Composite Spasticity Index (CSI)

Evidence Reviewed as of before: 25-04-2012
Author(s)*: Annabel McDermott, OT
Editor(s): Nicol Korner-Bitensky, PhD OT
Expert Reviewer: Mindy Levin, PhD PT

Purpose

The Composite Spasticity Index (CSI) provides a clinical measure of spasticity that can be used with patients with hemiparesis following stroke.

In-Depth Review

Purpose of the measure

The Composite Spasticity Index (CSI) s a measure of upper and lower extremity spasticity that is suitable for use with patients with hemiparesis following stroke.

The CSI measures the phasic stretch reflex by assessing the tendon jerk and clonus, and the tonic stretch reflex with assessment of resistance to passive movement of the limb (Calota & Levin, 2009).

Available versions

A Composite Spasticity Scale was originally proposed by Chan (1986). The CSI was subsequently developed by Levin & Hui-Chan (1992).

Features of the measure

Items:

The CSI is comprised of 3 items:

  1. Tendon jerk
  2. Resistance to passive flexion
  3. Clonus

Description of tasks:

The first item (tendon jerk) measures hyper-reflexia by applying taps to the biceps, triceps, patellar or Achilles tendon, depending on the location of the spasticity being measured. The therapist should apply enough force to evoke a ‘maximal’ reflex jerk. This can be compared with the maximum tendon reflex elicited on the unaffected side (Chan, 1986).

The second item (resistance to passive stretch) measures hyperactivity of the tonic stretch reflex by assessing the amount of resistance felt by the examiner when the passive muscle is stretched (Chan, 1986). This item incorporates the Modified Ashworth Scale 5-point ordinal scale, which is doubly-weighted (0 to 8), and measures the magnitude of the resistance to stretch at moderate speed (> 100 degrees per second) (Levin & Hui-Chan, 1992).

The third item (clonus) assesses the number of beats of clonus at the wrist (upper limb) or ankle (lower limb) when the hand or foot is rapidly flexed by the examiner (Chan, 1986).

What to consider before beginning:

Chan (1986) advised that the presence of contractures and/or a clasp-knife reflex should be noted when assessing resistance to passive stretch. Contractures can be detected by comparing the passive ROM on the affected side with that on the unaffected side. A clasp-knife reflex can be detected by the presence of a ‘melting away’ of the resistance with increased amount of passive stretch.

The tonic stretch reflex is influenced by the initial muscle length and the velocity of stretch; accordingly, the examiner should apply passive stretch to the affected muscles at the same speed (around 100 deg/sec) and with the limb positioned in the same posture to attain similar initial muscle length on readministration (Chan, 1986).

The CSI measures the magnitude of the stretch response rather than the threshold (Calota & Levin, 2009).

Scoring and Score Interpretation:

The first and second items of the CSI (tendon jerk and resistance to passive stretch) are scored on 5-point scales. The third item (amount and duration of ankle clonus) is scored on a 4-point scale (Levin & Hui-Chan, 1992).

Tendon jerks 0 No response
1 Normal response
2 Mildly hyperactive response
3 Moderately hyperactive response
4 Maximally hyperactive response
Resistance to passive stretch* 0 No resistance (hypotonic)
2 Normal resistance
4 Mildly increased resistance
6 Moderately increased resistance
8 Maximally increased resistance
Clonus 1 Clonus not elicited
2 1-3 beats of clonus elicited
3 3-10 beats of clonus elicited
4 Sustained clonus

* Note that resistance to passive stretch is double-weighted

The second item (resistance to passive stretch) is doubly weighted, as it most closely represents hypertonus.

The patient’s total spasticity score is calculated as the sum of scores (tendon jerk + resistance to passive stretch + clonus). The composite score is used to determine the severity of spasticity, which is defined by the following levels according to clinical experience:

0 – 9 = mild spasticity
10 – 12 = moderate spasticity
13 – 16 = severe spasticity

Time:

Time taken to perform the CSI has not been reported.

Training requirements:

There are no specific training requirements for the CSI. However, administrators should practice stretching the joint smoothly (without quick accelerations and decelerations) at a given speed to improve reproducibility of the results.

Subscales:

N/A

Equipment:

A reflex hammer is required for administration of the CSI.

Alternative Forms of the Composite Spasticity Index (CSI)

Jobin and Levin (2000) created the Modified Composite Spasticity Index to measure phasic (tendon jerks) and tonic (resistance to manual stretch) stretch reflex excitability in children with cerebral palsy. This measure does not include assessment of clonus. Accordingly, the composite score is the sum of the tendon jerk + resistance to passive stretch. Composite scores range from 1 to 12, where scores of 1 – 4 indicate mild spasticity; scores of 5 – 9 indicate moderate spasticity; and a score of 10 or greater indicates severe spasticity (Jobin & Levin, 2000; Scholtes et al., 2006).

Client suitability

Can be used with:

  • Patients with stroke (Levin et al. 2000)
  • Patients with spinal cord injury (Goulet et al., 1996)
  • Children with cerebral palsy (Jobin & Levin, 2000)

Should not be used with:

  • Not specified.

In what languages is the measure available?

English

Summary

What does the tool measure? Spasticity
What types of clients can the tool be used for? Patients with hemiparesis following stroke as well as cerebral palsy and spinal cord injury.
Is this a screening or assessment tool? Assessment
Time to administer Not reported
Versions
  • Composite Spasticity Index
  • Modified Composite Spasticity Index
Other Languages N/A
Measurement Properties
Reliability Internal consistency:
Two studies examined the internal consistency of the CSI: one study reported an excellent correlation between the total spasticity score and resistance to passive stretch and clonus; one study reported adequate internal consistency using Cronbach’s coefficient alpha.

Test-retest:
One study reported excellent test-retest reliability of the CSI; one study reported excellent test-retest reliability of the Composite Spasticity Scale.

Intra-rater:
No studies have reported on the intra-rater reliability of the CSI.

Inter-rater:
No studies have reported on the inter-rater reliability of the CSI.

Validity Content:
One study reported that the three items of the CSI are relevant to assess spasticity in stroke.

Criterion:
Concurrent:
Three studies have reported on the concurrent validity of the CSI: two studies reported excellent negative relationships between clinical spasticity and stretch reflex measures; one study reported no consistent significant correlations between the CSI and physiological measures of reflex function.

Predictive:
No studies have reported on the predictive validity of the CSI.

Construct:
Convergent/Discriminant:
Two studies examined the convergent validity of the Composite Spasticity Index and the Fugl-Meyer Assessment (FMA). One study found an excellent negative correlation using Spearman rank-order statistics, whereas the other study found no significant relationships between the two measures using Pearson product moment correlation statistics.
One study examined the convergent validity of the Composite Spasticity Scale with other measures using Spearman correlation coefficients and reported an adequate negative correlation between ankle plantar flexor spasticity of the affected leg (measured by the CSS) and stance time of the affected leg only. No significant correlations were found between the CSS and the Timed Up and Go Test (TUG), 6 Minute Walk Test (6MWT) or other gait parameters (velocity, cadence and step length and stance time of the unaffected leg).

Known Group:
One study examined known groups validity of the Composite Spasticity Scale and reported a significant difference in spasticity between patients with chronic stroke and healthy elderly subjects.

Floor/Ceiling Effects No studies have examined the floor or ceiling effects of the CSI in clients with stroke.
Sensitivity/ Specificity No studies have reported on the sensitivity or specificity of the CSI.
Does the tool detect change in patients? No studies have formally examined the responsiveness of the CSI.
Acceptability The CSI is commonly used in the assessment of spasticity in adults. It is an acceptable measure of spasticity in stroke.
Feasibility The time to administer the CSI has not been reported. Ease of administration and scoring is dependent on experience of the assessor.
How to obtain the tool? Click here to see the CSI.

Psychometric Properties

Overview

A literature search was conducted to identify all relevant publications on the psychometric properties of the Composite Spasticity Index (CSI). While this assessment can be used with other populations, this module addresses the psychometric properties of the measure specifically when used with patients with stroke. Four studies reported on the psychometric properties of the CSI; one study that reported on the psychometric properties of the Composite Spasticity Scale (CSS) has also been included in this review.

Floor/Ceiling Effects

No studies have reported on the floor/ceiling effects of the CSI when used with patients with stroke. However, it is unlikely that this tool would be appropriate for floor and ceiling evaluation.

Reliability

Internal Consistency:
Levin & Hui-Chan (1993) administered the CSI over three trials with 10 patients with stroke and hemiparesis and 7 healthy subjects. Data was calculated using Pearson correlation coefficients and was provided for two of three trials. The total spasticity score had an excellent correlation with resistance to passive stretch (r = 0.94, 0.89, p<0.05) and clonus (r = 0.85, 0.86, p<0.05) but not Achilles tendon reflex. There was a significant correlation between clonus and resistance to passive stretch on one occasion (r = 0.82, p<0.05), indicating that severity of clonus varied with that of the resistance to passive stretch. There were no other significant associations between items.

Nadeau et al. (1998) measured the internal consistency of the CSI with 19 adults with acute to chronic stroke. Internal consistency of the CSI, as measured using Cronbach’s coefficient alpha, was adequate (α = 0.7023).

Test-retest:
Chan (1986) advised that the examiner instructs the patient to relax when measuring resistance to passive stretch, or alternatively, for the examiner to exert the same effort from test to test. Further, the examiner should apply passive stretch to the affected muscles at the same speed and with the limb positioned in the same posture to attain similar initial muscle length on readministration.

Ng & Hui-Chan (2005) assessed the 1-week test-retest reliability of the CSS in 10 healthy elderly adults and 10 patients with chronic stroke and reported excellent test-retest reliability using Intraclass Correlation Coefficient (ICC) (ICC = 0.97 and 0.80 for affected and unaffected limbs, respectively, of patients with stroke; ICC = 0.80 for both left and right limbs of health adults).

Levin & Hui-Chan (1993) assessed the test-retest reliability of the CSI in 10 patients with spastic hemiparesis and stroke over three testing occasions and reported excellent test-retest reliability (r = 0.87) using Intraclass Correlation Coefficients (ICC).

Intra-rater:
No studies have examined the intra-rater reliability of the CSI.

Inter-rater:
No studies have examined the inter-rater reliability of the CSI.

Validity

Content :

Nadeau et al. (1998) used principle component analysis to determine the extent to which each item of the CSI contributed to the concept of spasticity. In a sample with 19 patients with acute to chronic stroke, the three items of the CSI combined into 1 factor that explained 67.9% of the total variance (eigenvalue = 2.035). This indicates that the 3 items of the CSI are relevant to assess spasticity in stroke.

Criterion :

Concurrent :
Levin et al. (2000) examined the correlation between clinical spasticity and the stretch reflex threshold range for elbow flexor and extensor muscles in 12 patients with hemiparetic stroke, using Spearman rank order statistics. An excellent negative correlation was found between clinical spasticity and the static thresholds of the flexors (r=-0.75, p<0.01) and extensors (r=-0.69, p<0.05). This indicates that patients with more severe spasticity from hemiparetic stroke have more severe limitations in the range in which flexor and extensor stretch reflex thresholds can be regulated. There was also an excellent negative correlation with the velocity dependencies of the stretch reflex thresholds of the flexor muscle (r=-0.70, p<0.05), the limitations in the ranges of reciprocal innervation (r=-0.84, p<0.01) and active joint motion (r=-0.67, p<0.05).

Levin & Feldman (1994) compared clinical spasticity with values of static stretch reflex thresholds in elbow flexors, using Pearson Product Moment correlation statistics. Clinical spasticity was measured using the CSI and stretch reflex thresholds were measured by static angular threshold and the slope of the relationship between the dynamic threshold angles and velocity. A significant excellent negative relationship was seen between clinical spasticity (CSI) and static stretch reflex excitability alone (r = -0.652, p<0.05). These results indicate validity for quantification of spasticity in elbow flexors (Jobin & Levin, 2000).

Levin & Hui-Chan (1993) examined the correlation between clinical spasticity and altered reflex function in 10 patients with stroke and spastic hemiparesis. Physiological measurement of reflex functions included: H-reflex latency; maximal amplitude of the H-reflex as a percentage of the maximal M response (H/M ratio); amount of inhibition of the H-reflex during vibration as a percentage of the control H-reflex amplitude (Hvib/Hctl); and excitability (latency, duration, magnitude) of the soleus stretch reflex. Measures were taken over three testing days, although statistical data were provided for only two days. There were no consistent significant correlations between the CSI and physiological measures of reflex function.

Predictive:
No studies have reported on the predictive validity of the CSI when used with patients with stroke.

Construct:

Convergent/Discriminant:
Ng & Hui-Chan (2005) investigated the associations between the CSS, Timed Up and Go Test (TUG), 6 Minute Walk Test (6MWT) and other gait parameters (velocity, cadence and step length and stance time of the affected and unaffected legs), using Spearman correlation coefficients. An

Levin et al. (2000) found an excellent negative correlation between clinical spasticity and the Fugl-Meyer Assessment (FMA) (r=-0.86, p<0.005), using Spearman rank order statistics.

Levin & Feldman (1994) compared clinical spasticity (measured using the CSI) and the FMA, using Pearson product moment correlation statistics. No significant relationships were seen between clinical spasticity and clinical motor function.

Known groups:
Ng & Hui-Chan (2005) investigated the known-group validity of the CSS in 10 healthy elderly subjects and 11 patients with chronic stroke. Ankle plantar flexor spasticity, measured by the CSS, was significantly higher in the affected leg of patients with stroke than the mean scores of both legs in healthy elderly subjects (p<0.001). Within the group of patients with stroke, ankle plantar flexor spasticity was significantly higher in the affected leg than the unaffected leg (p<0.001).

Responsiveness

The responsiveness of the CSI has not been formally assessed, however it has been used in several studies as a measure of within-group and between-group differences.

Levin & Hui-Chan (1992) examined the effect of repetitive low-threshold afferent stimulation (TENS) on clinical spasticity in 13 patients with spastic hemiparesis following stroke. Patients received stimulation to the peroneal nerve for 60 minutes, 5 days a week for 3 weeks. Patients were assessed using the CSI at baseline and after 2 and 3 weeks of stimulation. The CSI detected a change in spasticity among patients who received TENS (at 2 weeks treatment only), and a significant difference in spasticity between patients who received TENS and the control group who received placebo stimulation (p<0.05).

Sensitivity/ Specificity:
No studies have reported on the sensitivity or the specificity of the CSI when used with patients with stroke.

References

  • Calota, A. & Levin, M.F. (2009). Tonic stretch reflex threshold as a measure of spasticity: Implications for clinical practice. Topics in Stroke Rehabilitation, 16(3): 177-188.
  • Chan, C.W.Y. (1986). Motor and sensory deficits following a stroke: Relevance to a comprehensive evaluation. Physiotherapy Canada, 38, 29-34.
  • Goulet, C., Arsenault, A.B., Bourbonnais, D., Laramee, M.T., & Lepage, Y. (1996). Effects of transcutaneous nerve stimulation on H-reflex and spinal spasticity. Scandinavian Journal of Rehabilitation Medicine, 28, 169-76.
  • Goulet, C., Arsenault, A.B., Bourbonnais, D., & Levin, M.F. (1994). Topographical effects of transcutaneous electrical nerve stimulation on the H-reflex of the triceps surae muscles. Journal of Electromyography & Kinesiology, 4, 116-125.
  • Jobin, A. & Levin, M. (2000). Regulation of stretch reflex threshold in elbow flexors in children with cerebral palsy: a new measure of spasticity. Developmental Medicine & Child Neurology, 42, 531-540.
  • Levin, M.F. & Feldman, A.G. (1994). The role of stretch reflex threshold regulation in normal and impaired motor control. Brain Research, 657, 23-30.
  • Levin, M.F. & Hui-Chan, C.W.Y. (1992). Relief of hemiparetic spasticity by TENS is associated with improvement in reflex and voluntary motor functions. Electroencephalography and Clinical Neurophysiology, 85, 131-142.
  • Levin, M.F., & Hui-Chan, C. (1993). Are H and stretch reflexes in hemiparesis reproducible and correlated with spasticity? Journal of Neurology, 240, 63-71.
  • Levin, M.F., Selles, R.W., Verheul, M.H.G., & Meijer, O.G. (2000). Deficits in the coordination of agonist and antagonist muscles in stroke patients: Implications for normal motor control. Brain Research, 853, 352-69.
  • Nadeau, S., Arsenault, A.G., Gravel, D., Lepage, Y., & Bourbonnais, D. (1998). Analysis of the spasticity index used in adults with a stroke. Canadian Journal of Rehabilitation, 11, 219-20.
  • Ng, S.S., & Hui-Chan, C.W. (2005). The Timed Up & Go Test: Its reliability and association with lower-limb impairments and locomotor capacities in people with chronic stroke. Archives of Physical Medicine and Rehabilitation, 86, 1641-7.
  • Scholtes, V.A.B., Becher, J.G., Beelen, A., & Lankhorst, G.J. (2006). Clinical assessment of spasticity in children with cerebral palsy: a critical review of available instruments. Developmental Medicine and Child Neurology, 48, 64-73.

See the measure

How to obtain the Composite Spasticity Index?

Click here for a copy of the Composite Spasticity Index.

Table of contents
We need your feedback