Response to Heijboer et al. Ann Clin Biochem 2009;46(3):263–4

Dear Editor,

Heijboer et al. ¹ highlighted the potential interference in the Roche free thyroxine (fT4) assay (Roche Diagnostics GmbH, Mannheim, Germany) resulting from ruthenium antibodies. Our laboratory performs approximately 100,000 thyroid profiles per annum. In 2005 we reported to Roche our concerns regarding the high number of increased fT4 results with normal thyroid-stimulating hormone (TSH) concentrations we were observing in euthyroid patients, with fT4 concentrations being normal when repeated using an Auto Delfia assay (Wallac Oy, Turku, Finland) in a neighbouring laboratory. In one case, serum was referred to Roche for further investigation. Roche reported fT4 concentrations of 17.1 and 24.3 pmol/L with and without a blocking reagent, respectively, confirming the presence of interference against ruthenium.

Between 2005 and June 2007, we identified 11 cases of falsely elevated fT4 concentrations in euthyroid patients. Due to this, in collaboration with Roche, we installed the TOSOH AIA360 analyser (Tosoh Corp., Tokyo, Japan) in June 2007 to confirm fT4 concentrations >22 pmol/L locally. Between June 2007 and October 2008 (introduction of lot 151161), we identified a further 33 fT4 results that were significantly elevated compared with the TOSOH assay.

We posted our concerns regarding falsely raised fT4 concentrations on the ACB mail base in February 2007 (Inc 017967 and 017509 found at https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0702&L=ACB-CLIN-CHEM-GEN&P=R20296) and were surprised that few other laboratories seemed to be experiencing the same problem. Roche responded to global customer complaints with the release, in 2008, of a new generation of their fT4 assay with reduced susceptibility to antiruthenium interference. Subsequently, we have detected only one falsely increased fT4. The sample in question came from a female patient and was analysed during the period of evaluation of the new generation fT4 assay. The fT4 results obtained by the old generation (lot 150353) and new generation assays (lot 151161) were 42.6 and 40.3 pmol/L, respectively, with a TSH of 1.28 mIU/L. Roche confirmed these findings and noted that lot 151161 resulted in only a slight decrease in the measured fT4 concentration. Analysis with a research assay using a modified conjugate produced an fT4 concentration in the reference range. It was concluded that the sample contained sufficiently high levels of the interfering factor to ruthenium to swamp the blocking agent in the latest generation fT4 assay.

In conclusion, we agree with Heijboer and colleagues that the incidence of antiruthenium antibodies is relatively common. From our experience the introduction of the new generation fT4 assay by Roche has significantly reduced, although as our most recent case indicates not eliminated, susceptibility to interference from antibodies. We therefore echo Heijboer and colleagues call for constant vigilance for interference in thyroid assays by both the laboratory and the clinician.

DECLARATIONS

Competing interests: None.

Guarantor: PS.

Contributorship: DMCK wrote the manuscript, DT retrieved records and data, DMCK, DT and PS contributed to discussion, reviewed and edited manuscript.

Acknowledgements: We thank Roche Diagnostics for measuring interference against ruthenium label, and our colleagues in the Royal Victoria Hospital, Belfast for measuring fT4 by the Auto Delfia assay.