Endocrine Abstracts

Introduction: Physicians in our hospital notified the laboratory staff of a number of patients at the outpatient clinic with increased free T4 (fT4) concentrations without (complete) suppression of thyroid stimulating hormone (TSH). This phenomenon appeared to occur more frequently following implementation of a new automated fT4 immunoassay. The discordance between fT4 and TSH concentrations may be explained by analytical issues (not further explained here), incorrect reference intervals, or patient-related factors (e.g. medication, population). We aimed to establish the contribution of the possible factors involved.

Methods: First, reference intervals of the current Cobas (Roche) and former Delfia (Perkin Elmer) fT4 immunoassay were re-evaluated using blood samples of healthy volunteers. Second, TSH (Cobas, Roche) and fT4 requests and the frequency of discordant pairings (i.e. fT4 above the upper limit of normal and TSH 0.02− 5.0 mU/l) of patients from Amsterdam UMC were retrospectively analysed using a Delfia fT4 and Cobas fT4 cohort. Third, we performed a literature search to assess whether time of blood draw and time of levothyroxine (L-T4) ingestion may contribute to higher fT4 concentrations in L-T4 users.

Results: The original reference intervals belonging to the Delfia and Cobas assay were confirmed. The Delfia (n=176, 5.5%) and Cobas cohort (n=295, 8.6%) showed comparable frequencies of discordance. Interestingly, approximately 80% of the discordant results belonged to L-T4 users. Review of the literature showed that fT4 concentrations may vary depending on time of blood draw and, therefore, time of L-T4 intake. Besides, fT3/fT4 ratios are different in L-T4 users vs healthy controls and indicate an adapted regulation of the thyroid axis in those patients.

Conclusion: Discordance between fT4 and TSH concentrations was not related to the introduction of a new fT4 immunoassay. The increased fT4 concentrations with discordant TSH could not be explained by analytical issues or incorrect reference intervals, but may be explained by L-T4 intake. Physicians and laboratory specialists should be aware that patients treated with L-T4 may have fT4 concentrations above the reference interval with normal or slightly decreased TSH concentrations, to avoid questioning the assay’s performance, or worse, adapting L-T4 dose in patients.