An objective measurement of brace usage for the treatment of adolescent idiopathic scoliosis

Medical Engineering & Physics 33 (2011) 290–294

Contents lists available at ScienceDirect

Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy

An objective measurement of brace usage for the treatment of adolescent idiopathic scoliosis Edmond Lou a,b,∗ , Doug Hill a , Douglas Hedden b , Jim Mahood b , Marc Moreau b , Jim Raso a a b

Alberta Health Services – Glenrose Rehabilitation Hospital, Edmonton, AB, Canada, T5G 0B7 Department of Surgery, University of Alberta, Edmonton, AB, Canada, T6G 2B7

a r t i c l e

i n f o

Article history: Received 12 May 2010 Received in revised form 12 October 2010 Accepted 15 October 2010 Keywords: Orthotic treatment Compliance Load measurement Data logging Adolescent idiopathic scoliosis

a b s t r a c t Effectiveness of orthotic treatment for scoliosis depends on how much time and how well the orthosis is worn. Questionnaires and clinical judgment are subjective methods to wear compliance. Even though using a temperature sensor can objectively record how long the orthosis has been used, it may not be able to answer the orthosis effectiveness without knowing the wear tightness. Custom made thoracolumbosacral orthoses (TLSO) were instrumented with low power wireless data acquisition systems to measure the time and loads imposed by the pressure pad during daily activities. Force measurements were recorded at 1 sample/min and the system was able to record data up to 4 months without patient-involvement. Ten subjects (9F, 1M), age between 9 and 13.5 years, average 11.6 ± 1.3 years, who prescribed a new TLSO and full-time brace wear were took part in this study over 4.4 ± 1.0 months. Longterm logging of loads within a spinal orthosis is a reliable method to measure compliance objectively. The monthly quantity of brace wear ranged from 33% to 82%, average 60.0 ± 4.3%. The monthly average loads imposed by the pressure pads varied from 39% to 78% relative to the reference level, average 64.3 ± 4.6%. There was a statistically significant decrease in force, but increase in wear time over the period after the brace fitting session. This information may help to better understand the effectiveness of bracing and to predict the brace treatment outcomes. © 2010 IPEM. Published by Elsevier Ltd. All rights reserved.

1. Introduction Adolescent idiopathic scoliosis (AIS) is a three-dimensional lateral curvature of the spine associated with vertebral rotation. It is the most common type of scoliosis and affects 2–3% of the adolescent population [1]. Girls tend to progress more often than boys [2]. Although scoliosis is rarely life threatening, the long-term results of untreated scoliosis are still controversial [3–8]. Patients with untreated curves usually have more back pain [2,5] later in their life. According to the Scoliosis Research Society, adolescents who have mild curves and are still growing should be monitored until skeletal mature. Brace (orthotic) treatment is recommended for growing children with curves of 25–45◦ Cobb angle. Surgery is the final treatment option for curves greater than 45◦ and its goals are to obtain safe correction, to produce a solid spinal fusion of the curve region, and to bring the spine and body into a more balanced position.

∗ Corresponding author at: Department of Research and Technology Development, Alberta Health Services – Glenrose Rehabilitation Hospital, 10105 – 112 Ave, Edmonton, AB, Canada, T5G 0B7. Tel.: +1 780 735 8212; fax: +1 780 735 7972. E-mail address: [email protected] (E. Lou).

Although brace treatment for scoliosis has been used for more than fifty years, its effectiveness is still debatable [9–12]. Some researchers believe that brace treatment does not alter the natural history while others believe that braces (orthoses) can help stop some curves from progression [32]. The controversial opinions and results from research studies may be partially due to inconsistent inclusion criteria, different wear patterns, and different definitions of brace effectiveness. Some studies combined both male and female patients during analysis [13,14]. Some studies only included the compliant patients [15,16]. Most studies used the amount of curve progression (as measured by the Cobb angle) to determine the effectiveness of brace treatment. Some defined success as 5◦ or less curve progression; and 6◦ or more to count as failure; and some used patients who required surgery at the end of growth to count as failure. Without a consistent definition, it is difficult to evaluate the effectiveness of the brace treatment. Recently, standardization of inclusion and evaluation criteria for AIS brace studies has been set out by the Scoliosis Research Society Committee on Bracing and Non-operative Management [17]. Although the standardization of the brace study has been set, to answer the effectiveness of brace treatment, the accurate measurement of compliance (how tight – quality, and how much time – quantity) is still a fundamental issue. Wearing a brace is not fashionable; it creates a lot of stress, restriction and is uncomfort-

1350-4533/$ – see front matter © 2010 IPEM. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.medengphy.2010.10.016

E. Lou et al. / Medical Engineering & Physics 33 (2011) 290–294

able for the wearer. Even though the AIS patients are instructed to wear their braces for a prescribed amount of time, they may choose not to follow this advice. Historically, the time that the brace has been worn during daily activities is measured only subjectively. This has been done by questionnaires, asking patients or their parents, or assessing the brace for signs of wear. As technology became more advanced, electronic devices were developed to measure compliance [18–23]. Pressure switches and temperature sensors have been used to monitor how long the brace has been worn during the brace treatment. Recent reports of compliance (quantity) were around 65% [21–23]. However, these studies did not measure how well (tightly) the brace was worn. Studies were also conducted to investigate how pressures or loads were applied while patients wore their braces in the laboratory [24–29]. Cote et al. [28] used 256 thin polymeric Force Sensing Resistor sensors to measure the entire pressure distribution inside the Boston brace. Jiang et al. [27] and Cote et al. [28] reported that the average maximum pressure level inside the brace was between 4 and 9 kPa. More recently, Perie et al. [24] analyzed the Boston brace biomechanics, through pressure measurements and finite element simulations and reported that mechanisms other than the main pressure pad area also produced correction and contributed to the force equilibrium within the brace. The maximum pressure level at the major pad area was found to be 7.1 ± 1.8 kPa. Wong et al. [29] also developed a tension transducer to measure the tension of the strap. They reported that the standing Cobb angle was correlated with the pressure applied by the pad and the strap tension. However, because all the force measurement studies were performed in a laboratory environment, the actual forces or pressures applied to the body during daily activities were still not clear. Only knowing how much time a brace has been worn during daily activities is insufficient to judge the effectiveness of orthotic treatment because the quality of brace usage has not reported. In the past, researchers demonstrated that both good compliance and proper brace wear were essential to achieve effective treatment [30,31]. This paper presents work on a reliable compliance monitor that logs the quantity and quality of brace usage for at least 4 months without requiring interaction from the patients. 2. Materials and methods The “brace dosage” was determined by monitoring the wear time with the interface forces between a brace and the body during daily living. A wireless monitor was developed to collect the quantity (how much time the brace has been worn) and quality (how tightly the brace has been worn) of orthotic usage. The monitor consisted of a force transducer (Honeywell, FS01) and a wireless data acquisition unit. Its dimensions were 54 mm × 35 mm × 15 mm and its weight was 27 g. Fig. 1a shows the acquisition unit embedded inside a brace and Fig. 1b shows the sensor embedded underneath the pressure pad. The brace modification did not affect its function because there was no structural change on the plastic shell of the brace. Also, the sensor embedded underneath the force pad did not change the pad contact area. 2.1. Force sensor The sensing area of the sensor was 113 mm2 (12 mm), force range was 0–6.8 N and its thickness was 7 mm. The maximum pressure that the transducer could measure was 60 kN/m2 (60 kPa), which was suitable to this application based on other studies [24,27,28]. This sensor provided a linear output result with reliable and repeatable response (r = 0.99). Its linearity and hysteresis were ±1.0% FSS (Full Scale Span) and ±0.5% FSS, respectively. The full scale span was the algebraic difference between the output volt-

291

Fig. 1. (a) A wireless data acquisition installed into a TLSO. (b) A force sensor embedded underneath the major pressure pad.

age at full scale load and the output at no load. The transducer was only able to measure the normal forces but not the shear forces. The shear force component would not affect the normal force magnitude. The sensor gave an accurate and stable output over the temperature range from 5 ◦ C to 50 ◦ C. The loading and unloading tests (3 repetitions) of the force sensor were reported in a previous paper [31]. The sensor returned to zero value when loads were removed from the sensor. It was also tested before and after being used for a 4-month period. There was no change on its sensitivity and accuracy (Fig. 2). The resolution (sensitivity) of the force sensor was 0.01 N and its maximum error was ±0.02 N. If a patient bent away from the sensor area, the system might detect a no load condition. However to minimize false negative conditions, the algorithm defined the brace as not worn only when there was a no load condition for 5 continuous minutes. 2.2. Wireless data acquisition unit The acquisition unit consisted of a system-on-chip (SOC) wireless processor, a real-time clock (RTC), a flash memory and a battery management circuit. This specific SOC (CC2430, Texas Instrument, USA) was selected because of its small size (7 mm × 7 mm), low power consumption and minimum external components requirement. It had a built-in 12-bit analog-to-digital converter and used the ZigBee wireless communication protocol. The ZigBee wireless protocol operates at the 2.4 GHz frequency which is in a worldwide license-free frequency band. The wireless SOC provided a real-time monitoring function by which the force value could be measured and displayed during the brace fitting session. The wireless function also allowed setting the parameters and downloading the data

292

E. Lou et al. / Medical Engineering & Physics 33 (2011) 290–294

3.00

prescribed tightness level (1.03 ± 0.17 N, range 0.78–1.25 N) was recorded as the individualized reference value. Each patient may have a different force partially due to the location of the curve, tolerance of the patient and flexibility of the body. The normalized force value (measured force magnitude/the individualized reference value) was used for analysis. The sample rate of the clinical trial was set to be 1 sample/min. Data was downloaded at the patients’ routine clinical visits. The time required to download 4 months of data was approximately 1 min.

2.00

2.5. Statistical analysis

1.00

A result is deemed to be statistically significant if it is unlikely to have occurred by chance. The significance level is usually denoted by the Greek symbol ˛. Popular levels of significance are 5% (0.05), 1% (0.01) and 0.1% (0.001). If a test of significance gives a p-value lower than the ˛-level, the null hypothesis of no difference between groups is rejected. Such results are referred to as ‘statistically significant’. Choosing a level of significance is an arbitrary task, but for this study, a level of 5% was chosen. In this paper, the two-tailed paired Student’s t-test was used to perform the statistical analysis.

7.00 Average (n = 3) (prior study) Average (n = 3) (aer Study)

Measured Force (N)

6.00 5.00 4.00

0.00 0

1

2

3

4

5

6

7

Applied Force (N) Fig. 2. The force sensor measurements before and after a 4-month study.

easily. A RTC was installed inside the system to save power by waking up the SOC from the low power mode only to acquire data and to provide the time stamp at data collection. A small coin sized backup battery (12 mm diameter × 2 mm thick, could last for 5 years) was connected to the RTC to ensure data was retained even if power failure occurred. A 4-Mb flash memory was used to store acquired data and it provided 524,288 force-time-stamped data. Power management circuitry was used to convert a single 1.2 V NiMH AAA-sized rechargeable battery to a 3 V regulated supply voltage so that the size and weight of the system could be minimized. For a sample rate of 1 sample/min, the monitor could store 1 year’s worth of data and the power could last for 130 days.

3. Results The first in-brace follow-up visit was approximately 2 months after the brace fitting session, by which time the patient should be wearing her brace full-time and the routine clinical follow-up visit was usually scheduled every 4 months. Up to now, subjects 1–6 have 4 months of brace wear data from months 3 to 6, but subjects 7–10 have 6 months of data collected from months 1 to 6. No data was lost from the systems. The average time of data collection was 4.4 ± 1.0 months. All the candidates were still in their brace treatment.

2.3. Interface program 3.1. Quality of brace usage The custom user interface was written in the Microsoft C# programming language to run on a PC with Windows XP and Windows 7 operating systems. The developed program simplified the set up and data analysis procedures. It allowed researchers to store patient demographics, set the sample rate, check the tightness level while the patient puts on the brace and perform data analysis. Daily, weekly and monthly summaries of the time wearing and wear tightness were generated which were useful to check for any changes in wearing patterns. This information was also helpful to determine if there were any relationships with the changes in the scoliosis curve severity. 2.4. Clinical trials The selection criteria of this study followed the brace study requirement guidelines provided by the SRS committee [17]. All participants had a diagnosis of idiopathic scoliosis, ages between 9 and 15 years and new to the brace treatment. Ten subjects (9F, 1M), age between 9.8 and 13.5 years, average 11.6 ± 1.3 years, who were prescribed a new TLSO and full-time brace wear (23 h/day) were recruited for this study. This study was approved by the local ethics board. All subjects signed their consent forms before participating. The Cobb angles of the major scoliosis curve prior to bracing, in-brace and at the first follow up out of brace were 31.8 ± 6.6◦ , 18.6 ± 7.4◦ and 32.8 ± 7.4◦ , respectively. All braces were made by the same orthotist, who has more than 10 years experience. The monitor was installed either after the first in-brace visit (subjects 1–6) or at the first brace fitting session (subjects 7–10). After the system was installed, loads were measured during standing while the orthotist adjusted the strap tightness. The force value at the

Because of the high sensitivity of the force sensor, the force value gave a clear distinction between the brace being worn and not worn. The mean reference force value was 1.04 ± 0.20 N (9.2 ± 1.6 kPa) with the range of 0.78–1.40 N. Unless the brace was worn very loosely, or a patient had a significant change of their body shape, it was unlikely to detect a no load value on the force sensor when the brace was actually worn. The force imposed by the brace varied from 0 to 5.1 N during the treatment period. The force magnitude distinguished when the brace was not worn. Fig. 3 shows the data from subject 9 during a 24 h period. This subject took the brace off 3 times during that day, around 10:00–11:00, 16:30–18:30 and 21:30–22:00 h. The reference force value of that subject was 0.78 N. For subjects S7–10, the average of the force value on the first and second months after the brace fitting session were 70.8 ± 5.6% and 67.5 ± 6.6%, respectively. The average of the force values of all subjects (S1–10) on the third to sixth months after the brace fitting session were 65.4 ± 9.3%, 63.2 ± 10.3%, 61.0 ± 10.3% and 58.0 ± 11.1%, respectively. There was a statistically significant decrease in force over the period after the brace fitting session (p = 0.03, month 1 vs month 6). For the S7–10 group, the decrease was from 70.8 ± 5.6% in the first month to 57.5 ± 13.6% in the sixth month, which was a 13% drop from the initial month (Fig. 4). Similarly for the S1–6 group, the decrease was from 65.3 ± 10.8% in the third month to 58.3 ± 10.4% in the sixth month; which was a 7% drop from the third month (Fig. 4). The mean force value was statically lower (p = 0.02) in the night time (10pm–7:59am) than during the day time (8–21:59 h). The average force values at night time and day time were 41.2 ± 10.6% and 79.5 ± 14.6%, respectively normalized to the reference load level. The day time force was almost twice

E. Lou et al. / Medical Engineering & Physics 33 (2011) 290–294

293

100 S7-S10

90

S1-S6

% of Brace Worn

80 70 60 50 40 30 20 10 0 1

2

3

4

5

6

Month Fig. 5. The quantity of brace wear during the first 6 months from subjects 1 to 10.

wear between the two patients groups in Fig. 5 was because one of the patients in group 1 used the brace very little (about 35%) and one of the patients in group 2 used the brace a lot (about 80%). Since the number of participants in this study was small, it was not enough power to perform the statistical analysis on these groups. Fig. 3. A daily brace wear pattern from subject 9.

3.3. Brace wearing patterns

the night time force. The average force value over the first 6 months was 62.9 ± 9.6%. 3.2. Quantity of brace usage

Force relative to the reference (%)

The quantity of brace usage was determined by dividing the measured hours worn (force value above zero) by the time specified by the clinician for the brace to be worn. The average quantity compliances from S7 to 10 for the first and second months after the brace fitting session were 51.2 ± 5.1% and 59.8 ± 8.5%, respectively. The average of the quantity compliance from all subjects (S1–10) on the third to sixth months after the brace fitting session were 60.9 ± 11.5%, 61.9 ± 13.6%, 62.4 ± 15.3%, and 62.3 ± 16.7% respectively. The overall average value was 60.0 ± 11.9% closely matching to previous studies [17–19,26,27]. There was an increase in wear time after the brace fitting session (the 1st and 2nd months), but the trend stabilized after month 3. The S7–10 group showed an increase from 51.2 ± 5.1% on the first month to 68.3 ± 11.5% on the third month, which was a 17% increase from the initial month (Fig. 5). Similarly, for the S1–6 group, the quantity compliance stabilized from 56.0 ± 10.4% on the third month to 55.3 ± 18.1% on the sixth month (Fig. 5). There was no correlation found on the wear time to the absolute force (r2 = 0.11). The difference in quantity of brace

90 S7-S10

80

S1-S6

70 60 50 40 30 20 10 0 1

2

3

4

5

6

Month Fig. 4. The quality of brace wear during the first 6 months from subjects 1 to 10.

Time-stamped force data revealed a personal wearing pattern during a 24 h period for each subject. The time wear pattern was consistent after the first in-brace follow-up clinic. Thus it may be possible to reliably predict the future brace wear pattern based after month 3. The difference was 1.4 ± 8.2% (ranged 1–21%). Only the non-compliance patients gave bigger errors on the prediction pattern. The qualities of the brace wear decreasing right after the brace fitting. The variation increases when subjects increase their wearing time during day time. 4. Discussion Even though bracing for treatment of scoliosis has been used for many years, there are still many unknowns on how a brace affects the spine. Brace treatment effectiveness is limited and outcomes need to be improved. The amount of time that a brace is worn is generally based on intuition. The most commonly recommended wear time is 23 h/day; this number is not based on any objective data. The measure of both quantity and quality of brace usage gives a good indication of whether a patient is properly following their treatment regimen. Measuring daily brace usage over a long period of time can provide information that how the brace has been used. When poor compliance is found, the surgeon, orthotist or a nurse practitioner can have a discussion with the patients to understand the reasons. In this study, no subject truly wore the brace full time; the maximum quantity of brace usage was 82% of the prescribed time over a single month. The average was 60.0 ± 11.9%. Comparing this study to a blinded study by Hasler et al. [33], there was no significant difference on the compliance (p = 0.04) (54 ± 22.3% vs 60.0 ± 11.9%). Also, Hasler et al. study only monitored the subjects for 2 weeks. This was similar to our previous study [31] that subjects who knew they were monitored did not increase the brace wear time compared to other studies. Technically it was possible to hide the system and sensor in a brace without being noticeable by the patients, however, this is not part of the ethics application in this study. The force value varied a lot during the day time and lower at night time. The force data profile over the logging period easily shows the time the brace is worn during the course of treatment.

294

E. Lou et al. / Medical Engineering & Physics 33 (2011) 290–294

This identifies how a brace is being used, especially when a patient is adapting to a new brace. Difference in growth velocity could cause differences in force measurements, but the visco-elastic properties of soft tissue would only have small effect on the measured forces. There are mainly two sources of elasticity in soft tissues: 1) it is due to the change of internal energy and 2) it is due to changes of entropy. Change of entropy occurs in tissues whenever changes of orientation or waviness of fibers during loading or unloading occur. Eventually, a steady force state should occur. However, the force value was consistently decreasing during the study period even though the wear time was stable. The reduction of the force over the treatment period suggests that the brace should be examined for proper fitting during every clinical visit. If the reduction is due to growth or the adjustment of the body shape related to the pressure pad, adding more padding may provide a gradual correction to the curve. Otherwise, the orthotist can re-educate the patient on brace wear, replace worn out components or alter the brace shape. A new brace may be required if excessive body shape or growth has occurred. By combining the quantity and quality of the brace usage and correlate these with the treatment outcomes may allow prediction of an effective “dosage” that is required for successful brace treatment. A preliminary study of prediction of brace treatment outcome based on quantity and quality of brace usage has been reported [35], which demonstrated that brace wear tightness might have a significant role on the treatment outcome. In this study, no relationship was found between the reference force value and the initial correction of the Cobb angle (r2 = 0.17). The decrease in force did not result in increase of wear time as the wear time was stabilized after month 3. Since only one sensor system was used in this study, we were not able to determine if a decrease of measured force could be caused by a redistribution of forces on the brace in time. Our group also developed a multi-sensor system [34] that could synchronize the load measurements at different spots during daily activities. Clinical trials on that system will be performed soon. A relationship was found between wear time with a change in Cobb angle (pre brace – the first follow up out of brace) (r = 0.75). The more time that brace was worn; it more likely to have a better treatment result. The current monitor is a low cost and low profile device. It can be adapted into any type of brace. The data can be downloaded wirelessly and is easily being uploaded to a data server to perform the analysis in real-time. More clinical trials will be conducted to try to answer the effectiveness of the brace treatment for scoliosis. Acknowledgements The work was supported by the Natural Sciences and Engineering Research Council of Canada and the Edmonton Orthopaedic Research Committee. Conflict of interest There are no conflicts of interest for the authors of this study. References [1] Lonstein JE, Carlson JM. The prediction of curve progression in untreated idiopathic scoliosis during growth. Journal of Bone and Joint Surgery 1984;66A:1061–71. [2] Lonstein JE. Adolescent idiopathic scoliosis. The Lancet 1994;344:1407–12. [3] James JIP. Idiopathic scoliosis. The prognoses, diagnosis, and operative indications related to curve patterns and the age at onset. Journal of Bone and Joint surgery 1954;36B:36–49. [4] Nachemson A. A long term follow-up study of non-treated scoliosis. Acta Orthopaedica Scandinavica 1968;39(4):446–76. [5] Weinstein SL, Zavala DC, Ponseti IV. Idiopathic scoliosis: long term followup and prognosis in untreated patients. Journal of Bone and Joint Surgery 1981;63A(5):702–12.

[6] Weinstein SL, Ponseti IV. Curve progression in idiopathic scoliosis. Journal of Bone and Joint Surgery 1983;65A:447–55. [7] Edgar MA, Mehta MH. Long-term follow-up of fused and unfused idiopathic scoliosis. Journal of Bone and Joint Surgery – British Volume 1988;70(5):712–6. [8] Dickson JH, Mirkovic S, Noble PC, Nalty T, Erwin RW. Results of operative treatment of idiopathic scoliosis in adults. Journal of Bone and Joint surgery – American Volume 1995;77(4):513–23. [9] VanRhijn LW, Plasmas CV. Changes in curve pattern after brace treatment for idiopathic scoliosis. Acta Orthopaedica Scandinavica 2002;73:277– 81. [10] Karol La. Effectiveness of bracing in male patients with idiopathic scoliosis. Spine 2001;26:2001–5. [11] Goldberg CJ, Moore DP, Fogarty EE, Dowling EE. The effect of brace treatment on the incidence of surgery. Spine 2001;26:42–7. [12] Dickson RA, Weinstein SL. Bracing (and screening) – yes or no? Journal of Bone and Joint Surgery 1999;81:193–8. [13] Price CT, Scott DS, Reed FR, Sproul JT, Riddick MF. Nighttime bracing for AIS with the Charleston bending brace: Long-term follow-up. Journal of Practical Orthodontics 1997;17:703–7. [14] Wiley JW, Thomson JD, Mitchell TM, Smith BG, Banta JV. Effectiveness of the Boston brace in treatment of large curves in AIS. Spine 2000;25:2326–32. [15] Willers U, Normelli H, Aaro S, Svensson O, Hedlund R. Long-term results of Boston brace treatment on vertebral rotation in Idiopathic Scoliosis. Spine 1993;18:432–5. [16] Goldberg CJ, Dowling FE, Hall JE, Emans JB. A statistical comparison between natural history of AIS and brace treatment in skeletally immature adolescent girls. Spine 1993;18:902–8. [17] Richards BS, Bernstein RM, D-Amato CR, Charles R, Thompson GH. Standardization of criteria for adolescent idiopathic scoliosis brace studies: SRS Committee on Bracing and Nonoperative Management. Spine 2005;30:2068–75. [18] Houghton RG, McInerney A, Tew T. Monitoring true brace compliance. In: Proceedings of the 21st meeting of the Scoliosis Research Society. 1986. p. 101. [19] Lavelle JR, Smith K, Platts R, Morley TR, Ransford AO, Edgar M. An assessment of brace compliance in adolescent idiopathic scoliosis using a new brace timer. Journal of Bone and Joint Surgery 1996;78B:162. [20] Havey R, Gavin T, Patwardhan A, Pawelczak S, Ibrahim K, Andersson G, et al. A reliable and accurate method for measuring orthosis wearing time. Spine 2002;27:211–4. [21] Nicholson G, Ferguson-Pell M, Smith K, Edgar M, Morley T. Development of instrumented spinal brace for measuring compliance and skin microclimate in the conservative treatment of adolescent idiopathic scoliosis. In: Proceeding of British Scoliosis Society, Silver Jubilee Meeting. 2002. p. 17. [22] Nicholson G, Ferguson-Pell M, Smith K, Edgar M, Morley T. The objective measurement of spinal orthosis use for the treatment of adolescent idiopathic scoliosis. Spine 2003;28:2243–51. [23] Helfenstein A, Lankes M, Ohlert K, Varoga D, Hahne H, Ulrich HW, et al. The objective determination of compliance in treatment of adolescent idiopathic scoliosis with spinal orthoses. Spine 2006;31:339–44. [24] Perie D, Aubin CE, Petit Y, Beausejour M, Dansereau J, Labelle H. Boston brace correction in idiopathic scoliosis: a biomechanical study. Spine 2003;28:1672–7. [25] Cochran GVB, Wangh TR. The external forces in correction of idiopathic scoliosis’. Journal of Bone and Joint Surgery 1969;51A:201. [26] Chase AP, Bader DL, Houghton GR. The biomechanical effectiveness of the Boston brace in the management of adolescent idiopathic scoliosis. Spine 1989;14:636–42. [27] Jiang H, Raso VJ, Hill DL, Durdle NG, Moreau M. Interface pressures in the Boston Brace treatment for scoliosis. In: Proceedings Int’l Symposium on 3-D Scoliotic Deformities. 1992. p. 395–9. [28] Cote B, Dansereau J, Labelle H, Reinecke S. A method for the measurement of the Boston brace biomechanical action on scoliotic deformities. In: Proceedings Int’l Symposium on 3-D Scoliotic Deformities. 1992. [29] Wong MS, Mak AF, Luk KD, Evans JH, Brown B. Effectiveness and biomechanics of spinal orthoses in the treatment of adolescent idiopathic scoliosis (AIS). Prosthetics and Orthotics International 2000;24:148–62. [30] Lou E, Raso VJ, Hill DL, Durdle NG, Mahood JK, Moreau MJ. The daily force pattern of spinal orthoses in subject with adolescent idiopathic scoliosis. Prosthetics and Orthotics International Journal 2002;26:58–63. [31] Lou E, Raso JV, Hill DL, Mahood JK, Moreau MJ. Correlation between quantity and quality of orthosis wear and treatment outcomes in adolescent idiopathic scoliosis. Prosthetic and Orthotic International 2004;28:49–54. [32] Bulthuis GJ, Veldhuizen AG, Nijenbanning G. Clinical effect of continuous corrective force delivery in the non-operative treatment of idiopathic scoliosis: a prospective cohort study of the triac-brace. European Spine Journal 2008;17:231–9. [33] Hasler CC, Wietlisbach S, Büchler P. Objective compliance of adolescent girls with idiopathic scoliosis in a dynamic SpineCor brace. Journal of Children’s Orthopaedics 2010;4(3(June)):211–8. [34] Lou E, Hill DL, Raso VJ. A wireless sensor network system to determine biomechanics of spinal orthotic during daily living. Journal of Medical and Biological Engineering and Computing 2010;48(3):235–43. [35] Lou E, Hill DL, Parent E, Raso VJ, Moreau MJ, Mahood JK, et al. Preliminary validation of curve progression model. research into spinal deformities 6 series studies in health technology and informatics, vol. 140. Oxford: IOS Press; 2008, 294–298.