: Split the sentence into two

Page 6, Lines 134-137: Cut the sentences a bit, similar to: “Experimental design, acquisition methods and measurement tools differed between studies. The MWC was propelled overground… …ergometer. Kinematics were either recorded…”.

Page 6, Lines 137-138: What is the use of “on the contrary”?

Page 6, Lines 140-142: Try to rephrase the sentence a bit, it is unclear what is part of the tables and what of the supplementary material. Maybe similar to: “Table 1 represents the synthesis of all studies, the remaining tables (2-5) summarize the study review for the four categories (slope, cross-slope, curb, ground type). … are appended as supporting information (S2 Table)”.

Page 18-19, Methods on slopes: See general comment

Page 19, Line 188-189: Sentence can be removed (“Results on cycle frequency…) and combined with the next sentence.

Page 19, Lines 191-194: Try to shorten the sentence: “…increase with the grade in SCI subjects, but unaffected with AB subjects.”

Page 19: Lines 201-203: What kind of differences were found?

Page 20: Line 212: Remove the part “in the study where they were reported”.

Page 20: Lines 219-220: Remove the part: “…in all studies reporting these data”.

Page 21: Line 233: Change to past sense

Page 21-22, Methods on cross-slopes: See general comments

Page 22: Lines 266-267: Is an interpretation sentence, better be part of the discussion

Page 22, Lines 270-274: Try to shorten this sentence, parts at the beginning of the sentence can be removed.

Page 23-24, Methods on curbs: See general comments

Page 23, Line 296: Remove the word “real”

Page 25-26, Methods on ground types: See general comments

Page 27, Lines 380-382: The references are missing

Discussion

Page 27-28, Lines 401-406: This paragraph only repeats the aims, without actually giving any relevant information, combining some of this information with the next paragraph might be a solution.

Page 28, Lines 417-425: This part can be shortened a bit, there is some repetition, especially with the last sentence.

Page 29, Lines 428-430: Future and past sense are both used in this sentence

Page 29-30, Lines 443-454: The paragraph starts with summarizing the number of articles investigating kinematics/kinetics. Despite main focus of this paragraph seems on the differences in acquisition methods. Try to get the most relevant information at the start/end of the paragraph.

Page 30, Lines 469-476: This feels like a separate paragraph, since the main focus of this paragraph was on the use of instrumented measurement wheels on both side of the MWC.

Page 31, Lines 486-490: It is unclear why this sentence is presented in between brackets, would make the argument stronger to present it as an example.

Page 32, Lines 505-508: Try to combine these sentences into one strong sentence.

Page 32, Lines 509-516: This paragraph does not add too much to the whole review and could, feels like lose paragraph at the end of the discussion, see also general comments

Conclusion:

Page 32-34: A conclusion is preferably short and to the point. The conclusion now consists of around five paragraphs in which also a lot of recommendations are given. Some of these recommendations are already discussed previously and don’t necessarily strengthen the paper.

Page 32, Lines 519-524: The first paragraph summarizes/repeats what has already been repeated multiple times

Page 33, Line 526: This is one of the main lines to be used for the conclusion.

Page 33, Lines 530-536: This all can be said in one sentence in the discussion part.

Page 33, Lines 537-540: Interesting point, also combine with some of the paragraphs in the discussion part (Future recommendation)

Page 33, Lines 541-543: Repeating the fact that cross-slopes and curbs should be studies, but now focused on asymmetrical propulsion.

Page 33-34, Lines 544-551: This seems like one of the main parts of the conclusion of the paper

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Ursina Arnet

Reviewer #2: Yes: Riemer Vegter & Thomas Rietveld

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

10 Feb 2022

Dear reviewers, Editor,

Thank you for your encouraging comments while reviewing the manuscript and for allowing us to improve its content.

Please find below how we have addressed each of the reviewers’ comment. The reviewers’ comments are reported in black, whereas the answers are reported in blue.

A version of the revised manuscript has also been uploaded. Changes are highlighted in yellow in the revised version of the manuscript.

Kindest regards,

Théo Rouvier, on behalf of all the authors of the manuscript entitled "Manual wheelchair biomechanics while overcoming various environmental barriers: a systematic review”

Reviewer 1:

General comments:

This review gives a very good and extensive overview on the existing literature of manual wheelchair propulsion while overcoming different barriers in daily life of wheelchair users. The results are presented in detail in the tables which is very convenient for the interested reader to extract the information of interest. However, the presentation of the results in the text is sometimes lengthy and a repetition of what is presented in the table. This makes the review difficult to read. In my view, the review would benefit from shortening the text in the results section and referring to the tables instead. Detailed comments to the manuscript are written below.

Thank you for your appreciation and for your detailed review of our work. You will find hereafter responses to your detailed comments. Following your advice, the text in the “results” section was shortened and references to the tables were added.

Introduction:

Line 55: delete (MWU)

“(MWU)” was deleted from line 47 but the abbreviation was redefined at line 130 of the revised manuscript since it is used in the “results” and “discussion” sections.

Line 55-57: This sentence is not clear. What is the link between ICF and physical demands associated with barriers? Please rephrase.

We believe that ICF could be better implemented for MWC users if clinicians were able to adapt training programs in accordance with the difficulty of a barrier with regards to a user’s capabilities. The quantification of the physical demands associated with barriers could be used as a quantification of a barrier’s difficulty. The paragraph was rephrased. (l. 55-60)

Line 67-68: Sentence is not complete.

The sentence was rephrased.

Line 69: replace "in the same study" with "in each study".

This was corrected (line 67 of the revised manuscript).

Line 78-80: these two sentences are not well placed, they should be placed earlier in the introduction not just at the end where it's out of context.

The sentences have been moved up in the text at the beginning of the “Methods” section (lines 82-86 of the revised manuscript).

Line 80: to be specific, add "manual" to handrim propulsion

The adjective “manual” was added to be more specific, following the reviewer’s comment. (line 86 of the revised manuscript).

Methods:

Line 83: replace "consisted of" with "consisted in". There are other small language flaws in the manuscript, please let it check by a native English speaker.

The sentence was rephrased. The manuscript was checked by an English speaker and language flaws have been corrected throughout the manuscript.

Line 95-98: did you also try to include "inclination" as an addition to "slope"? This might have resulted in more studies.

Following the reviewer’s suggestion, we performed a new search on PubMed with the addition of the term “inclination”, which resulted in 21 new results compared to the initial PubMed search, among which some had already been retrieved with the initial Scopus search. Based on the screening on the title, most are related to seat or backrest inclinations, and not to the ground inclination (slope grade). Finally, only 1 reference appeared to be relevant, but it did not meet the inclusion criteria based on the screening on the abstract.

We agree with the reviewer that it remains possible that we missed some papers relative to very specific situations due to the absence of very specific keywords, but we believe that the global conclusions drawn on methodologies will not be challenged.

Line 118: S1 Appendix: The appendix does not include all parameters mentioned here, and not all parameters from the appendix are mentioned here. Please make it consistent.

Thank you for your comment, the parameters defined in the appendix and those mentioned in the “methods” section are now consistently the same.

Results:

Line 133: include a reference to table 1.

A reference to table 1 was added.

Table 2: van Drongelen 2005: information on kinematics is missing. I did not check the whole table on missing information, but just spotted this one. So please check again whether everything is included here.

Thank you for noticing this mistake. Information on kinematics recording in van Drongelen 2005 was added to Table 2. All the tables were double checked following your comment.

Table 3: - what does * mean?

* Represents data that was extrapolated by the authors using the data given by the study. The signification of “*” was added to the tables, in the caption.

- what does sEMG mean?

sEMG means Surface EMG, this was detailed in the tables’ captions.

- Since it is a new table, explain the abbreviations below. Same accounts for the following tables.

All abbreviations are now defined in the caption of each table.

Table 5: why is in this table height and weight of the participants included, but not in the others? Keep it consistant over the tables.

It is true that data reported in tables was not consistent. We initially chose to report height and weight of subjects for the studies focusing on ground types because some of these studies normalized kinetic data on subject height and weight. But it is indeed confusing to report such data only for those studies and we don’t consider it crucial to the comprehension of our systematic review. For these reasons we normalized the tables and removed data on subject height and weight.

Line 156 and further: As already indicated in the general comments, all the information you give in here is already displayed in the table. The first part of the methods, where you summarize the range of participant number, the studied population and the range of slopes measured, is ok (although also not really needed since it is already indicated in the table). The second part on how the measurements were performed is obsolete, since you list again what is mentioned in the table (i.e. 19 articles measured kinematics, 15 article measured kinetics,..). Please shorten this part a lot or even delete it and refer to the table. The same accounts for cross-slope, curb and ground types.

Thank you for your comment. Following advice from both reviewers, the presentation of the experimental methods of the reviewed studies was shortened in the corpus of the manuscript. Full detail of the methods can be found in tables 2-5, and only key information was retained in the text.

Line 157: it should be referenced to table 2 instead of table 1.

This was corrected.

Line 187 and further: In general, I like the overview on the results in the appendix, it gives a good overview. If possible it would be good to take the tables out of the appendix and place it in the manuscript Then you could also shorten the text of the results part.

We agree with the reviewer’s that the tables provide a nice overview of the results, and, initially, it was our aim to place them in the manuscript.

However, given the lack of consensus in the literature on the outcome parameters of interest, it is difficult to provide tables that would fit over a page. The tables have indeed a large number of columns and few rows, which compromise their insertion in the text. Therefore, we chose to summarize the results in the paper and to refer to appendix table S2.

Line 187: Please refer to appendix Table S2, also further on in the results section.

We added a reference to the S2 Table in the slope sections, as well as in other sections (l.180-181 of the revised manuscript).

Line 188: replace "speed decreases with the grade of the slope" with "speed decreases with increasing slope"

The sentence was rephrased following the reviewer’s suggestion (l.184).

Line 193: replace "increase with the grade" with "increase with increasing slope"

The sentence was rephrased following the reviewer’s suggestion (l. 188).

Line 215: replace "increase in internal rotation with the slope" with "increase in internal rotation during propulsion on a slope"

This was changed to “increase in internal rotation with increasing slope”.

Line 241: should be referred to table 3 instead of 2.

The reference to the table was corrected. Thank you for spotting this error.

Line 250: before you always called it instrumented wheel instead of handrim dynanometer. Please use one of the labels conistently throughout the manuscript

Thank you for noticing this inconstancy. We replaced three usages (all of them) of “handrim dynamometer” by “instrumented wheel”.

Line 270: "on speed, cycle frequency and duration of push and recovery phase" can be deleted. In general, write more concisely and omit the information that is not really necessary or is a repetition. This makes it easier and more convenient for the reader.

Thank you for your comment which allowed us to increase the readability of our manuscript.

Line 274-275: give reference to the study and Appendix

Reference to the study and to the appendix were added (appendix reference l. 262-263 and study reference l. 268).

Line 296: should be referred to table 4 instead of 3.

This was corrected (l.293).

Line 296: "real" MWC sounds odd, please delete.

This was deleted.

Line 313: add reference to appendix.

A reference to the appendix table was given (l. 306-307).

Line 344: should be referred to table 5 instead of 4.

This was corrected.

Line 365: this is another example of sentences that are not needed, a reference to the appendix is more meaningful.

These sentences were deleted (reference to the appendix l. 352-353).

Discussion

Line 408: "investigated in literature" instead of "from literature"

The sentence was rephrased following the reviewer’s suggestion (l. 392).

Line 409-414: This information belongs to the results, there you already listed how many studies investigated which barrier. It does not have to be mentioned again, especially not with numbers of studies. You might want to say in one sentence what has been studied most.

The sentence was shortened following the reviewer’s suggestion (l. 393-394).

Line 431: safety belts were not used in all of the studies conducted on treadmills

Thank you for your comment. The sentence was changed, and now states that safety belts were sometimes used in studies conducted in treadmills (l. 444-447).

Line 443: Number of studies reporting kinematics and kinetics is for the results section, not for the discussion.

The sentence was shortened following the reviewer’s suggestion.

Line 467: include reference to the study.

The reference to the study was added (l. 451).

References

Reference 1: what kind of a reference is this? if it is a website, please indicate the url.

The reference was corrected to display authors, organisation, and website url.

Reference 11: Title is written twice in the reference

This was corrected.

Figure 1: indicate in the figure why the records (n=1098) they are excluded

The 1098 records were excluded after either a screening on title or abstract. These records were excluded based on our inclusion/exclusion criteria. These criteria are detailed in section (2.2 – Article selection) of the manuscript. “The inclusion criteria were original study or systematic review, study written in English and experimental work. Exclusion criteria were articles about electric wheelchairs, powered-assisted wheelchairs, sports wheelchairs, other propulsion systems than manual handrim, hemiplegia-pattern propulsion”.

It was added to the PRISMA flow-chart that these records were excluded after title or abstract screening.

Reasons detailing why studies were excluded in the full-text screening were also added to the flow-chart.

S2 Table: Check the table, units are not always indicated. For example, in Slope: MWC speed, recovery phase duration, contact angle, sometime it's indicated after the numbers instead of in the the title line (i.e. body kinetics), or both in the title line and after numbers (Peak un-normalized EMG). Please make it consistent.

Thank you for this comment. It is true the reporting of units was inconsistent. All units are now consistently reported. Units for all values except those that are normalized are reported in the title line. Units for values that are normalized are always reported in each cell, because studies did not always use the same normalization criterion, even on the same biomechanical parameter, and therefore had different units for the same parameters.

Appendix

- Contact angle: you define it as "angle distance travelled by the non-dominant hand on the handrim during the push phase". Why is it limited to the non-dominant hand?- Rate of rise: your definition is not correct, not detailed. Is it the mean resultant force divided by the contact time of the whole cycle?

Besides that, in literature there are different definitions:

• taking the derivative of FR with respect to time and then determining the maximum value during the first third of the stroke

• (first) peak of the resultant force, divided by the time to reach the peak.

Therefore, write it more precisely in your definition.

- Fraction of effective force: There are also different definitions used, either FEF = (Mwheel·r−1)·Ftot−1 or Ftan2/Fres2

Thank you for your comment, the definition of these biomechanical parameters was detailed. We presented both definitions for the two latter parameters, and indicated which one was more present in the reviewed articles.

Reviewer 2:

A nicely written systematic review, which presents the work in a very detailed and extensive way. In more detail:

Thank you for the appreciation of our work and your thorough review of the manuscript.

Introduction: the introduction is written well, only the ending could use some finetuning.

The last sentences of the introduction were moved up in the introduction so as to provide a better ending of the introduction, following the reviewer’s comment.

Methods: Methods section is well presented and only lacks a clear inclusion/exclusion overview.

Inclusion and exclusion criteria were separated in the Methods section following the reviewer’s suggestion.

Results: The results were separated for the four areas (slopes, cross-slopes, curbs and ground type), as well as methods and results.

Regarding the methods on slopes, cross-slopes, curbs and ground type: It is unnecessary to present all results on methods this extensively. Most information can also be found in the presented tables, which already cover a lot of information. Removing parts in these sections might help with the readability of the whole paper.

On the Results on slopes, cross-slopes, curbs and ground type part: Adding subheadings for the subcategories: spatio-temporal, kinematics, kinetics and muscle activity are recommended.

Thank you for your comment. Following your recommendations, the presentation of the reviewed methods in the “Results” section was shortened to only contain information we deemed key for the reader.

Also, as suggested, subheadings were added in the “Results” section for the different biomechanical parameters (spatio-temporal, kinematics, kinetics, muscle activity).

Discussion: In the discussion section a lot of important information is discussed. It feels like a lot of separate interesting discussion points are raised but the structure is a bit lacking. The transitions from one paragraph to another, as well as the general structure should use some finetuning.

Thank you for your comment, we have worked on improving the flow of the “Discussion” section. Paragraphs were re-arranged and subheadings were added to place a greater emphasis on the different key points addressed in the section. We think that this new general structure of the discussion improves its readability compare to the previous draft.

Conclusion: In this section a lot of information is discussed, which can either be removed or combined with some of the discussion points. Try to get the conclusion short and to the point.

Thank you for your comment and suggestion, the conclusion indeed felt too long. Points addressed in the discussion were either shortened or moved up into the “Discussion” section. The conclusion is now only two paragraphs short.

Smaller general points: The consistency in the use of references, sometimes all references are listed extensively, but there are cases in which some the references are missing, examples are Lines 380-382 and Lines 426-442.

Thank you for noticing these missing references. References were added in both paragraphs as suggested (l. 365, 443-45 of the revised manuscript).

Maybe a figure with the 4 environmental barriers might be a good addition to have a clear overview and explanation of the 4 barriers (cross-slope might be unclear to some naïve readers).

Thank you for your suggestion. Thank you for your comment. Pictures of the recreated environmental barriers were moved from the supplementary materials (S1-3 Figs) to the corpus of the manuscript, and fused in one figure

Abstract:

Page 2, Lines 33-35: Make the conclusion at the end a bit stronger, something similar to: “It is recommended to standardize the procedure, studying a wide set of…”

Following the reviewer’s comment, the concluding sentence of the abstract was rephrased to “It is recommended to standardize the procedure when studying various physical environmental situations and to systematically report MWC configuration(s). Furthermore, a wider set of situations should be studied” (l. 33-36).

Introduction:

Page 3, Lines 45-46: The sentence seems confusing, maybe something similar to: “Despite the improvement of overall accessibility of public areas…”

The sentence was rephrased following the reviewer’s suggestion (l. 45-47).

Page 3, Lines 46: Recommended to start a new paragraph when ICF is introduced

A new paragraph dedicated to the ICF was started (l. 48-60).

Page 3, Lines 61-62: Last sentence of this paragraph feels a bit redundant

Indeed, the last sentence was removed following the reviewer’s comment.

Page 4, Line 75: Remove the word “therefore” in this sentence

The word “therefore” was removed from the sentence.

Page 4, Lines 78-80: This paragraph is unnecessary, the end of the previous paragraph is a more proper ending of the introduction

The sentences of the last paragraph were not deleted as some studies focusing on other means of propulsion were retrieved from the literature and are not discussed in the present review. However, following both reviewer’s comments, they were moved in the Methods section. The introduction now ends on: ” To fill this gap, the purpose of this study was to identify and synthesize data and experimental methods from the literature on the biomechanics of MWC propulsion for various and frequent environmental barriers that are daily encountered by MWC users.” (paragraph moved to l. 82-86).

Methods:

Page 4, Line 83: Change the sentence a bit: “The present study conducted a systematic review to identify…”

The sentence was changed following the reviewer’s comment (l. 82).

Page 5, Lines 105-106: Remove “Appended to this paper…” and add “(S1 table)” add the end of “…(PRISMA) Checklist [24]”

This was changed (l. 108).

Page 5, Lines 106-110: Separate into inclusion/exclusion criteria, inclusion: biomechanical data, experimental work, English languages, exclusion: electric and power assisted wheelchairs etc.

The criteria for article selection into the present systematic review were separated into inclusion and exclusion criteria following the reviewer’s comment (see l. 109-112 of the revised manuscript)

Page 5, Line 111: Remove the word “three” from the sentence

Following the reviewer’s suggestion, “Three” was removed from the sentence as it was already cited earlier in the manuscript that three authors participated in the article selection process.

Page 5, Lines 114-116: Very nice and detailed description of the parameters

Thank you for your comment

Page 6, Line 117: Change towards: “A more detailed definition of the biomechanical parameters can be found in S1 Appendix”

The sentence was changed (l. 119).

Page 6, Line 118-119: Sentence feels a bit redundant

The sentence was moved to the section 2.1 of the manuscript. Indeed, it was a bit out of context in section 2.2 “Article selection” but it seemed relevant to the authors to insist of the methodological aspects of the studies as well as on the obtained results (l. 95-96).

Results:

Page 6, Lines 129-132: Split the sentence into two

The sentence was split into two sentences and a reference to Table 1 was added following both reviewers’ comments (l. 129-133).

Page 6, Lines 134-137: Cut the sentences a bit, similar to: “Experimental design, acquisition methods and measurement tools differed between studies. The MWC was propelled overground… …ergometer. Kinematics were either recorded…”.

The sentence was cut into three (l. 134-138).

Page 6, Lines 137-138: What is the use of “on the contrary”?

“On the contrary” was removed from the sentence.

Page 6, Lines 140-142: Try to rephrase the sentence a bit, it is unclear what is part of the tables and what of the supplementary material. Maybe similar to: “Table 1 represents the synthesis of all studies, the remaining tables (2-5) summarize the study review for the four categories (slope, cross-slope, curb, ground type). … are appended as supporting information (S2 Table)”.

Following the reviewer’s suggestion, the sentence was rephrased to clarify the organization of the review: “An overview of the retrieved studies is provided in Table 1. A dedicated subsection to each investigated environmental barrier (slope, cross-slope, curb, ground type) synthesizes the experimental methods used in these studies (also reported in Tables (2-5) ) as well as the obtained biomechanical results with summarizing the experimental method. A compilation of the studies' numerical results is appended to this document as supplementary material (S2 Table).” (l. 139-143).

Page 18-19, Methods on slopes: See general comment

As answered before in general comments, the presentation of the experimental methods of the reviewed studies was shortened. Full detail of the methods can be found in table 2, and only key information was retained in the text.

Page 19, Line 188-189: Sentence can be removed (“Results on cycle frequency…) and combined with the next sentence.

This sentence was removed.

Page 19, Lines 191-194: Try to shorten the sentence: “…increase with the grade in SCI subjects, but unaffected with AB subjects.”

The sentence was shortened as suggested.

Page 19: Lines 201-203: What kind of differences were found?

Thank you for this comment, we indeed did not detail which differences were found. A sentence detailing these differences was added to the manuscript: ”Contact angle was higher on the same slope when experimenting on AB subjects, and seemed to remain constant with different grades of slope in AB subjects, whilst contact angle tended to decrease with increasing slope in MWU.” (l. 198-200).

Page 20: Line 212: Remove the part “in the study where they were reported”.

This part was removed.

Page 20: Lines 219-220: Remove the part: “…in all studies reporting these data”.

This part was removed.

Page 21: Line 233: Change to past sense

The verb “give” was changed to the past tense, thank you for spotting that error.

Page 21-22, Methods on cross-slopes: See general comments

See general comments for response: the review of the studies’ methods was shortened in the text. While tables still give full detail on the methods of the reviewed studies, the text now holds only key information.

Page 22: Lines 266-267: Is an interpretation sentence, better be part of the discussion

Thank you for your suggestion. The sentence was moved into the discussion section (4.1) (l. 395-397 of the revised manuscript).

Page 22, Lines 270-274: Try to shorten this sentence, parts at the beginning of the sentence can be removed.

The sentence was rephrased (l. 265-268).

Page 23-24, Methods on curbs: See general comments

See general comments for response: the review of the studies’ methods was shortened in the text. While tables still give full detail on the methods of the reviewed studies, the text now holds only vital information.

Page 23, Line 296: Remove the word “real”

This word was removed.

Page 25-26, Methods on ground types: See general comments

See general comments for response: the review of the studies’ methods was shortened in the text. While tables still give full detail on the methods of the reviewed studies, the text now holds only vital information.

Page 27, Lines 380-382: The references are missing

References were added (l. 365).

Discussion:

Page 27-28, Lines 401-406: This paragraph only repeats the aims, without actually giving any relevant information, combining some of this information with the next paragraph might be a solution.

Thank you for your comment. The paragraph was shortened to avoid repetition and allows for a better introduction of the review’s focus on experimental design issues (l. 386-388).

Page 28, Lines 417-425: This part can be shortened a bit, there is some repetition, especially with the last sentence.

This part of the paragraph was split into two, creating one more paragraph to present our point better. The penultimate sentence of the part was changed to get rid of repetition (l. 406-415).

Page 29, Lines 428-430: Future and past sense are both used in this sentence

The sentence was changed so the verbs are conjugated accordingly (l. 417-418).

Page 29-30, Lines 443-454: The paragraph starts with summarizing the number of articles investigating kinematics/kinetics. Despite main focus of this paragraph seems on the differences in acquisition methods. Try to get the most relevant information at the start/end of the paragraph.

Thank you for this comment. Along with rewiever 1’s suggestions, the first sentence of the paragraph was shortened to skip the summary of articles investigating kinematics/kinetics (l. 471-472).

Page 30, Lines 469-476: This feels like a separate paragraph, since the main focus of this paragraph was on the use of instrumented measurement wheels on both side of the MWC.

Indeed, not starting a new paragraph felt odd. This was corrected and a new paragraph was created discussing joint coordinate systems usage (new paragraph l. 497-507).

Page 31, Lines 486-490: It is unclear why this sentence is presented in between brackets, would make the argument stronger to present it as an example.

Thank you for your remark, the part in the brackets was moved out of brackets to be used as an example (l. 529-531).

Page 32, Lines 505-508: Try to combine these sentences into one strong sentence.

These two sentences were combined into one (l.466-468 of the revised manuscript).

Page 32, Lines 509-516: This paragraph does not add too much to the whole review and could, feels like lose paragraph at the end of the discussion, see also general comments

The paragraph was moved up in the discussion, where it was more relevant (section 4.3) and was slightly reformulated (l. 508-517).

Conclusion:

Page 32-34: A conclusion is preferably short and to the point. The conclusion now consists of around five paragraphs in which also a lot of recommendations are given. Some of these recommendations are already discussed previously and don’t necessarily strengthen the paper.

Thank you for your comment. Following your recommendations, the conclusion was shortened and is now straighter to the point.

Page 32, Lines 519-524: The first paragraph summarizes/repeats what has already been repeated multiple times

The first paragraph was deleted.

Page 33, Line 526: This is one of the main lines to be used for the conclusion.

Page 33, Lines 530-536: This all can be said in one sentence in the discussion part.

The paragraph was moved to the discussion section and split into different sentences adding weight to their respective new paragraphs.

Page 33, Lines 537-540: Interesting point, also combine with some of the paragraphs in the discussion part (Future recommendation)

The paragraph was moved in the discussion, in the section 4.1 discussing the environmental barriers studied in the retrieved literature (l. 406-409).

Page 33, Lines 541-543: Repeating the fact that cross-slopes and curbs should be studies, but now focused on asymmetrical propulsion.

This paragraph indeed felt like a repetition of what was already discussed previously, for this reason, it was deleted.

Page 33-34, Lines 544-551: This seems like one of the main parts of the conclusion of the paper

This paragraph is now the conclusion’s and our manuscript’s end (l. 545-552 of the revised manuscript).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

7 Apr 2022

25 Apr 2022

Dear reviewers,

Thank you for your encouraging comments while reviewing the revised manuscript and for allowing us to keep improving its content.

Please find below how we have addressed each of the reviewers’ comment. The reviewers’ comments are reported in black, whereas the answers are reported in blue.

A version of the revised manuscript has also been uploaded. Changes are highlighted in yellow in the revised version of the manuscript.

Kindest regards,

Théo Rouvier, on behalf of all the authors of the manuscript entitled "Manual wheelchair biomechanics while overcoming various environmental barriers: a systematic review”

Reviewer 1:

The authors have addressed my comments well and the manuscript is now much clearer and easier to read.

I only have some small comments to this revised version:

Thank you for your endearing comments. We will try to take the latter comments into account as best as we can.

Abstract:

In my view the conclusion stated in the abstract does not reflect the conclusion of your whole paper. Please include all recommendations given in the conclusion section (i-v), or state it more general. Furthermore, do you suggest to standardize the measurements or the reporting of it?

Thank you for your comment. We added the five recommendations made in the conclusion to the abstract. These recommendations include suggestions to both standardize the measurements (ii) and the reporting of it (i, iii, iv, v).

Methods:

Line 107: You refer to the PRISMA checklist when describing article selection. The prisma Checklist does not specify how to select articles but specifies what should be reported in a systematic review. So better refer to figure 2 or rewrite the first sentence.

Indeed, we should not have referred to the prisma checklist. The sentence now refers to figure 2, which was moved up in the manuscript to appear after this paragraph.

Line 110: You mention systematic review as a inclusion criteria, but another inclusion criteria is experimental work. If experimental work is a inclusion criteria, systematic reviews (as mentioned in the line above) will not be included.

Thank you for your comment. This error was fixed by modifying the phrasing of the inclusion criteria, which now state “features experimental results” instead of “experimental work”

Results:

Figure 2: In the description of figure 2 you indicate classification, but this is not addressed in figure 2.

Thank you for your comment. This was from an earlier draft of the figure. Classification was removed from the figure’s title.

Line 286: Please indicate whether EMG was measured on both sides or just on up-/ or downhill side.

EMG was measured only on the left side. This precision was added in the presentation of the methods and results of muscle activity in cross-slopes.

Line 344: include "wheel" after instrumented

This was corrected.

Discussion:

Line 440: include "due" in front of to.

This was corrected.

Conclusion:

Line 548-553: here you state what you recommend for further studies. Is this what you miss in the current studies, or is this all you would like to have reported in studies? And what for example about the participant characteristics?

Thank you for this question. These recommendations are based on what we miss in most studies of the literature, but also what we would like see reported in future studies. For example, some studies -but not all- reported barrier’s lengths and other characteristics, and we deem it is important that this practice be generalized. On the other hand, no reviewed study reported measurement uncertainties or model reconstruction errors, and we think it would be a truly interesting metric for researchers trying to compare their results with the state of the art. Participant characteristics are also a very interesting metric, but we feel that the state of the art has reached closer to a consensus, hence why we did not include them in our recommendations.

Reviewer 2:

Thank you for addressing all comments and the changes made to the manuscript. The changes have improved the readability of the manuscript. Only some minor adjustments are recommended.

A small note: Sometimes a tab is used to begin a new paragraph, sometimes not, there is no real consistency throughout the whole paper.

Thank you for your encouraging comments. Tabulations are now systematically used at the start of the first paragraph of a section.

Abstract:

Page 2, Lines 33-36: It would be recommended to make 1 strong final sentence: “It is recommended to standardize the procedure when studying various physical environmental situations, systematically report MWC configuration(s) and study a wider set of situations.”

Thank you for your comment. Alongside reviewer 1’s comments, we decided to end the abstract by the five recommendations listed in the conclusion. The sentence “Furthermore a wider set of situations should be studied” that was used to close the abstract was moved up to let our recommendations close the abstract.

Introduction:

The introduction now indeed ends with a logical flow towards the purpose of the study, well done.

Thank you very much for your comment.

Methods:

Page 5, Lines 109-112: The exclusion criteria are now nicely stated, only the inclusion criteria are still a bit too general (English language, original study, experimental work). It would be recommended to also add something similar to: “Studies investigating slope, cross-slope, curb and ground type in manual wheelchair users.”

Thank you for your comment. We changed Inclusion criteria and replaced “experimental work” by “features experimental results on slopes, cross-slopes, curbs, and ground types during MWC locomotion”.

Results:

Removing parts on the ‘methods’ section for all barriers improved the readability of the results part of the manuscript. Adding subheading really helped for the structure as well, great work.

Thank you for your comment.

Page 19, Line 180-182: These two sentences can be removed or moved a bit up, it is already previously stated on Page 7, Line 144 that the numerical results can be found in S2 Table. Now the same sentence is stated for all barriers on Page 22, Lines 263-264, Page 24, Lines 307-308 and Page 26, Lines 353-354. Stating that information once around line 144 will probably be enough, similar to: ‘Supplementary material on the detailed results on all environmental barriers (slopes, curbs etc.) can be found in S2 Table.’.

Thank you for your comment. Duplicates were removed, and the reference to Table S2 is now only listed in the overview of general results.

Page 22, Line 265: The subheading ‘3.2.2.1. Spatio-temporal parameters’ is too big, should be similar to the ‘3.2.2.2. Joint Kinematics’.

This was corrected.

Discussion

Page 27, Lines 387-389: This introductory statement might not necessarily be needed, in the next paragraph it is also summarized that environmental barriers were investigated.

The introductory statement was deleted.

Page 28, Lines 411-422: One nice combined paragraph can be made out of these three separate paragraphs with the last one only being one sentence.

Thank you for your comment. The three paragraphs were merged into one

Page 31, Lines 483-497: This has become a rather long paragraph to only state one thing: ‘Instrumented wheels should be used on both sides’. Especially lines 490-495 are repetitions of the first sentences of the paragraph.

Thank you for your comment. The paragraph indeed felt long. We shortened the paragraph to restrain from repeating ourselves, and focused the latter half of the paragraph on the fact that the only study that investigated cross-slopes using two instrumented wheels, reported results on the dominant and non-dominant sides of the user without expressing which one was downhill and uphill, which is also an important factor in the biomechanics of cross-slope locomotion

Page 32, Lines 509-511: Which studies are you referring to?

We added the precision as to which studies we are referring to: all studies presented in this review.

Page 32, Line 511: ‘… might have resulted in significant uncertainties’.

This was corrected

Conclusion:

The conclusion is a lot more shortened and to the point, well done.

Thank you for your comment.

Page 33, Line 538: The first sentence is a bit unclear, seems like there is missing some information.

The first sentence was rephrased: “This review highlighted discrepancies in focus given to each environmental situation in the literature”.

Page 33, Line 545: ‘… parameters retrieved from the relevant literature.’

This was corrected

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

26 May 2022

Manual wheelchair biomechanics while overcoming various environmental barriers: a systematic review

PONE-D-21-13523R2

Dear Dr. Rouvier,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Lucas van der Woude

Academic Editor

PLOS ONE

31 May 2022

PONE-D-21-13523R2

Manual wheelchair biomechanics while overcoming various environmental barriers: a systematic review

Dear Dr. Rouvier:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Lucas van der Woude

Academic Editor

PLOS ONE

Table 3

Study review for cross-slope investigation.

Study	Population		Experimental condition	Speed	slope grade (°), length	Kinematics	Kinetics	Muscle activity	Model
Holloway et al., 2015 [30]	7 (7M)	SCI1	overground	self-selected	0, 1.4° (length: not reported)	IMU2	Instrumented wheel (left side, 50Hz)	Surface EMG3 (3 Muscles)	Holzbaur et al., 2005 [56]
Richter et al 2007 [14]	25 (NA)	MWU4	motorized treadmill	self-selected	0, 3, 6° (35m*)	Motion capture system (100 Hz)	Instrumented wheel (downhill side, 200Hz)
Hurd et al., 2008 [15]	12 (11M 1F)	SCI	overground	self-selected	2° (length: not reported)		Instrumented wheel (both sides, 240Hz)
Symonds et al., 2016 [50]	13 (1F, 12M)	7 SCI, 6 AB5	overground	self-selected	0, 1.4° (8.4m, 7.2m)	IMU (50Hz)	1 Instrumented wheel (left side)	EMG (3 muscles)

1SCI: Spinal cord-injured;

2IMU: Inertial measurement unit;

3EMG: Electromyography;

4MWU: Manual wheelchair user;

5AB: Able-bodied.

*Data estimated by the authors of this review.

Table 4

Study review for curb investigation.

Study	Population		curb height (cm)	Kinematics	Kinetics	Muscle activity	Model
Lalumiere et al., 2013 [37]	15 (14M 1F)	SCI1	4, 8, 12cm	opto-electronic (4 cameras, 30Hz)	instrumented wheels (both sides, 240Hz)	Surface EMG2 (4 muscles)	Desroches et al., 2010 [59]
van Drongelen et al., 2005 [52]	5	SCI	10cm	opto-electronic (3 cameras, 100Hz)	instrumented wheel (left side)		Delft Shoulder and Elbow Model

1SCI: Spinal cord-injured;

2EMG: Electromyography.