Abstract
Laboratory tests are important for the diagnosis of both hypothyroidism and hyperthyroidism because of the lack of specificity of the typical clinical manifestations of these disorders. Thyroid function is usually monitored by measuring thyroid-stimulating hormone (TSH) in combination with free thyroxine (fT4) and sometimes free triiodothyronine (fT3). In our hospital, these hormones are measured with immunoassays from Roche Diagnostics (Mannheim, Germany) using monoclonal antibodies with a ruthenium complex.
Two female patients are described here, who had results that were difficult to explain. The first patient was taking levothyroxine because of hypothyroidism. We measured a very high fT4 concentration of 85 pmol/L. Her general practitioner doubted the validity of this result given that the patient did not appear thyrotoxic. He therefore referred her to another laboratory for thyroid hormone measurement using a different method (Architect, Abbott, IL, USA). A fT4 concentration of 24 pmol/L was reported (Table 1). After deliberation, the second sample was also measured with our Roche immunoassay, and again a very high fT4 concentration of 75 pmol/L was observed. The second patient was referred within our hospital to the endocrinologist with elevated thyroid hormone concentrations and a normal TSH: thyroid function tests had been measured because of hair loss. This raised the possibility of two rare disorders, i.e. thyroid hormone resistance and a TSH-secreting pituitary adenoma. However, analysis of that sample with another immunoassay system (Autodelfia, Perkin-Elmer, MA, USA) demonstrated a completely normal TSH, fT4 and fT3 (Table 1).
Thyroid hormone concentrations found with different methods
Locally derived reference values appropriate for the assays cited are given in brackets after the results. nd: not determined
*Determined by Architect
†Determined by Autodelfia
There are many causes of interference in immunoassays 1–3 including heterophilic antibodies, antithyroid hormone antibodies, rheumatoid factor (RF) and interfering drugs. 4 As far as possible these were excluded in these two samples using a heterophilic blocking tube, measuring RF, and screening for drug use. The difference between our findings and the immunoassays in other laboratories led to the conclusion that the results were due to another interfering factor. Because the fT3 assay is known to be sensitive to antiruthenium antibodies, 5,6 the samples of these two women were analysed in Roche Diagnostics Laboratory. In both patients, an interference against ruthenium label was detected, explaining the interference in our fT3 and fT4 assays.
As far as we know, this is the first report that demonstrates interference of antiruthenium antibodies in the Roche fT4 assay. Interference of antiruthenium antibodies in the Roche fT3 assay has been described previously. 5,6 Sapin et al. described that the antiruthenium antibodies in their study did not interfere with the fT4 assay, and in the paper of Ando et al. antiruthenium antibodies seemed to interfere at least partially in the fT4 assay, but could not be confirmed. In our patients, the interference of the antiruthenium antibodies was similar in the fT4 and the fT3 assay. Allthough ruthenium was also present in the TSH assay, we did not observe its interference in that assay. Although our results are in accord with the previous studies, 5,6 we do not have an explanation of why the interference is not seen in the TSH assay.
Antiruthenium antibodies are relatively common (prevalence of 0.2% 6 ), and may therefore be an important laboratory and clinical problem. In earlier papers 5,6 it was assumed that Roche overcame this problem in the fT3 assay using a new blocking protein in reagents from 2006. However, our samples were measured between April 2007 and August 2008, suggesting that at least not all problems with antiruthenium antibodies had been solved. Roche claims to have new reagents for fT4 and fT3 assays that are not sensitive to antiruthenium antibodies and will be available soon. In the laboratory constant suspicion for interference is necessary to produce high quality. Moreover, the clinician should look critically at results and discuss these with the laboratory staff if necessary.
DECLARATIONS