BIOMECHANICAL EVALUATION OF TRANSVERSE PROCESS FIXATION OF THE THORACIC SPINE

J. E. Hale, S. Thanapipatsiri, D. Fenton, D. P. K. Chan
Department of Orthopaedics, University of Virginia,
Charlottesville, VA 22908

Presented at the 20th Annual Meeting of the American Society of Biomechanics
Atlanta, Georgia. October 17-19, 1996

INTRODUCTION

Posterior segmental instrumentation of the thoracic spine can be accomplished by means of pedicle screw fixation, laminar fixation, spinous process fixation, or transverse process fixation. Given the potential for direct spinal cord injury, pedicle screw fixation has been employed sparingly in the thoracic region. Likewise, the use of sublaminar wire or supralaminar hooks carries the risk of neurological complication. In addition, laminar and spinous process fixation cannot be accomplished in the absence or deficiency of the laminae (e.g., spina bifida, post-laminectomized spine, fractures of the posterior bony elements).

REVIEW AND THEORY

Transverse process fixation provides three major advantages over other techniques: (a) implants do not invade the spinal canal thereby decreasing the risk of spinal cord injury, (b) implant attachment is possible even in the absence of the laminae or spinous process, and (c) a greater moment arm is provided for correction of rotational deformity in scoliosis. However, the strength of the transverse process has not been thoroughly investigated. The objectives of this study were to assess the strength of thoracic transverse process fixation constructs using stainless steel wire and Mersilene tape, and to compare the strength of transverse process fixation constructs with that of spinous process button-wire fixation construct.

PROCEDURES

Eight human thoracic spinal columns (C7-T11) with costo-transverse joints and proximal portion of the ribs intact were obtained from 5 male and 3 female fresh unembalmed cadavers (average age: 68 yrs, range: 61 to 72 yrs). Radiographic studies were performed on all specimens to exclude metastatic spinal disease and other destructive lesions of the spine. Prior to mechanical testing, specimens were disarticulated and embedded in self-curing dental acrylic to the level of the vertebral body-pedicle junction, with the posterior bony elements exposed.

For each vertebra, three fixation constructs were evaluated: 1) Transverse process wire fixation (TW). Double strands of 18 gauge stainless steel wire were placed around the left transverse process and fastened with all wires twisted together. 2) Transverse process tape fixation (TT). Mersilene tape (Ethicon, Inc.), 5 mm wide, was placed around the right transverse process and fastened with a surgical knot as advocated by Gaines and Leatherman [1981]. The tape was used to investigate the effect of implant-bone interface area on the strength of the transverse process fixation construct. 3) Spinous process wire fixation (SW). An 18 gauge button-wire (Wisconsin wire) was passed through a drill hole at the base of the spinous process and fastened with a symmetrical twist as recommended by Drummond [1988]. The order of testing was the same for all specimens (TW, TT, SW) and was chosen to minimize the effect of a particular test on the results of any subsequent tests.

Tensile loads were applied to each construct using a materials testing machine (MTS Model 858 Bionix, Minneapolis, MN) operating at a constant load rate of 10 N/sec. Loading was applied until either the bony structure or the means of fixation failed. Specimens were positioned such that the direction of the applied load was perpendicular to the surface of the bony structure being tested. Ultimate tensile force (UTF) was determined from the load-displacement data (maximum load) for each test and a multi-variate analysis of variance performed to determine statistical significance between groups.

RESULTS

A total of 73 thoracic vertebrae were studied (Figure 1). Eleven transverse process-tape constructs (16%) failed by rupture of the Mersilene tape. All other constructs failed by bony fracture. Specimens exhibited varying degrees of osteoporosis as indicated by bone mineral density values at L3 (mean = 0.72, range 0.53 to 0.86, n = 6). Weak positive correlations between bone mineral density and ultimate tensile force were noted for each fixation construct.

The mean ultimate tensile force for each fixation construct was calculated for the upper (T1-2), middle (T3-7), and lower (T8-10) thoracic spine (Table 1) and differences between groups assessed at p<0.01, unless otherwise noted. For the upper thoracic spine, UTF values for TW and TT constructs, while not different from each other, were both statistically different from UTF values for the SW construct. For the middle and lower thoracic spine, statistical differences were noted between SW and TW (p<0.05, T8-10) and between TW and TT, but not between SW and TT.

Figure 1. Representative force-displacement response for spinous process wire (SW), transverse process wire (TW), and transverse process tape (TT) fixation constructs attached at the mid-thoracic spine level.

Table 1. Mean ultimate tensile force (std.dev.) of fixation constructs for upper, middle, and lower thoracic spine in Newtons.

DISCUSSION

Mean UTF values were reasonably consistent with corresponding data from previous studies. Wenger et al. [1982] determined the strength of segmental fixation constructs using 16 gauge stainless steel wire attached to the transverse process and spinous process of human vertebrae. The strength of transverse process constructs of the upper and mid-thoracic region were 331.5 N and 237.4 N, respectively. Strengths of 286.7 N and 421.1 N were reported for attachment to the spinous processes of the mid- and lower thoracic region, respectively.

Based on experimental measurements of ultimate tensile force, transverse process wire fixation was inferior to spinous process wire fixation at all levels in the thoracic spine. In the upper levels of the thoracic spine, transverse process tape fixation does not provide any advantage over transverse process wire fixation. Interestingly, the highest mean ultimate tensile forces occurred in this region. However, in the lower thoracic spine, transverse process tape fixation proved superior to wire fixation and provided a suitable alternative to spinous process wire fixation.

REFERENCES

Gaines and Leatherman, Spine 6:483-8, 1981.

Drummond, Orthop Clin North Am 19:281-90, 1988.

Wenger et al., Orthop Trans 6:23-4, 1982.