This erratum corrects the following:
Hailer NP, Garland A, Gordon M, Kärrholm J, Sköldenberg O, Eriksson N, Garmo H, Holmberg L. No generally increased risk of cancer after total hip arthroplasty performed due to osteoarthritis. Int J Cancer. 2020 Jul 1;147(1):76‐83. doi: 10.1002/ijc.32711. Epub 2019 Nov 5. PMID: 31595487.
Going through their database and syntax in the context of a novel project the authors found that the translation of ICD‐7‐codes into specific cancer diagnoses was erroneous in the initial analysis with respect to prostate cancer. This error led the authors to identify fewer cases with prostate cancer than were actually present in the cohort.
After recalculation, the adjusted hazard ratio for being affected by prostate cancer in individuals exposed to a hip arthroplasty is 1.07 (1.03‐1.11), whereas the published estimate was 0.97 (0.90‐1.05). Since prostate cancer is a relatively common cancer the estimate for the overall risk of cancer for individuals exposed to a hip arthroplasty was also affected, and it is now 0.99 (0.97‐1.01) whereas the published estimate was 0.97 (0.95‐0.99).
Taken together, the key message that hip arthroplasty does not increase the overall risk of cancer is still valid and unchanged. A slight increase in the risk of being affected by prostate cancer is however observed in men exposed to hip arthroplasty. Under the widespread use of prostate‐specific antigen (PSA) testing also on wide indications in men in Sweden since 1997, the estimate for prostate cancer is highly likely attributable to a diagnostic bias in men in frequent contact with health care providers, especially when the symptoms have involved pain in hips and the pelvic region.
The authors sincerely apologise for any confusion or inconvenience this error may have caused.
For all authors:
Nils P. Hailer and Lars Holmberg