The aim of this clinical study was to determine the optimal time-window for surgical treatment of spontaneous intracerebral hemorrhage (ICH). From January 1998 to September 2000, 17 hospitals in Shanghai participated in a prospective randomized controlled trial. Among a consecutive series of 500 patients with spontaneous ICH, 234 underwent medical treatment and 266 patients received surgical treatment. According to the interval from initial onset to treatment, they were divided into 3 stages: ultra-early (< or =7 h), early (7-24 h), and delayed (> 24 h). Perioperative evaluation (Glasgow Outcome Score), long-term outcome (the activities of daily living [ADL] score), mortality, as well as incidence of associated complications were compared respectively. We found that: a) in the ultra-early and early stages, both the perioperative and long-term outcome of surgical treatment was definitely better than medical treatment; b) for the outcome of surgical treatment, there was no significant difference between ultra-early and early stages; c) in ultra-early stage, risk of postoperative rebleeding was significantly higher, and decreased henceforth; d) in delayed stage, incidence of associated respiratory, urinary, and gastrointestinal system complications was higher in surgery group than in medication group. In summary, our study yielded conclusive evidence that the early stage (within 7-24 h) was the optimal time-window for surgical intervention of spontaneous ICH.