The properties of the endothelium differ between the brain and the remainder of the body. In most non-CNS tissues the size of the junctions between endothelial cells averages 65 A. Proteins do not cross these gaps, while sodium does. In the brain, the junction size is only 7 A, which is too small to allow crossing by sodium. Investigations with changes in osmotic and oncotic pressure have demonstrated that: (1) reducing osmolality results in edema formation in all tissues including normal brain; (2) a decrease in oncotic pressure is only associated with peripheral edema but not in the brain; (3) in case of brain injury, a decrease in osmolality elicits edema in the part of brain which remained normal; (4) similarly, a decrease in oncotic pressure does not cause an increase in brain edema in the injured part of the brain. The determinant factor of water exchange in the brain is mediated through the osmolality and not the oncotic pressure. The use of hypertonic solutions (Ringer lactate or NaCl) for intravascular fluid resuscitation of patients suffering from hypovolemic head trauma has gained popularity. A research survey in regard with this observation can be summarized as follows: NaCl 7.5% (2400 mOsm/l) is becoming the most popular hypertonic solution because of its favorable systemic and cerebral effects. It improves myocardial contractility, precapillary dilatation, and reactive venoconstriction, and it has a plasmatic expansion factor of 3.8. In regard to the brain tissue, it improves the PO2 and the cerebral blood flow (CBF) as a result of decreasing cerebrovascular resistance. Finally, it reduces the cortical water content of intact blood-brain barrier area. The overall consequence is reduction of intracranial pressure (ICP). Although the homeostasis of the cerebral intracellular compartment remains unknown, it is possible that brain cells are able to resist important osmolar overload. NaCl 7.5%/dextran 70.6% is clinically at this moment the most studied hypertonic/hyperoncotic agent in prehospital emergencies. Its effects on cerebral homeostasis are identical to NaCl 7.5%. However, the addition of a colloid agent has the advantage of prolonging the systemic effects without affecting the brain. The plasmatic expansion factor is 4.5, which is slightly superior to NaCl 7.5%. Mannitol improves CBF by maintaining autoregulation as a result of changes in viscosity and reactive cerebrovascular constriction. It generates an osmotic gradient which reduces the cerebral volume and subsequently the ICP. In the presence of a cryogenic cerebral lesion, its reductive effects on brain water are superior to the hypertonic/hyperoncotic solution. Because mannitol has less spectacular systemic responses than the other solutions, it is not indicated for resuscitation following hemorrhagic shock. In conclusion, it is important to note that hypotension and hypoxemia represent the determinant factors of secondary cerebral insults. Therefore, in the presence of patients with head injury and especially hemorrhagic shock, it is essential to ensure a cerebral perfusion pressure (CPP) of > 80 mm Hg. Hypertonic solutions have gained popularity in these clinical situations because of their combined effects on ICP, mean arterial pressure (MAP) and CPP. However, the therapeutic approach to polytraumatized patients with small intravascular volume (4-6 ml/kg) of hypertonic solutions should not be a substitute for the usual volemic resuscitation technique. The clinical indication for these solutions should be limited to the initial resuscitation maneuvers in traumatized patients. Prolonged use of hypertonic solutions for the purpose of intravascular resuscitation would only contribute to increasing the side effects and eventually counteract the initial beneficial advantages.