An evaluation of the effect of clotting and processing of blood samples on the recovery of copper from bovine blood R.A. Laven a, * , C.T. Livesey b a Scottish Agricultural College Veterinary Services, St. MaryÕs Industrial Estate, Dumfries, DG1 1DX, United Kingdom b Veterinary Laboratories Agency, New Haw KT15 3NB, United Kingdom Accepted 4 November 2004 Abstract The reference ranges for blood copper (Cu) concentration used by SAC and the UK Veterinary Laboratories Agency have been developed for use with plasma only. Copper sequestration into the clot, primarily in the form of caeruloplasmin, means that serum Cu concentrations are significantly lower than plasma Cu concentrations in the same animals. Serum Cu is commonly used as a surrogate for plasma Cu by applying a proportional conversion factor. However the results of these two studies, which involved a total of 104 pairs of serum and Cu samples from 84 Holstein cattle in one herd, suggest that serum Cu concentrations should not be used in this way. First, the reduction in Cu concentration during clotting was unrelated to initial plasma concentration (vary- ing from 8–59%), which suggests that using a proportional conversion factor is incorrect. Secondly, the correlation between plasma and serum Cu concentration, while significant, was relatively low (r 2 µ 0.6), with the individual variation in Cu loss during clotting ranging from 1.5 to 10.8 lmol/L (mean 4.9 lmol/L). Although this study was undertaken using samples from normocupraemic lac- tating Holstein cattle in one herd, the data strongly suggest that plasma rather than serum Cu should be used to estimate Cu status unless the laboratory used has developed a reliable serum reference range. We also found that factors other than plasma Cu concentration had a significant effect on loss of Cu during the clotting process. However, no significant effect on the loss of Cu during clotting or by varying the incubation/storage time and temperature of bloods prior to clot removal was noted, nor was there any significant effect of age or Cu intake. Furthermore, when the loss of Cu during clotting was examined in the same heifers at two different timepoints, no correlation between the results was found. Further research is required to establish the factors that affect the loss of Cu during clotting. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Cattle; Copper; Serum; Plasma; Correlation 1. Introduction In the UK, copper (Cu) deficiency is the most com- monly diagnosed nutritional deficiency in cattle (Anon, 2002) and Cu toxicity is being increasingly recognised in adult dairy cows (Livesey et al., 2002). The majority of diagnoses of these two diseases, particularly defi- ciency, are made on the basis of blood Cu concentra- tions, but diagnosis is not simple as whole blood, plasma, and serum Cu concentrations are only approx- imate measures of Cu status. Liver Cu concentration may appear to be a better measure, but even this reflects Cu status of the previ- ous diet and not necessarily functional Cu status at critical sites. Furthermore, the correlation between liver and blood, plasma or serum Cu is irregular (Under- wood and Suttle, 1999). Whole blood Cu includes that bound in erythrocytes and therefore whole blood Cu concentration reflects historical Cu status because of the extended biological half-life of erythrocytes com- 1090-0233/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tvjl.2004.11.008 * Corresponding author. Tel.: +44 1387 267260; fax: +44 1387 255028. E-mail address: richard.laven@sac.co.uk (R.A. Laven). www.elsevier.com/locate/tvjl The Veterinary Journal 171 (2006) 295–300 The Veterinary Journal