Condition Definition

Scoliosis is characterized by anatomic structural alteration, a lateral curvature of the spine in the coronal plan that is usually accompanied by rotation. The direction (right or left) is defined by the curve's convexity. The location is defined by the vertebra that is most deviated and rotated from midline, called the apical vertebra. By convention, scoliosis is defined as a curvature of at least 10° (as measured by the Cobb angle¹ on an anteroposterior x-ray of the spine); curves with a Cobb angle of less than 10° are referred to as “spinal asymmetry.”

Idiopathic scoliosis is a diagnosis of exclusion for cases with no definite etiology and is categorized based on age of presentation:

• Infantile: Presents between birth and age 3 years
• Juvenile: Presents between ages 4 and 9 years
• Adolescent: Presents at age 10 years or older

Adolescent idiopathic scoliosis (AIS) is the most common form of idiopathic scoliosis, and accounts for 80 to 85 percent of cases of idiopathic scoliosis.^2-4 The clinical course associated with infantile and juvenile idiopathic scoliosis appears to be different than that seen in AIS; therefore, these conditions generally are considered as separate entities.

Nonidiopathic scoliosis is scoliosis that is attributable to or associated with other underlying conditions. This is usually categorized by cause, such as neuromuscular scoliosis (secondary to nervous system or peripheral neuromuscular dysfunction; typically seen in persons with underlying neurologic and/or musculoskeletal conditions such as cerebral palsy) and congenital scoliosis (resulting from anatomic abnormalities of the vertebra that lead to progressive spinal deformity as a child grows).⁵ The clinical course of nonidiopathic scoliosis varies by etiology, and is also quite often different than that seen in persons with AIS.

Prevalence and Burden

Commonly cited estimates of prevalence of AIS vary, but usually are around 1 to 3 percent (both U.S. and non-U.S. studies) for AIS with a Cobb angle of at least 10° in children ages 10 to 16 years.⁶ Prevalence estimates for curves of greater severity are somewhat lower: a retrospective cohort study conducted to characterize school-based screening for scoliosis in the United States found a cumulative incidence of 1.8 percent (95% confidence interval [CI], 1.2 to 2.3) for curves of more than 10°, 1.0 percent (95% CI, 0.6 to 1.5) for curves of at least 20°, and 0.4 percent (95% CI, 0.1 to 0.6) for curves of 40° or more.⁷

Prevalence also varies by sex. Based on school screening studies conducted internationally, the prevalence of AIS in children ages 10 to 18 years ranges from 0.15 to 0.66 percent in boys⁸ and 0.24⁸ to 3.1 percent in girls.⁹ However, the discrepancy between sexes is greatly affected by increased degree of curvature: males and females have a similar prevalence of scoliosis with a Cobb angle of 10°, but females are 10 times more likely than males to have progression of Cobb angle to 30° or more.¹⁰

The prevalence of AIS and the female-to-male ratio, by degree of Cobb angle severity, is approximately as follows:

• Cobb angle of at least 10°: Prevalence of 2 to 3 percent in adolescents; female-to-male ratio of 1.4–2.4 to 1
• Cobb angle of at least 20°: Prevalence of 0.3 to 0.5 percent in adolescents; female-to-male ratio of 5.4 to 1
• Cobb angle of at least 30°: Prevalence of 0.1 to 0.3 percent in adolescents; female-to-male ratio of 10 to 1
• Cobb angle of at least 40°: Prevalence of 0.1 percent in adolescents; female-to-male ratio not available⁶

Etiology and Natural History

The etiology of AIS is, by definition, unknown, although there is some evidence suggesting a possible genetic contribution.¹¹ Studies show a higher concordance in incidence and degree of AIS in monozygotic twins,¹² and an increased prevalence of AIS in siblings, children, and some close relatives of those with AIS.^{4, 13} The inheritance pattern of familial AIS is not clear,^{14, 15} although some research suggests that expression of familial AIS may be linked to the X chromosome (with dominant inheritance),¹⁶ and genetic loci for AIS have also been mapped to certain chromosomes.^17-20 Other possible etiologies for AIS include abnormalities of the growth and structure of vertebral bodies and discs, abnormal spinal mechanics with secondary spinal instability, body asymmetry, neurologic dysfunction, abnormal ribcage anatomy, abnormal platelet microstructure, and melatonin secretion (as it relates to growth). None of these conditions have been found to be universally associated with development of AIS, suggesting that this most likely is a multifactorial condition.²¹

AIS curves typically progress most rapidly during the adolescent growth spurt before skeletal maturity. Skeletal maturity is associated with decreased growth rate and a decrease in the likelihood of progression of the scoliosis curve. Clinically, skeletal maturity is most often assessed in AIS patients by the Risser sign (the stage of ossification of the iliac apophysis as seen on x-ray; measured on a scale of 1 to 5, with 5 indicating the full ossification seen in developmentally mature adolescents and adults); however, other measures of developmental maturity (e.g., age at menarche in girls) are used as well.²²

Curves progress in approximately two-thirds of skeletally immature patients before they reach skeletal maturity (defined in most studies as a Risser sign of ≥4 in females or 5 in males); however, only one-third of patients with scoliosis will experience more than a 10° increase in curve magnitude, and less than 10 percent will have an increase of 30° or more.²³ The likelihood of progression varies depending on sex, curve magnitude, curve location, and maturity or remaining growth potential.²⁴ One study followed 123 skeletally immature adolescents with AIS (mean age, 14 years; Cobb angle <50º) without treatment until skeletal maturity.²³ In this study, the average curve measured 33° (range, 10° to 49°) at the time of diagnosis and 49° (range, 12° to 97°) at skeletal maturity. Curves remained unchanged (i.e., progressed by <5º) in 32 percent of patients, progressed by 5° or greater in 68 percent, progressed by greater than 10° in 34 percent, progressed by greater than 20° in 18 percent, and progressed by greater than 30° in 8 percent.

Older studies, such as those involving a cohort of 444 patients in Iowa,^25-30 have also examined curve progression in untreated patients. These studies show curves can continue to progress after skeletal maturity in untreated patients, especially in those with curves measuring greater than 40° at the end of growth.^{27, 28, 31, 32} Curves greater than 50° are thought to progress 1° per year after skeletal maturity,^{30, 32, 33} while curves less than 30° at skeletal maturity have a low likelihood of significant progression during adulthood.^{6, 27} However, the likelihood of continued curve progression in any individual is affected by other factors such as curve location and direction, apical vertical rotation, and trunk imbalance.⁶

The extent to which AIS is associated with other adverse health outcomes is not well understood. Most individuals with AIS curves of mild severity do not appear to have clinical symptoms during adolescence, although recent research suggests a higher likelihood of back pain at age 18 years in persons with curves of 6° to 10° at age 15 years compared with persons without spinal curvature, as well as an increased likelihood of missed school and avoidance of activities at age 18 years in adolescents with slightly larger curves.³⁴ Older studies with long-term followup of untreated cohorts suggest that back pain and cardiopulmonary compromise, with associated disability, are common. However, many of these studies included subjects with nonidiopathic scoliosis and/or scoliosis with onset before adolescence;³⁰ newer studies composed exclusively of persons with AIS suggest a more benign natural history.

Adults with AIS may be at higher likelihood of having back pain and possibly degenerative disc changes than unaffected adults. Reports are mixed with regard to whether this significantly affects functioning. Some studies suggest the presence and severity of back pain are greater in adults with AIS than in the general population, and cause a significant impact on function.³⁵ However, other studies did not find excessive disability in adults with AIS despite increased prevalence of back pain,^{30, 36} or found the frequency of back pain to be similar in adults with AIS and in the general population.³¹ Back pain does not appear to be correlated with the severity of scoliotic curve;³⁷ and studies have not shown that treatment of AIS affects the likelihood of development of back pain.^{27, 35, 36, 38, 39}

Abnormal pulmonary function is strongly associated with thoracic curve size,³⁷ but clinically significant cardiopulmonary problems are seen only with severe scoliosis. Adolescents with curves of greater than 50° are at increased risk for shortness of breath in later adulthood,³⁰ and those with curves greater than 70° have diminished lung volumes,⁴⁰ but pulmonary function appears to be most significantly affected in those with curves greater than 100°.²⁷

The extent of the psychosocial impact of scoliosis during adolescence and adulthood is unclear. In addition to concerns about body image and deformity, adults with AIS may have a poorer perception of their own health status and of their ability to interact socially compared to unaffected adults, although the presence and severity of psychological problems does not necessarily correlate with the severity of the scoliosis curve.^{6, 39, 41, 42}

There are few data on mortality in untreated scoliosis. Observational data from a long-term (50-year) cohort study of persons with untreated AIS does not suggest an increase in mortality compared with the general population.³⁰ However, the loss to followup in this cohort was substantial (roughly 40% of participants could not be located).

Risk Factors

Scoliosis of nonidiopathic etiology (i.e., neuromuscular or congenital scoliosis) is often associated with other clinical findings and/or symptoms that should prompt evaluation of the spine. As noted above, however, AIS is most often asymptomatic during adolescence, and is not typically associated with clinical findings other than body asymmetry (which itself may be subtle, with mild degrees of spinal curvature and/or trunk rotation).

Sex is not predictive of development of AIS, although the risk of curve progression is 10 times higher in females than in males, and females therefore are more likely to require treatment of AIS.¹⁰ As noted previously, evidence exists of a possible genetic contribution to AIS development, with studies showing increased prevalence of AIS in siblings and children of affected individuals and in monozygotic (as compared to dizygotic) twins; see the section on “Etiology and Natural History” for details and references. Skeletal immaturity, and by association younger age, is associated with greater risk for curve progression, as is the magnitude of curvature at the time of detection of scoliosis.²⁴

Screening

The need for and benefit derived from universal population screening to identify adolescents with mild or moderate idiopathic scoliosis (i.e., scoliotic curves of <40°–50°) has been a subject of debate and disagreement in the medical community for several decades. Curves of this degree are often asymptomatic in adolescence, with the exception of cosmetic deformity;⁴³ the majority of such curves will not progress significantly during adolescence,²³ and the likelihood of continued progression in adulthood is low for curves less than 30° at skeletal maturity.²⁸ However, the ability to identify which cases of AIS are likely to worsen significantly during adolescence is limited. Therefore, the rationale behind screening for milder degrees of scoliosis is that if early, effective treatment can be instituted for persons with AIS, then curve progression can be slowed or halted before skeletal maturity, which theoretically could improve long-term outcomes.

Most AIS screening methods are low-cost and noninvasive; however, because they measure trunk rotation or trunk asymmetry rather than actual spinal curvature, and because interexaminer error precludes reliable correlation of screening results with a specific degree of spinal curvature, a confirmatory x-ray is needed to quantify severity of AIS.^{44, 45}

Treatment

The goal of AIS treatment is to slow or halt the progression of the scoliotic curve during the adolescent growth period. Options for treatment include observation, bracing, surgery, or nonsurgical intervention such as physiotherapy. Exercise therapy is recommended for mild scoliosis in some countries but has not been routine in the United States. The choice of therapy depends primarily on the degree of curvature and potential for further growth (both of which determine the risk for progression). Because most cases of AIS will not have symptoms other than spinal curvature during adolescence,⁴³ the typical approach to treatment of AIS is informed by a patient's Cobb angle and developmental maturity: higher Cobb angles and lower levels of maturity are generally felt to warrant more aggressive intervention.⁵⁵

There are no guidelines for management of AIS published by professional societies in the United States.⁴³ International organizations have published such guidelines,⁵⁵ and current practice in the United States is relatively uniform with regard to basic elements of management. However, there is some variability among primary care providers with regard to imaging and referral for treatment.⁵⁶ Persons with a Cobb angle of less than 20° (or an ATR <7º) are usually observed without treatment; this is often done by the primary care provider, with referral to a specialist in the event of continued curve progression. Those who recommend exercise treatment often direct it at mild scoliosis of this magnitude. According to some guidelines for management of adolescents with substantial growth remaining (Risser sign, 0 to 2),⁴⁹ those with a Cobb angle of 20° to 29° are braced if they exhibit curve progression (i.e., increase in Cobb angle ≥5° over 3 to 6 months), those with a Cobb angle of 30° to 40° should usually be braced, those with a Cobb angle of 40° to 50° may be managed with bracing or surgery, and those with a Cobb angle greater than 50° usually require surgical intervention. Some research suggests a correlation between severity of curve and the risk and complexity of surgical treatment.⁵⁷

Brace Treatment

There are several types of braces used for treatment of AIS (Table 1). Braces fall into three general categories: full-time rigid bracing, nighttime rigid bracing, and soft bracing. Brace selection is based on curve location and characteristics and the anticipated tolerance of the patient.⁵⁸ Most rigid braces are prescribed for use 20 to 24 hours per day.^{22, 55, 59} Thoracolumbosacral rigid braces, such as the Boston brace, are the most frequently used in North America.⁶⁰ Nighttime braces are worn 8 to 12 hours while sleeping; they are used for certain types of curves.^{55, 61} Soft braces are adjustable, flexible, and noninvasive compared to other braces.^{58, 62} The most widely used brace in the United States until the 1970s was the Milwaukee brace; this brace may still be used for very high thoracic curves, but most cases of AIS are now managed with braces that do not rise to the neck and therefore are more cosmetically acceptable.^{58, 60, 63, 64}

Table 1

Types of Braces for AIS.

Brace treatment is not intended to correct curvature but rather to slow or halt curve progression; bracing therefore is primarily indicated for skeletally immature patients (Risser sign, 0 to 2) at high risk of rapid curve progression. Skeletal immaturity has traditionally been defined as a Risser sign of 0 to 2; newer measures of skeletal maturity, such as digital skeletal age, are increasingly being used, as they correlate better with acceleration of curve progression.⁶⁵ Treatment is generally continued until skeletal maturity (Risser sign of 4 in girls and 5 in boys). Skeletally mature patients with a Cobb angle of less than 30° to 40° are thought to be at low risk for continued progression, and typically are not monitored in adulthood.

Current Clinical Practice in the United States

Recommendations of Other Groups

Several specialty groups have published recommendations or statements in support of screening, but none are based on a systematic review of evidence (Table 2).

Table 2

Other Recommendations on Screening for AIS.

Routine screening for AIS has been recommended since the 1980s by the American Academy of Orthopedic Surgeons,⁷⁶ the American Academy of Pediatrics,⁷⁷ and the SRS,⁷⁸ and either was required by law or established voluntarily in more than half of U.S. states at that time.⁷⁹ However, contemporaneous recommendations from other countries either recommended against screening⁸⁰ or acknowledged the poor evidence base in support of it.⁸¹

The U.S. Preventive Services Task Force (USPSTF) found insufficient evidence to recommend for or against routine screening in 1993,^{82, 83} and recommended against screening in 2004 (see the section on “Previous USPSTF Recommendations” for details).⁸⁴ However, routine screening for AIS continues to be endorsed by the American Academy of Orthopedic Surgeons, the American Academy of Pediatrics, the SRS, and the International Society on Scoliosis Orthopedic and Rehabilitation Treatment.⁴³

Previous USPSTF Recommendation

In 2004, the USPSTF recommended against the routine screening of asymptomatic adolescents for AIS (D recommendation), based on the results of a brief evidence update.⁸⁴ This constituted a change from its previous 1993 C recommendation (equivalent to an I statement under current methodology), in which the USPSTF found insufficient evidence to recommend for or against routine screening.^{82, 83}

The USPSTF did not find good evidence that screening asymptomatic adolescents detects idiopathic scoliosis at an earlier stage than detection without screening. The accuracy of the most common screening test—the FBT, with or without a scoliometer—in identifying adolescents with idiopathic scoliosis is variable, and there is evidence of poor followup of adolescents with idiopathic scoliosis who are identified in community screening programs.

The USPSTF found fair evidence that treatment of idiopathic scoliosis during adolescence leads to health benefits (decreased pain and disability) in only a small proportion of persons. Most cases detected through screening will not progress to a clinically significant form of scoliosis. Scoliosis needing aggressive treatment, such as surgery, is likely to be detected without screening.

The USPSTF found fair evidence that treatment of adolescents with idiopathic scoliosis detected through screening leads to moderate harms, including unnecessary brace wear and unnecessary referral to specialty care. As a result, the USPSTF concluded that the harms of screening adolescents for idiopathic scoliosis exceed the potential benefits.