Abstract
Background: Patients being treated for hypothyroidism with levothyroxine are typically monitored by measuring serum thyroid stimulating hormone (TSH) and if needed, free thyroxine (FT4) concentrations. However, FT4 reference intervals (RIs) established using healthy subjects may not be applicable to patients taking synthetic thyroid hormone. Establishing method-specific FT4 RIs for patients on thyroxine replacement would be valuable. Objective: The objectives were to: 1) Compare central 95(th) percentile (%) ranges for serum FT4 concentrations in patients with no thyroid disease and normal TSH with ranges in patients prescribed levothyroxine with normal TSH and 2) Determine if there are differences in serum FT4 concentrations between patients taking levothyroxine with normal TSH who have had Hashimoto’s disease, thyroid lobectomy, or thyroidectomy. Methods: One year of TSH and FT4 results reported on the same day for individual patients were extracted from the electronic medical record (10/1/2017-10/1/2018, n=27,534 unique patients). Patients with TSH results outside of the RI (n=7,082), hospitalized patients (n=4,833) and pregnant patients (n=223) were excluded. Results from patients without documented thyroid disease for whom thyroid testing was ordered due to symptoms suggestive of thyroid dysfunction (no thyroid disease (NTD) patients, n=6,774; male=39% and female=61%) and those from patients taking levothyroxine for established thyroid disease (levothyroxine-treated (LT) patients, n=4,294; male=21% and female=79%) were analyzed. The LT group was subsequently divided into patients with: Hashimoto’s disease (n=150), thyroid lobectomy (n=128) and thyroidectomy (n=381). TSH (RI: 0.3-4.2 mIU/L) and FT4 (RI: 0.9-1.7 ng/dL) tests were performed on Roche Cobas 8000 immunoassays. Central 95(th) percentile (%) was determined using non-parametric analysis. Results: FT4 values for the NTD and LT patients had a median(interquartile range) of 1.2(1.1-1.3) ng/dL and 1.4(1.2-1.6) ng/dL, respectively. The central 95(th) % for FT4 was 0.9-1.6 ng/dL for NTD and 0.9-2.0 ng/dL for LT patients using the Roche FT4 assay. In LT patients with Hashimoto’s disease, thyroid lobectomy, or thyroidectomy, the median (interquartile range) FT4 values were 1.4(1.2-1.6) ng/dL, 1.5(1.3-1.7) ng/dL, and 1.6(1.4-1.8) ng/dL, respectively. The central 95(th) % for FT4 in LT patients was 0.8-1.9 ng/dL for Hashimoto’s disease, 0.8-2.3 ng/dL for thyroid lobectomy and 1.0-2.1 ng/dL for thyroidectomy patients. Conclusion: In levothyroxine-treated patients with normal TSH, the FT4 results are shifted higher compared to the population without thyroid disease. Awareness of this finding should prevent inadvertent changes in levothyroxine dose in patients whose TSH is at goal.