Smoke inhalation injury is a major cause of morbidity and mortality in thermally injured individuals. There is evidence of increased oxygen free radical activity, e.g., superoxide, in association with smoke inhalation injury. Because superoxide dismutase converts the reactive superoxide radical to peroxide, we hypothesized that nebulization of manganese superoxide dismutase (Mn-SOD) into the airway might attenuate pulmonary dysfunction secondary to smoke inhalation injury. The present study was designed as a prospective, controlled, and randomized laboratory experiment to determine the effects of aerosolized Mn-SOD on lung fluid balance, as indexed by changes in pulmonary microvascular permeability, lung lymph flow (Q(L)), and gas exchange in an established and clinically relevant ovine model of smoke inhalation injury. Fifteen female Merino sheep were chronically instrumented with a femoral arterial, a Swan-Ganz, and a left atrial catheter. In addition, the right caudal mediastinal lymph node was cannulated to measure Q(L) (mL.h(-1)). Pneumatic occluders were placed around the right pulmonary veins for the determination of the reflection coefficient (sigma). After 7 days of recovery, sheep were randomly allocated to (a) an untreated control group (4 groups of 12 breaths of cotton smoke), (b) an injured group treated with nebulized Mn-SOD (5 mg/kg), and (c) an injured group that received only the vehicle (nebulized saline). Nebulization was performed 1 h and 12 h after smoke inhalation. Mn-SOD nebulization attenuated the increase in both filtration coefficient and sigma and significantly decreased lung tissue conjugated dienes. However, there were no differences in Q(L), PaO2/FiO2 ratio, and bloodless lung wet/dry weight ratio between groups. Although Mn-SOD nebulization attenuated the loss of protein, it failed to improve lung edema and pulmonary gas exchange, thereby limiting its clinical use.