Abstract
The study by Chon et al. utilizes pharmacokinetic (PK) data to estimate the impact of co-ingesting 2% milk on the bioavailability (BA) of levothyroxine (LT4) tablets (1). The authors state that they followed a modified Food and Drug Administration (FDA) protocol for measuring the relative BA of the orally ingested LT4. From the description of statistical methods and the appearance of figure 1, it is not apparent that a correction for baseline endogenous total thyroxine (TT4) has been utilized in the calculation of the areas under the curves (AUCs) used to make this comparison as the FDA now stipulates (2). The 8.4% difference in AUC is consistent with the authors' premise. The sensitivity of FDA-mandated PK measurements (3) is greatly enhanced by baseline TT4 correction (4), but uncorrected AUCs are very insensitive and of uncertain clinical significance. I would simply ask the authors to clarify if a baseline correction for TT4 was made in calculating their PK data.