COMPARISON OF PULLOUT CHARACTERISTICS FOR TWO TAPERED BONE SCREWS

J. E. Hale, D. V. Carmines, C. L. Baker
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

Screw fixation is one of the most widely used means of affecting internal stabilization. However, implant related complications, such as screw loosening, compromise the usefulness of such procedures. To improve the screw/bone interface strength, a self-tapping bone screw with a tapered minor diameter and a buttress thread was developed. Compared with constant minor diameter screws, a tapered design increases the interface area and cancellous bone volume between the distal screw threads.

Although previous studies have reported a "small but significant effect" of major/minor diameter on pullout strength [DeCoster, 1990], the effect of tapered minor diameter screws has not been widely studied. The purpose of this study was to compare the screw/bone interface strength of a new tapered screw with that of a similar, commercially-available device.

PROCEDURES

The strength of the screw/bone interface was investigated in the pedicle of lumbar vertebrae. Fifteen lumbar specimens (L1-L5) were obtained from four human cadaveric lumbar spines (3 male, 1 female), dissected down to the ligamentous tissues, and subjected to non-invasive bone density measurement (DEXA). Prior to testing, the pedicle dimensions were measured and a pedicle screw was inserted while measuring insertion torque. Specimens with screws inserted were mounted in the test fixture and a monotonic tensile load was applied along the axis of the screw through a circular opening (1.25 cm diameter) in the fixture. Loading to failure was performed using a materials test system( MTS Model 858 Bionix, Eden Prairie, MN) under a controlled displacement rate of 1 mm/sec. Specimens were randomly X-rayed with screws inserted and then sectioned following pullout testing to inspect screw purchase and alignment. Two self-tapping, cannulated screws were evaluated, each with major diameter=7 mm, thread length=40 mm, and pitch=2.8 mm. One screw (Danek, Cannulated bolt: P/N 823-131) featured a proximal taper (PT) reducing the minor diameter from 6 mm to 5 mm within 4 threads, followed by a constant diameter of 5 mm to its distal tip. The second screw provided a full taper (FT), with minor diameter decreasing from 6 mm proximally to 4.2 mm distally. Insertion was performed according to the standard surgical procedure. Pilot holes were drilled using a straight drillbit (PT screw) or a drillbit with matched taper (FT screw), each undersized by 0.3 mm diameter, respectively. To obtain paired data for the two screw designs, one screw of each type was inserted into the right and left pedicles of each vertebral body tested. Assignment of the screws to the right or left pedicle was reversed in adjacent levels to minimize the effect of specimen asymmetry.

RESULTS

Mechanical characteristics were computed from the load-displacement curve for each screw test, including: pullout force (maximum load), loading stiffness (slope of load-displacement curve over linear region), and loading energy (area under load-displacement curve up to maximum load) (Figure 1).

The FT screw provided greater pullout forces than the PT screw in 12 of 15 vertebrae tested (range: 6% to 80%, p=0.083 for paired t-test). The loading energy was also higher for the FT screw in 10 of 15 vertebrae, while comparable loading stiffnesses were observed for the two screws. Means and standard deviations were computed for population test data concerning pullout force, loading stiffness, and energy to failure (Table 1).

Table 1. Population mean (standard deviation) for pullout characteristics of two tapered screws.

Bone mineral density measurements for the fifteen specimens ranged from 0.672 - 0.904 gm/cm2 (mean = 0.761).

DISCUSSION

Commonality of the proximal screw design standardized the purchase of the two screws within the posterior cortical bone and pedicle isthmus, thereby providing a comparison of the effect of the major/minor diameter ratios within the vertebral body. The increased major/minor diameter ratio of the FT screw provided increased holding strength in cancellous bone compared with the PT screw.

Correlation analysis revealed a weak positive correlation between pullout force and bone mineral density (CORR=0.25, both screws). However, the pullout force correlated more strongly to insertion torque for the PT screw (CORR=0.65) as compared to the FT screw (CORR=0.37), indicating that the FT screw may be less sensitive to insertion torque errors.

Direct comparison of the pullout force results with those of previous studies is complicated by differences in bone density, screw diameter, insertion torque, etc. Soshi et al. [1991] presented a linear correlation between bone mineral density and pull-out force for a 7.0 mm major diameter screw with a constant minor diameter. Based on the mean bone mineral density for specimens in this study, a pullout force of 720.2 N is obtained, less than that of either tapered screw. Prior evaluation of a tapered screw (major diameter=6.0 mm, minor diameter=3.9-5.0 mm) by Yamagata et al. [1992] demonstrated improved endurance limit in bending fatigue tests, but did not report pullout strength.

Results of this study indicate that the full tapered pedicle screw design provides comparable, if not improved, pullout strength when compared to the proximally tapered screw. It also provides comparable mean stiffness and loading energy up to ultimate load. Based on these findings, additional testing and design development may lead to an improved screw design based upon minor diameter thread taper.

REFERENCES

DeCoster, T.A. et al., J Orthop Trauma 4(2):169-74, 1990.

Soshi, S. et al., Spine 16(11):1335-41, 1991.

Yamagata, M. et al., Spine 17(3S):51-4, 1992.

ACKNOWLEDGMENTS

This work was supported by a grant from the Sofamor-Danek Corporation (Memphis, TN). The authors also gratefully acknowledge the contributions of E. Jang, S. Thanapipatsiri, C. McLaurin and G. Wang to this work.