Résumé : We challenged the current management of uncontrolled haemorrhagic shock (UHS) and put forward a hypothesis that therapeutic mild hypothermia combined with delayed fluid resuscitation will improve the survival rate. After an initial blood withdrawal of 3 ml/100 g for 15 min, the rat's tail was amputated up to 75% to induce UHS phase I. The mean arterial blood pressure (MAP) was maintained at 40 mmHg or 80 mmHg, according to the assigned study group. This was followed by homeostasis of the tail wound and increase of the MAP up to 100 mmHg during resuscitation phase II. Finally, phase III was an observation of phase up to 72 h. Rats were anaesthetised and randomised into four groups. Group 1 received immediate fluid resuscitation and normothermia. Group 2 received immediate fluid resuscitation and therapeutic mild hypothermia. Group 3 received limited fluid solutions to maintain MAP at 40 mmHg and normothermia. Group 4 also received limited fluid solution, but the rats were subjected to therapeutic mild hypothermia. In groups 2 and 4, the body temperature was kept at 34°C throughout the UHS phase I and resuscitation phase II. At the end of the observation phase III, the brains of the animals were fixed and analysed histologically. The blood loss from the tail during the UHS phase I was significantly higher in groups 1 and 2. The survival rate was 33.3, 83.3, 58.3 and 91.7%, respectively in groups 1-4. In all surviving rats, no histological brain damage was observed. These results indicate that therapeutic mild hypothermia or delayed fluid resuscitation increase the survival rate in this model. However, when mild hypothermia and limited fluid resuscitation were combined, the survival rate was the highest. © 2004 Elsevier Ireland Ltd. All rights reserved.