Investigation of the Providence TLSO for Treatment of Scoliosis

Hannah G. Godlewski, Resident Orthotist and C. Leigh Davis, CPO, MSPO


Adolescent idiopathic scoliosis (AIS) has been defined by the Scoliosis Research Society (SRS) as a lateral curvature of the spine greater than ten degrees as measured using the Cobb method on a standing radiograph, with no clear underlying cause. 1 Scoliosis is present in 2 to 4% of children aged between 10 and 16 years of age. Idiopathic scoliosis is the most common cause of spinal deformity in the adolescent population. 2, 3

SRS guidelines recommend orthotic treatment for adolescents diagnosed with AIS with curves between 25 and 40 degrees. The goal of orthotic management in idiopathic scoliosis is to alter the natural history of the patients' curve. Natural history data refer to "estimates" and "likelihoods" derived from findings in large population samples. They do not tell us what will happen to an individual child; therefore each patient should be managed on a strict case-by-case basis. 7 An orthosis is the only potentially effective non-operative treatment for preventing curve progression in adolescent idiopathic scoliosis. 5 However, the effectiveness of orthotic management remains controversial.

Lonstein and Carlson found that skeletally immature patients (Risser 0 or 1) with curves 20 to 29 degrees who were not treated had a 68% risk of curve progression. 4 Factors associated with an increased risk of progression include curve magnitude, skeletal immaturity, and double curve patterns. Nachemson and Peterson also reported that 66% of untreated patients with curves between 25 and 35 degrees will progress by 5 degrees or more. 5 Bunnell reported progression of at least 5 degrees in 68% of patients, 10 degrees in 24% of patients, and 20 degrees in 18% of patients. 6

Many studies have researched the effectiveness of the use of full-time thoracolumbar orthoses (TLSOs) and night-time TLSOs in the conservative treatment of AIS. Emans et al has suggested that part-time bracing may be as effective as full-time bracing. 10 The Charleston and Providence orthoses are the two most common orthoses worn only at night. The Providence orthosis, with recommended wear time of 8-10 hours per night, has been used to control progressive AIS curves. In previous retrospective studies, the Providence orthosis seems to alter the natural history, but they are generally found inferior to or no different from full-time orthotic strategies in comparison studies. 9, 11, 12 In a recent study, D'Amato et al found excellent results from treatment of scoliosis with the Providence, with a success rate of 79%, if the apex was at or below T9. 13 Other studies have concluded that the Boston brace is more effective in preventing curve progression, but night bracing should be considered in the treatment of thoracolumbar and lumbar curves less than 35 degrees. 9 Janicki's 2007 study compared treatment with a full time Boston TLSO to a night time Providence TLSO. He showed that 15% of Boston TLSO patients and 31% of Providence patients did not progress more than 5 degrees. This research concluded that the Providence orthosis was more effective for avoiding surgery and preventing curve progression than a Boston, when the primary initial curves at bracing was 35 degrees or less. 8

Some suggest that the compliance with the Providence brace may be greater, which may improve results. The fundamental idea of nighttime-only bracing is to decrease the negative effects of brace wear to the quality of life without compromising the final results of treatment. This is theoretically possible with better brace correction and better compliance with wear. These are probably the only two factors which we can influence during brace treatment of scoliosis. 14 Climent and Sanchez found the nighttime only bracing had the least negative effect on psychological functioning, sleep disturbances, back pain, body image, and flexibility. 15

Research indicates that the use of bracing with any scoliosis orthosis, including the Providence, in curves greater than 35 degrees to control AIS is questionable. The results using the Providence orthosis in smaller curves (25-35 degrees) were more favorable. These results support the need for a larger study using the SRS inclusion and assessment criteria. This study will examine a previously unobserved group of 80 patients (female and male), all treated by the same organization with the same type of Providence bracing, but with varying degrees of scoliosis, including high and low magnitude curves. Success of bracing treatment will be determined by curve magnitude at the conclusion of treatment compared to initial curve magnitude and natural history data.


Study related data was gathered from CHOA's electronic patient chart system (EPIC and Chartmaxx), Children's Orthopedics of Atlanta's patient transcriptions, and from the records of Spinal Technology, Inc, who manufacturers the Providence. Data was collected from the patient notes from each visit with their orthopedist and reviewed by HG. Radiographs were measured using the Cobb technique and Risser sign assessed by the prescribing orthopedist and when available for retrospective review, by LD. Patient information recorded and reviewed included:

  1. Name/Patient ID
  2. Date of Birth
  3. Gender
  4. Age at scoliosis diagnosis
  5. Medical history
  6. Physical findings
  7. Radiograph at initial visit
  8. Menarcheal status at brace initiation
  9. In-brace radiograph
  10. Length of follow-up
  11. Radiograph at brace discontinuation
  12. Radiograph from most recent follow-up.

435 patients were identified who received a Providence orthosis from the CHOA Department of Orthotics and Prosthetics between 2006 and present.

Then the following inclusion criterion was applied:

  • At least 4 years of age
  • Risser 0, 1, or 2 at brace initiation
  • Initial curve between 20 and 45 degrees
  • No prior orthotic treatment
  • Pre menarche or less than 1 year post menarche
  • Diagnosed with idiopathic scoliosis.

Only patients who had completed bracing treatment were included in the analysis. All patients with intent to treat were studied, regardless of compliance during treatment. A failure criterion was set at greater than 5 degrees of curve progression and/or spinal fusion.

Eighty patients from the original 435 patients met the criteria for inclusion. A large percentage of those excluded are still undergoing orthotic treatment for their scoliosis, and thus did not meet the criteria for having completed treatment. The range of treatment time was 6 months to 39 months. Nine of the patients were male, 71 were female. The patients ranged from 8 to 16 years of age at treatment start, with median age of 12 years old. This median age was the same for the group whose treatment was successful and the group whose treatment failed. All patients were diagnosed by one of five orthopedists at Children's Orthopedics of Atlanta. All of these patients were treated between 2006 and 2010 by Certified Orthotists at CHOA who were trained in Providence treatment. Selection of orthotic treatment with the Providence was based on physician preference.

The Providence night orthoses utilized for this study were fabricated by Spinal Technology, Inc., using the principles as described by D'Amato et al. Each orthosis was designed for the particular patient's presentation by a trained orthosis using an acrylic frame to apply direct corrective forces to the patient as directed by the Providence manual. The orthoses thus apply controlled, direct, lateral and rotational forces on the trunk to move the spine toward the midline or beyond the midline; they do not bend the spine. 16

A statistical analysis on the data collected was completed to review the success and failure rates of the patients included in the study. The patient data was sorted into 4 groups:

  1. All 80 patients studied.
  2. Patients that met SRS inclusion criteria (age 10 or older at brace initiation, Risser 0, 1, or 2, primary curve 25-40 degrees, previous orthotic treatment, and premenarche or <1 year post-menarche).
  3. Patients with a primary curve of less than 25 degrees.
  4. Patients with a primary curve greater than or equal to 35 degrees.

All analyses were performed using SAS 9.2 (Cary, NC). Statistical significance was assessed at the p<0.05 level of significance. The mean and median Cobb angle measurements were calculated for each group. Chi-square tests were used to detect differences between categorical variables and the outcome variable (success or failure of brace). Finally, the Wilcoxon-rank sum test was used to compare the distribution of age in each outcome group.


For all 80 patients who met inclusion criteria, 46.2% had successful treatment with curves progressing only 5 degrees or less. Forty-nine patients met the SRS inclusion criteria, and 23 (46.9%) had a successful treatment outcome. Eighteen patients had initial curves of less than 25 degrees, and in this group there was a 55.6% success rate. Only 2 (15.4%) of the 13 patients with initial curve magnitude greater than 35 had successful treatment.

The distribution of the pattern of curve types for all 80 patients ( Figure 1 ) showed that the majority (48.75%) were double major curves. The mean initial curve degree for the total sample was 29 degrees with standard deviation of +/- 6.3. The range of initial curve for all 80 patients studied was 15 to 51 degrees. For the forty-nine patients that met SRS inclusion criteria, the mean initial curve was 30 degrees with standard deviation of +/- 4.0, and range was 25 to 38 degrees. The average curve at initiation for small magnitude curves (<25 degrees) was 21.7 degrees with a standard deviation of +/- 2.5. The range of curves for the smaller magnitude group is 15 to 24 degrees. The high magnitude group of 13 patients had a mean initial curve degree of 39.4 degrees with a standard deviation of +/- 4.3, with a range of 36 to 51 degrees.

Figure 2 shows the distribution of skeletal age as assessed by Risser sign. 75% of the 80 subjects were Risser 0 at brace initiation.

The findings of Nachemson and Peterson showed that girls with AIS with curves between 25 and 35 degrees with apexes between T8 and L1 that wore a TLSO for 16 hours per day had a 74% success rate. A similar analysis of our study group comparing 35 patients with apexes between T8 and L1 and curves measuring 25-35 degrees found that 57.1% were successful with Providence nighttime treatment and unknown compliance level.


There were no known complications that occurred during treatment of the 80 patients studied.


When compared to natural history data, this investigation supports the treatment of scoliosis with the Providence orthosis. Lonstein and Carlson reported a 68% rate of progression in untreated boys and girls Risser 0 or 1 with curves 20-29. The 30 patients from our study that had these same demographics had a 50% rate of progression.

When compared to the two recent studies looking at outcomes with use of the Providence TLSO, the results of this investigation lie soundly in the middle of those previously reported. In all groups, we fell short of the 74% success rate reported by D'Amato et al, and surpassed the 31% success rate reported by Janicki, et al. D'Amato had the highest success rate (94%) for lumbar curves and 93% success rate for thoracolumbar curves. Similar to D'Amato, we had our highest success rates (based on curve type) with thoracolumbar curves (curves with apex T12 or L1) and lumbar curves. We had 100% success rate for lumbar curves, although sample size (3) was minute. Of the 16 subjects with thoracolumbar curves, 9 (56%) had successful treatment. This compares to success rates of 36% for thoracic curves and 44% for double major curves. Also similar to D'Amato, et al, is our lower success rate with higher apex curves. For curves with an apex of T8 or above, only 15% did not progress.

The question of why our results are much lower than D'Amato's groups is tough to answer. The patient populations were quite similar. Our patient population was slightly less mature as evidenced by a younger age at brace initiation (12 compared to 13 years) and a higher number of subjects at Risser 0 or 1 (87.7% compared to 82%). In addition, our average curve value at brace initiation was slightly higher (29 degrees compared to 27 degrees). However, it is unlikely that such small changes could account for the entire discrepancy in success rate. Two additional factors are worth noting. D'Amato et al represents the group who developed the Providence orthosis. Therefore, they are likely to be more skilled at treatment with the orthosis, although all treating orthotists at CHOA have been trained in the method. In addition, their study was a prospective study. As part of the study, the subjects likely were required regular follow up. CHOA's protocol is to follow up with scoliosis patients every 3 months. However, the data mined for these patients included did not show nearly that rate of follow up. Some patients were only seen twice - for orthosis prescription and discontinuation. The amount of follow up could influence compliance and success with the orthosis.

When comparing to results of full time bracing, our success rate is slightly lower than typical success rate reported. Nachemson and Peterson found a success rate of 74% with their prospective study of girls with curves between 25 and 35 degrees and curve apex between T8 and L1 treated with a 16 hour/day TLSO. Comparing the subjects from our study with the same demographics, a 57% success rate was found. The main difference between our findings and Nachemson and Peterson's findings is the required wear time, which was 16 hours per day for the full time TLSO compared to 8-10 hours per night with the Providence TLSO. Although the success rate is lower, some families may wish to trade off a higher success rate for a higher quality of life with a night-time only TLSO.

This study also reviewed high magnitude curves treated with the Providence, an area with few subjects found in the literature. This study had only 13 patients (16.25% of the patients studied) with initial curves greater than 35 degrees. The rate of progression was 84.6%, with a success rate of 15.4%. This study supports that it is recommended to proceed with caution in using the Providence as the primary treatment for larger curves. As this is still too small of a sample size to draw conclusions, more studies of curves >35 degrees are needed.

In conclusion, our results support that the outcome that treatment with the Providence night time TLSO can change the natural history of scoliosis progression, as it was effective in preventing curve progression in 46% of our patients with AIS. The rate increases for smaller magnitude curves and decreases for high magnitude curves. Thoracolumbar curves, with a success rate of 56% are most likely to succeed with this orthosis compared to other curve types.


Very special thanks to Catherine Leigh Davis, CPO, for her invaluable assistance in this study. Thanks to Courtney McCracken, M.S., PH.D of Emory University for providing assistance in the statistics of this study. Also, special thanks to Rian Thornton, Clinical Research Coordinator at CHOA and Dr. Dennis P. DeVito, M.D. of Children's Orthopedics of Atlanta.

Children's Healthcare of Atlanta Orthotics and Prosthetics Department and Children's Orthopedics of Atlanta, Atlanta, GA


    1. Kane WJ. Scoliosis prevalence: a call for a statement of terms. Clin Orthop . 1997;126:43-6.


    1. Shindle MK, Khanna AJ, Bhatnagar R, Sponseller PD. Adolescent idiopathic scoliosis: Modern management guidelines. J Surg Orthop Adv 2006;15(1):43-52.


    1. Howard A, Wright JG, Hedden D. A comparative study of TLSO, Charleston, and Milwaukee braces for idiopathic scoliosis. Spine 1998;23(22)2404-2411.


    1. Lonstein JE, Carlson JM. The prediction of curve progression in untreated idiopathic scoliosis during growth. J Bone Joint Surg Am . 1984;66:1061-1071.


    1. Nachemson AL, Peterson LE, Members or Brace Study Group of the SRS. Effectiveness of treatment with a brace in girls who have adolescent idiopathic scoliosis: a prospective controlled study based on data from the Brace Study of the SRS. J Bone Joint Surg Am . 1995;77:815-822.


    1. Bunnell WP. The natural history of idiopathic scoliosis before skeletal maturity. Spine . 1986;11:773-776


    1. Grivas, TB, Rodopoulos GI, Bardakos NV. Night-time braces for treatment of Adolescent Idiopathic Scoliosis. Disability and Rehab: Assistive Tech . 2008;3 (3):120-129.


    1. Janicki JA. A comparison of the thoracolumbosacral orthoses and Providence orthosis in the treatment of adolescent idiopathic scoliosis. J Pediatr Orthop . 2007;27(4)369-374.


    1. Katz DE, Richards BS, Browne RH, Herring JA. A comparison between the Boston brace and the Charleston bending brace in adolescent idiopathic scoliosis. Spine . 1997;22:1302-1312.


    1. Emans JB, Kaelin A, Bancel P, et al. The Boston bracing system for idiopathic scoliosis. Follow up results in 295 patients. Spine . 1986;11:792-801.


    1. Gepstein R, Leitner Y, Zohar E, et al. Effectiveness of the Charleston bending brace in the treatment of single-curve idiopathic scoliosis. J Pediatr Orthop . 2002;22:84-87.


    1. Karol LA. Effectiveness of bracing in male patients with idiopathic scoliosis. Spine . 2001;26:2001-2005.


    1. Price CT, Scott DS, Reed FE Jr, et al. Nighttime bracing for adolescent idiopathic scoliosis with the Charleston bending brace: long-term follow-up. J Pediatr Orthop . 1997;17:703-707.


    1. Yrjonen T, Ylikoski M, Schlenzka D, Kinnunen R, Poussa M. Effectiveness of the Providence nighttime bracing in adolescent idiopathic scoliosis: a comparative study of 36 female patients.Eur Spine J ;15:1139-1143.


    1. Climent JM, Sanchez J. Impact of the type of brace on the quality of life of adolescents with spine deformities. Spine . 1999;24:1903-1908.


  1. D'Amato CR, McCoy B. The Providence Scoliosis System Manual. Milwaukee, WI: Scoliosis Research Society ; 2003.