Subscription is FREE for qualified healthcare professionals in the US.
11 Articles in Volume 10, Issue #9
Activated Glia: Targets for the Treatment of Neuropathic Pain
Acute Herpes Zoster Neuritis and Postherpetic Neuralgia
Acute Treatment of Cluster Headache
Chronic Overuse Sports Injuries in the Adolescent/Pediatric Population
Clinical Recognition of Central Abnormal Neuroplasticity
H-Wave® Stimulation: A Novel Approach In Electromedicine
Homeopathy Enters Contemporary Pain Practice
Immune-modulating Effects of Therapeutic Laser
Pain and Addiction: Words, Meanings, and Actions in the Age of the DSM-5
Partial Plantar Fasciectomy With Autologous Platelet Concentrate
Tethered Spinal Cord Syndrome: Pathophysiology and Radiologic Diagnosis

Tethered Spinal Cord Syndrome: Pathophysiology and Radiologic Diagnosis

The diagnosis of Tethered Cord Syndrome (TCS) relies primarily on clinical and radiologic criteria. Similar to a variety of other spinal pathologies, TCS may manifest as diffuse or localized lumbar pain, lower extremity motor or sensory deficits, muscular atrophy, and urinary incontinence. TCS has been broadly sub-categorized as either primary/childhood or acquired/adult-onset. Recent research efforts have focused on at unraveling the underlying pathological mechanisms that differentiate these two forms of TCS. While positive-contrast myelography and computerized tomography (CT) with intrathecal contrast have previously been used to diagnose intrinsic spinal cord pathology, magnetic resonance (MR) imaging is now considered the imaging modality of choice due its ability to portray anatomic details in high resolution. While a thickened filum terminale and a low-lying conus medullaris have classically characterized TCS on MR imaging, observational studies have revealed that a significant population of TCS patients may have atypical radiographic findings. In this paper, we provide an updated review on the pathophysiology behind TCS and its characterization on MR imaging.

A comprehensive literature search was performed using the PubMed and Google Scholar database for all journal articles published prior to and including July, 2010. Key words used in this search were tethered cord syndrome, MR imaging, pathophysiology, filum terminale, conus medullaris, and diagnosis. All terms were searched alone or in combination. The articles selected were based upon the quality of the scientific investigations and included clinical case studies, retrospective analysis analyses and experimental studies using animal models. All papers meeting these criteria and relevant to either the pathophysiology or the radiologic diagnosis of TCS were included in this report.

The purpose of this review is to report the experimental and clinical data that were aimed towards uncovering the pathophysiology of TCS. Similarly, the diagnostic findings of TCS on MR imaging will also be reviewed using post-mortem anatomic studies, patient-based observations and retrospective analyses. The results reported will include a brief discussion of caudal spinal cord development followed by an etiological description of TCS in the adult and pediatric populations. Experimental studies describing the pathologic basis for, precipitation of, and the events leading to neuronal damage in TCS will also be discussed. Finally, we report the various MR imaging findings of TCS and their clinical significance.

Embryology of the Caudal Spinal Cord

Development of the caudal spinal cord is initiated as the neural tube elongates. The caudal portion of the neural tube, also termed the “caudal neurpore,” ultimately forms the CM, the ventriculus terminalis and the FT. Until the eighth week of gestation, the spinal cord terminates at the thecal sac. The FT then develops through canalization and retrograde differentiation of the caudal neural segments. Extending from the tip of the intradural CM to its extradural position at the first coccygeal segment, the FT elongates proportionally with the spinal cord. As development progresses, the differential growth rate of the vertebral bodies and the spinal cord ultimately positions the CM at the level of L4 by the seventeenth week of gestation.1 Both the vertebral bodies and spinal cord will continue to grow, with the CM reaching the level of L2 at approximately two months of age in 98% of the population.2-6

Tethered Cord Syndrome

Tethered Cord Syndrome (TCS) is a condition in which the filum terminale (FT), the anchoring segment of the cauda equina, is prevented from hanging freely within the spinal canal. Originally described in the pediatric population, it was hypothesized that a tight FT may result in prolonged spinal cord traction leading to caudal displacement of the conus medullaris (CM).7 Neurologic deficits, especially incontinence, were thought to be attributed to the effects of the tethered FT because these patients often improved clinically after surgical un-tethering. In 1981, Yamada et al described children with tethering of the FT who were responsive to surgical un-tethering as having “primary” tethered cord syndrome.8-10 This population of patients includes those whose neurologic deficits were related to spinal cord traction, as well as those patients with a tethered FT due to a congenital malformation—such as myelomeningocele, lipomyelomeningocele, diastematomyelia, or spinal lipomas.2,5,11 In both patient populations, the diagnosis of TCS pre-operatively depends primarily on MR imaging. A low-lying CM or a thickened FT is often used as sensitive indicators of TCS.8-10,12,13

Currently, MR imaging remains the standard of care in the diagnosis and characterization of TCS in both the adult and pediatric populations. MR imaging is extremely useful in visualizing the shape and location of the CM and FT in high resolution anatomic detail. The majority of patients in the adult population with clinically-suspected TCS will exhibit suggestive radiographic findings of TCS, such as the low-lying CM or thickened FT. However, a small subset of adult patients with TCS will have a normally positioned CM with the presence or absence of a thickened FT. Hence, additional radiographic findings, such as craniocaudal movement of the CM, have been studied as a means to diagnosis TCS definitively in these situations.12,14-17

Pediatric Versus Adult TCS Onset

Children with documented TCS often present clinically with neurologic deficits beginning in childhood and so contributes to early diagnosis. Due to the relative differential growth rates of the vertebral bodies and spinal cord during childhood, the FT elongates to allow ascent of the CM.2,4,14,18,19 Hence, lesions that prevent the FT from elongating or tether the FT at an inappropriate spinal level will cause a significant amount of traction to be placed on the CM. Embryologic defects that disrupt neurulation of the caudal spinal cord can inevitably anchor the FT prematurely.2,4,5,20-22 As mentioned earlier, TCS is common in children who present with myelomeningocele or meningocele—with both caused by a defect in the closure of the caudal neuropore resulting in spinal dysraphism. Exposure of paraxial mesoderm, secondary to defects within the neural ectoderm, allows infiltration of fat leading to spinal lipomas or even lipomyelomeningoceles. Diastematomyelia refers to total separation of the spinal canal due to an osseous, fibrous or fibrocartilaginous septum. Both legs of the spinal cord reunite caudally, tapering to form a shortened, pathologic FT. Although genetic abnormalities associated with TCS have been described, the molecular details involving its pathogenesis remain unclear.1,10,15,22,23

While adults with TCS may be asymptomatic initially, when TCS becomes clinically evident, it often presents with lumbrosacral pain before neurological deficits.4,5,10,24 Based on their different clinical presentations, it is unclear whether or not adult TCS shares a common pathophysiology with childhood TCS. Because adult onset TCS has a somewhat insidious onset, Pang hypothesized that precipitating events— i.e., disk protrusion, additional tugging in the lithotomy position, severe flexion of the spine, or direct trauma—can produce additional traction on the spinal cord thus eliciting the symptomatology of TCS.4,10,15,23,25 These observations suggest that, in adults, either the spinal cord is under a minimal amount of traction to allow blood flow to neurons or the tethering process is slow and gradual (i.e., scar tissue formation). When these precipitating events occur, traction on the spinal cord is increased, blood flow to neurons is diminished, neurons enter a state of hypoxia, and cellular injury occurs.4,14,25

TCS in adults has been described as causing a constellation of neurological complaints but lower back pain remains the most common symptom (although rare as a presenting symptom in the pediatric population).5,25,26 Unlike its pediatric counterpart, TCS in adults is considered secondary (or acquired) and can occur in the absence of the traditional radiologic criteria. Hence, adult onset TCS is often referred to as a tethered cord-tight filum syndrome, occult tethered cord, and normal conus-tight filum syndrome.8,9,10 Similar to primary TCS, releasing of the tethered FT in adults has been shown to be efficacious in the resolution of symptoms and, perhaps, prevention of further deterioration of the spinal cord7-10,12,27 Because of their different clinical presentations and etiologies, the question remains whether TCS in the adult population and pediatric population represent the same underlying disease process or two pathologically different processes.

In both pediatric and adult TCS populations, tethering of the FT ultimately leads to spinal cord ischemia.8-10,14 In children, ischemia is produced early and abruptly leads to early signs and symptoms of autonomic neuropathy. In adults, ischemic insults are produced gradually, with symptoms manifesting as a result of a precipitating event. Yamada and Dolan et al demonstrated that the neuronal cell bodies within the grey matter of the spinal cord may first be affected by hypoxia.8-10,14 In children, this may account for the signs and symptoms associated with lower motor neuron injury that is manifested as incontinence, hyporeflexia, and muscle atrophy. This is in contrast to signs and symptoms associated with injury of white matter, upper motor neuron corticospinal tracts such as hyperreflexia, clonus and the Babinski’s sign.1,5,10,14,15,20,28

Etiology of Tethered Cord Syndrome in the Pediatric Population
Many congenital anomalies predispose to the formation of TCS in the pediatric population. Spinal dysraphism, or defects in spinal fusion, can present in such forms as myelomeningocele or meningocele. Although these defects will often be surgically corrected, spontaneous resolution of the meningocele has been described as “meningocele manquΘ,” a healing process often leading to a tethered FT.1,2,8,29 Other anomalies associated with TCS have been described as well; diastematomyelia, caudal agenesis, imperforate anus, and dermal sinus tumors are common examples. TCS has also been categorized as a part of various congenital syndromes, such as the VATER syndrome (vertebral anomalies, imperforate anus, tracheoesophogeal fistula and renal anomalies) and the OEIS complex (omphalocele, extrophy of the cloaca, imperforate anus and spinal malformations with tethered cord). Genetic components involved in the pathogenesis of TCS have revealed associations with trisomy 21 (Down Syndrome), 22q11.2 deletion (DiGeorge Syndrome), trisomy 13q32 (Patau Syndrome), and trisomy 8 (Warkany Syndrome 2). Other genetically linked syndromes, such as neurofibromatosis type 1, the Dandy-Walker anomaly, and Fuhrman Syndrome have been associated with the presence of TCS.1 In many cases, the specific cause of TCS in the pediatric patient is made evident upon physical examination thru the presence of a sacral dimple, hypertrichosis, subcutaneous lipomas, or even scoliosis. Neurologic symptoms predominate in most clinical presentations, as they are often severe and occur abruptly.1,2,6,9,10,30

Etiology of Tethered Cord Syndrome in the Adult Population
Similar to children, many adults with TCS have structural lesions that contribute to the tethering of the FT. Such lesions include tumors, myelomeningocele, lipomyelomeningocele, and even post-surgical or inflammatory adhesions.1,21,29 However, the exact mechanisms that precipitate TCS in adulthood and lead to its symptomatology remain unknown. Pang, et al hypothesized, through retrospective analysis, that three different mechanical processes precipitated adult TCS in 61% of their patient cases.25 These mechanisms included: 1) spinal trauma, 2) constriction of the spinal canal and 3) transient stretching of the spinal cord. In their case report, examples of spinal trauma precipitating TCS included traumatic falls on the buttocks or direct impact to the lumbar spine. Spinal canal stenosis could be seen in cases of spondylolisthesis, vertebral osteophytes, lumbar disk herniation, and facet arthritis. Transient stretching of the spinal cord could occur during birth or during periods of forced flexion of the leg, such as the case in certain motor vehicle accidents. In addition, Pang theorized that the severity of any of these precipitating events would inversely correlate with the age of symptom onset.2,19,21,25,29

The Role of Traction and Mitochondrial Dysfunction

Yamada et al assessed the oxidative function of neuronal mitochondria during periods of spinal cord traction by measuring the reduction/oxidation (or redox) potential of cytochrome a,a3 with dual wavelength spectrophotometry. The oxygen dependent cytochrome a,a3 is the terminal oxidase of the electron transport chain in mitochondria, and functions in the generation of cellular energy by catalysis of adenosine diphosphate into its triphosphate form. Yamada measured the redox ratio of cytochrome a,a3 as an indirect measurement of local oxygen availability. During periods of hypoxia, cytochrome a,a3 favors reduction, and therefore the redox ratio is relatively increased. Yamada showed that these high levels of reduced cytochrome a,a3 were present within spinal interneurons during periods of transient spinal cord traction. High levels of the reduced form of cytochrome a,a3 reflected an impairment of oxygen delivery to the spinal cord during periods of traction, depleting intracellular energy stores and leading to cellular injury.8-10,15 Based on Yamada’s observations in human and animal models of TCS, placing traction on the spinal cord impaired energy metabolism as evidenced by the high levels of reduced cytochrome a,a3 within neuronal mitochondria.8-10 As in hypoxic conditions, transient traction and reduced blood flow to spinal neurons increased the reductive capacity of cytochrome a,a3.8-10,14

Schneider observed in vivo that high redox ratios and reduced blood flow to the caudal cord in pediatric patients with TCS could even be reversed with surgical release.29 Similar results using animal models of TCS were reported when Yamada described that low to moderate spinal cord traction led to neuronal ischemia that was reversible upon release. Pre-surgical studies in pediatric patients with TCS showed ischemic changes within neuronal mitochondria as evidenced by disruptions in the redox ratio of cytochrome a,a3. Post-surgical analysis exhibited that release of the tethered FT not only allowed the redox ratio to trend towards normal, but it also was followed by immediate resolution of symptoms. However, these findings were seen mostly in patients with mild ischemic neuronal insults. Patients with severe oxygenation impairments exhibited only mild improvements in the redox ratio following surgery, and it often took longer for their symptoms to resolve.8-10,15,20,22

Additional studies examined the correlation between the amount of traction exerted on the caudal spinal cord, the severity of symptoms and the metabolic state of neuronal mitochondria.8,9,11,15,31 Yamada showed that low to moderate isometric traction on the FT produced hind limb motor deficits in cats, analogous to those seen clinically in adults with TCS. Redox ratios were increased during periods of moderate isometric traction, and reversed to normal once the traction was removed. When large amounts of isometric traction were applied to the spinal cord, motor deficits appeared to be more severe, and the redox ratio increased as expected. Once the traction was removed, however, there was incomplete resolution of the oxidative state of cytochrome a,a3. Hence, large amounts of traction on the spinal cord may lead to irreversible neuronal injury and severe motor deficits. Yamada suggested that severe traction, as opposed to low grade traction, caused extensive blood flow disruptions resulting in diffuse neuronal hypoxia and irreversible cell injury8-10 and only showed partial neuronal recovery upon reversal. These observations suggest that the ability to reverse neuronal injury upon surgical release may be dependent upon the severity of the FT tethering.

In addition, Yamada showed that after several months of chronic low grade traction, cats were not only able to adapt to the experimental conditions and regain hind limb function, but also neuronal mitochondria showed no statistical difference in the redox ratio of the experimental group verses control. Although more traction applied to the cord lengthened the amount of time needed for adaptation, chronic spinal cord traction (9 months) did not create metabolic shifts within the mitochondria.8-10,20,22,23

Blood Flow Studies

As traction is applied, the luminal radius of arteries, veins and capillaries that supply the spinal cord is diminished thus decreasing blood flow and increasing the redox potential of cytochrome a,a3 and, ultimately, resulting in neuronal cell injury. Yamada first showed that high redox ratios could be reproduced in spinal interneurons by occluding the aorta. After 15 minutes of occlusion, the redox ratios could be reversed to normal by the restoration of blood flow. If this period of ischemia lasted for longer than 15 minutes, only a partial recovery was achieved. Yamada observed that when blood flow decreased, neuronal cell bodies of the grey matter were preferentially affected as demonstrated by spinal evoked potentials portraying ischemic patterns.8-10,15,23

Using spinal evoked potentials, Dolan et al showed that prolonged periods of traction can cause ischemic insult to neurons within the spinal cord. Releasing traction within a window of two to eight weeks significantly reduced ischemic injury as evidenced by reversal of blood flow and normalization of spinal evoked potentials.14 Further studies using Doppler flowmetry showed significantly reduced blood flow at the level of the caudal spinal cord in patients with TCS intraoperatively.15,23,31 Reversal of blood flow was obtained following surgical release of the tethered FT. These results suggest that both traction and vascular compromise contribute to the pathophysiology of TCS.

Biomechanics of the Conus Medullaris and Filum Terminale

Hoffman et al suggested that elasticity and mobility of the spinal cord are critical components in the pathogenesis of TCS.4,23 Hoffman and colleagues observed that the most caudal part of the spinal cord, the FT, undergoes the greatest amount of elongation when caudal traction is applied. They noted a linear relationship between the amount of force applied to the spinal cord and the length of the FT. Areas above the lowest dentate ligaments (level of T11) of the spinal cord did not move due to their stabilization.4,15 During periods of prolonged traction, such as in TCS, neurologic deficits would then be isolated to the caudal spinal cord where stabilizing ligamental connections are absent. Clinically, this correlated to the motor, sensory and autonomic deficits that typically manifested as a result of TCS. Thus, factors such as fibrosis, scarring, or external compressive forces—e.g., tumors—could inevitably alter the inherent mobility of the FT and precipitate TCS.4 Spinal lipomas and diastematomyelia were found to be the two most prevalent lesions precipitating TCS in a study of 71 patients conducted by Lassman et al.4,15,29,32,33

Retrospective Analysis of MR Imaging Findings in TCS

Raghaven et al described the MR imaging findings via retrospective analysis in 25 patients diagnosed with TCS.24 MR imag-ing was obtained in all patients using a 1.5 T whole body superconductive magnet, with spin-echo sequences acquired at 128 or 256 views in the phase encoding direction and 256 views in the read out direction. Sagittal T1-weighted images using short TE/TR with 3-5mm thickness were obtained in all patients. Tethering of the spinal cord was identified readily on T1-weighted MR images as the high signal intensity of fat was separable from the signal of the cord itself and the CSF. In addition, all patients with TCS were documented as having visible evidence of a lumbro-sacral anomaly (sacral dimple or fatty lump) or neurogenic symptoms associated with TCS—i.e., urinary incontinence, gait, motor or sensory deficits. Raghaven described the level of the CM in 84% of his patients. Most commonly, the tip of the CM was located below the mid L2 vertebral body. In 16%, the tip of the CM was located at or above the level of the mid-vertebral body of L2. In order to diagnose TCS in this small subset, Raghaven noted the presence or absence of a thickened FT or tethering lesion. A thickened FT (>2mm) was noted in 94% of patients in whom this measurement could be made. Raghaven also noted the presence of syringohydromyelia/myelomalacia in 45% patients with identifiable TCS. This was described as a linear, low intensity cord lesion that extended from the CM to the FT or tethering lesion and best seen on axial views. Because these lesions were not biopsied for definitive pathologic diagnosis, these linear low intensity lesions were termed syringohydromyelia/myelomalacia, by Raghaven.24

Movement of the Conus Medullaris in Patients with TCS and a Normally Positioned Conus
Witkamp et al studied the movement of the CM in patients with and without TCS using MR imaging.11 This consisted of obtaining sagittal and axial T1-weighted images (500/20, 3–5-mm section thickness, 350-mm field of view) and T2*-weighted gradient-echo (615/27, 25░ flip angle) images through a region of interest, depending on the abnormality found in the sagittal place. Due to the support of the dentate ligaments until it terminates caudally at the level of the 11th thoracic vertebrae, the CM is freely suspended within the spinal canal at the level of the 1st and 2nd lumbar vertebrae.32-34 Hence, Witkamp and colleagues hypothesized that they would be able to visualize movement of the CM as patients moved from supine to prone positioning under MR imaging. In patients unaffected by TCS, CM motion ranged on average 33%. In patients who showed radiographic evidence of TCS on supine MR imaging, motion of the CM was unchanged compared to prone imaging. In patients who were clinically suspected of having TCS but showed no evidence of TCS on supine MR imaging, movement of the CM was markedly diminished (13% movement). TCS was then confirmed in these patients intraoperatively. Due to respiratory effort, prone positioning can create movement artifacts. As Witkamp showed, these variables can be accounted for by supporting the shoulder and pelvic region. This allows for more space for the abdomen to move and thus, reduces the anterior posterior movement of the spine.11

The Fatty Filum Terminale

Whereas increased intradural fat attenuation may indicate a mass-like lesion such as a spinal lipoma, the significance of a fatty FT in the diagnosis of TCS is still undetermined. Raghaven et al stated that fat in the FT may not be a reliable indicator of TCS since many of the patients in his study with a fatty FT did not meet the radiographic criteria for TCS.11 Fatty infiltration of the FT was noted in select cases where the CM lay in normal position, the FT was not thickened (<2mm), and the symptoms could not be explained by disk herniation. In fact, Raghaven proposed that fatty infiltration of the FT may even be a result of faulty retrogressive differentiation during the development of the caudal spinal cord.11 Bulsara et al identified patients who were revealed to have a fatty FT under MR imaging, during the evaluation of metastasis, infection, neurologically intact low back pain, and neurological impairment. While all groups demonstrated a CM in normal position, the extent of fat infiltration along the FT to the tip of the CM itself was the only statistically significant difference amongst these groups. Specifically, the variable of fatty infiltration of the FT within 13mm of the CM was most predictive of future neurological deficits.3 Uchino et al described the presence of a fatty FT in a series of 1,691 patients undergoing lumbrosacral MR imaging for a variety of causes. These authors approximated the incidence of finding a fatty FT as 0.24% in this group of patients. In addition, this small subset of patients showed no evidence of tethering or abnormal CM positioning. Interestingly, degenerative disk disease was noted as a common finding in these patients.13 Lizuka et al described the presence of a fatty FT in 0.01% of patients who also underwent lumbrosacral MR imaging for a variety of indications. 85% of the patients with a fatty FT had concurrent disk degeneration. However, Lizuka found no statistically significant differences in the clinical symptoms, location, distance and diameter in patients with a fatty FT compared to those without.27

Anatomic Considerations of the Filum Terminale

Pinto et al described the intradural anatomic features of the FT in a human cadaveric study analyzing the morphologic characteristics as they pertain to TCS. Pinto characterized the vertebral level at which the CM intersected with the FT, the diameter and length of the FT, as well as the level at which the FT fused with the dural sac. In his report, the FT/CM junction occurred most commonly at the upper one-third of the vertebral body of L1 (19.51%). CM/FT junctions occurring at the middle and lower one-third of the vertebral body of L1 were the next most common locations at 17.07% and 12.20%, respectively. The mean length of the FT was 156.4mm, and the mean thicknesses of the initial FT and the FT midpoint were 1.38mm and 0.76mm respectively. Four FT specimens were thicker than 2mm at their initial point, but no specimens were thicker than 2mm at their midpoint. The mean level of FT fusion reported was at the S1 level in 46.34% of specimens, although 29.27% showed fila that fused at the sacral level of S2.16

Radiologic Criteria

Currently, MR imaging is used to confirm the presence of TCS in patients who are clinically stable and without any contraindications to electromagnetic exposure. The inherent advantages of MR imaging over the previous use modalities of CT myelography are: 1) no radiation exposure, 2) is non-invasive, 3) shows superior anatomic definition, and 4) can characterize scarring or space-occupying lesions—such as a spinal lipoma—causing the tethered FT.5,7,12,20,23,24,28,32 Although the use of CT still remains superior to MR imaging in the characterization of bony details and vertebral arch malformations, relatively simple bone spurs or dysraphic changes may be interpreted with ease using MR imaging, as noted by Raghaven. Raghaven also described the presence of a thickened FT (>2mm) in 16% of patients with TCS who also have a normally positioned CM. Due to the complex and intricate anatomy of the spinal cord, MR imaging is extremely useful in the diagnosis of TCS, as structures such as the CM and FT must be identified.24

While either axial or sagittal views can be used to identify the CM, axial views are far superior in the ability to differentiate the CM from the cauda equine.24 Axial views tend to be unaffected by midline shifts that would inherently make medial structures difficult to discern on sagittal sections.24 Recent research using diffuse tensor imaging (DTI) and tractography has gained attention due to its ability to delineate networks of communicating spinal tracts within the nervous system. Filippi et al described the particular benefit of using tractography and DTI in the characterization of diastematomyelia. In the future, DTI and tractography may play a more central role in defining the risks of neurological impairment as well as aiding in presurgical planning.35

After two months of age, the normal position of the CM lies at or above the mid vertebral level of L2, most commonly at the T12-L1 vertebral interspace.1,17,36 The radiographic criteria that define TCS are the presence of a CM below the level of L2 by the age of two months and a thickened FT (>2mm).8-10,37 The uses of axial T1-weighted images are most accurate in determining the CM level.18 Raghaven described a low-lying CM below the L2 vertebral body in 84% of his patient population with surgically-identifiable tethered FTs. However, in a small population of patients with TCS, the CM may lie in a normal position. In fact, Raghaven described the presence of a thickened FT (>2mm) in 16% of patients with TCS who had a normally positioned CM.24

The radiological diagnosis of TCS thus relies primarily on the presence of a thickened FT or lesion potentially causing the tethering.38 Often viewed as a protective function, the spinal cord is inherently an elastic structure contributing most of its compliance to the ability of the FT to stretch in response to stress. When the FT is thickened, its compliance, and thus the compliance of the spinal cord, is markedly decreased even in the absence of a caudally displaced CM.4

In TCS, the thickened FT is attached to the dura or extradural connective tissue that anchors the cord. In some patients, the cord may be bound by lipomatous tumors. Lipomas may be sessile or pedunculated and may be attached to the dorsal and/or caudal aspect of the CM.1,3,5,12,13,20,23,24,29,31 Fatty infiltration of the FT may signal the presence of an extradural fibrolipoma or an intradural spinal lipoma.3,5,13,20,27 An intradural spinal lipoma—more commonly associated with tethering—exists at the distal aspect of the thecal sac and results from small amounts of mesenchyme passing into the neural canal prior to closure of the caudal neuropore.1,3,39 Intradural lipomas are usually located within the dural sac and have no anatomic connection to subcutaneous fat. An extradural fibrolipoma tends to infiltrate the FT and is the result of a caudal cell mass malformation during embryogenesis.1,3,13,20,27,39 Raghaven observed that the most common causes of tethering in his studies were in fact, spinal lipomas.24 In such cases, a low-lying CM was concurrently present. The diagnosis of a spinal lipoma is generally made with MR imaging showing characteristic hyperintense lesions on T1-weighted images. T1-weighted images have high sensitivity in the detection of lipomas because of the short relaxation time of hydrogen protons in fat. In addition, the fat content of the FT may also be confirmed through the use of fat-saturation techniques.1,3,5,8-10,12,13,20,23,24,29,33


Currently there is debate regarding the clinical significance of small spinal lipomas in the absence of other anatomic anomalies. Fatty infiltration of the FT, when the CM is in normal position and the FT is not thickened, may be considered a normal variant of the spinal cord and may even be reported as an incidental finding on MR imaging.3,8-10,12,13,17,27,40 Raghaven proposed that fatty infiltration of the FT may result from retrogressive differentiation where a small portion of mesenchyme remains on the FT.24 Both Uchino and Lizuska described the presence of a fatty FT as an incidental finding in patients undergoing routine lumbrosacral MR imaging. No statistically significant correlations were identified between fat content and tethering of the FT. Interestingly, in both the Uchino and Lizuska studies, patients noted to have a fatty FT had concurrent degenerative disk disease.13,27 Bulsara concluded that only the distance of the fatty FT to the CM (<13mm) correlated with an earlier onset of neurological deficits in patients with degenerative disk disease.3 Degenerative disk disease can produce edema, neovascularization, fibrosis and even fat infiltration at the intervertebral disk end-plate—pathological processes termed Modic changes.41 Chronic degenerative disk disease may be associated with fatty infiltration of the vertebral end-plate.41 Thus, fatty infiltration of the FT secondary to vertebral end-plate infiltration may correlate with chronic, advanced degenerative disk disease. However, more research is needed to support these findings.

In his study, Raghaven described TCS patients who had a CM in normal position, a minimally thickened FT (3mm) and a linear low-intensity lesion that represented syringohydromyelia/myelomalacia.24 Myelomalacia is a chronic pathological process that involves softening of the spinal cord usually associated with atrophy or chronic neuronal injury. Myelomalacia is usually determined by a low intensity signal on T1-weighted images resembling cerebrospinal fluid (CSF) with irregular margins representing focal gliosis. Hydromyelia is a tubular cavitation of the spinal cord, or syrinx, that becomes distended by CSF.42 Commonly, margins of the cavity are well defined due to the ependymal lining of the canal. Syringomyelia refers to the process whereby excessive amounts of CSF that may dissect into the surrounding white matter to form a paracentral cavity in which there is no ependymal lining and thus no margin regularity. In contrast, diastematomyelia presents as a local widening of the spinal cord at the level of the septum with hypointense CSF filling the smooth, contoured borders of the subarachnoid space.12,42

When hydromyelia and syringomyelia occur together, they are referred to as syringohydromyelia, which may result from scarring, tumors or vascular insults.42 Blockage of normal CSF flow through the spinal cord and focal areas of cavitating necrosis may result in syringohydromyelia. Due to their similar appearance on T1-weighted MR imaging, Raghaven described these findings together as “syringohydromyelia/myelomalacia.”24 Because a variety of pathological processes can produce syringohydromyelia/myelomalacia, it is relatively non-specific to the diagnosis of TCS. However, the presence of these findings may suggest underlying spinal cord injury in those patients who minimally meet the criteria for TCS.24 As previously discussed, it is common for adults with TCS to remain subclinical until a precipitating event creates further traction on the spinal cord. As indicated by Yamada and Dolan, transient ischemic episodes may occur chronically in patients with TCS in which spinal cord neurons may fully or partially recover.8-10,14 If only partial neuronal recovery occurs, the accumulation of damaged cellular functions may ultimately lead to areas of weakened spinal cord integrity—i.e., myelomalacia. Further studies, however, are needed to elucidate the utility of these findings in the diagnosis and clinical significance of TCS.

By convention, MR imaging is performed with the patient in the supine position.18,40 In this position, the natural kyphosis and lordosis of the thoracic and lumbar spine, respectively, are reduced. The CM and FT are therefore free from additional traction contributed by curvatures of the spine. In the prone position, curvatures of the spine will inherently alter the shape and position of the CM and FT.18,40 Witkamp showed that movement of the CM when switched from the supine to prone position was on the order of 33% in patients without TCS. In patients with TCS, Witkamp explains that supine and prone MR imaging is only useful in patients for whom there is a high clinical suspicion of TCS but who exhibit normal supine MR imaging. In these patients, diminished movement of the CM on the order of 13% movement was seen on prone MR imaging.11 As TCS progresses, supine MR imaging may be more likely to show advanced pathology of the CM and FT in which case movement would inevitably be restricted in both the supine and prone positions.

Pinto et al described the normal morphological features of the FT in a randomized adult population. The anatomic variations in the FT thickness have accounted for much debate, as 3-6% of the normal population will have FT thickness >2mm and surgical “release” has shown marked clinical improvement in patients with FT’s <2mm in thickness. Nazar et al described a series of 32 pediatric patients with TCS who did not exhibit a CM below L2 or an FT thicker than 2mm.28 Pinto observed in his study that in most cadavers, the CM lay at the level of L2 with the initial FT segment beginning at L1. This represents the most common location of the CM in a healthy population greater than two months of age.2,36 When the CM was below the level of L2, no cadavers had thickened FT or a history of TCS. The most common level for the FT to fuse with the dural sac is S1. Hence, surgical sectioning of a tethered FT is often performed in conjunction with an S1 laminectomy. Surgically relevant, Pinto observed that in a significant portion of his specimens (30%), the fila fused at S2.16 In addition, the FT decreases in diameter as it extends from its initial junction with the CM to its fusion with the dural sac. Thus, the initial segment of the FT may have a greater thickness than the FT at its midpoint. Pinto described an initial FT thickness of > 2mm in 10% of his specimens.16 Thus, multiple measurements of FT thickness at different levels would confer a greater predictive value of the presence of TCS in light of these normal anatomic variations.


Tethered Cord Syndrome is a well-documented clinical entity that is becoming increasingly recognized as a significant cause of adult onset lower back pain and pediatric autonomic neuropathy. Since its discovery by Garceau in 1953,44 many scientists and clinicians have begun to unravel the etiology and pathophysiologic mechanisms leading to its clinical presentation. Through numerous clinical observations, animal-based research studies and MR imaging procedures, the ability for physicians to recognize and detect TCS has greatly improved. Advances in imaging technology, such as diffuse tensor imaging (DTI) and tractography, may prove to be useful in the characterization of lesions in the spinal cord and pre-surgical planning prior to intervention. In this field of evolving neuroradiological research, clinicians can begin to now integrate diagnostic information obtained through imaging and clinical presentation in order to ensure a definitive diagnosis and optimize medical intervention.

Last updated on: March 7, 2011
close X