Hoffmann reflex parameters as neurophysiological biomarkers for quantifying spasticity: A systematic review

INTRODUCTION

Spasticity is a clinical phenomenon characterized by a velocity-dependent increase in muscle tone accompanied by exaggerated tendon reflexes due to hyperexcitability of the stretch reflex arc.^[1] It is a common manifestation after central nervous system (CNS) injury, particularly following stroke, spinal cord injury (SCI), and cerebral palsy (CP). Spasticity critically impairs mobility, self-care, and independence, often resulting in additional complications such as contractures, pain, and secondary musculoskeletal changes. Understanding and quantifying spasticity is therefore fundamental for developing targeted rehabilitation strategies and improving quality of life for affected patients.^[2]

Stroke is one of the leading causes of adult disability worldwide, with an estimated global incidence of 13.7 million annually.^[3] It can occur due to ischemia or hemorrhage, leading to focal or global interruption of cerebral blood flow and subsequent neuronal damage. Spasticity develops as part of the upper motor neuron syndrome, with studies reporting post-stroke spasticity prevalence ranging from 17% to 43% depending on the phase after stroke, lesion location, and assessment tools employed.^[4] Spasticity most commonly affects the upper limb flexors and lower limb extensors, resulting in functional limitation, abnormal posture, and considerable morbidity. Approximately one in five stroke survivors experiences clinically meaningful spasticity within 3–6 months post-event, with higher rates observed in those with severe hemiparesis or earlier functional impairment.^[4] SCI is another devastating neurologic event with an annual incidence rate estimated at 15–40 cases per million worldwide.^[5] Approximately 70% of individuals with SCI develop spasticity, which may manifest as increased tone, spasms, or exaggerated reflexes.^[6] After SCI, loss of descending inhibitory input from the brain results in disinhibition of spinal circuitry, increased motoneuron pool excitability, and changes in synaptic transmission, all contributing to spasticity.^[7] Both incomplete and complete injuries can give rise to spasticity, with presentations ranging from regional involvement to generalized patterns depending on injury characteristics and chronicity. Up to 78% of individuals with chronic SCI develop spasticity, which can complicate care, increase pain, and impair wheelchair mobility, transfers, and self-care.^[6] CP is the most common cause of childhood physical disability, affecting roughly 2–3/1,000 live births globally.^[8] The spastic form of CP accounts for 75–80% of cases, resulting from static, non-progressive insult to the developing brain that occurs pre-, peri-, or postnatally.^[9] Spasticity is especially prominent in the pyramidal-type CP, gravely impacting gross and fine motor functions, gait, and daily living activities. The presence and severity of spasticity in CP are influenced by lesion site, extent of brain damage, age, and coexisting motor and sensory impairments.^[10] The majority of children with CP have spastic forms, with moderate-to-severe spasticity present in nearly half, contributing to contracture risk and functional loss.^[9] The most widely used scale for clinical estimation of spasticity is Modified Ashworth Scale (MAS). It is simple and quick but limited by subjectivity and poor sensitivity for small changes.^[11] Tardieu Scale incorporates both the amplitude and velocity of muscle reaction, improving reliability but is somewhat examiner-dependent.^[12] Penn Spasm Frequency Scale and Spasticity Frequency Score is a self-reported tools for SCI, more appropriate for quantifying spasms than true spasticity.^[13] Electrophysiological tests include electromyography readings, F-wave analysis, and stretch reflex testing. These provide objective data but require specialized equipment and expertise.^[14]

While clinical rating scales remain standard, objective quantification methods such as neurophysiological testing have gained attention for their sensitivity to underlying changes in CNS excitability and reflex modulation.

The Hoffmann’s reflex (H-reflex) is an electrically-evoked monosynaptic reflex analogous to the spinal stretch reflex, typically recorded in lower limb muscles (soleus, gastrocnemius) after stimulation of the corresponding peripheral nerves.^[15] It is elicited by submaximal stimulation of IA sensory afferents in a peripheral nerve, resulting in synaptic excitation of alpha motoneurons and measurable efferent activity in the target muscle. The H-reflex is favored for assessment because it bypasses muscle spindles, directly exploring central synaptic and motoneuron pool properties.^[16]

The key parameters recorded in H-reflex studies are latency which is time from stimulus artifact to initial response. Prolonged latency may indicate slowed conduction or synaptic delay.^[17] Amplitude corresponds to the number of activated motor units. Increased amplitude reflects heightened motoneuron pool excitability, while reduced amplitude indicates inhibition or pre-synaptic depression.^[18] The Hmax/Mmax ratio is ratio of maximal H-reflex to maximal direct muscle response (M-wave) amplitude, used to normalize findings and account for peripheral variability.^[19] The recovery cycle assesses excitability modulation after conditioning pulses, revealing properties such as post-activation depression and presynaptic inhibition.^[20] Slope and threshold ratios provide insight into the gradation and recruitment of motoneurons.^[21] These parameters can be modulated by voluntary activity, body position, pharmacologic agents, and various disease states. Post-stroke spasticity is frequently accompanied by elevated H-reflex amplitude and Hmax/Mmax ratios in affected limbs, consistent with increased motoneuron excitability.^[22] Latency may be prolonged depending on lesion chronicity and severity. Downstream changes, such as recovery cycle and post-activation depression, are often blunted, indicating impaired inhibitory processes.^[23] Studies have correlated changes in H-reflex slope or amplitude with clinical improvement post-intervention (e.g., botulinum neurotoxin-A, physiotherapy).^[24] Post-activation depression of spinal reflexes is increasingly recognized as a key physiological marker linking segmental circuitry to functional recovery in people with spasticity.^[25,26] Interventions such as botulinum neurotoxin-A injections and task-specific locomotor training can induce plastic changes in spinal transmission, including reduced reflex amplitude and partial normalization of post-activation depression, indicating that spinal circuitry remains modifiable even in chronic stages.^[27–29] Given the multiple H-reflex parameters available for interpreting spasticity, selecting a specific one for prospective studies remains challenging due to the lack of a dedicated review addressing this question. Hence the purpose of this review is to identify the commonly used H-reflex parameter to quantify spasticity in the three most commonly treated neurological conditions.

METHODS

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.^[30]

Literature search

A comprehensive literature search was conducted in PubMed (2016–2025). All searches included available records until May 2025 using the following keywords and Medical Subject Headings (MeSH terms): (1) Stroke OR Cerebrovascular Accident OR Hemiplegia, (2) H-reflex, (3) Spasticity, (4) SCI, and (5) CP. The searches in all databases were combination as follows: 2 AND 3, 2 and 3 and 1, 2 and 3 and 4, 2 and 3 and 5. No limits were applied to the search. Figure 1 illustrates the PRISMA flow diagram of study selection process.^[30]

Close

Figure 1:

PRISMA 2020 flow diagram showing the study selection process for the systematic review on H-reflex parameters as neurophysiological biomarkers for quantifying spasticity.

Export to PPT

DISCUSSION

The studies analyzed commonly assessed spasticity using H-reflex parameters such as amplitude, Hmax/Mmax ratio, and latency, often coupled with clinical scales like the MAS. Table 2 illustrates details of the studies included in the review among the included articles, the most widely reported and overlapping parameters were H-reflex amplitude and the Hmax/Mmax ratio, measured in at least eight of the studies involving patients with conditions such as stroke, SCI, and CP. Latency measurements, though less frequently reported, were also present in several studies and linked to spasticity as a marker of altered reflex excitability and alpha motor neuron function.

Table 3 illustrates studies that used H-reflex parameters as an outcome measure. The Hmax/Mmax ratio is widely considered an index of peripheral reflex excitability which is typically elevated in patients with spasticity in conditions such as stroke, SCI, and CP compared to healthy controls, reflecting increased motor neuron pool excitability, while normal values are observed in rigid but not spastic conditions (Kshirsagar et al., 2022); (Qin et al., 2021) (Sangari et al., 2019) (Agarwal et al., 2023).^[42-45] The Hmax/Mmax ratio measures reflex excitability by comparing the maximum H-reflex amplitude (Hmax) to the maximum direct muscle response (Mmax). An increased ratio signifies heightened spinal motor neuron excitability, which is associated with spasticity (Qin et al., 2021). [43] It serves as a quantitative electrophysiological marker that can complement clinical spasticity scales such as the Modified Tardieu Scale (MTS) and the MAS (Kshirsagar et al., 2022). [42] The ratio shows a moderate correlation with spasticity severity but sometimes modest relationships with functional outcomes, reflecting that spasticity is multifactorial and not fully captured by this measure alone (Pisano et al., 2000).^[46] Unlike healthy individuals, the stroke group shows less modulation of the ratio between postures (e.g., prone vs. standing), indicating altered spinal reflex adaptation after stroke (Qin et al., 2021).^[43] The Hmax/Mmax ratio correlates moderately with clinical spasticity measures such as MTS in chronic stroke, validating its relevance as a biomarker for post-stroke spasticity assessment (Kshirsagar et al., 2022). ^[42]

In the study, Shen et al. (2022)^[38] in post-stroke patients stated that the Hmax/Mmax ratio was significantly higher on the spastic side compared to the unaffected side in both upper and lower limbs.^[33] This elevation was associated with severe spasticity and higher MAS scores. Over time, the ratio showed early elevation with partial normalization in chronic stages, tracking motoneuron pool excitability after stroke. Saito et al. (2021) stated in stroke, the Hmax/ Mmax ratio increased on the affected side compared to both unaffected side and healthy controls, regardless of body position.^[41] This increase was interpreted as disrupted adaptive spinal control and heightened excitability underlying spasticity in post stroke patients. Fawzi et al. (2023)^[31] found that following botulinum toxin injection for post-stroke spasticity, there was a marked reduction in Hmax/Mmax ratio, H-reflex amplitude, and increased latency, interpreted as decreased motoneuron excitability and improved spasticity outcomes.^[31]

Hmax/Mmax ratios do not consistently distinguish spastic from non-spastic SCI patients due to complex underlying mechanisms such as disrupted descending inhibition and altered afferent inputs (Sangari and Perez 2019).^[44] The ratio correlates with spasticity severity and varies by CP subtype and age, reflecting developmental changes in spinal excitability (Agarwal et al., 2023).^[45] Increased Hmax/Mmax ratio supports that disrupted descending inputs in CP cause secondary spinal cord excitability changes contributing to muscle tone abnormalities (Agarwal et al., 2023).^[45] Studies consistently note that H-reflex amplitude and Hmax/Mmax ratio correlate with clinical measures of spasticity, but this correlation may be moderate because these parameters reflect neurophysiological excitability rather than subjective tone alone.

Latency, another common electrophysiological parameter, is often shortened in spastic individuals, particularly children with CP or adults’ post-stroke, suggesting increased spinal excitability and reduced inhibitory control. H-reflex latency represents the time for conduction through the monosynaptic reflex arc and reflects the speed of neural transmission and excitability of spinal motor neurons. A decreased H latency often indicates heightened excitability contributing to spasticity (Tekgül et al., 2013).^[47] In children with spasticity, H latencies were significantly shorter compared to controls, indicating increased excitability, which correlated partially with MAS scores (Tekgül et al., 2013). H latency has clinical utility in quantifying alpha motor neuron excitability changes in mild spasticity, though its sensitivity declines with increased muscle contracture and severity (Tekgül et al., 2013).^[47]

In our study, an interventional study done by Fawzi et al. (2023)^[31] in post-stroke patients reported a significant increase in H-reflex latency after botulinum neurotoxin-A injection. The interpretation was that increased latency reflected decreased motoneuron excitability and improved spasticity outcomes after the intervention. In a randomized controlled trial by Chantanachai et al. (2025)^[35] of telerehabilitation with tDCS and exercise for SCI, H-reflex latency was assessed as a neuroplasticity biomarker. Improved amplitude and modulation were found post-intervention, although specific latency change details were not highlighted.^[41] Saito et al. (2021)^[41] reported that in post stroke spasticity, H-reflex latency was measured across body positions.^[36] Modulation of latency reflected disrupted adaptive spinal control mechanisms, important for understanding functional impairment, and motor recovery in stroke patients. Chen et al. (2016)^[37] stated that in chronic SCI, soleus H-reflex latency was measured but found to show no significant differences between SCI and controls, indicating that while other reflex measures may be impaired or altered, conduction speed as measured by latency remained stable.^[32] Interpretations from these articles indicate that increased H-reflex latency post-intervention (such as botulinum toxin in stroke) correlates with clinical improvement, reflecting slowed reflex conduction and reduced neural excitability. In stroke, altered modulation of latency is linked to adaptive control deficits, while in SCI, the parameter may remain unchanged, reflecting preserved conduction despite changes in excitability or spasticity severity. Thus, H-reflex latency serves as a complementary neurophysiological marker in the assessment and monitoring of spasticity, with its significance most notable in interventions and recovery contexts in post-stroke populations.

Post-stroke spasticity is associated with significantly decreased H-reflex latency compared to healthy subjects, reflecting enhanced spinal motor neuron excitability (Fawzi et al., 2023).^[31] Longer chronicity of stroke can affect reflex latency patterns, but early post-stroke stages show pronounced latency shortening (Phadke et al., 2012).^[48] In SCI, H-reflex latency is initially preserved or slightly altered but becomes shorter over time as spasticity develops, indicating increased excitability of spinal circuits (Little and Halar, 1985); (Hiersemenzel et al., 2000).^[49,50] Reduced latency is associated with decreased presynaptic inhibition and spinal reorganization explaining spasticity onset post-injury (Schindler-Ivens and Shields, 2000).^[51] Changes in latency post-SCI reflect pathological plasticity and may vary with injury phase and therapeutic interventions.

Children with spastic CP show significantly shortened H-reflex latency relative to healthy children, reflecting enhanced motor neuron excitability due to corticospinal tract damage (Tekgül et al., 2013).^[47] Latency shortening correlates moderately with clinical spasticity measures and varies with age and CP subtype (Agarwal et al., 2023).^[45] Electrophysiological studies indicate altered conduction velocities and excitability in CP children, reinforcing H latency as a useful marker for spasticity assessment (Tekgül et al., 2013).^[47]

The interpretation of H-reflex changes across the included studies consistently emphasizes its role as a marker of spinal motoneuron excitability and inhibitory control in spasticity. In CP, Chuang et al. (2022)^[36] reported that ankle continuous passive motion (CPM) increased presynaptic inhibition, reflected by decreased Hmax/Mmax ratio and amplitude, which aligned with reduced spasticity on the MAS.^[31] In SCI, Chen et al. (2016)^[37] found that although reflex-induced torque and stretch reflexes were increased, soleus H-reflex amplitude showed no significant difference compared to controls, indicating limitations in H-reflex for detecting neural contributions to stiffness in some cases.^[32] In post-stroke populations, Shen et al. (2022)^[38] observed higher Hmax/Mmax and developmental slope ratios on the spastic side, correlating with more severe spasticity, while Son et al. (2019)^[39] demonstrated prolonged excitatory postsynaptic potentials (EPSP) alongside increased H-reflex amplitude, indicating impaired reflex inhibition. Akbas et al. (2020)^[40] showed that rectus femoris (RF) H-reflex amplitude was elevated in post-stroke stiff-knee gait, correlating with reduced knee flexion and impaired motor control. Saito et al. (2021)^[41] noted reduced modulation of soleus H-reflex amplitude across body positions in post-stroke spasticity, further highlighting disrupted spinal adaptive control.^[41]

Botulinum neurotoxin-A treatment studies (Fawzi et al., 2023)^[31] documented decreased H-reflex amplitude, latency increases, and reduced Hmax/Mmax ratio post-injection, indicating decreased motoneuron excitability concurrent with clinical spasticity improvement. Operant conditioning in SCI (Thompson et al., 2019)^[33] led to down-conditioning of H-reflex amplitude during gait, which correlated with improved walking, supporting neuroplastic rehabilitation potential. In chronic incomplete SCI, H-reflex cyclic modulation across gait phases was diminished, reflecting impaired spinal control mechanisms (Thompson et al., 2019).^[33] Studies examining posterior root-muscle (PRM) reflexes found blunted inhibition in SCI, with unchanged H-reflex recovery cycles, suggesting differential modulation of spinal circuits (Hofstoetter et al., 2019).^[34] Recent telerehabilitation combining tDCS and exercise demonstrated enhanced H-reflex amplitude and modulation in SCI, consistent with neuroplasticity and functional gains (Chantanachai et al., 2025).^[35]

Overall, these studies collectively confirm that altered H-reflex parameters such as amplitude, Hmax/Mmax ratio, latency, and modulation are sensitive indicators of altered spinal excitability and inhibitory dysfunction in spasticity across neurological conditions. While context-specific variations exist, the H-reflex remains a valuable neurophysiological biomarker for both assessment and monitoring of spasticity and its response to interventions in systematic clinical evaluation.

The articles assessed various outcome measures besides H-reflex parameters, which often involved clinical, biomechanical, neurophysiological, and functional assessments with variable co-relevance to H-reflex findings. MAS was used in multiple studies including Chuang et al. (2022)^[36] on CP, Shen et al. (2022)^[38] on stroke, and Fawzi et al. (2023)^[31] on post-stroke spasticity.^[31,38] MAS scores generally showed some correlation with H-reflex measures such as Hmax/Mmax ratio, reflecting clinical spasticity severity, though correlation was sometimes moderate or inconsistent, indicating MAS as a subjective complement to objective H-reflex data. Plantar flexor stiffness and reflex-induced torque was studied by Chen et al. (2016)^[37] in SCI, these biomechanical measures correlated with increased reflex excitability and spasticity but did not always correlate well with H-reflex amplitudes, highlighting the multifactorial nature of spasticity involving neural and mechanical components.

Electrophysiological measures such as M-wave amplitude, EPSP decay time constant (Son et al., 2019), F-waves, motor-evoked potentials, and PRM reflexes were studied in stroke, SCI.^[39] Some studies showed parallel changes or discrepancies between H-reflex and these measures, indicating complementary but distinct aspects of motor neuron excitability and reflex pathway integrity.

Functional and gait parameters in Akbas et al. (2020) correlated RF H-reflex amplitude with knee flexion angles in post-stroke stiff-knee gait, demonstrating that elevated H-reflex relates to impaired motor control.^[40] Thompson et al. (2019)^[33] showed H-reflex modulation linked to gait phase in SCI, supporting functional relevance of H-reflex in locomotion. Alongside spasticity scales, measures such as the medical research council scale (strength) and quality-of-life assessments were included in some interventional studies, with H-reflex used as an objective biomarker for neuroplasticity and treatment response.

The studies included in this systematic review utilized a variety of interventions aimed at reducing spasticity, each demonstrating effects on H-reflex parameters as markers of neurophysiological changes. In CP, daily ankle CPM for 4 weeks was shown to decrease the Hmax/Mmax ratio and H-reflex amplitude, indicating increased presynaptic inhibition alongside clinical reduction in spasticity (Chuang et al., 2022).^[36] In SCI, although passive muscle stiffness and reflex torque were elevated, interventions were not applied directly in the cited cross-sectional study; however, neuromuscular electrical stimulation has been explored in other contexts for modulating reflex excitability. Post-stroke interventions included botulinum neurotoxin-A injections, which significantly reduced H-reflex amplitude, increased latency, and lowered Hmax/Mmax ratios, reflecting decreased motoneuron excitability concurrent with spasticity improvement (Fawzi et al., 2023).^[31] Operant conditioning approaches to down-regulate H-reflex amplitude during gait were employed in SCI patients, demonstrating sustained reduction in reflex excitability coupled with improved locomotor function (Thompson et al., 2019).^[33] In addition, tele-rehabilitation involving tDCS combined with exercise enhanced H-reflex amplitude modulation and clinical outcomes in SCI (Chantanachai et al., 2025).^[35] Studies measuring H-reflex modulation across body positions in post-stroke spasticity elucidated impaired adaptive spinal control mechanisms that may benefit from targeted positioning and rehabilitation strategies (Saito et al., 2021).^[41] Collectively, these interventions underscore the capacity of various therapeutic modalities – from physical to neuropharmacological and neuromodulatory – to induce quantifiable changes in H-reflex parameters, thereby providing objective evidence of altered spinal excitability and improved motor control. Such neurophysiological markers complement clinical scales and functional assessments, enriching the evaluation and optimization of spasticity treatments in neurological rehabilitation.

Physiotherapy interventions play a critical role in reducing spasticity and improving function in individuals with neurological disorders. Recent advancements in the literature highlight a variety of effective techniques used to manage spasticity. Extracorporeal shock wave therapy has emerged as a promising non-invasive modality, with studies demonstrating significant reduction in muscle tone and improved range of motion in conditions such as CP and post-stroke spasticity Khan et al. (2025); Zhang et al. (2022).^[52,53] In addition, invasive physiotherapy techniques such as acupuncture, electroacupuncture, and dry needling have shown efficacy in spasticity reduction, particularly when combined with conventional physiotherapy (JavierOrmazábal et al., 2022).^[54] Neuromodulation approaches including repetitive transcranial magnetic stimulation and electrical stimulation have also been reported to decrease spasticity and enhance motor performance (Di Ludovico et al., 2023).^[55] Furthermore, robotic-assisted gait training, hydrotherapy, and balance rehabilitation represent important components of physiotherapy regimens targeting spasticity by improving muscle strength, coordination, and functional mobility (Di Ludovico et al., 2023).^[55] Recent controlled trials underscore the safety and efficacy of selective nerve transfer surgeries combined with physiotherapy in addressing severe spasticity when conservative methods are insufficient Khan et al. (2025).^[52] Overall, a multimodal approach integrating these advanced physiotherapy interventions is endorsed for optimized spasticity management tailored to individual patient needs.

CONCLUSION

The reviewed literature shows that H-reflex amplitude and Hmax/Mmax ratio are the most consistently used parameters and correlate moderately with clinical spasticity scales such as the MAS and MTS. Elevated Hmax/Mmax ratio reliably reflects increased spinal motoneuron excitability observed in spasticity across stroke, SCI, and CP. However, its correlation with functional outcomes may be modest due to the multifactorial nature of spasticity. H-reflex latency is often shortened in spastic individuals, indicating increased spinal excitability. Post-intervention changes in latency (such as after botulinum neurotoxin-A) are indicators of treatment response and neural excitability reduction. A variety of physiotherapy and neuromodulatory interventions demonstrate measurable impacts on H-reflex parameters, supporting their use as objective outcome measures in both research and clinical rehabilitation. Overall, the H-reflex serves as a valuable tool in both research and clinical practice for the assessment, monitoring, and neurophysiological understanding of spasticity, aiding tailored rehabilitation and improved patient outcomes in neurological rehabilitation