Authors’ Response to Letters to the Editor Re: Interpreting cobalt blood concentrations in hip implant patients

To the Editor:

We appreciate the opportunity to respond to the letters by Dijkman et al., as well as Devlin and Pomerleau regarding our recent paper on cobalt published in this journal in February.¹

Dijkman et al. have asked whether we have changed our recommendations regarding the interpretation of blood cobalt concentrations in hip-implant patients, stating, “What evidence is there (with a time difference of only 6 months between these articles!) to support this increase from 100 to 300 μg/L?” Our views on interpreting blood cobalt concentrations have not changed. The following text appeared in our original submission of the manuscript to this journal:

“Based on currently available data, when blood cobalt concentrations over 100 μg/L are discovered, we feel it would be useful to also obtain testing for thyroid function and elevated RBC levels.”

This text was removed at the request of a peer reviewer and the journal editor after two rounds of peer review. They believed it was inappropriate to recommend routine screening in potentially asymptomatic patients, and asked that we did not include the recommendation in our text. We respected their request. Nonetheless, since most physicians will likely be measuring cobalt in blood to monitor for excessive wear, we do believe that it would be useful to monitor patients for thyroid effects and polycythemia (the most sensitive systemic effects of cobalt) when blood cobalt concentrations exceed 100 μg/L.

Devlin and Pomerleau have asked whether we have considered certain factors that they believe might impact the applicability of our research. The authors first state that “metal nanoparticles and metal ions do not behave similarly.” We acknowledge that nanoparticles and ions are not synonymous, and that they may produce different biological responses. Based on the currently available evidence, it appears that the majority of the systemic cobalt measured in hip-implant patients is present in the ionic form. The nanoparticles from wear debris that enter systemic circulation are predominantly chromium because of the relatively high solubility of cobalt. In support of this conclusion, we refer the reader to:

1. Numerous studies showing that cobalt dissolves from particles in-vitro.^2,3

2. Studies characterizing wear debris particles in hip-implant patients, which indicate that the particles are largely chromium-rich with little or no cobalt.^4,5

3. The passivation layer, which ranges in depth from 18 nm to 85 nm, is primarily comprised of oxidized chromium, so it is thought that the chromium oxide nanoparticles that have been described in some metal-on-metal hip implant patients originated from the chromium-rich passivation layer.^3,6

4. Studies showing that the distribution of cobalt in various blood compartments such as plasma and whole blood is similar in hip-implant patients and human volunteers ingesting a cobalt dietary supplement.^7,8

In addition, Devlin and Pomerleau also note that there may be “latency between exposure, peak blood values and the onset of symptomatic disease in metal toxicology.” We believe that blood cobalt concentrations can be used as a marker for potential systemic effects in hip-implant patients, since the case studies of hip-implant patients in the current literature identify the same systemic responses to cobalt as those identified in human and animal studies (following ingestion). We would not expect a latency response to cobalt in blood, since such an outcome has not been reported in the animal and human studies conducted thus far.

As we have noted here, we agree with Devlin and Pomerleau that monitoring implant patients for signs of hypothyroidism and polycythemia starting at blood or serum cobalt concentrations above 100 μg/L would be prudent. And, we agree that a collection of these data and a thoughtful interpretation of what was learned would be helpful in validating our recommendations.

Respectfully,

Dennis Paustenbach, David Galbraith, and Brent Finley

Cardno ChemRisk, San Francisco, CA, USA

Declaration of interest

All of the authors are employed or contracted by Cardno ChemRisk, a consulting firm providing scientific advice to the government, regulatory agencies, corporations, law firms, and various scientific and professional organizations. Cardno ChemRisk has been engaged by DePuy Orthopedics, Inc., a manufacturer of prosthetic devices. It is likely that this work will be relied upon in future medical research, nutrition research, and in litigation proceedings. Some of the authors may be called upon to serve as expert witnesses and one of the authors (Paustenbach) has previously testified on behalf of DePuy. Funding for the paper was provided by DePuy.