journal of neuroscience nursing clinical measurement of...

Copyright @ 201 American Association of Neuroscience Nurses. Unauthorized reproduction of this article is prohibited.1

Clinical Measurement of Limb Spasticity inAdults: State of the Science

Rozina H. Bhimani, Lisa C. Anderson, Susan J. Henly, Sarah A. Stoddard

ABSTRACTSpasticity is a neuromuscular dysfunction characterized by tight or stiff muscles. Spasticity occurs acrossthe spectrum of upper motor neuron disease and complicates the course and quality of life of thoseaffected. Accurate and precise assessment of spasticity is the first step in providing safe and effectivetreatments to patients for management of spasticity. Examiner evaluations (Ashworth Scale, ModifiedAshworth, and Visual Analog Scale) and patient self-reports (Visual Analog Scale and Numeric RatingScale) are used to assess spasticity in clinical practice. We reviewed the biology of spasticity andsummarized research that assessed properties of scores obtained from clinical scales when used in avariety of upper motor neuron diseases. The definition of spasticity was inconsistent. Rater reliability oragreement on clinical scales varied widely. Correspondence with electromyogram results was mixed.There was dissimilarity in patient reports and examiner assessments. Scores from clinical scales areresponsive (decrease after initiation of treatment with known effectiveness), but the utility of scores forindexing individual change associated with the natural history of upper motor neuron disease is unknown.Future research incorporating patient reports and examiner findings over time will help to clarify thedefinition and capture the essence of spasticity.

Upper motor neuron (UMN) lesions are centralnervous system impairments that diminishmotor control and create emergence of patho-

logical signs such as spasticity, rigidity, clonus, andhyperreflexia (Leonard, Gardipee, Knootz, Anderson,& Wilkins, 2006). Spasticity, characterized by tightor stiff muscles, occurs in UMN diseases such as spi-nal cord injury, multiple sclerosis, stroke, and trau-matic brain injury (Pettibone, 1988). Spasticitycontributes to pain, insomnia, and fatigue and caninterfere with mobility, transfers, self-care, activities ofdaily living, and social functioning (Bhimani, 2008).Spasticity increases caregiver burden. A person ex-periencing spasticity requires passive range of motionto the limbs upon awakening and at bedtime so thatpersonal care activities such as toileting and dressingcan be carried out with some ease. Untreated spasticity

can lead to permanent muscle contractures (Ashworth,Satkunam, & Deforge, 2006).

Clinical factors such as urinary tract infectionsand decubitus ulcers sometimes increase spasticity(Nuyens et al., 1994; Skold, 2000). Spasticity varies indynamic (moving) and static (sustained contraction)states (Leonard et al., 2006; Pandyan et al., 2005).Repeated stretching and posture changes may alsoaffect spasticity (Bakheit, Maynard, Curnow, Hudson,& Kodapala, 2003; Skold, Levi, & Seiger, 1999;Wood et al., 2005).

Measurement of spasticity is an important part ofpatient care. Members of the clinical team quantifyspasticity with a variety of standardized approacheswhen they assess patient status, select interventions,and evaluate intervention of effectiveness over time.Obtaining accurate, precise measurements is a chal-lenge because the nature of spasticity is elusive andassessment is subjective even when evaluated usingestablished clinical protocols.

This article reviews the biology of spasticity andevaluates approaches to measurement of spasticityused in every day clinical assessment. Specific aims are(a) to review anatomy and pathophysiology of spas-ticity and link them to electromyogram (EMG)findings as a foundation for evaluating clinical mea-surement issues, (b) to identify scales used to measurespasticity in clinical practice and research, (c) to sum-marize information about measurement properties ofscores obtained from examiner and self-report scales inclinical populations, and (d) to judge the state of thescience of spasticity measurement and comment onimplications for practice and future research.

Journal of Neuroscience Nursing104

Questions or comments about this article may be directed toRozina H. Bhimani, PhD RN CNP, at [email protected] is an assistant scientist at Sister Kenny Research Center, andan associate professor at St. Catherine University, Minneapolis,MN.

Lisa C. Anderson, PhD, is a lecturer in the Department ofIntegrative Biology and Physiology, University of Minnesota,Twin Cities, MN.

Susan J. Henly, PhD RN, is a professor in the School of Nursing,University of Minnesota, Twin Cities, MN.

Sarah A. Stoddard, PhD RN CNP, is a research fellow in theSchool of Nursing, University of Michigan, Ann Arbor, MI.

Copyright B 2011American Association of Neuroscience Nurses

DOI: 10.1097/JNN.0b013e31820b5f9f


The Biology of SpasticityDefinitionsThe definition of spasticity is evolving. There is agree-ment that it is characterized by increased stiffnessin skeletal muscles. In a narrow sense, spasticity isobserved as an increased velocity-dependent responsein the tonic stretch reflex (Lance, 1980). The SPASMproject (Burridge et al., 2005) proposed a broaderdefinition. This new definition adds effects of ab-normal sensory input to motor control that result inhyper reflexivity, increased tone over extended periodsof rest, or disorderedmotion occurring during activitiessuch as walking. Other abnormalities of neuromuscu-lar functioning (e.g., clonus, ataxia, akathisia, atheto-sis, hypertonia, rigidity, restless leg syndrome) mayoccur with spasticity but are distinct from it (see Table1 for definitions and Web links for video demonstra-tion; Larsen & Stensaas, 2009; We Move, 2009).

Anatomy and PhysiologyKnowledge about the relevant anatomy and physiol-ogy provides the foundation for theory that guides

measurement of spasticity in research and clinicalpractice (Levin, 2005). The stiff, tight, overstimu-lated muscles and disordered sensory-motor controlin patients with spasticity result from disinhibitionof the tonic stretch reflex by impaired function ofUMNs (Burridge et al., 2005, p. 72; Lance, 1980,p. 485). The interconnected structures and dynamicfunctioning of the nerves and muscles are all in-volved in production and maintenance of spasticity.A discussion of muscle anatomy in normal sensory-motor physiology is essential.

TABLE 1. Terms Describing Neuromuscular Anatomy, Physiology, and Dysfunction

Term Definition

Action potential A large depolarization conducted along the membrane of muscle fiber or nerve axona

Akathisia Subjective feeling of unease and inability to remain still; compulsion to move with little reliefin the absence of anxiety; objectively seen as restlessness and inability to sitb

Ataxia Gross incoordination of muscle movement when standing and walkingc

Athetosis Complex worm-like irregular nonpurposeful movements that involve limbs and facec

Clonus One form of phasic stretch where muscles rapidly contract and relaxc

Contractures Tendon and soft tissue of limbs in a fixed position due to increased muscle tone leading tomuscle shorteningc

Extrafusal fiber Part of muscle fiber that is innervated by alpha motor neurons; they aid in musclecontractiona

Hypertonia Simultaneous co-contraction of agonists and antagonists muscles where non-velocity-dependent resistance to passive movement experienced by the examiner as increasedmuscle tone and rigidityc

Intrafusal fiber Type of muscle fiber that is innervated by gamma motor neurons and sensory Ia neurons;they aid in sensory proprioceptiona

Load An applied force or effort to move a resistancea

Muscle spindle Encapsulated receptor in a skeletal muscle cell that is sensitive to muscle stretcha

Muscle stiffness Condition where muscles feel tight and are not in relaxed or normal supple statec

Proprioceptors Receptors that provide information about posture, muscle tone and movement usually foundin muscles, joints, and tendona

Restless leg syndrome A sensation to move limbs to reduce restlessness sensation when at restb

Rigidity Abnormal muscle stiffness and resistance to movementc

Spasm Hyperexcitability of muscles induced by stimulationc

Stretch reflex Predictable, rapid motor response initiated by a muscle spindle when muscles are stretched

Stretch velocity Direction and speed of the muscle stretch when an external stimulus is applieda

Note. aWidmaier et al. (2007). bAllen and Earley (2001). cWe Move (2009).

A broader definition of spasticity

includes the effects of abnormal

sensory input into motor control in

the evaluation of increased tone and

disordered motion during activity.

Volume 43 & Number 2 & April 2011 105


Motor SystemA motor system consists of higher centers (motorcortex), middle centers (brain stem), and lower cen-ters (spinal cord). Components of a motor systemactivate muscles; they also receive sensory informa-tion about muscle length, muscle tension, and jointposition to activate muscles with the intensity, se-quence, and timing needed for smooth execution of aparticular movement.

The muscles are contractile in nature and are joinedtogether by connective tissue known as fascicle toprovide shape. These muscle structures are made upof muscle fibers that contract whole muscles. Themuscle fibers are innervated via nerve receptors suchas muscle spindles, which control proprioceptors(body position sensors; see Figure 1). As the muscleis stretched (e.g., by a tendon tap or in the course ofdaily activities), the muscle spindle is also stretched,generating a faster frequency of action potentials inthe sensory neuron (see Figure 2a). As the musclecontracts, the stretch on the spindle are relieved,decreasing the frequency of action potentials in thesensory neuron (see Figure 2b). Muscle spindles notonly participate in reflexes but also provide essentialdynamic sensory information about the length andtension of muscles during as voluntary movementoccurs (see Figure 3). For example, as a person per-forms a motor task like reaching for an elevator but-ton, the motor system requires information about thestarting condition of the muscles before a trajectory

of movement can be planned and executed. In thisway, information about body position (propriocep-tion) is supplied by muscle spindles (Nielsen, Crone,& Hultborn, 2007; Widmaier, Raff, & Strang, 2007).

Muscle FibersA skeletal muscle consists of two kinds of musclefibers. First, the bulk of the muscle is made of ex-trafusal fibers (synonymous with skeletal musclefiber), which are innervated by alpha motor neurons;when extrafusal fibers contract, the muscle shortens.The second type of muscle fiber, intrafusal fibers,functions within muscle spindles and is innervatedby gammamotor neurons and sensory Ia neurons (seeFigure 1). When intrafusal fibers contract, there is nodirect contribution to muscle tension or muscle short-ening. Instead, activation of gamma motor neuronsand contraction of intrafusal fibers exert stretch onmuscle spindles making them more sensitive to lengthchanges in the muscle. This unique receptor systempermits the brain to control the length information re-ceived from the muscle spindle (see Figure 2; Widmaieret al., 2007).

Motor NeuronsOn the basis of the location of motor neurons in thebody, a neuron is either a UMN or a lower motorneuron. AUMN arises in the motor cortex of the brainand extends to the spinal cord. An LMN extends fromthe spinal cord to the muscle that it innervates. The

FIGURE 1 Overview of Stretch Reflex Pathways and the Influence of HigherMotor Centers

FIGURE 1

Note. Figure modified from freely available content (copyleft) at: http://thebrain.mcgill.ca/flash/i/i_06/i_06_cl/i_06_cl_mou/i_06_cl_mou.html.



UMNs project to and synapse on lower motor neuronsin the spinal cord.

An LMN also receives sensory information aboutmuscle position and stretch from the muscle spindles.Thus, the lower motor neurons are the Bfinal com-mon pathway[ to coordinate voluntary movementbecause the lower motor neurons receive informationfrom both UMNs and sensory neurons from theperiphery. The lower motor neurons integrate sensoryand motor information for the activation of musclesduring both reflexive and voluntary movement(Widmaier et al., 2007).

The Stretch ReflexThe physiologic role of the stretch reflex is to helpa muscle maintain a steady state. If a person picksup a weight with his or her hand, the load pulls on thebiceps muscle. The stretch depolarizes muscle spin-dles which send action potentials to the spinal cordvia sensory neurons. The sensory neurons synapse onthe alpha motor neurons within the spinal cord thatinnervate the muscle that was stretched. The bicepscontracts and maintains the load. This is the mono-synaptic stretch reflex. Concurrently, the triceps, anantagonist muscle to the biceps, is inhibited byinterneurons activated by the same reflex and thetriceps muscle relaxes. Inhibition of the triceps elim-inates opposition to contraction of the biceps and theload is maintained (Widmaier et al., 2007).

UMN Lesion EffectsAn UMN lesion is a disease or injury that disruptsthe anatomical integrity and/or physiological func-tioning of the UMNs. Multiple sclerosis, spinal cordinjury, and stroke are examples of UMN lesions. UMNdisease, that produces spasticity, affects both reflexiveand voluntary motor function.

Reflexive Function and SpasticityUMN lesions alter the activation of the LMN, pro-ducing a state of net disinhibition of spinal reflexes.In spasticity, the negative feedback system betweenmuscle spindles and alpha motor neurons is disruptedbecause of the UMN lesion, and the abnormal responseof tight muscles is obtained. For example, the UMNlesions decrease the inhibitory drive in the cortico-spinal tract, affecting alpha-motor neuron excitabilityand causing increased muscle contraction. Also, dis-ruption of inhibition of the antagonist muscle or in-creased action potentials in the sensory neurons fromthe muscle spindle can lead to muscle tightness(Nielsen et al., 2007).

Voluntary Movement and SpasticityFor voluntary movement, impulses are generated inthe brain and relayed from UMN to lower motor neu-ron. This impulse coactivates gamma and alpha motorunits. Contraction of muscle relieves the stretch on themuscle spindle such that the unloaded muscle spindle

FIGURE 2 Tonic Stretch Reflex Action Potential Mechanism by Which Message Is Sentto Spinal Cord

FIGURE 2




and its sensory neuron stop sending action potentials(information indicating muscle length) to the centralnervous system. By coactivating the alpha and thegamma motor neurons, the muscle spindles continueto provide information about length to the centralnervous system throughout the entire range of motion(see Figure 3).

Gamma motor neurons can also be activated be-fore alpha motor neurons. Gamma activation exerts astretch on the muscle spindle. Sensory neurons fromthe muscle spindle synapse on alpha motor neurons inthe spinal cord and activate them. Thus, gamma acti-vation and contraction of intrafusal fibers can drivealpha activation and the contraction of entire muscle(Nielsen et al., 2007).

As in reflex responses, UMN lesions decrease theinhibitory drive in the corticospinal tract to producespasticity during voluntary movement and createabnormalities of posture and tone associated withspasticity. The consequences of spasticity during vol-untary movement depend on the severity and source ofthe UMN syndrome. Patients may experience muscle

weakness and lose dexterity; weakness of an agonistmuscle and excessive coactivation of an antagonistmuscle may reduce range of motion or slow the speedat which a patient can complete a task (Bhakta,Cozens, Chamberlain, & Bamford, 2001; Leonardet al., 2006). Spasticity during voluntary movement isgenerated through local activation of muscle spindles,but the propagation and manifestation of spasticityrequires involvement of the central nervous system(Nielsen et al., 2007; Pandyan et al., 2005).

The EMGElectromyography is an electrical summary of neu-romuscular functioning. Spasticity generates charac-teristic EMG tracings, so it can serve as a criterion forassessing decisions about spasticity made on the basisof clinical examination or self-report.

EMG is a common neurophysiologic measure thatuses surface electrodes to monitor electrical activitygenerated from skeletal muscle. Surface measuringelectrodes are placed on the skin over the musclesof interest. EMG activity can then be recorded in

FIGURE 3 Alpha Motor Neuron Activation of Extrafusal Fibers and SimultaneousGamma Motor Neuron Activation of Intrafusal Fibers Maintains the SensoryFunction of the Muscle Spindle System Throughout a Movement

FIGURE 3




response to mechanical or electrical stimuli. For exam-ple, reflex EMGactivity can bemeasured in response toa tendon tap. The time from stimulus to EMG responseis the latency, and the size of the EMG response isthe amplitude. Patients with spasticity exhibit shorterlatency periods and greater amplitudes as comparedwith healthy subjects (see Figure 4).

A passive stretch can also elicit contraction andEMG activity, albeit the velocity of the stretch stim-ulus is critical for detection of spasticity (Voerman,Gregoric, & Hermens, 2005). Surface electrodes moni-tor the EMG response as a clinician or researcherquickly stretches the patient’s muscle. Healthy controlsubjects demonstrate little or no EMG response re-gardless of the stretch stimulus velocity, whereaspatients with spasticity do exhibit EMG activity. Theamplitude of the EMG responses increases in am-plitude with the increased velocity of the stretch(Sorionola, White, Rushton, & Newham, 2009).

The Hoffman reflex (H-reflex) uses electrical stim-ulation of a peripheral nerve to elicit EMG activity. Alow-intensity current from the stimulating electrode tothe mixed peripheral nerve brings sensory neurons tothreshold, such that action potentials are propagatedtoward the spinal cord. As described earlier in thestretch reflex, these sensory neurons synapse on andactivate alpha motor neurons that innervate skeletalmuscle. Action potentials in alpha motor neuronsactivate the muscle and excitation of the muscle canbe detected in the form of an H-reflex wave on theEMG tracing. A high-intensity current stimulus di-rectly brings alpha motor neurons to threshold, andaction potential is propagated toward the muscle;muscle activation subsequent to alpha motor neuronstimulation is detected as anM-wave. As with a tendontap or a stretch stimulus, the latency and the amplitudeH and M waves are recorded. The M-wave typically

has a larger amplitude and shorter latency than does theH-reflex wave (Voerman et al., 2005).

The H-reflex is a variable measure because thelimb and head position of the patient, any othersensory input experience by the patient (e.g., visualstimulation) and stimulus duration, and frequency canall influence EMG activity. However, the H-reflex isa helpful tool for detecting increases in neuronexcitability and decreases in higher center inhibi-tion in patients with spasticity (Burridge et al., 2005;Voerman et al., 2005). For example, the increasedexcitability in patients with spasticity is shown asdecreased latency and increased amplitude in H-reflexwaves as compared with those elicited in healthycontrol subjects. Another EMG measure of motorneuron excitability is the Hmax/Mmax ratio. This ratioalso increases in patients with spasticity and has theadvantage of less variability and greater reproduc-ibility over time in people with UMN disease.

Clinical Scales for Spasticity MeasurementTable 2 lists clinical scales for measurement ofspasticity that are easily administered by an examinerat the bedside or by patients themselves. Examinersuse a prescribed maneuver to elicit limb spasticity andthen grade severity based on response to the exami-nation stimulus. Self-reports of spasticity require thatpatients discriminate among various manifestations oftheir neurological dysfunction, identify spasticity, andrate severity using a scale with standardized anchors.Detailed assessment technique, rater, and scoring cri-teria are described for three examiner scales and twoself-report scales used with adults.

The Ashworth Scale (AS) was developed more than45 years ago to evaluate the clinical efficacy of theantispasticity medication carisoprodol in multiple scle-rosis patients (Ashworth, 1964). The AS is a clinician’ssubjective interpretation of the resistance or catch givenby limbs when a quick stretch is applied during passiverange of motion. During the examination, patients areasked not to assist with any voluntary movement of thelimbs. A grade (score) is assigned by the examiner onthe basis of felt resistance to passive movement. TheAS is an ordinal scale with five levels, ranging from0 (no increase in tone) to 4 (affected part rigid inflexion or extension). It is assumed that resistance topassive movement is exclusively due to spasticity(Pandyan et al., 1999).

The Modified Ashworth Scale (MAS) includesan additional grade termed 1+ (Bohannon & Smith,1987), intended as a mid-classification between aslight increase in tone and a marked increase in tone.The purpose of this addition was to enhance precisionin the clinical measurement of spasticity at lower

FIGURE 4 Schematic EMG TracingsDemonstrating the DecreasedLatency and IncreasedAmplitude of the H WaveAssociated With Spasticity

FIGURE 4



levels. The assessment technique is the same as theAS, but the additional scoring requires that examinerspossess greater awareness of nuances in patient re-sponse to quick stretch during passive movement.

The Modified Modified Ashworth Scale (MMAS)deleted the additional grade of 1+ between 1 and 2from MAS and changed evaluation criteria (Ansariet al., 2006) to enhance the validity and reliability ofthe MAS. Like the original AS, the MMAS has fivecategories scored 0Y4. The scoring criteria for cate-gories 1 and 2 reflect slight or marked increase intone in the AS and MMAS, but the MMAS includesadditional stipulations about catch and resistance,which is similar to the MAS.

AVisual Analog Scale (VAS) is used by patients forself-report of perceived spasticity. A 100-mm straightline marked at 0 (no spasticity) and 100 (worst spas-ticity imaginable) is used. Patients are asked toindicate their level of spasticity by identifying a pointon that straight line that corresponds to their mo-mentary current experience of spasticity.

The Numeric Rating Scale (NRS) has also beenused to report patient perception of spasticity (Farrar,Troxel, Stott, Duncombe, & Jensen, 2008). Com-monly, an NRS of 0 (no spasticity) and 10 (worstpossible spasticity) is used, and patients are asked toindicate their level of spasticity by identifying awhole number between those two anchors.

TABLE 2. Clinical Scales for Measurement of Spasticity

Scale Technique Rater Scoring

AS Resistance or catchgiven by limb withquick PROM

Examiner 0 No increase in tone

1 Slight increase in tone

2 Marked increase in tone, but affected part(s) easily flexed

3 Considerable increase in tone-passive movement difficult

4 Affected part(s) rigid in flexion or extension

MAS Resistance or catchgiven by limb withquick PROM

Examiner 0 No resistance or catch

1 Slight increase in tone (catch and release at end of ROM)

1+ Slight increase in tone, manifested by a catch, followed byminimal resistance throughout the remainder (less than half ofthe ROM)

2 Marked increase in tone through most of ROM but affectedpart(s) easily moved

3 Considerable increase in tone-passive movement difficult


MMAS Resistance or catchgiven by limb withquick PROM

Examiner 0 No increase in muscle tone

1 Slight increase in tone, manifested by a catch and release or byminimal resistance at the end of the ROM when the affectedpart(s) is moved in flexion or extension

2 Marked increase in muscle tone, manifested by a catch in themiddle range and resistance throughout the remainder of theROM, but affected part(s) easily moved

3 Considerable increase in muscle tone-passive movement difficult


NRS Patient self-reportof spasticity

Patient 0 No spasticity

10 Worst possible spasticity

VAS Patient self-reportof spasticity

Patient 0 No spasticity

100 Worst imaginable spasticity

Note. AS = Ashworth Scale; MAS = Modified Ashworth Scale; MMAS = Modified Modified Ashworth Scale; NRS = Numeric RatingScale; VAS = Visual Analog Scale.

s

s



Reliability, Validity, and Responsivenessto ChangeMeasurement in clinical practice is the systematic pro-cess of assigning numbers to represent characteristics ofpeople related to their healthYillness status. Numbers arearrived using carefully prescribed procedures, such asthe techniques described for the clinical and self-reportscales for measurement of spasticity (see Table 2).Reliability (repeatability), validity (accuracy), and re-sponsiveness (to change over time and because of in-tervention) are all desired qualities (Platz et al., 2005).

ReliabilityReliability is concerned with repeatability of scores or,conversely, with error of measurement. Spasticity as-sessment using the AS, MAS, and MMAS involvessubjective classification into distinct, ordered catego-ries by individual examiners, so agreement indices areused to evaluate repeatability. Obtaining a high levelof agreement in spasticity assessment requires thatprescribed technique be followed under the sameconditions for the same patients; for example, the samelevel of stretch applied at the same speed is needed toobtain the same scores on examiner assessments. Thesense of catch or resistance must be the same. Further,examiner differences such as experience or tendency torate high or low must be minimal to avoid systematicdistortion of scores and to obtain high interrater agree-ment. Cohen’s kappa is the proportion of agreementcorrected for chance, which has values ranging fromj1 (complete disagreement) to 0 (chance agreement)to 1 (perfect agreement) (Viera & Garrett, 2005), and

is appropriately used to assess agreement in spasticityscores by different examiners.

Table 3 shows values for kappa for the AS, MAS,and MMAS when used in spinal cord injury, stroke,and traumatic brain injury populations. In these stud-ies, spasticity was evaluated by two to three raters inthe upper and/or lower extremities (including digits)of 15Y50 patients. Raters were PTs, PT students, orMDs. (No nurse raters were included in any of thestudies.) More information was available for theMAS than that for the AS or the MMAS. In general,agreement was low to moderate. On the basis ofavailable findings, it was not possible to determinewhether agreement was higher or lower when testswere used in different patient populations.

ValidityValidity, or accuracy, refers to confidence in makinginferences from test scores. Concurrent validity re-flects relationships between spasticity scores obtainedusing two or more approaches or between spasticityscores and some theoretically associated variable ob-tained on the same occasion. Correlations, group dif-ferences, and regression techniques are used to evaluateconcurrent validity (Brink & Wood, 1997). EMG andpatient self-reports are both used to assess concurrentvalidity of examiner scores obtained using the AS,MAS, and MMAS.

Correspondence With EMGEMG results are often selected as the criterion in val-idity assessments of clinical scales because the latencyand the amplitude of the H-reflex provide directinformation about spasticity under controlled condi-tions. In addition, a stretch-induced reflex response canbe measured simultaneously as a clinical scale such asthe MAS is administered. In general, associationsbetween EMG and clinical examination are mixed.

AS scores were associatedwith EMG findings in fivelower extremity muscle groups in patients with spinalcord injury (Sherwood, Graves, & Priche, 2000).Among patients with stroke, AS scores and EMGwere significantly associated when upper extremitieswere tested (Patrick & Ada, 2006; Starsky, Sangani,McGuire, Logan, & Schmitt, 2005) but not in thelower extremities (Patrick & Ada, 2006). The patternwas opposite when MAS scores were used with strokepatients, where correlations with scores from lowerextremity examinations were statistically significant(Cooper, Musa, van Deursen, & Wiles, 2005; Kimet al., 2005; Lamontagne, Malouin, & Richards, 2001),but those in the upper extremities were not (Leonardet al., 2006; Naghdi et al., 2008). Sample sizes weresmall (range = 13Y31), except for Sherwood et al.(2000) (n = 97). The EMG response was stimulated

TABLE 3. Rater Agreement UsingCohen’s Kappa

Condition

Test

AS MAS MMAS

SCI .21Y.61a .14Y.35b

.20Y.62a

CVA .17c .21c .63d

.51e .72e

.84f .81g

TBI .16Y.42h

.49Y.54h

Note. Ranges are given when multiple sites were tested. Tablecitations list first author’s last name only. AS = Ashworth Scale;MAS = Modified Ashworth Scale; MMAS = Modified ModifiedAshworth Scale.aHaas, Bergstrom, Jamous, and Bennie (1996). bSmith, Jamshdi,and Lo (2002). cAnsari, Naghdi, Moammeri, and Jalaie (2006).dNaghdi et al. (2007). eAnsari, Naghdi, Arab, and Jalaie (2008).fGregson et al. (1999). gAnsari, Naghdi, Younesian, andShayeghan (2008). hMerholz et al. (2005).



electrically at the nerve rather than with passive stretch;this procedure is inconsistent with the definition ofspasticity as a movement-related, velocity-dependentresponse.

Sorionola et al. (2009) studied EMG response ofwrist flexors in the hemiplegic wrist of 10 personswith stroke who met the inclusion criterion of ascreening MAS score of 1 or more. They used aprocedure consistent with definitions of spasticity andthe protocol for clinical exams. Experimenters com-pleted the MAS. They also applied passive manualstretches that were measured objectively using a rigapplied to the forearm and hand. Target velocitiesranged from slow (60 degrees/second) to quick(360 degrees/second). The EMG outcome was thepercentage of maximal voluntary contraction. Theyfound that experimenters had difficulty reachingabsolute target stretch velocities but were able tosuccessfully apply Bslower[ (50 degrees/second) orBfaster[ (240 degrees/second) stretch. Average EMGresponse and variability in response showed acceler-ating increase as velocity increased. The correlationof EMG response and MAS score was significant(r = .72, p G .05) only at the lowest measured veloc-ity. The authors urged caution because of the smallsample size but suggested that the MAS measure themechanical compliance of the muscle rather than thevelocity-dependent response of spasticity. The studyis important because it goes beyond asking simplequestions about agreement between raters or correla-tion between standard EMG and clinical ratings toexplore biological mechanisms that can explain someof the disparate results in other research.

Examiner and Patient CorrespondenceAnother line of evidence important to validity as-sessment is correspondence between patient self-report of spasticity and examiner assessment. Amongpatients participating in the Stockholm Spinal CordInjury Study, Byes[ or Bno[ answers to a query aboutwhether they experienced spasticity was associatedwith examiner identification of any spasticity usingthe MAS; this association held for most joints andmovements tested (Skold et al., 1999). The correla-tion between self-ratings using the VAS with MASscores ranged from .44 to .62 (p G .001) when mea-sured in 45 people with spinal cord injury who weretaking part in an intervention study for spasticity(Skold, 2000). The cross-sectional correlation be-tween VAS and AS scores was .70 among 47 pa-tients with spinal cord injury but was much lower(r = .36) when a question about general level ofspasticity was used for self-report (Lechner, Frotzler,& Eser, 2006). In a subset of 8 of the 47 patients,correlations between VA and AS over time were

weak but significant for three patients and low forthe remaining five patients. Pain and other symp-toms seemed to interfere with self-report of spasticityspecifically. Taken together, findings about corre-spondence between self-report and examiner ratingsare mixed. The longitudinal studies suggest intra-individual variations in the experience of spasticity aswell as between-individual differences in the ability topinpoint spasticity as a specific unpleasant symptom.

ResponsivenessResponsiveness, a change in observed score that occurswith true change over time as disease progresses or afterintervention, has long been a key concern in spasticitymeasurement (Dekker, Dallmeujer, & Lankhorst,2005). The AS (Ashworth, 1964) was originally de-veloped to assess responsiveness to treatment of spas-ticity to treatment with carisoprodol in multiplesclerosis. More recent studies have demonstrated thatAS and MAS scores decrease in response to treatmentwith intrathecal baclofen (Delhaas, Beersen, Redekop,& Klazinga, 2008; Pohl et al., 2003) and botuliniumneurotoxin NT 201 (Caty, Detrembleur, Bleyenheuft,Deltombe, & Lejeune, 2008; Kanovsky et al., 2009).Neuromuscular function changes after insults such asstroke (Mirbagheri, Tsao, & Rymer, 2009) and spinalcord injury (Skold et al., 1999), but we were not ableto locate any studies that formally measured spasticityover time in conjunction with progression of UMNdisease.

State of the ScienceSpasticity is a neuromuscular alteration that occursconcurrently with other manifestations of dysfunc-tion in the presence of UMN disease. The unpleasantexperience of spasticity is associated with complica-tions such as pain, insomnia, difficulties in perfor-mance of activities of daily living, and persistentspasticity contributes to development of permanentcontractures. Thus, accurate and precise measurementof spasticity over time is critical to effective clinicalpractice but continues to be challenging. The evolvingdefinition of spasticity (Burridge et al., 2005; Lance,1980) makes the conceptual basis for measurement amoving target and validity difficult to achieve withboth clinical examinations and patient self-reports.

The underlying biology is complex. There is anexpansive body of research on spasticity but muchof the literature is technical and can be difficult fornonspecialists to interpret. Spasticity results from theinjury and dysfunction of UMNs. Loss of inhibitoryUMNs results in unchecked synaptic transmission inthe spinal cord and exaggerated stretch reflexes. Withthe loss of UMN control, new synaptic connectionsappear in the spinal cord, but the nature of these new



synaptic connections is poorly understood (Nielsenet al., 2007). In other words, although UMN dys-function is essential to spasticity, the effects mayoccur at many points within the nervous system,creating a myriad of effects. The exact pathophysi-ology of spasticity in traumatic brain injury may bedifferent than spasticity in multiple sclerosis, forexample, making it difficult to generalize study con-clusions from one population to another. Spasticity inupper limbs may be different from spasticity in lowerlimbs, also depending on the type of UMN injury.Variation in clinical examinations (AS, MAS, andMMAS) combined with variation in the way muscleswere stimulated for EMGs (PROM, tendon tap,voluntary contraction, and electrical current) creates acomplex literature. Simons and Mense (1998) empha-sized the importance of using electrical stimulation todistinguish the stiff muscles of spasticity from therigidity characteristic of other motor diseases likeParkinson’s disease, thus indicating the unique role ofthe H-reflex in assessing the validity of scores ob-tained from clinical assessment of spasticity.

Spasticity changes in response to different stimuli(e.g., electrical, mechanical tendon tap, velocity ofpassive range of motion, voluntary movement) andconditions (e.g., posture, fatigue, stress, disease pro-gression), making measurement context dependent(Woolacott & Burne, 2006). Thus, true intraindivid-ual variation rather than stability should be antici-pated when spasticity is measured. The intraindividualvariation may explain some reported instances of lowlevels of interrater reliability in clinical scores such asthe AS, MAS, or MMAS, even when well-trained, ex-perienced examiners are used.

Spasticity management is an important patient careissue as it can impact quality of life. Spasticity in-creases the tone of the muscles, and some spasticity ina UMN syndrome may be beneficial if it provides themuscle tone needed to maintain body posture, mo-bility, and ADLs in otherwise weak patients. Whenproblematic spasticity outweighs any beneficialeffects of spasticity, treatments such as use of oralmedications, Botox injections, and intrathecal baclo-fen pumps are indicated. To manage spasticity and toevaluate effectiveness of a treatment plan, cliniciansneed valid and reliable scales that can be easily ad-ministered and interpreted and are sensitive to changein spasticity. Using the AS or its variations is a start,but information can be augmented by systematic,concurrent utilization of patient self-reports.

Both the VAS and the NRS allow incorporation ofpatient perceptions of their own spasticity to comple-ment examiner findings, but little is known aboutpatient ability to sort spasticity from related dysfunc-tions (such as clonus or hypertonia) to produce an

accurate (valid) report. At this time, protocols forVAS and NRS self-report scales for spasticity do notinclude education about exactly what is meant byspasticity. Because patients with spasticity are proneto a wide array of unpleasant disturbing symptoms,this step is essential and should be included to op-timize accuracy whenever self-repot is used. Concor-dance between self-report and examiner scores mayimprove when this step is taken, contributing to thebelievability of patient reports (NIH PROMIS, 2007).Inclusion of patient report and self-assessment scalesin research is of paramount importance. New scalesor modification of existing scales must account forpatients’ lived experiences of spasticity.

SummarySpasticity is a movement-related velocity-dependentneuromuscular dysfunction experienced as tight orstiff muscles that affects quality of life for people whoexperience this phenomenon. Spasticity is manifestedas increased amplitude or shortened latency of EMGactivity in response to stretch or electrical stimula-tion. Systematic examiner assessment (AS,MAS, andMMAS) and patient-reported approaches are usedclinically to measure spasticity. Discrepancy betweenself-and examiner evaluation of spasticity complicatesdecision making about treatment and evaluation oftreatment effectiveness. Understanding of the patho-physiology, definition, and patient conceptions of spas-ticity is the starting point to translate research intobetter clinical practice for spasticity.

AcknowledgmentGrant funding for this review was provided by theSister Kenny Foundation of Sister Kenny Institute,Minneapolis, Minnesota, Grant No. 08-RS-1.

ReferencesAllen, R. P., & Earley, C. J. (2001). Restless leg syndrome: A

review of clinical and pathophysiological features. Journalof Clinical Neurophysiology, 18, 128Y147.

Ansari, N. N., Naghdi, S., Arab, T. K., & Jalaie, S. (2008). Theinterrater and intrarater reliability of the Modified AshworthScale in the assessment of muscle spasticity: Limb andmuscle group effect. NeuroRehabilitation, 23, 231Y237.

Ansari, N. N., Naghdi, S., Moammeri, H., & Jalaie, S. (2006).Ashworth Scales are unreliable for the assessment of musclespasticity. Physiotherapy Theory and Practice, 22, 119Y125.

Ansari, N. N., Naghdi, S., Younesian, P., & Shayeghan, M. (2008).Inter- and intrarater reliability of the Modified ModifiedAshworth Scale in patients with knee extensor poststrokespasticity. Physiotherapy Theory and Practice, 24, 205Y213.

Ashworth, B. (1964). Preliminary trial of carisoprodol inmultiple sclerosis. The Practitioner, 194, 540Y542.

Ashworth, N. L., Satkunam, L. E., & Deforge D. (2006). Treat-ment for spasticity in amyotrophic lateral sclerosis/motor



neuron disease. Cochrane Database of Systematic Review, 2,1469Y1493.

Bakheit, A. M., Maynard, V. A., Curnow, J., Hudson, N., &Kodapala, S. (2003). The relation between Ashworth Scalescores and the excitability of the alpha motor neurones inpatients with post-stroke muscle spasticity. Journal ofNeurology, Neurosurgery, and Psychiatry, 745, 646Y648.

Bhakta, B. B., Cozens, J. A., Chamberlain, M. A., & Bamford, J. M.(2001). Quantifying associated reactions in the paretic armin stroke and their relationship to spasticity. Clinical Re-habilitation, 15, 195Y206.

Bhimani, R. (2008). Intrathecal baclofen therapy in adults andguideline for clinical nursing care. Rehabilitation Nursing, 33,110Y117.

Bohannon, R. W., & Smith, M. B. (1987). Interrater reliabilityof a modified Ashworth Scale of muscle spasticity.Physical Therapy, 67, 206Y207.

Brink, P. J., & Wood, M. J. (1997). Advanced design in nursingresearch (2nd ed.). Newbury Park, CA: Sage.

Burridge, J. H., Wood, D. E., Hermens, H. J., Voerman, G. E.,Johnson, G. R., Wijck, V., et al. (2005). Theoretical andmethodological considerations in the measurement of spas-ticity. Disability and Rehabilitation, 27, 69Y80.

Canadian Institute of Health Research. (2008). The brain fromtop to bottom. Retrieved May 19, 2010, from http://thebrain.mcgill.ca/flash/i/i_06/i_06_cl/i_06_cl_mou/i_06_cl_mou.html

Caty, G. D., Detrembleur, C., Bleyenheuft, C., Deltombe, T., &Lejeune, T. (2008). Effect of simultaneous botulinum toxiinjections into several muscles on impairment, activity,participation, and quality of life among stroke patientspresenting with a stiff knee gait. Stroke, 38, 2803Y2808.

Cooper, A., Musa, I. M., van Deursen, R., & Wiles, C. M. (2005).Electromyography characterization of stretch responses in hemi-paretic stroke patients and their relationship with the modifiedAshworth Scale. Clinical Rehabilitation, 19, 760Y766.

Dekker, J., Dallmeujer, A. J., & Lankhorst, G. J. (2005).Clinimetrics in rehabilitation medicine: Current issues indeveloping and applying measurement issues. Journal ofRehabilitation Medicine, 37, 193Y201.

Delhaas, E. M., Beersen, N., Redekop, W. K., & Klazinga, N. S.(2008). Long-term outcomes of continuous intrathecal bac-lofen infusion for treatment of spasticity: A prospective multi-center follow-up study. Neuromodulation, 11, 227Y236.

Farrar, J. T., Troxel, A. B., Stott, C., Duncombe, P., & Jensen,M. P. (2008). Validity, reliability, and clinical importanceof change in a 0Y10 numeric rating scale measure ofspasticity: A post hoc analysis of a randomized, double-blind, placebo controlled trial. Clinical Therapeutics, 30,974Y985.

Gregson, J. M., Leathley, M., Moore, A. P., Sharma, A. K.,Smith, T. L., & Watkins, C. L. (1999). Reliability of thetone assessment scale and the Modified Ashworth Scale asclinical tools for assessing poststroke spasticity. Archives ofPhysical Medicine and Rehabilitation, 80, 1013Y1016.

Haas, B. M., Bergstrom, E., Jamous, A., & Bennie, A. (1996).The inter rater reliability of the original and of theModified Ashworth Scale for the assessment of spasticityin patients with spinal cord injury. Spinal Cord, 34,560Y564.

Kanovsky, P., Slawek, J., Denes, Z., Platz, T., Sassin, I.,Comes, G., et al. (2009). Efficacy and safety of botulinumneurotoxin NT 201 in poststroke upper limb spasticity.Clinical Neuropharmacology, 32, 259Y265.

Kim, D. Y., Park, C. I., Chon, J. S., Ohn, S. H., Park, T. H., &Bang, I. K. (2005). Biomechanical assessment withelectromyography of post-stroke ankle plantar flexor spas-ticity. Yonsei Medical Journal, 46, 546Y554.

Lamontagne, A., Malouin, F., & Richards, C. L. (2001).Locomotor-specific measure of spasticity of plantarflexormuscles after stroke. Archives of Physical Medicine andRehabilitation, 82, 1696Y1704.

Lance, J. W. (1980). The control of muscle tone, reflexes andmovement: Robert Wartenberg Lecture. Neurology, 30,1303Y1313.

Larsen, P., D., & Stensaas, S., S. (2009). Neurological exam.Retrieved April 18, 2010, from http://library.med.utah.edu/neurologicexam/html/home_exam.html

Lechner, H. E., Frotzler, A., & Eser, P. (2006). Relationshipbetween self- and clinically rated spasticity in spinal cordinjury. Archives of Physical Medicine and Rehabilitation,87, 15Y19.

Leonard, C. T., Gardipee, K. A., Koontz, J. R., Anderson, J. H.,& Wilkins, S. A. (2006). Correlation between impairmentand motor performance during reaching tasks in subjectswith spastic hemiparesis. Journal of Rehabilitation Medi-cine, 38, 243Y249.

Levin, M. F. (2005). On the nature and measurement ofspasticity. Clinical Neurophysiology, 116, 1754Y1755.

Merholz, J., Major, Y., Meissner, D., Sandi-Gahum, S., Koch, R.,& Pohl, N. (2005). The influence of contractures and varia-tion in measurement stretching velocity on the reliability ofthe Modified Ashworth Scale in patients with severe braininjury. Clinical Rehabilitation, 19, 63Y72.

Mirbagheri, M. M., Tsao, C., & Rymer, W. Z. (2009). Naturalhistory of neuromuscular properties after stroke: a longitu-dinal study. Journal of Neurology, Neurosurgery, and Psy-chiatry. 80, 1212Y1217.

Naghdi, S., Ansari, N. N., Mansouri, K., Olyaei, G. R, Asgari,A., & Kazemnejad, A. (2008). The correlation betweenModified Ashworth Scale scores and the new index of alphamotoneurones excitability in post-stroke patients. Electro-myography and Clinical Neurophysiology, 48, 109Y115.

Naghdi, S., Ebrahimi, I., Asgari, A., Olyaei, G. R., Kazemnejad,A., Mansouri, K., et al. (2007). A preliminary study into thecriterion validity of the Modified Modified Ashworth Scaleusing the new measure of the alpha motoneuron excitabilityin spastic hemiplegia. Electromyography and ClinicalNeurophysiology, 47, 187Y192.

Nielsen, H. B., Crone, C., & Hultborn, H. (2007). The spinalpathophysiology of spasticityVFrom a basic science pointof view. Acta Physiologica, 189, 171Y180.

NIH PROMIS (2007). Patient-reported outcomes measure-ment information system. Retrieved January 19, 2007, fromhttp://www.nihpromis.org/default.asp

Nuyens, G., Weerdt, W. D., Ketelaer, P., Feys, H., Wolf, L. D.,Hanston, L., et al. (1994). Inter-rater reliability of theAshworth Scale in multiple sclerosis. Clinical Rehabilita-tion, 8, 286Y292.

Pandyan, A. D., Gregoric, M., Barnes, M. P., Wood, D., Wijck,V., Burridge, J., et al. (2005). Spasticity: Clinical percep-tions, neurological realities, and meaningful measurement.Disability and Rehabilitation, 27, 2Y6.

Pandyan, A. D., Johnson, G. R., Price, C. I., Curles, R. H.,Barnes, M. P., & Rodgers, H. (1999). A review of theproperties and limitations of the Ashworth and ModifiedAshworth Scales as measures of spasticity. Clinical Re-habilitation, 13, 373Y383.



Patrick, E., & Ada, L. (2006). The Tardieu scale differentiatescontracture from spasticity whereas the Ashworth Scale isconfounded by it. Clinical Rehabilitation, 20, 173Y182.

Pettibone, K. A. (1988). Management of spasticity in spinalcord injury: Nursing concerns. Journal of NeuroscienceNursing, 20, 217Y222.

Platz, T., Eickhof, C., Nuyens, G., & Vuadens, P. (2005).Clinical scales for the assessment of spasticity, associatedphenomena, and function: A systematic review of the lit-erature. Disability and Rehabilitation, 27, 7Y18.

Pohl, M., Rockstroh, G., Ruckriem, S., Mehrholz, J., Pause, M.,et al. (2003). Time course of the effect of a bolus dose ofintrathecal baclofen on severe cerebral spasticity. Journal ofNeurology, 250, 1195Y2000.

Sherwood, A. M., Graves, D. E., & Priche, M. M. (2000).Altered motor control and spasticity after spinal cord injury:Subjective and objective assessment. Journal of Rehabilita-tion Research and Development, 37, 41Y51.

Simons, D. G., & Mense, S. (1998). Understanding and mea-surement of muscle tone as related to clinical muscle pain.Pain, 75, 1Y17.

Skold, C. (2000). Spasticity in spinal cord injury: Self and clini-cally rated intrinsic fluctuations and intervention-inducedchanges. Archives of Physical Medicine and Rehabilitation,81, 144Y149.

Skold, C., Levi, R., & Seiger, Å. (1999). Spasticity after traumaticspinal cord injury: Nature, severity and location. Archives ofPhysical Medicine and Rehabilitation, 80, 1548Y1557.

Smith, A. W., Jamshidi, M., & Lo, S. K. (2002). Clinical mea-surement of muscle tone using a velocity-corrected modifiedAshworth Scale. Spinal Cord Injury, 81, 202Y206.

Sorionola, I. O., White, C. M., Rushton, D. M., & NewhamD. H. (2009). Electromyographic response to manual pas-sive stretch of the hemiplegic wrist: Accuracy, reliability,and correlation with clinical spasticity assessment andfunction. Neurorehabilitation and Neural Repair, 23,287Y294.

Starsky, A. J., Sangani, S. G., McGuire, J. R., Logan, B., &Schmitt, B. D. (2005). Reliability of biomechanical spasticitymeasurements at the elbow of people poststroke. Archives ofPhysical Medicine and Rehabilitation, 86, 1648Y1654.

Viera, A. J., & Garrett, J. M. (2005). Understanding interob-server agreement: The kappa statistic. Family Medicine, 37,360Y363.

Voerman, G. E., Gregoric, M., & Hermens, H. J. (2005).Neurophysiological methods for the assessment of spastic-ity: The Hoffmann reflex, the tendon reflex and the stretchreflex. Disability Rehabilitation, 27, 33Y68.

We Move. (2009). Life in motion. Retrieved April 11, 2010,from http://www.life-in-motion.org/

Widmaier, E., Raff, H., & Strang, K. (2007). Vander’s humanphysiology: The mechanisms of body function (11th ed.).NewYork, NY: McGraw-Hill Science.

Wood, D. E., Burridge, J. H., Wijck, V., Mcfadden, C.,Hitchcock, R. A., Pandyan, A. D., et al. (2005). Bio-mechanical approaches applied to the lower and upperlimb for the measurement of spasticity: A systematicreview of the literature. Disability and Rehabilitation, 27,19Y32.

Woolacott, A. J., & Burne, J. A. (2006). The tonic stretchreflex and spastic hypertonia after spinal cord injury.Experimental Brain Research, 174, 386Y396.


journal of neuroscience nursing clinical measurement of...

Documents