OTA 1997 Posters - Pelvic & Acetabular Fractures

*Stability of Open Book Pelvic Fractures Using a New Biomechanical Model of Single Limb Stance

Michael C. MacAvoy, BE, R. Trigg McClellan, MD, Chi-Ray Chien, ME, William A. Allen, BS, Marjolein C. H. van der Meulen, PhD, Stuart Goodman, MD

Santa Clara Valley Medical Center, San Jose, California, USA; Stanford University Medical Center, Stanford, California, USA

Introduction: "Open book" pelvic fractures may be stabilized by internal fixation, but it is not clear which methods provide the best stability. Biomechanical models of double-limb stance are very common, but are of limited clinical relevance. Single-limb stance is a more rigorous test of pelvic stability and more closely resembles the asymmetric stances involved in walking, stair-climbing, and running. Our model of single-limb stance solves the problems of applying the forces of muscles and body mass to cadaver pelves, and was used to compare single versus double plate fixation of the pubic symphysis after open book pelvic fracture.

Materials And Methods: Specimen Preparation: Nine embalmed cadaveric pelves (6 female and 3 male, mean age 79), without any gross structural abnormalities were dissected of all soft tissue, sparing the hip joint capsules and the ligaments of the pelvic ring and floor. To interface with the testing equipment, the L3 vertebra and proximal third of one femur on each pelvis were potted in polyester resin. Correct stance was achieved by alignment of the anterior superior iliac spines and the pubic symphysis in a coronal plane, the ischial tuberosities in a horizontal plane, and the shaft of the femur in a coronal plane. Adjustable cables were attached to the specimen to mimic the muscle forces of gluteus medius and minimus (hip abductors), psoas major (hip flexor), and gluteus maximus (hip extensor). The lengths of the cables were adjusted prior to mounting each pelvis in the testing machine.

Joint Micromotion Measurement: Shear strain at the posterior surface of the pubic symphysis and gapping at the anterior face of the sacroliac joint were measured with extensometers mounted across each joint surface. The knife edges of the extensometer at the sacroiliac joint were oriented parallel to the local joint plane, while those of the extensometer at the pubic symphysis were oriented perpendicular to the joint plane of the pubic symphysis. Data was sampled at 50 Hz.

Force Application: The spinal and femoral fittings on each pelvis were attached to ball-and-socket moment releases to eliminate moments and horizontal force reactions. The applied loading was based on a body mass of 53 kg (400 N), which was confirmed through testing to be suitably taxing on the pelves. The choice of body mass was intended to avoid failure loads, which are lower than usual for the following reasons: 1) the application of muscle forces in a single-limb model produces potentially damaging joint forces and point loads on the pelvis exceeding twice body mass. 2) most of the pelves were from elderly women, and 3) the experimental design necessitated loading each specimen multiple times. 60% (32 kg) of the body mass was applied at the spine, and 17% (9 kg) was applied at the contralateral femur to simulate the free hanging leg. The spinal force was applied cyclically, ramping from zero to the peak load three times at a frequency of 1 Hertz. Data from the third cycle was used for subsequent analysis.

Experimental Sequence: Intact, fixed, and injured-unfixed pelves were tested sequentially. After testing the intact pelvis, an "open book" injury was simulated by disrupting the pubic symphysis, anterior sacroiliac, sacrotuberous, and sacrospinous ligaments. After creation of the injury, surgical fixation was achieved by either a single or double plating method, in randomized order, and each specimen was re-tested. The single plate was a curved 6-hole 3.5 mm reconstruction plate across the superior pubic symphysis. The double plating method consisted of the 6-hole plate plus a 4-hole 3.5 mm reconstruction plate placed anterior to it. Finally, the plates were removed and the tests repeated on the injured-unfixed pelves.

Statistical Analysis: Differences in displacement among the groups of pelves were compared based on method of fixation as a single factor. A one-way ANOVA (F-test), followed by a corrected multiple comparisons test (Bonferroni/Dunn) in addition to descriptive statistics were performed. Statistical power calculations were based on sample size, measured standard deviation, size of detectable effect, and the t statistic for a significance level of p=0.05.

Results: Motions at the symphysis and sacroiliac joints were significantly greater than intact in both plated and unfixed groups (p<0.005). The unfixed pelves were grossly unstable, exhibiting motion on the order of centimeters (mean: 6.5 cm), while fixed and intact specimens were stable. In the plated groups, mean symphyseal shear was <0.74 mm, and the mean intact symphyseal shear was <0.19 mm. Although the double plated group showed less shear displacement at the symphysis than the single plated group, the mean difference was only 0.08 mm, which was not statistically significant. Power calculations based on our sample size, measured standard deviation, and p=0.05 indicate that this study has 90% power to detect a difference 0.56 mm of shear at the pubic symphysis.

Conclusions: The instability of the open book injury is severe under single-limb stance, in contrast to double-limb behavior. The pubic symphysis is crucial for maintaining pelvic stability in the open book injury, in which the posterior sacroiliac ligaments are intact. Fixation of the pubis symphysis alone with one or two contoured 3.5 mm reconstruction plates is sufficient to reestablish stability in the open book injury. The difference in stability between single and double plating may be clinically insignificant.