Posture
Posture is the relationship of parts of the body to a vertical line passing through the center of gravity. In practice, the term posture is almost always used to describe spinal relationships usually in the sagittal plane, although certain nonpathologic conditions unrelated to the spine (e.g., flatfeet, tibial torsion) could be considered postural variations. Posture in the upright or erect position develops from the action of antigravity muscles, such as the erector spinae and gluteus maximus, on the axial skeleton of the infant and toddler.
The spine is held in generalized flexion in the newborn, and does not change significantly in the prewalking stage. In a sitting infant, the thoracolumbar spine continues to be flexed as the axial weight-bearing line falls anterior to the axis of rotation of the spine in the sagittal plane. The pelvis and hips are flexed.
As the child begins standing upright, the antigravity muscles produce postural spinal curves, and the normal sagittal contours of the spine begin to appear. Because the head is anterior to the axis of rotation, a cervical lordosis develops to move the center of gravity posteriorly. Because the hips and pelvis are in a flexed position, the erector spinae muscles must act against these flexed parts to put the lower extremities in a more vertically aligned weight-bearing position. The result is an increase in lumbar lordosis. The thoracic portion of the spine remains in a kyphotic position, unchanged from the infantile flexed position.
The normal range of thoracic kyphosis is considered to be 20 to 45 degrees (Cobb angle). Depending on age of evaluation and ethnicity, 50 degrees may be the upper limit of normal. Hyperkyphosis exists when this upper limit is exceeded. This range usually is exceeded in postural round-back deformity and Scheuermann kyphosis.
Causes of Hyperkyphosis
Round-back deformity in adolescence has been known in the medical literature since the nineteenth century. Before radiography became available, this deformity was believed to be secondary to muscular deficiencies or congenital anomalies. In addition to round-back deformity and Scheuermann kyphosis, other conditions that can cause hyperkyphosis include congenital kyphosis, which is diagnosed by radiographic demonstration of defects in vertebral body formation or segmentation, and the rarer variation, progressive noninfectious anterior fusion. Other causes of hyperkyphosis that can be identified and distinguished on the basis of the history and physical examination findings include trauma, infection, postlaminectomy and postirradiation conditions, metabolic disease, and skeletal dysplasias ( Box 13-1 ). Because many of these conditions may threaten a patient’s overall health and spinal cord function, correct and timely diagnosis is important.
Postural
Infantile resolving
Adolescent
Scheuermann disease
Congenital
Failure of formation
Failure of segmentation
Mixed failure
Progressive noninfectious anterior fusion
Traumatic
Structural bone or ligament failure
Secondary to paralysis
Neuromuscular
Myelomeningocele
Developmental (secondary to paralysis)
Congenital
Postlaminectomy
Postradiation therapy
Metabolic
Osteoporosis
Osteomalacia
Osteogenesis imperfecta
Skeletal dysplasia
Achondroplasia
Mucopolysaccharidosis
Neurofibromatosis
Collagen disease
Marfan syndrome
Neoplastic
Benign
Malignant
Primary
Metastatic
Postinfectious
Bacterial
Fungal
Tuberculous
Postural Kyphosis
Postural round-back deformity of adolescence is a benign condition that is correctable by means of passive and active forces. Patients with postural kyphosis have a flexible deformity much different than those with Scheuermann kyphosis ( Fig. 13-1 ). They frequently can reduce their kyphosis by actively contracting the erector spinae muscles and can flatten their lumbar lordosis with the abdominal muscles. In addition, the vertebrae do not exhibit the radiographic changes described in Scheuermann disease.
Postural kyphosis usually responds to parent and patient education and nonoperative treatment. A series of muscle-strengthening and stretching exercises may be helpful for the trunk, abdomen, shoulder girdle, and lower extremities. Orthotic management also can be used for extreme cosmetic deformity (see Fig. 13-1 ). Because the natural history and long-term sequelae of postural hyperkyphosis generally are benign, more aggressive treatment (including surgery) is never indicated for this condition.
A form of infantile thoracolumbar kyphosis associated with hypoplasia of L1 or L2 vertebra has been described. It is discussed here because of its benign natural history and resolution of the vertebral body “wedging” once the sitting infant becomes fully ambulatory and the infant’s postural kyphosis undergoes transition to a normal upright posture. Apparently, however, this benign condition can persist long enough to require posterior spine fusion. Thus a period of observation is indicated to avoid unnecessary surgery on the benign, resolving form.
Scheuermann Kyphosis
In 1920, Scheuermann described a specific type of fixed angular kyphosis with anterior wedging of the vertebral bodies and irregularities of the vertebral apophyses. When seen with the typical clinical appearance of a sharper angular kyphosis, the condition bears his name, Scheuermann kyphosis. This deformity, initially described only for the thoracic spine, also can occur in the thoracolumbar and lumbar spine ( Fig. 13-2 ). Thoracic Scheuermann kyphosis has an apex between T7 and T9, whereas the apex of the thoracolumbar form is between T10 and T12 (see Fig. 13-2 ). Lumbar Scheuermann disease, characterized primarily by the typical radiographic changes associated with Scheuermann kyphosis, lacks the clinical deformity because of its location (see Radiographic Findings ).
Scheuermann kyphosis occurs in 0.4% to 10% of adolescents between 10 and 14 years of age. Some investigators report an equal number of male and female cases, whereas others report either a male or female preponderance. The condition has its onset in the prepubertal growth spurt and becomes apparent at around 10 to 12 years of age. The typical vertebral wedging, the sine qua non of the diagnosis, has not been reported in children younger than 10 years.
Originally, the wedging was thought to be due to lack of development of the vertebral ring apophysis, and that after ossification (approximately 10 years of age), the deformity—a trapezoid-shaped wedging—could be seen radiographically (see Fig. 13-1 , C ). However, the vertebral apophysis does not actually contribute to the longitudinal growth (height) of the vertebral body, and thus damage to it by vascular or mechanical mechanisms would not cause the characteristic wedging. Disorganized endochondral ossification in the vertebral end-plates, the true site of growth, is consistently found and has been likened to the histopathology of Blount disease or slipped capital femoral epiphysis. The true cause of the wedging deformity and the reason why it does not appear until after 10 years are unknown.
Etiology
Based on radiographic findings, Scheuermann and others considered the kyphosis to be a form of osteochondritis. Mechanical, vascular, hormonal, nutritional, traumatic, and metabolic causes have all been proposed over the years; a summary is provided by Robin, to which the reader is referred. More recent evidence, including reports of monozygotic male twins, suggests a genetic cause for Scheuermann kyphosis. Despite significant investigation, however, the cause of Scheuermann disease remains unknown. There have been many discrepancies and nonconfirmatory findings in histopathologic specimens obtained during surgery. Of interest is that no signs of juvenile osteoporosis or other bony metabolic problems have been documented, and the typical disorganized endochondral ossification probably is a result rather than a cause of the condition. Patients with Scheuermann are taller, heavier, and have an increased body mass index compared with age-matched controls, although these body habitus parameters have no correlation to the severity of the kyphosis, and are thus likely the result of the hormonal or whatever genetic disturbance is associated with Scheuermann rather than the cause. Typical thoracic or thoracolumbar Scheuermann kyphosis remains idiopathic, whereas a traumatic origin is accepted for the lumbar form.
Clinical Features
Beside the clinical deformity—a sharp-angled kyphosis best viewed from the side during an Adams forward-bending test ( Fig. 13-3 )—patients with Scheuermann kyphosis often present with pain at the kyphotic apex, and also at the cervicothoracic and thoracolumbar junctions. Apical pain is ascribed to discogenic origin based on MRI imaging showing narrowed and degenerated end-plates and disk contents. Junctional pain probably represents stresses in segments that are mobile (cervical, lumbar) adjacent to the rigid kyphotic segment. These segments develop compensatory hyperlordosis, which in itself can be painful because of facet impingement, and in the neck, forward protrusion of the head develops ( Fig 13-4 ) when the cervical lordosis maximum is exceeded. Anterior shoulder/pectoral muscle and hamstring contracture may be present. The patient’s inability to correct the kyphosis by “standing at attention” completes the clinical picture. A small scoliosis may accompany the hyperkyphosis.
In the lumbar form of the condition, patients generally are male and backache is typically the initial complaint. There usually is no deformity of the lumbar spine. However, the typical radiographic findings seen in the thoracic and thoracolumbar forms are present in the lumbar form as well. They include irregular vertebral end-plates, a reduction in disk height, and anterior vertebral defects (Schmorl nodes), the latter which can be dramatic ( Fig. 13-5 ). These findings are diagnostic of the condition. The radiographic picture together with the invariable complaint of back pain is consistent with a repetitive traumatic cause. Patients typically are athletic or perform hard labor in a repetitive, overuse manner. Because Schmorl nodes and the anterior defects are seen over several segments of the lumbar spine, the association of the lumbar form of Scheuermann disease is well accepted with overuse syndromes.
Although degenerative osteophytes have been noted on radiographs, the area of osteophytic degeneration frequently is mobile rather than ankylosed. Pain in apical areas does not seem to correlate with the magnitude of the kyphosis. Indeed, the most recent long-term natural history study of Scheuermann patients from Finland found increased risk of pain and disability in patients averaging only 45-degree thoracic kyphosis, barely enough angular deformity to fall within the Scheuermann definition.
Finally, progression of the kyphosis, which might be expected to have a high correlation with increasing symptoms, is poorly related to pain. Patients in one series had no increase in symptoms with progression. Those with greater deformities chose less strenuous jobs and activities and seemed to have more pain than those with lesser deformities, but the pain did not limit their daily work activities.
Late painful sequelae are uncommon in patients with kyphosis of less than 75 degrees, although there may be an increase in the incidence of degenerative disk disease and spondylolysis in the lumbar spine caudal to the kyphotic deformity. Hyperlordosis caudal to the kyphosis is probably the precursor to these conditions.
However, quality of life evaluations in hyperkyphotic adolescents (mean Cobb 70 degrees) clearly identify deficits in self-image, general function, and activity levels compared with normal controls. Kyphosis exceeding 75 degrees (“severe”) had a significant activity decrease compared to those with “moderate” kyphosis, suggesting this degree of deformity as a “tidemark” for possible active treatment.
Radiographic Findings
The diagnostic criteria of Sörensen define the condition ( Box 13-2 ; see Figs. 13-1 and 13-2 ). In some patients these findings are seen only at the apex of the kyphosis, whereas in others the findings may be spread throughout the entire extent of the curve. Any evidence of actual bony ankylosis or fusion, especially in the more anterior portion of the disk spaces, changes the diagnosis to a form of congenital kyphosis or progressive noninfectious anterior fusion.
- •
More than 5 degrees of anterior wedging of three consecutive adjacent vertebral bodies in the apex of the kyphosis
- •
Irregular vertebral apophyseal lines combined with flattening and wedging
- •
Narrowing of the intervertebral disk spaces
- •
Schmorl nodes variably present
Additionally, patients with Scheuermann kyphosis who are undergoing surgery (and thus presumably have greater disruption of their sagittal balance) demonstrate negative sagittal balance on standing radiographs—that is, the C7 plumb line falls more than 2 cm posterior to the sacral promontory (see Fig. 13-2 ). Normally, the combination of cervical lordosis, thoracic kyphosis, and lumbar lordosis produces a C7 plumb line falling within 2 cm anterior or posterior to the posterior edge of the sacral promontory. Patients with significant thoracic hyperkyphosis often have lumbar hyperlordosis, and the combination of increased dorsal displacement of the thoracic apex because of kyphosis and the compensatory lumbar hyperlordosis produce the overall negative sagittal balance.
In lumbar Scheuermann disease, wedging of the vertebral bodies is usually minimal and produces little disturbance in the sagittal plane. Late degenerative changes may result, especially when diskography has demonstrated tears in the annular ligaments of involved lumbar disks. Spondylolysis at the lumbosacral junction (with or without mild degrees of spondylolisthesis) has been reported to be more prevalent in patients with Scheuermann disease than in the normal population, although the Iowa long-term study disputes this commonly accepted finding. The possibility of spondylolysis must be considered when a Scheuermann patient complains of low back pain. Although hyperlordosis itself is considered a cause of low back pain in Scheuermann kyphosis, other causes, such as diskogenic and spondylolytic pain, must be ruled out.
Nonoperative Treatment
Indications
There is a serious lack of natural history data on Scheuermann kyphosis. Suffice it to say that the common belief that patients have few symptoms and the deformity remains cosmetically acceptable is poorly documented. Patients in the Iowa review (32-year follow-up) had increased back pain and fatigue, selected less strenuous employment, and had greater interference with activities of daily living than controls, yet were “not disabled” by their condition. The more recent Finnish study (37-year follow-up) had similar findings regarding pain, daily disability, quality of life, and general health deficits despite a smaller degree of deformity. The associated low SRS-24 scores for adolescents with 75 degrees of deformity suggests that the untreated natural history is more pessimistic than the long-term data might predict.
An exercise program to counteract the decreased back extensor and abdominal muscle strength is generally recommended for most patients with Scheuermann kyphosis. More aggressive treatment is appropriate for individuals with significant symptoms or progressive deformity or when the deformity occurs in an immature adolescent.
Indications for nonoperative treatment include relative skeletal immaturity (Risser grade 2 or less) and a deformity that is already cosmetically or functionally unacceptable (usually >60 degrees) and is known to be progressive. Some authors believe that virtually any Scheuermann deformity can be managed nonoperatively in a skeletally immature patient. The goal of nonoperative management is twofold: to control the deformity and to attempt to reconstitute the anterior vertebral height by applying hyperextension forces (by means of a brace, for example). Without evidence of such reconstitution, loss of correction after discontinuation of brace therapy is almost certain.
Brace Therapy
For a thoracic apex, the Milwaukee brace is recommended because it is the only orthosis that can effectively apply a three-point corrective force to a midthoracic apical vertebra ( Fig. 13-6 ). The brace should also decrease the lumbar lordosis and help correct the negative sagittal balance. For a thoracolumbar deformity, a thoracolumbo-sacral orthosis (TLSO) can be tried, usually supplemented with anterior sternal or infraclavicular outriggers to provide an extension moment cephalic to the apex. With a decrease in lumbar lordosis, the patient is encouraged to actively hyperextend the spine to maintain the head in a more upright position. An advantageous feature of the Milwaukee brace that is not available with the TLSO is the ability to progressively bend more correction into the posterior kyphosis pads over time.
Initially, bracing should be done full-time, with the patient allowed to remove the brace 1 to 2 hours per day to perform exercises. Radiographs may be obtained every 4 months, with progressive correction bent into the posterior kyphosis pads as tolerated. Brace treatment should continue until skeletal maturity is achieved, which for boys may require that they wear the orthosis until Risser grade 5. Although a weaning period from brace wear usually is recommended, there is no evidence that a particular weaning schedule is more efficacious than another.
The results of orthotic management of Scheuermann disease show that the deformity can be effectively improved during brace wear; however, when brace wear is terminated, loss of correction occurs. Sachs and associates reported that larger deformities (>74 degrees) at the start of treatment showed the most significant loss of correction after brace discontinuance, with the result that there was little overall correction. This finding suggests that orthotic management is perhaps not indicated for a deformity of large magnitude without some previous correction of the deformity—by casting, for example. Although smaller deformities can be maintained at a smaller magnitude with orthotic treatment, greater deformities need greater initial correction to be stabilized, thus requiring a more aggressive treatment approach.
Brace treatment is used relatively infrequently at our institution because of the generally benign larger magnitude of deformity at presentation, and the problem of brace fit and compliance in an adolescent already with self-image deficit. For immature patients, the Milwaukee brace is used most frequently, regardless of the curve pattern, because we have found the use of a TLSO less effective.
Cast Treatment
When passive correction is less than 40%, as determined from a hyperextension lateral radiograph over a bolster or on clinical examination, brace treatment is not likely to be effective. Antigravity or localizer casts can be applied in serial fashion to produce more correction of the kyphosis ( Fig. 13-7 ). This treatment regimen, used extensively in Europe, entails applying two or three casts (changed every 2 to 3 months) in an attempt to progressively correct the deformity. After the 6- to 9-month period of casting, the patient is treated with a Milwaukee or other type of retention brace to maintain the correction during the remainder of growth. With such a regimen, not only is the deformity improved by as much as 40%, but there also is less loss of correction. In a 2-year follow-up French study, the loss of correction averaged just 7 degrees, whereas in a 3-year follow-up series reported by Ponte and associates, only 4 degrees of correction was lost.
The European experience with cast treatment has led to the following observations: (1) The long-term goal of controlling the deformity so that the curve ultimately ends up in a physiologic range (<50 degrees) is achievable. (2) The later that treatment is begun, the lower the probability of reconstituting the anterior vertebral height and maintaining any significant angular correction. (3) For patients initially seen after puberty, with little growth remaining, nonoperative treatment cannot correct vertebral wedging and thus probably should not be attempted. In such cases it may be preferable to correct the deformity surgically, if indeed treatment is indicated.
The use of cast treatment in an adolescent rests largely on the patient’s desire to achieve maximal correction without resorting to surgery. Because of the prolonged and relatively inconvenient treatment period (exceeding 1 year, with 6 to 9 months in casts and a minimum of 6 additional months in a brace), such therapy will never succeed without the total compliance and desire of the adolescent.
Operative Treatment
Indications
Surgical treatment of Scheuermann kyphosis is reserved for patients with pain, a rigid deformity, a curve of more than 70 to 75 degrees, and an unacceptable cosmetic appearance with a deteriorating self-image. Although some have advocated surgical correction for a kyphotic deformity of as little as 50 degrees, the less serious natural history and lack of symptoms in patients with 60 degrees of kyphosis strongly argue against “prophylactic” surgery. Furthermore, as the long-term results of surgical treatment are lacking, the case for such prophylaxis is totally undocumented. If the deformity is painful, nonoperative pain management (e.g., general fitness, extension, and shoulder- and hamstring-stretching exercises) probably should be attempted for a minimum of 6 months before resorting to surgery. Patients with complaints about cosmesis but a nonprogressive deformity should be referred for psychological consultation before the physician acquiesces to surgical management. Although operative treatment may be appropriate for a patient with significant self-image concerns, it is probably inappropriate for a patient whose postural “problem” is exacerbated by depression, anxiety, and low self-esteem.
Surgical Goal
The goal of surgical correction of hyperkyphosis is to achieve a stable, balanced spine in the sagittal plane by obtaining stable correction without neurologic complications. The degree of curve reduction should be planned relative to the patient’s overall sagittal balance. The deformity may require more or less reduction of the curve to achieve balance in any individual patient. According to Lowe, a kyphosis should never be reduced more than 50% of the preoperative deformity, both to prevent neurologic complications and to avoid junctional kyphoses at the ends of the fusion. Maximum correction of kyphosis, exceeding 50% for example, is actually contraindicated due to the likelihood of a junctional kyphosis. Historically, the biomechanical principles of kyphosis correction have included elongating the anterior column of the spine, providing some form of anterior column support, and shortening the posterior column of the spine ( Fig. 13-8 ). Cantilever maneuvers pushing the kyphotic apex ventrally, and direct segmental compression to shorten the posterior column, are current mainstays of treatment. Because of the first two principles, the use of anterior release and fusion has been accepted as part of a standard two-stage corrective procedure. However, the need for the anterior procedure can now be questioned when adequate correction and fusion are achievable by a posterior procedure alone.
Preoperative Evaluation
Preoperative imaging should include a posteroanterior radiograph of the spine to assess for mild scoliosis, common with Scheuermann kyphosis. The lateral radiograph is taken with the patient standing and the arms comfortably resting on a shoulder level support or flexed with fingertips on the clavicles. A lateral radiograph with the patient positioned over a bolster will provide some assessment of the flexibility of the spine and may assist in determining the need for anterior release, especially when the curves are extremely stiff. The need for routine evaluation of the neural axis, determination of the status of the thoracic disks, and assessment for spinal stenosis with MRI of the spine is debatable. Some suggest that MRI is necessary to prevent the rare but significant neurologic problems reported during surgical correction of these patients. A recent multicenter study reported 4 of 79 patients had the surgical plan changed based on preoperative MRI findings.
As with every surgical patient, a patient with kyphosis must have a careful neurologic examination, due to the rare incidence of spastic paraparesis, from either a herniated anterior cord effacement, thoracic disk, or intradural mass (syrinx). Patients with atypical pain that is not mechanical, such as night pain or dysesthetic pain, should undergo MRI to exclude intradural abnormalities (e.g., syrinx). Discogenic sources of pain should be ruled out in patients who have significant radiculopathy, low back pain that is not easily controlled by standard nonsurgical measures, or a history of sciatica or extremely tight hamstrings. A precarious preoperative respiratory status is rare except in the most severe and neglected cases of kyphosis (>100 degrees). Only in such a patient would the choice of surgical approach (anterior and posterior versus posterior alone) be modified or dictated by the preoperative pulmonary status.
Surgical Approach
Patients younger than 16 years treated by posterior fusion for Scheuermann kyphosis can have a gradual postoperative improvement in kyphosis attributed to remodeling of the anterior vertebral wedging. The posterior-only approach originally was recommended for immature patients (<Risser 3) whose kyphosis was relatively flexible (corrected to <50 degrees on hyperextension radiograph) and thus could be expected to gain adequate intraoperative correction plus some additional improvement from remaining anterior growth potential, which could be realized by Hueter-Volkmann unloading of the anterior column. On the other hand, mature patients without such growth potential would presumably lose correction with posterior surgery alone , and it was this observation by Bradford that led to the recommendation that anterior release and fusion was essential to avoid postoperative loss of correction.
A close reading of the two reports by Bradford, however, shows little actual difference in the outcomes of the two sets of patients. The 1975 report after posterior-only fusion, which was concerned primarily with postoperative loss of correction, reported preoperative kyphosis of 72 degrees with a final correction to 47 degrees. In the 1980 report the same surgeons used the two-stage approach to treat patients with an average preoperative kyphosis of 77 degrees; the final average correction was also 47 degrees.
Other reports using earlier-generation posterior compression rod instrumentation have documented adequate correction with posterior-only surgery. Modern series using segmental pedicle screw instrumentation and posterior shortening (Ponte) osteotomies have further confirmed that anterior/posterior surgery provides no advantage over a posterior-only approach when the two methods are directly compared, and thus the additional morbidity of an anterior procedure can be safely avoided.
Determining Fusion Levels
Fusion levels are determined on the standing lateral radiograph. The upper limit of fusion must include the most proximal vertebra that is tilted into the kyphosis. This generally means fusion to T2, especially in patients younger than 15 years. If the fusion stops distal to this level, there is a risk that a postoperative junctional kyphosis will develop at the upper end of the instrumentation (see Fig. 13-4 ). Similarly, the caudal extent of the fusion should include the first lordotic disk space, which commonly includes one level distal to the measured end vertebra of the kyphosis. This usually incorporates the stable sagittal vertebra concept, which determines the “neutral” vertebra in the sagittal plane. Failure to extend into the lumbar lordosis similarly risks creation of a caudal junctional kyphosis. As the apex of the kyphosis is displaced ventrally by deformity correction, the C7 sagittal axis moves ventrally to produce a kyphotic moment at any lumbar level not already in lordosis. If the caudal extent of fusion does not include all nonlordotic segments, junctional kyphosis can result.
Surgical Technique Options
In rigid kyphosis of large magnitude (especially in skeletally mature individuals), anterior release and fusion of the apical portion of the deformity may be advisable to increase prospects for correction through posterior instrumentation. This allows the surgeon to balance the spine more harmoniously and probably improve the rate of fusion. Anterior column fusion with a supporting bone graft is required to maintain any additional correction achieved. However, as noted earlier, recent series have not confirmed the need for anterior release. * Thus, anterior release and fusion may be indicated only in rigid, large magnitude deformity.
Anterior Release and Fusion Technique
* References .
If an anterior release and fusion procedure is elected, it usually is performed as the first stage of a two-stage approach, with both stages generally performed on the same day because the complication rate and morbidity are lower than if the staged procedures are performed 10 to 14 days apart. The release should include the rigid apical segments (as determined on a hyperextension lateral radiograph). The release is performed through a right-sided thoracotomy or thoracoscopically, unless the patient has a left convex scoliosis of sufficient magnitude that a left-sided release is indicated to approach the convexity of the scoliosis. The right side generally is more approachable because the cardiac structures and great vessels fall to the left of the spine.Thoracotomy Approach.
The transthoracic approach should parallel the rib leading to the most cephalic segment to be released and fused. For example, a fifth rib thoracotomy would be used to reach the T5-6 disk space if the latter was the most cephalic extent of the rigid segment requiring anterior release and fusion. The chest wall in patients with rigid Scheuermann kyphosis can be inflexible, and in such a situation double thoracotomies may be necessary to reach the more distal segments.
Once the chest has been entered, the pleura over the spine is opened longitudinally and a flap is created by posterior dissection to expose the costovertebral joints. Segmental vessels are preserved if only an anterior release and fusion procedure is being performed. It is recommended that vessels in the so-called watershed area of the midthoracic space (T4-9) be ligated and divided only after they have been temporarily occluded for 20 minutes and neuromonitoring shows no signal degradation from ischemia. Ligation of these vessels can result in paraplegia as a result of cord ischemia, a condition known as anterior spinal artery syndrome. This clinical recommendation is based on experimental and clinical evidence of a delay in degradation of evoked potentials of up to 20 minutes after anterior spinal artery occlusion. The segmental vessels can be temporarily occluded with rubber vessel loops, which can also be used for retraction during the diskectomy and fusion.
The costovertebral joints should be exposed and the rib heads resected to increase mobility of the spine and to enhance visualization of the posterior disk space. The contents of the disk space should be evacuated, with rongeurs used to remove the nucleus pulposus and curets or elevators used to dissect the vertebral apophysis off the end-plates. Resection should include division of the anterior longitudinal ligament and should proceed posteriorly to the annulus guarding the posterior longitudinal ligament. The rib that has been resected for the thoracotomy approach is morselized and used as an interbody bone graft. The pleura is then closed with running suture to achieve hemostasis and maintain the rib graft in the interbody spaces. Enhancement of the anterior fusion process by creating an osteoperiosteal flap has been described. This can be accomplished after the diskectomy but before closing the spinal pleura. The flap should be created in a lateral-to-anterior direction with an osteotome or other suitable elevator. It provides a bed for a solid anterior column of bone to develop in the apical portion of the kyphosis. In Scheuermann disease, modern posterior instrumentation is capable of sufficiently correcting kyphosis, but a need for anterior column grafting arises if the disk space opens and an anterior defect is created (see “ Posterior Instrumentation and Fusion ,” and Fig. 13-8 , C ). In severe cases of non-Scheuermann kyphosis, such anterior grafting is a crucial part of obtaining a solid arthrodesis.
Routine short segment anterior instrumentation to treat Scheuermann kyphosis has been described by Gaines. For other uses of anterior instrumentation for kyphosis, see “ Postlaminectomy or Postirradiation Kyphosis .”
Thoracoscopic Approach.
Following the success of video-assisted thoracic surgery (VATS) for intrathoracic procedures (see Video 12-1 
, Anterior Thoracoscopic Spinal Fusion and Release), Mack and associates in 1993 reported the application of this new endoscopic technique for spinal surgery. There are both functional and cosmetic advantages to VATS techniques. Functionally, there is less incisional pain and less chest wall and shoulder girdle dysfunction, which hypothetically leads to faster recovery, fewer postoperative respiratory complications in susceptible patients, and shorter hospitalization and rehabilitation times. Cosmetically, the VATS technique requires only small 2-cm incisions, whereas standard thoracotomy requires longer incisions. Few complications were reported in an early series that combined the experiences at three separate centers.
Anterior thoracoscopic diskectomy plus fusion has been reported as part of the two-stage treatment of Scheuermann kyphosis. Using three or four portals, the operator removes the apical five or six disks, with or without segmental vessel ligation. Morselized rib or other bone graft is inserted after curetting of the vertebral end-plates. Clinical series and animal studies have demonstrated that the anterior release technique (i.e., mobilizing the spine to improve correction by posterior instrumentation) results in a correction similar to that achieved with thoracotomy. Fusion rates of 75% are usually achieved although in the presence of a posterior instrumentation and fusion, the significance of an unfused disk space is uncertain.
The theoretical advantages of thoracoscopic release for kyphosis have not always been realized. Recovery times (i.e., shorter hospitalization) were not reduced, and blood loss, volume of chest tube drainage, and operative time were actually greater in the thoracoscopic group than in patients treated by open thoracotomy. However, there are no recent series directly comparing the two techniques because most thoracoscopic surgeons no longer perform open thoracotomy.
Posterior Instrumentation and Fusion
Posterior instrumentation and fusion, alone or in sequence after anterior release and fusion, is performed with the patient on a standard four-poster spinal frame, with the abdomen free. Because exposure of the upper thoracic segments will be necessary, the patient’s head is slightly flexed to facilitate access to T1 if necessary. The spine is exposed through a standard midline incision in the subperiosteal plane out to the tips of the transverse process, at which point segmental instrumentation is inserted according to the intended corrective maneuver.
A three-point cantilever mechanism (see Fig. 13-8 ) corrects the hyperkyphosis by direct pressure on the kyphotic apex. Depending on the length of the instrumentation and the stiffness of the apex, the ventrally-directed force at the apex may meet resistance and produce unwanted dorsally directed forces at the end vertebrae producing the junctional kyphosis, which has been reported in up to one third of patients. Junctional kyphosis can be best avoided by instrumenting long to ensure complete inclusion of proximal end vertebrae (usually T2) and avoiding the use of sublaminar anchoring implants (e.g., wires) where the interspinous ligament is removed and the next proximal segment is destabilized. Distally, junctional kyphosis is best avoided by including the segment below the first lordotic disk, or if any doubt exists, including the sagittal stable vertebra. Finally, overcorrection of the kyphosis (>50%) has also been identified as a source of junctional kyphosis, emphasizing that sagittal balance rather than maximum correction should be the goal of surgery ( Fig. 13-9 ).
