Plasma free T4 (FT4) concentrations could be increased during hemodialysis in patients with chronic renal failure (CRF) because an increase in non-esterified fatty acids (NEFA) could interfere with the binding of T4 to thyroxine-binding globulin. To evaluate the effect of hemodialysis on the FT4 concentration in patients with CRF, we measured the FT4 in 39 patients with CRF by four assay methods including equilibrium dialysis, the I-125-T4 analog method and enzyme immunoassay. The addition of the fatty acid sodium oleate to normal pooled sera led to a marked increase in FT4 as measured by equilibrium dialysis (Mode! FT4). A moderate increase in the serum FT4 concentration also was observed with an IMX enzyme immunoassay kit, whereas the Coat-A-Count analog method demonstrated no interference by sodium oleate. The mean serum FT4 prior to hemodialysis measured by equilibrium dialysis did not differ significantly from that in the normal control, although those measured by analog methods (Coat-A-Count and Amerlex) and IMX were subnormal. The FT4 by IMX were albumin-dependent, and the values decreased as the samples were serially diluted, but Model FT4 was not affected by the albumin level or the serial dilution. FT4 by Model FT4 showed a marked increase beginning 10 min after the start of dialysis, and it correlated well with the plasma concentration of NEFA and the NEFA/albumin molar ratio. The other three assay methods, including one which is not affected by NEFA, did not show a change in FT4 at 10 min, but a significant increase of 11 to 17% was observed by the end of dialysis. The TSH concentration decreased significantly during hemodialysis. These data suggest that (1) the low serum FT4 in hemodialysis patients measured by some immunoassay methods may be an underestimation due to the low albumin level; (2) FT4 actually increases during hemodialysis due to the actual increase in NEFA, although the marked increase in FT4 during hemodialysis as measured by equilibrium dialysis is an overestimation due to the in vitro generation of NEFA; and (3) one should beware of abberations in thyroid hormone parameters during hemodialysis and potential complications.
