Supplementary Material for: Cause of Death after Traumatic Brain Injury: A Population-Based Health Record Review Analysis Referenced for Nonhead Trauma

Introduction: Traumatic brain injury (TBI) is a leading cause of disability and is associated with decreased survival. Although it is generally accepted that TBI increases risk of death in acute and postacute periods after injury, causes of premature death after TBI in the long term are less clear. Methods: A cohort sample of Olmsted County, Minnesota, residents with confirmed TBI from January 1987 through December 1999 was identified. Each case was assigned an age- and sex-matched non-TBI referent case, called regular referent. Confirmed TBI cases with simultaneous nonhead injuries were identified, labeled special cases. These were assigned 2 age- and sex-matched special referents with nonhead injuries of similar severity. Underlying causes of death in each case were categorized using death certificates, International Classification of Diseases, Ninth Revision, International Statistical Classification of Diseases, Tenth Revision, and manual health record review. Comparisons were made over the study period and among 6-month survivors. Results: Case-regular referent pairs (n = 1,257) were identified over the study period, and 221 were special cases. In total, 237 deaths occurred among these pairs. A statistically significant difference was observed between total number of deaths among all cases (n = 139, 11%) and regular referents (n = 98, 8%) (p = 0.006) over the entire period. This outcome was not true for special cases (32/221, 14%) and special referents (61/441, 14%) (p = 0.81). A greater proportion of deaths by external cause than all other causes was observed in all cases (52/139, 37%) versus regular referents (3/98, 3%) and in special cases (13/32, 41%) versus special referents (5/61, 8%) (p < 0.001 for both). Among all case-referent pairs surviving 6 months, no difference was found between total number of deaths (p = 0.82). The underlying cause of death between these 2 groups was significantly different for external causes only (p < 0.01). For special cases surviving 6 months versus special referents, no difference was observed in total number of deaths (p = 0.24) or underlying causes of death (p = 1.00) between groups. Discussion/Conclusion: This population-based case-matched referent study showed that increased risk of death after TBI existed only during the first 6 months after injury, and the difference was due to external causes.

引言：创伤性脑损伤（Traumatic brain injury, TBI）是导致残疾的主要病因之一，且与生存率降低密切相关。尽管学界普遍认为，创伤性脑损伤会增加损伤急性期及伤后亚急性期的死亡风险，但长期来看，创伤性脑损伤患者过早死亡的具体诱因仍不甚明确。
方法：本研究纳入1987年1月至1999年12月期间，美国明尼苏达州奥尔姆斯特德县的确诊创伤性脑损伤居民作为队列样本。为每例病例匹配年龄与性别均匹配的非创伤性脑损伤对照，命名为常规对照。同时纳入合并同期非头部损伤的确诊创伤性脑损伤病例，标记为特殊病例，并为每例特殊病例匹配2例年龄、性别匹配且非头部损伤严重程度相似的特殊对照。通过死亡证明、《国际疾病分类第九版》（International Classification of Diseases, Ninth Revision, ICD-9）、《国际疾病分类第十版》（International Statistical Classification of Diseases, Tenth Revision, ICD-10）以及手动查阅健康档案，对所有研究对象的根本死亡原因进行分类。随后在整个研究周期内，以及针对伤后存活6个月的亚组开展对比分析。
结果：本研究共纳入病例-常规对照配对样本1257对，其中221例为特殊病例。所有配对样本中共发生237例死亡。全研究周期内，所有创伤性脑损伤病例组的总死亡数为139例（占比11%），显著高于常规对照组的98例（占比8%），差异具有统计学意义（p=0.006）。但该组间差异未在特殊病例组（32/221，14%）与特殊对照组（61/441，14%）中出现（p=0.81）。与常规对照组（3/98，3%）相比，所有创伤性脑损伤病例组因外部因素导致的死亡占比更高（52/139，37%）；特殊病例组（13/32，41%）的外部因素死亡占比亦显著高于特殊对照组（5/61，8%），两组对比差异均具有统计学意义（均p<0.001）。在伤后存活6个月的配对样本亚组中，两组总死亡数无显著差异（p=0.82）；仅在外部因素导致的死亡构成比上存在显著组间差异（p<0.01）。针对存活6个月的特殊病例组与特殊对照组，无论是总死亡数（p=0.24）还是根本死亡原因构成（p=1.00），均未观察到显著组间差异。
讨论与结论：这项基于人群的病例配对对照研究表明，创伤性脑损伤患者仅在伤后前6个月存在死亡风险升高的情况，该差异主要由外部因素导致的死亡所贡献。