摘要： 目的 总结北京市通州区颅脑创伤随社会经济快速发展的变化特点，以为调整疾病的预防策略和高危人群的治疗策略提供依据。方法 计算机自动提取首都医科大学附属北京潞河医院电子病历系统中2004年12月1日至2018年10月22日记录的颅脑创伤患者的病历首页，回顾分析其临床特点，单因素和多因素前进法Logistic回归分析筛查出院时病死相关危险因素。结果 共筛选出4058例颅脑创伤患者，出院时生存3683例（90.76%），病死375例（9.24%）。不同年龄（χ²=51.526，P=0.000）、不同致伤原因（χ²=82.349，P=0.000）、不同治疗方法（χ²=186.270，P=0.000）、不同住院时间（χ²=9.842，P=0.000）出院时病死率差异具有统计学意义，其中，45~64、65~74和≥ 75岁患者病死率高于15~24和25~44岁（均P<0.01），≥ 75岁患者病死率亦高于45~64岁（P=0.001）；交通事故伤病死率高于摔伤（P=0.000），交通事故伤、摔伤、其他致伤原因病死率高于暴力打击伤（均P=0.000）；接受神经外科手术和非神经外科手术患者病死率高于保守治疗（均P=0.000）；住院时间1~14 d患者病死率高于15~30 d和>30 d（均P=0.000）。Logistic回归分析显示，45~64岁（OR=2.285，95% CI：1.428~3.657；P=0.001）、65~74岁（OR=4.025，95% CI：2.334~6.941；P=0.000）、≥ 75岁（OR=5.757，95% CI：3.317~9.991；P=0.000），2007-2010年入院（OR=1.856，95% CI：1.376~2.504；P=0.000），接受神经外科手术（OR=6.465，95% CI：4.991~8.376；P=0.000）和非神经外科手术（OR=3.389，95% CI：1.828~6.283；P=0.000），住院时间1~14 d （OR=6.023，95% CI：3.822~9.490；P=0.000）是颅脑创伤患者出院时病死的危险因素，而暴力打击伤是出院时病死的保护因素（OR=0.245，95% CI：0.112~0.537；P=0.000）。结论 颅脑创伤病死率仍较高，45岁以上、2007-2010年入院、交通事故伤、接受神经外科手术和非神经外科手术、住院时间1~14 d的患者出院时病死率较高。

关键词: 脑损伤,创伤性, 医院信息系统, 死亡率, 危险因素, Logistic模型

Abstract: Objective To summarize the characteristics and changes of traumatic brain injury (TBI) in Tongzhou district of Beijing with the rapid development of social economy, in order to provide the theoretical basis for adjustment of the prevention and treatment strategies in high risk population. Methods The home pages of electronic health record (EHR) of patients with TBI from December 1, 2004 to October 22, 2018 were automatically extracted from the EHR of Beijing Luhe Hospital, Capital Medical University. The clinical characteristics were retrospectively analyzed, and the risk factors related to death at discharge were screened by univariate and multivariate forward Logistic regression analysis. Results A total of 4058 patients with TBI were screened out, of which 3683 cases (90.76%) survived and 375 cases (9.24%) died at discharge. The mortality at discharge was significantly different among different age (χ²=51.526, P=0.000), causes of injury (χ²=82.349, P=0.000), treatment modalities (χ²=186.270, P=0.000) and hospital length (χ²=9.842, P=0.000). The mortality of patients aged 45-64, 65-74 and ≥ 75 years was higher than that of patients aged 15-24 and 25-44 years (P<0.01, for all), and the mortality of patients aged ≥ 75 years was higher than that of patients aged 45-64 years (P=0.001). The mortality of patients with traffic accident injury was higher than that of fall injury (P=0.000), and the mortality of patients with traffic accident injury, fall injury and other cases was higher than that of violent injury (P=0.000, for all). The mortality of patients received neurosurgery and non-neurosurgery was higher than that of conservative treatment (P=0.000, for all). The mortality of patients with hospital lengh of 1-14 d was higher than that of 15-30 d and > 30 d (P=0.000, for all). Logistic regression analysis showed that 45-64 years old (OR=2.285, 95%CI:1.428-3.657; P=0.001), 65-74 years old (OR=4.025, 95%CI:2.334-6.941; P=0.000), ≥ 75 years old (OR=5.757, 95%CI:3.317-9.991; P=0.000), hospital admission in year 2007-2010 (OR=1.856, 95%CI:1.376-2.504; P=0.000), treatment with neurosurgery (OR=6.465, 95%CI:4.991-8.376; P=0.000) and non-neurosurgical procedures (OR=3.389, 95%CI:1.828 to 6.283; P=0.000), hospital length 1- 14 d (OR=6.023, 95%CI:3.822-9.490; P=0.000) were risk factors for death at discharge of TBI patients, and violent injury was a protective factor for death at discharge (OR=0.245, 95%CI:0.112-0.537; P=0.000). Conclusions The mortality of TBI is still high, patients over 45 years, hospital admission in year 2007-2010, traffic accident injury, treatment with neurosurgery and non-neurosurgery, and hospital length of 1-14 d had a higher mortality at discharge.

Key words: Brain injuries, traumatic, Hospital information systems, Mortality, Risk factors, Logistic models