Methods

At our unit all patients undergoing revision of a MoM hip prosthesis have periprosthetic tissue samples graded for ALVAL. Explants undergo volumetric wear testing of the bearing and taper surfaces. We identified patients with moderate/severe ALVAL who had been exposed to lower than the median wear rate of all recorded patients who had developed ALVAL (<3mm³/year). This was termed the “ALVAL” group. We then identified all patients whose tissues had shown no signs of ALVAL. The patients in the two groups were sent buccal DNA collection kits. DNA was examined using next generation sequencing. Alleleic frequencies in the two groups were compared using Fisher's test and compared to a background UK population group (n=8514). We then conducted binary logistic regression with patient age, sex, primary source of debris (taper/bearing) and HLA genotype as the predictors. With the hypothesis that a cobalt/albumin metalloprotein acts as the epitope, we used validated binding prediction software to determine the relative affinities of the binding grooves created by different DQA1/DQB1 genetic combinations for albumin derived peptides. Given the protection that male sex and younger age appears to confer against ALVAL, we hypothesized that testosterone peptides may compete for these binding sites.