To compare lung injury induced by a hemorrhagic shock resuscitated with normal saline or with small volumes of a hypertonic/hyperoncotic solution.

Randomized, controlled, laboratory study in an animal research laboratory.

Nineteen pigs (43 ± 4 kg).

After anesthesia and mechanical ventilation animals were bled to induce a 2-h deep shock and resuscitated for 2 h using normal saline (NS, 2 ml/kg per minute, n = 7) or the association of 7.2% NaCl with 6% hydroxyethylstarch 200/0.5 (HSHES, 4 ml/kg in 10 min followed by 0.2 ml/kg per minute, n = 7) to reach cardiac index and mixed venous oxygen saturation goals. Lungs were removed 6 h after the initiation of hemorrhage. Five animals were used as controls without hemorrhage.

Resuscitation goals were achieved using 90 ± 17 ml/kg NS or 6.8 ± 1.9 ml/kg HSHES. Lung injury was noted in both hemorrhage groups but was not influenced by the type of resuscitation. Extravascular lung water was measured at 9.6 ± 1.8 ml/kg in the NS group, 9.2 ± 1.6 ml/kg in the HSHES, group and 6.4 ± 1 m/kg in the control group. The degree of histological alveolar membrane focal thickening and interstitial neutrophil infiltration were significantly more pronounced in the hemorrhage groups with no difference between the two types of fluid loading. Finally, pulmonary levels of IL-8 were higher after hemorrhage regardless of the type of resuscitation.

When included in an optimized and goal directed resuscitation, the use of normal saline or a small volume of hypertonic/hyperoncotic solution does not result in a different early hemorrhage-induced lung injury.