Cervical Myelopathy

Etiology

Myelopathy refers to the neurological deficits that result from spinal cord compression, whereas stenosis describes the anatomical narrowing of a normally patent canal. In the cervical spine, individuals with a congenitally narrow spinal canal are at a higher risk of developing myelopathy. Progressive stenosis or the occurrence of a cervical disc herniation is more likely to produce symptomatic cord compression at levels where the canal diameter is already reduced.

Degenerative changes are typically observed at the C5-C6 or C6-C7 levels, where increased segmental mobility places greater mechanical stress on the cervical spine. The infolding of the ligamentum flavum, spondylolisthesis, osteophytes, and facet hypertrophy also impact canal narrowing. Myelopathy develops in approximately 100% of patients with a space of less than 6 mm available for the cord, and in none of the patients with a space of more than 14 mm.[7] Age is the strongest predictor of perioperative morbidity and unfavorable neurologic recovery.

DCM can be caused by any combination of the following conditions: cervical disc herniations, cervical spondylosis, cervical degenerative disc disease, congenital stenosis, and ossification of the posterior longitudinal ligament (OPLL).[8][9][10][11] OPLL was first described by Key in 1838  and is characterized by replacement of the ligamentous tissue with ectopic bone.[12] OPLL has been recognized as one of the main causes of cervical myelopathy in adults. This pathological ossification progressively narrows the cervical spinal canal, leading to chronic cord compression. OPLL involves an abnormal endochondral ossification process within the posterior longitudinal ligament, characterized by the proliferation of fibrocartilage-like cells that differentiate into osteoblasts, along with associated neovascularization.

Notably, genetic predispositions also contribute, as results from recent genome-wide association studies have identified multiple susceptibility loci implicating dysregulated bone formation pathways in OPLL.[13] Clinically, OPLL may remain asymptomatic until ossification is advanced, at which point it often manifests as DCM. The severity of neurological deficits generally correlates with the extent of spinal cord compression, although dynamic factors (eg, repetitive cervical motion or segmental instability) can further exacerbate cord injury beyond what static stenosis alone would produce.[12]

Epidemiology

DCM arises from cervical spinal cord compression (SCC), a radiographic finding that becomes increasingly common with age. A pooled analysis of 19 magnetic resonance studies reported an SCC prevalence of 24% in neurologically healthy adults.[14] Age strongly influences these figures. Among individuals younger than 60, the prevalence of SCC was 7%, whereas in those aged 60 and older, it climbed to 35%—only a minority of persons with SCC exhibit clinical myelopathy. The same meta-analysis estimated that 2.3% of otherwise healthy adults already fulfilled clinical criteria for DCM, suggesting that roughly 1 in 10 individuals with SCC has symptomatic disease at any given time. The incidence of DCM not only rises steeply with age, but is also consistently higher in men. Results from a study in Taiwan showed a predominance in men across every age band, escalating to almost double the incidence rate in men aged 70 years or older, compared with women.[15]

Administrative data confirm that diagnosed DCM represents only a fraction of its likely population burden. Analysis of United Kingdon Hospital Episode Statistics between 2012 and 2019 revealed a mean treated prevalence of 0.19%, peaking at 0.42% in the 50–54 age group and declining thereafter.[16] When the same population was modelled using published SCC conversion rates, the expected prevalence averaged 2.2% and exceeded 4% beyond 79 years of age. The widening gap between predicted and observed figures suggests substantial underdiagnosis, particularly among adults aged 60 and above.

Pathophysiology

DCM results from a sequence of mechanical, vascular, cellular, and genetic events that culminate in spinal cord injury and neurological decline.[17] Structural narrowing of the canal delivers the primary insult. Fixed, or static, contributors include congenital canal stenosis, intervertebral-disk degeneration, osteophytic ridging, OPLL, and ligamentum flavum hypertrophy. Superimposed dynamic forces, such as repetitive flexion, extension, or rotational motion, produce vertebral translation and ligamentous buckling that further constrict the canal and strain the axons. Compression of the anterior spinal artery, penetrating arterioles, and venous plexus impairs perfusion to the spinal cord. The ensuing hypoxia contributes to neuronal loss within the gray horns and demyelination of lateral and posterior columns. Mechanical and ischemic insults trigger inflammation, excitotoxicity, apoptosis, and disruption of the blood-spinal cord barrier.

Genetic factors influence degenerative spine disease. Candidate variants in collagen, growth factor, and vitamin-D receptor genes modulate susceptibility to canal degeneration and cord vulnerability. A recent transcriptomic-imputation study extended these observations, identifying 16 tissue-specific genes associated with DCM. Upregulated spinal cord genes, such as HES6 and PI16, are associated with neurogenesis and neuroinflammation, whereas downregulated genes, such as TTC12 and CDK5, influence nociception and neuronal survival. These findings suggest that inherited regulation of nociceptive, proliferative, and immune pathways influences the onset and trajectory of disease.[18]

Cervical spondylotic myelopathy commonly involves compression of the lateral corticospinal tracts, affecting voluntary motor control, and the spinocerebellar tracts, impairing proprioception. These combined deficits contribute to the hallmark features of cervical myelopathy, including gait disturbances and clumsy upper extremity function. Additional commonly involved spinal cord regions include the spinothalamic tracts, which are responsible for contralateral pain and temperature sensation; the posterior columns, which are responsible for ipsilateral position and vibration sense; and the dorsal nerve root, which is responsible for dermatomal sensation.[19][20][21]

A scoping review of 61 studies on the symptoms in DCM found unspecified paresthesias in 86% of cases, hand numbness in 82%, and hand paresthesias in 79%.[22] Gait disturbance follows closely, occurring in approximately 72% of cases, and sometimes precedes arm complaints, especially when patients describe their legs as “heavy” or “dragging.”[23] Neck or shoulder pain is present in approximately half of the patients, whereas axial or lower extremity pain is less common. Autonomic symptoms appear late; bladder urgency or retention affects 38% of individuals, and bowel dysfunction affects 23%.

History and Physical

Begin the history with broad, open-ended questions to identify atypical presentations of DCM. Targeted symptom inquiry should be organized into 4 key domains: upper extremity function, lower extremity function, bladder or bowel concerns, changes in sexual function, and pain. Specifically, ask about hand clumsiness and difficulty with fine motor tasks such as buttoning, using utensils, or writing, as these may indicate cervical cord compromise even in the absence of pain. Inquire directly about gait instability, difficulty navigating stairs, falls, and reliance on handrails. Assess for urinary urgency, frequency, or episodes of incontinence.

When pain is present, clarify its onset, duration, radiation, and any potential triggering events. A presentation of radiating arm pain suggests concomitant radiculopathy.[24] Radiating pain, as the primary issue, typically has a more predictable surgical outcome than nonspecific neck pain, which is often related to muscle fatigue and strain. Additionally, ask questions that may suggest alternative diagnoses, such as headaches, vision changes, swallowing or voice issues, facial numbness, hearing problems, ringing in the ears, memory changes, or seizures.[25]

Quantify baseline disability with the modified Japanese Orthopaedic Association (mJOA) score. Scores of 15–17 denote mild, 12–14 moderate, and 11 or less denotes severe impairment. These empirically derived thresholds correlate with gait speed, quality-of-life metrics, and the number of neurological findings.[26] Recording the mJOA at each visit allows objective monitoring. Perform a complete neurological exam, assessing tone, strength, sensation, reflexes, coordination, and gait. Additional pertinent assessments include:

• Gait: Examine for balance issues or spasticity. Observe tandem gait because many patients cannot maintain a narrow base even before weakness develops.
• Reflex examinations for hyperreflexia: Examine the biceps, triceps, brachioradialis, patellar, and Achilles tendons.
• Lhermitte sign: If neck flexion causes an electric shock-like feeling, a positive exam is denoted.
• Look for myelopathic hand wasting (thenar eminence).
• Hoffman sign: The examiner holds the patient’s hand in a relaxed position and flicks the distal phalanx of the middle finger downward. A positive response is indicated by involuntary flexion of the thumb and/or index finger, suggesting dysfunction of the corticospinal tract.
• Trömner sign: The examiner taps the volar surface of the distal phalanx of the slightly flexed middle finger. A positive response is reflexive flexion of the thumb and/or index finger, serving as an alternative to the Hoffmann sign for assessing corticospinal tract involvement.
• Crossed radial reflex: This includes the contraction of both the biceps and wrist following a biceps reflex test.
• Inverted radial reflex: This includes the contraction of both wrist extension and finger flexion following tapping the brachioradialis.
• Finger escape sign: The patient is unable to hold the ulnar digits in extension and adduction.
• Grip test: Failure to repetitively make a tight fist denotes a positive exam.
• The Tromner sign and general hyperreflexia carry the greatest sensitivity, while Babinski response, clonus, and an inverted supinator sign provide high specificity.[27] 

Results of a study showed that 3 of 5 findings—Hoffmann, inverted supinator, Babinski, gait deviation, age older than 45—increase posttest probability of myelopathy above 90%.[28] Brachial plexus variants can alter the expected distributions of symptoms; remember that deficits do not always correspond to classic dermatomes.[29]

Grading Systems

The Nurick grading system is a simple scale used to quantify the functional severity of cervical myelopathy by assessing gait disturbance and its impact on employment. Grade 0 indicates root symptoms only or normal findings. Grade 1 denotes signs of cord compression on examination (such as hyperreflexia or pathological reflexes) but with a normal gait. Grade 2 reflects mild gait difficulty while the patient remains fully employed. Grade 3 describes gait difficulty severe enough to prevent gainful employment, although the patient can still walk unassisted. Grade 4 means the patient is unable to walk without assistance (eg, using a cane, walker, or human support). Grade 5 signifies that the patient is wheelchair or bed-bound.

The mJOA score is the most widely used grading system for DCM. The mJOA quantifies the severity of cervical myelopathy by assigning points across 4 domains—upper extremity motor function, lower extremity motor function, upper extremity sensory function, and sphincter (bladder) function—for a total possible score of 18. Lower scores indicate greater dysfunction. Below is a breakdown of each domain and its corresponding point values:

• Upper extremity motor0 = No hand movement1 = Moves hands but cannot eat with a spoon2 = Eats with a spoon but cannot button a shirt3 = Buttons shirt with great difficulty4 = Buttons shirt with slight difficulty5 = Normal function

• Lower extremity motor0 = No leg movement or sensation1 = Sensation only, no voluntary movement2 = Moves legs but cannot walk3 = Walks on flat floor with a cane or crutch4 = Walks stairs using a handrail5 = Walks stairs unaided but with significant instability6 = Walks on level ground with mild instability7 = Normal function

• Upper extremity sensory0 = No hand sensation1 = Severe loss or constant pain2 = Mild loss3 = Normal sensation

• Sphincter0 = Cannot initiate micturition1 = Marked difficulty, requires straining or manual aid2 = Mild to moderate difficulty, hesitancy or slow stream3 = Normal micturition 

Total mJOA score

By summing the points from all 4 domains (5 + 7 + 3 + 3 = 18 maximum), the mJOA score provides an overall measure of functional impairment.

• 0–11 = Severe myelopathy
• 12–14 = Moderate myelopathy
• 15–17 = Mild myelopathy
• 18 = Normal (no functional impairment) [30]

Evaluation

Evaluating patients with suspected cervical myelopathy requires a structured approach that integrates clinical findings, radiographic imaging, and, occasionally, laboratory studies and electrophysiological testing to confirm the diagnosis, assess disease severity, and exclude alternative etiologies. Laboratory studies primarily aim to exclude systemic or metabolic conditions that can mimic myelopathy. These laboratory investigations are indicated when history or examination raises concern for noncompressive causes of cord dysfunction.[25] 

Based on clinical suspicion, potential testing may include a complete blood count, inflammatory markers (erythrocyte sedimentation rate and C-reactive protein), and panels assessing vitamins B1, B6, B12, E, and thyroid function. Serological screening for infectious or autoimmune myelopathies, such as syphilis serology, human immunodeficiency virus, antinuclear antibodies, and aquaporin-4 antibodies, can be tailored based on clinical suspicion. Serum copper and zinc levels should be measured if nutritional deficiency is suspected. Creatine kinase and autoantibody panels help exclude primary myopathic processes.

Plain radiographs of the cervical spine constitute the initial imaging modality. Standard anteroposterior and lateral views evaluate spinal alignment, disc height, and gross bony anatomy. Flexion-extension radiographs identify dynamic instability or spondylolisthesis. Oblique views may reveal foraminal narrowing and uncovertebral joint spurs. Radiographs also detect OPLL and osteophytic encroachment that contribute to canal stenosis. Load-bearing imaging (upright views) is crucial for assessing physiological alignment, including cervical lordosis and sagittal balance.

Computed tomography (CT) offers high-resolution visualization of bony structures and is indicated when magnetic resonance imaging (MRI) is contraindicated or when detailed assessment of OPLL or osteophyte morphology is required. CT helps quantify ossification, classify OPLL patterns, and guide surgical planning by defining bone-quality and screw trajectories. CT angiography may be employed selectively to assess vertebral artery anatomy in cases with suspected vascular compromise or anomalous anatomy. CT myelography is helpful in cases where MRI findings are equivocal (eg, due to the presence of magnetic artifacts from instrumentation) or when MRI is contraindicated.

MRI is the gold standard for evaluating cervical myelopathy. MRI yields multiplanar, high-contrast images of the spinal cord, soft tissues, and osseous elements. T2-weighted sequences estimate the degree of cord compression and detect changes in intramedullary signal. High T2 signal intensity often correlates with edema, gliosis, or myelomalacia, whereas low T1 signal intensity indicates chronic neural tissue loss.[31]

Standard classification systems used to quantify MRI findings include:

• Muhle grading: Categorizes spinal canal stenosis from grade 0 (no subarachnoid space narrowing) to grade 3 (spinal cord compression or displacement).
• Kang modification: Expands on Muhle grading by incorporating T2-weighted signal changes—grade 0 (no stenosis), grade 1 (subarachnoid space obliteration >50%), grade 2 (cord deformation), and grade 3 (intramedullary T2 signal change).

Quantitative MRI metrics enhance diagnostic precision, but none of them has reached widespread acceptance in clinical practice. Dynamic (or kinematic) MRI obtained in flexion and extension can detect stenotic changes that static neutral-position scans may overlook. In a series of 81 patients with degenerative cervical disease, extension views revealed new areas of cord impingement in 27% of subjects, while flexion views did so in 5%.[32] In a surgical cohort, incorporating flexion–extension sequences altered the intended decompression levels in 46% of cases and was associated with better postoperative outcomes (ΔmJOA +3.9 vs +2.4; ΔVAS –4.1 vs –2.7) compared to planning based solely on neutral MRI.[33]

Diffusion-weighted MRI offers quantitative, microstructural parameters that outperform simple canal-diameter measurements for assessing the severity of cervical spondylotic myelopathy. In a prospective trial, diffusion basis spectrum imaging demonstrated 81% accuracy in distinguishing between healthy controls, mild, and severe cases. These diffusion metrics correlated more strongly with postoperative improvements in the mJOA score than did conventional imaging parameters, underscoring their prognostic utility.[34]

Electrophysiological testing is not routinely obtained in DCM, but in select cases it can help evaluate functional integrity and exclude mimics. Somatosensory evoked potentials monitor dorsal column pathways; abnormalities in N13 are found in DCM. Motor evoked potentials assess corticospinal tracts and can help differentiate DCM from motor neuron disease.[25] Electromyography and nerve conduction studies differentiate radiculopathy, peripheral neuropathy, and anterior horn cell diseases. 

OPLL

The imaging evaluation of OPLL in DCM begins with a lateral cervical radiograph, on which the K-line is drawn by connecting the midpoints of the spinal canal at C2 and C7. When the ossified ligament remains entirely anterior to this line (K-line positive), sufficient posterior shift of the spinal cord typically occurs after a posterior decompressive procedure. Conversely, if the ossification extends beyond (posterior to) the K-line (K-line negative), a posterior approach alone is unlikely to relieve cord compression, indicating a need for an anterior or combined decompressive strategy. CT provides precise delineation of OPLL morphology and canal occupancy.

On sagittal computed tomography reconstructions, OPLL is subclassified into:

• Segmental type: Multiple, discrete ossified foci located posterior to individual vertebral bodies
  • These focal lesions often represent early or limited OPLL involvement and can produce localized spinal canal narrowing.
• Continuous type: A confluent ossified mass extending across 3 or more adjacent vertebral bodies without interruption
  • Continuous OPLL frequently results in more severe and multilevel canal stenosis.
• Focal type: Ossification confined to the posterior surface of 1 intervertebral disc level, often appearing between segmental and continuous forms
• Mixed type: Combination of the above patterns, wherein discrete segmental ossifications coexist with longer spans of continuous ossification.

Axial computed tomography sections allow measurement of the canal-occupancy ratio (COR), defined as the maximum thickness of OPLL divided by the anteroposterior diameter of the spinal canal. A higher COR correlates with more pronounced static stenosis and is predictive of symptom severity. MRI assesses cord compression and intramedullary signal changes: T2-weighted sagittal images reveal hyperintense myelomalacia at levels of maximal ossification, and axial T2 views help distinguish OPLL from other compressive lesions.[35]

Treatment / Management

Surgical intervention remains the cornerstone of treatment for DCM once functional impairment extends beyond mild symptoms. For patients classified as having moderate (mJOA 12–14) or severe (mJOA ≤11) DCM, current evidence supports prompt decompression to prevent irreversible neurological decline and optimize recovery of function.[36] In these cohorts, surgery has demonstrated large, sustained improvements in mJOA scores, Nurick grades, and disability indices, with low rates of major complications over short-, intermediate-, and long-term follow-up.(A1)

For patients with mild DCM (mJOA 15–17), management may begin with a supervised trial of structured rehabilitation, including immobilization (eg, cervical orthosis), physical therapy focused on cervical stabilization and strengthening, and nonsteroidal anti-inflammatory medications. Such nonoperative measures have not consistently produced gains exceeding minimal clinically important differences in mJOA scores, and 20% to 62% of patients may deteriorate without surgical decompression. Therefore, when mild DCM is managed conservatively, clinicians should maintain close serial follow-up to monitor progression. An operative intervention is recommended in the event of neurologic worsening and may be considered if functional status fails to improve after a defined trial of structured rehabilitation.

The AO Spine (Arbeitsgemeinschaft für Osteosynthesefragen, Germany) 2025 recommendations make a conditional recommendation that surgical decompression be offered as a valid option for mild DCM. However, in the absence of randomized comparisons against continued observation, structured nonoperative follow-up (with regular clinical reassessment) remains reasonable for patients without progressive deficits.[37] A lifetime cost-utility analysis demonstrated that early surgery in mild DCM yields an average incremental gain of 1.26 quality-adjusted life-years at an incremental cost-utility ratio well below accepted willingness-to-pay thresholds, supporting the cost-effectiveness of intervention.[38]

Nonmyelopathic individuals with radiographic cord compression but without radiculopathy should not undergo prophylactic decompression. They should be counseled on symptom vigilance and monitored clinically, given the low absolute risk of progression. Conversely, those with asymptomatic cord compression who present with radiculopathy (especially those with electrophysiologic confirmation) bear a higher risk of developing myelopathy. They may be offered either early decompression or a structured nonoperative protocol with close follow-up. If myelopathy ensues, management should follow the recommendations for mild, moderate, or severe DCM based on the mJOA grading score. Factors predictive of clinical decline during nonoperative management include circumferential cord compression on axial MRI, segmental hypermobility, angular-edged vertebral deformity, spinal instability, diminished lordosis, and the presence of OPLL.[39] Surgical intervention can involve either an anterior, posterior, or a combined approach.

Cervical decompression can be approached via the anterior, posterior, or combined routes, with the choice largely driven by the location and extent of pathology, cervical alignment, and patient comorbidities. Among anterior techniques, anterior cervical discectomy and fusion (ACDF) is commonly employed for 1–3 level ventral compression, such as disc herniations or focal osteophytes. By directly removing offending disc material or osteophytes and restoring segmental height with a bone graft and plate, ACDF reliably improves neurologic function while minimizing intraoperative blood loss and preserving posterior musculature.

ACDF drawbacks include a higher risk of dysphagia or recurrent laryngeal nerve irritation, adjacent-segment degeneration over time, and increasingly fragile constructs in osteoporotic bone, particularly when more than 2 levels are fused. When pathology extends behind the vertebral bodies (such as extensive OPLL or large central osteophytes) or when kyphosis requires correction, anterior cervical corpectomy and fusion (ACCF) may be preferred. By excising 1 or more vertebral bodies and reconstructing the column (with a strut graft and plate or a titanium cage), ACCF achieves more comprehensive ventral decompression and greater correction of kyphosis, albeit at the expense of increased blood loss, longer operative time, and a higher incidence of graft-related complications (eg, subsidence, pseudarthrosis) and postoperative dysphagia.

Hybrid approaches—combining corpectomy at 1 level with discectomy and fusion at adjacent levels—or skip corpectomies can tailor the decompression to irregular patterns of pathology while preserving intervening vertebral bodies, potentially reducing graft stress. However, these techniques are technically demanding, carry similar risks of graft failure, and usually prolong operative time. The role of cervical arthroplasty in established cervical myelopathy remains somewhat debated because fusion reliably immobilizes the affected segment, thereby minimizing any dynamic component of cord compression. However, when myelopathy is caused solely by a focal, noncalcified disc herniation (without significant osteophytes, OPLL, or segmental kyphosis), disc replacement is broadly accepted as a valid, ‘relative’ indication without major controversy.

Posterior decompression techniques become advantageous when 3 or more contiguous levels require treatment, when there is no rigid cervical kyphosis, or when patients have significant dorsal compressive elements, such as ligamentum flavum hypertrophy or ossification. A simple laminectomy provides direct dorsal canal expansion and is relatively straightforward; however, sacrificing the posterior tension band often leads to progressive postlaminectomy kyphosis, late instability, and a risk of C5 palsy. To mitigate these sequelae, laminectomy is frequently paired with instrumented fusion using lateral-mass screws and rods. This combination maintains alignment and prevents kyphotic collapse but prolongs operative time, increases blood loss, and elevates the incidence of hardware-related complications.

Laminoplasty, whether in the “open-door” or “French-door” configuration, offers indirect decompression by hinging and expanding the lamina while preserving the posterior elements and motion. In multilevel stenosis without significant preoperative kyphosis, laminoplasty yields neurological improvement rates comparable to laminectomy. However, it may not adequately relieve severe axial neck pain, and is contraindicated when preoperative kyphosis exceeds approximately 10° or in the presence of instability. Performing dome laminotomies at the laminae adjacent to the laminoplasty directly decompresses and prevents static or dynamic spinal cord kinking that can occur when the cord dorsally migrates after decompression.[40]

For patients with combined ventral and dorsal compression (such as those with extensive OPLL or severe kyphotic deformity), neither anterior nor posterior decompression alone may suffice. In these complex cases, a staged or single-sitting combined anterior–posterior procedure is often advocated. Anterior surgery (typically ACCF or multilevel ACDF) removes the principal ventral lesion and corrects alignment; this is followed by posterior laminectomy (or laminoplasty) with fusion to maintain stability, address residual dorsal compression, and prevent postoperative deformity. Although this circumferential approach offers the most reliable neurologic recovery and long-term correction of alignment in rigid deformities, it entails the greatest operative time, blood loss, perioperative morbidity, and prolonged hospitalization. Thus, combined surgery is generally reserved for patients with fixed kyphosis, multilevel OPLL, overt instability, or those undergoing revision decompression.[41][42]

In addition to surgical and rehabilitation strategies, several pharmacological agents have been investigated as adjuncts or standalone therapies in DCM, aiming to mitigate secondary injury pathways and promote neuroprotection. Overall, these pharmacological approaches target multiple secondary injury mechanisms, including excitotoxicity (riluzole), neurotrophic support (cerebrolysin, erythropoietin [EPO]), microcirculatory enhancement (limaprost, cilostazol), anti-inflammation (glucocorticoids), stem cell differentiation (granulocyte colony-stimulating factor), and anti-apoptosis (EPO, anti–Fas ligand). However, clinical evidence remains sparse, often limited by small sample sizes, lack of blinding, or inconsistent results. At present, robust, large-scale randomized trials are needed to determine whether any pharmacological adjunct can reliably delay progression in mild DCM or improve outcomes after surgical decompression.[43]

Regardless of initial strategy, all patients should receive education on activity modification to reduce cervical stress, avoid high-risk trauma (eg, contact sports), and recognize early myelopathic signs (eg, hand clumsiness, gait instability). Nutritional optimization, smoking cessation, and management of comorbidities (eg, diabetes, osteoporosis) enhance surgical candidacy and may favorably influence fusion rates and recovery. For individuals with mild deficits, observing and periodically reassessing (both clinically and radiographically) at intervals of 3 to 12 months is prudent to detect any subclinical progression.

OPLL

The extent and continuity of OPLL (segmental versus continuous) and its relationship to the K-line determine whether a posterior-only approach (laminoplasty or laminectomy with fusion) will allow sufficient posterior cord shift or whether an anterior decompression (discectomy or corpectomy with fusion) is required to remove the ossified mass directly. Specifically, anterior decompression is generally indicated when the ossified ligament extends beyond the K-line (K-line negative) or, in K-line–positive individuals, when the center of rotation (COR) is 60% or greater, a direct anterior resection is required to relieve severe focal compression.

In contrast, a posterior approach is preferred when the ossification remains entirely anterior to the K-line (K-line positive) with a COR below 60% or when multilevel OPLL produces more diffuse stenosis that can be adequately addressed by expanding the canal posteriorly. Finally, patients with severe, multilevel disease often benefit from a combined anterior–posterior strategy to ensure comprehensive decompression and maintain alignment.[35] An additional factor in surgical planning is that the dura is usually calcified and integrated into the PLL calcified mass, increasing the risk of cerebrospinal fluid leak and cord injury through an anterior approach that aims to remove all calcified mass. For this reason, in cases where either an anterior or posterior approach is equally valid for cord decompression, some surgeons might choose the posterior approach.(B3)

Differential Diagnosis

Cervical myelopathy may be mimicked by a range of compressive and noncompressive spinal cord disorders as well as peripheral nerve and neuromuscular conditions. Careful clinical evaluation, imaging studies, and neurophysiological testing are crucial for distinguishing these entities from true cervical myelopathy.[25]

Compressive MyelopathiesConditions that directly compress the cord can present similarly to cervical myelopathy. Central cord syndrome (often from trauma or acute disc herniation) produces bilateral weakness and sensory changes, sometimes preceding frank myelopathy. Chiari malformation can also cause cord compression at the foramen magnum, manifesting with gait ataxia or hand clumsiness. Epidural abscess typically presents with severe localized neck pain, systemic signs of infection, and rapidly progressive neurologic deficits. Intradural or extradural neoplasms (such as meningiomas, schwannomas, or metastases) may cause insidious onset of motor and sensory deficits, often accompanied by constitutional symptoms (fatigue, weight loss). Syringomyelia, whether idiopathic or associated with Chiari, can produce a cape-like distribution of pain and temperature loss, frequently sparing vibration sense until advanced stages.

Noncompressive MyelopathiesIntrinsic cord disorders without focal compression may closely resemble cervical myelopathy. Demyelinating conditions like multiple sclerosis or neuromyelitis optica spectrum disorder often present with sensory disturbances, motor weakness, and spasticity. Acute transverse myelitis typically evolves over hours to days and may be associated with antecedent viral illness or systemic autoimmune disease. Spinal cord infarction (due to aortic surgery or embolic phenomena) manifests with rapid motor/sensory loss below the lesion level. Nutritional deficiencies (vitamin B12, vitamin E, or copper) produce subacute dorsal column dysfunction, often accompanied by peripheral neuropathy. Infectious myelitis (caused by human immunodeficiency virus, human T-cell lymphotropic virus type 1, Lyme disease, syphilis, or schistosomiasis) can also mimic compressive myelopathy. Still, it may also include fever, neck stiffness, or systemic symptoms. Radiation-induced myelopathy typically arises months after therapy and presents with gradual onset of spastic weakness and sensory changes. Metabolic and toxic myelopathies (eg, nitrous oxide–induced B12 depletion) may feature predominant dorsal column signs with minimal or no cord compression on imaging.

Neurodegenerative DisordersMotor neuron diseases require differentiation from cervical myelopathy, especially when imaging suggests cord compression. Amyotrophic lateral sclerosis (ALS) can mimic myelopathy by producing a mix of upper and lower motor neuron signs in the limbs without sensory loss or sphincter involvement. Unlike compressive myelopathy, ALS typically demonstrates normal spinal cord imaging and preserved dorsal column function. Peripheral nerve stimulation in ALS yields reduced M-wave amplitudes and decreased F-wave frequency; these findings are not seen in compressive myelopathy.[44]

Peripheral Nerve and Radicular SyndromesFocal nerve root compression or peripheral neuropathies can masquerade as myelopathy. Cervical radiculopathy typically presents with dermatomal pain, sensory loss, and isolated muscle weakness, often without signs in the lower extremities or hyperreflexia. Carpal tunnel syndrome (median neuropathy at the wrist) may cause thenar atrophy and hand numbness, but it typically does not cause gait disturbance or hyperreflexia. Brachial plexopathies, peripheral polyneuropathies (eg, diabetic neuropathy), and neuromuscular junction disorders (eg, myasthenia gravis) often present with asymmetric or distal weakness and preserved upper motor neuron signs. Electromyography and nerve conduction studies help localize lesions to the peripheral nerve or root level, showing slowed conduction velocities or denervation potentials, thereby ruling out central cord pathology.

Prognosis

Prognostic evaluation in cervical myelopathy relies on specific clinical and radiological variables. Age, symptom duration, and baseline myelopathy severity exert the greatest influence on outcomes after surgical intervention. Older patients, prolonged symptom duration, and more severe deficits at presentation are associated with less favorable recovery.[45]

In addition to these descriptive factors, examination findings (such as hand muscle atrophy, lower extremity spasticity, clonus, and a positive Babinski sign) have demonstrated predictive value. Patients exhibiting pronounced hand atrophy or extensor plantar responses often experience poorer postoperative improvement. Intramedullary signal changes on T2-weighted imaging carry important prognostic weight. Patients with a fuzzy margin of T2 hyperintensity may have active cord edema and inflammation, which could respond to timely decompression. In contrast, well-defined focal hyperintensity suggests gliosis or necrosis, indicating limited potential for neurological recovery.[46]

Diffusion tensor imaging parameters, such as fractional anisotropy and the apparent diffusion coefficient (ADC), have been shown to have prognostic utility. Fractional anisotropy can be useful for gauging the severity of myelopathy; however, it does not reliably predict postdecompression recovery. In contrast, ADC correlates with surgical outcomes but not with initial symptom severity. Clinically, this means that high fractional anisotropy values reflect more severe symptoms without necessarily indicating postoperative improvement. In contrast, ADC appears to be a more dependable predictor of functional recovery after decompression.[47]

Delays in diagnosis are common in DCM, and such delays contribute to worse recovery of neurological function after surgery. In a prospective study, presurgical serum levels of neurofilament light chain (NfL), interleukin-6 (IL-6), and brain-derived neurotrophic factor (BDNF) were significantly higher in DCM patients than in healthy controls, and the combined panel of NfL, IL-6, and BDNF demonstrated high diagnostic accuracy (area under the curve: 0.83). Presurgical serum NfL was significantly associated with improvement in pinch strength after decompression, indicating that higher circulating NfL (which reflects ongoing axonal injury) paradoxically portends better postsurgical hand function recovery when decompression is performed promptly.[48]

Complications

Untreated cervical myelopathy may lead to irreversible neurological deficits. Patients can develop progressive motor weakness and sensory loss. Autonomic dysfunction may include disturbances of the bladder or bowel. Delayed surgical intervention often reduces the likelihood of functional recovery. Overreliance on physical therapy can postpone definitive decompression and worsen long-term outcomes. Forceful, rapid, or extreme passive neck range-of-motion maneuvers carry a significant risk of spinal cord injury in patients with preexisting cord compression. Moreover, accidents and falls carry a risk for spinal cord injury and paralysis in the setting of cord compression.

All surgeries carry the risk of paralysis, infection, bleeding, dural tears, and cerebrospinal fluid leakage. In the case of fusion, the added risks are pseudarthrosis, hardware complications, and adjacent segment disease. Complications of anterior cervical surgery include dysphagia, hoarseness, and airway compromise. Esophageal or tracheal injury is rare but may be devastating. Vertebral artery injury, hematoma formation, and surgical site infection are rare but can occur. Graft subsidence, nonunion, and hardware failure may compromise spinal stability.

Posterior decompression may result in C5 nerve root palsy and persistent neck pain. Wound infections are more common with posterior approach procedures than with anterior approach procedures. Postoperative kyphotic deformity (especially following multilevel laminectomy without fusion) can undermine neurological improvement. Explainable machine learning models revealed that postoperative complications in surgeries commonly performed for cervical myelopathy are associated with the number of operated levels, patient age, body mass index, preoperative hematocrit, and American Society of Anesthesiologists physical status.[49][50][51][52]