Adult Thoracic and Lumbar Deformity

Clinical Presentation and Classification

There are two general categories of adult patients with thoracolumbar spinal deformity. The first category includes patients who previously had a spinal deformity at a much younger age as an infant, during early childhood, or as an adolescent ( Fig. 172.1 ). Patients may become more symptomatic during adulthood as a result of degenerative changes. Patients with previous idiopathic scoliosis often have more complicated curvature that spans more segments and involves both the lumbar and thoracic spine. In addition, these curves may have significant rotational components to them. The second category is patients with de novo deformity development. Patients with de novo deformity may present with degenerative spine disease symptoms including neurogenic claudication and painful or non-painful radiculopathy. The addition of progressive sagittal or coronal deformity can add intractable axial back pain to the list of symptoms. De novo curves are often centered in the lumbar region, focally span fewer segments, and may involve listhesis or focal rotation. Differentiating between these two major categories of deformity can help determine a treatment strategy in either surgical or conservative management.

It is also important to understand the underlying pathogenesis behind thoracic and lumbar deformity. The cause is likely multifactorial, with numerous degenerative changes that occur and accumulate over time, including asymmetrical disc degeneration, dehydration, and collapse, resultant progressive facet degeneration, and ligamentous laxity. ^, The above processes may occur symmetrically or asymmetrically, which can lead to more rapid progress of deformity. An important consideration is iatrogenic causes of spinal deformity, which can happen after laminectomy, decompression, or short-segment lumbar fusion procedures, resulting in a loss of normal or physiologic lumbar lordosis (LL). These two factors are a common and important cause of sagittal malalignment in patients with spinal deformity.

Numerous classification schemes have been devised for discussion and treatment planning for thoracolumbar deformity. In 2005, Aebi described a cause-based classification system that included three types: (I) deformity caused by asymmetric degenerative changes; (II) adolescent idiopathic scoliosis (AIS); and (III) scoliosis secondary to extravertebral abnormalities including pelvic inclination or resulting from osteoporotic compression fractures. This classification was useful in describing natural history but not detailed features of individual deformity. In 2006, the Scoliosis Research Society (SRS) devised a system based on radiographic features. In the same year, Schwab et al. devised a system to describe adult degenerative scoliosis that focused on the apex of the curve, evaluation of LL, and factoring in vertebral subluxation. Importantly, this was the first report that lower curve apex and loss of LL resulted in lower health-related quality-of-life (HRQOL) scores. In 2012, the Schwab classification was combined with the SRS classification, and this is the current classification scheme used today, which has been validated on an independent cohort ( Fig. 172.2 ).

The Lenke Classification was devised to describe operative AIS with the goals of being comprehensive for all curve types while providing a 2D description of the deformity. There are six curve types, three lumbar spine modifiers, and a single sagittal thoracic modifier ( Fig. 172.3 ) in the overall classification. This classification has better interobserver and intraobserver reliability than previously described classification schemes for AIS.

Epidemiology

The prevalence of adult spinal deformity is increasing in North America because of an aging population with increased life expectancy and increased recognition and detection. Adults with de novo, or degenerative, curvature present with pain in up to 90% of cases. The prevalence of adult scoliosis ranges from 1% to 30%. ^, Degenerative scoliosis typically presents at a mean age of 70.5 years, in the lumbar spine, and with coronal curvature greater than 10 degrees with the apex of the curve at L3 and an associated distal fractional curve between L4 and S1. With the primary lumbar curvature, there is often a compensatory (secondary) thoracic curvature. More severe structural deformities can be associated with degenerative scoliosis that can manifest as lateral listhesis or rotatory subluxation, most commonly at L3–L4. The prevalence of these degenerative lumbar curves is 63% for 10-degree curves, 44% for curves 10 to 20 degrees, and 24% for curves greater than 20 degrees, and they can occur equally between males and females. ^, Although the diagnosis of degenerative scoliosis is commonplace, emphasis has been placed on those curves that are most likely to progress: those with a Cobb angle greater than 30 degrees, apical rotation greater than Nash-Moe grade II, lateral listhesis >6 mm, and an intercrest line through L5. Interestingly, there has been no association between age or sex and curve progression in this group.

Gait difficulty and progressive functional decline can be secondary to hyperkyphosis. The prevalence ranges from 20% to 40% in elderly patients, although the differences among normal and abnormal thoracic spine alignment and hyperkyphosis have not been well defined. Thoracic kyphosis increases approximately 3 degrees per decade. Although there is a trend toward increased prevalence in older age groups, other risk factors have not been well elucidated. The most commonly referenced cause for hyperkyphosis is vertebral compression fracture, which can occur in approximately 40% of older patients with the greatest degree of kyphosis.

Coronal Alignment

Alignment in the coronal plane is measured as the horizontal distance between a line drawn vertically through the center of the sacrum (central sacral vertical line; CSVL) and a plumb line dropped from the center of C7. The C7 plumb line that is situated to the left or right of the CSVL indicates a negative or positive coronal alignment, respectively ( Fig. 172.5A ). When addressing coronal curvature on posterior–anterior radiographs, it is important to define the apex of the curve, which is the disc or vertebra maximally displaced from the midline and minimally angulated. Thoracic, thoracolumbar, and lumbar curves have their apex between T2 and T11–T12, between T12 and L1 vertebra, and caudal to the L1–L2 disc, respectively. When discussing curvature in the coronal plane, the nomenclature dextroscoliosis is defined as a rightward curve, and a levoscoliotic curve is a leftward curve. By convention, the largest curve, or major curve, is measured using the Cobb technique from the maximally angulated (end) vertebra above and below the apex.

Three important terms in relation to global deformity should be considered when planning thoracic and lumbar deformity correction ( https://www.srs.org/professionals/online-education-and-resources/glossary/revised-glossary-of-terms ). The SRS defines a neutral vertebra as a vertebra without axial rotation, which is in reference to the most cephalad and caudal vertebrae that are not rotated in a curve; the apical vertebrae as the one most deviated laterally from the vertical axis (CSVL); and the end vertebra as the one at the end of a curve in a coronal or sagittal projection. The cephalad end vertebra is the first vertebra in the cephalad direction from the curve apex whose superior surface is tilted maximally toward the concavity of the curve. Conversely, the caudal end vertebra is the first vertebra in the caudal direction from a curve apex whose inferior surface is tilted maximally toward the curve concavity.

Adjacent curves are minor and may be compensatory or structural. To distinguish these two phenomena, side bending radiographs must be obtained in the supine position. Compensatory curves will reduce to less than 25 degrees and structural curves will not reduce to less than 25 degrees. Compensatory curves will often correct upon correction of the primary curve, whereas structural curves should be incorporated into the surgical construct.

Pelvic Obliquity

Pelvic obliquity is an offset of the pelvis and can be a primary deformity issue, a result of leg-length discrepancy, or secondary to lumbar scoliosis. It is important to identify pelvic obliquity and, more importantly, understand the relationship between pelvic obliquity and deformity to prevent improper correction of malalignment, specifically suboptimal coronal realignment (see Fig. 172.5B ).

Thoracolumbar and Lumbar Alignment

Thoracolumbar and lumbar alignment are measured on lateral 36-inch scoliosis films with the patient facing to the right. Measurement of thoracic kyphosis comes from the superior endplate of T2 or T5 and the inferior endplate of T12—this is reported as a positive value (see Fig. 172.5C ) whose normal range is 20 to 40 degrees. LL is an additional measure of importance in deformity and is measured from the inferior endplate of T12 or L1 to the superior endplate of S1 (see Fig. 172.5C ); its normal range is 20 to 45 degrees.

The sagittal vertical axis (SVA) is a measure of global sagittal alignment and is determined by dropping a plumb line from the body of C7 to the posterior superior corner of the sacrum (see Fig. 172.5D ). If the C7 plumb line falls anterior to the sacrum, the SVA is positive, and if it falls posterior to the sacrum, the SVA is negative. An additional measure is the T1 or T9 spinopelvic inclination (T1SPI and T9SPI), which is not dependent on the image scale (see Fig. 172.5D ). The T1 pelvic angle, introduced by Protopsaltis et al., is the angle between lines connecting the femoral head axis to the center of T1 and the midpoint of the superior endplate of S1: it is the sum of the T1SPI and pelvic tilt (PT). The T1 pelvic angle does correlate with pain and disability, with an angle greater than 20 degrees corresponding to severe disability (Oswestry Disability Index score >40).

Pelvic geometry and alignment have been previously demonstrated as strong contributors to overall sagittal spinal alignment. ^, The three pelvic parameters that have the most effect are pelvic incidence (PI), PT, and sacral slope (SS) ( Fig. 172.6 ). The PI is a fixed parameter that does not change after development and is independent of pelvic position. It is measured as the angle between the femoral heads and a line drawn perpendicular to the midpoint of the sacral endplate (see Fig. 172.6 ). This can help in calculating lumbopelvic mismatch by subtracting LL from the PI, a discrepancy that should be restored during deformity correction surgery within 10 degrees. Another important measure is PT, which is a marker of retroversion of the pelvis, which is a compensatory response to positive sagittal alignment. Thus, the increasing retroversion equates to increased PT. PT is measured as the angle between a perpendicular line drawn from the femoral heads to the midpoint of the sacral endpoint and a vertical line upwardly taken from the femoral head axis (see Fig. 172.6 ). The SS is the angle between the endplate of the sacrum and a horizontal line (see Fig. 172.6 ). The relationships among the three parameters are as follows: