Assessment of the strength of the lower limb muscles in subjects with stroke with portable dynamometry : a literature review

Introduction: Weakness of the lower limb muscles, which are the main impairments after stroke, is associated with reduced mobility and decreased performance in functional tasks. Therefore, the assessment of strength of these muscles is necessary, which is commonly assessed with portable dynamometry. Aims: To perform a literature review regarding the methods used to assess lower limb strength with portable dynamometry in subjects with stroke and to describe its investigated measurement properties with this population. Materials and Methods: An extensive search was performed on the MEDLINE, SCIELO, LILACS, and PEDro databases, by combining specific key words, followed by active manual search by two independent researchers. Results and Discussion: Thirty studies were included, and the muscular groups of the knee (90%) were the most assessed, followed by the ankle (66.7%) and hip (63.3%) joints. In 5% of the studies, * JCM: MSc, e-mail: julia_caetano@yahoo.com.br LTA: MSc, e-mail: larissatavaresaguiar@gmail.com EML: MSc, e-mail: izamlara@yahoo.com.br JBM: grad, e-mail: bmoura.juliana@gmail.com LACS: MSc, e-mail: lucasouza16@yahoo.com.br LFTS: PhD, e-mail: lfts@ufmg.br CDCM: PhD, e-mail: cdcmf@ufmg.br


Introduction
Stroke is defined as a brain injury produced by changes in the blood supply, which causes a set of neurological symptoms, which last for at least 24 hours (1).The neurological deficits caused by the stroke may lead to disabilities, which can last for months or remain for years, resulting in high burden to the patients, their families and to the health systems (2).Amongst the impairments caused by the stroke, muscular weakness is the most commonly observed (3).
Weakness of the lower limb (LL) muscles may lead to limitations in the ability to perform functional tasks, such as gait (4,5), stair ascent and descent (6), and sit-tostand transfers (7) and increases in energy expenditure to perform these tasks (6).Strength impairments of the LL muscles in subjects with stroke can increase the risk of falls 2.9 times, when compared with healthy subjects (8).Muscular strengthening programs (7,9,10) may modify these strength deficits, that affect gait speed (5) and functional mobility (7,9,10).Therefore, muscular strength must be carefully assessed, to guide clinical decision-making in stroke rehabilitation.
Nowadays, the method mostly used for the assessment of strength within clinical settings is the Manual Muscular Test (MMT).However, the MMT has some limitations: it is inaccurate (11,12), subjective, when muscular strength is rated as good or normal (12,13,14), and shows low responsiveness (12,15).Therefore, to accurately assess strength, it is necessary to apply a valid (16), reliable (11,15), and sensitive (11) method, that provides objective measures (11,17), such as portable dynamometry.Portable dynamometers are commonly used in research (18) and, in some situations, within clinical settings.They are easy-to-use devices and to perform the tests, the device is positioned between the examiner's hand and the muscular group under assessment, similar to the MMT assessment (12,18).Some factors may influence the measures obtained with portable dynamometers (19), such as positioning of the subjects and the device, number of trials, contraction and resting times, prior demonstration and familiarization with the procedures, and supply of verbal or visual encouragements.Some factors do not directly influence the acquisition of the strength measures, nevertheless, they could be important for the analyses of the results, such as unilateral or bilateral assessments, and the measurement properties of portable dynamometry for the assessment of strength in subjects with stroke (11,15).
In this context, the aims of the present study were: to perform a review of the literature, to verify if there were standardized protocols for the assessment of the strength of the LL muscles in subjects with stroke with portable dynamometry, as well as to verify which measurement properties were already investigated using this device with this population.Standardized protocols, employed for the assessment of muscular strength with portable dynamometry, would facilitate the test reproducibility within clinical and research contexts, which are important for comparisons between studies and evaluations.

Methods
Searches were performed on the MEDLINE (via PubMed), SCIELO, LILACS, and PEDro databases.The MEDLINE search strategy followed the recommendations of the Cochrane group (20), which was modified to suit the other databases.To select the studies related to the purpose of this review, the following descriptors related to LL and portable dynamometry, were used: "lower limb", "lower extremity", "membrum inferius", feet, foot, ankle, knee, hip, shank, leg, and thigh, dynamometer and "hand-held dynamometer".To be included, the studies should report in their method sections the evaluation of the strength of the lower limb muscles with portable dynamometers in subjects with stroke and be published until August, 2015.There were no restrictions regarding language of data of publication.Two independent examiners selected the studies, following three steps.The first step consisted of screening the titles of all studies found in the databases and excluding those that clearly did not meet the previously established criteria, followed by critical analyses of the abstracts, and the full papers.From the references of the selected studies, an active manual search was also performed, which followed the same criteria and procedures above described.Furthermore, when there was a disagreement between the examiners, a third reviewer resolved by consensus.

Results
The electronic search identified 808 studies.In the first step, 672 were excluded, for not meeting the inclusion criteria.In the second step, 63 studies were excluded and in the third step, eight studies were excluded.The 65 studies that met the inclusion criteria, 42 were duplicates.Thus, 23 studies were included from the electronic search.From the active manual search of these 23 studies, nine others were included.However, two could not be retrieved.Therefore, a total of 30 studies were included in this systematic review (Figure 1).The main reasons for the exclusion of the studies were the use of isokinetic dynamometers and/or assessment of strength in healthy subjects or in subjects with other diseases.
All of the 30 studies reported at least one clinical and demographic information of the included sample.In total, 965 subjects of both sexes, with ages ranging between 17 and 88 years, were evaluated.Moreover, the time since the onset of the stroke was also reported by the majority of the studies, ranging from acute (three days) to chronic (4934 days) phases, as shown in Table 1.

Equipment positioning:
Hip Flex: on the distal third and anterior part of the thigh.Knee Flex/Ext: on the distal third and posterior/anterior part of the thigh.Ankle DF: on the head of the metatarsal bones, dorsal part of the foot.Lin et al. (31) Bilateral Ankle DF/PF Ankle DF: SP, hip and knees extended.5 trials, Ankle PF: SP, hips and knees flexed at 90° (LL supported on a block).

Equipment positioning:
Hip Flex: on the distal third and anterior part of the thigh, trunk stabilization with strap.Knee Flex: on the distal third and posterior part of the leg, segment stabilization with strap.Knee Ext: on the distal third and anterior part of the leg, waist stabilization with strap.Ankle DF/PF: on the head of the metatarsal bones over the anterior/posterior part of the foot, waist stabilization with strap.
(To be continued)

Equipment positioning:
Hip Flex/Ext: on the anterior/posterior part of the thigh, proximal to the knee.Hip Abd: on the lateral part of the thigh, proximal to the knee.Knee Flex/Ext: on the posterior/anterior part of the leg, proximal to the ankle, stabilization on the distal and anterior part of the thigh.Ankle DF/PF: on the anterior/posterior part of the foot, proximal to the metatarsophalangeal joints, stabilization on the distal and anterior part of the leg.
(To be continued) Of the 30 included studies, five (16.7%) 27,34,38,45) did not provide information regarding the procedures of stabilization and positioning of the subjects and the dynamometer.Table 2 shows the positioning and stabilization procedures adopted in the included studies.
Of the studies that assessed the strength of the hip flexor muscles, 16 (53.3%)(6, 15, 21, 23, 24 -26, 28, 30, 32, 33, 37, 39, 40, 44, 46) described the positioning of the subjects and the supine position, with the hips and knees in extension, was adopted in six (20%) studies (23 -25, 28, 30, 39).All the eight studies, which assessed the strength of the hip extensor muscles, reported the positioning of the subjects and half (50%) (15,21,40,44), adopted the supine position with the hips and knees flexed.The seven studies (15,21,22,32,41,43,44) that assessed the strength of the hip abductor muscles, all adopted the supine position.The only study that assessed the strength of hip adductor muscles (15) adopted the supine position with the hip in neutral in the frontal plane and knees extended.
Few studies reported the use of visual or verbal feedback, to motivate the participants during the performance of maximal isometric contractions.Only only seven (23.3%) (11,15,23,27,33,35) reported some data and reliability was the only measurement property investigated.Three studies (10%) (23,33,46) assessed intra-rater reliability; one (14.3%)(11) inter-rater reliability, two (28.6%)(15.35) test-retest reliability, and one (14.28%)(27) internal consistency.All of these studies reported significant values with correlation coefficients above 0.70, indicating moderate to high reliability levels, based upon the classification adopted by Portney and Watkins (48) (Table 3).All muscular groups of the LL assessed with portable dynamometry had some type of reliability investigated.(11,39,44) reported the use of immediate verbal feedback.The demonstration and familiarization procedures were also rarely reported.
Six studies (20%) reported that the demonstration procedures were carried out and used verbal instructions (15,28,31,37,40,44) and two studies used movement instructions (15,44).Six studies (20%) (15,30,35,37,39,44) reported the familiarization procedures with the participants performing the same test procedures, prior to data collection.Of the 30 included studies, which investigated the measurement properties of portable dynamometry for the assessment of strength in subjects with stroke,

Discussion
The aims of this study were to perform a review of the literature to verify if there were standardized protocols for the assessment of the strength of the LL muscles of subjects with stroke with portable dynamometry, as well as to verify which measurement properties were already investigated using this equipment with this population.There was found large sample variability, including adults and elderly, male and female, at the acute, subacute, and chronic phases of stroke.The muscular groups of the knee were the most commonly assessed (90%), followed by ankle (66.7%), and hip muscles (63.3%).Over half of the studies provided information about the subjects' positioning employed for the assessment the strength of these muscular groups in subjects with stroke, since the weakness of the LL muscles is often observed in this population and may be associated with limitations in performing some functional activities (3,4,5,6).
Most of the studies (80%) evaluated the strength of the LL muscles, bilaterally.Muscular weakness in subjects with stroke is observed in both paretic and non-paretic limbs (49).The primary reason for the weakness of the non-paretic limb is related to the neuroanatomical characteristics, since approximately 10% of the descending motor fibers do not cross to the contralateral side, also leading to changes in strength in the muscles of the ipsilateral side of the brain injury (50).In addition, muscular atrophy that results from prolonged inactivity, enhances the weakness (51).Hamrin et al. (52) found that, in general, the torque of the knee flexor and extensor muscles in subjects with stroke is lower in faster movements and during flexion than during extension, when compared with healthy subjects.To determine the most affected side, the differences in strength between the paretic and non-paretic sides for a specific muscular group, can be calculated.Lower the differences, better will be the strength symmetry between the limbs of subjects with stroke (35).
When comparing the positions used in the studies included in the present review with those most commonly described for the clinical assessment of strength in healthy subjects using the MMT (13,53), some differences were observed.To measure the strength of the knee flexor and extensor muscles, the majority of the included studies adopted the seated position with the legs pending, and knee flexed at 90º.Kendall et al. (13), however, adopted the prone position with the thigh supported on a stretcher, hip with slight external rotation and knees flexed between 50º and 70º for the flexor muscles, and the seated position for the knee extensor muscles.In relation to the assessment of the hip flexors and extensors, and ankle dorsiflexors and plantar flexors, most studies adopted the supine position.For Kendall et al. (13), however, the positioning of the subjects was different, except for the hip flexor and ankle dorsiflexor muscles.According to them, the strength of the hip extensor and ankle plantar flexor muscles should be measured in prone position.Magee (53), however, evaluated the isometric strength of the LL muscles in the supine position, varying the hip and knee flexion angles, according to the muscular group to be tested.Only one study measured the strength of the of strength of the LL muscles and the supine position was the most used.Five studies did not describe the positioning of the subjects and the dynamometer, neither the stabilization procedures during data collection.Of the 50% of the studies, which reported the number of trials of muscular contraction, 73.3% performed three trials.Only 46.7% reported duration of the maximal isometric contractions and 42.9% used five seconds.Few studies reported the use of immediate verbal feedback (10%) and demonstration (23.3%).Few studies (23.3%) investigated the measurement properties of the portable dynamometer and reliability was the only property assessed, with significant results showing moderate to high reliability levels.
Strength measures were shown to be predictive of functional capacity and motor skills, length of hospital stay, and rehabilitation time (30).Muscular weakness of the LL muscles in subjects with stroke, for example, may be associated with reduced walking speed (21).In addition, strength deficit of the extensor muscles of the LL may be a limiting factor for the sit-to-stand performance and gait.The leg muscles play an important role to support impacts of high magnitude (34).The selection of the muscular groups assessed by the studies included in this review could be explained by the fact that these muscular groups can be more or less recruited for the task being performed (33), and these muscles are involved in many activities, such as walking, ascending and descending stairs, and making transfers (30,33).
Although extensive search for studies that evaluated the strength of the LL muscles with portable dynamometry, only one study measured the strength of the hip adductor muscles in subjects with stroke and none assessed the strength of the external and internal hip rotator muscles.According to Kendall et al. (13), weakness of the hip adductor muscles can compromise the efficiency in performing hip flexion, since they also act as hip flexors.As a result, this could lead to decreases in mobility (10), walking speed (6), and the ability to ascend and descend stairs (5).Weakness of the external hip rotators may be associated with medial rotation of the femur, followed by foot pronation, which causes knee valgus (13).Moreover, weakness of the internal hip rotators may laterally rotate the femur during the standing position and gait (13).Muscular weakness of the hip rotators will negatively affect mobility of these subjects (13).Therefore, it is also necessary to assess with both subacute and chronic stroke subjects showed that only one trial, after familiarization, was sufficient to provide consistent results (54,55,56).Therefore, although the majority of the previous studies used three trials for the assessment of strength with portable dynamometry in subjects with stroke, the use of fewer trials, specifically one after familiarization, can be applied, which enhance the applicability of the tests and decrease the evaluation time and the effect of muscle fatigue during strength tests in subjects with stroke (54,55).
Considering the isometric contraction time with portable dynamometry in subjects with stroke, most studies adopted five seconds.However, other studies adopted contraction times ranging between three and six seconds, except for one study, which adopted 15 seconds.According to Brum et al. (57), isometric contractions can increase blood pressure, since there is a mechanical obstruction of blood flow in response to isometric contractions, followed by accumulation of metabolites that activate chemoreceptors of the autonomic nervous sympathetic system, associated with increases in the peripheral vascular resistance.Fernandes and Marins (58) showed that the best isometric contraction time for healthy individuals was three seconds, in order to avoid this effect.Considering that the majority of subjects with stroke also has hypertension, it is necessary to carefully control the contraction time.
Few studies provided information regarding the rest time between the measurements.Moreover, the rest interval between trials ranged from 10 seconds to two minutes.Although rest time has not been commonly reported, it is important to reduce fatigue during strength tests (39,59).According to Nogueira et al. (59), adequate rest intervals can reverse the fatigue mechanism and provide time for energy recovery of the assessed muscular group.If this rest interval is insufficient, the muscular group can fatigue and this will negatively influence the values obtained with the measurements of strength.In the absence of a standardization procedure regarding the rest interval time to be used when evaluating individuals with stroke, it is always important to provide some rest intervals between the measurements of strength and to verify signs of muscle fatigue.Futures studies aiming at establishing the best rest intervals for this population during portable dynamometry strength tests may help the evaluation process.
hip and knee flexor/extensor muscles of subjects with stroke in the lateral position, reducing the action of the gravity on the tested segment (33).Further studies of this nature are needed to establish the best positioning of the individual and to facilitate the reproducibility of strength tests of the LL muscles within clinical and research contexts.
The stabilization of the segment to be tested, which varied with the adopted positioning, is also an important factor to assure that the subjects with stroke do not use compensatory strategies that may affect the results.Considering this, when assessing the strength of the hip flexor/extensor/abductor/ adductor muscles in the supine position, most studies provided stabilization of the trunk, while Kendall et al. (13) stabilized the pelvis contralateral for the hip flexors, in the supine position, and ipsilateral, for the hip extensors, in the prone position.For the knee flexor and extensor muscles, in the seated position, the distal third and anterior aspects of the thigh were stabilized in most studies.However, Kendall et al. (13) provided stabilization on the ipsilateral pelvis and posterior third and medial aspects of the thigh of the respective muscular groups in the prone position.Finally, for the ankle dorsiflexor and plantar flexor muscles, in the supine position, stabilization was provided at the ankle, whereas Kendall et al. (13) did not adopt any stabilization procedures.However, stabilization may be particularly difficult for the clinician, who is not always physically strong (15) and often requires the use of straps or belts, to ensure that the test is performed in the standard position, without interference of compensatory movements.
All studies, which reported the positioning of the device, positioned the dynamometer perpendicularly to the distal third of the assessed segment, so that the evaluator applied a force contrary to the direction of the movement.Although Kendall et al. (13) and Magee (53) performed sub-maximal strength tests of the LL muscles by means of isometric contractions, without using the hand-held dynamometer, both provided manual resistance on the distal third of the segment to be tested.This illustrates the positioning pattern of the applied resistance by the examiner.
Most of the studies included in the present review used three trials of muscular contractions, and the number of trials ranged from one to four.Recent studies that investigated if the number of trials (first trial, means of two and three trials) could affect the strength measurements with portable dynamometry the LL muscles in subjects with stroke, so that they can be used for this purpose.

Final Considerations
The use of portable dynamometry in subjects with stroke was most commonly employed for the assessment of the strength of the knee joint muscles, followed by the ankle and hip joint muscles.The majority of the studies provided some information regarding the positioning of the subjects during the tests, being the supine position mostly used.Some studies reported the procedures of stabilization of the tested segment and the distal third was the predominant site.The data collection protocols, regarding the number of trials, contraction time, and resting intervals were described in some studies; however, they were not standardized.Few studies reported procedures related to demonstration, familiarization, and/or incentives, which motivate maximal muscular contractions.Few studies investigated the measurement properties of portable dynamometry and the only assessed property was reliability, but with questionable statistical methods.Portable dynamometry seems to provide reliable measures of strength of the LL muscles in subjects with stroke and although there were not found any standardized protocols, it is a useful method to be employed within clinical contexts, since it provides objective measures of muscular strength.The demonstration and familiarization procedures were scarcely reported, although they are important to minimize the learning effects during data collection (39).Considering that individuals with stroke have difficulties in performing muscular contractions, particularly on their paretic side, in addition to difficulties in understanding the procedures, the provision of incentives are even more essential to obtain adequate measures of strength (49).However, only two studies included in the present literature review reported the use of incentives during the evaluation of strength.

References
Only seven studies (23.33%) investigated the measurement properties of portable dynamometry for the assessment of strength of the of LL muscles with subjects with stroke.Reliability was the only measurement property investigated.The test-retest and intra-rater reliabilities were the most investigated, probably due to the fact that is easier to collect data related to repeated measurements obtained by the same examiner, which is commonly used within clinical settings (measures of the same professional are compared before and after an intervention).Although the best statistical method to investigate the reliability of measurements obtained at different sessions or different examiners is the calculation of Intraclass Correlation Coefficients, as they reflect both the associations and the agreement levels between two or more quantitative measures (23,48), the majority of the studies reported the Pearson Correlation coefficients, which is not the most adequate method, since it only assesses the degree of association between the measures, regardless of the degree of agreement (23,48).
None of the studies included in this review investigated the validity of portable dynamometry for the assessment of strength in subjects with stroke.Portable dynamometers are devices comprised of strength cells and, therefore, have adequate face validity for strength measurements.In addition, adequate concurrent-related validity was reported by a previous study that compared the measures provided by portable dynamometers with those provided by isokinetic dynamometers in subjects with various health conditions (16).Considering that the validity of an instrument depends upon the context and the population (48), it is necessary to verify the criterion-related validity of portable dynamometers for the assessment of strength of

Table 1 -
Clinical and demographic characteristics of the participants included in the 30 studies that assessed the strength of the lower limb muscles with portable dynamometry in subjects with stroke

Table 1 -
Clinical and demographic characteristics of the participants included in the 30 studies that assessed the strength of the lower limb muscles with portable dynamometry in subjects with stroke

Table 2 -
Data extraction of the 30 studies, which assessed the strength of the lower limb muscles with portable dynamometry in subjects with stroke and provided information regarding the positioning or the data collection protocol Knee Flex/Ext: on the distal third and posterior/anterior part of the thigh, segment stabilization with strap.Ankle DF/PF: on the head of the metatarsal bones over the anterior/posterior part of the foot, ankle stabilization with strap.

Table 2 -
Data extraction of the 30 studies, which assessed the strength of the lower limb muscles with portable dynamometry in subjects with stroke and provided information regarding the positioning or the data collection protocol SP, hip flexed at 90° and knees relaxed.Hip Ext: SP, hip and knees flexed at 90°.Hip Abd: SP, hips in neutral of abduction and knees extended.Knee Flex/Ext: Sitting with hips and knees flexed at 90°.Hip Flex/Ext: on the distal third and anterior/posterior part of the thigh, trunk stabilization with strap.Hip Abd: on the distal third and lateral part of the thigh, trunk stabilization contralateral to the LL to be tested.Knee Flex/Ext: on the distal third and posterior/anterior part of the thigh, segment stabilization with strap.Ankle DF/PF: on the head of the metatarsal bones over the anterior/posterior part of the foot, leg stabilization with strap.

Table 2 -
Data extraction of the 30 studies, which assessed the strength of the lower limb muscles with portable dynamometry in subjects with stroke and provided information regarding the positioning or the data collection protocol Hip Flex/Ext: on the distal third and anterior/posterior part of the thigh.Hip Abd: on the distal third and lateral part of the thigh, trunk stabilization contralateral to the tested LL tested.Knee Flex/Ext: on the distal third and posterior/anterior part of the thigh.Ankle DF/PF: on the head of the metatarsal bones, over the anterior/posterior part of the foot.

Table 2 -
Data extraction of the 30 studies, which assessed the strength of the lower limb muscles with portable dynamometry in subjects with stroke and provided information regarding the positioning or the data collection protocol Hip Flex/Ext: on the distal third and anterior/posterior part of the thigh, trunk stabilization with strap.Hip Abd: on the distal third and lateral part of the thigh, trunk stabilization contralateral to the tested LL.Knee Flex/Ext: on the distal third and posterior/anterior part of the thigh, segment stabilization with strap.Ankle DF/PF: on the head of the metatarsal bones over the anterior/posterior part of the foot.
Knee Ext: Seated, hips and knees flexed at 90°.Equipment positioning:Hip Abd: on the lateral part of the thigh, 5 cm proximal to the lateral femoral epicondyle.Knee Ext: on the anterior part of the leg, 5 cm proximal to the medial malleolus.

Table 2 -
Data extraction of the 30 studies, which assessed the strength of the lower limb muscles with portable dynamometry in subjects with stroke and provided information regarding the positioning or the data collection protocol Note: NI = note informed; Flex=flexors; Ext=extensors; Abd= abductors; Add=aductors; DF= dorsiflexors; PF= plantar flexors; LL=lower limb; SP = supine position; PP = prone position; LP = lateral position

Table 3 -
Results of the seven studies which assessed the measurement properties of portable dynamometry 1. Sacco RL, Kasner SE, Broderick JP, Caplan LR, Connors JJ, Culebras A, Elkind MS, George MG, Hamdan AD, Higashida RT, Hoh BL, Janis LS, KaseCS,Kleindorfer DO, Lee JM, Moseley ME, Peterson ED, Turan TN, Valderrama AL, Vinters HV; American Heart Association Stroke Council, Council on Cardiovascular Surgery and Anesthesia; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular and Stroke Nursing; Council on Epidemiology and Prevention; Council on Peripheral Vascular Disease; Council on Nutrition, Physical Activity and Metabolism.An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American Heart Association/American Stroke Association.Stroke.2013 Jul;44(7):2064-89.