Mechanics of bone/PMMA composite structures: An in vitro study of human vertebrae

Abstract

The goal of this study was to provide material property data for the cement/bone composite resulting from the introduction of PMMA bone cement into human vertebral bodies. A series of quasistatic tensile and compressive mechanical tests were conducted using cement/bone composite structures machined from cement-infiltrated vertebral bodies. Experiments were performed both at room temperature and at body temperature. We found that the modulus of the composite structures was lower than bulk cement ($p<0.0001$). For compression at $37{}^{\circ }\mathrm{C}$: $\text{composite}=2.3\pm 0.5\mathrm{GPa}$, $\text{cement}=3.1\pm 0.2\mathrm{GPa}$; at $23{}^{\circ }\mathrm{C}$: $\text{composite}=3.0\pm 0.3\mathrm{GPa}$, $\text{cement}=3.4\pm 0.2\mathrm{GPa}$. Specimens tested at room temperature were stiffer than those tested at body temperature ($p=0.0004$). Yield and ultimate strength factors for the composite were all diminished (55–87%) when compared to cement properties. In general, computational models have assumed that cement/bone composite had the same modulus as cement. The results of this study suggest that computational models of cement infiltrated vertebrae and cemented arthroplasties could be improved by specifying different material properties for cement and cement/bone composite.

Introduction

Percutaneous filling of vertebral bodies with PMMA bone cement is a common treatment for osteoporotic compression fractures (Jensen et al., 1997, Rao and Singrakhia, 2003). In vitro tests of whole vertebra have shown that cement infiltration increased their strength (Tohmeh et al., 1999, Heini et al., 2001), but data for the material properties of the resulting cement/bone composite are sparse. These data are necessary to understand load transfer mechanisms and for accurate computational models of cement augmented vertebral bodies. Other workers have reported contradictory findings for cement/bone composites formed using cancellous bone from the proximal tibia. Jofe et al. (1991), who used human proximal tibias, reported that the cemzent/bone composite was weaker and less stiff than cement alone and that its properties did not correlate with bone volume-fraction (BVF). In contrast, Williams and Johnson (1989) found that, for composites derived from the bovine proximal tibia, modulus in the direction of major trabecular alignment correlated strongly with BVF. The present study sought to quantify the bulk material properties of cement infiltrated human vertebral cancellous bone formed using a realistic clinical procedure. Specifically, our research questions were: (1) How does the modulus and strength of the cement/bone composite differ from bulk cement? (2) Are the tensile and compressive properties of the cement/bone composite different? (3) Does BVF influence the properties of the cement/bone composite?