A delayed consequence of a lesion affecting the upper motor neuron pathways is the appearance of some forms of motor overactivity, including spasticity . Spasticity is an increase in resistance to the passive movement of a joint due to an abnormal increase in muscle tone, which is a range of amplitude and velocity dependent  .
Spasticity can be distinguished from other types of hypertonus, such as rigidity, by its sensitivity to velocity of joint motion and reaching maximum early in the range of motion. The magnitude and the pattern depend on the injury site and are usually higher in medullar lesions than cortical ones .
The pathophysiology underlying spasticity indicates increased stretching reflexes, excitability of either alpha and/or gamma motor neurons and abnormal reflex transmission. The role of neurotransmitters in the spasticity pathogenesis remains unclear  .
In scientific research, a broad range of methods and tools are used to assess muscle strength in individuals with spinal cord injury. The reality in clinical practice is otherwise dissimilar, where the tools available and currently used show important limitations .
Clinical scales such as the Ashworth Scale or the modified Ashworth are commonly used to quantify spasticity  . These are simple and practical instruments, however, outcome measurements remain highly subjective and personal factors related to the patient may affect the results. Some studies have found Ashworth Scales to be unreliable instruments for the assessment of muscle spasticity, showing poor interrater reliability and leaving the accuracy of the outcome measurements questionable . This lack of understanding or consensus may contribute to difficulties in measuring spasticity. The need for objective measurements concerning spasticity has been widely accepted.
In a literature review about instruments for measuring spinal injury muscle responses, Capelari et al. (2017) found many studies have confirmed the use of equipment like the isokinetic dynamometer and the portable dynamometer, ensuring more accurate results . This computer running device enables the investigator to standardize both velocity and angle of motion, and objectively record the amount of force generated by the patient’s muscles. The operation and interpretation of this method are simple and can be applied to a variety of joints and muscles .
This technique, which was initially developed to assess voluntary motor strength, was considered as an accurate and reliable possibility for the evaluation of spasticity in several joints .
The purpose of this review is to develop a critical view concerning the application of isokinetic dynamometer in neurological disorders with hypertonia as well as overview of measurement protocols applied. In this review, focus was defined on knee muscles because of their important role in maintaining upright posture in standing and walking, as well as in other functional activities such as squatting and stair climbing .
2.1. Data Sources and Searches
The electronic search using pre-defined search terms was restricted to English, French and Portuguese language publications retrieved from the following databases: PubMed, SciELO, Academic Search Complete and Scopus databases.
The initial research was limited to human studies published between 2000 and June 2020, however, it was extended before that because of the lack of studies to include respecting our inclusion criteria. Combinations of the following key words were used: “assessment”, “isokinetic dynamometer”, “spasticity”, “muscle tone” and “knee”. PubMed was searched using MeSH terms. The reference lists of retrieved articles were also screened for reports not identified through electronic searches.
2.2. Inclusion and Exclusion Criteria
To be included in this review studies must have:
1) Been published in peer-review journal as a full article or an abstract with sufficient detail to extract the main attributes of the study.
2) Adult subjects.
3) Used isokinetic dynamometer as an instrument to assess spasticity.
4) Applied a defined protocol to measure spastic knee muscle groups with sufficient detail.
Studies were excluded if:
• Data extraction was impossible;
• Measurements were not related to knee muscle groups.
2.3. Identification of Studies
Titles and abstracts of the retrieved articles were screened independently by RS and ALC against the eligibility criteria: Potentially eligible studies were identified and their respective full reports obtained. Full reports were then assessed separately by the two authors against the eligibility criteria. Discrepancies in judgement were resolved by consensus with consulting of CO. The percentage of agreement between both authors was calculated.
2.4. Assessment of Methodological Quality of Studies
Two authors (CO and ALC) independently evaluated the quality of the studies according to PEDro scale for RCTs and CCTs. Discrepancies were solved by consensus. PEDro scale is a valid measure for methodological quality of the studies and its scores can confidently be subjected to parametric statistical analysis .
2.5. Data Extraction, Synthesis and Analysis
One reviewer (RS) extracted relevant data from the included studies. These included information on:
- Sample characteristics.
- Assessment protocol.
- Outcome measures.
- Results and main conclusions.
A total of 151 studies were retrieved from the literature search and 69 duplicates were removed. From the remaining 82 studies, 66 were excluded base on title and abstract and 16 complete manuscripts were retrieved and assessed for inclusion by RS and ALC against the eligibility criteria (Figure 1). After this, 8 articles were included in this review        . The percentage of agreement between both reviewers was 92% and discrepancies were solved by consensus.
Figure 1. Flow chart for the systematic review.
3.1. Assessment of Methodological Quality of Studies
Concerning external validity, all studies with the exception of one  define and present inclusion criteria for participants. Related to internal validity, only two studies   randomized the sequence for the measurements. Blinding procedures were absent in all the studies. Although relevant for methodological quality, fact remains that the type of protocol we are considering for inclusion does not allow rigorous blinding procedures, nevertheless, data measuring is processed in a computer-based instrument, opaque to influences. Considering result reports, all studies present the results initially proposed in a clear and objective way (Table 1).
Table 1. Methodological quality checklist.
√ yes/× no.
The retrieved 8 articles included a total of 205 participants. Sample size varied between a minimum of 12  and a maximum of 33 participants .
3.3. Inclusion Criteria for Participants
Most of the studies established inclusion criteria in the studies based on spasticity evaluation with Ashworth scale     or Modified Ashworth Scale   while one study used the Gross Motor Function Classification System  or ASIA  . Some studies mentioned absence of knee orthopaedic problems      and absence of antispastic medication   .
3.4. Outcome Variables and Measurement Instruments
All studies used the isokinetic dynamometer to collect data regarding spasticity        . The Ashworth scale was used, in most studies      , and two studies used the ASIA scale  . EMG was used in three    of the studies analysed.
3.5. Type of Protocol
The assessment was performed with the patient seated in the chair of the isokinetic dynamometer, with the hips fixed between 70˚ and 90˚ flexion, using straps across the chest, the pelvis and upper thigh in all but one study . The rotational axis of the dynamometer was placed in line with the axis of flexion/extension movement of the knee. The lower leg was fixed to the lever of the dynamometer between 2 cm to 3 cm above the lateral malleolus. Akman et al. (1999) only refer to seated position lacking more detail.  Kakebeeke et al. (2002) measured every patient in two defined, stretched positions. In the seated position, the same protocol of the other studies was used. In the supine position, patients were positioned with the lower leg hanging over the edge of the seat; the pelvis was fixated with a strap over the anterior superior iliac spine; the lower leg was fixed with straps 4 cm above the lateral malleolus. The other leg was placed with flexed knee and hip with the foot on the table. This leg was held in position by one of the researchers .
Gravity correction was performed to account for the weight of the limb by all the researchers.
Previous to spasticity assessment, participants were distributed between control and experimental groups using assessment instruments like ASIA or Ashworth scale and assessment protocols were predefined in the isokinetic dynamometer (Table 2).
3.6. Comparisons and Outcome
Of the analysed studies, four      found a correlation between isokinetic dynamometer measures and other instruments, like Ashworth scale.
Table 2. Studies content summary.
n.d. ? not defined.
Concerning fatigue as a factor affecting spasticity, as measured by isokinetic dynamometer, two studies found no difference  . One study  compared two measuring protocols (seated and supine position) and found the excitability of the extensors of the knee was higher in the supine than in the sitting position, with a reverse result for the flexors In the sitting position, being easier to elicit spasms in the hamstring muscles.
Spasticity assessment using isokinetic dynamometer has been widely accepted as accurate and practical. Many authors have recurred to this particular form of outcome measurement associated with other instruments (like Ashworth scale) in order to increase outcome measurement reliability. Accuracy in outcome measurement is very important to allow objective interpretation along with more efficient planning for treatment programs.
Studies assessing hypertonia rely on an accurate definition and selection for participants, as well as on precisely defined assessment protocols considering all outcomes to measure with the isokinetic dynamometer (passive range of motion, angular velocity for PROM, number of repetitions for each velocity as well as methods for measuring spasticity through data interpretation for peak torque).
Boudarham et al. (2014), studying the relation between spasticity and fatigue found that repeated concentric contraction induced fatigue does not affect spasticity as assessed through peak torque resistance in any angular velocity.
This conclusion refutes previous conclusions by Franzoi, Castro & Cardone (1999), that found fatigue decreases stretching reflex, analysing a series of 5 repetitions for angular movement at any angular velocity in the knee.
Kakebeeke et al., (2002) measure spasticity on knee muscular groups in two different positions, sitting and in supine. The purpose of this study was to understand the importance of muscle length as a factor to consider when assessing spasticity. For both groups, flexors and extensors, torques obtained where significantly higher in elongated position leading to the assumption that measures are more reliable when the muscle is in this position.
This fact is logical if considered that the elongated muscle offers higher resistance to the passive mobilization, hence, that spasticity is increased. Therefore, the outcome assessment position influences the results, once the muscle position affects the response to passive movement.
Considering hip position influences the assessment and, consequently, the outcomes, when measuring knee muscle spasticity, studies must define protocols where hip position is considered and registered when performing isokinetic dynamometer testing. Perell, Scremin & Kunkel. (1996) referred the subjects were positioned with 90˚ hip flexion, Franzoi, Castro & Cardone (1999) leaves the register of sitting position with the back of the chair at 85˚ and Moreau, Li, Geaghan & Damiano, (2009) consider a sitting position, reclined, in a 70˚ hip flexion angle. More recent studies, measured spasticity in hip position of 85˚  or reclined back in a 15˚ vertical angle . The improved definition of hip position while measuring spasticity raises the importance of considering this factor as a possible influence on studies results.
Generally, isokinetic dynamometry has been showed to be a useful method when quantifying spasticity. In order to easily apply this assessment in a clinical context, Perell, Scremin & Kunkel., (1996) found the measurement at 120˚/s is sufficient to assess spasticity. Franzoi, Castro & Cardone (1999) reenforce this when describing the highest difference between torque averages occurrs at 120˚/s, highlighting that lower velocities (30 and 60˚/s) are not accurate to detect mild degrees of spasticity.
Considering Ashworth scale supplies data about spasticity assessment, although with low accuracy, most studies stablished this scale as a reference to compare results measured by the isokinetic dynamometer. Considering the final analysis of the outcome measurements, one may considerer the dynamometer is a reliable instrument to asses spasticity with good test-retest reliability allowing a clinical application to assess patient outcomes throughout a rehabilitation program.
More studies are suggested, particularly comparing measurements for spastic muscles with resource to other instruments like EMG to better define reliability for isokinetic dynamometer.
Assessing spasticity with the resource of scales is possible, however, lacks the necessary accuracy. The resource of an isokinetic dynamometer allows more precise and effective measurements for quantifying spasticity. This procedure offers good teste-retest reliability and it is easily applicable in a clinical context. Thus, the results of a clinical intervention can be verified and quantified making it possible to adapt treatment programs to the results obtained. Treatment efficacy can be measured calculating torque values before and after clinical interventions.
Spasticity assessment requires a definition of protocols considering both sampling and methods. The position of the participant, his understanding of the instructions, as well as specific details regarding PROMs, angular velocity and number of repetitions are definitive for the accuracy of measured values.
Concluding, isokinetic dynamometer is a valuable tool to address difficulties in assessing abnormal muscle tone, showing a high correlation with Ashworth scale results.
 Sheean, G. and McGuire, J.R. (2009) Spastic Hypertonia and Movement Disorders: Pathophysiology, Clinical Presentation, and Quantification. PM and R, 1, 827-833.
 Capelari, T.V., Borin, J.S., Grigol, M., Saccani, R., Zardo, F. and Cechetti, F. (2017) Evaluation of Muscle Strength in Medullar Injury: A Literature Review. Coluna/Columna, 16, 323-329.
 Ansari, N.N., Naghdi, S., Moammeri, H. and Jalaie, S. (2006) Ashworth Scales Are Unreliable for the Assessment of Muscle Spasticity. Physiotherapy: Theory and Practice, 22, 119-125. https://doi.org/10.1080/09593980600724188
 Akman, M.N., Bengi, R., Karatas, M., Kilinç, Ş., Sözay, S. and Özker, R. (1999) Assessment of Spasticity Using Isokinetic Dynamometry in Patients with Spinal Cord Injury. Spinal Cord, 37, 638-643. https://doi.org/10.1038/sj.sc.3100892
 Boudarham, J., et al. (2014) Relationship between Neuromuscular Fatigue and Spasticity in Chronic Stroke Patients: A Pilot Study. Journal of Electromyography and Kinesiology, 24, 292-299. https://doi.org/10.1016/j.jelekin.2013.11.006
 Moreau, N.G., Li, L., Geaghan, J.P. and Damiano, D.L. (2009) Contributors to Fatigue Resistance of the Hamstrings and Quadriceps in Cerebral Palsy. Clinical Biomechanics, 24, 355-360.
 De Morton, N.A. (2009) The PEDro Scale Is a Valid Measure of the Methodological Quality of Clinical Trials: A Demographic Study. Australian Journal of Physiotherapy, 55, 129-133.
 Kakebeeke, T.H., Lechner, H., Baumberger, M., Denoth, J., Michel, D. and Knecht, H. (2002) The Importance of Posture on the Isokinetic Assessment of Spasticity. Spinal Cord, 40, 236-243.
 Franzoi, A.C., Castro, C. and Cardone, C. (1999) Isokinetic Assessment of Spasticity in Subjects with Traumatic Spinal Cord Injury (ASIA A). Spinal Cord, 37, 416-420.
 Perell, K., Scremin, A., Scremin, O. and Kunkel, C. (1996) Quantifying Muscle Tone in Spinal Cord Injury Patients Using Isokinetic Dynamometric Techniques. Paraplegia, 34, 46-53.
 Hameau, S., Bensmail, D., Robertson, J., Boudarham, J., Roche, N. and Zory, R. (2014) Isokinetic Assessment of the Effects of Botulinum Toxin Injection on Spasticity and Voluntary Strength in Patients with Spastic Hemiparesis. European Journal of Physical and Rehabilitation Medicine, 50, 515-523.