Effect of Spinal Extension on Restoring Geometric and Loading Alignment of the Thoracic Spine with a Vertebral Compression Fracture

Avinash G. Patwardhan, PhD1,2; Ioannis N. Gaitanis, MD1,2,3; Gerard Carandang, MS1; Frank M.

Phillips, MD4; Alexander J. Ghanayem, MD1,2; Leonard I. Voronov, MD1,2; Robert M. Havey, BS1,2; Thomas M. Gavin, CO1,2

1 Department of Veterans Affairs, Edward Hines Jr. VA Hospital, Hines, Illinois;
2 Loyola University Medical Center, Maywood, Illinois;
3 University Hospital of Crete, Heraklion, Crete, Greece;
4 Rush University Medical Center, Chicago, Illinois.

Address correspondence to:
Avinash G. Patwardhan, PhD,
Department of Orthopaedic Surgery and Rehabilitation,
Loyola University Medical Center, 2160 S. First Avenue, Maywood, IL 60153, USA.
Phone: (708) 202-5804; Fax: (708) 202-7938; E-Mail: apatwar@lumc.edu

INTRODUCTION

Vertebral compression fractures (VCFs) are the most common complication of osteoporosis in the aging population. Wedge type VCFs occur about the thoracolumbar junction (T12-L1) and the mid-thoracic spine (T7-T8) [10]. These fractures lead to increased thoracic kyphosis, which may worsen over time in persons with prevalent VCFs [4]. Furthermore, the presence of VCFs with the resulting kyphotic deformity is a strong risk factor for additional vertebral fractures [8]. This is thought to be the result of altered loading caused by the kyphotic deformity [8,10].

The balloon kyphoplasty procedure has been developed to reduce the kyphotic deformity caused by a VCF. In this procedure, inflatable bone tamps (IBTs) are introduced into the fractured vertebral body and then inflated in an attempt to reduce the fracture deformity. Clinical studies of kyphoplasty have reported kyphosis reduction of up to 65% and restoration of vertebral body height of up to 68% [5-7,11]. Others [9] have reported that spinal extension alone can correct the vertebral body height loss caused by the VCF in fractures that have "dynamic mobility."

Restoration of the vertebral height and sagittal alignment are believed to mitigate adverse biomechanical consequences of osteoporotic VCFs; however, there are no biomechanical data on how the VCF deformity and its correction affect the loads in the adjacent vertebral bodies. The goal of this biomechanical study was to investigate the effects of balloon inflation vs. spinal extension on the restoration of spinal alignment following a thoracic VCF.

MATERIALS AND METHODS

Specimens

Six thoracic specimens from five donors (age: 81±10 years) were used. Each test specimen consisted of three adjacent vertebrae with all discs, ligaments and bony structures intact. Four specimens were from T10-T12 and two were from T7-T9.

Experimental Creation of VCF

A technique was developed for reproducibly creating a VCF in the middle vertebra of the test specimen. This technique involved disruption of the cancellous bone in the vertebral body, thereby creating a "stress-riser" in the target vertebra. Under A-P and lateral fluoroscopic control transpedicular channels were created in the middle vertebra of the test specimen for the insertion of IBTs (Kyphon, Inc., Sunnyvale, CA ). Next, IBTs were inflated in the vertebral body to cause a disruption of the cancellous bone (Fig. 1).

The IBTs were removed and the specimen was loaded in compression using bilateral loading cables such that the load path was perpendicular to the vertebral endplates and the application of the load did not change the sagittal alignment of the specimen. The compressive load on the bilateral loading cables was gradually increased until a fracture with at least 25% loss of the anterior vertebral body height was observed on fluoroscopy (Fig. 2A-B). When the fracture was established, the specimen remained under a physiologic preload of 250 N.

Reduction of VCF Deformity Using Spinal Extension

Effect of spinal extension on the VCF deformity was investigated with the specimen maintained under a compressive preload of 250 N. Spinal extension was simulated by applying extension moments to the test specimen in increments of 1 Nm. The extension moment was increased until the loading cables were aligned with the radiopaque markers placed in the adjacent vertebrae (corresponding to the pre-fracture alignment) (Fig. 2C). The magnitude of the extension moment that was necessary to achieve the restoration of the compressive load path to its pre-fracture alignment was recorded.

Correction of VCF Deformity Using Balloon Inflation

After removal of the extension moment, the specimen was again subjected to a compressive preload of 250 N. Correction of the VCF deformity was performed using two inflatable bone balloons (IBTs) inserted into the fractured vertebral body through the same pedicle channels that were created for cancellous bone disruption. The IBTs were inflated with radiopaque contrast medium in 0.5 mL increments bilaterally to reduce the fracture deformity (Fig. 2D) [7]. Vertebral body augmentation with cement was not performed.

Measurements and Statistical Analysis

Radiographic measurements were performed to determine the heights of the fractured vertebral body (Fig. 3). Vertebral kyphosis angle was measured between the two endplates of the fractured vertebra. Segmental kyphosis angle was measured between the endplates of adjacent vertebrae. Location of the compressive load path was measured in all three vertebrae (Fig. 3). The shift of the compressive load path was expressed as a percent of the total A-P endplate width. The measurements of geometric and loading alignment were performed on digital VF images by three observers. The measurements were averaged and analyzed using repeatedmeasures analysis of variance with Bonferroni correction for multiple comparisons.

RESULTS

The thoracic VCF caused significant changes in the pre-fracture geometric alignment. The fractured vertebra lost 37±15% of the intact height at the anterior wall (p<0.01). The middle height (measured at the mid-point between the anterior and posterior walls) lost 34±16% (p<0.05) (Fig. 4). Balloon inflation significantly reduced the VCF deformity (p<0.05). The anterior wall height was restored to 91±8.9% and middle-height to 91±14% of intact values. Application of extension moments alone restored the anterior wall height to 85±8.6% of intact. However, spinal extension was unable to produce a significant correction of the middle height of the fractured vertebra. The middle height of the fractured vertebra after the application of extension moments was smaller than that achieved after IBT inflation, and remained significantly smaller as compared to the pre-fracture value (p<0.05).

The VCF increased vertebral kyphosis by 13±5.5 degrees (p<0.01) and segmental kyphosis increased by 14±7.0 degrees (p<0.05) (Fig. 5). IBT inflation corrected the segmental kyphosis by 8.7±5.5 degrees (66±50% correction of the increase caused by the VCF) (p<0.05). The segmental kyphosis was restored to within 5.6±5.9 degrees and vertebral kyphosis to 2.4±4.0 degrees of the corresponding pre-fracture values. The application of extension moments was able to restore the segmental kyphosis to within 0.8±3.1 degrees of the pre-fracture value. However, the vertebral kyphosis of the fractured vertebra under the application of extension moment remained significantly larger than the pre-fracture value (p<0.05).

The thoracic VCF induced significant anterior shift of the compressive load path in the fractured vertebral body as well as in the adjacent vertebrae (p<0.05) (Fig. 6). Inflation of IBTs under a compressive preload of 250 N shifted the compressive load path posteriorly relative to the post-fracture state (p<0.01), moving it closer to the pre-fracture alignment. The location of the load path after IBT inflation still remained anterior to the pre-fracture location in all three vertebrae by about 10% of the A-P endplate width. Spinal extension was able to fully restore the load path to its pre-fracture alignment in the adjacent vertebrae. The mean extension moment needed to fully restore the loading alignment was 6.3±2.2 Nm.

DISCUSSION

A new technique was developed for reproducibly creating a VCF in the middle vertebra of the test specimen. The method allowed for experimental creation of a typical osteoporotic wedge fracture in a reproducible manner with a control over the amount of height loss at the anterior wall.

Each test specimen consisted of three adjacent vertebrae with all discs, ligaments and bony structures intact, with an experimentally created VCF in the middle vertebra. This is in contrast to previous studies [1-3,12] that investigated the response of individual vertebrae without posterior elements. The motion response of adjacent vertebrae under physiologic loads is affected by the load sharing in the discs, ligaments and intact facet joints. Therefore, preservation of the discs, ligaments, and facet joints is essential when investigating the effects of a vertebral fracture and its reduction on the kinematics and kinetics of the adjacent segments. The present study simulated the reduction of recent vertebral compression fractures. The restoration of the anterior vertebral body height after balloon inflation under a compressive preload of 250 N was 6.0±2.6 mm (76±23% recovery of the lost height) and the correction of kyphosis was 8.7±5.5 degrees (68±38%). These results are within the range of values reported in clinical studies [5-7,11].

The effects of a thoracic VCF and its reduction on the changes in the loading alignment of adjacent segments have not been investigated. A thoracic VCF caused a significant anterior shift of the compressive load path in the fractured vertebral body and in the adjacent vertebrae. An anterior shift of the compressive load path, caused by the VCF and the resulting kyphotic deformity, will produce an eccentric loading on the adjacent vertebrae, inducing additional flexion moments that are not present in the normal alignment of the spine. This may contribute to increasing the risk of new fractures in osteoporotic vertebrae adjacent to an uncorrected VCF deformity as the physiologic compressive load is now placed over the anterior region of the vertebrae weakened by a loss in trabecular bone mass.

A significant restoration of the load path was observed after IBT inflation. The anterior shift of the compressive load path caused by the VCF deformity was recovered by approximately 56% in the fractured as well as adjacent vertebral bodies. This suggests that a correction of the geometric spinal alignment using IBT inflation has the potential to mitigate some of the adverse biomechanical loads at the adjacent vertebrae induced by the VCF.

Spinal extension alone also had a positive effect in correcting the anterior wall height loss of the fractured vertebra and segmental kyphotic deformity, and in realigning the compressive load path to its pre-fracture alignment. However, the middle height of the fractured vertebra could not be restored to the pre-fracture value with spinal extension alone. This suggests that the application of spinal extension moments may not be helpful in correcting the vertebral deformity in concave or bi-concave osteoporotic vertebral fractures.

CONCLUSIONS

Application of extension moments, analogous to using a hyperextension posterior shell TLSO, was effective in restoring the pre-fracture geometric and loading alignment of adjacent segments, however the middle height of the fractured vertebra and vertebral kyphotic deformity were not fully corrected with spinal extension alone. This implies that the correction of segmental kyphosis with spinal extension occurred in part through the intervertebral discs adjacent to the fractured vertebra. The results suggest that it is feasible to restore the anterior height of the fractured vertebra and the geometric and loading alignment of adjacent segments using a hyperextension posterior shell TLSO without damage, since the extension moment needed to achieve this was well below the elastic limit of the soft tissues. The results support further clinical studies of extension orthoses in the management of the elderly patients with osteoporotic wedge compression fractures of the thoracic spine.

REFERENCES

BelkoffSM, Jasper L, Stevens S. An ex vivo evaluation of an inflatable bone tamp used to reduce fractures within vertebral bodies under load. Spine 2002;27:1640-3.
BelkoffSM, Mathis JM, Deramond H, Jasper L. An ex vivo biomechanical evaluation of a hydroxyapatite cement for use with kyphoplasty. Am J Neuroradiol 2001;22:1212-6.
BelkoffSM, Mathis JM, Fenton DC, et al. An ex vivo biomechanical evaluation of an inflatable bone tamp used in the treatment of compression fracture. Spine 2001;26(2):151-6.
Cortet B, Roches E, Logier R, Houvenagel E, Gaydier-Souquieres G, Puisieux F, Delcambre B. Evaluation of spinal curvatures after a recent osteoporotic vertebral fracture. Joint Bone Spine. 2002;69:201-8
Gaitanis I, Hadjipavlou A, Katonis P, Tzermiadianos P, Kontakis G. Balloon Kyphoplasty for the treatment of pathological vertebral fracture of the thoracic and lumbar spine. Spine Across the Sea, Hawaii, USA, July 27-31, 2003.
Garfin R, Yuan HA, Reily M. New technologies in spine. Kyphoplasty and vertebroplasty for the treatment of painful osteoporotic compression fractures. Spine 2001; 26:1511-5.
Lieberman IH, Dudeney S, Reinhardt MK, Bell G. Initial outcome and efficacy of "kyphoplasty" in the treatment of painful osteoporotic vertebral compression fractures. Spine 2001;26:1631-8.
Linville DA. Vertebroplasty and kyphoplasty. Southern Medical Journal 2002;95:583-7.
McKiernan F, Jensen R, Faciszewski T. The dynamic mobility of vertebral compression fractures. J Bone Miner Res. 2003;18:24-9.
Nevitt MC, Ross PD, Palermo L, Musliner T, Genant HK, Thompson DE. Association of prevalent vertebral fractures, bone density, and alendronate treatment with incident vertebral fractures: effect of number and spinal location of fractures. The Fracture Intervention Trial Research Group. Bone. 1999;25:613-9.
Phillips FM, Ho E, Campbell-Hupp M, McNally T, Wetzel FT, Gupta P. Early radiographic and clinical results of balloon kyphoplasty for the treatment of osteoporotic vertebral compression fractures. Spine 2003;28:2260-5
Tomita S, Kin A, Yazu M, Abe M. Biomechanical evaluation of kyphoplasty and vertebroplasty with calcium phosphate cement in a simulated osteoporotic compression fracture. J Orthop Sci. 2003;8(2):192-7.

FIGURE LEGENDS

Fig. 1. Intra-vertebral body space preparation for the creation of a VCF. (A) Intact specimen. (B) Inflation of IBT inserted into the vertebral body through transpedicular channels. (C) Disruption of the cancellous bone in the vertebral body seen after the removal of the IBTs.

Fig. 2. Effect of VCF and fracture reduction on the geometric and loading alignment. The VF images are for a T10-T12 specimen. (A) Intact specimen. (B) After creation of a compression fracture in the middle vertebra. (C) Under an extension moment with the specimen under 250 N compressive preload. (D) After IBT inflation under 250 N compressive preload.

Fig. 3. Radiographic measurements of geometric and loading alignment. The vertebral body heights were measured at three locations: at the anterior wall (Ha), at mid-point between the anterior and posterior walls (Hm), and at the posterior wall (Hp). The segmental kyphosis angle (è) was measured between the inferior endplate of the upper vertebra and superior endplate of the lower vertebra. The vertebral kyphosis angle (Ö) was measured between the two endplates of the fractured vertebra. The locations of the load path were measured along vertebral body endplates visible in the lateral x-ray images (arrows labeled A-D).

Fig. 4. Vertebral heights of thoracic vertebrae before and after a reduction of the compression fracture. The heights measured at the anterior wall and at the mid-point between the anterior and posterior walls are expressed as a percentage of the corresponding pre-fracture heights. Mean (±SEM) values are shown. Page 12

Fig. 5. Segmental and vertebral kyphosis (in degrees) before and after a compression fracture and its reduction. Please see Fig. 3 for the definitions of kyphosis angles. Mean (±SEM) values are shown.

Fig. 6. Location of the physiologic compressive load path relative to the pre-fracture (intact) spine in the fractured and adjacent vertebral bodies. The change in the location of the load path is expressed as a percent of the A-P endplate width. Mean (±SEM) values are shown.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6