Question:
Do patients with primary metal-on-metal hip arthroplasty have an increased risk of cancer?
Design:
Cohort study with database linkage of national registries.
Setting:
Finland.
Patients:
The study cohort consisted of 10,728 patients (59% men) who underwent metal-on-metal total hip arthroplasty (THA) between 2001 and 2010. The comparison group comprised 18,235 patients who underwent conventional metal-on-polyethylene, ceramic-on-polyethylene, or ceramic-on-ceramic total hip arthroplasties during the study period. The mean follow-up was 3.6 years in the metal-on-metal cohort and 5.1 years in the non-metal-on-metal cohort.
Risk factors:
Data from the Finnish Arthroplasty Register were linked with data from the Finnish Cancer Registry with use of personal identity codes. Patients were followed up from the date of the hip surgery until death or the end of 2010. Analyses were adjusted for sex, calendar period, and five-year age group and follow-up time since surgery.
Main outcome measures:
The primary outcome was overall cancer incidence. Risk was also assessed for the following individual cancers: soft-tissue sarcoma, basal cell carcinoma, stomach, colon, lung, uterus, prostate, kidney, bladder, non-Hodgkin lymphoma, Hodgkin lymphoma, multiple myeloma, leukemia, and melanoma.
Main results:
The risk of development of any cancer in patients with metal-on-metal total hip replacement was similar to that in the Finnish population (378 observed versus 400 expected cases, standardized incidence ratio [SIR] 0.95, 95% confidence interval [CI] 0.85 to 1.04). The cancer risk in the metal-on-metal cohort also did not differ from the non-metal-on-metal cohort (relative risk [RR] 0.92, CI 0.81 to 1.05). Among the individual cancers, the risk for soft-tissue sarcoma was not significantly higher in the metal-on-metal group than it was in the Finnish population (SIR 2.18, CI 0.71 to 5.09) or than the non-metal-on-metal group (RR 2.69, CI 0.89 to 6.71). The incidence of basal cell carcinoma in the metal-on-metal group was higher than that of the Finnish population (SIR 1.37, CI 1.15 to 1.61) and the non-metal-on-metal group (RR 1.32, CI 1.06 to 1.66). The incidence of lung cancer was lower than that of the Finnish population, both in the metal-on-metal cohort (SIR 0.58, CI 0.37 to 0.86) and the non-metal-on-metal cohort (SIR 0.68, CI 0.55 to 0.81).
Conclusion:
Patients with primary metal-on-metal hip prostheses were not at increased overall risk for cancer at a follow-up of 4 years.
The study by Mäkelä and colleagues addresses the concern that metal ion production from metal-on-metal hip prostheses may increase the incidence of cancer. While there is little support in the literature to date1, the high incidence of local problems from certain types of metal-on-metal prostheses with preliminary research and the associated media interest2 have led to heightened interest in this topic.
This study, which compared over 10,000 patients in each cohort who received arthroplasty and which linked their records to a national cancer registry, did not show an overall increased risk of cancer at a mean follow-up of four years. While the authors reported an increased risk of basal cell carcinoma, offset by a decrease in lung cancer, there is no obvious explanation for why these should happen in such a short time frame beyond chance sampling variation, and the inability of the authors to adjust for important potential confounders such as smoking and body-mass index. The study was observational and had short follow-up. Some cancers may have a long latency period, and longer follow-up is required.
In conclusion, this study should reassure the worried patient that sound epidemiological analysis has been unable to demonstrate an increased incidence of malignant tumors, and that, in the midterm, the risk of cancer is nonexistent or minimal.