Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Coronal Plane Trunk Shifts and Decompensational Perspectives in a New Design of an Asymmetrical TLSO Module

Keith M. Smith, CO, LO

ABSTRACT

Currently, two styles of thoracolumbosacral orthoses (TLSOs) are used in the conservative management of adolescent idiopathic scoliosis: molded or nonmolded (modules fabricated from measurements). The molded type allows the practitioner to shift coronal deviations in asymmetrical molds, and it is this shift principle that is being applied to a new design of a module made to measurements. The author presents the use of commercially available modules made with this shift applied to the trunk to decrease decompensation and trunk shift. The prototype was fabricated with CAD/CAM technology at Atlantic Rim Brace Manufacturing (Nashua, NH); a series of measurements is fed into the computer to create a mold of the patient. The mold is modified on screen to translate or shift the trunk on the pelvis in an overcorrected or asymmetrical module. Radiographs were taken to determine the efficiency of the shift on coronal plane deviations and overall spine balance. This article explains the importance of overall spine balance in the conservative management of adolescent idiopathic scoliosis and emphasizes the need to pay particular attention to coronal plane deviations, such as decompensation and trunk shift.

Current consensus among the orthopedic community and orthotists is to first balance the spine and then achieve Cobb angle corrections. In other words, a well-compensated spine should be achieved before any attempt is made to correct the magnitude of curves. For example, a curve that has remarkable in-orthosis correction to the Cobb angle but has a marked or increased decompensated spine or trunk shift should not be accepted in an upright scoliosis thoracolumbosacral orthoses (TLSO). Figure 1 shows a well-corrected Cobb angle measurement to the lumbar curve but an unbalanced spine. Without paying attention to the overall balance of the spine, this would be accepted as excellent in-orthosis correction.

The upright TLSOs are examined in this study. Nocturnal orthoses are worn in the supine position only and decompensation is not an issue because of the absence of gravity. Current treatment in modules such as the Boston (Avon, MA) and Atlantic Rim Brace (Nashua, NH) style TLSOs suggests using a trochanteric pad and ipsilateral side axillary extension to induce a medially directed force from a tilting of the orthosis in an attempt to decrease decompensation. Winter and Carlson ¹ emphasized, if there is a need to shift the upper spine, the use of a trochanteric extension and pad it until the orthosis tilts at the superior lateral axillary extension to push the spine over. Bassett and Bunnell ² studied not only decompensation, but also the concept of trunk shift using the method of Floman et al ³ , in which a horizontal line is drawn halfway between the seventh cervical and the first sacral vertebrae from one side of the rib cage to the other. A perpendicular line is then drawn at the horizontal line's midpoint. The distance between this perpendicular line and the center of the first sacral vertebra represents the lateral trunk shift. Watts ⁴ emphasized the importance of the mass of the body superior to the pelvis being centered over the pelvis. He defined decompensation as the horizontal distance between two vertical lines, one bisecting the sacrum and the other bisecting the highest vertebra seen on a standing posteroanterior roentgenogram. He also noted that even some curves exhibit the seventh cervical vertebra being centered over the sacrum, but the mass of the body still being out of balance or shifted. Bassett and Bunnell² were able to show in a molded TLSO on 80 patients that lateral trunk shift was improved in 58 percent of patients with thoracic curves, 65 percent with thoracolumbar curves, and 88 percent of patients with double major curves. In these molded TLSOs a force is applied to shift the trunk back in balance with the pelvis before curve correction forces are applied. This molded Wilmington style (originally developed at Alfred I Dupont Institute, Wilmington, DE) implements bilateral trocanteric extensions to serve as foundational support for the endpoint control of the orthosis. A force is applied to the trunk to create a shift of the trunk back into alignment in the coronal plane.

The purpose of this study is to adapt this trunk shift principle to symmetrical modules commercially in use. Because of a lack of a Risser table, a lack of experience, ease of use, or physician preference, symmetrical modules are more prevalent than Risser table-molded Wilmington TLSOs. The author currently uses Risser table-molded Wilmington TLSOs and modules fabricated at Atlantic Rim Brace Manufacturing. The current article presents the implementation of the trunk shift principle to these symmetrical modules to create an asymmetrical module TLSO used to address spine balance in the coronal plane.

It is important at this point to address the concept of the central sacral line (CSL) in relation to the curves. The CSL is found by bisecting the sacrum and measuring to the side of the radiograph from this point. This measurement is then used to measure from the side of the radiograph to a point at the height of around T1. These two points are connected with a line, giving the central sacral reference line. The question the orthotist needs to ask when trying to gain a well-balanced spine is where each curve is in respect to the CSL. Is one curve opposite the CSL from the other or are both curves on the same side of the CSL? Are both curves structural or is one structural and the other nonstructural? Using the Lenke classification system, ⁵ type 5 lumbar curves with modifier C are structural lumbar curves in which the thoracic curve is nonstructural. On the other hand, Lumbar type 6 modifier C has both structural lumbar and thoracic components. Whether the thoracic component is structural determines whether it is treated orthotically. Lenke et al ⁵ identify structural minor curves in thoracic regions as those that have a minimum of 25 degrees coronal curve and/or thoracolumbar kyphosis (T10-L2) of at least +20 degrees. Only structural minor curves and the major curves are included in spinal arthrodesis and thereby orthotic management as well. Modifier C refers to the CSL falling medial to the lateral aspect of the lumbar apical vertebral body. The curves being presented by the author are of the lumbar type, with the trunk above being shifted out of balance to the same side of the CSL as the lumbar curve.

It is in the trunk shift presented by Bassett and Bunnell² that the shifting principle is so important, but so often overlooked. Practitioners focus on Cobb angle measurements and decompensational issues without regard for trunk shifts presented. In a molded Wilmington TLSO, trunk shifts can easily be addressed by the use of a cast mold taken on a Risser table in which a patient is stabilized at the pelvis and then the trunk is shifted to the opposite direction on the pelvis. Therefore, an asymmetrical orthosis is created to achieve balance to the spine (Figure 2) . At this point, the application of forces is used to gain in-orthosis correction to the respective curves. Traditional TLSO modules made to measurements have limited this ability to gain a good correction to trunk shift because of their symmetrical nature. As mentioned, decompensation and trunk shift typically have been treated with added trochanter and axillary pushes. Figure 3 depicts the proposed action of the shift in an asymmetrical orthosis from a scoliosis pattern of two curves, both being on the same side of the CSL because of the trunk shift created by the lumbar curve. Figures 3B and C show the possibilities that may exist after applying the shifted orthosis. First, the top curve will remain but will shift back to the opposite side of the CSL, which would then be treated as a Lenke type 6 lumbar curve modifier C. Or, as in Figure 3C , the top curve will correct itself back on the CSL and be considered only a compensatory curve. The spine in Figure 3B then would be treated with two pads as long as both curves were above 25 degree Cobb angle measurements, whereas that in Figure 3C would be treated with one pad as a lumbar only curve. The important concept here is in Figure 3C . Without attention to trunk shifts, a practitioner would be left with the option of erroneously treating the upper curve with a pad or push that could lead to increased decompensation or trunk shift. The resultant action would then lead the orthotist to add a trocanteric pad in an attempt to decrease the decompensation. It is proposed that a better approach would have been to first work on spinal balance and then pad placement. These illustrations depict the goal of achieving overall spine balance with curve correction of the Cobb angles and corrected trunk shift.

FABRICATION

The patient is measured in the typical manner using a series of medial/lateral, anterior/posterior, and circumference measurements. The author prefers to have the TLSO made from a CAD/CAM mold to ensure the ability to take an intimate fitting module and apply the shift to it. Atlantic Rim Brace Manufacturing uses the CAD/CAM system, which allows the user to see the mold before it is produced. Figure 4 shows the computer-generated views, and Figure 5 shows the molds created from the CAD view. As is evident in these figures, the trunk is shifted to the right in relation to the pelvis in the coronal plane. This is the proposed treatment for the coronal imbalance of a typical lumbar primary curve with marked decompensation or lateral trunk shift. An asymmetrical module is then manufactured to which the pads can be added for Cobb angle corrections. It is important to note here that bilateral trocanteric extensions are used to give a firm grip on the pelvis to ensure good endpoint control. Remember the orthosis is not using the typical trocanteric extension and pad to tilt the brace or rotate it within the coronal plane, but rather to shift the trunk back into balance first. The author reserves the trochanter pads for fine-tuning of decompensation. The axillary extension is left as superior on the trunk as possible on the ipsilateral side to ensure good leverage on the shifting trunk. Anterior and posterior trim lines, left unchanged from the typical modules in use today, are inferior to the xiphoid and most inferior angle of the scapulae, as well as superior to the symphysis pubis anteriorly and 1 cm above the seating surface posteriorly.

Figure 6 shows the finished orthosis manufactured from the shifted molds. A closer look at the molds and the orthosis shows that it is as if we bisected the mold across the superior edge of the iliac crest and lifted the superior portion and translated it over on the pelvis. Figure 7 shows this orthosis applied clinically. It is evident in the figure that the mold shifts the trunk on the pelvis into better alignment. It should be noted that it is often necessary to flare the superior lateral edge on the shifting side to prevent edge pressure. This pressure is another reason the author prefers to use the modules fabricated from the CAD/CAM. The increased measurements required for the Atlantic Rim modules allow for a more anatomical fit. Traditional modules are fabricated without medial/lateral and circumference measurements above the xiphoid process. In this module, the medial/lateral and circumference measurement at the axilla region is taken into account. The anatomy of the latissimus dorsi muscles often increases the medial/lateral slightly in the axilla region, which is then taken into account in this fabrication process.

INITIAL PATIENTS TREATED

Tables 1 to 6 show pre- and postdistances of the center of each vertebral body from the CSL in six patients on posteroanterior radiographs. Lines were drawn from the center of each of the bodies to meet the CSL, and these were measured and recorded. The objective with the first two patients' evaluations ( Table 1 and Table 2 ) was to determine the efficacy of trunk shift correction in a TLSO to the overall spine balance.

The first patient had what initially appears to be a second curve in the thoracic region. However, if we look at the radiograph in Figure 8 and Table 1 , it becomes evident that the entire spine is to the left of the CSL, with the lumbar curve causing a marked trunk shift and decompensated spine. This patient was treated with the shift of the trunk and then a trocanteric and ipsilateral axillary extension, as seen in the second radiograph. Without curve correction forces applied, it is evident that the trunk shift alone improved drastically the spinal balance in relation to the CSL and at the same time decreased Cobb angle measurements. If attention is not given to spinal alignment in relation to the CSL, this is the type of curve that could be erroneously treated with forces of correction to both curves. As presented, the thoracic curve is only compensatory, so any applied forces to the thoracic curve would serve only to increase decompensation because of its direction of force.

Table 2 and Figure 9 represent a patient with a small curve who is being treated for spinal imbalance solely in the coronal plane. The original Cobb angle measurement was a 19-degree left thoracolumbar curve with a marked trunk shift and decompensation, as is evident in the radiographs. The patient was treated with a shift of the trunk on the pelvis to the right. The shift of the mold decreased decompensation as well as trunk shift.

Trunk shift in Patient 3 ( Table 3 and Figure 10 ) was attempted because of a lack of a favorable response from axillary and trocanteric forces applied to the spine in her symmetrical module on the coronal plane. The symmetrical nature of the TLSO was the limiting factor. Therefore, with the author's experience with the molded Wilmington principles, a new module was created with the trunk shift and then force pads applied. Table 3 shows a marked decrease in trunk shift of the lumbar curve back to the midline, resulting in a decrease of Cobb angles from 19 and 30 degrees to 13 and 12 degrees, respectively. Figure 10 shows radiographs of this shift and better balance to the overall spine. The shift was used to correct the trunk shift, and a trocanteric pad was still used to help with the decompensation.

Patient 4 can be seen in Table 4 and Figure 11 . This patient's curve type presented with marked trunk shift and decompensation to the left. This curve is a Lenke type 6 with a thoracic component, so along with the shift principle being applied, force pads were applied for both the thoracic and lumbar components. This particular patient is interesting to observe because her older sister was treated in the past by the author with a traditional module. This older sister had marked trunk shift as well and was treated with axillary and trocanteric forces. Postorthosis radiographs will make a nice comparison in the future of the two styles. However, this patient also needed a trochanter push to help decrease decompensation.

Patient 5 presented with extreme decompensation and trunk shift to the left. Figure 1 shows this patient with a lumbar pad and no shift to the trunk in-orthosis. Figure 12 depicts the same patient with the lumbar pad as well as the shift applied. It is evident from Table 5 and Figure 12 that there was great overall spine balance and Cobb angle correction. However, it is important to note that the increased spine balance in Figure 12 resulted in slightly less Cobb angle correction than in Figure 1 . However, our goal of achieving spine balance is primary, and Figure 12 depicts a better alignment. Cosmetically, the patient has gone from a marked lean to the left and extremely asymmetric stature to a better aligned trunk over the pelvis. Clinically, the spine balance is greatly improved, and practically the appearance to the patient is improved as well.

Patient 6 is represented in Figure 13 and Table 6 . These again show a patient with a type 5 lumbar curve modifier C and extreme decompensation and trunk shift to the left. The patient was fitted with a heel lift to level the pelvis as well as the shifted module to improve overall spine balance. As is evident in the radiograph, a lumbar-only pad was applied and the end result was a well-compensated spine and balanced alignment of the trunk over the pelvis.

As is evident in Tables 1 to 6, a marked decrease in trunk shift and decompensation was achieved with the use of the shifting principle. Figure 8 and Figure 9 depict the resultant action of decreased Cobb angle measurement from simply decreasing the trunk shift. These two patients had no force pads added directly to the curves. The tables show a strong correlation between the shifting principle, decreasing the distance from the center of each vertebral body to the CSL. It also is interesting to note that levels as high as T1 showed marked decrease in shift away from the CSL. In the six patients treated, the highest coronal plane correction occurred in the high thoracic region and the thoracolumbar region. Figures 8 to 13 suggest that the compensatory curve gained correction to the Cobb angle as well, thereby making the segment above and below this compensatory curve appear to have more correction to the distance from the CSL. Tables 1 to 6 confirm this observation numerically. However, the important concept to see is that the trunk shift in all cases increased the overall spine balance by decreasing the distance from the center of each vertebral body to the CSL. Patients 2 and 6 had remarkable correction in the coronal plane, bringing the extremely shifted trunk not only back to the midline, but also past it to the contralateral side. The author paid particular attention to Patient 6 to ensure that there was not a second structural thoracic curve. This upper curve still is treated as only compensatory. Our original goal with these six patients was to decrease trunk shift and decompensation, thereby giving better spinal balance. Collectively, by shifting the module, preorthosis and in-orthosis measurements to the trunk shift, decompensation and Cobb angles decreased.

DISCUSSION

The concept of shifting the trunk on the pelvis to address trunk imbalance and decompensation is not new. It has been done for years with the Wilmington concept. However, recent CAD/CAM technology has allowed the practitioner to use a module made from measurements in an asymmetrical shape patterning the molded style. Because the modules are symmetrical, this ability to increase correction to trunk shift has been overlooked. Radiographs allow the practitioner to implement as well as monitor this principle. As with the molded Wilmington style, a posteroanterior radiograph in-orthosis is required to accept in-orthosis spine balance and correction. It also provides the practitioner the proof and peace of mind that the shift is indeed improving the spinal balance. The concept presented here allows the practitioner to follow the progress of the patient's curve. The shifting principle is another tool in the arsenal of the conservative management of idiopathic scoliosis to ensure spinal balance. It is important to note that the shifting principle presented here is proposed treatment for spinal imbalance. The author reserves its use for curves in which the entire spine is on the same side of the CSL. When the thoracic and lumbar components are on opposite sides of the CSL, the shifting principle would improve one curve while having an adverse effect on the opposite. Keep in mind the shifting principle is used solely for the purpose of moving the shifted trunk closer to the midline. As opposed to the tilting of the orthosis created by axillary and trocanteric extensions, this design holds the pelvis while the trunk is translated back toward the midline. The author does not intend to disregard the trochanter/axillary extension relationship, but rather to expand upon the concept of coronal plane spinal balance. The author reserves the trochanter pad for further improved decompensation.

The tables presented along with the radiographs show a strong link to the shifting principle, clinically as well as cosmetically. The measurements are used to represent the general difference between the translation of the trunk to the midline. Error does exist because radiographs are blurry in some levels, leaving the center point to judgment. The author uses the numbers to illustrate the notion of the change. Future study will compare patients who have been treated in the past without the shifting principle to those who have and will be treated with the shift. The curve pattern studied will be limited to Lenke classification type 5 curves with lumbar modifier C. The author proposes to look long-term at the results in-orthosis and postorthosis of trunk shifts specifically on coronal plane motion.

In 1986, Bassett and Bunnell² reported on the first look at orthotic management on lateral trunk shift and its efficacy. As mentioned earlier, favorable results were found, and the authors suggested using TLSOs to improve existing lateral trunk shift. They used the Wilmington, in which trunk shifts could be implemented. Rudicel and Renshaw⁶ studied the Milwaukee brace and its effect on decompensation. They found that despite being compliant, there was no predictable improvement. Bassett and Bunnell² recognized that there is very little research on the topic of spinal balance in relation to orthotic management. In 1998, Raso et al⁷ reported that little research has been done on trunk distortion; they proposed that the best treatment would improve both spine alignment and body deformity, not simply rely on the single-plane measurements of the Cobb angle. Research is focused on Cobb angle measurements. With this new module being commercially available, the orthotist is better able to address coronal plane deviations. New research can focus on not simply in-orthosis Cobb angle corrections but also overall spine balance and these coronal plane deviations.

ACKNOWLEDGMENTS

The author thanks Dr. Bassett, Dr. Otis, and Dr. Meyers of Mid-County Orthopedics in St. Louis, Missouri, for radiographic assistance, introduction of patients into the study, and informative expertise. The author also thanks Atlantic Rim Brace Manufacturing for assistance in developing the prototype.

Correspondence to: Keith M. Smith, CO, LO, 777 South New Ballas Road, 116W, St. Louis, MO 63141; e-mail: Keithorth@aol.com.

KEITH M. SMITH, CO, LO, is affiliated with Orthotic & Prosthetic Lab, Inc., St. Louis, Missouri.

References:

1. Winter RB, Carlson MC. Modern orthotics for spinal deformities. Clin Orthop 1977;126:74-86.
2. Bassett GS, Bunnell WP. Effect of a thoracolumbar orthosis on lateral trunk shift in idiopathic scoliosis. J Pediatr Orthop 1986;6:182-185.
3. Floman Y, Denny N, Riseborough EJ, et al. Osteotomy of fusion mass in scoliosis. J Bone Joint Surg [Am] 1982;64:1307-1316.
4. Watts HG. Bracing in spinal deformities. Orthop Clin North Am 1979;10:769-785.
5. Lenke LG, Betz RR, Harms J, et al. Adolescent idiopathic scoliosis: a new classification system to determine extent of spinal arthrodesis. J Bone Joint Surg [Am] 2001;83:1169-1181.
6. Rudicel S, Renshaw TS. The effect of the Milwaukee brace on spinal decompensation in idiopathic scoliosis. Spine 1983;8:385-387.
7. Raso JV, Lou E, Hill DL, et al. Trunk distortion in adolescent idiopathic scoliosis. J Pediatr Orthop 1998;18:222-226.