A scoping review of cane use in people who have had a stroke

Abstract

BACKGROUND:

Various types of canes are frequently utilised in stroke rehabilitation to support walking ability by compensating for hemiplegic gait. However, Randomised Controlled Trial (RCT) design varies considerably in the area of cane use following stroke and there is no scoping review published to date on this topic. Thus, it is crucial for better evidence informed clinical care and future research that RCT evidence regarding specific assistive walking devices is examined.

OBJECTIVES:

Identify and map the types of canes that currently exist and review their impact on the rehabilitation of functional mobility post-stroke.

METHOD:

The following databases were searched: PubMed/MEDLINE, Web of Science, and CINAHL in Ebsco. Two authors independently screened 425 titles, identifying 16 RCT studies for inclusion in the review.

RESULTS:

16 studies were selected for review. Five different cane designs were identified, including one-point cane, three-point cane, quad cane, weight-supported feedback cane, and rolling cane. Twelve studies were crossover RCT and four were parallel RCT. Stroke phase varied widely among subjects included in the studies and outcome measures utilised and resultant findings are heterogeneous.

CONCLUSION:

More evidence is required to clearly indicate the effect of canes on people who have had a stroke and parallel long-term RCTs with follow-up assessment are lacking. Furthermore, there is a lack of research examining variations in new cane designs that specifically target motor function in people who have had a stroke at specific stages. Future research should aim to address the identified inconsistencies and knowledge gaps to facilitate novel and evidence-informed clinical care regarding assistive walking devices for post stroke rehabilitation.

Keywords

Stroke cane gait Randomised Control Trails (RCTs)scoping review

1 Introduction

The 2022 WHO Global Stroke Factsheet estimates that people who have had a stroke worldwide numbered 101 million [1]. Stroke remains the second leading cause of death and the third leading combined cause of death and disability in the world [2]. It has been observed that 65% of people who have had a stroke have reduced ambulation during the acute stage of stroke [3]. Only 30–50% of stroke survivors are capable of community ambulation [4].

Ambulation after stroke is altered by balance deficits, reduced sensation, decreased ability to properly time and grade normal levels of voluntary muscle contractions, involuntary increases in muscle tone, and changes in muscle mechanical properties [5]. These changes result in a hemiplegic gait characterised by reduced walking velocity, asymmetrical movement, shortened stride length, reduced standing phase and prolonged swing phase on the affected side [6]. Hemiplegic gait is associated with high instability and falls [7], decreasing cardiovascular fitness, and increased energy demand [8]. Thus, walking aids are frequently utilised to support walking ability by compensating for hemiplegic gait [9]. Approximately 32% to 76% of people with hemiplegia use at least one walking aid within three months post stroke [10].

In clinical practice, various walking aids, such as parallel bars, walkers, and different types of cane (single-point cane, 3-point or 4-point cane, weight-support feedback cane, cane with laser illuminators), are used for balance and gait training in people who have had a stroke [11]. Investigation of walking aids in stroke survivors have focused on cane use, as canes can help to increase the base of support for stroke survivors to provide postural stability and improve weight transfer ability in the standing position and walking. Moreover, canes are less detrimental to the temporal-distance characteristics of gait, more functional for ambulation in a challenging environment, and are less cumbersome and more aesthetically pleasing compared to other walking aids [11]. Additionally, walking frames demand bilateral upper-extremity function, which may be impaired in people who have had a stroke [12].

To date, the evidence on the effectiveness of cane use in stroke survivors is inconsistent. Laufer et al. reported that walking with a cane reduces postural sway and the chance of falling and improves independent walking [13]. According to the Choi et al. incorporating a cane into post-stroke gait training can improve muscle activation in the lower limbs [14]. In contrast, Neumann et al. demonstrated that the use of a cane increased weight-bearing in the nonparalyzed lower extremity by 40%, which interfered with weight training on the affected side and led to long-term gait asymmetry and inefficient gait [15]. Hamzat et al. showed that stroke survivors who used canes on a long-term basis had lower balance scores and less social participation than those who did not [16]. Due to broad study characteristic variations, there is little consensus or guidelines on how the prescription of canes should be individualised for patients at different stages of stroke [10].

A recent systematic review by Avelino and colleagues examined the immediate effects of canes on spatiotemporal parameters of walking [47]. However, study inclusion for this review was limited to cross-sectional within-group design, with outcomes primarily concentrating on the spatiotemporal parameters of walking. This review was also primarily focused on assessing study findings. Consequently, specific Randomised Controlled Trial (RCT) evidence was not assessed and the longitudinal effects and other aspects of cane research on walking remain indeterminate. To the author’s knowledge, no prior scoping review has been published regarding the incorporation of RCTs assessing cane use. Given that scoping reviews are designed to identify and broadly map a topic area [41], the current scoping review objectives are to assess the extent of parallel and cross-over RCT literature on the topic to identify and map the types of canes that currently exist and review their impact on the rehabilitation of functional mobility post-stroke. The following research question was investigated:

What are the key characteristics of included studies regarding cane type, participant rehabilitation phase, outcome measures and resultant findings?

2 Methods

This scoping review was conducted in accordance with Preferred Reporting Items for Systematic Reviews and Meta-analyses extension for scoping review (PRISMA-ScR), which was developed according to published guidance by the EQUATOR (Enhancing the Quality and Transparency Of health Research) Network for the development of reporting guidelines [17].

2.1 Search strategy and eligibility criteria

Searches were conducted in PubMed/MEDLINE (1965 to February 2024), Web of Science (1993 to February 2024), and CINAHL in Ebsco (1989 to February 2024). The text words contained in the titles and abstracts of relevant articles, and the index terms used to describe the articles were used to develop a full search strategy for CINAHL (see Table 1). The search strategy, including all identified keywords and index terms, was adapted for each included database. The reference list of all included sources of evidence were screened for additional studies.

Table 1
Example search strategy (CINHAL /EBSCO)

Search Query Results

S1 (MH “Stroke+”) 78,786

S2 TI (stroke* or “cerebral vascular accident” or “cerebrovascular accident” or “post-stroke”) OR AB (stroke* or “cerebral vascular accident” or “cerebrovascular accident” or “post-stroke“) 115,551

S3 S1 OR S2 132,336

S4 (MH “Canes”) 412

S5 (MH “Ambulation Aids+”) 1239

S6 TI (cane or canes or “walking aid” or “mobility aid” or “walking device” or “assistive device”) OR AB (cane or canes or “walking aid” or “mobility aid” or “walking device” or “assistive device”) 2073

S7 TI ((ambulation or ambulatory) N2 aid) OR AB ((ambulation or ambulatory) N2 aid) 149

S8 S4 OR S5 OR S6 OR S7 3226

S9 S3 AND S8 270

S10 (MH “Hemiplegia“) 6,804

S11 TI (hemipleg* or hemipare* or weak* or stiff* or spasticity or strength* or paralysis) OR AB (hemipleg* or hemipare* or weak* or stiff* or spasticity or strength* or paralysis) 229,531

S12 S10 OR S11 231,638

S13 S9 AND S12 111

Search	Query	Results
S1	(MH “Stroke+”)	78,786
S2	TI (stroke* or “cerebral vascular accident” or “cerebrovascular accident” or “post-stroke”) OR AB (stroke* or “cerebral vascular accident” or “cerebrovascular accident” or “post-stroke“)	115,551
S3	S1 OR S2	132,336
S4	(MH “Canes”)	412
S5	(MH “Ambulation Aids+”)	1239
S6	TI (cane or canes or “walking aid” or “mobility aid” or “walking device” or “assistive device”) OR AB (cane or canes or “walking aid” or “mobility aid” or “walking device” or “assistive device”)	2073
S7	TI ((ambulation or ambulatory) N2 aid) OR AB ((ambulation or ambulatory) N2 aid)	149
S8	S4 OR S5 OR S6 OR S7	3226
S9	S3 AND S8	270
S10	(MH “Hemiplegia“)	6,804
S11	TI (hemipleg* or hemipare* or weak* or stiff* or spasticity or strength* or paralysis) OR AB (hemipleg* or hemipare* or weak* or stiff* or spasticity or strength* or paralysis)	229,531
S12	S10 OR S11	231,638
S13	S9 AND S12	111

The titles, abstract, and then full texts were screened for articles which met the following selection criteria:

Related to the use of canes in people who have had a stroke;

Published randomised controlled trials (RCTs), including crossover RCTs;

Published in English.

Papers were excluded if they failed to meet any of the mentioned inclusion criteria.

2.2 Study selection

Following the search, all identified citations were collated and uploaded into EndNote20 (Clarivate Analytics, PA, USA) and duplicates removed. Titles and abstracts were then screened by two independent reviewers (RA, PB) for assessment against the inclusion criteria for the review. Following this, full text of selected citations was assessed in detail against the inclusion criteria by the two reviewers. The results of the search and the study inclusion process are presented in a Preferred Reporting Items for Systematic Reviews and Meta-analyses extension for scoping review (PRISMA-ScR) flow diagram (Fig. 1)

Fig. 1

PRISMA-ScR Flow chart illustrating the screening process.

2.3 Data extraction and synthesis

The full text of each included article was read, and key information was noted. Descriptive data from included studies were tabulated to compare content. Data was extracted on study design, study population, stroke phase of the patient, intervention and duration, inclusion/exclusion criteria, types of cane, height of cane, outcome measures and main findings. Following data extraction, a single reviewer independently synthesized data. All data were initially tabulated and subsequently reported descriptively to answer the key research questions of the review.

3 Results

All databases retrieved 206 potentially relevant articles after removing duplicates. After screening the titles and abstracts, 121 papers were excluded based on the inclusion criteria, and 85 full-text searchable papers were identified. Following this, another 68 papers were excluded, and 1 relevant paper was found from screening references of eligible articles, a total of 16 papers were included in the final scoping review. Full details regarding study characteristics of all included studies are included in(Table 2).

Table 2
Overview of selected studies

Reference Samle size Study design Types of cane Cane height level (with the subjects standing and the elbow extended) Inclusion/Exclusion criteria Intervention Outcomes Findings Phase

Li et al. 2018 30 Crossover RCT One-point cane, quad cane Patient dependent Inclusion criteria: (1) sustained unilateral hemiplegia caused by a first-time cerebral hemispheric stroke >3 months before the study; (2) could perform the STS movement with no assisting device and could ambulate with a cane for at least 10 m; (3) were able to stand with no external support for at least 15 s; (4) were able to understand and follow instructions; (5) had no history of major orthopedic operations within the previous 5 years; and (6) had no other orthopedic or neurological conditions that would affect the STS movement (1 day assessment) Sit-to-Stand (STS) test in seven situations in a random sequence: without a cane, three positions with a regular cane (parallel to the ankle, parallel to the 5th toe, and 10 cm in front of the 5th toe) Peak vertical ground reaction force (GRF) and maximum vertical cane support force during STS were recorded. Mean peak GRF: Significant differences between STS conditions with and without a cane (p = 0.000–0.040). Cane Support Force: Maximum support force varied by cane position, highest beside the ankle (8.62±4.01). Quad vs. Regular Cane: Maximum support force was significantly higher for all positions with a quad cane (p = 0.019 for 10 cm, p = 0.003 for toe, p = 0.000 for ankle). >3 months

Boonsinsukh 2009 39 Crossover RCT Adjustable-height cane, with force sensor Patient dependent Exclusion criteria: (1) had used a cane during walking before participating in this study, (2) had cognitive or language impairment, (3) had severe hemineglect, (4) had cerebral aneurysm, (5) had bilateral cerebral impairment, (6) had brainstem and cerebellar lesions, or (7) had impaired touch and pressure sensation on the nonparetic hand, (8) All subjects who had hemianopia, dizziness, or other symptoms indicating vestibular impairment; had taken medications that affect balance; had impaired sensation in the lower extremity; or had lower extremity deficits such as pain or contracture were also excluded from the study (1 day assessment) In the (light) touch contact condition, the subject’s goal was to use the cane while walking without the auditory buzzer being activated. A buzzer was activated whenever the subjects pressed on the cane with more than 400 g (4 N) of vertical force. In the force contact condition, the subject’s goal was to walk and push on the cane at the desired level. ML pelvic stability as measured by averaged peak-to-peak pelvic acceleration, muscle activation of bilateral tensor fascia latae (TFL), semitendinosus (ST), and vastus medialis (VM) using an electromyography system, and vertical cane force Cane Use: Force exerted on the cane was significantly higher in the force contact condition vs. touch contact (P < 0.001). Subjects with stroke applied force earlier and for a longer duration in the force contact condition (P < 0.001). Gait Speed &Pelvic Acceleration: No significant effect on walking speed or pelvic acceleration between force and touch contact conditions (P < 0.001). Temporal Gait Parameters: Single-limb support phase of the paretic leg was longer in the touch contact condition (P < 0.001), corresponding to longer muscle activation of the paretic vastus medialis. Muscle Activation: Tensor fascia lata, vastus medialis, and semitendinosus on the paretic side were activated more in the touch contact condition than the force contact condition (P < 0.001). Subacute

Tyson 2009 20 Crossover RCT One-point cane Greater trochanter Inclusion criteria: (1) Unable to walk for at least 2 weeks after their stroke, (2) Undergoing inpatient rehabilitation to restore walking, (3) Unable to mobilize independently without aids in every day life on the ward, (4) Able to step and practice walking during physiotherapy sessions although they may need support, (5) Able to walk 5 m without physical support (the minimum walking distance for the test procedure), (6) Able to give informed consent (1 day assessment) (1) Walking with no device (the control condition), (2) walking with a walking cane, measured using a five-meter walk test Functional mobility (functional ambulation categories FAC), walking impairments (speed 5MWT, step length of the weak leg 5MWT), and patients’ opinions self-report questionnaire. Walking with cane showed a significant improvement in functional mobility compared with walking with no device (P = .0001) with large effect size. Subacute

Polese 2011 19 Crossover RCT One-point Cane Not described Inclusion criteria: (1) Were ≥20 years old; (2) had a mean time of over 6 months since the onset of a unilateral stroke, (3) were regular users of walking sticks, but were able to walk without them for 10 m; (4) had no cognitive deficits, as determined by the Mini-Mental State Examination scores N24; (5) had no other neurological or orthopedic disorders. (1 day assessment) Four walking conditions: (i) with walking sticks at a comfortable speed; (ii) with walking sticks at a fast speed; (iii) without walking sticks at a comfortable speed; and (iv) without walking sticks at a fast speed. Linear kinematics (walking speeds) and angular kinematics (maximum joint angles), power, and work of the paretic hip, knee and ankle joints in the saggital plane Walking Speed: Increased during fast (P < 0.001) and comfortable (P = 0.001) walking with walking sticks. Joint Angles: No significant changes (P = 0.24–0.83). Power Generation: Increased in ankle plantar flexion, knee extension, and hip flexion (P < 0.01). Comparison of Walking Speeds: No greater changes between fast and comfortable walking for any outcome. Chronic

Beauchamp 2009 14 Crossover RCT One-point cane, quad cane Not describe Inclusion criteria: (1) first clinical diagnosis of stroke; (2) less than 3 months post-stroke; (3) ability to understandand follow oral commands; (4) ability to give informedconsent; (5) ambulatory without a gait aid before current stroke; (6) ambulatory without a gait aid for 6 m with close, non-contact supervision; and (7) “naïve” to use of an aid (i.e., patient had not integrated a cane into daily living) (1 day assement) Patinets were instructed to walk a distance of 6 m. Task conditions were as follows: (1) no cane task; (2) standard single-point cane task; (3) quad cane task. Gait-symmetry and walking velocity Symmetry Improvement: Standard cane improved symmetry in asymmetric patients (p = 0.028); quad cane did not (p = 0.36). No effect on symmetry in symmetric patients with either cane (standard cane p = 0.88, quad cane p = 0.32). Mean Velocity: No change with standard cane in asymmetric patients (p = 0.18); decreased with quad cane (p = 0.035). No change in velocity for symmetric patients with either one-point cane (standard cane p = 0.08, quad cane p = 0.17). Subacute

Boonsinsukh, 2011 62 Crossover RCT Force-sensing instrumented cane Radial styloid Iinclusion criteria: who had a stroke less than 1 year prior to the study. Exclusion criteria:Subjects were excluded from the study if they had (1) regularly used a cane during walkingbefore participating in this study; (2) cognitive or language impairments (Thai Mini-Mental State Examination score <21); (3) a cerebral aneurysm; (4) bilateral cerebral impairment; (5) brainstem and cerebellar lesions; (6) impaired touch and pressure sensation on the nonparetic hand, as measured by Semmes Weinstein Monofi lament (SWM) at the thenar eminence (fi lament <2.83 MN); or (7) severe hemineglect. Hemineglect was assessed by clinical observation; if there was any uncertainty, standard neuropsychological assessments, such as Bells Test, 18 were conducted. (1 day assessment) Subjects were instructed to walk along a 5-m walkway while using the cane in the 3 conditions: (1) no contact (NC), in which the cane was not allowed to touch the ground; (2) touch contact (TC) required subjects to exert less than 4 N of force through a cane that was held in the nonparetic hand. (3) force contact (FC), where the subject was instructed to exert as much force on the cane as needed. Trunk stability, percentages of gait cycle gait speed, muscle activation of bilateral tensor fascia latae (TFL) and vastus medialis (VM) using an electromyography system, and vertical cane force Trunk Acceleration: Greater in NC condition (P < 0.001); no difference between FC and TC conditions. Gait Speed and Percentage of Gait Cycle: No difference in gait speed or stance-swing durations among FC, TC, and NC conditions. Muscle Activation: Reduced in VM (P < 0.001) and TFL (P < 0.001) on the paretic leg during FC versus TC and NC conditions. No significant difference between TC and NC for paretic VM and TFL activation. <1 year

Buurke 2005 13 Crossover RCT One-point cane, quad cane Radial styloid Inclusion criteria: (1) Aged between 40 and 75 years (2) Had a first unilateral ischemic stroke and were able to walk without physical assistance: Functional Ambulation Categories (FAC≥3). (3) Sufficient cognitive abilities:Mini-mental State Examination (MMSE)≥22 (1 day assemement) Patients were tested in three different conditions: (1) walking without a walking aid, (2) walking with a cane and (3) walking with a quad stick Five patients used an ankle foot orthosis (AFO) Muscle activation of erector spinae, gluteus maximus, gluteus medius, vastus lateralis, semitendinosus, gastrocnemius and tibialis anterior muscles on the affected side, walking speed. Walking speed: Significantly lower with quad stick vs. cane Erector spinae burst duration: Decreased with cane vs. no aid. Tibialis anterior on-times: Later with quad stick vs. cane. Tibialis anterior burst duration: Decreased with both cane and quad stick vs. no aid. Vastus lateralis on-times: Later with quad stick vs. cane and no aid. Erector spinae average amplitude: Decreased with quad stick vs. no aid and cane. Gluteus maximus, medius, and vastus lateralis amplitude: Decreased with both cane and quad stick vs. no aid. Tibialis anterior amplitude: Decreased with quad stick vs. no aid and cane. They had a mean time since an ischaemic stroke of 205 days (range 47–385 days).

Deltombe 2020 32 Crossover RCT Quad cane, Wheeleo^® rolling cane The height can be adapted from 74 to 98 cm. Inclusion criteria were: age over 18 years; a history of hemiparesis as a consequence of a stroke or traumatic brain injury; use of a quadripod cane for walking; the ability to walk independently without additional help or support; and the capability to use the rolling cane for a few steps. (2 days assessment) Participants were assessed using a quad-ripod cane and a rolling cane. Walking speed during a 10-m walk test and a 6-min walk test; frequency of 2-step gait; physiological cost index; number of therapist interventions to control the balance; perceived exertion; and participant satisfaction Walking Speed: Increased with rolling cane during 10-m walk test (comfortable + 22%, p < 0.001; maximal+30%, p < 0.001) and 6-min walk test (+50%, p < 0.001). Distance: Increased with rolling cane by +49% in 6-min walk test (p < 0.001). 2-Step Gait Frequency: Improved with rolling cane. Physiological Cost Index, Perceived Exertion, Therapist Interventions: No change. Participant Satisfaction: Improved with rolling cane. NA

Jeong 2015 30 Crossover RCT One-point cane, quad cane, and a hemi-walker Ulnar process Inclusion criteria (1) first cerebrovascular accident verified by computed tomography or magnetic resonance imaging, (2) hemiplegic patient with gait disturbance, (3) independent ambulation with a cane over 10 meters, with or without orthosis, or a score of 4 or 5 on the Functional Ambulation Category (FAC) scale10 (no need for assistance), and (4) sufficient cognition to participate in the training, confirmed by a mini-mental state examination 11 (MMSE) score of 25 or higher. The exclu-sion criteria were: (1) neurologic co-morbidity other than stroke, (2) cardiovascular, musculoskeletal, or other medical condition that could influence gait or oxygen consumption, (3) refusal to participate in the study, or (4) recurrent stroke. A single-point cane, a quad cane, and a hemi-walker. At approximately the same time of day for three consecutive days, each subject completed a walk with a different type of walking aid. Energy expenditure (O₂ rate, mL/kg/min), energy cost (O₂ cost, mL/kg/m), and heart rate (HR) via a portable gas analyzer, a 10-meter walk test (10MWT), and a 6-minute walk test (6MWT). Energy Expenditure, Gait Endurance, Gait Velocity: Higher with single-point cane (p < 0.001 or p = 0.005) compared to other canes. Energy Cost: Lower with single-point cane (0.5±0.2 mL/kg/m) vs. others (0.6±0.2 mL/kg/m, p = 0.001). Heart Rate: No significant difference after Bonferroni correction (p≥0.05). Chronic

Tyson 1999 20 Crossover RCT One-point cane and three-point cane Radial styloid Inclusion criteria: Subjects with a chronic hemiplegia due to stroke (more than three months’ duration), able to walk at least 10 m independently without a walking aid, and aged over 50 years were recruited. (1 day assessment) Subjects were instructed to walk along a 5-m walkway in 3 different conditions: without an aid, with a walking stick (one point cane), and with a tripod. (three point cane) Subjects’ trunk movements, symmetry and velocity No significant differences in any of the parameters (trunk movements, symmetry or velocity) between the one-point cane, the three-point cane or no aid were found >3 month

Allet 2009 25 Crossover RCT Quad cane, one-point cane with ergonomic handgrip, and Nordic stick Ulnar process Inclusion creteria:Patients with a first stroke confirmed by computed tomography or magnetic resonance imaging scan at an early stage of gait rehabilitation (as soon as thepatient was able to walk 10 m holding a handrail without therapeutic support) Exclusion creteria: if they were capable of walking more than 5 m independently without a walking aid, if they could not understand instructions, if they were not able to hold a walking stick with their nonparetic hand, or if they had orthopedic problems or another pre-existing neurologic condition resulting in an asymmetric gait pattern On 3 consecutive days subjects performed 6 minutes walk test (6MWT) used, in random order, 1 of 3 walking aids:quad cane, one-point cane with ergonomic handgrip, and Nordic stick. Temporo-spatial gait parameters (velocity, cadence, step length, step time), subjective benefit with a scale ranging from 0 to 10.2 and gait symmetry Walking Distance: Greatest with one-point cane, followed by quad cane and Nordic walking stick. Walking Velocity: Highest with one-point cane. Patient Preference: one-point cane was preferred. Step Length Symmetry: No significant difference NA

Maguire 2009 13 Crossover RCT Not described Radial styloid Inclusion criteria: (1) All patients scored at least level 3 on the Functional Ambulation Category, (2) All patients were independent walkers prior to insult, (3) had a Mini Mental State score of 22 or above, (4) had no orthopaedic or other neurological conditions that could limit walking ability, (5) Patients had no gross visuospatial or visual field deficits and had no medical contraindications to walking. (1 day assessment) Data collection over six gait cycles as subjects walked at self-selected speed during: (1) hip abductor taping, (2) TheraTogs, (3) cane in non-hemiplegic hand. Muscle activation of gluteus medius and tensor fascia lata. Self-selected walking speed in m/s, spatial asymmetry – calculated as step length asymmetry ratio, and temporal asymmetry – calculated as mean single limb support time ratio. Gluteus medius peak EMG: Significant difference between cane vs. TheraTogs and tape (P < 0.001). Tensor fascia lata peak EMG: Significant for cane vs. TheraTogs (P < 0.009), non-significant for tape. Gait speed: No significant difference between cane, TheraTogs, and tape. Step length symmetry: Greatest with TheraTogs, followed by tape, then cane. Step length: Cane and TheraTogs increased hemiplegic leg length; taping kept non-hemiplegic leg longer but reduced the difference. Temporal asymmetry: Reduced only with cane; increased with TheraTogs and taping. Single support phase: Longer on non-hemiplegic leg at baseline, increased with TheraTogs and tape; cane increased support on hemiplegic leg. Not fully specified sub acute (n = 12) and chronic (n = 1)

Jung 2014 18 Parallel RCT pilot Weight Support Feedback Cane (WSFC) Greater trochanter Inclusion criteria: (1) first episode of unilateral stroke with hemiparalysis caused by hemicerebrum damage; (2) a Mini-Mental State Examination score of 24 or higher; (3) able to walk with a cane (functional ambulation classification, 2 3); (4) loaded more than 7% body weight of peak vertical force on the cane. Exclusion criteria included: (1) cerebral aneurysm; (2) hemianopia, dizziness, or other symptoms indicating vestibular impairment; (3) impaired touch and pressure sensation on the nonaffected hand; (4) hemineglect; (5) orthopedic disease influencing gait. 4 week training period. Subjects in the experimental group underwent gait training by using a cane that provided auditory feedback. The training lasted for 30 minutes a day, five times a week for four weeks. Subjects in the control group received gait training without auditory feedback for the same amount of time. Peak vertical force of cane, muscle activation (EMG) of gluteus medius, and vastus medialis oblique, single limb support phase Significant improvement in the peak force of cane, muscle activation of gluteus medius and vastus medialis oblique, single limb support phase of the affected leg, and gait speed were observed in the experimental group (13.5±9.5 cm/s) compared with the control group (3.7±8.4 cm/s) (p < 0.05). NA

Kang 2021 30 Parallel RCT Weight Support Feedback Cane (WSFC) Not described Inclusion criteria: (1) residual hemiparesis at least 6 months after the onset of stroke, (2) those with adequate cognitive levels to perform this study (the score of the Korean version of the Mini-Mental State examination >24), (3) those who walked using a cane (functional ambulation category 2-3 levels), and (4) those who supported 7% or more of their body weight on a cane. We excluded subjects with (1) a potential musculoskeletal condition that could interfere with safe walking, (2) hemispatial neglect, and (3) severe heart disease or uncontrolled hypertension. Three subjects were excluded because they did not meet the selection criteria (uncontrolled hypertension and MMSE 20 points). The Weight support feedback cane (WSFC) group participated in weight support feedback cane gait training conducted for 30 min per day, 3 times per week, for 4 weeks, while the subjects from the conventional cane (CC) group participated in conventional cane gait training conducted for 30 min per day, 3 times per week, for 4 weeks. Muscle activation of rectus femoris, biceps femoris, medial gastrocnemius, and gluteus medius. Gait ability a (velocity, cadence, single-limb support phase of the affected side, and symmetry index) Between-group findings: WSFC group had significantly greater improvement than CC group in rectus femoris, medial gastrocnemius, gluteus medius, affected single-limb support phase, and symmetry index (P < 0.05). Within-group findings: Both groups showed significant improvements in rectus femoris, biceps femoris, medial gastrocnemius, velocity, cadence, affected single-limb support, and symmetry index (P < 0.05). Significant improvement in gluteus medius for WSFC only (P < 0.05). Chronic

Avelino 2021 50 Parallel RCT One-point cane with ergonomic handgrip Ulnar process Inclusion criteria: Participants were eligible if they were >20 years of age (2) >6 months and <5 years after their last episode of stroke; (3) had hemiparesis, that is, weakness of the knee flexor/extensor and/or hip flexor muscles, as determined by 15% strength differences between the paretic and non-paretic limbs, (4) were able to walk at least 14 m, independently, at speeds 0.8 m/s (5) and were naïve to the use of walking devices. Exclusion criteria: (1) cognitive deficits, as determined by their education-adjusted scores on the Mini-Mental State Examination (as follows: 18/19 for the individuals with illiteracy and 24/25 for those with basic education) (2) and/or any other conditions, which prevented measurement or training. For the control group, they are askes to perform stretching exercises of the lower-limb muscles for one month. For experimental group, physiotherapist asked them to use the cane whenever they needed for one month. walking speed (10MWT), step length (10MWT), cadence (10MWT), walking capacity (6MWT), walking confidence (Modified Gait-Efficacy Scale), and participation (SIS participation domain) When walking without the cane, after one month and at follow-up (two months), there were no between-group differences in any measures of walking or participation. Chronic

Jung 2019 20 Parallel RCT pilot Weight Support Feedback Cane (WSFC) Greater trochanter Inclusion criteria: (1) diagnosis of first unilateral hemispheric stroke; (2) ability to walk with a cane (functional ambulation classification, 2–3); (3) ability to understand and follow simple verbal instructions; (4) no impaired touch and pressure sensation on the un-affected hand; (5) no hemineglect; and (6) no orthopedic disease influencing gait. Experimental group participated in gait training with auditory feedback for 30 minutes, 5 times a week, for 4 weeks. Control group received conventional gait training for 30 minutes, 5 times a week, for 4 weeks. Subjects in the experimental group underwent gait training with auditory feedback. Subjects in the control group received conventional gait training without auditory feedback. All participants underwent a conventional exercise program provided by the rehabilitation hospital for 30 minutes, 5 times a week, for 4 weeks. Muscle activation of erector spinae muscle, Trunk Impairment Scale (TIS). Dynamic balance was measured using the Timed Up and Go (TUG) test Muscle activation was significantly higher in the experimental group than in the control group (p = 0.028). No significant difference was found in the TIS score between the experimental and control groups (p > 0.05). A significant improvement was observed for the TUG test in the experimental group compared to the control group (p = 0.046). NA

Reference	Samle size	Study design	Types of cane	Cane height level (with the subjects standing and the elbow extended)	Inclusion/Exclusion criteria	Intervention	Outcomes	Findings	Phase
Li et al. 2018	30	Crossover RCT	One-point cane, quad cane	Patient dependent	Inclusion criteria: (1) sustained unilateral hemiplegia caused by a first-time cerebral hemispheric stroke >3 months before the study; (2) could perform the STS movement with no assisting device and could ambulate with a cane for at least 10 m; (3) were able to stand with no external support for at least 15 s; (4) were able to understand and follow instructions; (5) had no history of major orthopedic operations within the previous 5 years; and (6) had no other orthopedic or neurological conditions that would affect the STS movement	(1 day assessment) Sit-to-Stand (STS) test in seven situations in a random sequence: without a cane, three positions with a regular cane (parallel to the ankle, parallel to the 5th toe, and 10 cm in front of the 5th toe)	Peak vertical ground reaction force (GRF) and maximum vertical cane support force during STS were recorded.	Mean peak GRF: Significant differences between STS conditions with and without a cane (p = 0.000–0.040). Cane Support Force: Maximum support force varied by cane position, highest beside the ankle (8.62±4.01). Quad vs. Regular Cane: Maximum support force was significantly higher for all positions with a quad cane (p = 0.019 for 10 cm, p = 0.003 for toe, p = 0.000 for ankle).	>3 months
Boonsinsukh 2009	39	Crossover RCT	Adjustable-height cane, with force sensor	Patient dependent	Exclusion criteria: (1) had used a cane during walking before participating in this study, (2) had cognitive or language impairment, (3) had severe hemineglect, (4) had cerebral aneurysm, (5) had bilateral cerebral impairment, (6) had brainstem and cerebellar lesions, or (7) had impaired touch and pressure sensation on the nonparetic hand, (8) All subjects who had hemianopia, dizziness, or other symptoms indicating vestibular impairment; had taken medications that affect balance; had impaired sensation in the lower extremity; or had lower extremity deficits such as pain or contracture were also excluded from the study	(1 day assessment) In the (light) touch contact condition, the subject’s goal was to use the cane while walking without the auditory buzzer being activated. A buzzer was activated whenever the subjects pressed on the cane with more than 400 g (4 N) of vertical force. In the force contact condition, the subject’s goal was to walk and push on the cane at the desired level.	ML pelvic stability as measured by averaged peak-to-peak pelvic acceleration, muscle activation of bilateral tensor fascia latae (TFL), semitendinosus (ST), and vastus medialis (VM) using an electromyography system, and vertical cane force	Cane Use: Force exerted on the cane was significantly higher in the force contact condition vs. touch contact (P < 0.001). Subjects with stroke applied force earlier and for a longer duration in the force contact condition (P < 0.001). Gait Speed &Pelvic Acceleration: No significant effect on walking speed or pelvic acceleration between force and touch contact conditions (P < 0.001). Temporal Gait Parameters: Single-limb support phase of the paretic leg was longer in the touch contact condition (P < 0.001), corresponding to longer muscle activation of the paretic vastus medialis. Muscle Activation: Tensor fascia lata, vastus medialis, and semitendinosus on the paretic side were activated more in the touch contact condition than the force contact condition (P < 0.001).	Subacute
Tyson 2009	20	Crossover RCT	One-point cane	Greater trochanter	Inclusion criteria: (1) Unable to walk for at least 2 weeks after their stroke, (2) Undergoing inpatient rehabilitation to restore walking, (3) Unable to mobilize independently without aids in every day life on the ward, (4) Able to step and practice walking during physiotherapy sessions although they may need support, (5) Able to walk 5 m without physical support (the minimum walking distance for the test procedure), (6) Able to give informed consent	(1 day assessment) (1) Walking with no device (the control condition), (2) walking with a walking cane, measured using a five-meter walk test	Functional mobility (functional ambulation categories FAC), walking impairments (speed 5MWT, step length of the weak leg 5MWT), and patients’ opinions self-report questionnaire.	Walking with cane showed a significant improvement in functional mobility compared with walking with no device (P = .0001) with large effect size.	Subacute
Polese 2011	19	Crossover RCT	One-point Cane	Not described	Inclusion criteria: (1) Were ≥20 years old; (2) had a mean time of over 6 months since the onset of a unilateral stroke, (3) were regular users of walking sticks, but were able to walk without them for 10 m; (4) had no cognitive deficits, as determined by the Mini-Mental State Examination scores N24; (5) had no other neurological or orthopedic disorders.	(1 day assessment) Four walking conditions: (i) with walking sticks at a comfortable speed; (ii) with walking sticks at a fast speed; (iii) without walking sticks at a comfortable speed; and (iv) without walking sticks at a fast speed.	Linear kinematics (walking speeds) and angular kinematics (maximum joint angles), power, and work of the paretic hip, knee and ankle joints in the saggital plane	Walking Speed: Increased during fast (P < 0.001) and comfortable (P = 0.001) walking with walking sticks. Joint Angles: No significant changes (P = 0.24–0.83). Power Generation: Increased in ankle plantar flexion, knee extension, and hip flexion (P < 0.01). Comparison of Walking Speeds: No greater changes between fast and comfortable walking for any outcome.	Chronic
Beauchamp 2009	14	Crossover RCT	One-point cane, quad cane	Not describe	Inclusion criteria: (1) first clinical diagnosis of stroke; (2) less than 3 months post-stroke; (3) ability to understandand follow oral commands; (4) ability to give informedconsent; (5) ambulatory without a gait aid before current stroke; (6) ambulatory without a gait aid for 6 m with close, non-contact supervision; and (7) “naïve” to use of an aid (i.e., patient had not integrated a cane into daily living)	(1 day assement) Patinets were instructed to walk a distance of 6 m. Task conditions were as follows: (1) no cane task; (2) standard single-point cane task; (3) quad cane task.	Gait-symmetry and walking velocity	Symmetry Improvement: Standard cane improved symmetry in asymmetric patients (p = 0.028); quad cane did not (p = 0.36). No effect on symmetry in symmetric patients with either cane (standard cane p = 0.88, quad cane p = 0.32). Mean Velocity: No change with standard cane in asymmetric patients (p = 0.18); decreased with quad cane (p = 0.035). No change in velocity for symmetric patients with either one-point cane (standard cane p = 0.08, quad cane p = 0.17).	Subacute
Boonsinsukh, 2011	62	Crossover RCT	Force-sensing instrumented cane	Radial styloid	Iinclusion criteria: who had a stroke less than 1 year prior to the study. Exclusion criteria:Subjects were excluded from the study if they had (1) regularly used a cane during walkingbefore participating in this study; (2) cognitive or language impairments (Thai Mini-Mental State Examination score <21); (3) a cerebral aneurysm; (4) bilateral cerebral impairment; (5) brainstem and cerebellar lesions; (6) impaired touch and pressure sensation on the nonparetic hand, as measured by Semmes Weinstein Monofi lament (SWM) at the thenar eminence (fi lament <2.83 MN); or (7) severe hemineglect. Hemineglect was assessed by clinical observation; if there was any uncertainty, standard neuropsychological assessments, such as Bells Test, 18 were conducted.	(1 day assessment) Subjects were instructed to walk along a 5-m walkway while using the cane in the 3 conditions: (1) no contact (NC), in which the cane was not allowed to touch the ground; (2) touch contact (TC) required subjects to exert less than 4 N of force through a cane that was held in the nonparetic hand. (3) force contact (FC), where the subject was instructed to exert as much force on the cane as needed.	Trunk stability, percentages of gait cycle gait speed, muscle activation of bilateral tensor fascia latae (TFL) and vastus medialis (VM) using an electromyography system, and vertical cane force	Trunk Acceleration: Greater in NC condition (P < 0.001); no difference between FC and TC conditions. Gait Speed and Percentage of Gait Cycle: No difference in gait speed or stance-swing durations among FC, TC, and NC conditions. Muscle Activation: Reduced in VM (P < 0.001) and TFL (P < 0.001) on the paretic leg during FC versus TC and NC conditions. No significant difference between TC and NC for paretic VM and TFL activation.	<1 year
Buurke 2005	13	Crossover RCT	One-point cane, quad cane	Radial styloid	Inclusion criteria: (1) Aged between 40 and 75 years (2) Had a first unilateral ischemic stroke and were able to walk without physical assistance: Functional Ambulation Categories (FAC≥3). (3) Sufficient cognitive abilities:Mini-mental State Examination (MMSE)≥22	(1 day assemement) Patients were tested in three different conditions: (1) walking without a walking aid, (2) walking with a cane and (3) walking with a quad stick Five patients used an ankle foot orthosis (AFO)	Muscle activation of erector spinae, gluteus maximus, gluteus medius, vastus lateralis, semitendinosus, gastrocnemius and tibialis anterior muscles on the affected side, walking speed.	Walking speed: Significantly lower with quad stick vs. cane Erector spinae burst duration: Decreased with cane vs. no aid. Tibialis anterior on-times: Later with quad stick vs. cane. Tibialis anterior burst duration: Decreased with both cane and quad stick vs. no aid. Vastus lateralis on-times: Later with quad stick vs. cane and no aid. Erector spinae average amplitude: Decreased with quad stick vs. no aid and cane. Gluteus maximus, medius, and vastus lateralis amplitude: Decreased with both cane and quad stick vs. no aid. Tibialis anterior amplitude: Decreased with quad stick vs. no aid and cane.	They had a mean time since an ischaemic stroke of 205 days (range 47–385 days).
Deltombe 2020	32	Crossover RCT	Quad cane, Wheeleo^® rolling cane	The height can be adapted from 74 to 98 cm.	Inclusion criteria were: age over 18 years; a history of hemiparesis as a consequence of a stroke or traumatic brain injury; use of a quadripod cane for walking; the ability to walk independently without additional help or support; and the capability to use the rolling cane for a few steps.	(2 days assessment) Participants were assessed using a quad-ripod cane and a rolling cane.	Walking speed during a 10-m walk test and a 6-min walk test; frequency of 2-step gait; physiological cost index; number of therapist interventions to control the balance; perceived exertion; and participant satisfaction	Walking Speed: Increased with rolling cane during 10-m walk test (comfortable + 22%, p < 0.001; maximal+30%, p < 0.001) and 6-min walk test (+50%, p < 0.001). Distance: Increased with rolling cane by +49% in 6-min walk test (p < 0.001). 2-Step Gait Frequency: Improved with rolling cane. Physiological Cost Index, Perceived Exertion, Therapist Interventions: No change. Participant Satisfaction: Improved with rolling cane.	NA
Jeong 2015	30	Crossover RCT	One-point cane, quad cane, and a hemi-walker	Ulnar process	Inclusion criteria (1) first cerebrovascular accident verified by computed tomography or magnetic resonance imaging, (2) hemiplegic patient with gait disturbance, (3) independent ambulation with a cane over 10 meters, with or without orthosis, or a score of 4 or 5 on the Functional Ambulation Category (FAC) scale10 (no need for assistance), and (4) sufficient cognition to participate in the training, confirmed by a mini-mental state examination 11 (MMSE) score of 25 or higher. The exclu-sion criteria were: (1) neurologic co-morbidity other than stroke, (2) cardiovascular, musculoskeletal, or other medical condition that could influence gait or oxygen consumption, (3) refusal to participate in the study, or (4) recurrent stroke.	A single-point cane, a quad cane, and a hemi-walker. At approximately the same time of day for three consecutive days, each subject completed a walk with a different type of walking aid.	Energy expenditure (O₂ rate, mL/kg/min), energy cost (O₂ cost, mL/kg/m), and heart rate (HR) via a portable gas analyzer, a 10-meter walk test (10MWT), and a 6-minute walk test (6MWT).	Energy Expenditure, Gait Endurance, Gait Velocity: Higher with single-point cane (p < 0.001 or p = 0.005) compared to other canes. Energy Cost: Lower with single-point cane (0.5±0.2 mL/kg/m) vs. others (0.6±0.2 mL/kg/m, p = 0.001). Heart Rate: No significant difference after Bonferroni correction (p≥0.05).	Chronic
Tyson 1999	20	Crossover RCT	One-point cane and three-point cane	Radial styloid	Inclusion criteria: Subjects with a chronic hemiplegia due to stroke (more than three months’ duration), able to walk at least 10 m independently without a walking aid, and aged over 50 years were recruited.	(1 day assessment) Subjects were instructed to walk along a 5-m walkway in 3 different conditions: without an aid, with a walking stick (one point cane), and with a tripod. (three point cane)	Subjects’ trunk movements, symmetry and velocity	No significant differences in any of the parameters (trunk movements, symmetry or velocity) between the one-point cane, the three-point cane or no aid were found	>3 month
Allet 2009	25	Crossover RCT	Quad cane, one-point cane with ergonomic handgrip, and Nordic stick	Ulnar process	Inclusion creteria:Patients with a first stroke confirmed by computed tomography or magnetic resonance imaging scan at an early stage of gait rehabilitation (as soon as thepatient was able to walk 10 m holding a handrail without therapeutic support) Exclusion creteria: if they were capable of walking more than 5 m independently without a walking aid, if they could not understand instructions, if they were not able to hold a walking stick with their nonparetic hand, or if they had orthopedic problems or another pre-existing neurologic condition resulting in an asymmetric gait pattern	On 3 consecutive days subjects performed 6 minutes walk test (6MWT) used, in random order, 1 of 3 walking aids:quad cane, one-point cane with ergonomic handgrip, and Nordic stick.	Temporo-spatial gait parameters (velocity, cadence, step length, step time), subjective benefit with a scale ranging from 0 to 10.2 and gait symmetry	Walking Distance: Greatest with one-point cane, followed by quad cane and Nordic walking stick. Walking Velocity: Highest with one-point cane. Patient Preference: one-point cane was preferred. Step Length Symmetry: No significant difference	NA
Maguire 2009	13	Crossover RCT	Not described	Radial styloid	Inclusion criteria: (1) All patients scored at least level 3 on the Functional Ambulation Category, (2) All patients were independent walkers prior to insult, (3) had a Mini Mental State score of 22 or above, (4) had no orthopaedic or other neurological conditions that could limit walking ability, (5) Patients had no gross visuospatial or visual field deficits and had no medical contraindications to walking.	(1 day assessment) Data collection over six gait cycles as subjects walked at self-selected speed during: (1) hip abductor taping, (2) TheraTogs, (3) cane in non-hemiplegic hand.	Muscle activation of gluteus medius and tensor fascia lata. Self-selected walking speed in m/s, spatial asymmetry – calculated as step length asymmetry ratio, and temporal asymmetry – calculated as mean single limb support time ratio.	Gluteus medius peak EMG: Significant difference between cane vs. TheraTogs and tape (P < 0.001). Tensor fascia lata peak EMG: Significant for cane vs. TheraTogs (P < 0.009), non-significant for tape. Gait speed: No significant difference between cane, TheraTogs, and tape. Step length symmetry: Greatest with TheraTogs, followed by tape, then cane. Step length: Cane and TheraTogs increased hemiplegic leg length; taping kept non-hemiplegic leg longer but reduced the difference. Temporal asymmetry: Reduced only with cane; increased with TheraTogs and taping. Single support phase: Longer on non-hemiplegic leg at baseline, increased with TheraTogs and tape; cane increased support on hemiplegic leg.	Not fully specified sub acute (n = 12) and chronic (n = 1)
Jung 2014	18	Parallel RCT pilot	Weight Support Feedback Cane (WSFC)	Greater trochanter	Inclusion criteria: (1) first episode of unilateral stroke with hemiparalysis caused by hemicerebrum damage; (2) a Mini-Mental State Examination score of 24 or higher; (3) able to walk with a cane (functional ambulation classification, 2 3); (4) loaded more than 7% body weight of peak vertical force on the cane. Exclusion criteria included: (1) cerebral aneurysm; (2) hemianopia, dizziness, or other symptoms indicating vestibular impairment; (3) impaired touch and pressure sensation on the nonaffected hand; (4) hemineglect; (5) orthopedic disease influencing gait.	4 week training period. Subjects in the experimental group underwent gait training by using a cane that provided auditory feedback. The training lasted for 30 minutes a day, five times a week for four weeks. Subjects in the control group received gait training without auditory feedback for the same amount of time.	Peak vertical force of cane, muscle activation (EMG) of gluteus medius, and vastus medialis oblique, single limb support phase	Significant improvement in the peak force of cane, muscle activation of gluteus medius and vastus medialis oblique, single limb support phase of the affected leg, and gait speed were observed in the experimental group (13.5±9.5 cm/s) compared with the control group (3.7±8.4 cm/s) (p < 0.05).	NA
Kang 2021	30	Parallel RCT	Weight Support Feedback Cane (WSFC)	Not described	Inclusion criteria: (1) residual hemiparesis at least 6 months after the onset of stroke, (2) those with adequate cognitive levels to perform this study (the score of the Korean version of the Mini-Mental State examination >24), (3) those who walked using a cane (functional ambulation category 2-3 levels), and (4) those who supported 7% or more of their body weight on a cane. We excluded subjects with (1) a potential musculoskeletal condition that could interfere with safe walking, (2) hemispatial neglect, and (3) severe heart disease or uncontrolled hypertension. Three subjects were excluded because they did not meet the selection criteria (uncontrolled hypertension and MMSE 20 points).	The Weight support feedback cane (WSFC) group participated in weight support feedback cane gait training conducted for 30 min per day, 3 times per week, for 4 weeks, while the subjects from the conventional cane (CC) group participated in conventional cane gait training conducted for 30 min per day, 3 times per week, for 4 weeks.	Muscle activation of rectus femoris, biceps femoris, medial gastrocnemius, and gluteus medius. Gait ability a (velocity, cadence, single-limb support phase of the affected side, and symmetry index)	Between-group findings: WSFC group had significantly greater improvement than CC group in rectus femoris, medial gastrocnemius, gluteus medius, affected single-limb support phase, and symmetry index (P < 0.05). Within-group findings: Both groups showed significant improvements in rectus femoris, biceps femoris, medial gastrocnemius, velocity, cadence, affected single-limb support, and symmetry index (P < 0.05). Significant improvement in gluteus medius for WSFC only (P < 0.05).	Chronic
Avelino 2021	50	Parallel RCT	One-point cane with ergonomic handgrip	Ulnar process	Inclusion criteria: Participants were eligible if they were >20 years of age (2) >6 months and <5 years after their last episode of stroke; (3) had hemiparesis, that is, weakness of the knee flexor/extensor and/or hip flexor muscles, as determined by 15% strength differences between the paretic and non-paretic limbs, (4) were able to walk at least 14 m, independently, at speeds 0.8 m/s (5) and were naïve to the use of walking devices. Exclusion criteria: (1) cognitive deficits, as determined by their education-adjusted scores on the Mini-Mental State Examination (as follows: 18/19 for the individuals with illiteracy and 24/25 for those with basic education) (2) and/or any other conditions, which prevented measurement or training.	For the control group, they are askes to perform stretching exercises of the lower-limb muscles for one month. For experimental group, physiotherapist asked them to use the cane whenever they needed for one month.	walking speed (10MWT), step length (10MWT), cadence (10MWT), walking capacity (6MWT), walking confidence (Modified Gait-Efficacy Scale), and participation (SIS participation domain)	When walking without the cane, after one month and at follow-up (two months), there were no between-group differences in any measures of walking or participation.	Chronic
Jung 2019	20	Parallel RCT pilot	Weight Support Feedback Cane (WSFC)	Greater trochanter	Inclusion criteria: (1) diagnosis of first unilateral hemispheric stroke; (2) ability to walk with a cane (functional ambulation classification, 2–3); (3) ability to understand and follow simple verbal instructions; (4) no impaired touch and pressure sensation on the un-affected hand; (5) no hemineglect; and (6) no orthopedic disease influencing gait.	Experimental group participated in gait training with auditory feedback for 30 minutes, 5 times a week, for 4 weeks. Control group received conventional gait training for 30 minutes, 5 times a week, for 4 weeks. Subjects in the experimental group underwent gait training with auditory feedback. Subjects in the control group received conventional gait training without auditory feedback. All participants underwent a conventional exercise program provided by the rehabilitation hospital for 30 minutes, 5 times a week, for 4 weeks.	Muscle activation of erector spinae muscle, Trunk Impairment Scale (TIS). Dynamic balance was measured using the Timed Up and Go (TUG) test	Muscle activation was significantly higher in the experimental group than in the control group (p = 0.028). No significant difference was found in the TIS score between the experimental and control groups (p > 0.05). A significant improvement was observed for the TUG test in the experimental group compared to the control group (p = 0.046).	NA

3.1 Cane characteristics

Three studies used one-point canes [20 , 32], while one study [27] compared the differences between three-point canes and one-point canes, and five reported the differences between one-point canes and quad canes [18 , 28], weight supported feedback canes (WSFC) designed with a pressure sensor that detects the peak vertical force on the cane during gait for enhancement of weight-bearing on the affected lower limb were used in five studies [19 , 33], and one used a Wheeleo^® rolling cane [25] which has four small wheels allowing the patient to roll the cane instead of lifting it. Four different cane heights were used in the studies, four studies adjusted the height of cane at the level of each subject’s radial styloid process of the nonparetic hand with the arm hanging straight down [23 , 29]; three studies adjusted to the height of the ulnar process of each participant when in a standing position with the elbow extended [26 , 32], three studies adjusted at the level of the participants’ greater trochanter [20 , 33], three experiments in which the height of the cane was dependent on patient [18 , 25]. Three studies failed to provide the height of cane used [21,22 , 31].

3.2 Study design and intervention

Among the sixteen selected RCT studies, twelve [18 –29] were crossover and four were parallel [30 –33]. Nine crossover studies [18–24 , 29] implemented one-day interventions to assess walking ability with or without a cane and under different types of cane conditions. One crossover study [25] implemented a two-day intervention, and two [26, 28] crossover studies used three consecutive days assessment. All four included parallel RCTs [30 –33] implemented a 4-week intervention. The experimental groups in three parallel studies used weight supported feedback canes (WSFC), while the control group used ordinary canes without feedback. All three studies used cane training sessions lasting 30 minutes per day, 3–5 times per week, for 4 weeks for both experimental and control groups with no follow-up assessment [30–31 , 33]. One study was an attention controlled randomised trial with 2-month follow-up, the experimental intervention was to provide a single-point cane at any time when walking was required for a period of one month, while the control group engaged in stretching exercises of the affected lower-limb muscles daily for one month at home [32].

3.3 Participant characteristics

The mean age of subjects recruited in all studies was greater than 55 years. Nine studies [19–21 , 33] included participants who were regular users of canes, six [22–24 , 32] included non-regular users, and one study [18] did not report whether participants were non-regular or regular users. According to Julie et al. [34], time points for stroke recovery are defined as: Hyper acute (0–24 hours), Acute (1–7 days), Early sub-acute (7 days–3 month), Late sub-acute (3–6 months), Chronic (>6 months). One study recruited subjects with acute to sub-acute stroke [19], four [20 , 29] with sub-acute stroke, seven [18 , 33] with sub-acute to chronic, three [21 , 32] with chronic stroke, and one study [23] with all the stages of stroke (from acute to chronic).

3.4 Outcome measures

Twelve studies measured gait related outcomes that included spatiotemporal gait parameters (speed, step length, cadence, symmetry, etc.) and kinematic as well as kinetic gait parameters [20–28 , 30–32]. Balance assessment included trunk and pelvic stability as well as the number of times the therapist assisted the subjects to maintain balance while walking [19 , 27]. Trunk control was evaluated using Trunk Impairment Scale (TIS) [33]. In addition, the peak vertical cane force [18 , 23], muscle activation patterns (amplitudes and timing) were measured using surface electromyography (sEMG) [19 , 33], energy expenditure and energy cost were reported [26]. Only four of sixteen studies included psychological tests in their outcome measures. Assessment of walking confidence through perceived exertion utilising the modified Gait-Efficacy Scale, as well as participant satisfaction and opinions regarding the ease of use and safety of the cane. [20 , 32].

3.5 Effectiveness of canes

(a) Walking speed

Six studies [20–22 , 29] examined walking speed, when walking with either a single-point cane, three-points cane, or quad cane versus walking without a cane. One study [21] showed the use of a one-point cane resulted in significant increases in gait speed during both fast and comfortable walking compared to walking without a cane. Four studies [20 , 29] reported no significant difference in walking speed between walking with and without a cane. One study [22] showed that mean walking speed significantly decreased during the quad cane task. Four studies conducted comparative analyses of walking speed utilising various cane types. Three studies [24 , 28] showed a significantly lower walking speed when walking with a quad cane compared to walking with a one-point cane. One study [27] reported no significant differences in walking speed between a one-point cane and a three-point cane. One study [25] reported the significant improvement in walking speed and distance during a 6-min walk test with the use of Wheeleo rolling cane compared to a quad cane. Three studies [30 , 33] showed significant improvement in walking speed with the use of WSFC compared to ordinary canes. Two studies [19, 23] that examined two methods of cane use, light touch contact which required subjects to exert less than 4 N of force through a cane and force contact, where the subject was instructed to exert as much force on the cane as needed, reported that walking speed did not differ between the force contact and touch contact condition.

(b) Symmetry

Three studies [22 , 29] examined symmetry when walking with canes versus walking without canes. One study reported [22] that compared to walking without a cane, the use of a one-point cane significantly improved symmetry in stroke survivors with asymmetry, while the quad cane did not improve symmetry. One study showed [27] that no significant differences in symmetry were found between the conditions. One study [29] demonstrated that both spatial asymmetry (calculated as step length asymmetry ratio), and temporal asymmetry (calculated as mean single limb support time ratio) reduced during one-point cane use. Two studies [27, 28] reported that there was no significant difference in symmetry between a one-point cane and a quad cane. One study [31] showed a significant improvement in symmetry index with the use of WSFC.

Two studies [24, 29] examined muscle activation, when walking with canes versus walking without canes. Both studies showed that cane use significantly reduced EMG activity in gluteus medius, tensor fascia lata [29], erector spinae, tibialis anterior, vastus lateralis [24]. One study [24] reported that the EMG activities (amplitudes and burst duration) were significantly lower in erector spinae, tibialis anterior, vastus lateralis when compared to walking with a quad cane and walking with a one-point cane. Two studies [19, 23] reported that muscles (bilateral tensor fascia latae, vastus medialis) on the paretic side during stance phase were activated more in the light touch contact condition than in the force contact condition. Three studies [30, 31] reported that muscle activation was significantly higher in the WSFC group compared to the ordinary cane group in the affected lamb.

(d) Single limb support phase

One study [19] reported that the duration of the single-limb support phase of the paretic leg in the light touch contact cane condition was significantly longer than in the force contact condition. While one [23] showed that stance-swing durations did not differ among the two conditions. Two studies [30, 31] demonstrated the significantly greater improvement in the affected single-limb support with the use of WSFC compared to ordinary canes.

(e) Trunk and pelvic movements

Two studies [23, 27] examined trunk movements, when walking with canes versus walking without canes. One study [27] found that there was no significant difference in trunk movements when using a cane or no aids. While another study [23] reported trunk stability was significantly greater when walking with a one point-cane. One study [30] reported no significant difference in trunk stability between the WSFC group and an ordinary cane group. A significant improvement was also observed for the dynamic balance in the WSFC group compared to the ordinary cane group. One study [19] demonstrated that the comparison of the average peak-to-peak pelvic acceleration showed no difference between the force contact and light touch contactcondition.

(f) Additional outcomes

One study [20] examined step length when walking with a single-point cane versus walking without a cane. Additionally, one study [26] showed that energy cost was significantly lower for the single-point cane compared to a quad cane. One study [30] reported the significant improvement in the peak vertical force on the cane during gait with the use of WSFC.

4 Discussion

This scoping review has demonstrated that the RCT research literature on cane use in post-stroke rehabilitation is heterogeneous in nature regarding varied research and cane design, varied subject characteristics, and mixed results.

This review has also identified specific knowledge gaps in the current literature. The impact and practical implications of these review findings for clinical care or future research are discussed in detail below. Among the sixteen included studies, only three studies [30 , 33] examined evolutions of traditional cane design, utilising biofeedback canes and one study reported the effects of a rolling cane [25]. The biofeedback studies evaluated the effect of weight support feedback cane gait training. The results demonstrated greater improvement in the experimental group in lower limb muscle activity and gait ability compared to the traditional cane group. This potentially promising approach to novel cane design utilising technological application would benefit from further RCT research. Individuals with poor balance are frequently required to use a quad cane, which has a larger base of support and reduces sway and increases stability during the stance phase [40]. However, the cane must be lifted and advanced. This pattern causes a 3-step gait, which slows walking speed [13]. Deltombe et al added wheels to a quad cane, thus the participants rolled the cane instead of lifting it and this study found that the rolling cane improved walking speed and distance without additional risk of falls and patient satisfaction was increased compared to a quad cane [25]. Taking this evidence into account, the development of new types of canes that can prevent patients from developing abnormal gait and potentially improve neuromuscular control or recovery of balance and gait has the potential to be explored further due to the small number of RCT studies currently published in the area. Additionally, two studies [19, 23] suggested that light touch cue can be provided during walking with a cane, this augmented somatosensory information provides lateral stability during walking for people who have had a stroke by facilitating the activations of weight-bearing muscles on the paretic leg during the stance phase. Regarding overall study design, only four of the sixteen RCTs presented in this review were parallel studies, with each intervention lasting four weeks [30 –33], and only one study implemented a follow-up assessment process [32]. The remaining twelve crossover trials had intervention periods of one to three days. In short-term crossover trials, reduced intervention dosage and potential carry-over effects can impact treatment responses [39]. Recent clinical guidelines [45] emphasise 150–300 minutes of moderate-intensity aerobic activity per week for adults with chronic conditions like stroke. To enhance clinical understanding, future research should conduct more RCTs exploring higher therapy dosages and long-term outcomes with extended follow-up assessments.

Early walking is crucial in stroke rehabilitation, addressing psychological well-being, cardiovascular fitness, and preventing secondary muscle weakness and abnormal muscle tone [29]. However, limited research [46] focuses on the acute to sub-acute phase, a critical period for post-stroke clinical care. Study criteria are diverse, encompassing individuals who have experienced a stroke, from acute to chronic phases [18 , 30–33]. Limited RCT research on acute or subacute stroke phases underscores the need for further investigation. Current findings on cane impact on stroke patients’ mobility are mixed, with many studies reporting neutral (no change) or positive conclusions. Up-to-date statistical analysis in a systematic review with meta-analysis is recommended for a clearer understanding.

The Biopsychosocial model, the foundation of the International Classification of Function (ICF) [35], underscores the importance of biological, psychological, and social aspects in defining health and disability [44]. However, studies on walking aids primarily focus on kinetic, kinematic, and neuromuscular effects, neglecting satisfaction and quality of life impact. Given the psychological importance in device acceptance [43], future research should explore the psychological effects of cane types on stroke survivors for improved clinical care.

In conclusion, this scoping review has mapped cane use for people who have had a stroke to assess study characteristics, design and device variations and to identify potential knowledge gaps. The impact and practical implications of this review for clinical care and future research include the identification that parallel trials, long-term crossover trials and follow-up assessments are lacking. Furthermore, more RCT research is required to examine variations in novel cane design that specifically target motor function in stroke patients at specific stages due to the potential of such devices to improve mobility and motor function. As well as this, the current scoping review may act as an important precursor to more specific future systematic reviews by enhancing the clarification of research in the area. Future research should aim to address the identified inconsistencies and knowledge gaps to facilitate novel and evidence-informed clinical care regarding assistive walking devices for post stroke rehabilitation.

Conflict of interest

The authors have no conflict of interest to report.

References

Feigin

, Brainin

, Norrving

, et al. World Stroke Organization (WSO): Global stroke fact sheet 2022. Int J Stroke. 2022;17(1):18–29. doi:10.1177/17474930211065917

GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990-2019: A systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol. 2021;20(10):795–820. doi:10.1016/S1474-4422(21)00252-0

Danielsson

, Willén

, Sunnerhagen

. Physical activity, ambulation, and motor impairment late after stroke. Stroke Res Treat. 2012;2012:818513. doi:10.1155/2012/818513

Harvey

. Predictors of functional outcome following stroke. Phys Med Rehabil Clin N Am. 2015;26(4):583–98. doi:10.1016/j.pmr.2015.07.002

Woolley

. Characteristics of gait in hemiplegia. Top Stroke Rehabil. 2001;7(4):1–18. doi:10.1310/JB16-V04F-JAL5-H1UV

Balaban

, Tok

. Gait disturbances in people who have had a stroke. PM R. 2014;6(7):635–42. doi:10.1016/j.pmrj.2013.12.01

Estrada-Barranco

, Molina-Rueda

, Cano-de la Cuerda

. Validez de la Wisconsin Gait Scale para predecir el riesgo de caídas en pacientes con ictus [Validity of the Wisconsin Gait Scale to predict the risk of falls in stroke patients]. Rev Neurol. 2021;73(2):50–6. doi:10.33588/rn.7302.2020281

Michael

, Allen

, Macko

. Reduced ambulatory activity after stroke: The role of balance, gait, and cardiovascular fitness. Arch Phys Med Rehabil. 2005;86(8):1552–6. doi:10.1016/j.apmr.2004.12.026

Jeong

, Jeong

, Myong

, Koo

. Which type of cane is the most efficient, based on oxygen consumption and balance capacity, in chronic stroke patients? Gait Posture. 2015;41(2):493–8. doi:10.1016/j.gaitpost.2014.11.016

10.

Stinear

, Byblow

, Ackerley

, Smith

, Borges

, Barber

. Proportional motor recovery after stroke: Implications for trial design. Stroke. 2017;48(3):795–8. doi:10.1161/STROKEAHA.116.016020

11.

Jutai

, Coulson

, Teasell

, et al. Mobility assistive device utilization in a prospective study of patients with first-ever stroke. Arch Phys Med Rehabil. 2007;88(10):1268–75. doi:10.1016/j.apmr.2007.06.773

12.

Laufer

. The use of walking aids in the rehabilitation of stroke patients. Reviews in Clinical Gerontology. 2004;14(2):137–44. doi:10.1017/S0959259805001449

13.

Laufer

. Effects of one-point and four-point canes on balance and weight distribution in patients with hemiparesis. Clin Rehabil. 2002;16(2):141–8. doi:10.1191/0269215502cr481oa

14.

Choi

, Yang

, Jung

, Na

, Kim

, Cho

. Changes in lower limb muscle activation and degree of weight support according to types of cane-supported gait in hemiparetic stroke patients. Biomed Res Int. 2020;2020:9127610. doi:10.1155/2020/9127610

15.

Neumann

. An electromyographic study of the hip abductor muscles as subjects with a hip prosthesis walked with different methods of using a cane and carrying a load. Phys Ther. 1999;79(12):1163–76.

16.

Hamzat

, Kobiri

. Effects of walking with a cane on balance and social participation among community-dwelling post-stroke individuals. Eur J Phys Rehabil Med. 2008;44(2):121–6.

17.

Tricco

, Lillie

, Zarin

, et al. PRISMA extension for scoping reviews (PRISMA-ScR): Checklist and explanation. Ann Intern Med. 2018;169(7):467–73. doi:10.7326/M18-0850

18.

, Tu

, Shiue

, Lin

. Effects of cane use and position on performance of the sit-to-stand task in stroke patients. Am J Phys Med Rehabil. 2018;97(7):476–81. doi:10.1097/PHM.0000000000000898

19.

Boonsinsukh

, Panichareon

, Phansuwan-Pujito

. Light touch cue through a cane improves pelvic stability during walking in stroke. Arch Phys Med Rehabil. 2009;90(6):919–26. doi:10.1016/j.apmr.2008.12.022

20.

Tyson

, Rogerson

. Assistive walking devices in nonambulant patients undergoing rehabilitation after stroke: The effects on functional mobility, walking impairments, and patients’ opinion. Arch Phys Med Rehabil. 2009;90(3):475–9. doi:10.1016/j.apmr.2008.09.563

21.

Polese

, Teixeira-Salmela

, Nascimento

, et al. The effects of walking sticks on gait kinematics and kinetics with chronic stroke survivors. Clin Biomech (Bristol, Avon). 2012;27(2):131–7. doi:10.1016/j.clinbiomech.2011.08.003

22.

Beauchamp

, Skrela

, Southmayd

, et al. Immediate effects of cane use on gait symmetry in individuals with subacute stroke. Physiother Can. 2009;61(3):154–60. doi:10.3138/physio.61.3.154

23.

Boonsinsukh

, Panichareon

, Saengsirisuwan

, Phansuwan-Pujito

. Clinical identification for the use of light touch cues with a cane in gait rehabilitation poststroke. Top Stroke Rehabil. 2011;18(Suppl 1):633–42. doi:10.1310/tsr18s01-633

24.

Buurke

, Hermens

, Erren-Wolters

, Nene

. The effect of walking aids on muscle activation patterns during walking in stroke patients. Gait Posture. 2005;22(2):164–70. doi:10.1016/j.gaitpost.2004.09.003

25.

Deltombe

, Leeuwerck

, Jamart

, Frederick

, Dellicour

. Gait improvement in adults with hemiparesis using a rolling cane: A cross-over trial. J Rehabil Med. 2020;52(7):jrm00078. Published 2020 Jul 29. doi:10.2340/16501977-2705

26.

Jeong

, Jeong

, Kim

, et al. A randomized comparison of energy consumption when using different canes, inpatients after stroke. Clin Rehabil. 2015;29(2):129–34. doi:10.1177/0269215514543932

27.

Tyson

. Trunk kinematics in hemiplegic gait and the effect of walking aids. Clin Rehabil. 1999;13(4):295–300. doi:10.1191/026921599666307333

28.

Allet

, Leemann

, Guyen

, et al. Effect of different walking aids on walking capacity of patients with poststroke hemiparesis. Arch Phys Med Rehabil. 2009;90(8):1408–13. doi:10.1016/j.apmr.2009.02.010

29.

Maguire

, Sieben

, Frank

, Romkes

. Hip abductor control in walking following stroke – the immediate effect of canes, taping and TheraTogs on gait. Clin Rehabil. 2010;24(1):37–45. doi:10.1177/0269215509342335

30.

Jung

, Kim

, Cha

, In

, Hur

, Chung

. Effects of gait training with a cane and an augmented pressure sensor for enhancement of weight bearing over the affected lower limb in people who have had a stroke: A randomized controlled pilot study. Clin Rehabil. 2015;29(2):135–42. doi:10.1177/0269215514540923

31.

Kang

, Oh

, Cho

. Walking training with a weight support feedback cane improves lower limb muscle activity and gait ability in patients with chronic stroke: A randomized controlled trial. Med Sci Monit. 2021;27:e931565. Published 2021 May 30. doi:10.12659/MSM.931565

32.

Avelino

, Nascimento

, Ada

, de Menezes

KKP

, Teixeira-Salmela

. Using a cane for one month does not improve walking or social participation in chronic stroke: An attention-controlled randomized trial. Clin Rehabil. 2021;35(11):1590–8. doi:10.1177/02692155211020864

33.

Jung

, Bang

, In

, Cho

. Gait training with auditory feedback improves trunk control, muscle activation and dynamic balance in patients with hemiparetic stroke: A randomized controlled pilot study. J Back Musculoskelet Rehabil. 2020;33(1):1–6. doi:10.3233/BMR-170852

34.

Bernhardt

, Hayward

, Kwakkel

, et al. Agreed definitions and a shared vision for new standards in stroke recovery research: The Stroke Recovery and Rehabilitation Roundtable taskforce. Int J Stroke. 2017;12(5):444–50. doi:10.1177/1747493017711816

35.

World Health Organisation. International Classification of Functioning, Disability and Health. [Internet]. 2001 [cited 2023 Nov 27]. Available from: https://icd.who.int/dev11/licf/en

36.

Choi

, Yang

, Jung

, Na

, Kim

, Cho

37.

Sung

, Kim

, Yu

, Kim

. A hippotherapy simulator is effective to shift weight bearing toward the affected side during gait in people who have had a stroke. NeuroRehabilitation. 2013;33(3):407–12. doi:10.3233/NRE-130971

38.

Knikou

. Neural control of locomotion and training-induced plasticity after spinal and cerebral lesions. Clin Neurophysiol. 2010;121(10):1655–68. doi:10.1016/j.clinph.2010.01.039

39.

Sibbald

, Roberts

. Understanding controlled trials. Crossover trials. BMJ. 1998;316(7146):1719. doi:10.1136/bmj.316.7146.1719

40.

Jeong

, Jeong

, Myong

, Koo

. Which type of cane is the most efficient, based on oxygen consumption and balance capacity, in chronic stroke patients? Gait Posture. 2015;41(2):493–8. doi:10.1016/j.gaitpost.2014.11.016

41.

Munn

, Peters

MDJ

, Stern

, Tufanaru

, McArthur

, Aromataris

. Systematic review or scoping review? Guidance for authors when choosing between a systematic or scoping review approach. BMC Med Res Methodol. 2018;18(1):143. Published 2018 Nov 19. doi:10.1186/s12874-018-0611-x

42.

Kuan

, Tsou

, Su

. Hemiplegic gait of stroke patients: The effect of using a cane. Arch Phys Med Rehabil. 1999;80(7):777–84. doi:10.1016/s0003-9993(99)90227-7

43.

samuelsson

, Wressle

. User satisfaction with mobility assistive devices: An important element in the rehabilitation process. Disabil Rehabil. 2008;30(7):551–8. doi:10.1080/09638280701355777

44.

Mosey

. An alternative: The biopsychosocial model. Am J Occup Ther. 1974;28(3):137–40.

45.

Amanzonwé

, Tedesco Triccas

, Codjo

, Hansen

, Feys

, Kossi

. Exercise dosage to facilitate the recovery of balance, walking, and quality of life after stroke. S Afr J Physiother. 2023;79(1):1846. doi:10.4102/sajv79i1.1846

46.

Maguire

, Sieben

, de Bie

. The influence of walking-aids on the plasticity of spinal interneuronal networks, central-pattern-generators and the recovery of gait post-stroke. A literature review and scholarly discussion. J Bodyw Mov Ther. 2017;21(2):422–34. doi:10.1016/j.jbmt.2016.09.012

47.

Avelino

, Nascimento

, Menezes

KKP

, Ada

, Teixeira-Salmela

. Canes may not improve spatiotemporal parameters of walking after stroke: A systematic review of cross-sectional within-group experimental studies. Disabil Rehabil. 2022;44(10):1758–65. doi:10.1080/09638288.2020.1808088