Adolescent idiopathic scoliosis: Clinical features, evaluation, and diagnosis

INTRODUCTION — Scoliosis, lateral curvature of the spine with associated rotation of the spinal column, is a structural alteration that occurs in a variety of conditions. Progression of the curvature during periods of rapid growth can result in significant deformity, which may be accompanied by cardiopulmonary compromise. Adolescent idiopathic scoliosis (AIS) is the most common type of scoliosis. Other types include congenital scoliosis, neuromuscular scoliosis, and syndromic scoliosis.

The clinical features, diagnosis, and initial evaluation of AIS will be reviewed here. The management and prognosis of AIS are discussed separately. (See "Adolescent idiopathic scoliosis: Management and prognosis".)

TERMINOLOGY

●Scoliosis – Scoliosis is defined as curvature of the spine in the coronal plane (image 1). It is accompanied by a variable degree of rotation of the spinal column. By convention, >10° of curvature (as measured by the Cobb angle (image 2)) defines scoliosis [1]. The direction (right or left) of a scoliotic curve is defined by the curve's convexity (image 1). The location is defined by the apical vertebra (the one that most deviated and rotated from midline) [2]. (See 'Radiographic evaluation' below.)

●Spinal asymmetry – Curves with Cobb angle ≤10° are within the normal limits of spinal asymmetry and have no long-term clinical significance.

●Early onset scoliosis – Early onset scoliosis is a heterogenous diagnosis referring to any type of scoliosis in children ≤9 years of age. It is defined solely by age and includes idiopathic, neuromuscular, syndromic, congenital, and any other type of scoliosis.

●Congenital scoliosis – Congenital scoliosis results from asymmetry in the vertebrae secondary to congenital vertebral anomalies (eg, hemivertebrae (image 3), congenital fusion). Congenital scoliosis usually manifests before 10 years of age and is often associated with abnormalities of other organ systems (eg, cardiac, renal) [3,4].

●Idiopathic scoliosis – Idiopathic scoliosis is scoliosis for which there is no definite etiology, unlike neuromuscular, congenital, or syndromic types. (See 'Differential diagnosis' below.)

Idiopathic scoliosis is divided into three subcategories based upon the patient's age at presentation:

•Infantile – 0 to 3 years

•Juvenile – 4 to 9 years

•Adolescent – ≥10 years

AIS is the most common form of idiopathic scoliosis, accounting for between 80 and 85 percent of cases [5-7]. AIS is the focus of discussion in the remainder of this topic review.

EPIDEMIOLOGY — The prevalence of AIS with a Cobb angle >10° is approximately 3 percent, but only 10 percent of adolescents with AIS require treatment (0.3 percent of the population) [8-11]. Males and females are affected equally. However, the risk of curve progression (and therefore the need for treatment) is 10 times higher in females than in males [10].

The prevalence and female-to-male ratio (F:M) of AIS of according to severity are as follows [11]:

●Cobb angle >10° – 2 to 3 percent; F:M 1.4 to 2.4:1

●Cobb angle ≥20° – 0.3 to 0.5 percent; F:M 5.4:1

●Cobb angle ≥30° – 0.1 to 0.3 percent; F:M 10:1

●Cobb angle ≥40° – ≤0.1 percent

The severity of AIS on initial presentation appears to be increased in patients who are overweight or obese [12-14]. In a retrospective cohort of 150 patients (≥10 years of age) referred for spinal asymmetry, body mass index ≥85^th percentile was associated with greater average curve (approximately 24 versus 18°) at presentation, increased risk of having a curve ≥40° at presentation, and more advanced skeletal maturity at presentation [12]. Increased severity of AIS patients who are overweight/obese at presentation may be related to delayed detection for a variety of reasons, primarily the limited utility of scoliometer readings in such patients [13]. (See 'Use of a scoliometer' below.)

ETIOLOGY — The etiology of AIS is unclear. A genetic contribution is supported by twin and family history studies [7,15-19]. Familial AIS may be linked to the X chromosome, with a dominant inheritance pattern [20]. Other genetic loci for AIS have been mapped to chromosomes 8, 9, 17, 18, and 19 [21-25], but the inheritance pattern is unclear. It may be polygenic or a dominant major gene diallel with incomplete sex- and age-dependent penetrance [19,26,27].

Proposed but unproven factors in the pathogenesis of AIS include abnormalities in growth hormone secretion, connective tissue structure, paraspinal musculature, vestibular function (which affects axial posture), melatonin secretion (which affects growth [28]), and platelet microstructure (which has a contractile system similar to that of skeletal muscle) [10,29-31].

CLINICAL PRESENTATION — Patients with AIS usually come to medical attention as a result of truncal asymmetry noted by the patient or caregivers, during school-based scoliosis screening, or as an incidental finding during physical examination or on chest radiograph or other imaging study [32]. Patients with severe thoracic curves (Cobb angle ≥70°) may present with restrictive pulmonary disease [33,34], but curves of this severity usually have onset before age 10 years [35,36]. Patients with idiopathic scoliosis also may have obstructive lung disease; in a retrospective review of 176 patients with idiopathic scoliosis and Cobb angle ≥40°, 39 percent had obstructive lung disease [37].

CLINICAL EVALUATION

Goals — Evaluation of the adolescent with scoliosis has several objectives, including:

●Identification of an underlying etiology (table 1A-C), thereby excluding AIS.

●Assessment of the magnitude of the curve and need for radiographs.

●Determining the risk of progression, which influences management decisions. (See "Adolescent idiopathic scoliosis: Management and prognosis", section on 'Approach to surveillance and management'.)

History — Questions in the history focus on ruling out an underlying cause and the risk for progression (based upon estimation of remaining potential for linear growth). Important aspects of the history include (table 2) [2,38]:

●When was the deformity first noted, and who noted it? (Curves noted by the family tend to be greater than those noted by the primary care provider or detected through school screening.)

●What is the rate of progression? Rapid progression of the curve is suggestive of an underlying cause (table 1A-C).

●Does the patient have significant back pain (pain that limits activities, wakes a patient at night, or requires frequent analgesia [39])? Significant pain increases the likelihood of an underlying cause [40]. In a review of 2442 patients, 23 percent complained of pain at the time of presentation; among patients complaining of pain, 9 percent had an underlying pathologic condition (spondylolysis, spondylolisthesis, Scheuermann kyphosis, syrinx, herniated disc, hydromyelia, tethered cord, intraspinal tumor) [41]. (See "Back pain in children and adolescents: Causes".)

Patients with significant pain may require additional evaluation, particularly those who have neurologic symptoms or signs and/or a left thoracic curve. (See 'Radiographic evaluation' below and "Back pain in children and adolescents: Evaluation".)

●Are there symptoms suggestive of a neuromuscular condition (eg, muscle weakness, bowel or bladder problems, headache, neck pain)? Neurologic symptoms increase the likelihood of an underlying cause and require additional evaluation. (See 'Radiographic evaluation' below.)

●Does the patient have shortness of breath or difficulty breathing? Severe thoracic scoliosis may affect pulmonary function.

●What is the patient's growth trajectory and has the pubertal growth spurt begun? This information helps to estimate remaining linear growth and risk for progression. (See "Normal puberty", section on 'Growth spurt'.)

●Has the patient entered puberty? Tanner grade 2 (the onset of puberty) precedes peak height velocity (figure 1A-B).

●For female patients, has menarche occurred? If so, when? Peak height velocity usually occurs approximately 0.5 years before menarche (figure 1A) [42]. However, females continue to grow (with decreased velocity) for approximately 24 months after menarche [43]. (See "Normal puberty", section on 'Growth spurt'.)

●For male patients, have they started shaving (or do they need to)? Skeletal growth in males is completed when they are (or should be) shaving every day.

●Is there a history of lower-limb fracture, joint infection, or arthritis (which may result in leg-length discrepancy)?

●Is there a family history of scoliosis? AIS tends to run in families; the risk of scoliosis in the sibling of an affected patient is approximately 7 percent but may be increased if one of the parents was also affected [7,17]. (See 'Etiology' above.)

Scoliosis examination

Inspection — Examination for scoliosis is facilitated by proper exposure of the trunk (ie, an examination gown that is open in the back, with undergarments that expose the iliac crests and posterior and anterior superior iliac spines) [2].

The first step in the scoliosis examination is simple inspection.

●When inspecting the standing patient from behind, findings of scoliosis may include (table 3 and picture 1):

•Curvature of the spine with thoracic or lumbar asymmetry; left thoracic curves have been associated with nonidiopathic causes (table 1A-C) [44-47] and may require additional evaluation (see 'Radiographic evaluation' below)

•Differences in the level of the shoulders or scapulae

•Asymmetry of the waistline

•Asymmetry in the distance that the arms hang from the trunk

•Head shifted to one side and not centered over the sacrum, sometimes called trunk shift (in patients without a trunk shift, a plumb line dropped from the spinous process of the seventh cervical vertebra should pass through the gluteal cleft)

●When viewed from the side, a patient with idiopathic scoliosis may have decreased thoracic kyphosis. Increased thoracic kyphosis may suggest other etiologies [48].

Forward bend test — The Adams forward bend test is performed by observing the patient from the back while they bend forward at the waist until the spine becomes parallel to the horizontal plane, with feet together, knees straight ahead, and arms hanging free. Thoracic (rib) or lumbar (loin) prominence on one side (figure 2) is a sign of scoliosis.

The Adams forward bend test demonstrates the rotational component of scoliosis; the rib prominence is the result of the ribcage rotating along with the spine [49,50]. Asymmetry in the upper thoracic, midthoracic, thoracolumbar, or lumbar region should be measured with a scoliometer (picture 2) [38,51]. The scoliometer measures the angle of trunk rotation (ATR). The ATR provided by the scoliometer is not the same as the Cobb angle measurement. (See 'Use of a scoliometer' below.)

Inability to bend forward at the waist or decreased range of motion with forward or side bending may be secondary to pain, lumbar muscle spasm, and/or tightness of the hamstrings [8,38]. Any of these findings may indicate an underlying cause of scoliosis (table 1A-C).

The Adams forward bend test is the most sensitive of the clinical examination findings for scoliosis (using the radiographic Cobb angle as the reference standard) [52]. However, the sensitivity and specificity vary with the skills of the examiner, location of the curve, and Cobb angle of the curve that is used as the reference standard [52-55].

The range of sensitivity and specificity of the forward bend test without a scoliometer (performed by school-based screeners in most cases) for varying degrees of scoliosis are as follows:

●Thoracic scoliosis with Cobb angle ≥10° – Sensitivity 74 to 84 percent, specificity 78 to 93 percent [54,55]

●Thoracic scoliosis with Cobb angle ≥20° – Sensitivity 92 to 100 percent; specificity 60 to 91 percent [52,55]

●Lumbar scoliosis with Cobb angle ≥20° – Sensitivity 73 percent; specificity 68 percent [52]

●Scoliosis with Cobb angle ≥40° – Sensitivity 83 percent; specificity 99 percent (with repeat screening by a more experienced screener in some cases) [53]

Use of a scoliometer — A scoliometer (sometimes called an inclinometer) is a device used for scoliosis screening and quantification of trunk rotation [56,57]. Scoliometer measurements can help in determining which patients need radiographs but should not be used without radiographs to decide that a patient needs bracing or surgery. (See "Adolescent idiopathic scoliosis: Management and prognosis", section on 'Approach to surveillance and management'.)

A scoliometer is basically a version of a carpenter's level that measures the ATR. The scoliometer is run along the patient's spine from cephalad to caudad while the patient is in the position assumed for the Adams forward bend test (picture 2). If a rotational prominence (ie, rib or loin prominence) is present, the ball in the scoliometer deviates from the center of the device. If there is a right rib prominence, for example, the right side of the scoliometer tips upward, and the ball deviates to the left. Inexpensive scoliometer applications for use on mobile devices have been developed. Small validation studies of one such application suggested good correlation with standard scoliometers [58-60].

The degrees on a scoliometer do not correspond one-to-one to the degrees of curvature as measured using Cobb angles on a radiograph [61]. As a general guideline, an ATR of 7° corresponds to a Cobb angle of 20° [62]. However, not all patients with radiographic scoliosis have rotation of the trunk, and not all patients with trunk rotation have radiographic scoliosis [38]. In overweight children, the traditional correspondence of a 7° ATR to a 20° Cobb angle may not be accurate. Scoliometer readings in overweight children are typically lower than in nonoverweight children [13]. For this reason we suggest using an ATR cutoff of 5 rather than 7° for patients with body mass index ≥85^th percentile for age and sex.

Scoliometer use is, to some extent, operator dependent. In a small observational study, the interrater errors for thoracic and lumbar curves were 2 and 2.2°, respectively (compared with an error of 1.2 and 1.6°, respectively, when the measurements were performed by a single rater) [57]. Potential sources of error in scoliometer measurements include the size of the ball in relation to the markings for degree increments, inconsistent identification of the apex of the curve, and inconsistent performance of the forward bend test [52].

The sensitivity and specificity of the scoliometer to identify patients with scoliosis varies depending upon the threshold scoliometer value (ie, 5 versus 7 versus 10°) and location of the curve [52,61,63]. A scoliometer threshold of 7° has a sensitivity of 83 percent and a specificity of 87 percent (using Cobb angle >10° as the reference standard) [63]. Lowering the threshold increases sensitivity and decreases specificity; increasing the threshold decreases the sensitivity and increases specificity [61]. The sensitivity of the scoliometer appears to be greater for thoracic than lumbar curves, but specificity does not appear to be affected by curve location [52].

Evaluation for leg length discrepancy — Asymmetry on bilateral palpation of the iliac crests and posterior inferior iliac spines with the patient in the standing position with the hips and knees fully extended suggests leg-length discrepancy. In patients with leg-length discrepancy, clinically apparent scoliosis may be compensatory.

If leg length discrepancy is suspected, measuring blocks of various sizes can be placed under the shorter leg until the pelvis is level. A compensatory scoliosis usually disappears when the leg-length discrepancy is "corrected" with this maneuver. Next, the patient's leg lengths should be measured individually. Leg length is measured with the patient supine and the legs extended and held together in the midline; the measurement is taken from the anterior superior iliac spine to the medial malleolus, with the tape running medial to the patella.

Many people have leg length discrepancy of <1 cm (0.4 inches); the magnitude of difference that results in compensatory scoliosis varies from person to person but typically is >2.5 cm (1 inch). Children with clinically significant leg-length discrepancy should be evaluated by an orthopedist. Even experienced examiners may have difficulty differentiating leg-length discrepancy from structural scoliosis, which may coexist. Radiographs may be necessary to determine whether the patient has one or both conditions.

General examination — Important aspects of the general examination include (table 3) [2,38]:

●Measurement of the patient's height, which should be plotted on a growth curve standardized for age and sex (figure 3A-B). This helps to estimate the remaining growth potential. Linear growth is near completion when there has been <1 cm of change in standing height over a six-month period [43]. Height velocity curves (figure 4A-B) may provide more accurate information about attainment of peak height velocity. (See "Normal puberty", section on 'Growth spurt'.)

●Measurement of the patient's arm span – An increased arm span to height ratio may be a clue to Marfan syndrome (picture 3), a cause of syndromic scoliosis. (See "Normal growth patterns in infants and prepubertal children", section on 'Arm span to height' and "Genetics, clinical features, and diagnosis of Marfan syndrome and related disorders", section on 'Skeletal findings'.)

●Assessment of Tanner stage (picture 4 and picture 5) – The risk for curve progression is greatest during the adolescent growth spurt, the onset of which precedes Tanner stage 2 (figure 1A-B). (See "Normal puberty".)

●Examination of the skin for:

•Café-au-lait macules (picture 6A-B) and axillary freckling (picture 7) (suggestive of neurofibromatosis). (See "Neurofibromatosis type 1 (NF1): Pathogenesis, clinical features, and diagnosis", section on 'Clinical manifestations'.)

•Vascular lesions, hypopigmented lesions, hyperpigmented lesions, dimpling, or a patch of hair overlying the spine ( (picture 8), which may be associated with spinal dysraphism). (See "Closed spinal dysraphism: Clinical manifestations, diagnosis, and management", section on 'Clinical manifestations'.)

●Excessive skin or joint laxity may be associated with Ehlers-Danlos syndromes, Marfan syndrome, or osteogenesis imperfecta. (See "Clinical manifestations and diagnosis of Ehlers-Danlos syndromes", section on 'Clinical manifestations and diagnosis' and "Genetics, clinical features, and diagnosis of Marfan syndrome and related disorders", section on 'Clinical manifestations of MFS' and "Osteogenesis imperfecta: An overview", section on 'Clinical manifestations'.)

●Examination of the hands for intrinsic wasting (which can be associated with cervical syringomyelia or hereditary motor sensory neuropathy) and arachnodactyly (long, slender, curved fingers, which can be associated with Marfan syndrome).

●The feet should be examined for high arches (pes cavus) and hammer or claw toes, which are suggestive of neuromuscular disease. (See "Friedreich ataxia" and "Overview of hereditary neuropathies" and "Forefoot and midfoot pain in the active child or skeletally immature adolescent: Overview of causes", section on 'Pes cavus (high arch)'.)

●A full neurologic examination should be performed, including examination of the reflexes. Balance and strength of the lower extremities can be assessed by watching the patient walk normally, toe and heel walk, squat deeply, and hop on each leg [38]. Weakness or decreases/absent deep tendon reflexes may indicate neuromuscular disease (table 1A). (See "Detailed neurologic assessment of infants and children", section on 'Neurologic examination'.)

The abdominal reflex is especially important. An asymmetric abdominal reflex may indicate intraspinal pathology [64-66]. Elicitation of the abdominal reflex involves lightly stroking the skin on one side of the upper, middle, and lower abdomen (above, at, and below the umbilicus, respectively); the normal response is contraction of the muscles, pulling the umbilicus and midline to the stimulated side. (See "Detailed neurologic assessment of infants and children", section on 'Superficial reflexes'.)

RADIOGRAPHIC EVALUATION

Radiographs — Radiographs are required to confirm the diagnosis of scoliosis, evaluate the etiology (congenital, neuromuscular, idiopathic), determine the curve pattern, measure curve magnitude (Cobb angle), and to evaluate skeletal maturity (to determine the risk for progression) [2,38,67,68].

Low-dose three-dimensional spine imaging is increasingly available [69]. If it is available, digital radiography is preferred because it reduces the cumulative radiation dose over the course of treatment [70,71].

●Indications – Indications for radiographic evaluation in the patient with scoliosis include [72,73]:

•Scoliometer reading of ≥7° in children with body mass index (BMI) <85^th percentile or ≥5° in children with BMI ≥85^th percentile, since curves of this magnitude may require treatment [51,68]

•Clinically evident scoliosis (eg, a large, unambiguous curve) on physical examination

•Thoracic or lumbar asymmetry on physical examination in skeletally immature children or in combination with a family history of scoliosis [72]

•Monitoring progression in patients with previously diagnosed AIS

●Views – The initial radiographic evaluation for scoliosis should include standing, full-length posteroanterior (PA) and lateral views of the spine (C7 to the sacrum and iliac crest) [68,74,75].

PA views minimize radiation to the breasts and thyroid [68]. Supine spine radiographs underestimate curve magnitude (by eliminating the effect of gravity). An air fluid level in the stomach confirms a radiograph was taken with the patient upright. Leg-length discrepancy (if present) should be corrected by placing an appropriately sized block under the shorter leg before radiography or obtaining the radiographs with the patient in the sitting position [74].

The lateral view assesses spinal curvature in the sagittal plane. Spondylolisthesis (image 4) and Scheuermann kyphosis (image 5) are sagittal plane deformities that may be associated with scoliosis. Thoracic kyphosis is usually reduced in patients with AIS; increased thoracic kyphosis may suggest underlying pathology [48,76,77]. The lateral view need not be repeated at subsequent visits if the initial film shows only the normal findings of scoliosis (eg, thoracic kyphosis within the typical range and lumbar lordosis).

Lateral bending views are not necessary for diagnosis but are used for planning surgery. (See "Adolescent idiopathic scoliosis: Management and prognosis", section on 'Surgery'.)

Long film cassettes may not be available at all facilities. If long film cassettes are not available, sending the patient to a facility where long film cassettes are available is preferred to obtaining suboptimal radiographs on inappropriate cassettes.

●Radiographic findings – The PA radiograph is viewed with the patient's heart on the examiner's left (image 1), as if the examiner is standing behind the patient. In addition to assessing for variations in the number of thoracic or lumbar vertebrae, which are present in approximately 10 percent of patients with AIS [78,79], spinal radiographs should be assessed for abnormalities suggestive of congenital, neuromuscular (table 1A), or developmental scoliosis, including:

•Soft tissue abnormalities (eg, paraspinal mass)

•Congenital vertebral anomalies such as wedge vertebra (image 6) hemivertebrae (image 3)

•Developmental wedging of the spine (Scheuermann kyphosis) (image 5)

•Vertebral body lucency or erosion of the pedicles (suggestive of bone tumor)

•Widening of the interpedicular distance (suggestive of a spinal cord tumor, syringomyelia, diastematomyelia, or spinal dysraphism)

•Increased thoracic kyphosis [48]

The typical curve in AIS is a right thoracic, left lumbar double curve, but many other curve configurations occur. The direction of the curve (right or left) is defined by its convexity (image 1). The location is defined by the vertebra that is most deviated and rotated from midline (the apical vertebra) [2]:

•Cervical – C2 to C6

•Cervicothoracic – C7 to T1

•Thoracic – T2 to T11

•Thoracolumbar – T12 to L1

•Lumbar – L2 to L4

•Lumbosacral – L5 or below

Classifications for curve patterns that are used in making decisions regarding surgery (eg, King-Moe, Lenke) are beyond the scope of this review [80,81].

Cobb angle — The Cobb angle is the reference standard for quantitative monitoring of scoliosis. The Cobb angle is formed by the intersection of a line parallel to the superior end plate of the most cephalad end vertebra in a particular curve, with the line parallel to the inferior end plate of the most caudad end vertebra of the curve (image 2).

Although the Cobb angle is the accepted standard for measuring scoliosis on radiographs, it is has limitations [68,82]. It describes only one plane of the three-dimensional deformity. It is not directly proportional to the severity of scoliosis (ie, a curve with a Cobb angle of 40° is more than twice as severe as a curve with a Cobb angle of 20°). Finally, variation in inter- and intraobserver measurement error is approximately 5° [83-88].

When using digital images, one cannot accurately measure the Cobb angle by simply placing a goniometer on the computer screen. However, many digital imaging systems have an "annotation" feature with which the clinician can measure the Cobb angle semimanually (ie, by using the mouse to draw the lines, which takes practice to master) [89]. To use these features, a large high-resolution monitor is helpful. Alternatively, the Cobb angles can be drawn manually on a paper printout of the digital image.

Clinicians who treat children with scoliosis must have training and experience in measuring Cobb angles and should have a digital system that includes measurement functions, access to traditional printed films, or, at the very least, a good-quality paper printout of the digital image to manually measure the Cobb angle.

Skeletal maturity — Skeletal maturity is assessed to determine the risk for progression. The Risser sign and Sanders skeletal maturity rating system are two frequently used methods of assessing remaining growth using radiographs. (See "Adolescent idiopathic scoliosis: Management and prognosis", section on 'Risk for progression'.)

Risser sign — The Risser sign is a visual grading of the degree to which the iliac apophysis has undergone ossification and fusion and is used to assess skeletal maturity on PA spinal radiographs. The iliac apophysis ossifies in a stepwise fashion from anterolateral to posteromedial along the iliac crest.

The Risser grades are as follows (figure 5) [30]:

●0 – No ossification

●1 – Up to 25 percent ossification

●2 – 26 to 50 percent ossification

●3 – 51 to 75 percent ossification

●4 – Greater than 76 percent ossification

●5 – Full bony fusion of the apophysis

Lower Risser grades signify greater growth remaining and greater risk for curve progression [90]. However, Risser 1 usually occurs after peak height velocity, and it should not be used as the only indicator of skeletal maturity [91-93].

Sanders skeletal maturity staging system — The Sanders skeletal maturity assessment is based on the radiographic appearance of the bones of the hand (figure 6) [94,95]. In observational studies, peak height velocity occurred during Sanders stage 3 (figure 6) [93,96].

Other methods — Other methods of assessing skeletal maturity on radiographs that may be used to supplement the Risser sign include:

●The method of Tanner and Whitehouse (or scoring systems adapted from Tanner and Whitehouse), which is based on the radiographic appearance of the epiphyses of the distal radius, ulna, and small bones of the hand [97-99].

●Assessment of the triradiate cartilage openness or closure (peak height velocity occurs in females with open triradiate cartilage before attainment of Risser 1) [91,100].

●Assessment of proximal humeral ossification on the standard scoliosis radiograph (obtained with the patient's hands by the sides and the palms facing forwards) [101]. Ossification proceeds in five stages (figure 7), with peak height velocity at the beginning of stage 3. This method can be used in combination with the Sanders method [102].

●Assessment of ossification centers in the elbow. Two methods have been described:

•The Sauvegrain method, in which the appearance of four different elbow ossification centers (lateral condyle, trochlea, olecranon apophysis, and proximal radial epiphysis) are scored on a 27-point scale, which is then plotted on a graph [103].

•The olecranon method, in which only the olecranon apophysis is evaluated. The olecranon apophysis progresses through five radiographic stages at six-month intervals: appearance of two ossification nuclei, half-moon-shaped ossification center, rectangular-shaped ossification center, beginning of fusion, and complete fusion [104]. The appearance of the rectangular-shaped ossification center corresponds to closure of the triradiate cartilage.

MRI of the spine — Magnetic resonance imaging (MRI) of the spine may be indicated in patients with scoliosis and clinical or radiographic findings suggestive of intraspinal pathology (tumor, dysraphism, infection) [45,46,73,105-109]. These findings include:

●Neurologic symptoms or signs, including headache, neck pain, intrinsic wasting of the hands (which can be associated with syringomyelia), asymmetric abdominal reflex, asymmetric lower-extremity atrophy, pes cavus, and midline skin lesions (vascular, pigmentary, hair patch); asymmetric abdominal reflex is particularly important as a possible manifestation of underlying spinal cord abnormality.

●Significant pain (pain that limits activities, wakes a patient at night, or requires frequent analgesia); computed tomography and/or bone scan also may be indicated in these patients if bone tumors or infection are suspected [40,41]. (See 'History' above and "Bone tumors: Diagnosis and biopsy techniques" and "Hematogenous osteomyelitis in children: Evaluation and diagnosis", section on 'Advanced imaging'.)

●Abnormalities on radiographs that suggest congenital scoliosis (eg, wedged vertebrae (image 6) or hemivertebrae (image 3)) or intraspinal pathology (eg, vertebral body lucency, erosion of the pedicles, or widening of the interpedicular distance).

●Left thoracic scoliosis (unless they have dextrocardia) or scoliosis with increased kyphosis are considered atypical scoliosis deformities (which can be associated with Arnold-Chiari malformation or syringomyelia) [48,76,77]. These are relative indications for obtaining an MRI of the spine, but this is controversial [73,109-112].

DIAGNOSIS — The diagnosis of AIS is made clinically and radiographically (algorithm 1). Criteria for the diagnosis include:

●Age ≥10 years

●Curvature of the spine in the coronal plane with a Cobb angle >10° (see 'Cobb angle' above) with associated clinical rotation of the spinal column

●Absence of other etiologies for scoliosis (see 'Differential diagnosis' below)

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of AIS includes the nonidiopathic forms of scoliosis: neuromuscular scoliosis (table 1A), congenital scoliosis, and syndromic scoliosis (table 1B).

Neuromuscular scoliosis — Neuromuscular scoliosis occurs in patients with neurologic or musculoskeletal problems, such as cerebral palsy, myelomeningocele, or muscular dystrophy. It is the result of muscle imbalance and lack of trunk control. Most patients with neuromuscular scoliosis have additional findings related to the underlying disorder that help to differentiate neuromuscular from idiopathic scoliosis (table 1A).

Neuromuscular scoliosis may be structural or nonstructural. Nonstructural scoliosis has no rotational component; it may be related to postural abnormalities, leg-length discrepancy, or pain (eg, splinting in patients with pneumonia or empyema).

Congenital scoliosis — Congenital scoliosis results from asymmetry in the vertebrae secondary to congenital vertebral anomalies (eg, hemivertebrae (image 3) [failure of formation], congenital fusion [failure of segmentation], or a combination). Congenital scoliosis usually manifests before 10 years of age (the lower threshold for AIS) and is often associated with abnormalities of other organ (cardiac and renal) systems [3,4]. The congenital vertebral anomalies, apparent on plain radiographs, differentiate congenital scoliosis from idiopathic scoliosis.

Syndromic scoliosis — Syndromic scoliosis occurs as part of certain genetic disorders, including connective tissue disorders (eg, Marfan syndrome, osteogenesis imperfecta) and more generalized tissue disorders (eg, neurofibromatosis). Most patients with syndromic scoliosis have additional features of the syndrome that help to establish the diagnosis (table 1B). However, in some cases, the spinal deformity may be the first clinical manifestation of the disorder.

INDICATIONS FOR REFERRAL — Referral to a specialist (neurosurgeon, oncologist, orthopedic surgeon, neurologist) may be warranted for patients with clinical or radiographic findings suggestive of congenital or neuromuscular scoliosis.

For patients with AIS, indications for referral to an orthopedic surgeon are discussed separately. (See "Adolescent idiopathic scoliosis: Management and prognosis", section on 'Indications for referral'.)

SCOLIOSIS SCREENING — We perform a spinal examination, including the forward bend test, during the physical examination of children and adolescents at routine health supervision visits, as recommended by several professional societies [113,114]. Spinal deformity may be the presenting sign of nonidiopathic conditions, including heritable collagen disease, neurologic conditions, and skeletal dysplasia, in addition to AIS [115]. Spinal examination is particularly important before the pubertal growth spurt (at approximately 10 years) [116,117]. In addition, pediatric health care providers should be prepared to evaluate patients for scoliosis when it is discovered incidentally or through a school-based screening program, or when the adolescent or caregivers express concern about scoliosis.

Many states in the United States have mandated or voluntary school-based scoliosis screening programs [74]. However, the efficacy of such programs in preventing curve progression and/or the need for scoliosis surgery is unproven [118-120].

Supporters of screening programs suggest that earlier diagnosis permits conservative therapy (ie, bracing), preventing the need for surgery and potential surgical complications [67,115,121-125]. Opponents of screening programs suggest that AIS lacks the characteristics that make a disease a good candidate for screening (eg, high prevalence, substantial morbidity in untreated patients, preclinical phase that can be detected by a screening test that is highly sensitive and specific, benefit of preclinical detection, and availability of effective treatment) [53]. (See "Screening tests in children and adolescents", section on 'General principles'.)

Limitations of school-based screening for AIS in asymptomatic patients include [67,125]:

●The low prevalence of AIS requiring intervention (0.3 percent) [9,54,116,126,127] increases the likelihood that a positive screen will be a false positive. In a systematic review, rates of false positive screening ranged from 0.8 to 21.5 percent [120].

Potential harms of false positive screening include unnecessary anxiety, time lost from school or work for follow-up or specialty care, radiation exposure, and adverse psychosocial effects (particularly related to brace wear) [119].

●Untreated AIS does not result in substantial morbidity or mortality in most cases [128,129]. (See "Adolescent idiopathic scoliosis: Management and prognosis", section on 'Outcome'.)

●The Adams forward bend test has variable sensitivity and specificity depending upon the examiner and the curve magnitude used as the reference standard. The interrater error of scoliometer measurement is approximately 2°. In addition, there is not a one-to-one correlation between the scoliometer measurement and the Cobb angle (the reference standard for diagnosis). (See 'Scoliosis examination' above.)

●The benefit of preclinical detection is unproven [118]. Most curves detected through screening do not progress to the point of needing surgical intervention; most cases requiring surgical treatment are detected without screening [53,130,131].

●Although a systematic review found adequate evidence that bracing may slow curve progression in patients with curves with Cobb angle <40 to 50°, evidence about the association between decreased curve during adolescence and health outcomes in adulthood is lacking [120].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Idiopathic scoliosis in adolescents".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or email these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient education" and the keyword[s] of interest.)

●Basics topic (see "Patient education: Scoliosis (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Terminology – Scoliosis is a lateral curvature of the spine with associated rotation of the spinal column (image 1). Adolescent idiopathic scoliosis (AIS, also called late-onset idiopathic scoliosis) has onset after age 10 years and is the most common type of scoliosis. (See 'Terminology' above.)

●Evaluation – The evaluation for AIS centers on looking for an underlying etiology (table 1A-C), assessing the magnitude of the curve, and determining the risk for progression (table 2 and table 3). (See 'Clinical evaluation' above.)

•Examination – Examination for scoliosis includes inspection (picture 1), the Adams forward bend test (figure 2), and, if asymmetry is noted, measurement of the angle of trunk rotation (ATR) with a scoliometer (picture 2). (See 'Scoliosis examination' above.)

•Imaging – Radiographs are indicated for patients with an ATR as measured with the scoliometer of ≥7° in children with body mass index (BMI) <85^th percentile, an ATR of ≥5° in children with BMI ≥85^th percentile, and in children with clinically obvious scoliosis. Radiographs also may be indicated in skeletally immature patients with asymmetry on examination and a family history of scoliosis. (See 'Radiographs' above.)

The initial radiographic evaluation for scoliosis should include standing, full-length posteroanterior and lateral views of the spine (C7 to the sacrum and iliac crest). (See 'Radiographs' above.)

Important radiographic findings include assessment of the paravertebral soft tissues and the bony spine, the direction and magnitude of the curve(s) (as measured by the Cobb angle) (image 2), and skeletal maturity (ie, the Risser sign (figure 5)). (See 'Radiographs' above and 'Risser sign' above.)

The Sanders skeletal maturity staging system, based on radiographs of the hand (figure 6), also may be used to assess skeletal maturity. (See 'Sanders skeletal maturity staging system' above.)

•MRI is indicated in patients with associated neurologic signs or symptoms (asymmetric abdominal reflex) and abnormalities on radiographs that suggest congenital scoliosis or intraspinal pathology. MRI also may be indicated in patients with complaints of significant pain, left thoracic curves (unless they have dextrocardia), and/or scoliosis with increased kyphosis. (See 'MRI of the spine' above.)

●Diagnosis and differential diagnosis – The diagnosis of AIS is made clinically and radiographically (algorithm 1). Criteria for the diagnosis include age ≥10 years at onset, Cobb angle >10°, and absence of other etiologies for scoliosis (eg, congenital, neuromuscular, syndromic (table 1A-C)). (See 'Diagnosis' above and 'Differential diagnosis' above.)