Early Detection of Scoliosis—What the USPSTF “I” Means for Us

In the current issue of JAMA is the latest recommendation statement by the US Preventive Services Task Force (USPSTF),¹ and accompanying evidence report,² on the value of screening asymptomatic adolescents for adolescent idiopathic scoliosis. Since the topic was last reviewed in 2004, important new evidence has emerged that potentially supports scoliosis screening. In this update, the USPSTF changed its grade of the evidence from a “D” (discourage the use of screening programs) to an “I” (uncertainty about the balance of benefits and harms of the service).¹ The recommendation highlights the high sensitivity, specificity, and predictive value and the low false-positive rate of screening programs when the Adams forward bend test is paired with scoliometer measurements and Moiré topography. It is important to recognize that a properly implemented screening program will identify potential patients who can benefit from brace treatment, possibly avoiding surgery. In addition, spinal deformity may be the presenting sign of a variety of conditions, including heritable collagen disease, neurological conditions, or skeletal dysplasia unrecognized until adolescence. Even if surgery cannot be averted, early diagnosis of progressive curves allows for surgical intervention at the most opportune time.

The most important development since the last task force evaluation was the publication of prospectively controlled studies on the efficacy of brace treatment, which found a benefit in brace treatment over observation when comparing curve progression rates.^3-5 The National Institutes of Health (NIH)–funded Bracing in Adolescent Idiopathic Scoliosis Trial (BrAIST)⁴ evaluated brace treatment and observation in adolescents with idiopathic scoliosis between 20 and 40 degrees. It showed that brace treatment provided for a 56% reduction in the relative risk of scoliosis curve progression past 50 degrees, which was the defined measure of bracing failure, and also found a high correlation between bracing success and the number of hours of brace wear. There was limited evidence of treatment harms. Skin problems on the trunk (under the brace) and nonback body pains were more frequently reported in braced participants than in observed controls, but anxiety and depression rates were low and similar between groups. The study’s interim analysis showed that braced patients had significantly better rates of treatment success than nonbraced patients; therefore, the NIH Data Safety and Monitoring Board concluded the study prior to full enrollment.

The BrAIST study⁴ highlights several challenges of generating level 1 evidence on screening and brace treatment programs. Insufficient enrollment in the randomized arm of the study led to the inclusion of the patient preference arm of the BrAIST study in which the patients/parents showed a greater than 2 to 1 patient enrollment preference for brace treatment over observation. The ethics of future observational studies, given the strength of evidence in favor of brace treatment, is in question. Similarly, prospectively evaluating the adult impact of untreated idiopathic scoliosis with a 30-year or longer observational study would present an enormous challenge.

Undetected scoliosis deformity that advances to a surgical level carries its own significant consequences, which were not addressed by the Task Force recommendation.¹ Whereas brace treatment in adolescence is relatively low cost with little risk of adverse events, surgical treatment is extremely costly. In uncomplicated surgical cases, the family burden and loss of school time are measurable, but more concerning are the risks of surgical complications, including paralysis and deep infection, in the more severe deformities.⁶ Furthermore, the need for long-term revision surgery is as high as 22%.^7,8 Curves that progress past adolescence and require adult reconstruction are associated with dramatically increased risks of operative complications and even higher reoperation rates.^9-12 The USPSTF does not consider costs for any of the preventive services in its recommendations, as a matter of scope. By not giving adequate consideration to the risks and costs of surgical treatment in adolescence and adulthood compared with those of brace treatment, the analysis underestimated the value of early diagnosis that can be achieved through screening programs.

The USPSTF recommendation¹ referenced a number of studies evaluating the association of adults’ patient-reported health outcomes with the severity of their adolescent spinal curves. However, these studies were limited either by short-term follow-up (to adolescence) or by patients having in fact received bracing or surgical treatment. None of the study groups referenced in the USPSTF report represented the adult health burden of untreated progressive scoliosis, as none documented a natural history of untreated adult disease. The results were thereby biased against demonstrating the effect of pediatric bracing programs on adult health-related quality-of-life (QOL) measures. Several adult spinal deformity study groups have published extensively on the effect of untreated progressive deformity on adult health-related QOL measures and their effects on health economics.^13-18 Their studies indirectly underscore the long-term treatment effects of a pediatric bracing program on adult QOL. These studies, however, did not appear in the USPSTF report.

The USPSTF should be applauded for the change in recommendation from a “D” to an “I”. But a potential unintended consequence of the USPSTF recommendation is how clinicians will use it in their everyday practice. The evidence report addresses 6 key questions specific to a diagnosis of adolescent idiopathic scoliosis. For key question No. 2, “what is the accuracy of screening for adolescent idiopathic scoliosis?”, 6 of the 7 fair-quality prospective cohort studies included in the evidence review² were conducted in the setting of school screening. Medical societies may interpret the USPSTF “I” statement, specific to school-based screening, as a directive not to perform any examination of the spine during a routine office examination. For example, the American Academy of Family Physicians in their Choosing Wisely list state “don’t screen adolescents for scoliosis.”¹⁹ However, adolescents may have various other spine diagnoses that the report does not address that could benefit from early detection by a clinician in the medical home. While the “I” statement is very different than a “D” recommendation, it is not certain how schools and individual clinicians will interpret the new recommendation.

Although the current analysis by USPSTF¹ represents an upgrade in evidence from its prior recommendations against screening programs, clinicians who care for adolescents have expressed disagreement with USPSTF on this issue. The Scoliosis Research Society, American Academy of Orthopaedic Surgeons, Pediatric Orthopaedic Society of North America, and American Academy of Pediatrics recommend that girls be screened twice, at ages 10 and 12 years, and boys once, at age 13 or 14 years.²⁰ Further research into the value of early scoliosis detection during the adolescent medical home preventive health visit needs to continue.

The critical concept is that early detection of scoliosis and nonoperative treatment are inseparable. If presurgical curves are not clinically detected, patients cannot benefit from the potential surgical risk reduction of a brace treatment program. The recent bracing data convincingly support the ability of brace treatment to minimize the risk of progressing to the point of requiring surgical treatment. We continue to believe strongly that early recognition and diagnosis is important so that family-centered, shared decision making in the treatment of scoliosis can be offered to our patients and their families.

Corresponding Author: M. Timothy Hresko, MD, Department of Orthopaedic Surgery, Boston Children’s Hospital, 300 Longwood Ave, Boston, MA 02467 (timothy.hresko@childrens.harvard.edu).

Published Online: January 9, 2018. doi:10.1001/jamapediatrics.2017.5585

Conflict of Interest Disclosures: Dr Hostin reports serving as consultant to DePuy Spine; he received research support from NuVasive, Seeger, Don Joy Orthopedics, DePuy, and K2M. Dr Schwend reports serving as consultant to K2M and Medtronics; on the medical advisory boards for Miracle Feet and Project Perfect; he has served as president of the Pediatric Orthopedic Society of North America (POSNA), committee chair of the Scoliosis Research Society Health Policy committee, and as executive committee member for the American Academy of Pediatrics. Dr Hresko has served as committee chair for the Scoliosis Research Society Awards committee and the POSNA Health Policy Council; he has received travel/education support from Medtronics and NuVasive and an honorarium from Boston Orthotics and Prosthetics. No other disclosures are reported.