It has belatedly come to the authors’ attention that an inadvertent error in data manipulation has led to an erroneous association between OPD-163 and hip arthroplasty failure.

The raw data have now been re-examined and Table 1 now should read as follows:

Table 1 SNP allele and genotype frequencies in controls, aseptic and septic THA failures (numbers in parentheses)

The previously specified association between RANK+575 and aseptic failure remains unchanged.

The authors apologise for this error.