One Word, "Spasticity" — Reading the Upper Motor Neuron Syndrome Apart

In rehabilitation wards, few words are used as often yet remain as fuzzy in meaning as "spasticity is severe." Different things get mixed into this single word. There is the resistance caught when stretched quickly, and the posture that remains bent even when resting. There are also times when the muscle moving the joint in the opposite direction when trying to straighten the arm—namely, the antagonist muscle—contracts together, and there is contracture, a joint that has already stiffened.

The primer article "Paralysis and Spasticity — Where the Command Breaks Shapes What Follows" covered the first part of this story. When the upper motor neuron, the nerve cell that sends movement commands from the brain down to the spinal cord, is cut, spinal reflexes that have lost regulation grow larger. That article defined spasticity as a state where the reflex of contracting back when a muscle is stretched becomes excessively large, making the resistance stronger the faster the muscle is stretched. And it went as far as showing that other things besides spasticity are mixed into the resistance that an injured arm puts up.

This article is the next step. It names and separates the mixed elements one by one, and addresses the problem that the commonly used scale does not actually measure speed, which is the defining feature of spasticity. This problem is not a criticism from the outside, but something the field itself is pointing out.

Why We Must Know This — A Single Diagnostic Label Blurs Treatment

The starting point is a 2026 critique. Because this critique is an argument leaning on the authors' clinical experience and collaborative work rather than research results, the sentences below should be read as the authors' perspective.

The authors' main point is this. "Spasticity" has long been used as a general term grouping together distinct positive motor signs arising from central nervous system lesions. As a result, it has become ambiguous to discern the impairments present in the patient one by one (PMID 42424540).

What this critique separates is four things (PMID 42424540).

• Spasticity sensu stricto is spasticity in the narrow sense defined by Lance in 1980, where the faster the muscle is stretched, the greater the tonic stretch reflex (muscle tone) becomes, with hyperactive tendon reflexes. The cause is hyperexcitability of the stretch reflex. This is precisely the definition from the primer.
• Spastic dystonia is a state where muscles remain continuously contracted even without being stretched. An elbow posture remaining bent even at rest is close to this.
• Spastic co-contraction is a state where even antagonist muscles contract together when trying to move, blocking the intended direction of movement on one's own.
• Spasms are contractions where limbs suddenly bend or extend on their own regardless of intent.

The authors continue thus. They view that using spasticity as a broad diagnostic label obscures the specific impairments that actually contribute to the clinical presentation, which can undermine treatment decisions and outcomes (PMID 42424540). The logic is that treatment targets impairments one by one, and when it is obscured which impairment is present, the targeting also becomes blurred.

However, the conclusion of this critique does not end with dismantling the word "spasticity." The authors propose "spasticity syndrome" as a new umbrella term to capture both the clinical presentations of upper motor neuron lesions and the underlying pathology beneath them. This term additionally includes non-neural peripheral changes and patient-reported symptoms on top of the positive signs above (PMID 42424540). Non-neural peripheral changes refer to changes that occurred not in the nerve, but in the muscle and its surrounding tissues themselves. The demand to separate and the proposal to bundle again exist together within one paper. The side that separates aims to view impairments precisely one by one, and the side that bundles aims to include even extra-neural changes and symptoms felt by the patient into a single picture.

The Big Picture — The Two Faces of Upper Motor Neuron Syndrome

Upper motor neuron syndrome is a bundle of symptoms and signs that appear together after an upper motor neuron is injured. This bundle contains both things that have been added and things that have been subtracted.

Positive signs are symptoms that newly appeared after the damage where nothing was before. Spasticity, spastic dystonia, co-contraction, and spasms, which we distinguished earlier, are all positive signs. Negative signs are things that disappeared after the damage that were there before. Muscle weakness, the loss of selective motor control to move only the desired muscle, and fatigue fall into this category.

Which of the two actually blocks movement varies from patient to patient. And the two sides move independently. As we saw in the primer, the contraction of spasticity is generated by spinal reflexes regardless of one's own will, so it does not become a force that can be used to pick up a cup. Therefore, reducing spasticity does not bring back muscle strength. The animal experiment in the next section shows how movement impairment remains even after suppressing reflexes.

Spasticity Depends on Speed — That Is the Definition

The core of Lance’s definition is speed dependence. If stretching slowly meets little resistance and stretching quickly meets catch, that resistance comes from the hyperexcitability of the stretch reflex (PMID 42424540).

Why speed is the core becomes clear when looking at the pathway the reflex travels. Inside the muscle are sensory organs that sense when the muscle is stretched, and the faster the muscle stretches, the stronger the signal these sensory organs send to the spinal cord. Ia fibers are the sensory nerve fibers that send this stretch signal from the muscle up to the spinal cord. In the spinal cord, this signal excites alpha motor neurons, which directly send contraction commands to the muscle. Alpha motor neurons are the cells that the primer called lower motor neurons—namely, nerve cells that carry commands directly from the spinal cord to the muscle. When the regulation that used to press down on this loop from above is cut off, the loop runs too easily, and faster stretches return stronger signals as stronger contractions. This is why resistance that varies with speed is the hallmark of reflex-generated resistance.

There is an animal experiment that directly interfered with this loop. In a mouse model with a hemisected spinal cord, the activity of Ia fibers making up the stretch reflex was pharmacologically suppressed for two weeks (PMID 40610652). The results were measured in two ways.

One is the H-reflex. The H-reflex is a test that artificially triggers the stretch reflex circuit by electrically stimulating a peripheral nerve, and it is normal for the reflex to gradually shrink when stimulation is repeated quickly. This property is called frequency-dependent depression. In this experiment, the index of frequency-dependent depression of the H-reflex related to spasticity partially improved from 0.6 to 0.2. The other is the synapse. Synaptic boutons are the contact points where the endings of Ia fibers touch alpha motor neurons, and the number of Ia fiber synaptic boutons per alpha motor neuron decreased from 4.2 to 2.6 (PMID 40610652). The authors took this to conclude that maladaptive synaptic connections within the spinal cord were "partially" improved and spasticity-related reflex hyper-excitability was alleviated (PMID 40610652).

What the same paper reported alongside is more important. The activity of alpha motor neurons remained higher than that of the control group, and motor impairments in ladder walking and single-pellet reaching tasks were still observed in the drug-administered group (PMID 40610652). The introduction of this paper summarizes the background as follows: spasticity arises from plastic neural changes following injury, and current treatments that block spastic muscle contractions do not promote recovery from motor function impairments (PMID 40610652).

In this experiment, suppressing the reflex did not bring back movement. Although this is from a single mouse model and cannot be directly transposed to humans, it points in the same direction as the background summarized by the paper. Eliminating spasticity and promoting recovery are different tasks.

But the Scales We Use Do Not Measure Speed

This is the most uncomfortable part of this installment. The scales commonly used to measure spasticity in hospitals belong to two families. The Ashworth scale and its modified version (MAS) have an examiner move a limb by hand and grade the resistance felt into several tiers. The Tardieu scale and its modified version stretch a muscle at slow and fast speeds respectively to see where the reaction catches.

The 2026 critique mentioned earlier evaluates these two families like this (PMID 42424540).

• The Tardieu and modified Tardieu scales lack sufficient evidence for validity and reliability.
• The Ashworth and modified Ashworth scales do not measure the speed-dependent nature of spasticity, nor do they reliably tell us what mechanism creates the resistance.

The authors write that relying on such tools only deepens the ambiguity we saw earlier (PMID 42424540). The Tardieu scale, created to measure speed, lacks evidence, while the Ashworth scale, the most widely used, does not measure speed.

To sum it up: We are measuring a speed-dependent phenomenon with a tool that does not control speed, and reporting the result under the single word "spasticity." A high MAS score means the resistance is large, but that score alone cannot tell whether that resistance came from a reflex or somewhere else.

Still, the MAS Grade Moves Alongside Some Functional Tests

Criticizing a scale and saying the scale contains no information at all are two different things.

The only two tests where performance differed according to the spasticity level were tests that required changing direction: the figure-of-eight walking test, and the four-square step test, which involves stepping forward, backward, and sideways across four squares on the floor in a set sequence. The group with the most severe spasticity performed worse on these two tests than the group without spasticity. On the other hand, the Activities-Specific Balance Confidence (ABC) scale, where people rate their own confidence in performing daily activities without falling, and the five-times sit-to-stand test, which involves standing up and sitting down from a chair five times in a row, showed no differences across spasticity levels (PMID 41156227). This is a cross-sectional observational study of 54 chronic stroke patients, which divided spasticity of the muscles that bend the ankle toward the sole (plantarflexors) into three groups (0 / 1–1+ / 2 or more) using the MAS.

The authors concluded that tests requiring direction changes (the figure-of-eight walk and four-square step tests) captured differences in spasticity levels, whereas balance confidence and the five-times sit-to-stand test did not (PMID 41156227). The starting point of this study was the idea that spasticity might restrict movement when turning or shifting the feet in multiple directions, and these findings match that direction. This means that even though the MAS grade is rough, there are tasks where performance is split depending on that grade.

This study also calculated the test-retest reliability, measurement error, and minimal detectable change of each test according to spasticity level (PMID 41156227). These are concepts covered in "How Do We Measure Recovery — An Introduction to Rehabilitation Outcome Measures". Even when measuring the same patient twice with the same tool, the values fluctuate slightly; the magnitude of that fluctuation is the standard error of measurement (SEM), and the minimum amount of change required to say that a change has exceeded the error is the minimal detectable change (MDC). The authors suggested choosing outcome measures based on spasticity levels and using SEM and MDC as benchmark values when interpreting change (PMID 41156227). Because these benchmark values also came from chronic stroke patients, when reading a particular patient's change, one must look together at which patient group and spasticity level those benchmark values were derived from.

Contracture Is Not Spasticity

If spasticity persists, joints can also become stiff. The primer defined contracture as a state where the muscles and tissues around a joint shorten and stiffen, reducing the range of motion and leaving the joint rigid. Therefore, stiffness itself is not a hyperactive reflex, but a change in tissue.

Conflating the two leads to misguided treatments. This is because treatments targeting reflexes (e.g., botulinum toxin) and approaches targeting shortened tissue length are different. The previous critique also notes that the clinical presentation of upper motor neuron damage includes not only neurological components but also non-neurological peripheral changes together, and specifies them as components of the "spasticity syndrome" (PMID 42424540).

The clue that separates the two is speed again. Resistance created by reflexes increases the faster you stretch it, but shortened and stiffened tissue resists just the same even when stretched slowly. Therefore, clinicians stretch it twice while varying the speed, and if it fails to straighten equally whether stretched quickly or slowly, they suspect tissue issues rather than reflexes. However, this diagnostic trick does not come directly from the cited evidence, but is an explanation relying on clinical convention.

Attempts to Step Outside the Scale

Dissatisfaction with hand-felt scales is giving birth to new measurement methods.

Mechanomyography (MMG) is a method that uses sensors to measure the tiny vibrations generated when a muscle contracts. There have been attempts to align this signal with the range of the MAS to create a quantitative scale. In a pilot study looking at 34 limbs of 22 volunteers, polynomial regression fitting yielded an R² = 0.987, and the mean difference from the MAS was 0.001G (PMID 39204970). R² is a value indicating how well the fitted curve explains the data, and this value means it matched almost perfectly. The authors concluded that this algorithm and equipment made it possible to evaluate spasticity quantitatively and semi-automatically (PMID 39204970). However, because this index was aligned to fit the limitations and range of the MAS, it is closer to meaning that it reproduced the scale well rather than stepping outside of it.

The pendulum test involves lifting and releasing the lower leg in a sitting position to let it swing, and observing the spasticity of the knee muscles through that swing. There are methods that assist this test with pose estimation from smartphone video to evaluate knee spasticity after stroke (PMID 41302184), and attempts to measure knee extensor spasticity after spinal cord injury using body-worn portable devices (PMID 38594696). Both studies verified their new measured values by comparing them with MAS scores (PMID 41302184, PMID 38594696).

A scoping review addressing spasticity evaluation methods for children with cerebral palsy analyzed 30 papers out of 1,971. It compared established clinical scales like the MAS and Tardieu with new technologies such as real-time sonoelastography, which looks at tissue stiffness using ultrasound, and inertial sensors, which measure acceleration and rotation of movement (PMID 38992701). It pointed out that traditional methods are convenient and widely used but often lack reliability and objectivity, and that dynamic evaluations of range of motion or stiffness tools have the potential to provide better information. The conclusion was that there is a need for more reliable methods to measure the degree of spasticity more accurately (PMID 38992701). The new tools have not yet replaced the old scales, but are at the stage of finding a path to measure what the old scales could not.

How It Actually Appears — When Looking at Treatments

This section is a description of what studies have observed, and is not content recommending any specific treatment to any specific patient.

Botulinum toxin is an injection given directly into the muscle, as seen in the primer, which makes the injected muscle contract less for a while. It sits at the center of spasticity treatment, and a wide body of related literature has accumulated. However, as seen below, the results are not consistent in a single direction. The point to watch from the perspective of this article is what each study set as its target.

An everyday clinical observational study examining the timing of treatment for upper limb spasticity after stroke found that MAS scores decreased in all groups, but found no significant difference between the early group and the late group (adjusted mean difference 0.15, p=0.546). The authors wrote, "We found no strong evidence for a significant benefit of early initiation" (PMID 32184753). The primary outcome of this study was the MAS score. On the other hand, there is a secondary analysis showing that patients who received injections earlier after stroke also recovered motor function faster, suggesting the possibility that early injections help motor recovery (PMID 41295873). This analysis set muscle tone as the primary outcome and looked at muscle strength, sensorimotor recovery, and overall disability together. The two studies showed mixed results, and the indices they focused on also differed from each other.

Combinations and comparisons with other treatments also follow. There are a systematic review and meta-analysis covering concurrent use with constraint-induced movement therapy, which involves tying down the less paralyzed arm and making the paralyzed arm used intensively (PMID 34646693), a comparison of efficacy with extracorporeal shock wave therapy, which sends shock waves from outside the body into the muscle (PMID 35412036), and a literature review dealing with injection dosage (PMID 36355984).

In multiple sclerosis spasticity, there is a study that used goal-setting itself as an evaluation axis (PMID 36136520). Goal Attainment Scaling (GAS) is a method where the patient and physician set goals together and score how well those goals were achieved a few weeks later. The background for this study choosing this method lay in the lack of evidence that function improves after botulinum injections in multiple sclerosis spasticity (PMID 36136520). It is a method of looking at results through the goals the patient aims to achieve, rather than spasticity scores.

Read in connection with the preceding sections, it comes down to this: a drop in the spasticity score and an improvement in the patient's function are separate events, and which side was looked at is told by the index the paper chose.

Common Misconceptions

"Spasticity is muscle weakness." The two directions are opposite. Spasticity is a positive sign added after damage, and muscle weakness is a negative sign that is missing (PMID 42424540).

"Reducing spasticity brings recovery." Current treatments that block spastic spasticity contractions are reported not to promote recovery from motor dysfunction (PMID 40610652).

"MAS measures spasticity." MAS does not measure the velocity-dependent nature of spasticity (PMID 42424540). What MAS measures is the magnitude of resistance felt by the hand.

"The Tardieu scale is more accurate." The Tardieu scale stretches while changing the velocity, but the same critical review notes that there is a lack of evidence for the validity and reliability of the Tardieu and modified Tardieu scales (PMID 42424540).

"Anything stiff is spasticity." Contracture is a tissue change, and the same review also separates this into non-neural peripheral changes (PMID 42424540). In clinical practice, stretching twice while changing the velocity is a practical starting point.

To Go Deeper

When reading a paper on spasticity, checking two things first makes what the results say clear. One is which sign is being dealt with. You look at whether it is spasticity in the strict sense, co-contraction, or dystonia. The other is what was used to measure it. When looking at the results, you separate whether the spasticity score changed or the functional indicator changed. The tier distinction from "Why Hold Several Rulers Against One Person — An Introduction to the ICF Framework" applies as is. Spasticity scores measure the body function tier, while walking tests measure the activity tier.

If you have read this series up to here, you will read the results tables of rehabilitation papers in this order: how recovery is possible ("How the Central Nervous System Recovers — A Primer on Neuroplasticity"), what was used to measure it ("How Do We Measure Recovery — An Introduction to Rehabilitation Outcome Measures"), how deep that improvement is ("How Deep Is the Recovery — Between Compensation and True Recovery"), and what precisely that symptom was. Where spasticity starts inside the body can be seen in pictures by returning to the primer piece "Paralysis and Spasticity — Where the Command Breaks Shapes What Follows".

Source Papers

• Is Spasticity a Syndrome? A Historical Perspective on Spasticity Definitions and Descriptions. PMID 42424540 (2026)
• Effect of continuous Ia fibre activity suppression on hyperreflexia-related spasticity and maladaptive synaptic connections in the spinal cord after injury. PMID 40610652 (2025)
• Differences in Functional Performance and Minimal Detectable Change According to Levels of Ankle Plantar Flexor Spasticity in Patients with Chronic Stroke. PMID 41156227 (2025)
• Mechanomyography-Based Metric Scale for Spasticity: A Pilot Descriptive Observational Study. PMID 39204970 (2024)
• Methods of muscle spasticity assessment in children with cerebral palsy: a scoping review. PMID 38992701 (2024)
• A Feasible Method for Evaluating Post-Stroke Knee Spasticity: Pose-Estimation-Assisted Pendulum Test. PMID 41302184 (2025)
• A portable system to measure knee extensor spasticity after spinal cord injury. PMID 38594696 (2024)
• Effectiveness of AbobotulinumtoxinA in Post-stroke Upper Limb Spasticity in Relation to Timing of Treatment. PMID 32184753 (2020)
• Early Botulinum Toxin Type A Injection May Improve Motor Recovery in Patients with Post-Stroke Spasticity: A Secondary Analysis from a Longitudinal Cohort Study. PMID 41295873 (2025)
• Constraint-Induced Movement Therapy Combined With Botulinum Toxin for Post-stroke Spasticity: A Systematic Review and Meta-Analysis. PMID 34646693 (2021)
• A systematic review on extracorporeal shock wave therapy and botulinum toxin for spasticity treatment: a comparison on efficacy. PMID 35412036 (2022)
• AbobotulinumtoxinA Doses in Upper and Lower Limb Spasticity: A Systematic Literature Review. PMID 36355984 (2022)
• Goal-Setting in Multiple Sclerosis-Related Spasticity Treated with Botulinum Toxin: The GASEPTOX Study. PMID 36136520 (2022)