Management of adolescent idiopathic scoliosis (AIS) with low-profile thoraco-lumbar-sacral orthoses (TLSOs) is a common treatment modality.1 Successful orthotic management requires patients to wear the orthoses for significant periods each day. Some controversy exists as to what the optimal wear schedule should be, but it is becoming clear that a dose-based relationship exists between use of the orthosis and outcomes.2–4 The precise number of hours required to positively affect curve progression is unclear; however, Katz et al.2 showed that those adolescents who wore their orthoses 12 or more hours per day had significantly better results than did those who wore their orthoses for fewer hours. Patients do require some time to reach their maximum daily wear, as shown in the report by Lou et al.5 showing that wear increased significantly between months 1 and 3. We also know that overestimation of orthosis wear time is common.6 Encouraging patients and families to use these orthoses for an adequate number of hours each day is a significant challenge to practitioners.
The clinical teams responsible for the treatment of patients with AIS at Texas Scottish Rite Hospital for Children (TSRHC) are completing a large multiyear study to address orthosis wear, patient/family counseling, and outcomes. Scoliosis treatment at TSRHC is carried out through seven orthopedic teams. Each team includes an orthopedist, a clerk, nurses, orthopedic residents, fellows, therapists, and an advanced practice nurse or physician assistant. A dedicated orthotist measures, fits, and follows up with the orthoses worn by that team’s patients. The orthotist spends the greatest amount of time directly with the patients and families because, after participating in the clinic visit, the orthotist measures and may scan the patient. The next visit is scheduled for 2 hours with the orthotist for delivery and final adjustments to the orthosis. However, the team is directed by the orthopedist, who follows his/her clinical judgment in prescribing spinal orthoses. The prescribed hours of wear vary from 16 to 22 hours per day, based on physician preference.
Two of these teams were used in this analysis because of a consistent and significant difference in the weaning protocol given at orthosis delivery. Both of these teams prescribe TLSO wear of 22 hours per day. However, one author counsels patients to rapidly begin full-time wear of the orthosis within 1 week (fast group). The subjective rationale for this approach includes the idea that it is easier to start new habits immediately. This avoids learning strategies to rationalize leaving the orthosis off. The parent and the child are given a clear idea of how many hours to wear the orthosis each day. Any variations from a fast-track increase in wear times are negotiated in advance.
The other team’s patients (slow group) are counseled to take 2 weeks or more to gradually acclimatize to the new orthosis. The subjective rationale includes allowing time for the skin to develop tolerance to the significant pressures of the orthosis. School wear is delayed until the child has demonstrated wearing the orthosis for an entire day without problems. This is because school is a more challenging environment than the home is, and inability to wear the orthosis for the entire school day could be considered a failure. The families are given broad guidelines for increasing wear of the orthosis. It is felt that by empowering families to fit orthosis use into their schedule, families will take ownership of the process. We also want to acknowledge the difficulties of wearing a scoliosis orthosis.
Our goal for either weaning program is to provide a transition from not having an orthosis to wearing an orthosis for the prescribed daily hours. All patients are given strategies for coping with the common concerns, including those about clothing and social acceptance. We understand, however, that some adolescents may never reach the prescribed hours of daily wear. Only through the use of an objective data logger can we understand how orthoses are truly worn. For the purposes of this study, we are looking at the maximum number of hours that an orthosis is worn on any single day, not a steady state.
This study came out of attempts to define a single optimal orthosis wean-in protocol. Understanding the outcomes of our current multiple protocols appeared to be the first piece of information needed. One hypothesis is that AIS patients can successfully wean into appropriate spinal orthosis wear within 1 week. The opposing hypothesis is that patients and families require 2 to 3 weeks to successfully wean into wearing a spinal orthosis. Each hypothesis makes the assumption that the instructions given influence families’ behavior, which has not previously been demonstrated. The unique features of this report include having objective data and adequate numbers to provide statistical evidence for this usually nebulous subject.
Patients diagnosed with AIS and prescribed full-time TLSO use to control progression were invited to participate in a study of orthosis wear. If the family agreed, informed consent was obtained and the patient was enrolled into an institutional review board (IRB)–approved study (UT Southwestern IRB: STU 082010–153) following the Scoliosis Research Society protocol. They then received standard treatment with either a Boston Brace or a custom TLSO manufactured with a computer-aided design/computer-aided manufacturing (CAD/CAM) system. The CAD/CAM orthoses mimicked the design of the Boston Braces with posterior openings and the same patterns of pressures and relief areas, except that some of the corrective forces were built into the plastic shape. The type of orthosis was determined by the timing of implementation of the CAD/CAM system in our facility.
Two iButton® thermochron temperature loggers (Maxim Integrated, www.iButton.com) were installed in each orthosis. Each logger holds 85 days of data, measured every 15 minutes. Two loggers were used to provide more than 5 months of continuous data. At each clinic visit, the temperature data were downloaded and the sensors were relaunched to start collecting data again. The captured data were processed to determine the hours of wear each day. This process has been previously reported.7 However, no standard software was available at the time to process data in this way. Micro-T version 5.06.032 from NexSens Technology (www.NexSens.com) was used to communicate with the thermochron data loggers. The data were then exported to Microsoft Excel, where a custom VBA macro of more than 750 lines in length was used to process the data and create the reports. The rules used to determine wear from the raw temperatures require rapid movement into or out of a range from 85°F to 97°F (29.4°C–36.1°C). Thermochron data are an estimate of wear, as the most accurate that it could be is within 15 minutes of the donning and doffing times. Also, our algorithm ignores either donning or doffing events that last less than an hour. This mechanism gives some patients credit for 24 hours of wear in a day. Conversely, a patient wearing the orthosis for less than an hour at a time will not get credit for that wear. Other groups have used simpler algorithms to convert temperature to wear.4,7
In this report, we focused on the initial wear period, before the first download of data. Therefore, the patients in this report had not yet seen the objective printouts of their wear. The reporting period ranged from 27 to 47 days, ending when each subject returned to the clinic. The number of subjects available for analysis dropped off after 30 days, with only 30 of 41 (73%) having data available at 40 days.
We used a convenience sample of patients enrolled in the larger study who had continuous data available from the orthosis delivery until the first clinic return visit. At the time of this analysis, there were 66 patients who had or were being treated by the two teams being analyzed. Of these, only 42 had complete data for the analysis. A number of the cases of missing data were early patients in the study, before we improved our data collection methods. The two patient groups did not differ significantly, except that the fast group had greater in-brace correction (Table 1). The day-by-day wear times were collected from the data sets to compare the rates of increasing wear time.
A p value of <0.01 was chosen to demonstrate statistical significance. The Fisher exact test was used to compare categorical variables, such as Risser sign and the rates of reaching 16 or 22 hours of wear. The two-sample t-test was used to compare the continuous variables, such as the mean hours of wear for weeks 1 through 4 and the number of days needed to achieve 16 or 22 hours of wear.
The fast group increased their wear time more rapidly than the slow group did (Figure 1). However, not all of these patients reached the prescribed 22 hours per day of wear. To provide a secondary analysis, we chose 16 hours of daily wear to be an effective level, as one of our teams prescribes this number of hours and it is well above what Katz et al.2 had shown as a minimum for successful orthosis treatment. The rate of reaching 16 hours per day of wear during the first month and a half after receiving an orthosis was 92.3% in the slow group, with one patient not reaching this level, and 92.9% in the fast group, with two patients failing to reach 16 hours of wear in a single day (p = 1.000). The success rate of getting to 22 hours per day was much lower in both groups with 69.2% in the slow group and 78.6% in the fast group, without a significant difference between them (p = 0.6979) (Table 2). However, these numbers reflect the rates of reaching the threshold number for at least 1 day. They do not necessarily indicate a steady state. For instance, 92% of patients successfully reached 16 hours of wear on at least 1 day, but during week 4, the slow group averaged only 13.6 hours per day.
Looking more closely at week 3, we find a total of 287 opportunities for patients in the combined cohort to wear their orthoses (patient-days). One patient failed to wear the orthosis at all during this week. Including that case, nine patients (22%) took a total of 26 “brace holidays” (9%) by not wearing the orthosis for a day. A total of eight patients (19%) never wore the orthosis 16 hours or more during week 3. Also, during this period, there were 60 patient-days (31%), representing 23 patients (56%), with 22 hours or more wear. Only two patients (5%) met the 22-hour prescription level for the entire 7 days.
Significant differences between the groups were found in the hours worn per day during the first 2 weeks. During week 1, the fast group averaged 9.9 hours per day, whereas the slow group managed only 5.2 hours per day (p = 0.0083). Similarly, during week 2, the fast group achieved 16.7 hours per day, on average, whereas the slow group remained a week behind, with an average of 9.6 hours per day (p = 0.0004). Week 3 approached significance, with the fast group wearing the orthoses, on average, 5 hours more each day than the slow group (p = 0.0123). By week 4, there was no longer a significant difference between the average hours of wear between the two groups (p = 0.1011). The total hours of wear during the study period also differed significantly between the groups, with the fast group averaging 683 hours versus the slow group patients, who totaled an average of 466 hours (p = 0.0088) (Table 2).
We set out to find an optimal weaning schedule by using objective data. The data demonstrate that differing approaches to orthosis weaning lead to different behaviors on the part of the patients and families. The results of both groups show that adolescents are able to develop new habits of orthosis wear in 1 to 3 weeks, on average. How rapidly the new habits are implemented is statistically significantly influenced by the counseling received. Some adolescents can accomplish this goal in 1 week, if so instructed. Others fail to wean into full-time orthosis wear during the first month regardless of the instructions given. The factors determining these outcomes and the clinical relevance of these results are still unknown. We will have to look at the long-term follow-up to see if the overall wear and outcomes differ between the groups with different weaning protocols. Thus, we have been unable to determine if either of these protocols is optimal.
The numbers involved in this study are small, especially in the slow group, which is a limitation of this report. As this was a very short-duration study, we do not know whether the weaning period is associated with long-term wear or outcomes. However, Lou et al.5 state that “The first 6 months is considered a critical time to evaluate whether AIS patients will commit to the treatment…” of scoliosis. Those patients who struggle with the initiation of treatment may never reach a therapeutic level of orthosis wear.
Also, our clinic teams are fairly large. Families discuss their cases with at least four clinicians during a typical clinic visit. We have chosen our two groups based on the orthotists’ approaches to implementing orthosis wear. However, the attitudes and words of the rest of the clinic team also have an impact on the “buy in” of the families. As each orthotist is part of a team, we are unable to control for differences in the counseling received from other clinicians.
An unexpected finding was a statistically significant difference in the in-brace correction between the two groups (28% vs. 47%; p = 0.0102). However, this study is not rigorously designed to compare in-brace correction. For instance, the time between donning of the orthosis and the radiographic procedure was not standardized. As Li et al.8 have shown, this time can make a difference to the in-brace correction. If the time is ignored, it is still not clear what effect in-brace correction might have on patient behavior. One possibility is that one group is more flexible in the aggregate than the other, which might lead to more comfortable wear of an orthosis. Another possibility is that the orthoses provided by one orthotist provide more aggressive corrective forces than do those from the other author. As these variables were uncontrolled, it is hard to draw any conclusions from the in-brace correction difference.
A further limitation of this study is that the orthoses were not standardized. The initial orthoses used custom-to-measurement Boston modules, whereas later orthoses were made on site using a CAD/CAM system. Some of those later braces were fully lined with 1.25-cm pressure pads, whereas others were quite asymmetrical and had 5-mm padding in the corrective pressure regions only. It is possible that brace design could influence wear behavior and no attempt was made to control for this variable.
The optimal instructions to provide to patients and families are unknown. However, these data show that the instructions given to families have a measurable effect on the behavior of patients. As clinicians, our counseling of patients and families will affect their use of an orthosis. We need to be thoughtful and purposeful when providing instructions to our clients.
The authors extend their appreciation to TSRHC for supporting this research. Numerous staff members have spent many hours, in addition to their clinical responsibilities, collecting and processing data.
1. Stokes OM, Luk KD. The current status of bracing for patients with adolescent idiopathic scoliosis. Bone Joint J
2013; 95-B: 1308–1316.
2. Katz DE, Herring JA, Browne RH Brace wear control of curve progression in adolescent idiopathic scoliosis. J Bone Joint Surg Am
2010; 92: 1343–1352.
3. Rahman T, Bowen JR, Takemitsu M, Scott C. The association between brace compliance and outcome for patients with idiopathic scoliosis. J Pediatr Orthop
2005; 25: 420–422.
4. Weinstein SL, Dolan LA, Wright JG, Dobbs MB. Effects of Bracing in adolescents with idiopathic scoliosis. N Engl J Med
2013; 369: 1512–1521.
5. Lou E, Hill D, Raso J Brace wear characteristics during the first 6 months for the treatment of scoliosis. Stud Health Technol Inform
2012; 176: 346–349.
6. Morton A, Riddle R, Buchanan R Accuracy in the prediction and estimation of adherence to bracewear before and during treatment of adolescent idiopathic scoliosis. J Pediatr Orthop
2008; 28: 336–341.
7. Benish BM, Smith KJ, Schwartz MH. Validation of a miniature thermochron for monitoring thoracolumbosacral orthosis wear time. Spine (Phila Pa 1976)
2012; 37: 309–315.
8. Li M, Wong MS, Luk KD Time-dependent response of scoliotic curvature to orthotic intervention: when should a radiograph be obtained after putting on or taking off a spinal orthosis? Spine (Phila Pa 1976)
2014; 39: 1408–1416.