BACKGROUND: Huangqi (Astragalus mongholicus), a Chinese herb, has already been included in the "Chinese Pharmacopoeia" for the treatment of ischemic cerebrovascular disease. Secondary injury following brain injury is associated with free radical production, and Huangqi possesses the ability to ameliorate free radical-mediated injury. OBJECTIVE: This study was designed to observe the correlation between anti-free-radical properties of Huangqi and early histological changes of brain tissues following traumatic brain injury. DESIGN, TIME AND SETTING: This study, a randomized, controlled, animal experiment, was performed from May 2006 to June 2007 at the Experimental Center of Science and Technology, School of Basic Science, Liaoning Medical University, Jinzhou City, Liaoning Province, China. MATERIALS: Healthy, adult, Sprague Dawley rats of either gender were included. Huangqi injection was purchased from Heilongjiang Provincial Zhenbaodao Pharmaceutical Co., Ltd., China (National License Medical Number: Z23020781). Na+-K+-adenosine triphosphatase (ATPase), Ca2+-ATPase, and Mg2+-ATPase, as well as kits to measure superoxide dismutase (SOD) activity and malondialdehyde (MDA) content, were purchased from Nanjing Jiancheng Biological Reagent Company, China. METHODS: Seventy-two rats were randomly divided into three groups, with 24 rats in each group: (1) sham-operated group: rats were only exposed, but not injured; (2) model group: brain focal laceration rat models were established by free-falling. These groups were intraperitoneally injected with saline, once every 10 hours; (3) Huangqi group: rats were intraperitoneally injected with 4 mL/kg Huangqi (2 g/mL), once every 10 hours, following brain focal laceration by free-falling. MAIN OUTCOME MEASURES: Ultrastructural changes in brain tissue were observed under an electron microscope 24 hours after injury. The water content of brain tissue was measured using the dry-wet weight method. In addition, the activity of ATPase and SOD, as well as MDA content, was analyzed using biochemical indicators at 4, 24, and 48 hours after injury. RESULTS: All 72 rats were included in the fmal analysis. At 4, 24, and 48 hours after injury, ATPase activity was significantly reduced in the model and Huangqi groups than in the sham-operated group (P 0.05). At 24 and 48 hours after injury, ATPase activity in the Huangqi group gradually decreased, but remained significantly greater than that in the model group (P