Secondary Logo

Journal Logo

Focus Paper

A Posterior Tether for Fusionless Modulation of Sagittal Plane Growth in a Sheep Model

Lowe, Thomas G. MD*; Wilson, Lucas BSME*; Chien, Jui-Teng MD; Line, Breton G. BSc*; Klopp, Lisa DVM, MS; Wheeler, Donna PhD§; Molz, Fred PhD

Author Information
doi: 10.1097/01.brs.0000175175.41471.d4
  • Free

Scheuermann’s disease has been reported to affect 1% to 8% of the U.S. population.1,2 It consists of both a radiographic and clinical structural hyperkyphosis of the thoracic or thoracolumbar spine, which develops during adolescence in both males and females. The radiographic criteria defined by Sorenson include a hyperkyphotic deformity with three or more adjacent vertebrae each with wedging of greater than 5°.3 Clinically, the deformity has a sharp, angular appearance, especially when viewed from the side when performing the Adam’s forward bend test. On hyperextension, it does not fully correct and is often associated with tight hamstring and pectoral muscles. Although one third of the patients have an associated mild scoliosis, the major deformity is always in the sagittal plane.

Brace treatment for Scheuermann’s disease has been shown to be effective in skeletally immature patients with reversal of vertebral body wedging in some patients in accordance with the Heuter-Volkmann principle. Unfortunately, compliance with brace wear has been a major issue in the United States, especially with males.1,2,4,5

For neglected cases or when brace treatment has been ineffective, the only alternative treatment for high magnitude deformities (>80°) is a long segment spinal fusion, which often entails both an anterior as well as a posterior approach. Indications for spinal fusion include loss of sagittal balance, functional limitations, pain, and cosmesis.

Modulation of bone growth with the use of epiphyseal staples has been successful for the treatment of tibia vara, leg length inequality, and more recently adolescent idiopathic scoliosis (AIS).6–8 Similarly, scoliosis has been created in skeletally immature animals using an asymmetric posterior as well as anterior tether.9,10 The success of fusionless treatment of AIS through a minimally invasive surgical approach has been variable, largely because of the three-dimensional nature of the deformity.11 However, fusionless correction of kyphosis may be more applicable since it is a single plane deformity.

Skeletally immature sheep appear to be a good model for fusionless modulation of growth because of their moderately kyphotic spine and short, rapid growth period. Spinal growth in sheep is near completion within 2 years. Therefore, the purpose of this study was to determine whether fusionless modulation of sagittal plane growth in a skeletally immature sheep model could be successfully performed using a posterior tether system.

Materials and Methods

Nine, 4-month-old sheep underwent a posterior spinal tether procedure, and five control animals of the same age were followed without treatment but with identical clinical, radiographic, biomechanical, and histologic follow-ups.


The procedure for the placement of the posterior tether consisted first of insertion of two bilateral pedicle screws at adjacent levels, proximally at the thoracolumbar junction and distally at the lower lumbar spine. This region proved to be the most kyphotic as well as the most flexible portion of the spine. The pedicle screws served as proximal and distal anchors for the tether. A polyethylene (UHMWPE) cord was then passed subcutaneously between the sets of pedicle screws on each side of the spine and tensioned to 20 lbs. The polyethylene cords were fixed within each pedicle screw with a sharp pointed set screw, which securely held the cords within the pedicle screws. Figure 1 shows the two incisions, at the cranial and caudal ends, where the anchor screws were placed and the tether after the setscrews have been tightened. One animal was lost to postoperative infection. All animals ambulated freely after surgery.

Figure 1
Figure 1:
Tethering procedure. Note the openings at the cranial and caudal ends with the pedicle screws anchoring the tethers.

Anteroposterior and lateral radiographs were obtained before surgery, immediately after surgery, and at 1, 2, 4, 6, 9, 12, and 13 months after surgery for all animals. The tether was released percutaneously, with a small portable cautery tool, in four of the test animals at 6 months (6MR group) through a small dorsal midline incision. In the other four tethered animals, the tether was released at 12 months (12MR group). Because of time constraints, all animals were sacrificed 2 weeks following the release of the tether in the 12MR group. CT scans, biomechanical analysis, and histology were all performed following sacrifice.

Radiographic Analysis.

Anteroposterior and lateral radiographs of the spine were obtained with the sheep anesthetized in dorsal recumbency and lateral recumbency, respectively. Three sagittal radiographic parameters were measured serially. Total kyphosis of the instrumented levels was obtained by measuring the angle subtended by the intersection of the perpendicular line from the superior endplate of the inner cranial instrumented level and the inferior endplate of the inner caudal instrumented level. Vertebral body wedging was measured from the angle subtended between the intersection of a line along the superior endplate with a line along the inferior endplate of each vertebra. Intervertebral disc wedging was obtained by measuring the angle subtended by lines along the inferior endplate of the suprajacent vertebra and the superior endplate of the infrajacent vertebra.

Biomechanical Analysis.

Immediately following sacrifice, the spines were removed of most soft tissue while leaving the intervertebral discs, surrounding ligaments, pedicle screws, and tethers. The spines were then potted in Dynacast (Kindt-Collins Co., LLC., Cleveland, OH), ensuring that proper alignment was achieved. For motion analysis, the spines were then instrumented with three 10-mm retro-reflective markers per vertebra. The biomechanical testing setup is demonstrated in Figure 2. The biomechanical analysis was performed by applying a moment, in both flexion and extension, to the spines while measuring the motion of each vertebra during the process. A moment of ±4 N-m was applied to the spines using a MTS 319.10 Axial/Torsional system (MTS Systems Corp., Eden Prairie, MN), whereas motion was measured using a Peak Performance motion analysis system (Peak Performance Technologies, Inc., Englewood, CO). The moment was applied to the spine with a 4-point bending mechanism. As each spine underwent flexion and extension, two video cameras simultaneously recorded their motion. The markers were then digitized and tracked using Peak Motus software (Peak Performance Technologies, Inc.). Three-dimensional coordinate data were then converted into angular motion using a program written by two of the authors (LW and BGL). Global motion (relative motion between the cranial and caudal most vertebrae), cranial motion (motion between the two cranial most vertebrae), apical motion (motion between the two apical most vertebrae), and caudal motion (motion between the two caudal most vertebrae) were all calculated.

Figure 2
Figure 2:
Four-point bending of a specimen with markers for motion analysis.

The global bending stiffness was calculated by determining the slopes of the linear regions on the moment versus angle graph, using global angular deflection measurements. The linear regions were located just before the minimum and maximum moments (there were two stiffness calculations per graph, one for flexion and one for extension). The global stiffness measurements were then standardized by multiplying the global stiffness by the number of levels tested in each group.

A statistical analysis of the data was performed using two-way ANOVA, performed by SAS statistical software version 9.1 (SAS Institute, Inc., Cary, NC) to evaluate the effects of treatment and time.


Instrumented Kyphosis

Mean initial kyphosis for each of the groups was similar with a range from 28° to 35°, which was followed by an initial sudden decrease in kyphosis in both the 6MR (2.8° ± 4.8°) and 12MR (10 ± 9.9°) groups. By 6 months, both groups had a reversal of kyphosis to mild lordosis (−3.4° ± 14.7°, 6MR and −1.2° ± 8.7°, 12MR). Following release of the tether in the 6MR group, there was a gradual rebounding, and by 12 months the overall mean kyphosis was about 60% of its mean preoperative value (18.6 ± 7.4° vs. 31.5° ± 6.8°, P = 0.092). Similarly, in the 12MR group there was a progressive reversal of kyphosis to lordosis occurring until the release of the tether (−3.3° ± 5.7°). At this time, there was a slight return of kyphosis with a mean of 5.9° ± 7.4° or about 17% of the mean preoperative value in the 12MR group (P < 0.0001). The control group, on the other hand, maintained a mean kyphosis of about 32° throughout the 13-month period. These changes can be seen radiographically in Figure 3 and graphically in Figure 4.

Figure 3
Figure 3:
Preoperative (A), immediate post tether insertion (B) 6-month prerelease (C), and final follow-up (D) radiographs of a sheep from the 6MR group.
Figure 4
Figure 4:
Graph of mean kyphosis in the 6MR, 12MR, and control groups at various measurement times.

Intervertebral Disc Angulation

There was no significant difference between the mean preoperative intervertebral disc angulation and the final follow-up measurements in the 6MR group (−1.7° ± 1.6° vs. −3.3° ± 3.4°, P = 0.193). There was, however, a significant difference between the mean preoperative measurement and final follow-up measurement of the 12MR group (0.02° ± 1.1° vs. −5.7° ± 2.3°, P < 0.0001). The control group had a mean initial disc angulation measurement of −1.8° ± 1.8°, and at 12 months was −2.9° ± 2.2°.

Vertebral Body Wedging

Significant differences in vertebral body wedging were not seen in either group until 6 months after surgery. In the 6MR group, the 12-month mean vertebral body wedging was 5.4° ± 1.6°, which was significantly less than the mean of the control (9.2° ± 1.9°, P < 0.002) and was significantly less than its preoperative value (8.2° ± 1.9°, P = 0.005). In the 12MR group, the 12-month mean vertebral body wedging was 6.6° ± 2.2°, which was significantly less than the control (P < 0.002) but not significantly less than its preoperative value, which was 8.0° ± 2.8° (P > 0.19). Four months after surgery, the mean vertebral body wedging of the control became significantly greater than its preoperative value. This trend continued through the final follow-up, suggesting that normal sheep vertebrae, in the lumbar region, become more wedge shaped with growth. These changes are shown in Figure 5.

Figure 5
Figure 5:
Graph of vertebral body wedging of instrumented levels at the various measurement times.

Imaging Studies

CT scans were performed on both instrumented and noninstrumented regions of the spine on all animals at 13 months. Although no degenerative changes were noted in the facets of either instrumented or noninstrumented levels, heterotopic ossification (HO) was noted in both the 6MR and 12MR groups within the instrumented levels. Figure 6 consists of representative CT scans of the control, 6MR, and 12MR groups (from left to right).

Figure 6
Figure 6:
CT scans of vertebrae from the control group (A), the 6MR group (B), and the 12MR group (C). Note the HO with ankylosis of the facets and incorporation of polyethylene cords within heterotopic bone mass in the CT scans from the 6MR and 12MR groups.

Biomechanical Evaluation

The mean stiffness of the 6MR group was significantly greater than the control group both in flexion (5.54 ± 2.38 Nm/° vs. 1.72 ± 0.29 Nm/°, P = 0.022) and extension (3.82 ± 1.60 Nm/° vs. 1.59 ± 0.22 Nm/°, P = 0.036). Likewise, the stiffness of the 12MR group was significantly greater than the control group in extension (5.36 ± 3.00 Nm/° vs. 1.59 ± 0.22 Nm/°, P = 0.046) but not in flexion (7.22 ± 5.96 Nm/° vs. 1.71 ± 0.29 Nm/°, P = 0.115). The stiffness of the 12MR group was greater than that of the 6MR group, in both flexion and extension, but not significantly different.

Associated with the stiffness, the mean global motion, in the control group, was significantly greater in both flexion (30.2° ± 4.0°) and extension (−36.1° ± 2.1°) than both the 6MR (13.6° ± 3.7° and −19.1° ± 8.3°, P < 0.01 each) and 12MR (13.1° ± 8.2° and −14.1° ± 9.9°, P < 0.01 each) groups. Comparisons were also made in the cranial, apical, and caudal segments between each group. There were no significant differences between the treatment groups and the control group in the cranial segment. There was, however, a significant difference between the 12MR group and the control group means within the apical segment in extension (−1.6° ± 1.1° vs. −6.2° ± 0.9°, P < 0.001). Within the caudal segment, the mean extension in the control group (4.6° ± 0.8°) was significantly greater than in the 6MR group (2.7° ± 1.0°) (P = 0.038). At no location were the 6MR and 12MR groups found to be significantly different; however, the general trend showed less motion in the 12MR group compared with the 6MR group.

Histology of Intervertebral Discs and Facet Joints

Tethered animals, both 6 and 12 month, exhibited normal growth plate and endplate cellular morphology. However, some thinning in the posterior region of the intervertebral discs was noted in both the 6MR and 12MR groups compared with the controls. In addition, intervertebral discs outside the tethered region also showed thinning in the posterior region. No inflammatory reaction was noted in either tether group, and at 13 months the discs in all animals appeared healthy and viable. No disc abnormalities were noted in the control animals.

Many of the facet joints within the tethered region were fused. However, fusion was more prevalent in the 12MR group than the 6MR group. Similar changes were also noted outside the tethered region. In addition, some cartilage fibrillation within the facet joint was noted both within and outside the tethered region. These changes were not noted in the controls.

Regional lymph nodes were studied from all animals. Normal lymphatic tissue with no inflammatory response was noted in all three groups.


Fusionless modulation of sagittal growth appears to be an achievable goal in skeletally immature sheep using a posterior tether. Preliminary radiographic data showed statistical differences in global sagittal plane angulation as well as vertebral body wedging between tethered and control animals following tether release at 13 months. The effects of the tethering were most apparent in the 12MR group. However, despite the desire for a 6-month post-release follow-up period, follow-ups for the 12MR group were obtained less than 1 month following tether release because of time constraints. Therefore, it is unknown whether an increase in kyphosis would have occurred if followed for a longer time after tether release, as in the 6MR group.

The CT data failed to demonstrate any obvious degenerative changes within the instrumented or adjacent to noninstrumented facets. It did, however, demonstrate migration of the polyethylene tethers into the paraspinous muscle, which presumably was a major factor in the development of HO and facet fusion, as well as increased stiffness of the tethered spines when tested biomechanically. Future studies will include modification of the tether to minimize migration into the paraspinous muscle.

The histologic data demonstrated some thinning in the posterior region of the discs both outside and within the tethered segments, but no inflammatory response, and normal endplate histology was noted. The changes within the facets are a concern but were found both inside and outside of the tethered segments and appear to be related to heterotopic bone formation.

Nonoperative treatment for Scheuermann’s disease continues to be a problem either because of noncompliance or failure of brace treatment to control the deformity. Modulation of growth using a posterior tether system appears to be a viable goal based on these preliminary data. HO was seen in several of the sheep both on CT scans and histologic evaluation, even though there was no periosteal or muscle stripping from bone, the tether was found to have penetrated the paraspinous fascia. It would have been interesting to consider the use of oral nonsteroidal anti-inflammatory drugs in the tethered sheep as a means of reducing the likelihood of HO. To the authors’ knowledge, there have not been any reported series of adolescent humans that have developed HO without muscle or periosteal stripping of bone. The authors think that the loss of motion and increased stiffness that occurred in the tethered animals were directly related to the HO, although mild degenerative facet changes were noted within or adjacent to the instrumented segments.


Fusionless modulation of sagittal growth in a skeletally immature sheep model was successfully performed. Instrumented kyphosis, vertebral body remodeling, and disc angulation all demonstrated significant changes when comparing the 12-month measurements between the 6MR and 12MR groups, and the control group. Increased stiffness within the tethered constructs following tether release was presumably related to HO, which is commonly seen in skeletally immature sheep following muscle disruption.

Key Points

  • Skeletally immature sheep appear to be a good model for studying fusionless modulation of sagittal plane growth.
  • Instrumented kyphosis, vertebral body remodeling, and disc angulation all demonstrated significant changes when comparing the 12-month measurements between the 12-month tether release group and the control.
  • Increased stiffness within the constructs was presumably related to heterotopic ossification, commonly seen in immature sheep.


1.Lowe T. Scheuermann disease. J Bone Joint Surg Am 1990;72:940–5
2.Lowe T. Scheuermann’s disease. In: Zeidman S, ed. Principles and Practice. St. Louis, MO: Mosby, 2002:681–94
3.Sorensen KH. Scheuermann’s Juvenile Kyphosis: Clinical Appearances, Radiography, Aetiology, and Prognosis. Copenhagen: Munksgaard, 1964
4.Montgomery S, Erwin W. Scheuermann’s kyphosis: long-term results of Milwaukee brace treatment. Spine 1981;6:5–8
5.Sachs B, Bradford D, Winter R, et al. Scheuermann’s kyphosis: follow-up of Milwaukee brace treatment. J Bone Joint Surg Am 1987;69:50–7
6.Betz RR, Cunningham B, Selgrath C, et al. Preclinical testing of a wedged-rod system for fusionless correction of scoliosis. Spine 2003;28(suppl):275–8
7.Stokes IA, Spence H, Aronsson DD, et al. Mechanical modulation of vertebral body growth. Spine 1996;21:1162–7
8.Akrkin A, Katz J. The effects of pressure on epiphyseal growth. J Bone Joint Surg Am 1956;5:1056–76
9.Braun J, Ogilvie J, Akyuz E, et al. Experimental scoliosis in an immature goat model: a method that creates idiopathic-type deformity with minimal violation of the spinal elements along the curve. Spine 2003;28:2198–203
10.Newton PO, Fricka KB, Lee SS, et al. Asymmetrical flexible tethering of spine growth in an immature bovine model. Spine 2002;27:689–93
11.Betz RR, Kim J, D’Andrea LP, et al. An innovative technique of vertebral body stapling for the treatment of patients with adolescent idiopathic scoliosis: a feasibility, safety, and utility study. Spine 2003;28(suppl):255–65.

hyperkyphosis; fusionless modulation; tether; biomechanics; motion analysis; Scheuermann’s disease; histology

© 2005 Lippincott Williams & Wilkins, Inc.