It is well known that serum T3 and T4 concentrations are often decreased in patients with nonthyroidal illnesses (NTI). The purpose of this study was to elucidate the possible role of serum thyroid hormone binding inhibitor (THBI) in lowering serum T4 in patients with NTI using the originally developed method for measuring THBI. 125I.T4 and 0.05 micromilligram of pooled normal serum was applied to the Sephadex G.25 column equilibrated in advance by 0.1N NaOH, and 4 micromilligram of 0.075M veronal buffer, pH 8.6 (Buffer) was added. 125I.T4 thus adsorbed on the column was then eluted by adding the charcoal treated thyroxine-free serum (used as thyroxine binding protein (TBP) solution) with THBI which had been extracted from the serum by diethyl ether or standard sodium oleate solution (0.125 approximately 1.0 mumol/column). When the radioactivity of the eluate was counted, the gradual decrease in radioactivity was observed with increase in sodium oleate due to the inhibition of 125I.T4 and TBP binding. Thus, THBI in serum could be expressed as mumol sodium oleate equivalent from the dose response curve for binding inhibition. Coefficients of intrassay and interassay variations in this assay were 6 and 17%, respectively when sera from patients with acute myocardial infarction (AMI) were measured. THBI from 15 normal controls ranged from 0 to 0.27 (mean +/- SD: 0.11 +/- 0.08) mumol sodium oleate equivalent, the normal range being less than 0.27 mumoles. In 12 patients with AMI and 12 with end-stage malignancy (cancer), serum T4 levels were below the normal range in 9 and 8 cases, respectively. In these patients, serum THBI was positive in as many as 10 (83%) and 11 (92%) cases, respectively. Serum T4 and THBI were serially measured after admission in patients with AMI. Serum T4 concentrations showed their lowest values on the first, second or third hospital day and they gradually increased to the normal range by the seventh day. On the other hand, serum THBI showed their highest value on the first hospital day in 8 cases of AMI, and they gradually decreased thereafter. A statistically significant negative correlation was observed between serum T4 and THBI in these patients (r = -0.63, p less than 0.01). Moreover, statistically significant positive correlations were observed between serum THBI and FFA in both AMI and cancer patients (r = 0.89, p less than 0.01; r = 0.88, p less than 0.01, respectively).(ABSTRACT TRUNCATED AT 400 WORDS)