Critical illness is often associated with reduced TSH and thyroid hormone secretion as well as marked changes in peripheral thyroid hormone metabolism, resulting in low serum T(3) and high rT(3) levels. To study the mechanism(s) of the latter changes, we determined serum thyroid hormone levels and the expression of the type 1, 2, and 3 iodothyronine deiodinases (D1, D2, and D3) in liver and skeletal muscle from deceased intensive care patients. To study mechanisms underlying these changes, 65 blood samples, 65 liver, and 66 skeletal muscle biopsies were obtained within minutes after death from 80 intensive care unit patients randomized for intensive or conventional insulin treatment. Serum thyroid parameters and the expression of tissue D1-D3 were determined. Serum TSH, T(4), T(3), and the T(3)/rT(3) ratio were lower, whereas serum rT(3) was higher than in normal subjects (P < 0.0001). Liver D1 activity was down-regulated and D3 activity was induced in liver and skeletal muscle. Serum T(3)/rT(3) ratio correlated positively with liver D1 activity (P < 0.001) and negatively with liver D3 activity (ns). These parameters were independent of the type of insulin treatment. Liver D1 and serum T(3)/rT(3) were highest in patients who died from severe brain damage, intermediate in those who died from sepsis or excessive inflammation, and lowest in patients who died from cardiovascular collapse (P < 0.01). Liver D3 showed an opposite relationship. Acute renal failure requiring dialysis and need of inotropes were associated with low liver D1 activity (P < 0.01 and P = 0.06) and high liver D3 (P < 0.01) and skeletal muscle D3 (P < 0.05) activity. Liver D1 activity was negatively correlated with plasma urea (P = 0.002), creatinine (P = 0.06), and bilirubin (P < 0.0001). D1 and D3 mRNA levels corresponded with enzyme activities (both P < 0.001), suggesting regulation of the expression of both deiodinases at the pretranslational level. This is the first study relating tissue deiodinase activities with serum thyroid hormone levels and clinical parameters in a large group of critically ill patients. Liver D1 is down-regulated and D3 (which is not present in liver and skeletal muscle of healthy individuals) is induced, particularly in disease states associated with poor tissue perfusion. These observed changes, in correlation with a low T(3)/rT(3) ratio, may represent tissue-specific ways to reduce thyroid hormone bioactivity during cellular hypoxia and contribute to the low T(3) syndrome of severe illness.

In critically ill patients who required more than 5 d of intensive care, rT3 and T3/rT3 were already prognostic for survival on d 1. On d 5, T4, T3, but also TSH levels are higher in patients who will survive. Serum rT3 and T3/rT3 were correlated with postmortem tissue deiodinase activities.

In patients with septic shock NTIS tissue responses are orientated to decrease production and increase degradation (muscle) or decrease uptake (adipose tissue) of T3, as well as to decrease thyroid hormone actions.

Our results suggest that, in addition to changes in the hypothalamus-pituitary-thyroid axis, tissue-specific mechanisms are involved in the reduced supply of bioactive thyroid hormone in critical illness.

Preterm infants have low serum T4 and T3 levels, which may partly explain the immaturity of their tissues. Deiodinase enzymes are important in determining the bioavailability of thyroid hormones: deiodinases D1 and D2 convert T4 to T3, whereas deiodinase D3 inactivates T3 and produces rT3 from T4. In human and ovine fetuses, plasma T3 rises near term in association with the prepartum cortisol surge. This study investigated the developmental effects of cortisol and T3 on tissue deiodinases and plasma thyroid hormones in fetal sheep during late gestation. Plasma cortisol and T3 concentrations in utero were manipulated by exogenous hormone infusion and fetal adrenalectomy. Between 130 and 144 d of gestation (term 145+/-2 d), maturational increments in plasma cortisol and T3, and D1 (hepatic, renal, perirenal adipose tissue) and D3 (cerebral), and decrements in renal and placental D3 activities were abolished by fetal adrenalectomy. Between 125 and 130 d, iv cortisol infusion raised hepatic, renal, and perirenal adipose tissue D1 and reduced renal and placental D3 activities. Infusion with T3 alone increased hepatic D1 and decreased renal D3 activities. Therefore, in the sheep fetus, the prepartum cortisol surge induces tissue-specific changes in deiodinase activity that, by promoting production and suppressing clearance of T3, may be responsible for the rise in plasma T3 concentration near term. Some of the maturational effects of cortisol on deiodinase activity may be mediated by T3.