Abstract
Background: During cemented hip arthroplasty revision removal of all the old cement mantle is a time staking process with multiple disadvantages. In some selected patients cementing revision stem into the old mantle is regarded as a highly attractive option. Contradictory evidence exists whether bond between two cement layers is strong enough, especially in the presence of interfering fluids.
Aim: analysis of the shearing strength of the interface between two layers of polymethylmethacrylate cement in the presence of fluid.
Methods: Cylindrical blocks of polymethylmethacrylate cement represented primary cement mantle. Its flat surface was machined to reproduce smooth old cement mantle surface comparable with that after removal of a highly polished stem (Ra=200nm). A second block was cast against the first and their junction represented the investigated interface. The influence of fluid was examined by injecting liquid onto the ‘primary’ surface prior to casting. Water or 2% water solution of carboxy-methyllcellulose (representing bone marrow viscosity of 400mPas) were used in two volumes: 0.02ml/cm2 (small) or 0.4ml/cm2 (large - surface submerged).
6 variants (control monoblock, dry surface, surface stained with small or large volume of water or highly viscous fluid) containing 7 repeats were exposed to a single shearing stress to failure at the speed of 1mm/min (Autograph AGS, Shimadzu, Japan).
Results were analyzed using 1-way ANOVA with post-hoc analysis (equal N HSD) and power calculations.
Results: Large volume of viscous fluid prevented bonding completely in two cases and significantly weakened the other samples showing mean failure stress of 5.53 MPa (95%CI:1.33–9.73 MPa). This was significantly lower compared with control monoblock (19.8–95% CI: 17.8–21.9 MPa), dry surface variant (16.9–95% CI: 15.9–18.0 MPa) and that stained with small amount of high viscosity fluid (16.01–95% CI: 15.12–17.0 MPa). Interestingly, presence of a large volume of low viscosity fluid (water) did not significantly reduce resistance to shear stress (17.05 – 95% CI:15.67–18.43 MPa).
Similar relations were observed when strain at failure and toughness were analyzed.
Conclusions: In all but large volume of viscous fluid variants, the failure occurred away from the interface between two cement layers. Large amount of viscous fluid weakened significantly this interface. If such a viscous fluid can be eliminated by copious water irrigation it is likely that strength of the cement-cement bond will be maintained. In the presence of low viscosity fluids (water, blood) careful use of gun technique is likely to allow for their escape as the cement is advanced within the femoral or the old mantle canal leading to a satisfactory bond. Our observations suggest that cement-in-cement technique seems to be biomechanically acceptable.
Correspondence should be addressed to: EFORT Central Office, Technoparkstrasse 1, CH – 8005 Zürich, Switzerland. Email: office@efort.org
