21

Thoracic disk herniation is not a common disease. In 1838, Key1 wrote the first report of a thoracic herniated disk causing spinal cord compression. Diagnosis can be challenging due to a variety of clinical presentations. With the advent of magnetic resonance imaging (MRI) the diagnosis of symptomatic disk herniation has been enhanced.^2–10 The natural history of thoracic disk herniation is not completely clear. Some authors propose a degenerative etiology,¹⁰ whereas others have shown a role for trauma.¹¹

With the advent of MRI, there has been an overdiagnosis of false-positive results because of asymptomatic thoracic herniated disks. In some cases surgical removal is regarded as the treatment of choice for a symptomatic herniated thoracic disk causing (1) myelopathy, (2) lower extremity weakness or paralysis, (3) bowel/bladder dysfunction, and (4) radicular pain refractory to conservative treatment. Diskectomy may be performed by a variety of surgical approaches: costotransversectomy, transthoracic, posterolateral, or video-assisted, each of which has its own merits and shortcomings.

The prognosis associated with surgical decompression has improved dramatically from the days of laminectomy and diskectomy to more refined approaches without cord manipulation. Although there have been numerous reports in the literature regarding thoracic disk herniations there is still controversy as to the most accurate mode of diagnosis and the best surgical approach to this problem.

Although there have been many reports in the literature regarding thoracic disks, there are several key issues that remain controversial: (1) What is the natural history for a patient with a thoracic herniated disk? (2) What is the best imaging modality/protocol to accurately identify symptomatic thoracic disk protrusion? and (3) What is the best surgical approach to address symptomatic thoracic disk herniations?

To assess these issues we performed a comprehensive review of the literature to determine the best evidence available on each of these topics. The search included Medline and Cochrane Central Register of Controlled Trials. Also a review of the references of these articles was performed for any additional studies. A search for thoracic disk herniation yielded 244 articles. Each article was individually reviewed to identify whether it would be useful in addressing these controversial issues. A majority of these were level IV case reports or review articles or older reports from the 1950s to the 1970s. Only 13 studies provided data on the key issues being discussed in this chapter. There was one level I study identified, nine relevant level II studies, and three level III studies. The level of evidence is summarized in Table 21.1.

What Is the Natural History for a Patient with a Thoracic Herniated Disk?

Level I Data

There are no level I data published regarding this topic.

Level II Data

There are no level II data published regarding this topic.

Level III Data

The study by Brown et al¹² reported on 55 patients with 72 thoracic disk herniations. Each patient’s record was retrospectively reviewed in an effort to ascertain the natural history of this disease. The treatment programs given to these patients were evaluated, and 15 (27%) of the 55 patients eventually required surgery. The majority, however, did not require surgery and have continued to perform activities of daily living, some even participating in vigorous sports activities (e.g., skiing) without any apparent neurological consequences. Thoracic disk herniations, similar to cervical and lumbar disk herniations, do not always lead to major neurological compromise. A less aggressive approach therefore can be considered.

Table 21.1 Evidence Levels of Published Studies

Sonnabend et al13 analyzed 127 cases of intervertebral disk calcification in children, including 11 previously unreported cases. A distinction was made between symptomatic and asymptomatic patients, whose age, sex, and spinal distributions of the calcifications have been shown to differ. Radiologically detectable protrusions and later resorption of the calcifications are common events in symptomatic children but are unrecorded in asymptomatic children. The spinal distribution of pediatric calcifications is quite different from that of adult calcifications of the nucleus pulposus.

Summary of Data

The literature shows that in the majority of cases herniated thoracic disks in both adults and children have a fairly benign course and may not lead to long-term symptoms of axial back pain, radiculopathy, or myelopathy. There is, however, a subset of patients that may have a more aggressive natural history and develop neurological compromise. Based on the grading scale of Guyatt et al¹⁴ this recommendation for following patients with a conservative protocol with the expectation that most will not need surgery would be 2C, or very weak. The data are summarized in Table 21.2.

Table 21.2 Summary of Data Regarding the Natural History of Herniated Thoracic Disks

Wood et al15 reported on an investigation whose objective was to determine the responses to thoracic diskography of asymptomatic individuals. Historically the literature regarding lumbar and cervical diskography reveals that even morphologically abnormal disks are often not painful, whereas painful disks typically exhibit anular or end plate disruption. Understanding this dilemma, they studied 10 adult lifelong asymptomatic volunteers, ages 23 to 45 years. The participants underwent MRI of the thoracic spine, followed by four-level diskography. Provocative responses were graded on a scale of 0 (no sensation) to 10 (extreme pain or pressure), and filmed disks were graded using a modified Dallas scheme. Concomitantly, 10 nonlitigious adults (6 men and 4 women, ages 31 to 55 years) experiencing chronic thoracic pain were similarly studied as a control group. The mean pain response in the asymptomatic volunteers was 2.4/10. Three disks were intensely painful (scores of 7/10, 8/10, 10/10), with all three exhibiting prominent end plate irregularities and anular tears typical of thoracolumbar Scheuermann disease. On diskography, 27 of 40 disks were abnormal, with end plate irregularities, anular tears, and/or herniations. Ten disks read as normal on MRI showed anular pathology on diskography. In the group with chronic thoracic pain, the average pain response was 6.3/10 (p < 0.05). Of the 48 disks studied, 24 were concordantly painful, with a pain response of 8.5/10 (p < 0.05); 17 had nonconcordant pain/pressure, with an average pain of 4.8/10 (p < 0.05); and five had no response. On MRI 21 of the 48 disks appeared normal. However, on diskography, only 10 were judged as normal. On diskography, thoracic disks with prominent Schmorl nodes may be intensely painful, even in lifelong asymptomatic individuals, but the pain is unfamiliar or nonconcordant. Thoracic diskography may demonstrate disk pathology not seen on MRI.

Ryan et al¹⁶ reviewed 270 computed tomographic (CT) scans of the thorax obtained over a 7-month period; four patients (1.5%) with calcified herniated thoracic disks were identified. Each of these patients presented with abnormal chest radiographs and had a CT examination for evaluation of suspected malignancy. None showed any signs or symptoms of thoracic spinal cord compression. The clinical significance of incidental thoracic disk protrusions is unknown.

Awwad et al¹⁷ retrospectively reviewed the myelograms of 433 patients and identified those who had no symptoms or signs referable to the thoracic cord, roots, or nerves. By postmyelography CT scan criteria, the frequency of asymptomatic thoracic herniated disks (ATHDs) was calculated. Postmyelography CT scans of 68 ATHDs were analyzed. Their imaging characteristics were compared with a series of five symptomatic thoracic herniated disks and symptomatic thoracic herniated disks in the literature. They were unable to identify any imaging features that could reliably classify a disk as an ATHD or a symptomatic thoracic herniated disk. Their results call into question the propriety of prophylactic surgery for ATHDs, even when the lesions are radiographically impressive.

Videman et al18 undertook a descriptive epidemiological study of MRI findings of the spine. Their goal was to describe the prevalence of MRI findings in a general population at spinal levels T6–S1, and to examine the relationships of these findings within each spinal level and between levels. They thought the prevalence of specific findings and the associations between findings and spinal levels could provide general insights into the etiopathogenesis of spinal degeneration. They reviewed MRIs of 232 men from a population sample (mean age 49.3 years). Signal intensity, disk bulging, disk herniation, and end plate irregularities were among 11 findings assessed from MRI. They found the disk signal intensities were lowest in the lumbar and middle thoracic regions. Disk bulging and disk height narrowing were most common in the lower levels of both the thoracic and the lumbar regions. All MRI findings except herniations and end plate irregularities were clearly associated with age. Osteophytes were most highly associated with disk bulging in levels T6–L3, and with end plate irregularities in the lower lumbar levels. Disk herniations were not consistently associated with any other findings. The disk levels that most highly correlated were grouped as follows: T6–T10, T10–L4, and L4–S1. With the exception of end plate irregularities and herniations, the MRI findings appeared to be associated with the same pathogenic process. The interaction of mechanical factors and spinal structures varied between spinal levels, and the degeneration common in the lower parts of the thoracic and lumbar spine could be an outcome of vulnerability for torsional forces. Some gross guidelines for grouping findings could be drawn from disk level correlations.

An early study by Wood et al¹⁹ reviewed MRI studies of the thoracic spines of 90 asymptomatic individuals to determine the prevalence of abnormal anatomical findings. This group included 60 individuals who had no history of any thoracic or lumbar pain and 30 individuals who had a history of low back pain only. In addition, they reviewed imaging studies of 18 patients who had an operatively proved herniation of a thoracic disk and studies of 31 patients who had been seen with thoracic pain. Sagittal T1-weighted spin-echo and axial multiplanar gradient refocused images at each disk level were interpreted by the authors (two neuroradiologists and two orthopedic spine surgeons); they were blinded to the clinical information about the patients. Sixty-six (73%) of the 90 asymptomatic individuals had positive anatomical findings at one level or more. These findings included herniation of a disk in 33 subjects (37%), bulging of a disk in 48 (53%), an anular tear in 52 (58%), deformation of the spinal cord in 26 (29%), and Scheuermann end-plate irregularities or kyphosis in 34 (38%). This study documents the high prevalence of anatomical irregularities, including herniation of a disk and deformation of the spinal cord, on the MRI of the thoracic spine in asymptomatic individuals. We emphasize that these findings represented roentgeno-graphic abnormalities only, and any clinical decisions concerning the treatment of pain in the thoracic spine usually require additional studies.

Arce and Dohrmann²⁰ reported on CT scanning with and without metrizamide in the subarachnoid space as a method with improved accuracy to diagnose thoracic disk herniations. CT scanning can demonstrate the type and level of the lesion even when the myelographic study is negative. They reviewed 280 cases; a peak incidence was noted in the fourth decade with 75% of the protruded disks occurring below T8. Back pain was the most common presenting symptom followed by sensory disturbances. By the time of diagnosis, 70% of the patients had signs of spinal cord compression. A small group of patients could be identified that invariably had a good prognosis. They had a history of trauma, symptoms lasting less than a month, and soft disk herniation. Regarding the results of surgical treatment, there was a success rate ranging from 57% for decompressive laminectomy to over 80% for the posterolateral, lateral, and transthoracic approaches.

Schellhas et al²¹ sought to evaluate the safety and reliability of diskography used to investigate thoracic disk degeneration observed on MRI studies in patients with clinical pain. They analyzed retrospectively the results of thoracic diskograms performed on 100 outpatients. After the MRI, clinically suspect, morphologically abnormal thoracic disks and at least one, nearby, control level were injected with either nonionic contrast or saline, filmed, and individually described by the patient as concordant versus nonconcordant relative to clinical pain and rated in intensity on a scale of 0 to 10. There were no serious complications in the series, and the authors resolved whether the injected disks related to pain in each case. They found disks with anular tears, intrinsic degeneration, and/or associated vertebral body end plate infractions to be painful ~75% of the time. Clinical concordance was ~50%. Control levels were usually painless. They concluded that thoracic diskography can be performed safely by experienced individuals as a reliable tertiary diagnostic procedure to determine if degenerated disks on MR studies are related to clinical complaints.

Level III Data

There are no level III data published regarding this topic.

Summary of Data

MRI has revolutionized the diagnostic evaluation of thoracic disk disease. Some centers rely on it almost exclusively, but others still perform myelography and CT when an operative treatment is being considered. Based on a thorough review of the best available evidence it appears that MRI combined with diskography provides the most accurate method for diagnosing symptomatic herniated thoracic disks. Given the general prevalence of degenerative thoracic disk pathology in the asymptomatic population, diskography appears reserved for those individuals with interdiskal pathology on MRI and thoracic axial pain unresponsive to an appropriate duration of nonsurgical treatment. There is some conflicting information with regard to the accuracy of CT myelogram, use of Metrizamide in the subarachnoid space, and MRI. At this point, there are no good comparative studies that demonstrate superiority of one diagnostic modality versus another. Based on the grading scale of Guyatt et al¹⁴ this recommendation for MRI and diskography would be 1C, or intermediate-strength recommendation. The data are summarized in Table 21.3.

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 Bone Morphogenetic Protein in the Setting of Infection: Indications, Risks, and Efficacy Vertebral Compression Fractures: Percutaneous Vertebral Augmentation Cervical Myelopathy: Meaning of Magnetic Resonance Imaging Signal Changes

Stay updated, free articles. Join our Telegram channel

Join