Tri-iodothyronine (T3) therapy in a pre-clinical model of septic shock.

Abstract

Thyroid hormone is essential for normal organ function. Tri-iodothyronine (T3) is the most active form of thyroid hormone, derived from the deiodination of the more abundant thyroxine (T4). T3 is considered to have prominent haemodynamic and metabolic effects. During illness, blood levels of T3 decline with a reciprocal increase of the inactive reverse-T3 and eventually, a fall of T4. This phenomenon is referred to as Non- Thyroidal Illness Syndrome (NTIS) and the extent of change in circulating thyroid hormones is proportional to severity of disease and survival. NTIS is particularly marked during sepsis. Sepsis is the most common diagnosis of patients requiring emergency admission to an Intensive Care Unit (ICU) and mortality rates remain high despite provision of all supportive therapies. Given the importance of T3 for normal function and the relationship between low T3 and poor outcome, NTIS may contribute to the multi-organ dysfunction of sepsis. Restoring T3 levels during sepsis may be beneficial but has received little attention. Concerns that NTIS may be an adaptive response and that T3 supplementation may provoke thyrotoxicity have limited the conduct of clinical trials in patients with septic shock. There is also uncertainty regarding the need to co-administer hydrocortisone (HC) with T3. Consequently, a pre-clinical study was required to test the safety and efficacy of T3 therapy with and without HC. An ovine model of septic shock was developed, applying many of the supportive care elements provided to humans in an ICU. Following a bolus of intravenous E.coli, sheep received 24 hours of protocol-guided sedation, ventilation, parenteral fluids and noradrenaline (NorA) infusion. The model was validated over time and replicated much of the human septic response, including NTIS. Following pharmaceutical and dose finding studies, a randomised, blinded, placebo-controlled trial of T3 with and without HC, was conducted in the ovine model. After two hours of sepsis, 32 sheep received a 24-hour infusion of: (i) T3 + placebo, (ii) HC + placebo, (iii) T3 + HC, or (iv) placebo + placebo. The primary outcome was the total amount of NorA required during the infusion of study drugs; while the secondary outcomes included haemodynamic, metabolic and parameters of organ function. Plasma T3 levels fell in placebo animals and were increased to supraphysiological concentrations by T3 infusion. The amount of NorA required was no different between the study groups (mean ± SEM μg/kg; T3 group, 501 ± 131; T3 + HC group, 466 ± 175; HC group, 167 ± 101; placebo group, 208 ± 160; p = 0.20). There was no significant treatment effect on any haemodynamic variable, temperature, pH, lactate or oxygen extraction. The same dose of T3 was subsequently tested in a group of non-septic sheep. Despite supra-physiological plasma levels, there was no change to any physiological endpoint. In conclusion, a 24-hour infusion of T3 (with or without HC) in an ovine model of septic shock did not reduce NorA requirements nor alter any other measured physiological variable. Acute T3 replacement appears to be safe, but the role of this therapy for intractable septic shock remains uncertain.