We compared canine systemic and cerebral hemodynamics after resuscitation from hemorrhagic shock with 4 mL/kg (a volume approximating 12% of shed blood volume) of 7.2% saline (HS; 1233 mEq/L sodium), 20% hydroxyethyl starch (HES) in 0.8% saline, or a combination fluid consisting of 20% hydroxyethyl starch in 7.2% saline (HS/HES). Eighteen endotracheally intubated mongrel dogs (18-24 kg) were ventilated to maintain normocarbia with 0.5% halothane in nitrous oxide and oxygen (60:40). After a 30-min period of hemorrhagic shock (mean arterial blood pressure = 40 mm Hg), extending from time T0 to T30, animals received one of three randomly assigned intravenous resuscitation fluids: HS, HES, or HS/HES. Data were collected at baseline, at the beginning and end of the shock period (T0 and T30), immediately after fluid infusion (T35), and at 60-min intervals for 2 h (T95, T155). After resuscitation, mean arterial blood pressure and cardiac output increased similarly in all groups, but failed to return to baseline. Intracranial pressure decreased during shock and increased slightly, immediately after resuscitation in all groups. During shock, cerebral blood flow (cerebral venous outflow method) declined in all groups. After resuscitation, cerebral blood flow increased, exceeding baseline in the HS and HS/HES groups but remaining low in the HES group (P less than 0.05 HS vs HES at T35). We conclude that small-volume resuscitation (4 mL/kg) with HS, HS/HES, or HES does not effectively restore or sustain systemic hemodynamics in hemorrhaged dogs. In dogs without intracranial pathology, the effects on cerebral hemodynamics are also comparable, except for transiently greater cerebral blood flow in the HS group in comparison with the HES group.