目的：评价标准化胰岛素强化治疗对重症脑出血患者预后的影响。方法：451例重症脑出血(GCS评分≤ 8)伴血糖异常升高患者，随机分为胰岛素强化治疗组和常规治疗组，观察治疗30天后病死率。结果：胰岛素强化治疗组和常规治疗组30天的病死率分别是47.79%和41.33%，差异没有统计学意义(P > 0.05)。接受胰岛素强化治疗的226例患者中，GHb升高组病死率明显高于GHb正常组(57.14% vs 40.63%，P = 0.01)。接受胰岛素常规治疗的225例患者中，GHb正常组病死率显著高于GHb升高组(49.65% vs 27.38%，P = 0.001)。结论：重症脑出血患者入院时就应该立即测定GHb，进而制定个体化血糖管理目标和胰岛素治疗策略。Objective: To evaluate the effect of insulin therapy on the prognosis of patients with severe cerebral hemor- rhage. Methods: 451 patients with severe cerebral hemorrhage (GCS score ≤ 8) were randomly divided into two groups, to receive intensive insulin therapy or conventional insulin therapy. The mortality was observed in the 30th day after treatment. Results: No significant difference in the mortality was found between the intensive insulin therapy group and conventional insulin therapy group (P > 0.05). In 226 patients receiving intensive insulin treatment, a higher mortality was observed in patients with higher GHb (57.14% vs 40.63%，P = 0.01). In 225 patients receiving conventional insulin treatment, a higher mortality was found in patients with normal GHb (49.65% vs 27.38%，P = 0.001). Conclusion: GHb should be tested in the patients with severe cerebral hemorrhage at admission, to provide information for adopting individualized glycemic management and setting glycemic control targets.