Author’s reply

The Journal of Bone and Joint Surgery. British volumeVol. 89-B, No. 7 CorrespondenceFree AccessAuthor’s replyJ. DANIEL, H. ZIAEE, C. PRADHAN, P. B. PYNSENT, D. J. W. McMINNJ. DANIELDirector of ResearchSearch for more papers by this author, H. ZIAEESearch for more papers by this author, C. PRADHANSearch for more papers by this author, P. B. PYNSENTSearch for more papers by this author, D. J. W. McMINNSearch for more papers by this authorPublished Online:1 Jul 2007https://doi.org/10.1302/0301-620X.89B7.19973AboutSectionsPDF/EPUB ToolsDownload CitationsTrack CitationsPermissionsAdd to Favourites ShareShare onFacebookTwitterLinked InRedditEmail Sir,We that Dr Vendittoli and colleagues for their interest in our paper.Our study is the first attempt at understanding in vivo metal release and systemic exposure in patients with modern metal-metal (MM) resurfacings through a prospective longitudinal assessment of the whole blood levels and daily output of cobalt and chromium. They suggest that we should not have used the inclusion criterion of selecting only two bearing diameters in our study. Including all sizes and genders is commendable, provided that adequate numbers of patients are recruited to represent every possible bearing, diameter and gender; and the temporal trend is assessed in each separately through adequately-sized cohorts. Inclusion criteria are designed to reduce confounding variability while allowing the sample to be representative of the larger population under investigation. It has been stated that metal release from surface corrosion is a function of surface area1 and clearly bearings of different diameters will have differing contribution to overall metal ion production. Being ad hoc and unselective with regard to this important variable would lead to unspecified numbers of subjects with each diameter and contribute to statistical confounding.Our choice of the two sizes (50 mm and 54 mm bearings) was based on several observations. First, these two sizes are by far the most commonly used sizes and represent more than twice as many resurfacings as all other sizes put together in our centre (over 70% need one of these two sizes). Second, young men with unilateral hip osteoarthritis managed with resurfacing have been shown2 to be the most active resurfacing group with 92% participating in sport and 62% participating in impact sport. Our purpose was to assess metal release in this extremely active patient group who are the worst-case scenario for metal ion release in correctly implanted devices. Furthermore, when attempting to assess timed excretion of metal in urine, patients need to collect urine in a specimen container without spillage or contamination and it is obvious that men are more efficient at complying with this task. They did not perform urine collection to assess 24-hour production of metal ions thus simplifying their study but reducing the overall scientific value of their work.Moreover, in their study they instructed their patients not to engage in new, strenuous activities in the week before specimen collection. Being prescriptive with regards to activities in any form has the potential to make patients restrict regular activities, rendering the investigation less representative of the day-to-day conditions these bearings are subject to in real life, and may reduce the validity of their results. Our patients were asked to live their lives as normal and collect their specimens without any restriction of physical activity at all.It appears to us that Vendittoli et al’s arguments stem from their assumption that we have presented a subset of results selected from a larger cohort. We carefully designed this investigation as a prospective longitudinal study with the stated inclusion criteria, amongst others, of using only two femoral head sizes in men to ensure the study has sufficient power. After informed consent we collected urine and whole blood specimens from all men (who were templated and found likely to need 50 mm or 54 mm femoral components; and consented to participate in this long-term study) pre-operatively but discarded specimens from patients who did not receive the stated head sizes immediately after the operation.In their letter, Vendittoli and colleagues select a subset of results from their publications. This post facto selection is unscientific and does not carry any relevance to the study. From an assessment of blood levels in a large (n = 152) cross-sectional group of men and women with unilateral Birmingham Hip Resurfacings (BHRs) with over five years follow-up we found no significant difference in either cobalt (p > 0.1) or chromium (p > 0.05) when men with 50 mm and 54 mm diameter bearings were compared with all other subjects. Neither was there a significant mean difference nor association between subjects with different bearing diameters or genders.The other possibility is that the design difference between the components used in the two studies accounts for the differences found in the two centres. Vendittoli et al state that the Durom resurfacing implant produces higher metal ion blood levels in women compared with men and higher blood levels in small implants compared with large. Also they note that the radial clearance on the Durom is 75 μm. Since it is not otherwise stated, we have assumed that this clearance is held constant across the range of sizes. The BHR, on the other hand, has a decreasing clearance with decreasing head diameter with the ratio of diameter to clearance held constant across the range. The fact that the blood metal ions with the BHR do not differ in women compared with men and across the range of head sizes may merely reflect a constant lubrication regimen across the range of BHR sizes. With a constant clearance across the range of Durom head sizes, clearly the lubrication regimen will be less favourable in smaller sizes and this design difference may account for our differing perspectives.In conclusion, we reiterate our assertion that our longitudinal metal ion study is carefully designed to minimise confounding factors and monitor in vivo metal ion release on a continuing time scale following implantation of a particular metal-metal resurfacing and forms a valid baseline for future studies. References 1 Black J. Does corrosion matter? J Bone Joint Surg [Br] 1988;70-B:517–20. Link, Google Scholar2 Daniel J, Pynsent PB, McMinn DJ. Metal-on-metal resurfacing of the hip in patients under the age of 55 years with osteoarthritis. J Bone Joint Surg [Br] 2004;86-B:177–84. Link, Google ScholarFiguresReferencesRelatedDetailsCited byAnalysis of cobalt deposition in periprosthetic bone specimens by high-resolution synchroton XRF in undecalcified histological thin sectionsMaterialia, Vol. 6Cobalt deposition in mineralized bone tissue after metal-on-metal hip resurfacing: Quantitative μ-X-ray-fluorescence analysis of implant material incorporation in periprosthetic tissue30 May 2016 | Journal of Biomedical Materials Research Part B: Applied Biomaterials, Vol. 105, No. 7 Vol. 89-B, No. 7 Metrics History Published online 1 July 2007 Published in print 1 July 2007 InformationCopyright © 2007, The British Editorial Society of Bone and Joint Surgery: All rights reservedPDF download