A randomized prospective study was done to assess the response of hypothyroxinemic patients with severe nonthyroidal illnesses to T₄ therapy. Patients admitted to a medical intensive care unit who had a total serum T₄ concentration less than 5 µg/dl were randomly assigned to a control (12 patients) or a T₄ treatment group (11 patients).L-T₄ in a dose of 1.5 µg/kg was given iv each day for 2 weeks. In the treatment group, serum T₄ and free T₄ concentrations significantly increased by day 3 and were normal on day 5. Serum TSH levels decreased significantly in the T₄ treatment group, as did the TSH response to TRH. A significant rise in serum T₃ occurred in the control group on day 7, but was delayed until day 10 in the treatment group. Mortality was equivalent in the 2 groups (75% control vs. 73% treatment). Regardless of group assignment, survivors and nonsurvivors were completely separable based on baseline T³ to T₄ ratios [17.0 ± 1.8 (± SE) ng/ug in survivors vs. 7.0 ± 0.7 in nonsurvivors; P < 0.001]. Angiotensinconverting enzyme was significantly reduced in the T₄ treatment group, but did not rise significantly in response to treatment.

T₄ therapy was not beneficial in this population of intensive care unit patients, and by inhibiting TSH secretion, it may suppress an important mechanism for normalization of thyroid function during recovery. (J Clin EndocrinolMetab 63: 1, 1986)