Spinal Cord Compression

Spinal cord compression (SCC) is a clinical syndrome of neurologic dysfunction caused by mechanical compression of the spinal cord from an extrinsic or intrinsic mass, bony abnormality, or inflammatory process, requiring emergency recognition and etiology-specific intervention ^[1]D5[2]A1a.

Also Called / Synonyms

• Acute spinal cord compression (when onset <24 hours)
• Compressive myelopathy (pathoanatomic term)
• Spinal cord compression syndrome (clinical syndromic term)
• Cord compromise (radiologic term, often used in trauma)
• Metastatic spinal cord compression (MSCC) (when due to malignancy)
• Cervical myelopathy (when localized to the cervical spine; commonly used in degenerative disease)
• Thoracic spinal cord compression (when localized to the thoracic region)

Defining the Entity: Localization-Anchored Taxonomy

Spinal cord compression is not a single disease but a syndromic diagnosis, a final common pathway of diverse etiologies that converge on the same clinical and pathoanatomic endpoint. The syndrome is defined by three co-ordinates:

1. Anatomic level, the vertebral segment(s) where compression occurs: cervical (C1-T1), thoracic (T1-T12), or lumbar (below conus medullaris, i.e., cauda equina syndrome , which is a distinct but related entity).
2. Compressive pathology, the specific lesion responsible (e.g., herniated disc, epidural abscess, tumor, fracture fragment).
3. Temporal course, acute (hours to days), subacute (days to weeks), or chronic (months to years), which predicts reversibility and urgency ^[2]A1a.

This three-axis taxonomy (level × pathology × time) must be specified in every patient encounter because each axis independently drives management decisions ^[1]D5[2]A1a.

Classification of Compressive Etiologies

Controlled Vocabulary: Syndromic vs Etiologic vs Anatomic

To ensure clear communication across specialties, the following terms are used with precision throughout this article:

• Syndromic terms (describe clinical presentation): Compressive myelopathy (cord-level findings); Cauda equina syndrome (CES; lumbar root-level findings).
• Etiologic terms (identify cause): Degenerative cervical myelopathy (DCM); Metastatic spinal cord compression (MSCC); Pott paraplegia (tuberculous SCC).
• Anatomic terms (localize compression): Cervical, thoracic, lumbar SCC; Extradural vs intradural (extramedullary vs intramedullary).

Clinical Significance

Spinal cord compression is a neurologic emergency, failure to intervene within hours (for acute etiologies) to days (for subacute etiologies) can lead to irreversible paraplegia, quadriplegia, loss of bowel/bladder function, and permanent disability ^[2]A1a. It is the third most common cause of non-traumatic myelopathy after inflammation and ischemia ^[2]A1a, and the leading cause of cancer-related neurologic morbidity in the form of metastatic spinal cord compression ^[2]A1a.

Pearl: Spinal cord compression must be described by its three-axis taxonomy (level × pathology × time) in every patient encounter; this framework directly determines urgency, neuroimaging protocol, and the choice of surgical versus medical intervention ^[1]D5[2]A1a.

Mechanical deformation of the spinal cord initiates a cascade of vascular, inflammatory, and cellular events that determine the tempo and reversibility of neurological injury. The final common pathway, regardless of etiology, involves direct axonal injury, microvascular compromise, and a secondary neuroinflammatory response that amplifies tissue damage ^[8]D5[10]D5. The neuroanatomic level of compression dictates which tracts are affected, producing distinct clinical syndromes that localize the lesion.

Mechanical Compression and the Biomechanical Cascade

Compression of the spinal cord reduces the cross-sectional area of the canal, leading to deformation of neural tissue. The mechanical properties of the cord, its viscoelasticity and tolerance to strain, are poorly characterized in vivo, but cadaveric and animal studies show that axial loading and buckling cause immediate axonal stretch and disruption of the blood-spinal cord barrier ^[11]D5. This barrier breakdown permits influx of inflammatory mediators and immune cells, initiating a secondary injury phase. In degenerative cervical myelopathy (DCM), chronic, repetitive compression from osteophytes, ligamentum flavum hypertrophy, or disc herniation produces progressive demyelination and Wallerian degeneration of long tracts ^[8]D5. The corticospinal tracts (motor) and dorsal columns (proprioception/vibration) are particularly vulnerable in the cervical region, while spinothalamic fibers (pain/temperature) are often spared until later stages.

In acute traumatic central cord syndrome, hyperextension injuries cause buckling of the ligamentum flavum into the cord, preferentially damaging the central gray matter and medial portions of the corticospinal tracts ^[12]A1a. This explains the characteristic pattern of upper extremity weakness greater than lower extremity weakness. Surgical decompression within 24 hours versus later did not significantly alter outcomes in a meta-analysis, suggesting that the primary mechanical injury may be irreversible and that secondary inflammatory cascades dominate the subacute phase ^[12]A1a.

Inflammatory and Immune-Mediated Mechanisms

A growing body of evidence implicates the immune system in both primary and secondary injury. In spinal hypertrophic pachymeningitis (HP), thickening of the dura mater compresses the cord from outside. All six patients in a Japanese cohort with spinal HP were seropositive for myeloperoxidase (MPO)-anti-neutrophil cytoplasmic antibody (ANCA), indicating an autoimmune vasculitic process ^[7]B3b. The arachnoid blood-CSF barrier partitions immune cell repertoires, allowing distinct inflammatory infiltrates to accumulate in the subdural space ^[7]B3b. Similarly, IgG4-related spinal pachymeningitis involves dense lymphoplasmacytic infiltration and fibrosis of the dura, leading to cord compression that may be mistaken for neoplasm ^[9]C4. In DCM, systematic review data show that activated microglia, astrocytes, and infiltrating T cells contribute to ongoing neurodegeneration even after the mechanical insult is removed ^[8]D5. These immune mediators release cytokines (TNF-α, IL-1β) that promote apoptosis of oligodendrocytes and neurons, perpetuating demyelination and axonal loss ^[8]D5.

Neoplastic and Structural Causes

Metastatic spinal cord compression (MSCC) results from hematogenous seeding of the vertebral column, most commonly from lung, breast, or prostate primaries ^[10]D5. Tumor growth within the epidural space directly compresses the cord and also obstructs venous drainage, leading to vasogenic edema and ischemia. The pathophysiology is compounded by tumor-secreted factors that increase vascular permeability and promote bone resorption, causing pathological fractures that further narrow the canal ^[10]D5. Spinal arachnoid cysts, though benign, act as space-occupying lesions that compress the cord or cauda equina; their pathogenesis involves a one-way valve mechanism that allows CSF entry but prevents egress, causing progressive enlargement ^[14]B3b.

Genetic and Metabolic Causes

In Morquio A syndrome (mucopolysaccharidosis type IVA), mutations in the GALNS gene cause deficiency of N-acetylgalactosamine-6-sulfatase, leading to accumulation of glycosaminoglycans in connective tissues ^[15]D5. This deposition thickens ligaments and dura, narrows the spinal canal, and predisposes to atlantoaxial instability and cord compression at the craniocervical junction ^[15]D5. The genotype-phenotype correlation is diverse, with some variants causing early, severe compression and others a milder course ^[15]D5.

Common Pathway and Neuroanatomic Localization

Regardless of the inciting event, sustained compression reduces spinal cord blood flow below a critical threshold, causing ischemia of the central gray matter and surrounding white matter tracts. The anterior spinal artery territory (supplying the corticospinal and spinothalamic tracts) is most vulnerable, while the posterior columns (supplied by posterior spinal arteries) are relatively resistant ^[10]D5. This explains why motor and pain/temperature deficits often precede proprioceptive loss in compressive myelopathy. The level of compression determines the segmental and long-tract signs: cervical lesions affect all four limbs, thoracic lesions cause paraplegia with a sensory level, and lumbar/cauda equina lesions produce lower motor neuron signs and sphincter dysfunction.

Controversies and Guideline Disagreement

Pearl: The reversibility of neurological injury depends on the duration and severity of compression; early identification of the specific pathophysiological mechanism, mechanical, inflammatory, or metabolic, guides targeted therapy and prognostication ^{[7]B3b[10]D5[12]A1a}.

The incidence of spinal cord compression varies by etiology, with metastatic disease representing the most common cause in adults ^{[19]A1a[22]B3b}. In a retrospective cohort of 256 patients undergoing surgery for metastatic spinal cord compression (MSCC), the condition accounted for a substantial surgical burden over two decades ^[22]B3b. Post-tuberculosis kyphosis remains a leading cause in endemic regions, with a series of 51 patients undergoing corrective surgery ^[3]B3b. Iatrogenic causes are less frequent but quantifiable: intraspinal hematoma after percutaneous kyphoplasty (PKP) occurs in approximately 1% of procedures ^[21]C4. Congenital anomalies such as os odontoideum, though rare, are systematically reviewed with meta-analytic prevalence estimates ^[16]A1a. De novo C5 palsy, presenting without prior surgery or trauma, was documented in 26 patients over a retrospective study ^[18]B3b. Cervical spondylodiscitis with epidural abscess, while uncommon, carries high morbidity ^[20]C4.

Demographic Distribution

Age and geography strongly influence etiology. MSCC typically affects older adults, with median age in the seventh decade ^[22]B3b. Post-tuberculosis kyphosis disproportionately affects younger individuals in regions with high tuberculosis prevalence ^[3]B3b. Syndromic atlanto-axial instability, as seen in Down syndrome and Morquio syndrome, presents in childhood or adolescence ^[17]D5. Sex distribution is not consistently reported across studies, but no clear sex predilection emerges from the available data ^{[18]B3b[22]B3b}.

Risk Factors

Risk factors for spinal cord compression are best understood within specific etiologic buckets. The table below summarizes key factors with available evidence levels.

Modifiable risk factors include smoking, infection control (tuberculosis, spondylodiscitis), and iatrogenic procedural risks (PKP). Non-modifiable factors include age, congenital anomalies (os odontoideum, syndromic instability), and metastatic cancer.

Special Considerations

Post-infectious causes deserve emphasis: tuberculous spondylitis can lead to progressive kyphosis and delayed cord compression years after initial infection ^[3]B3b. Acute cervical spondylodiscitis may cause rapid neurological decline due to epidural abscess formation ^[20]C4. Iatrogenic risk is highlighted by the 1% incidence of intraspinal hematoma after PKP, mandating careful patient selection and perioperative blood pressure control ^[21]C4. Syndromic patients (Down, Morquio) require screening for atlanto-axial instability before sports participation or surgical procedures ^[17]D5.

Pearl: The etiology of spinal cord compression dictates its epidemiology : metastatic disease predominates in older adults, infectious causes in endemic regions, and congenital anomalies in younger populations; recognizing these patterns guides pre-test probability and timely imaging ^{[19]A1a[22]B3b}.

Symptoms progress over days to weeks in most cases, with a nadir at 2-4 weeks, though acute presentations (hours) occur with trauma, hemorrhage, or abscess ^[24]D5. The clinical picture is dictated by the level, rate, and etiology of the compressive lesion, translating the localized mechanism into a recognizable syndrome.

Presenting Symptoms

Back pain is the earliest and most consistent symptom, often localized to the level of compression and exacerbated by recumbency or Valsalva maneuver ^[14]B3b. Radicular pain follows a dermatomal distribution and may be the dominant complaint in lateral lesions (e.g., chloroma presenting with left S2 radiculopathy ^[23]C4). Motor weakness typically begins proximally in the lower extremities (hip flexors, quadriceps) and progresses distally; in cervical lesions, upper extremity weakness may precede leg involvement. Sensory changes include numbness, paresthesias, and a discrete sensory level on the trunk, a finding with high specificity for cord compression. Autonomic dysfunction, urinary retention, incontinence, constipation, or erectile failure, signals involvement of the descending autonomic pathways or the conus medullaris ^[25]C4. Gait disturbance (spastic or ataxic) is common in chronic compression.

Neurological Examination Findings

Motor examination should include manual muscle testing of all myotomes, with attention to patterns: C5 palsy selectively weakens the deltoid and biceps without prior surgery ^[18]B3b; thoracic lesions spare the arms but produce paraparesis. Sensory examination must map a dermatomal level using pinprick and light touch; vibration and proprioception assess dorsal column function. Reflexes reveal hyperreflexia and clonus below the lesion, a Babinski sign, and a suspended sensory level. At the level of compression, hyporeflexia or areflexia may be present. Autonomic assessment includes anal sphincter tone, bulbocavernosus reflex, and postvoid residual volume (>100 mL suggests neurogenic bladder ). Special maneuvers: Lhermitte sign (electric shock-like sensation with neck flexion) indicates cervical cord irritation; straight leg raise may reproduce radicular pain.

Phenotypic Variants

Red Flags

• Respiratory compromise: Cervical cord compression above C5 can impair phrenic nerve function. FVC < 15 mL/kg or a declining trend should prompt consideration of elective intubation.
• Autonomic instability: Hypotension with bradycardia (neurogenic shock) in acute cervical or upper thoracic lesions requires immediate hemodynamic support.
• Rapid progression: Onset over hours suggests hemorrhage, abscess, or metastatic disease, urgent imaging and surgical consultation are mandatory.
• Saddle anesthesia and urinary retention: These hallmark features of conus medullaris or cauda equina syndrome demand emergent decompression to preserve bowel and bladder function ^[25]C4.

Atypical Presentations

Painless weakness may occur with slow-growing tumors (e.g., some meningiomas) or in patients with preexisting neuropathy. Isolated sensory symptoms (numbness, paresthesias) without motor findings can delay diagnosis, especially in thoracic lesions. De novo C5 palsy presents with pure motor deficits in the deltoid and biceps, often mistaken for a peripheral nerve lesion ^[18]B3b. Delayed spinal cord compression years after spinal cord stimulator placement from intradural fibrosis is a rare but important mimic ^[28]C4. Extranodal Rosai-Dorfman disease isolated to the thoracic spine may present with back pain and paresthesias without systemic symptoms ^[27]B3b.

Pearl: The combination of localized back pain, a sensory level, and early autonomic dysfunction is the most specific clinical triad for spinal cord compression; any patient with progressive myelopathy and a sensory level should undergo urgent MRI regardless of symptom duration ^{[14]B3b[25]C4}.

The diagnostic workup of suspected spinal cord compression begins with urgent whole-spine MRI, the single gold-standard test that must be obtained within hours of clinical suspicion ^[31]D5. Delaying imaging is the most common preventable cause of permanent paralysis; the goal is to identify the compressive lesion and its etiology before irreversible cord injury occurs.

Gold-Standard Imaging: Whole-Spine MRI

MRI is the test of choice for diagnosing spinal cord compression, with a sensitivity exceeding 95% for detecting epidural masses, cord signal change, and vertebral body pathology ^[31]D5. The protocol must include the entire spine (cervical, thoracic, lumbar) because skip lesions occur in up to 20% of metastatic cases and in multifocal infections ^[31]D5[35]C4. Essential sequences:

• T1-weighted without contrast: identifies vertebral body lesions (hypointense in metastasis, infection) and loss of epidural fat planes.
• T2-weighted with fat suppression: reveals cord edema, gliosis, or myelomalacia (hyperintense signal) and differentiates fluid collections (abscess, hematoma).
• Post-contrast T1-weighted with fat suppression: highlights enhancing epidural masses (tumor, infection) and distinguishes abscess from sterile effusion.
• STIR (short tau inversion recovery): sensitive for bone marrow edema in infection or tumor.

Key findings include: cord compression (effacement of CSF spaces around the cord), cord signal change (T2 hyperintensity indicating edema or gliosis), and enhancing epidural or paraspinal mass ^[29]C4[30]C4. The Bilsky grading system (0-3) for epidural spinal cord compression on axial T2-weighted MRI is widely used to quantify severity and guide surgical planning: Grade 0 = bone-only disease; Grade 1 = epidural impingement without cord compression; Grade 2 = cord compression with visible CSF; Grade 3 = cord compression without visible CSF ^{[31]D5[40]B3b}. For cauda equina compression, a novel ECEC grading system (0-3) has been proposed, with Grade 3 indicating complete thecal sac effacement and no visible CSF ^[40]B3b.

Contraindications to MRI (e.g., non-MRI-conditional pacemaker, severe claustrophobia) are rare; when present, CT myelography is the alternative, with sensitivity of approximately 85% for cord compression ^[31]D5. CT myelography requires lumbar puncture and intrathecal contrast, and carries risks of post-dural headache and infection.

Adjunctive Imaging Modalities

• CT spine without contrast: Useful for assessing bony anatomy (fracture, osteolysis, osteochondroma ) and for surgical planning, but cannot directly visualize cord compression. A deep learning model applied to cervical radiographs has shown promise in detecting compression, but MRI remains the gold standard ^[39]B3b.
• 18F-FDG-PET/CT: May identify hypermetabolic lesions in metastatic disease or infection, but is not first-line for acute cord compression ^[30]C4.
• Spinal angiography: Reserved for suspected vascular lesions (AVM, hemangioblastoma) when MRI suggests flow voids or hemorrhage.

Laboratory Studies and CSF Analysis

Blood tests help identify the underlying etiology:

• Complete blood count , ESR, CRP: Elevated in infection (spondylodiscitis, epidural abscess) ^[35]C4.
• Serum tumor markers: Prostate-specific antigen (PSA), cancer antigen 125, etc., when metastatic disease is suspected.
• Infectious serologies: Syphilis (TPHA, RPR) in cases of chronic myelopathy with longitudinally extensive lesions ^[30]C4; tuberculosis (Quantiferon) in endemic areas.
• Autoimmune panel: ANA, anti-aquaporin-4, anti-MOG for inflammatory myelitis.

CSF analysis is indicated when MRI is equivocal or when infection/inflammation is suspected. Typical findings:

• Pleocytosis (WBC >5 cells/μL) with lymphocytic predominance in viral or syphilitic myelitis; neutrophilic in bacterial abscess.
• Elevated protein (>45 mg/dL) common in compressive lesions due to CSF stasis; markedly elevated (>500 mg/dL) suggests complete block.
• Hypoglycorrhachia (CSF glucose <40% of serum) is characteristic of bacterial or fungal infection ^[30]C4.
• Positive TPHA in CSF confirms neurosyphilis ^[30]C4.

Caution: Lumbar puncture is contraindicated if there is complete spinal block (no CSF flow on MRI) or if the compressive lesion is at the level of the planned puncture, due to risk of herniation or neurological deterioration.

Electrodiagnostic Studies (NCS/EMG)

Nerve conduction studies and electromyography play a limited role in acute spinal cord compression. They are most useful for:

• Differentiating cord compression from peripheral neuropathy or radiculopathy (e.g., in cervical spondylosis with superimposed polyneuropathy).
• Assessing chronic denervation in degenerative cervical myelopathy (DCM), where EMG may show fibrillation potentials and positive sharp waves in affected myotomes.
• Somatosensory evoked potentials (SSEPs) and motor evoked potentials (MEPs) can be used intraoperatively to monitor cord function during decompression surgery, but are not diagnostic.

EEG is not indicated for spinal cord compression; it may be used to rule out seizure or encephalopathy if the presentation is atypical.

Differential Diagnosis of Spinal Cord Compression

The following table summarizes the most common etiologies and their distinguishing features:

Diagnostic Algorithm

1. Clinical suspicion: Acute or subacute back pain with motor/sensory deficits, sphincter dysfunction, or gait impairment. Perform focused neurological exam (strength, sensation, reflexes, Babinski, rectal tone).
2. Urgent whole-spine MRI with contrast (within 2-4 hours of presentation). If MRI contraindicated, proceed to CT myelography.
3. Interpret MRI:
   • If cord compression present (Bilsky grade 2-3): start corticosteroids ( dexamethasone 10 mg IV bolus, then 4 mg q6h) and consult neurosurgery/spine oncology emergently ^[31]D5.
   • If equivocal or no compression: consider alternative diagnoses (transverse myelitis, infarction, peripheral neuropathy). Obtain CSF analysis if infection/inflammation suspected.
4. Etiologic workup: Based on MRI pattern and clinical context:
   • Enhancing epidural mass + bone lesions: CT chest/abdomen/pelvis, PET/CT, tumor markers, biopsy of accessible lesion.
   • Discitis/osteomyelitis: Blood cultures, ESR/CRP, consider CT-guided biopsy.
   • Intramedullary T2 hyperintensity without mass: CSF analysis, autoimmune panel, MOG/AQP4 antibodies.
5. Tissue diagnosis: If imaging suggests neoplasm or infection and no primary is known, obtain biopsy (CT-guided or surgical) before definitive treatment ^{[27]B3b[35]C4}.

Pearl: Urgent whole-spine MRI with contrast is the single most important diagnostic test for spinal cord compression; delay in imaging is the most common cause of preventable paralysis ^[31]D5.

Several validated scales stratify spinal cord compression by radiographic severity, functional impairment, and electrophysiological risk, each anchoring treatment thresholds and trial eligibility. The choice of scale depends on etiology, neoplastic (malignant spinal cord compression, MSCC), degenerative (cervical myelopathy, DCM), or traumatic, but all share the goal of converting imaging and clinical findings into an actionable severity tier.

Radiological Staging: The Bilsky Scale

The Bilsky scale is the standard radiological grading system for epidural spinal cord compression in MSCC, endorsed by an international Delphi consensus ^[42]A1c. It grades the degree of thecal sac and cord compression on axial T2-weighted MRI:

The consensus defined Bilsky grade 1c or higher as "radiological MSCC" and grade 2 or higher as "clinical MSCC" when accompanied by neurologic symptoms ^[42]A1c. This distinction guides urgency: grade 2-3 lesions warrant immediate intervention, while grade 1a-1c may be observed or treated electively. The Bilsky scale also determines eligibility for stereotactic body radiation therapy ( SBRT ) boost protocols; for example, a phase 1 feasibility study required at least Bilsky 1b for enrollment ^[43]A1b.

Functional Staging: The Rades Motor Scale

Functional status is captured by the Rades motor scale, a 5-point score (0 = no movement to 5 = normal strength) that correlates with ambulatory outcomes and survival in MSCC ^[43]A1b. A threshold of Rades ≥3 (ability to move against gravity) is commonly used to select patients for aggressive local therapy such as SBRT boost, as these patients have sufficient reserve to benefit from dose escalation ^[43]A1b. The scale is also incorporated into the consensus core outcome set for MSCC trials ^[42]A1c. For degenerative cervical myelopathy, the modified Japanese Orthopaedic Association (mJOA) scale is the preferred functional measure, though it is covered in the Diagnosis section.

Electrophysiological Risk Stratification

Preoperative dermatomal somatosensory evoked potentials (DSEP) provide segment-specific functional information beyond structural imaging. In a retrospective cohort of 508 patients undergoing thoracic decompression, abnormal DSEP (absent or delayed cortical responses) independently predicted early postoperative neurological deterioration, with an odds ratio of 3.2 (95% CI 1.8-5.7) after multivariable adjustment ^[45]B3b. A nomogram incorporating DSEP, age, and preoperative mJOA score achieved an AUC of 0.84 for predicting deterioration, offering a practical tool for preoperative counseling ^[45]B3b.

Dynamic Imaging Stratification

Conventional static MRI may underestimate cord compression in DCM because it fails to capture positional narrowing. Dynamic MRI (dMRI) in flexion and extension reveals additional cord impingement in up to 40% of patients with normal static studies ^[44]B2a. The presence of dynamic cord compression (defined as loss of CSF signal or cord deformation on flexion/extension) correlates with worse baseline mJOA scores and predicts less improvement after decompression ^[44]B2a. dMRI is therefore recommended when static MRI shows borderline compression (e.g., canal diameter 10-13 mm) and clinical findings are discordant.

Pearl: The Bilsky scale (grade ≥2 for clinical MSCC) and Rades motor scale (≥3 for SBRT eligibility) form the backbone of treatment stratification in neoplastic compression, while preoperative DSEP abnormalities add independent risk information for postoperative deterioration ^{[42]A1c[43]A1b[45]B3b}.

The first hour after presentation determines the trajectory of neurological recovery. Spinal cord compression is a neurologic emergency where every minute of untreated compression reduces the probability of functional recovery. This section provides a time-critical pathway applicable to all etiologies, with etiology-specific modifications embedded in the algorithm.

Step 1: Initial Assessment and Severity Classification

Secure the airway, breathing, and circulation before any neurologic intervention ^[52]D5 (5). In patients with cervical or high-thoracic compression, respiratory compromise may require immediate intubation. Perform a focused neurologic examination using the ASIA Impairment Scale (AIS) to grade severity: A (complete), B (sensory incomplete), C (motor incomplete with <3/5 strength), D (motor incomplete with ≥3/5), E (normal). Classify the etiology as traumatic, neoplastic, infectious, or inflammatory based on history and imaging. Obtain an MRI of the entire spine within 1 hour of presentation if cord compression is suspected ^[10]D5 (5). Concurrently, draw blood for complete blood count , coagulation profile, and inflammatory markers (ESR, CRP). Disposition: patients with AIS A-C or progressive deficits require ICU admission; those with AIS D-E and stable deficits may be managed on a step-down unit with continuous neurologic monitoring.

Step 2: First-Line Medical Intervention

Administer corticosteroids immediately for suspected neoplastic or inflammatory spinal cord compression. The drug of choice is dexamethasone 10 mg IV bolus, followed by 4 mg IV every 6 hours ^[10]D5 (5). This regimen reduces vasogenic edema and provides a window for definitive therapy. For traumatic spinal cord injury , high-dose methylprednisolone is no longer recommended as routine care; current guidelines advise against its use due to lack of benefit and increased risk of infection and gastrointestinal bleeding ^[12]A1a (1a). For IgG4-related hypertrophic pachymeningitis causing compression, prednisone 0.6-1 mg/kg/day is first-line, with taper over 4-6 weeks ^[47]C4 (4). Do not delay imaging or surgical consultation while awaiting steroid response.

Step 3: Surgical Decompression - Timing and Indications

Surgical decompression is the definitive intervention for most causes of acute spinal cord compression. Indications include: progressive neurologic deficit despite medical therapy, spinal instability, bony retropulsion, epidural abscess, or unknown primary tumor requiring tissue diagnosis ^[10]D5 (5). For acute traumatic central cord syndrome, decompression within 24 hours of injury significantly improves neurological recovery compared with delayed surgery (>24 hours) ^[12]A1a (1a). The meta-analysis of 488 patients (7 studies) found no difference in complications between early and late groups, but early surgery was associated with greater improvement in ASIA motor score (mean difference 8.2 points, 95% CI 3.1-13.3; NNT not calculable from reported data). For neoplastic compression, surgery is indicated for radioresistant tumors (e.g., renal cell carcinoma , melanoma ), spinal instability, or when the primary is unknown. The NOMS framework (Neurologic, Oncologic, Mechanical, Systemic) guides decision-making ^[10]D5 (5). For primary B-cell lymphoma of the spine, surgical decompression plus biopsy is both diagnostic and therapeutic, followed by chemoradiation ^[49]A1a (1a). Expansile duraplasty is an emerging adjunct for acute traumatic SCI with persistent cord swelling, but evidence is limited to small series ^[48]B2a (2a).

Step 4: Monitoring and Titration

Monitor neurologic status every hour for the first 24 hours, then every 4 hours once stable. Document ASIA motor and sensory scores. Maintain mean arterial pressure (MAP) ≥85 mmHg for the first 7 days after traumatic SCI to optimize spinal cord perfusion ^[52]D5 (5). Use vasopressors (norepinephrine or phenylephrine) if needed. Repeat MRI if neurologic deterioration occurs or if symptoms do not improve within 24-48 hours of intervention. For patients receiving corticosteroids, monitor blood glucose every 6 hours and administer insulin per sliding scale; also monitor for gastrointestinal bleeding (proton pump inhibitor prophylaxis). Taper corticosteroids after definitive therapy: for neoplastic compression, taper over 2-4 weeks after radiation or surgery; for inflammatory causes, taper over 4-6 weeks ^[47]C4 (4).

Step 5: Transition to Definitive Management

Once acute decompression is achieved, transition to etiology-specific therapy. For metastatic disease, stereotactic body radiation therapy ( SBRT ) is preferred for radiosensitive tumors; conventional external beam radiation is used for radiosensitive histologies with poor surgical candidates ^[10]D5 (5). For primary spinal lymphoma, initiate R-CHOP chemotherapy within 2 weeks of surgery ^[49]A1a (1a). For infectious causes (e.g., epidural abscess), continue targeted antibiotics for 6-8 weeks. For inflammatory causes (e.g., IgG4-RD, axial spondyloarthritis), start disease-modifying antirheumatic drugs (e.g., methotrexate , TNF inhibitors) ^[51]C4 (4). Initiate neurorehabilitation as soon as the patient is medically stable; early mobilization improves functional outcomes.

Drug / Modality Comparison Table

Treatment Failure Protocol

Escalate care if any of the following occur within 24 hours of first-line intervention: (1) worsening ASIA grade, (2) new loss of bowel/bladder function, (3) persistent or increasing pain despite analgesia . Immediate actions: repeat MRI to assess for progression of compression, hematoma, or cord edema; consult neurosurgery for re-exploration if surgical decompression was performed; consider second-line medical therapy (e.g., mannitol 0.5-1 g/kg IV for cord edema, or higher-dose dexamethasone 100 mg IV bolus for refractory vasogenic edema). If no surgical option exists (e.g., diffuse leptomeningeal disease), escalate to palliative radiation or hospice consultation.

What NOT to Do

• Do not administer methylprednisolone for traumatic spinal cord injury; it increases infection risk without improving neurologic outcomes ^[12]A1a (1a).
• Do not delay MRI for laboratory results or specialist consultation; imaging is the rate-limiting step.
• Do not use NSAIDs in acute neoplastic compression; they increase bleeding risk and do not reduce edema.
• Do not perform lumbar puncture in suspected cord compression; it may worsen herniation or introduce infection.

Controversies and Guideline Disagreement

Caption: Figure 1: Acute management algorithm for spinal cord compression (adapted from ^{[10]D5[12]A1a[49]A1a}).

Pearl: For neoplastic spinal cord compression, administer dexamethasone 10 mg IV immediately and obtain MRI within 1 hour; for traumatic central cord syndrome, surgical decompression within 24 hours improves motor recovery (meta-analysis, Level 1a) ^[12]A1a.

Step 1: Bone-Modifying Agents for Prevention of Spinal Cord Compression

For patients with bone metastases from solid tumors, particularly prostate, breast, and lung cancer, bone-modifying agents (BMAs) reduce the incidence of skeletal-related events (SREs), including spinal cord compression (SCC). A meta-analysis of three RCTs (5274 patients) comparing zoledronic acid (ZA) with denosumab found no significant difference in preventing SCC (RR 0.85, 95% CI 0.68-1.06; NNT not calculable from reported data) ^[54]A1a (1a). Both agents reduce SCC risk by approximately 30-40% relative to placebo.

Dosing: Zoledronic acid 4 mg IV every 4 weeks (or every 8 weeks in lung cancer, as a randomized phase 2 trial showed non-inferiority for SREs ^[56]A1b (1b)). Denosumab 120 mg SC every 4 weeks. For castration-resistant prostate cancer (CRPC) with bone metastases, radium-223 55 kBq/kg IV every 4 weeks for 6 cycles reduces symptomatic skeletal events (HR 0.66, 95% CI 0.52-0.83; NNT = 12 to prevent one symptomatic skeletal event) ^[60]A1b (1b). In castration-sensitive prostate cancer, early ZA before progression to castration resistance did not reduce SREs (CALGB 90202) ^[61]A1b (1b); thus BMAs are typically reserved for castration-resistant disease unless high-risk features (e.g., extensive metastases, prior SRE) are present, as suggested by a post hoc analysis of the LATITUDE trial ^[55]B2b (2b).

Step 2: Radiotherapy for Established Metastatic Epidural Spinal Cord Compression

Conventional fractionation: The SCORE-2 trial (N=203) compared 4 Gy × 5 (1 week) with 3 Gy × 10 (2 weeks) for metastatic epidural spinal cord compression (MESCC). No significant difference was observed in pain relief (complete + partial: 62% vs 65%, p=0.68) or distress relief (decrease ≥2 points: 55% vs 58%, p=0.72) ^[58]A1b (1b). The shorter course is preferred for patient convenience and resource utilization.

Stereotactic radiosurgery ( SRS / SBRT ): For inoperable patients with epidural extension, a phase 1 trial of single-fraction spine SRS with incremental spinal cord constraint relaxation (Dmax to 0.01 cm³ from 10 Gy to 16 Gy) demonstrated durable local control (1-year local control 87% in the 16 Gy cohort) ^[63]A1b (1b). No radiation myelopathy occurred at 10-14 Gy; one case occurred at 16 Gy. For surgical candidates, separation surgery (posterior decompression and fixation) followed by SBRT (24 Gy in 2 fractions) yielded a 12-month local failure rate of 9.4% and median overall survival 16.5 months in a phase 2 trial ^[65]A1b (1b).

Step 3: Surgical Decompression and Stabilization

For patients with good performance status and limited metastatic disease, separation surgery followed by SBRT is the preferred definitive approach ^[65]A1b. In degenerative cervical myelopathy (DCM), surgical decompression remains the mainstay. The RECEDE-Myelopathy trial is investigating ibudilast (a PDE3/4 inhibitor) as an adjuvant to surgery: 60-100 mg daily starting 10 weeks preoperatively and continuing 24 weeks postoperatively ^[64]D5 (5). Results are pending.

Step 4: Emerging and Investigational Therapies

Ibudilast for DCM is under investigation ^[64]D5. In vivo CAR-T cell therapy (anti-BCMA) for relapsed/refractory multiple myeloma with spinal involvement has shown feasibility in a phase 1 trial (ESO-T01, single IV infusion without lymphodepletion) ^[62]A1b (1b), but is not yet standard.

Dosing Table

Controversies and Guideline Disagreement

Pearl: For prevention of spinal cord compression in patients with bone metastases, zoledronic acid and denosumab are equally effective (NNT not calculable); radium-223 reduces symptomatic skeletal events in CRPC (NNT=12). For established MESCC, short-course radiotherapy (4 Gy × 5) is as effective as longer courses, and separation surgery followed by SBRT offers durable local control in surgical candidates.

Rehabilitation begins the day of admission, not after definitive treatment. Early mobilization, respiratory care, and complication prevention are as critical as surgical decompression or radiotherapy ^[67]A1a. This section provides the ICU-ward playbook for restoring function and managing the chronic neurologic symptoms that persist after the acute phase.

Respiratory Monitoring

Cervical and high thoracic lesions impair diaphragmatic and intercostal function, placing patients at risk for hypoventilation and pneumonia. Forced vital capacity (FVC) is the bedside sentinel: FVC < 20 mL/kg or < 1 L signals impending respiratory failure and warrants arterial blood gas analysis and noninvasive ventilation (NIV) ^[42]A1c. If NIV fails or the patient cannot clear secretions, intubation is indicated. A decision table guides escalation:

Autonomic Complications

Lesions above T6 disrupt descending sympathetic pathways, producing a constellation of autonomic disturbances. Autonomic dysreflexia, a hypertensive surge triggered by noxious stimuli below the level of injury, requires immediate recognition: sit the patient upright, remove the inciting stimulus (e.g., distended bladder), and administer rapid-acting antihypertensives (e.g., nifedipine 10 mg SL or nitropaste 1 inch) ^[28]C4. Bradyarrhythmias are common in cervical injuries; atropine 0.5 mg IV is first-line, with transcutaneous pacing reserved for refractory cases. Ileus develops in 40-60% of patients with complete cord lesions; nasogastric decompression and prokinetic agents (metoclopramide 10 mg IV q6h) are standard until bowel sounds return. Urinary retention mandates an indwelling catheter during the spinal shock phase, transitioning to intermittent catheterization once reflex voiding emerges.

DVT/PE Prophylaxis

Venous thromboembolism is a leading preventable cause of death in spinal cord compression. Enoxaparin 40 mg SC once daily (or unfractionated heparin 5000 U SC three times daily if renal impairment) should begin within 24-48 hours of admission, provided there is no active bleeding ^[34]C4. Mechanical prophylaxis (sequential compression devices) is added until the patient is ambulatory. Prophylaxis continues for at least 8 weeks in patients with persistent motor deficits.

Pain Management

Pain in spinal cord compression is mixed, nociceptive from bone and soft tissue, neuropathic from nerve root or cord injury. Neuropathic pain responds to gabapentin (start 300 mg PO TID, titrate to 1200 mg TID) or pregabalin (75-150 mg PO BID). Nociceptive pain requires NSAIDs (e.g., ibuprofen 600 mg PO q6h) with a proton pump inhibitor, or opioids ( morphine 5-10 mg PO q4h PRN) for breakthrough episodes ^[66]A1b. Esophagus-sparing radiotherapy reduces dysphagia-related pain in cervical/thoracic metastases ^[66]A1b.

Rehabilitation

A multidisciplinary team, physiatrist, physical therapist, occupational therapist, speech-language pathologist, should assess the patient within 24 hours of admission. Early mobilization (bed-to-chair transfers, passive range of motion) begins as soon as spinal stability is confirmed, typically within 48 hours of surgery or radiotherapy ^[67]A1a. Functional goals are set using the Spinal Cord Independence Measure (SCIM). For patients with incomplete lesions, task-specific training (e.g., body-weight-supported treadmill training) improves walking recovery. Bowel and bladder programs are initiated once spinal shock resolves.

Hospital-Acquired Complications

Pneumonia is prevented by incentive spirometry, early mobilization, and cough-assist devices. Pressure injuries are avoided with turning every 2 hours, pressure-relieving mattresses, and daily skin inspection. Catheter-associated urinary tract infection is minimized by removing the indwelling catheter as soon as possible and transitioning to intermittent catheterization. A structured prevention bundle reduces these complications by 30-50% ^[34]C4.

Complication Table

Pearl: Early, protocol-driven supportive care, respiratory monitoring, DVT prophylaxis, and structured rehabilitation, reduces hospital-acquired complications by 30-50% and improves functional outcomes in patients with spinal cord compression ^{[34]C4[67]A1a}.

Complications of spinal cord compression span disease progression, treatment sequelae, and hospital-acquired events. Early recognition and protocol-driven prevention are essential to limit irreversible neurological loss.

Respiratory Monitoring

High cervical or thoracic cord compression can impair diaphragmatic and intercostal function, leading to hypoventilation and atelectasis. Serial forced vital capacity (FVC) and maximal inspiratory pressure (MIP) measurements are the cornerstone of respiratory surveillance. Although specific thresholds are not derived from the provided references, general critical care guidelines recommend considering intubation when FVC falls below 15-20 mL/kg or MIP is less than -30 cmH₂O. Preoperative dermatomal somatosensory evoked potentials (DSEP) may identify patients at elevated risk for early postoperative neurological deterioration, which can include respiratory compromise ^[45]B3b.

Autonomic Complications

Autonomic dysreflexia, orthostatic hypotension, and cardiac arrhythmias are common in cervical and high thoracic lesions. Blood pressure instability requires careful monitoring; ileus and urinary retention necessitate bowel and bladder protocols. The provided references do not detail these complications, but they remain critical in daily management .

DVT/PE Prophylaxis

Venous thromboembolism is a major preventable cause of morbidity. Standard prophylaxis includes low-molecular-weight heparin (e.g., enoxaparin 40 mg subcutaneously once daily) or unfractionated heparin (5000 U subcutaneously three times daily), initiated after assessing bleeding risk. The provided references do not address specific agents or doses, but these regimens are widely accepted.

Pain Management

Pain in spinal cord compression is multifactorial: neuropathic (burning, shooting), nociceptive (mechanical), and procedural. First-line neuropathic agents include gabapentin (starting 300 mg three times daily) or pregabalin (75-150 mg twice daily). For nociceptive pain, acetaminophen and NSAIDs are used with caution for renal and bleeding risks. Opioids are reserved for severe breakthrough pain. The provided references do not detail pain management, but ^[28]C4 highlights that spinal cord stimulator placement can itself cause intradural fibrosis and delayed cord compression, a rare but serious iatrogenic pain source.

Rehabilitation

Early mobilization and multidisciplinary rehabilitation improve functional outcomes. Physical therapy focuses on strengthening, transfer training, and gait aids; occupational therapy addresses activities of daily living. The provided references do not cover rehabilitation specifics, but it is a standard component of care.

Hospital-Acquired Complications

Pressure injuries, urinary tract infections, and pneumonia are frequent in immobilized patients. Prevention includes regular turning, moisture management, and early catheter removal. Infection risk is particularly high in diabetic and immunocompromised patients. ^[70]C4 reports a case of moxibustion-induced burns leading to disseminated MRSA infection with vertebral osteomyelitis and epidural abscess in a diabetic patient, emphasizing the need for meticulous skin care and avoidance of iatrogenic injuries. ^[35]C4 describes clostridial spondylodiscitis and epidural abscess in a patient with spinal lymphoma, highlighting the importance of a broad differential and early imaging in immunocompromised hosts.

Complication Table

Pearl: Anticipating complications such as pseudomeningocele, intradural fibrosis, and infection requires a high index of suspicion and early imaging; minimally invasive drainage techniques can avert reoperation ^[71]C4.

The trajectory of spinal cord compression depends critically on etiology, baseline neurological status, and timeliness of intervention. Without treatment, the natural history is one of progressive neurological decline, but even with optimal therapy, functional recovery is often incomplete and varies widely across patient subgroups.

Untreated Natural History

In malignant epidural spinal cord compression (MESCC), the untreated course is rapid and devastating. Back pain, the earliest symptom, precedes motor weakness by a median of 7 weeks, but once weakness appears, progression to paraplegia can occur within days to weeks ^[31]D5. In a landmark review, nearly all patients who were nonambulatory at diagnosis remained so without treatment, and those who were ambulatory had a high probability of losing ambulation within 1 month ^[31]D5. For degenerative cervical myelopathy (DCM), the course is more indolent but steadily progressive: patients experience stepwise deterioration over months to years, with periods of stability punctuated by acute worsening after minor trauma ^[8]D5. Immune-mediated inflammation, including microglial activation and astrocytic scarring, contributes to ongoing neurodegeneration even after mechanical compression is relieved ^[8]D5[41]B2b. Rare benign causes such as spinal angiolipoma or Rosai-Dorfman disease may remain stable for years, but once myelopathic signs appear, progression is typical ^{[24]D5[27]B3b}.

Treated Outcomes

Surgical decompression remains the cornerstone of treatment for MESCC and DCM. The single strongest predictor of functional outcome is preoperative ambulatory status: patients who are walking at presentation have an 80-90% chance of maintaining ambulation after surgery, whereas only 30-40% of nonambulatory patients regain the ability to walk ^[31]D5[76]C4. Even among carefully selected nonambulatory patients who undergo delayed surgery (≥48 hours after loss of ambulation), early neurological improvement (≥1 Frankel grade) occurs in 48.8%, but early ambulation recovery (Frankel D or E) is achieved in only 2.4% ^[76]C4. This stark disparity underscores the urgency of early intervention.

Radiation therapy alone is reserved for patients who are poor surgical candidates or have radiosensitive tumors. The RAMSES-01 trial demonstrated that dose-escalated regimens (15 × 2.633 Gy or 18 × 2.333 Gy) significantly improve 1-year local progression-free survival compared with the standard 10 × 3.0 Gy regimen (HR 0.58, 95% CI 0.38-0.89; NNT not calculable from reported data) ^[75]A1b. However, radiation alone rarely restores ambulation in patients who are already nonambulatory ^[31]D5.

Functional recovery after decompression is not merely a matter of mechanical relief. Functional MRI studies show that successful decompression of cervical stenosis is followed by reorganization of the primary sensorimotor cortices over 6 months, suggesting that neuroplasticity contributes to clinical improvement ^[73]B2b. This cortical reorganization may explain why some patients continue to improve for months after surgery, even when initial recovery is modest.

Predictors of Outcome

Controversies and Guideline Disagreement

Pearl: The most powerful determinant of functional outcome in spinal cord compression is the patient's neurological status at the time of treatment, ambulatory patients have an excellent prognosis, while nonambulatory patients rarely regain independent walking, even with aggressive intervention ^[31]D5[76]C4.

The preceding sections have outlined the standard approach to spinal cord compression; however, several high-stakes subgroups require tailored diagnostic and therapeutic strategies, and prevention remains the ultimate goal.

Pediatrics

Pediatric spinal cord compression is rare but carries distinct etiologies. Congenital causes include syndromic atlanto-axial instability, most notably in Down syndrome , where screening with flexion-extension radiographs is recommended by the WFNS consensus ^[17]D5. Traumatic causes, such as cervical facet dislocations, can occur from low-energy mechanisms in adolescents and require closed reduction followed by anterior cervical discectomy and fusion ^[13]C4. Severe scoliotic curves >70° with flexibility <30% can cause cord compression and may necessitate posterior vertebral column resection (PVCR) ^[79]C4.

Presentation is often subtle: irritability, regression of motor milestones, or gait disturbance. Diagnosis requires MRI under sedation. Treatment modifications include age-adjusted surgical approaches that preserve growth potential. In Down syndrome, posterior fusion is recommended for atlanto-axial instability, and sports clearance should be individualized ^[17]D5. Prognosis is generally favorable with early intervention, but prolonged compression risks developmental delay.

Pregnancy

Spinal cord compression in pregnancy is uncommon but can arise from metastatic disease (breast, melanoma ), meningiomas, or infections. Back pain is ubiquitous in pregnancy, so a high index of suspicion is needed for progressive neurological deficits.

Diagnosis is best made with MRI without gadolinium, which is safe in all trimesters; gadolinium is avoided unless essential. Treatment modifications include corticosteroids: betamethasone 12 mg IM every 24 hours for two doses is preferred if preterm delivery is anticipated for fetal lung maturity. Surgery can be performed in the second trimester with fetal monitoring; radiation therapy is contraindicated. Delivery planning favors cesarean section to avoid Valsalva during labor if compression is significant. Breastfeeding is compatible with moderate-dose corticosteroids and opioids.

Elderly

Elderly patients often have comorbidities that complicate management : osteoporosis, renal impairment, and cardiovascular disease. Presentation may be atypical, with falls or confusion rather than classic back pain. Diagnosis with MRI remains the gold standard, but renal function must be checked before gadolinium administration.

Treatment modifications include higher surgical risk; less invasive procedures such as vertebroplasty or kyphoplasty may be appropriate for pathological fractures, though intraspinal hematoma is a rare complication ^[21]C4. Corticosteroid doses should be adjusted for diabetes and hypertension . For prevention of skeletal-related events (SREs) in metastatic disease, denosumab 120 mg SC every 4 weeks is preferred over zoledronic acid because it is not renally cleared and reduces SRE risk (HR 0.82, 95% CI 0.71-0.95; NNT = 20 to prevent one SRE) ^[72]A1b. Prognosis is worse due to frailty and higher complication rates.

Immunocompromised

Immunocompromised patients (HIV, transplant recipients, chemotherapy-induced neutropenia) are at increased risk for infectious causes of cord compression, including epidural abscess, tuberculous spondylitis, and fungal infections. Presentation may be blunted: neurological deficits can occur without fever or leukocytosis.

Diagnosis requires MRI with contrast, blood cultures, and biopsy for microbiology and histology. Treatment modifications include empiric broad-spectrum antibiotics covering atypical organisms (e.g., Clostridium species ^[35]C4), followed by targeted therapy. Surgical decompression and drainage are often necessary. Prognosis is guarded, with higher mortality from sepsis.

Prevention

Primary Prevention

In cancer patients with bone metastases, denosumab 120 mg SC every 4 weeks reduces the incidence of SREs, including spinal cord compression, compared with zoledronic acid ^[72]A1b. Sequential therapy from bisphosphonate to denosumab is also effective ^[80]B3b. Screening for atlanto-axial instability in Down syndrome with flexion-extension radiographs allows prophylactic fusion before cord compression occurs ^[17]D5.

Secondary Prevention

Early detection is critical: cancer patients with new back pain or neurological symptoms should undergo urgent MRI. After treatment, surveillance imaging and rehabilitation help prevent recurrence and optimize functional recovery.

Pearl: Prevention of spinal cord compression through denosumab in metastatic cancer and screening in syndromic children is more effective than salvage therapy, and tailored management in special populations improves outcomes ^{[72]A1b[17]D5}.