Upper extremity asymmetry due to nerve injuries or central neurologic conditions: a scoping review

Abstract

Background

Peripheral nerve injuries and central neurologic conditions can result in extensive disabilities. In cases with unilateral impairment, assessing the asymmetry between the upper extremity has been used to assess outcomes of treatment and severity of injury. A wide variety of validated and novel tests and sensors have been utilized to determine the upper extremity asymmetry. The purpose of this article is to review the literature and define the current state of the art for describing upper extremity asymmetry in patients with peripheral nerve injuries or central neurologic conditions.

Method

An electronic literature search of PubMed, Scopus, Web of Science, OVID was performed for publications between 2000 to 2022. Eligibility criteria were subjects with neurological conditions/injuries who were analyzed for dissimilarities in use between the upper extremities. Data related to study population, target condition/injury, types of tests performed, sensors used, real-world data collection, outcome measures of interest, and results of the study were extracted. Sackett’s Level of Evidence was used to judge the quality of the articles.

Results

Of the 7281 unique articles, 112 articles met the inclusion criteria for the review. Eight target conditions/injuries were identified (Brachial Plexus Injury, Cerebral Palsy, Multiple Sclerosis, Parkinson’s Disease, Peripheral Nerve Injury, Spinal Cord Injury, Schizophrenia, and stroke). The tests performed were classified into thirteen categories based on the nature of the test and data collected. The general results related to upper extremity asymmetry were listed for all the reviewed articles. Stroke was the most studied condition, followed by cerebral palsy, with kinematics and strength measurement tests being the most frequently used tests. Studies with a level of evidence level II and III increased between 2000 and 2021. The use of real-world evidence-based data, and objective data collection tests also increased in the same period.

Conclusion

Adequately powered randomized controlled trials should be used to study upper extremity asymmetry. Neurological conditions other than stroke should be studied further. Upper extremity asymmetry should be measured using objective outcome measures like motion tracking and activity monitoring in the patient’s daily living environment.

Introduction

It is estimated that 795,000 people suffer a stroke in the United States yearly of which 70% are first time strokes [1]. Between 2015 and 2018, 7.6 million people suffered a stroke [2]. Cerebral Palsy (CP) affects more than 17 million people worldwide [3,4,5,6]. Each year, about 60,000 Americans are diagnosed with Parkinson’s disease (PD) [7]. In 2020, the number of known cases of Multiple Sclerosis (MS) increased to 2.8 million worldwide [8]. Amyotrophic Lateral Sclerosis (ALS) affected about 12,187 people in the United States between 2010 and 2011 [9].

Traumatic injuries resulting in peripheral nerve injury (PNI) (including brachial plexus injury (BPI)) and spinal cord injury (SCI) cause extensive disabilities in the upper extremity (UE). Motor vehicle accidents (MVA) are the predominant cause of PNI, where 5% of all MVAs result in a form of PNI [10]. About 8% of PNI patients have a BPI [11]. Adult traumatic BPI results in severe impairment following penetrating wounds, falls, and MVA or other high-energy trauma. Young male adults comprise a majority among patients with a BPI [12]. MVAs and falls are a leading cause of SCI. 38.1% of all SCI were caused by MVAs and 53% by falls between 2010 and 2014 [13]. The United States has an estimated annual SCI incidence of 17,000 [14]. The National Spinal Cord Injury Statistical Center estimated 282,000 people were living with a SCI in 2016 [15].

Conditions/injuries affecting the nervous system can be debilitating. Stroke victims suffer paretic limbs [16] and is the third-leading cause of disability [17]. CP accounts for most of the lifelong neurological disabilities [3,4,5,6]. Bimanual coordination impairments were found in children with CP [18,19,20]. BPI predominately affects young and otherwise healthy men resulting in paralytic upper extremities [21]. Patients with MS display reduced gross or fine motor capabilities, in addition to slowness, clumsiness, and dysmetria [22,23,24], while patients with PD displayed a deficit in inhibitory control [25,26,27,28,29,30]. Many of the UE PNI patients, despite rehabilitative efforts, never achieve satisfactory motor recovery [31,32,33,34]. All of these neurological conditions/injuries can affect the patient’s ability to perform daily tasks with their affected UE.

A wide range of outcome measures have been described in the literature as well as clinical practice to quantify the disabilities of the UE. These measures involve tests and surveys on dexterity, strength, pain, disability, amount of activity, etc. When the neurological condition/injury is unilateral, the differences in the outcome measures between the ipsilateral and contralateral sides can be used to gauge the patient’s recovery or the progression of the condition. Such a comparison between the affected and the unaffected sides is beneficial and provides patient specific, real-world evidence (RWE) [35] based data and information. Real world evidence is a combination of data collected outside of a traditional clinical setting. Asymmetry measures can either be subjective or objective, measured in a clinical setting or in the patient’s daily living environment. Clinical surveys to measure outcomes are either too extensive or too limited in scope, are completed without supervision and is time intensive for patients with a relatively high rate of failure to complete. This emphasizes the importance to understand the benefits and shortcomings of each type of test in relation to the treatment population.

Previous reviews have explored wearable systems [36,37,38], fine and gross motor tests [39], and various functional evaluation techniques [40] for UE assessment and rehabilitation. No currently available review or article explores the state of the art in assessing UE asymmetry caused by nerve injuries or central neurologic conditions. Hence, it is necessary to investigate the available literature and identify the present gaps in knowledge and redirect research focus onto such gaps.

The purpose of this review is to document the various tests/techniques/sensors used in clinical practice to assess UE asymmetry in a population with neurological conditions/injuries. This work was undertaken with the following goals:

• 1. Report trends in studies published from 2000 to 2022.

• 2. Classify the most widely used tests/techniques/sensors to assess UE asymmetry.

• 3. Classify the most widely studied neurological conditions/injuries.

• 4. Analyze the frequency of RWE based approaches compared to in-clinic approaches.

Methods

The PRISMA-ScR guidelines for reporting scoping reviews were followed [41, 42]. The project was registered with the Open Science Framework (https://0-doi-org.brum.beds.ac.uk/10.17605/OSF.IO/8PFUW). A protocol was created and followed for the review (https://osf.io/bk3at).

Literature search

Initial searches were performed on 1/26/2023 in PubMed, Scopus, and Web of Science. Date limits were set from 1/1/2000 forward. A follow-up search of multiple databases was performed on April 18, 2023. Results were also limited to 1/1/2000 forward. Databases for the follow up search were Ovid MEDLINE(R) (1946 + including epub ahead of print, in-process, and other nonindexed citations), Ovid Embase (1974 +), Ovid Cochrane Central Register of Controlled Trials (1991 +), Ovid Cochrane Database of Systematic Reviews (2005 +), and Scopus via Elsevier (1970 +). The initial search was performed by the study investigator SGB. The expanded and updated search strategies were designed and conducted by a medical librarian with input from the study investigators (SGB, AYS, KRK). The searched articles were limited to a publishing date between 1/1/2000 to 12/31/2022. Controlled vocabulary supplemented with keywords was used for this search. The actual strategies for each search that lists all search terms and how they are combined is available in the Table 1. Results included journal articles and peer reviewed conference proceedings in the English language and articles with available English translation. Duplicates were identified and removed from the main list using Endnote X9 (Clarivate, Philadelphia, PA).

Table 1 Search strategies

Eligibility criteria

A thorough review of the literature was performed based on the following criteria.

Inclusion criteria:

• The study involved analysis of the dissimilarities in activity between the upper extremities (right vs. left) and included any assessment method to analyze this dissimilarity.

• The study involved subjects with neurological conditions/injuries.

Exclusion criteria:

• The article was a systematic review, a case study, or a book chapter.

• The study included only unimpaired subjects (No treatment group).

• The study involved analysis of the UE during indirect periodic tasks (e.g., gait, etc.).

• The article did not focus on the asymmetry between the upper extremities (e.g., device design and validation articles).

• The study was performed on newborns or infants.

• The article explored the UE activity via mathematical modelling.

Selection process

The selection process included several steps (Fig. 1). Database search results (publication years 2000 forward) were checked for duplicates. These unique articles were then screened using their abstracts for relevance to the review topic. Full texts for the screened articles were accessed online. The articles that could not be retrieved (for any reason) were disregarded from the review. The retrieved articles were assessed and selected based on the eligibility criteria. Several systematic review articles [36,37,38,39] related to the current review topic were identified during the search. The references listed in these review articles were screened and the process described above was performed on these references to check for eligibility.

Fig. 1 

PRISMA flowchart for the literature search and exclusions. “n” is the number of articles in each given step

Data extraction

Selected articles were reviewed and the following data were extracted as per the protocol: study population, target condition/injury, types of tests performed, sensors used, real-world data collection as described in [35], outcome measures, and study results. Statistical significance was recorded. Data collected was classified as either objective, subjective, or a mixture of both. The search and mark functionality were used on pdf reader applications such as Adobe acrobat reader DC v2022.001.20142 (Adobe, San Jose, CA, USA) and Notability v11.3.1 (Ginger Labs, San Francisco, CA, USA). The extracted data were charted in an excel sheet (Excel, Microsoft 365 apps for enterprise, version 2211, Microsoft corporations, Redmond, WA, USA), and the counts were plotted using R version 4.2.0.

Critical appraisal

A modified Sackett’s Level of Evidence (LoE) [43] was used to judge the quality of the studies based on the information provided in the articles. Level I indicated the study was a randomized controlled trial (RCT), Level II were cohort studies, Level III were case-controlled studies, and Level IV were poorly designed case-controlled studies. The LoE value indicating quality of the articles decreased from Level I to IV.

All the above steps were performed by SGB under the supervision of AYS and KRK.

Results

Selection of sources

From a total of 7281 unique articles, 112 met the specified criteria and were included in this review. Details of the exclusions performed are listed in Fig. 1. Review articles or case studies [36,37,38,39, 44,45,46,47,48,49,50,51,52,53] and studies without a treatment population (conducted exclusively on healthy subjects) [54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105] were excluded. Eleven articles studied UE asymmetry during indirect periodic movements (e.g., gait, etc.) [106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121]. Some excluded studies were designed to assess a specific device’s design [122,123,124,125], focused on mathematical modelling and analysis of the asymmetries [111, 126,127,128,129,130], or did not study the UE asymmetry resulting from any specific neurological condition (e.g., [131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151]). Articles focused on the reliability or validity of methods were also excluded [152,153,154,155,156,157,158,159,160,161,162,163]. Exclusions were also performed due to the subject population studied (Amputees [164, 165], and newborns/toddlers [166,167,168,169,170,171]). The study population, objectives and significant results from the reviewed articles are listed in Table 2.

Table 2 Summary of the articles selected for review (sorted by the target condition and first author’s last name)

The selected articles concentrated on BPI, CP, MS, PD, PNI, SCI, Schizophrenia (SCZ), and stroke. It is important to note that even though SCZ is not a nervous system lesion or injury, it is a neurological disorder and presents with upper extremity behavioral asymmetry. Stroke was the most prominent condition (63 articles) followed by CP (27 articles). PNI, and SCZ were the least studied conditions/injuries (1 article each) (Fig. 2a).

Fig. 2

Number of articles divided by a target condition/injury, b level of evidence, c type of study, and d publication year (n = number of articles)

Critical appraisal of sources

The reviewed articles had a higher proportion of Level III and IV of evidence (44 and 35 counts respectively). Five articles were at Level I evidence, and 28 articles contained Level II evidence (Fig. 2b and Table 2). The number of Level II and III articles increased over the years (Fig. 2d), with the increase being concentrated over 2018 to 2021. 93 articles performed data collection in a clinical setting, while 19 articles collected complete or partial RWE (Fig. 2c). An upward trend was seen in the number of articles published over the years, with the highest number of articles published in 2020 and 2021 (11 articles each) (Fig. 2d).

Extracted data

The tests and sensors used in the selected studies were divided into thirteen categories based on the type of test and data collected (Table 3). SGB divided the tests into respective categories based on each test’s description (as stated in databases such as the Shirley Ryan Ability Lab rehabilitation measure database) and data collected. Kinematics tests were the most common types of tests performed (37 instances) followed by strength measures (36 instances). Motion reflex test was the least common type of test used (1 instance) (Fig. 3c). The articles contained 153 instances of objective tests, 48 instances of subjective tests, and 11 instances of mixed data type tests (Table 3). Objective data collection increased between 2000 to 2021 (from 6 to 18instances) (Fig. 3a). Stroke had the highest number of objective tests (105 instances) and subjective tests (25 instances), followed by CP (34 and 12 instances, respectively) (Fig. 3b). All activity monitoring, electromyography, kinematic, and motion reflex tests collected objective data, while amount of use, handedness, quality of life, and spasticity tests collected subjective data. Some disability tests of fine motor skills or strength collected mixed data. Strength measures were the most popular outcome measures for stroke (28 instances) followed by fine motor tests (20 instances), activity monitoring (19 instances) and disability measures (18 instances). Gross motor tests were used primarily on the CP treatment groups (15 instances).

Table 3 Categories of tests and sensors used in the selected articles

Fig. 3

Heat map distribution of the tests performed in the selected articles by target condition/injury and test categories; Number of articles divided by a publication year, b target condition/injury, c test categories

Most of the articles reviewed performed tests and collected data in the clinic (93 articles). Articles on stroke as the target condition saw most of the real-world data collection (16 articles) (Fig. 4a). Studies using RWE based tests had a higher treatment group population (median: 20) compared to studies using in-clinic tests (median: 16) (Fig. 4b). The number of studies utilizing real-world data increased between 2000 and 2021 with only 1 out of 6 articles in 2000 compared to 5 out of 11 articles in 2021 containing some form of RWE (Fig. 4c).

Fig. 4

Distribution of the data collection method for a target condition/injury, b treatment group population vs. type of study, and c publication year (n = number of articles)

Discussion

Summary of evidence

The evolution of the science for assessing UE asymmetry secondary to neurological conditions/injuries from 2000 to 2022 was reviewed. Most articles were Level III or IV of evidence. Stroke and CP were the most studied conditions. Tests collecting data on the kinematics and strength were widely used. RWE based testing and objective data collection increased between 2000 and 2021.

Stroke and CP were the most common conditions studied. Although CP and the other neurological conditions affect the patient chronically, stroke resulted in the most deaths, while also causing chronic paralysis in the patients who survived. Research funding allocated to stroke by the National Institute of Health (NIH) has seen a rise of 51% between 2008 and 2021, $296 million to $447 million respectively [283]. Within the same period, funding for CP increased by 7% ($28 to $30 million), PD by 67% ($152 to $254 million), SCI by 26% ($80 to $101 million). The funding for MS and SCZ reduced by 25% ($169 to $126 million) and 2% ($249 to $242 million) respectively [283]. No data was found for BPI and PNI. The neurological conditions other than stroke and PD are underfunded and under-studied. Hence, it is recommended that future research endeavors focus on a more diverse patient population affected by neurological conditions/injuries.

A higher proportion of the studies reviewed were case-controlled or poor cohort/ case-controlled studies (LoE Level III and IV). The quality of a study is largely dependent on its design. It is widely accepted that RCTs (LoE Level I) are the “gold standard”, but there are certain disadvantages associated with them (e.g., expensive to conduct, monitoring biases, quality is dependent on degree of randomization, etc.) [284]. Due to these disadvantages, cohort studies (LoE Level II) are often preferred. Case–control studies are retrospective (LoE Level III), hence do not explore the incidence of the outcome. Case–control studies also have many shortcomings as discussed by Lewallen et. al. [285]. Hence, future studies on UE asymmetry should be designed as either a RCT or a cohort study (LoE Levels I or II).

Most of the tests performed in the articles collected objective data. Qualitative tests often require less personnel training to collect, are less time consuming, and do not require expensive equipment or sensors. However, they often depend on a skilled observer or a dedicated patient to report the outcome based on a set guideline. The subjective data collected is dependent on perception or feeling at the time of the test. They also require skilled observers and if not attentive, they may inadvertently induce bias. Objective data provides a more unambiguous and unbiased representation of outcome. The use of objective data for quantifying UE asymmetry increased between 2000 and 2022, and further studies are necessary to determine their applicability and reproducibility.

The reviewed articles extensively employed tests to capture the kinematics of patients across all studied conditions and injuries. Among the various methods used for studying the subject's kinematics, Optical Motion Capture (OMC) stood out as the most prominent. According to an article that reviewed the state of the art in human motion tracking, Optoelectronic measurement systems (also known as OMC systems) were found to be more accurate than other systems, but they relied on proper calibration [286]. In contrast, Electromagnetic motion tracking systems were less accurate than OMC. They exhibited greater susceptibility to electromagnetic noise and had a lower sampling rate [286]. On the other hand, Inertial Measurement Units (IMUs) gained significance as a wireless and marker less motion tracking technology in recent years. IMU-based systems offered advantages such as being lightweight, cost-effective, and portable. However, there were certain concerns related to angle calculation that needs careful consideration [287]. Therefore, selecting the appropriate motion tracking system became crucial and depended on the specific use-case at hand.

Force transducers, primarily handheld dynamometers, were the dominant type of test used, appearing in 33 instances. As Mendoza et al. highlighted, handheld dynamometry (HHD) offered an efficient, objective, sensitive, and cost-effective method for quantifying strength [288]. Nevertheless, recent literature indicates that HHD is prone to intertester variability, meaning that different testers may produce varying results when using the same device [289]. Additionally, the reliability of HHD is influenced by the strength of the tester, particularly when assessing larger muscle groups. This aspect must be considered while interpreting the results obtained through HHD measurements.

Activity monitoring tests were one of the popular forms of outcome measure applied. Hollis et. al. supported the use of accelerometry by stating that it “…is not contaminated by learning and practice effects that may occur with repeated administration of standard measures.” [238]. When used to measure limb use asymmetry, accelerometry has been well correlated with standard clinical assessments [174, 290]. De Lucena et. al. stated “If quality of movement during daily life is an outcome important to people with a stroke, perhaps kinematic analysis of accelerometry provides a window to assess it.” [232]. Lakhani et. al. suggested the use of accelerometry in conjunction with other outcome measures to predict impairment in individuals with chronic stroke [250]. Toba et. al. emphasized the importance of the method used to analyze the actigraphy data [276]. Hence, when used appropriately, activity monitors have the potential to become a valid outcome measure in clinical practice.

Studies adopting RWE based outcome measures have increased in the recent years. Rehabilitation of patients with neurological conditions/injuries aims at improving the use of the affected UE in daily living tasks [233]. Webber et. al. noted that “…collection of real world data places minimal burden on subjects and provides quantitative arm usage information previously inaccessible to clinicians.” [174]. Similar observations were made by other articles [240, 252, 278]. The U.S. Food and Drug Administration (FDA) emphasized the importance of real world data and real world evidence to supplement clinical data in medical device clearance and best practices development [291]. RWE based approaches provide a better understanding of the patient’s condition in their day-to-day life as opposed to data collection in a research/clinical setting [35]. Pau et. al. pointed out the shortcoming of in-clinic tests, stating “…clinical tests capture only limited information about individuals’ actual upper limb dysfunction…” [201]. John Doyle stated “Real-world patients are fundamentally different than clinical trial patients.”, hence supporting the use of RWE in medical device testing [292]. All these testimonies recommend RWE and data be used in prognostic, diagnostic, and rehabilitative care of patients. There are number of limitations associated with RWE as stated by Kim et. al. (need for experts for data analysis, high possibility of bias, lack of standardization, etc.) [293]. RWE based tests often require an extended period of data collection. These limitations make it difficult to maintain a large subject size. All studies reviewed had a statistically significant result but might have been underpowered since a power analysis was not provided. Hence, it is recommended that a power analysis be performed for RWE based studies and the population size be large enough to ensure reliable statistical analyses.

Limitations

There are limitations inherent to retrospective, scoping reviews. Only articles published within the range of 2000 and 2022 were considered for the review. Articles were selected from a list of articles only published in English (including English language translations). Hence, there is a possibility of missing the knowledge from publications in other languages. The software used for identifying the duplicate articles in the master list was trusted and the results were not cross referenced. There is a chance that some articles could have been overlooked due to the terms used for the electronic search. Hence, the search involved using multiple synonyms to reduce the risk of data loss. This review focused on multiple conditions and injuries. The quality assessment tools used for these conditions/injuries, though similar, still have some differences. We used the LoE to judge all the articles to facilitate a fair comparison. Systematic reviews, case studies, and book chapters were excluded from our review.

Conclusions

The discussed limitations notwithstanding, this review demonstrated the following:

• More randomized controlled trials or cohort studies (LoE I or II) are needed in studies on UE asymmetry to improve the level of evidence being reported.

• Real-world outcome measures should be collected more frequently.

• Objective outcome measures should be given more importance.

• UE asymmetry for neurological conditions other than stroke need to be studied.

• Adequate power analysis must be performed to ensure reliable analyses.