14

Cervical spondylotic myelopathy (CSM) is spinal cord dys-function caused by age-related degenerative changes within the spinal column. CSM is reported to be the most common cause of spinal cord dysfunction in individuals older than 55 years of age.1 Frequency of degenerative magnetic resonance imaging (MRI) findings increases linearly with age. Radio-graphic features of CSM are frequently present in asymptomatic adults. Disk degeneration to some degree is present in ~25% of individuals younger than 40, 50% in individuals over 40, and 85% in individuals over 60.^2,3 The clinical signs and symptoms are often vague and may be masked by other concomitant conditions. The exact prevalence of disease is unknown due to the difficulty clinicians have in making an accurate diagnosis in early stages.

Pathophysiology

Mechanical compression of the spinal cord is thought to be the primary pathophysiological mechanism causing cervical myelopathy. The cascade of acquired cervical stenosis begins with loss of intervertebral disk integrity.⁴ With increased age the disk loses its load-bearing capacity and undergoes dehydration and disappearance of the distinction between the nucleus pulposus and the anulus fibrosus. The dysfunctional disk results in an increased load on the uncovertebral processes and alters the load-bearing function of the intervertebral joint.⁵ Alterations in normal biomechanics lead to formation of osteophytes (spondylosis) at margins of the end plates and facet joints. Osteophytes increase the weight-bearing surface area and help stabilize the hypermobile adjacent vertebra caused by disk degeneration.⁶ As the intervertebral disk collapses, the ligamentum flavum thickens and buckles into the spinal canal dorsally.⁷ Collectively these changes can ultimately lead to circumferential stenosis of the spinal canal and increased mechanical pressure on the spinal cord.

Despite the presence of cervical spondylosis, most patients can tolerate narrowing of the canal prior to developing cervical cord impingement. The normal sagittal diameter of the spinal canal is ~17 to 18 mm from C3 to C7, whereas the cord itself measures ~10 mm.⁸ This leaves ~6 to 7 mm of space in the anteroposterior (AP) dimension before there is risk of significant cord compression. Developmental stenosis and ossification of the posterior longitudinal ligament (OPLL) narrows this zone of safety and is correlated with predisposition to development of cervical myelopathy. Individuals with an AP canal diameter of < 13 mm are considered stenotic and at risk for development of CSM.⁹ A cross-sectional area of the spinal cord < 60 mm has been associated with development of signs and symptoms of myelopathy.³ A compression ratio of the AP to the transverse diameter of the spinal cord of less than 40% has also been associated with worse neurological function.¹⁰ Increasing the ratio to greater than 40% or increasing the cross-sectional area to more than 40 mm has a strong predictor of clinical recovery.¹¹

Besides static mechanical cord compression, dynamic cord compression may also contribute to pathogenesis of CSM. The spinal canal can change dimensions with either normal or abnormal segmental mobility of the neck. The AP diameter and volume of the canal are reduced with neck extension.¹² Hyperextension of the neck shingles the lamina and causes buckling of the ligamentum flavum.¹³ A postmortem study demonstrated that involution of the ligamentum flavum can contribute significantly to cervical stenosis in extension and is relieved in a flexed posture.¹⁴ Translation and angulation of the vertebra in flexion-extension can also transiently narrow the spinal canal. With preexisting stenosis and abnormal and excessive motion, there are increased strain and shear forces applied to the spinal cord, potentially causing localized axonal injury.15 Morphological changes to the cord itself occur with flexion and extension. The spinal cord stretches with flexion and shortens and thickens with extension of the cervical spine. Thickening in extension makes the cord more susceptible to pressure from the infolded ligamentum flavum or the lamina. With flexion the stretched cord can impinge against the disk or the vertebral body. These dynamic forces may render the axons more susceptible to secondary injury processes such as ischemia, excitotoxicity, free radical or reactive oxygen species–mediated oxidation, lipid peroxidation, and apoptosis.¹⁶ The majority of abnormal MRI signal changes in patients with myelopathy are visualized on neutral position static MRI scans. The increased utilization of dynamic MRI may provide useful data on the pathomechanics of cervical myelopathy.

Ischemia of the cord has been postulated to have a cumulative effect on the clinical manifestations of cervical myelopathy. Anterior compression can compromise perfusion by tenting transverse arterioles arising from the anterior sulcal arteries leading to ischemia of the anterior horn and lateral column. Posterior cord compression can reduce perfusion to the intramedullary branches of the central gray matter.^17,18 Abnormal biomechanics at a motion segment can also trigger a vasospastic response that can compromise the cord’s intrinsic blood supply.¹⁹ Neurological and histological changes caused by compression, ischemia, and their combination correlate with altered patterns of blood flow within the cervical spinal cord.²⁰ Pathophysiology of cervical myelopathy involves a complex interplay of static and dynamic factors that trigger a cascade of biochemical and molecular changes.

Natural History

Consensus regarding the natural history of CSM remains undefined. Clarke and Robinson published the first natural history study in 1956.²¹ Over time two thirds of their patients experienced deterioration, whereas one third remained unchanged. Seventy-five percent of the patients that deteriorated manifested symptoms in a series of episodes. Twenty percent had a slow steady progression from the onset of symptoms, whereas 5% had rapid onset. The authors concluded that it was uncommon for patients with neurological deficits to undergo spontaneous regression. Lees and Turner concluded that CSM followed a prolonged clinical course with long periods of relatively stable symptoms.²² Correspondingly, Nurick conducted a retrospective assessment that revealed patients may deteriorate early on, but their clinical presentation followed a static course for many years.²³ More recent studies confirm that the majority of patients with CSM deteriorate in neurological function. Sadasivan and colleagues performed a retrospective evaluation that suggests progressive deterioration without stabilization of symptoms as reported by Lees and Turner.²⁴ The mean delay of diagnosis was 6.3 years as a result of vague symptoms, and spontaneous regression did not occur in any case.²⁴ It is difficult to study the true natural history of disease due to the variability in clinical presentation.

Clinical Presentation

Cervical myelopathy is a syndrome of long-tract clinical findings in the upper and lower extremities arising from involvement of the spinal cord. The diagnosis of CSM during the early course of disease may be difficult even with a detailed neurological examination. The signs and symptoms are often subtle and vary depending on the relative degree of spinal cord involvement. Concomitant axial neck pain or radiculopathy or both can also be commonly present. Patients may have coexisting symptomatic nerve root compression that causes radicular symptoms such as pain radiating down the arm. Classically patients with myelopathy present with clumsiness or loss of fine motor skills in their hands. History may reveal diminished strength, inability to manipulate small objects such as coins or buttons, changes in handwriting, or nondermatomal numbness. Patient or family members may note gait instability, difficulty maintaining balance, and a history of falls. Subtle gait disturbance such as difficulty on uneven terrain is often the first physical symptom of CSM.²⁴ The Nurick grading system utilizes the degree of gait abnormality in formulating a disability classification in CSM.²⁵ Bowel and bladder symptoms, if present, typically arise in late stages of the disease and carry a poor prognosis. In an attempt to objectively quantify the degree of functional disability of patients with CSM, the Japanese Orthopaedic Association (JOA) developed a system that includes scoring upper and lower extremity function, sensory disturbance, and bladder function.²⁶

Myelopathy hand is a term coined by Ono and associates that describes a constellation of findings in the upper extremity.¹⁰ The motor examination can be normal or show only subtle signs. It may manifest as diminished dexterity or wasting of the hand intrinsic muscles. Deficiency of adduction can be detected by having patients hold their fingers extended and adducted. If the ulnar two digits drift into abduction and flexion they have a positive finger escape sign. Patients may also demonstrate the grip release sign, when they are unable to make and release a fist more than 20 times in 10 seconds. Myelopathy hand has been associated with spasticity in the lower extremities.¹⁰

Neurological assessment may reveal loss of pain and temperature, vibratory sensation, and proprioception below the level of the lesion depending on the anatomical location of compression.²⁷ The spinothalamic tract involves contralateral pain and temperature several levels below the compression. Posterior column tracts affect ipsilateral position and vibratory sense disturbance and dermatomal sensory loss. Loss of vibratory sensation and proprioception may be the earliest sign of myelopathy and can be detected by tuning fork examination of the great toes and perception of toe or ankle position.

Upright AP, lateral, and flexion-extension plain radiographs are routine in evaluation of patients with CSM. The AP radiograph helps identify degenerative changes in the uncovertebral joint and identifies the presence of scoliosis. The lateral view reveals disk space narrowing, end plate sclerosis, osteophytes, and the presence of spondylolisthesis and helps quantify sagittal plane alignment. The Torg-Pavlov ratio (AP canal diameter:AP vertebral body diameter) can also be measured and if less than 0.8 suggests congenital stenosis.28 Flexion-extension radiographs help determine cervical range of motion and identify ankylosed segments and the presence of instability.

Advanced diagnostic imaging such as MRI is necessary to confirm spinal cord compression in patients that present with myelopathy. This allows direct evaluation of intrinsic abnormalities of the spinal cord. The basic examination consists of sagittal and axial T1- and T2-weighted images. T1-weighted images (T1WI) have better spatial resolution, and fat has increased signal relative to the spinal cord. T2-weighted images (T2WI) tend to accentuate pathology, and cerebrospinal fluid (CSF) has greater signal intensity than the spinal cord. Patients with CSM may demonstrate abnormal intramedullary signal intensity. Mehalic et al²⁹ described a grading scale ranging from 0 (no increase in signal intensity) to 4 (very intense, focal increase in signal). The meaning of signal changes in patients with CSM is debated as to correlation on prognosis or outcome of treatment. The level and location of compression can be noted from the sagittal images. Axial MRI allows for circumferential evaluation of potential compressive structures and to quantify space available for the cord. To guide decompressive surgery the relative contribution of disk, facet joint, and ligamentum flavum causing cord compression can be studied in detail. The axial images also allow for accurate assessment of flattening of the cord and calculation of AP compression ratio discussed earlier. In cases that present with concomitant radiculopathy, foraminal stenosis is best evaluated on oblique views designed to give true cross-sectional views of the foramina. Computed tomographic (CT) myelography is valuable for patients in which MRI is contraindicated or prior cervical surgery precludes adequate visualization. CT can provide an accurate distinction between osseous and soft tissue compression of neural structures and may be preferable to MRI when seeking detailed resolution of osseous anatomy.

Treatment

Moderate to severe myelopathy is typically considered a surgical disorder. A trial of nonoperative treatment and observation may be reasonable in asymptomatic patients with cord compression, those with mild myelopathy, or if the risk of surgery is unacceptable due to medical comorbidities. Treatment of asymptomatic patients with evidence of cervical cord compression on MRI has not completely been elucidated. Patients who are asymptomatic or present with mild myelopathy may have a significantly prolonged course as discussed in the natural history section. Nonoperative treatment may include intermittent immobilization, antiinflammatory medications, and isometric flexion exercises to help reduce neural irritation. In patients with concomitant radicular symptoms, epidural steroids and cervical traction may provide relief. If signal changes on MRI could effectively predict outcome of conservative measures, patients may be better advised on whether it is prudent to postpone surgery. If nonoperative treatment is decided on, close and careful follow-up is recommended.

In light of the uncertain natural history and lack of clear prognostic factors the surgeon and patient together may decide for surgical intervention. The primary goal of surgical intervention is to halt the progression of disease. Improvement of symptoms or function occurs frequently but is a secondary goal and has been associated with the severity of myelopathy at the time of intervention.³⁰ When considering surgery, many options exist, including surgical approach, type of hardware, or choice of graft material. Irrespective of technique, the goal of treatment remains wide decompression of the spinal cord and affected nerve roots. Decompression can be achieved via either anterior or posterior approaches. Patients with one or two levels of canal stenosis and kyphotic alignment are often best approached anteriorly. Three or more levels of pathology may be addressed from either a multilevel anterior procedure or a posterior approach. Cervical lordosis is a prerequisite for posterior-based decompression and in the setting of kyphosis may lead to failure and progression of deformity.³¹ The posterior approach is often useful in patients with multilevel acquired stenosis, congenital stenosis, or the presence of OPLL.

Controversy exists as to the optimal timing and indications for surgical intervention. Often we attempt to correlate our history and physical examination findings with advanced diagnostic imaging for guidance in formulating a treatment plan. Abnormal MRI signal changes appear to correlate with the degree of histopathological alteration of the spinal cord.31,32 The clear prognostic value of MRI signal changes found in patients with either symptomatic or asymptomatic CSM remains controversial. To address this controversy we have attempted to provide a comprehensive review of the current evidence-based literature on this topic.

Meaning of MRI Signal Changes

Level I Data

There are no level I data published regarding this topic.

Level II Data

Yukawa et al³³ prospectively enrolled 104 patients with compressive myelopathy who were treated with cervical expansile laminoplasty. The diagnosis was CSM in 74 patients and OPLL with symptoms of myelopathy in 20 patients. They set out to study if postoperative MRI signal alteration reflected the severity of myelopathy and surgical outcome. MRI was performed preoperatively and an average of 39.7 months (minimum 12 months, range 12 to 90 months) postoperatively. They graded the amount of increased signal intensity on T2WI on a scale of 0 to 3. Signal changes on sagittal T1WI were not studied due to their low occurrence in the study population. The clinical severity of myelopathy was quantified using the JOA score. Increased signal intensity was present in 83% of cases preoperatively and 70% postoperatively. Patients with increased signal intensity on preoperative imaging were older and had longer duration of symptoms and improved less after surgery than those without signal change. The correlation between alteration in signal intensity from pre- to postoperative MRI and surgical outcomes/recovery rate did not reach statistical significance. Whether the signal intensity improved, worsened, or remained unchanged had no significant difference on postoperative clinical symptoms or surgical results.

Fernández et al³⁴ conducted a prospective case series study that included 67 patients with cervical cord compression with a mean follow-up of 39 months. The aim of the study was to determine if T2WI findings could determine the prognosis of disease. The surgical technique depended on the compression pattern. MRI was performed 3 months or less before surgery. Functional status was evaluated using the JAO scale pre- and postoperatively. Their results found that (1) focal changes on T2WI did not indicate poor functional recovery, (2) low-intensity changes on T1WI plus high intensity on T2WI had poor functional recovery after surgery, (3) multi-segmental high intensity on T2WI was a sensitive reflection of poor functional recovery. The greater the high-intensity changes on T2WI the higher the probability of irreversible neuronal loss. They felt multisegmental high signal intensity changes noted on T2WI may prove to be more sensitive than changes noted on T1WI because they were present in a larger group of their patients (21%) who had poor outcome.

Shimomura et al³⁵ prospectively enrolled 56 patients with mild myelopathy and treated them conservatively. They evaluated various potential prognostic factors, including presence of high-intensity area within the cord on T2WIs. Their analysis showed that the presence of signal change did not affect the clinical condition. The only factor that correlated with deteriorating clinical condition was circumferential cord compression on axial MRI. The signal abnormality was assumed to vary from acute edema to chronic myelomalacia. They concluded that high signal intensity area did not predict prognosis in mild forms of CSM.

Bednarik et al³⁶ reported on a group of 199 patients with MRI signal changes without clear clinical signs of myelopathy. They performed a prospective cohort study with a minimum of 2-year (range 2 to 12 years) follow-up recording various demographic, clinical, imaging, and electrophysiological parameters. They developed a predictive model to determine whether presymptomatic spondylotic cord compression would progress to symptomatic CSM. They performed a detailed clinical examination every 6 months for 2 years, then annually. In addition, patients were instructed about possible signs and symptoms and to follow up sooner if they suspected progression to myelopathy. The functional status of the patients was graded using the JOA scale and Nurick score. The primary end point of the study was a detection of clinical signs or symptoms of myelopathy. The results demonstrated that 25% of the patients progressed to symptomatic myelopathy within 4 years. The risk of early symptomatic myelopathy (≤ 12 months) involved 8% of the cases and was predicted by the presence of clinically symptomatic radiculopathy and abnormal somatosensory and motor evoked potentials. MRI intramedullary hyperintensity predicted later (> 12 months) development of CSM. Increased signal on T2WI indicated edema, inflammation, vascular ischemia, gliosis, or myelomalacia. They recommend electrophysiological evaluation in MRI-documented asymptomatic patients and in patients presenting with radiculopathy and back pain. They concluded that compression detected on MRI in presymptomatic patients was generally benign with a low risk of progression to CSM within 4 years, and less frequent clinical follow-up may be sufficient.

Mastronardi et al³⁷ prospectively followed 47 patients that underwent anterior decompression and fusion for progressive CSM. Patients were assessed with MRI preoperatively, intraoperatively (iMRI), and at a minimum 6-month follow-up. Severity of myelopathy was rated using the Nurick scale and JOA classification. They graded the signal intensity changes on MRI using the Mehalic grading system.29 Preoperative MRI showed intramedullary signal changes in 37 patients (79%). In 23 of those patients (62%) abnormal signals were present on both T1- and T2WI, whereas 14 cases (38%) had changes on T2WI alone. In 12 of the 23 cases (52%) regression of hyperintensity on T2WI were observed. Four of these patients had regression of hyperintensity noted during the intra-operative MRI. They concluded that in patients with CSM, hyperintensity on T2WI images was reversible, whereas hypointensity on T1WI was irreversible and therefore carried a worse prognosis. Clinical outcomes were related to the degree of intensity of signal alteration on preoperative MRI. Very intense signals (grade 3 or 4) showed better prognosis than slight or moderate intensity (grade 1 or 2). They did not find a correlation between times of signal intensity recovery and effect on overall outcome.³⁷

Suri and colleagues conducted a prospective study of 146 consecutive patients with CSM operated on during a 2-year period. The objective of the study was to correlate MRI findings with clinical presentation, prognosis, and postoperative outcome. They recorded clinical parameters and radiographic findings and graded their functional disability using the Nurick scale. Patients were assessed preoperatively and postoperatively at 3- and 6-month follow-up. Postoperative MRI was done in all patients that had preoperative MRI signal changes (36.4%). They found no significant difference in clinical presentations in patients with and without intramedullary signal changes (ISCs). They did find a correlation between age, duration of symptoms, number of prolapsed disks, preoperative ISCs, residual compression, and postoperative regression/persistence of ISCs. Patients without ISCs and those with ISCs on T2WI alone had better clinical outcome than those with ISC on both T1- (hypointensity) and T2- (hyperintensity) weighted images who had a poor prognosis. There was no significant correlation between clinical and radiographic variables with regression of ISCs.³⁸

Level III

There are several retrospective case series and case-control studies reported in the literature trying to determine a correlation between MRI signal changes and prognosis or outcome of surgery. Takahashi et al³⁹ reported that high signal intensity reflected myelomalacia or cord gliosis secondary to longstanding compression and indicated a poor prognosis. Morio et al⁴⁰ described three patterns of cord signal intensity on T1-and T2WI, respectively: normal/normal, normal/high, low/high. The postoperative recovery in patients with preoperative low/high signal was inferior to the patients in the normal/high group. Low-signal intensity changes on T1WI reflected pathologically irreversible changes and predicted poor prognosis. The authors speculated that the high-intensity changes on T2WI indicated a broad spectrum of myelopathic pathologies and reflected a broad spectrum of recuperative potentials. Ramanauskas et al⁴¹ divided myelomalacia into three stages: the early stage reflected cord edema, the intermediate stage reflected cystic necrosis of the central gray matter after prolonged cord edema, and the late stage represented cavity formation or syrinx. They reported that early and intermediate stages showed high signal intensity on T2WI sequences alone, whereas in the late stage there was also low-signal intensity on T1WI sequences that indicated poor prognosis. Mehalic et al²⁹ concluded that high-signal changes on T2WI sequences was nonspecific and indicated edema, inflammation, vascular ischemia, gliosis, or myelomalacia and may be reversible postoperatively. Al-Mefty et al⁴² reported that low-signal changes on T1WI sequences along with high signal intensity on T2WI indicated cystic necrosis or secondary syrinx formation. Papadopoulos et al⁴³ concluded patients with focal areas of high signal intensity would have the same surgical outcome as those without intramedullary high signal. Along with others they felt that multisegmental areas of high signal intensity on T2WI carried a poor surgical outcome.^34,43,44 We summarize select pertinent studies in Table 14.1 for review.

Summary of Data

The literature shows conflicting information regarding the significance of increased signal intensity seen on T2WI alone as it relates to predicting prognosis. In most current prospective studies available some authors report increased focal signal intensity on T2WI to have poor prognosis after surgery,³⁷ whereas others have found no such relationship^33,35 (Fig. 14.1). Multisegmental high signal intensity on T2WI does appear to have consistent support in predicting a poor result and functional recovery^34,44 (Fig. 14.2). Multiple studies also agree that the combination of increased signal on T2WI images and decreased signal on T1WI images carries a poor prognosis and is associated with irreversible injury^34,37,38 (Fig. 14.3). The reversal of high signal intensity changes on postoperative T2WI has conflicting data regarding outcome.^29,38,39,45 The predictive role of MRI signal changes in asymptomatic patients remains unclear. Current prospective data suggest increased signal intensity noted on T2WI in presymptomatic patients is generally benign³⁶ and even in patients with mild myelopathy may not predict prognosis.³⁵

Related posts:

Clearing Cervical Spine Injuries: MRI, Dynamic X-rays, CT Surgery for Axial Back Pain: ALIF versus PLIF or TLIF Avoidance of Cervical Adjacent Segment Disease Recurrent Lumbar Disk Herniation: Repeat Diskectomy versus Fusion Spine Infections: Medical versus Surgical Treatment Options Vertebral Compression Fractures: Percutaneous Vertebral Augmentation

Stay updated, free articles. Join our Telegram channel

Join