12 Thoracic Spinal Stenosis Abstract Thoracic spinal stenosis (TSS) is relatively uncommon compared to its cervical and lumbar counterparts. Perhaps as a consequence, it is an underrecognized cause of neurological sequelae and frequently misdiagnosed on account of the vague nature of its presentation and similarity in symptomatology to cervical and lumbar disorders. There are no standardized criteria for diagnosing TSS and no formal guidelines exist for management. However, surgical decompression is a mainstay in management and the treatment of choice once symptoms have progressed or become progressive. Considerable variation exists in the surgical approach for addressing thoracic stenosis, but the corridor employed is of little import, so long as there is adequate decompression of neural elements without introduction of iatrogenic instability. Several factors have been shown to influence treatment outcomes, most notably the shorter the duration from symptom onset to diagnosis. Timely decompression is therefore imperative to ensuring optimal neurological recovery. As the world’s population ages and life expectancy continues to rise, so too will the incidence of TSS and other degenerative spine pathologies. Thorough understanding of the nuances of TSS presentation and management will prove fundamental to successfully navigating a spine surgery practice in the years to come. Keywords: thoracic myelopathy, thoracic stenosis, thoracic spine, facet hypertrophy, spondylosis, degenerative spine disease, ossification of ligamentum flavum, ossification of posteri- or longitudinal ligament Clinical Pearls • Thoracic stenosis is a rare occurrence in the everyday practice of spine surgeons. • Symptoms can be vague and easily attributed to cervical and lumbar pathologies, hence there is a tendency for delayed diagnosis. • Identification of thoracic stenosis should prompt an investigation for other stenotic regions due to the high rates of tandem stenosis. • The shorter the interval from symptom onset to surgery, the higher the likelihood for optimal neurological recovery. • Surgical decompression is an effective treatment for thoracic stenosis. • Long-term follow-up is recommended because of the possibility for delayed neurological decline from recurrent or new areas of stenosis. Thoracic spinal stenosis (TSS) is a rare clinical entity when considered vis-à-vis cervical and lumbar degenerative spine pathologies. Of individuals suffering from spinal stenosis, thoracic disease is estimated to account for less than 1% of cases.1 Because of its rarity, and close presenting symptomatology to cervical and lumbar variants, TSS is often missed or misdiagnosed, which leads to delayed treatment and, perhaps as a consequence, suboptimal outcomes. As the U.S. population ages, however, the number of individuals aged 65 and above is expected to double by the year 2050.2 Such a demographic shift heralds increased societal burden from spinal disorders in which there is chronological deterioration with advancing age. Intimate knowledge of the operative and perioperative management of patients with TSS will therefore prove increasingly valuable in a surgeon’s armamentarium, as the field of spine surgery evolves around changing demographics. The term “spinal stenosis” can be used broadly to encompass any anatomical narrowing of the spinal canal irrespective of neurological sequelae. Nonetheless, it more commonly denotes a reduced canal dimension that results in cord or nerve root impingement with resulting neurological deterioration as well as detriment to daily functions and overall quality of life.3 Various TSS cutoffs have been reported in the literature, including an absolute anteroposterior diameter of 10 mm or less.1,4 In a study of 700 adult cadaveric specimens, detailed sagittal canal diameter and interpedicular distances were determined throughout the thoracic spine and the respective values falling two or more standard deviations below the mean equated with thoracic stenosis. It was discovered that an anteroposterior diameter of 15 mm and an interpedicular distance of 18.5 mm were predictive of congenital stenosis at all levels throughout the thoracic spine with a sensitivity and specificity of 80 and 100%, respectively.1 Despite such attempts at standardization, no universally accepted quantitative or radiographic criteria have been adopted to date. A formal diagnosis of TSS is therefore dependent on overlap between clinical presentation and radiographic evidence of canal compromise. By definition, TSS excludes compressive etiologies from thoracic malignancy (primary or metastatic), infection, kyphosis, scoliosis, and trauma.5 TSS can be classified as primary or secondary in origin. Primary stenosis constitutes a congenitally narrowed canal that over time encroaches on neural elements owing to one or more degenerative precipitants. Causative factors include namely disc extrusion, ossification of the ligamentum flavum (OLF), ossification of the posterior longitudinal ligament (OPLL), facet joint hypertrophy, and osteophytes at the posterior aspect of the vertebral endplates.5 Conversely, secondary stenosis occurs in the wake of rheumatological, metabolic, or orthopaedic disorders that narrow the spinal canal globally. Conditions commonly implicated include Paget’s disease, acromegaly, achondroplasia, rheumatoid arthritis, renal osteodystrophy, diffuse idiopathic skeletal hyperostosis (DISH), ankylosing spondylitis, osteofluorosis, osteochondrodystrophy, familial hypophosphatemic vitamin D–refractory rickets, and Scheuermann’s disease.6,7,8,9 The spine is subjected to repetitive stresses, over and above routine wear-and-tear, which accumulate throughout one’s lifetime and can result in spinal degeneration.10 The underlying pathophysiological events that account for this breakdown have been termed the degenerative cascade and were first proposed by Kirkaldy-Willis et al in the 1970s.11 It was postulated that repetitive microtrauma to mobile components of the axial spine (i.e., facet joints and intervertebral disc) led to disc tears, synovial reaction, osteophyte formation, and other spondylotic changes that culminated in mechanical deterioration and central stenosis. The original theory was devised with the lumbar spine in mind and has less implications for the thoracic region, which is relatively limited in motion and stable on account of the surrounding rib cage. However, as mobility increases toward the thoracolumbar junction so too does the predilection for TSS,12,13 underscoring the idea that recurring motion and microtrauma are at least contributory, if not causal, in the pathogenesis of thoracic stenosis as well. While TSS can exist anywhere along the thoracic spine, a disproportionate number of cases involve caudal segments, namely T10 to T12.14,15 This is especially true when ossification of the OLF is responsible for TSS.16,17 Higher incidence of spinal degeneration at the thoracolumbar junction is presumably related to increased mobility as aforementioned. Two factors account for this: first, there is gradual enlargement of the spinal canal in the vicinity of T10 to T12 with transition to floating ribs that lack thoracic articulation ventrally and, second, there is increased axial rotation due to orientation of the zygapophyseal joints.4 Apropos to the latter, Maigne et al demonstrated that the caudal thoracic spine constituted somewhat of a transitional zone in which vertebrae could take on either a lumbar or thoracic configuration. Those that were more thoracic in anatomical structure permitted greater degrees of rotational freedom and an increased likelihood for TSS from OLF. Conversely, those that were more lumbar-like in shape restricted rotation and were associated with decreased rates of TSS secondary to OLF degeneration.13 Similar reports detailing differential vertebral configuration in the distal thoracic spine have been confirmed by others albeit with different results: three-dimensional studies of cadaveric thoracic vertebrae by Punjabi et al determined that T10 to T12 vertebrae exhibiting a lumbar phenotype were more predisposed to spondylosis than their thoracic counterparts.4,12 OPLL and OLF are two of the most common causes of TSS. Whereas OPLL predominates in the mid- to upper thoracic spine, OLF and disc herniation account for the majority of cases in the lower thoracic spine.18,19 The genetic factors and cellular processes at play in spinal degeneration and abnormal ossification of spinal ligaments remain obscure, but there have been advances over the years. Recent work has identified bone morphogenetic protein-2 and tissue transglutaminase-2 as key regulators in heterotopic ossification of ligaments.20 Other key players in chondrogenesis, osteogenesis, and bone mineralization have also been implicated.21 In addition, roles for various proinflammatory cytokines, such as TGF-beta, IL-6, IL-12, IL-18, TNF, and VEGF22 have been established. Polymorphisms for various genes (vitamin D receptor and hyaluronan and proteoglycan link protein 1/HAPLN1) have been linked to intervertebral disc failure and spine degeneration.23,24 The finding that tensile train was shown to modulate expression of ossification signaling factors (beta-catenin, Runx2, Sox9, and osteopontin) in cultured human ligamentum flavum cells25 underscores this idea that biomechanical changes within the spine can, in and of themselves, lead to an overhaul of molecular expression at a cellular level. It is conceivable then that repetitive microtrauma, height loss from intervertebral disc deterioration, or some other precipitant could create perturbations in the strain experienced by facet joints and intraspinal ligaments, ultimately serving as a catalyst for their hypertrophy or ossification and overall spine degradation. The average age of presentation for thoracic stenosis is in the 50 s and prevalence is known to increase with age.18,26,27,28,29 TSS is more common in the male gender (~ 2:1) and people of Japanese or East Asian descent.18,27,28,30 Presenting symptoms vary with the anatomical level of compression and degree of canal compromise: some present acutely with paraplegia, whereas others exhibit myelopathy with a more insidious onset with vague symptoms persisting for months to years. Such diversity in presentation, coupled with the fact that neurological manifestations are not entirely dissimilar from symptoms encountered in cervical or lumbar disease, makes correct identification of TSS challenging and explains the tendency toward delayed diagnosis or misdiagnosis.31 Lesions in the upper thoracic spine can present with upper motor neuron symptoms, anywhere along a spectrum from radiculopathy, to myelopathy, to complete motor and sensory loss below the affected level. Conversely, lower thoracic involvement can produce a hybrid picture of both upper and lower motor neuron signs due to impingement on the conus medullaris and upper aspect of the cauda equina.32,33 General complaints in TSS include but are not limited to back pain, neurogenic claudication, torso and abdominal radiculopathy, lower extremity sensory and motor disturbance, gait abnormalities, bowel, bladder, and sexual dysfunction. Of these, lower extremity motor and sensory deficits are the most frequently reported on initial presentation.18,33 In one of the largest retrospective studies of a TSS cohort to date, 81% of patients exhibited motor dysfunction, followed closely by sensory disturbance in 64%.18 Sphincter dysfunction is evident in only a small proportion of patients and is not a prototypical feature of this disorder.34,35 However, sphincter dysfunction and saddle anesthesia are more likely when there is involvement of the conus medullaris.36 On physical examination, patients may exhibit upper motor neuron signs in the lower extremities, including increased spasticity, hyperactive knee and ankle reflexes, and pathological responses (e.g., Babinski, Chaddock signs). In fact, patellar and Achilles reflex changes have been observed in the majority of patients (70–85%) in some series.18,28 If stenosis involves the T10 to T12 segment, the physical examination may be complicated by a mixed picture in which there are both upper and lower motor neuron signs in which patients may exhibit patellar tendon hyperreflexia and concomitant Achilles’ tendon hyporeflexia.5 TSS symptoms can be intermittent if there is dynamic instability at a compressive segment.37 Several theories have been proposed to explain symptomatology: neurogenic compression theory and vascular compression theory. The former states that symptoms are the direct consequence of mechanical compression of neural elements, while the latter posits that impaired vascular flow at the stenotic level is the root cause.38 The higher incidence of severe motor rather than sensory deficits reported in the literature lends credence to a vascular hypothesis; the robust vasculature supplying posterior columns more readily resists compression than the tenuous network of anterior and lateral spinal artery perforators which create a watershed zone in the vicinity of anterior gray matter.4 Although there is no objective data to support one theory over another, it is plausible that both contribute to the eventual neurological decline seen in TSS. The advent of widely available imaging modalities like magnetic resonance (MR) and computed tomography (CT) have increased the detection rates for spinal stenosis. A diagnosis of TSS can thus be readily rendered so long as clinical suspicion prompts further imaging. Stenotic patterns vary from single to multiple, contiguous or remote, levels throughout the thoracic spine. While TSS can occur in isolation, it is usually accompanied by stenosis in other areas ( Plain radiographs of the thoracic spine with anteroposterior and lateral views can generally be obtained with relative ease and at low cost in a clinic setting and are useful as a screening test. Although somewhat limited in their visualization of the spinal canal because of obstruction from the shoulders and overlying ribs, lateral views can reveal changes that intimate underlying degeneration and stenosis, including disc height loss, shortened pedicles, exuberant hyperostosis, or abnormal ossification ( CT is the preferred technique for depicting osseous anatomy and calcific changes. OPLL, OLF, and calcified discs are accordingly readily identified ( MR remains the gold standard for diagnosing spinal stenosis because of its high sensitivity, noninvasive functionality, and clear delineation of canal contents, disc pathology, and marrow abnormalities. Additionally, assessment for the attendant consequences of compression, such as cord edema or volume loss, allows for predictions regarding the significance of canal narrowing and informs the timing of any surgical interventions. Evaluation for thoracic stenosis is best conducted by systematically reviewing axial and sagittal T2-weighted MR images level by level. Radiographic findings vary with the underlying process at play and degree of neurovascular compression, which spans the gamut from mild deformation of the thecal sac to severe cerebrospinal fluid (CSF) effacement, cord signal change, and myelomalacia. Osteophytes, calcified discs, and calcified ligaments are recognizably hypointense on T1- and T2-weighted fast spin-echo sequences (
12.1 Introduction
12.2 Definition of Thoracic Spinal Stenosis
12.3 Pathophysiology
12.4 Clinical Presentation
12.5 Diagnostic Modalities
Fig. 12.1). In a recent retrospective analysis of 427 cases of thoracic stenosis, coexisting disease was identified in the cervical and lumbar regions in 15 and 11% of patients, respectively.18 High clinical suspicion should therefore be maintained for tandem pathologies, particularly when the presenting symptoms that led to a diagnosis of TSS are not entirely explained by available radiographic evidence.
Fig. 12.2). If clinical symptoms are sufficiently concerning, the clinician may forego plain radiographs in favor of more comprehensive imaging modalities.
Fig. 12.1, 
Fig. 12.3a, b). Image acquisition is fast and multiplanar reformats can greatly aid surgical planning. CT is particularly helpful in distinguishing ligament hypertrophy from ossification. Use of intravenous contrast enhancement can also facilitate a diagnosis of disc herniation if there is sufficient thoracic stenosis to impair flow within the epidural venous plexus. However, because compression of neural elements is inferred from reduced canal or neuroforaminal cross-sectional areas, and not directly visualized, it makes CT an unreliable modality for detecting neural impingement. Moreover, it carries the unwanted risks of exposure to ionizing radiation. As a result, it is not ideal for diagnosing TSS and is reserved for scenarios in which MR is contraindicated or where images would otherwise be too degraded from susceptibility artifact caused by implanted hardware.
Fig. 12.3c, d, 
Fig. 12.4). Because the facet joints are oriented coronally in the thoracic spine, hypertrophy produces posterolateral compression of the thecal sac which can lead to a trefoil-shaped canal.32 Finally, gadolinium-enhanced T1 sequences are particularly useful in the detection of infection and neoplasm.
Thoracic Key
Fastest Thoracic Insight Engine
