We tested the hypothesis that full or ``standard resuscitation'' (SR) with lactated Ringer's solution (LRS) results in increased bleeding in uncontrolled hemorrhagic shock, compared with a ``limited prehospital resuscitation'' (LPR) regimen and a control group of ``no resuscitation'' (NR). Cardiac output was used as physiological endpoint for resuscitation. Twenty swine had 25 mL/kg of blood withdrawn during a 30-minute controlled hemorrhage, followed by a 20-minute uncontrolled hemorrhage (5 mm aortotomy). A 20-minute ``prehospital'' resuscitation regimen was conducted in three groups: the SR group (n = 6), LRS infused as needed to restore cardiac index (CI) to 100% baseline; the LPR group (n = 8), with resuscitation using LRS to 60% of baseline CI, with volume limited to 10 mL/kg; and the NR group (n = 6). After aortotomy repair, intraoperative resuscitation was continued for 120 minutes using LRS to achieve and maintain 80% of baseline mean arterial pressure. Blood pressure and cardiac index were greatly reduced, to 34% and 39% of baseline, respectively, by hemorrhage. During prehospital resuscitation, the SR group required 48.8 +/- 6.5 mL/kg of LRS, whereas the LPR group received 9.4 +/- 0.6 mL/kg (p < 0.05). Mean arterial pressure increased in all three groups during prehospital resuscitation (p < 0.05). Pulse pressures increased in the SR and LPR groups only (p < 0.05). The increment in oxygen delivery was significantly greater in the SR group, compared with the LPR group (p < 0.05), which in turn was significantly greater than the NR group (p < 0.05). Peritoneal blood volume was significantly higher in the SR group (20.6 +/- 5.6 mL/kg), versus the LPR (7.3 +/- 1.3 mL/kg; p < 0.05) and NR groups (3.0 +/- 0.9 mL/kg; p < 0.05). Crystalloid and whole blood requirements during the intraoperative resuscitation phase were significantly higher in the SR group (193 +/- 16.0 and 9.0 +/- 2.5 mL/kg), than in LPR (111.8 +/- 15.6 and 4.5 +/- 1.8 mL/kg; p < 0.05) and NR groups (128.5 +/- 32.3 and 3.9 +/- 2.3 mL/kg; p < 0.05). In the presence of uncontrolled hemorrhagic shock, LPR and NR can significantly reduce internal hemorrhage and subsequent intraoperative crystalloid and blood requirements.