Special Tests

As mentioned previously, the Adams forward-bending test, which is often used as a screening tool, is designed to identify the rotation of the chest wall that occurs in scoliosis. In addition to being used in school screenings, this test should be part of every well-child pediatric visit once a child is able to ambulate. Controversy exists about whether widespread screening leads to excessive specialist referrals for scoliosis.7–9 In the Adams forward-bend test, the bend should exhibit a smooth “spinal rhythm.” Any restriction or hesitancy of normal rhythm or motion, or lack of normal intersegmental motion (lumbosacral, midlumbar, thoracolumbar, thoracic), can usually be easily noted. Loss of spinal rhythm is a much more sensitive indicator of painful intersegmental disorders than is loss of range of motion (ROM), although both loss of rhythm and of ROM can clearly coexist. Abnormalities in spinal rhythm are not specific to any diagnosis, but will always occur when serious structural pathology exists.10 Most children can bend sufficiently forward to extend their fingertips down to within one hand’s length of touching the floor. Failure to extend to this level is abnormal. A child who upon repeat examination is still unable to bend adequately should be evaluated for hamstring contractures. A thorough neurological examination of such children is essential.

In having a child perform the Adams forward-bend test, the examiner should note any curvature of the spine or rib prominence on the side of the convexity of a spinal curve. This is best done by standing directly behind the child and looking in a straight line from the gluteal cleft of the buttocks to the neck. The procedure should be repeated with the examiner looking down from the head toward the buttocks. Previous surgical procedures such as a sternotomy, thoracotomy, or thoracoplasty may distort the chest wall and may cause the bend test to be “falsely” positive. The test is completed by evaluating the child’s bend from both sides. A normal bend of the spine should be smooth, without a sharp peak or a hollow in its midsection when viewed from the side. If these abnormalities are present, they may indicate excessive kyphosis or lordosis.

The scoliometer is an excellent screening device that can be used in conjunction with the Adams forward-bend test to evaluate truncal rotation ( Fig. 5.4 ). The device is a spirit level that when placed at different spinous processes can quantify rotation of the trunk. An angle of less than 7 degrees is considered within the limits of normal. When following a patient with scoliometer monitoring, the same vertebrae should ideally be used for each reading. The inter- and intra-user reliability of scoliometer testing has been evaluated in the literature.11–12 Although there is too much variation in inter-user reliability to permit substituting the Adams forward-bend test and scoliometer readings for routine radiography when a child is followed by multiple practitioners, the intra-user reliability of the test is sufficient to allow for extension of the time between radiographs as long as the patient is followed with frequent clinical examinations done by the same examiner.

Limb-Length Evaluation

Limb-length discrepancy may result in pelvic tilt, which can induce a “compenstory scoliosis.” A child with as much as a 3-cm leg-length discrepancy may have no functional difficulties, and these patients may go undetected until a positive Adams forward bending test prompts further investigation. Conventional teaching recommends absolute limb measurement from the anterior superior iliac spine (ASIS) to the medial malleolus, and relative limb measurement from the umbilicus to the medial malleolus. These measurements fail to include the foot, which in some cases may be up to an inch shorter than the other side due to anatomical differences or postsurgical changes. For this reason the authors recommend including the foot in clinical measurement of leg length. Measurements may be made from the ASIS to the lateral border of the sole of the foot just below the fibula. Conventional scanograms provide a radiographic means to assess limb length. Unfortunately, these studies often do not include the foot, and thus may not fully demonstrate a discrepancy. Standing blocks can also provide an effective means to evaluate limb length. They allow visual assessment of pelvic leveling. To use this technique, the ASIS is assessed from the front and the posterior superior iliac spine is assessed from the back. A difference of up to 2 cm in limb length is acceptable and should not be considered in the surgical correction equation if spinal fusion is considered.

In cases of limb-length discrepancy, X-ray films made with the patient in the standing position may show pelvic obliquity and a compensatory curve that is concave on the side of the longer limb. In addition to limb-length inequality, an unleveled pelvis may be caused by joint contractures in the lower extremity. It is extremely important to measure calf and thigh circumference for evidence of unilateral atrophy, which can be indicative of a neurological problem and thus possibly responsible for a compensatory scoliosis related to the limb-length discrepancy. A unilateral foot deformity, especially when associated with clawing of the toes or abnormal hair bearing, is highly associated with neurological disorders.

Psychosocial Implications

Children with idiopathic scoliosis most often are completely asymptomatic. The effect of having a formal diagnosis of scoliosis is unpredictable in this age group. Most patients with a family history of scoliosis tend to take the diagnosis in stride. However, children who have heard frightening stories of intense pain, neurological deficit, and hardship of an affected relative or peer may confront the diagnosis and its treatment with much apprehension. Additionally, patients involved in competitive sports or intramural activities may be afraid that the condition and its treatment will prevent them from continuing or markedly limit their participation. The method of intervention for scoliosis can have a profound effect on the psychological response of the child. There are essentially three options in the treatment of scoliosis (i.e., “the three O’s”): observation, orthotic, or operation. Observation, although passive, may cause significant anxiety because of the lingering possibility that bracing or surgery may be required. In the case of children for whom an orthotic device has been prescribed, the physician needs to be aware that they may be threatened by the thought of having to wear a device that might make them look different. Even though modern braces can be nearly completely disguised by garment modification and loose clothing, these patients may be sufficiently disturbed about wearing a brace as to either refuse to wear it in school or request home schooling. A randomized controlled study of brace effectiveness (Bracing in Adolescent Scoliosis Trial [BrAIST]) sponsored by the National Institutes of Health is currently underway to determine whether bracing truly has the ability to alter the natural history of idiopathic scoliosis. As expected, patients for whom surgical treatment is recommended are typically concerned that it may harm them or that their scar will be unsightly. A multivariate assessment of patients and parents considering surgery revealed that despite their stated concerns about surgery, the most prevalent issue among them was the fear of paralysis.13

Radiographic Evaluation

Medical imaging for scoliosis allows quantification of the patient’s spinal curvature and the diagnosis of underlying conditions that may have led to the deformity (e.g., sources of nonidiopathic scoliosis). Questions that radiographs may answer for the surgeon include the degree of deformity (in the coronal, sagittal, and axial planes), the flexibility of spinal curves, the levels of the spine requiring instrumentation, the quality of the pedicles, and the presence of associated spondylolisthesis.14,15

It is important to remember that patients with scoliosis will need multiple radiographic studies each year over a period of several years. Proper training of physicians in requesting radiographs and of technicians in obtaining radiographs can reduce the need for repeat studies and reduce overall radiation exposure for the patient. Proper radiation safety training is an essential part of a successful spinal-deformity practice.

Plain Radiographs

The two most common radiographs used to evaluate patients with scoliosis are the standing posteroanterior (PA) and lateral views, utilizing full-length cassettes (14 × 36 in.) or digital equipment that allows accurate splicing of images ( Fig. 5.5 ). The PA and lateral radiographs should include the lower cervical spine and shoulders, entire thoracolumbar spine, and pelvis. Properly made films allow assessment of the patient’s overall skeletal balance as well as skeletal maturity. PA images are used rather than anteroposterior (AP) images in an attempt to reduce radiation exposure to the breast. An association between diagnostic imaging for scoliosis and an increased risk of breast cancer in women has been established (see the section below on Radiation Hazards). PA radiographs of scoliosis, unlike most X-ray films, are displayed with the patient’s right side on the right. This allows the films to be viewed as if the examiner were clinically examining the patient’s spinal curve from a posterior position (e.g., with the patient standing in front). This is also the way in which the spine is viewed in the operating room when the patient is put in the prone position, which is the position used for a posterior spinal fusion.

An erect sitting position is an acceptable alternative to a standing view if the patient has a limb-length inequality or is only minimally ambulatory or wheelchair bound. It is well known that gravity can change the radiographic appearance of a spinal deformity, and a few minutes of sitting before the exposure allows a more accurate representation of the deformity. With supine radiographs there is an absence of the effect of gravity on the spine, and films made with the patient in this position can therefore show a very different curve magnitude and spinal balance as opposed to films made with the patient in a standing position. In the special case of oelisthetic scoliosis associated with high-grade spondylolisthesis, a remarkable decrease in the curve is often seen in the supine bending as compared with the standing position ( Fig. 5.6 ).

It is important to obtain serial radiographs with a consistent method to demonstrate the true deformity in scoliosis and its progression. Radiographs should always be marked by the technician for technique, laterality, and patient position (i.e., supine, sitting, erect). If the patient has a limb-length inequality, an appropriate block may be placed under the shorter limb to level the pelvis before obtaining X-ray films. For more significant limb-length discrepancies, the film may be made with the patient in an erect sitting position or in the supine position, to negate the effects of pelvic obliquity and gravity. In either view, the iliac crests should be visible ( Fig. 5.5 ). The offset of ribs with respect to the pelvic margins and to each other (including the double-rib contour sign) is important in the evaluation of scoliosis ( Fig. 5.7 ).16

When PA and lateral views are obtained, the patient should be instructed to stand in a relaxed manner but not to slouch. In the PA radiograph, the arms are held out slightly from the sides to avoid overlap with the body’s silhouette. Various studies have been done to evaluate the effect of arm positioning in acquisition of the lateral radiograph. The humeri need to be out of the way so that the spine can be visualized; however, holding the arms straight out from the body, as is done with a lateral chest radiograph, may influence the sagittal balance of the spine. When their arms are outstretched, patients tend to assume a “water skiing” position, with the spine leaning backward over the pelvis. One standard practice is to have the patient hold an intravenous infusion pole, or ski poles, to keep the arms at 45-degree angles from the trunk, with the poles supporting the weight of the arms.17 Other described methods include having the patient stand with the arms folded or with the hands on the shoulders or midclavicles. Regardless of which method is chosen, it is important to standardize the method of image acquisition so that radiographs are comparable from one visit to another and from one patient to another.

Congenital anomalies and endplate changes associated with Scheuermann’s kyphosis can sometimes be visualized only in a lateral view ( Fig. 5.8 ). Generally, at least three anteriorly wedged vertebral bodies are seen at the apex of a true Scheuermann curve.

If the patient has complained of low back pain, then oblique views, and radiography of a spot lateral of the lumbosacral junction (if the long lateral does not suffice), may be ordered to look for a spondylolysis. If the patient has persistent back pain in another area, or if there is a history that raises suspicion of a tumor or infection, radiographs with a metallic marker placed over the area to be investigated can be helpful. A specialized oblique image, the Stagnara (Leeds) view, aims at lateral visualization of the apical vertebral bodies when there is severe scoliotic rotational deformity. The amount of obliquity (of the X-ray machine) required to achieve this view is related to the magnitude of the apical rotation of the spine. In the absence of a computed tomography (CT) image through the plane of the vertebral body, the patient is positioned for the Stagnara view so that the X-ray beam forms a tangent to the right and left posterior rib cage as positioned by the technologist. The Stagnara view is helpful because the anatomy and morphology of the pedicle, which is distorted in severe curves on routine PA films, can often be visualized through this technique, with less radiation exposure than is required in CT scanning.18

An AP Stagnara view can also be helpful because it often allows better visualization of the pedicle anatomy and morphology, which is often distorted in severe curves. The AP Ferguson view of the lumbosacral junction is very helpful for assessing sacral obliquity and hemivertebrae, as well as for the quality of a postoperative fusion. The Ferguson view is an upwardly oriented AP exposure of the sacrum so that it is seen en face, with the beam perpendicular to the estimated sacral inclination ( Fig. 5.9 ). For the average sacrum this is about a 30-degree cephalad angulation of the beam.

Assessment of the flexibility of a scoliotic curve pattern is important for planning and predicting correction in a brace or Risser cast, determining fusion levels, and evaluating postoperative correction with growing rod instrumentation or instrumented fusion. The use of lateral or AP side-bending radiographs is a standard method of assessing the flexibility of a scoliosis. The patient is asked to make maximal effort when bending into and then away from the separate curves, and to hold those positions while the X-ray film is made ( Fig. 5.5 ). These films are typically made with the patient in the supine position, but some authors have advocated prone positioning. In making supine bending radiographs for which the patient bends toward the concavity, the rib or flank prominence may increase in size and distort the view of the adjacent segments of the spine. This bias may be reduced by making these images with the patient in the prone position, in which the rib prominence will not be in contact with the X-ray plate. Images of the patient in supine traction ( Fig. 5.5 ) have been proposed as an alternative to lateral bending views, but their usefulness is not yet fully established.

Luk et al have advocated fulcrum bending for evaluation of the flexibility of scoliotic curves, as this technique has been shown to be predictive of curve correction through posterior surgical techniques.19 The test is performed by having the patient lie in the lateral position over a fulcrum (cylindrical bolster) placed beneath the apex of the curve. The fulcrum flexibility is the percentage correction of the Cobb angle of the patient’s deformity as a result of this technique. This test, unlike erect side-bending films, does not require voluntary muscle activation by the patient.19, 20

Push prone radiographs are also a good method for assessing curve flexibility, especially for patients who are unable to make a full bending effort.21 In addition, these images can be obtained in the operating room after induction of anesthesia, to help predict the intraoperative correction of spinal curves. These radiographs require the assistance of additional persons to apply apical pressure and counterpressure above and below the specific curve for which surgery is being done. Radiographs made with supine traction show greater flexibility than side-bending films for patients with curves over 60 degrees.22 These images are also useful for patients who have paralytic curves and are unable to move or bend in the ways needed for other methods. Radiographs of patients in Cotrel dynamic traction may be made with the use of a Risser traction table. This technique often requires anesthesia, because most young children cannot tolerate traction and lying on the bars of the Risser table while awake. Previously published studies using traction films to determine the end-instrumented vertebra have found frequent postoperative coronal decompensation.23 Another study found that PA radiographs made during intraoperative traction showed better flexibility than radiographs made during supine bending, and this changed the operative plan for 11 of 13 patients, eliminating the need for anterior releases.24

Lateral radiographs made with backward bending over a bolster are appropriate for evaluating the flexibility of a kyphosis ( Fig. 5.8 ). These images are obtained with the patient in a recumbent position. The technician must be able to judge where the bolster is to be placed (i.e., the apex of the kyphosis), on the basis of the physician’s order and physical appearance of the patient. The technician must be experienced enough to ensure that the correct levels of the spine and ribs are captured on the image. If the initial lateral image does not reveal abnormality, lateral images on follow-up should be avoided unless needed for a full preoperative evaluation.