Fann Jesse R, Ribe Anette Riisgaard, Pedersen Henrik Schou, Fenger-Grøn Morten, Christensen Jakob, Benros Michael Eriksen, Vestergaard Mogens
Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, WA, USA.
The Research Unit for General Practice and Section for General Medical Practice, Department of Public Health, Aarhus University, Aarhus, Denmark.
Lancet Psychiatry. 2018 May;5(5):424-431. doi: 10.1016/S2215-0366(18)30065-8. Epub 2018 Apr 10.
Traumatic brain injury (TBI) has been associated with increased risk of dementia; however, large-scale studies with long follow-up have been scarce. We investigated the association between TBI, including severity and number of TBIs, and the subsequent long-term risk of dementia.
We did a nationwide population-based observational cohort study in Denmark using information on citizens from national registries. We used the Danish Civil Registration System to establish a population-based cohort consisting of all people born in Denmark who were living in the country on Jan 1, 1995, and who were at least 50 years old at some point during follow-up (between 1999 and 2013). We obtained information on TBIs from the Danish National Patient Register (NPR), and obtained information on dementia by combining data recorded in the NPR, the Danish Psychiatric Central Register, and the Danish National Prescription Registry (DNPR). The long-term risk of dementia after TBI was established using survival analysis. We used three prespecified models for each of the three analyses: different time periods since the TBI, multiple TBIs, and sex. The first model adjusted for sociodemographic factors, the second model added medical and neurological comorbidities, and the third added psychiatric comorbidities.
We used data from a cohort of 2 794 852 people for a total of 27 632 020 person-years (mean 9·89 years per patient) at risk of dementia. 132 093 individuals (4·7%) had at least one TBI during 1977-2013, and 126 734 (4·5%) had incident dementia during 1999-2013. The fully adjusted risk of all-cause dementia in people with a history of TBI was higher (hazard ratio [HR] 1·24, 95% CI 1·21-1·27) than in those without a history of TBI, as was the specific risk of Alzheimer's disease (1·16, 1·12-1·22). The risk of dementia was highest in the first 6 months after TBI (HR 4·06, 3·79-4·34) and also increased with increasing number of events (1·22, 1·19-1·25 with one TBI to 2·83, 2·14-3·75 with five or more TBIs). Furthermore, TBI was associated with a higher risk of dementia (1·29, 1·26-1·33) in people with TBI than in individuals with a non-TBI fracture not involving the skull or spine. The younger a person was when sustaining a TBI, the higher the HRs for dementia when stratified by time since TBI.
TBI was associated with an increased risk of dementia both compared with people without a history of TBI and with people with non-TBI trauma. Greater efforts to prevent TBI and identify strategies to ameliorate the risk and impact of subsequent dementia are needed.
Lundbeck Foundation.
创伤性脑损伤（TBI）与痴呆风险增加有关；然而，长期随访的大规模研究较少。我们调查了TBI（包括严重程度和TBI次数）与随后发生痴呆的长期风险之间的关联。
我们在丹麦进行了一项基于全国人口的观察性队列研究，使用来自国家登记处的公民信息。我们利用丹麦民事登记系统建立了一个基于人群的队列，该队列由1995年1月1日居住在丹麦、在随访期间（1999年至2013年）某个时间点至少50岁的所有丹麦出生的人组成。我们从丹麦国家患者登记处（NPR）获取TBI信息，并通过合并NPR、丹麦精神病学中央登记处和丹麦国家处方登记处（DNPR）记录的数据来获取痴呆信息。使用生存分析确定TBI后痴呆的长期风险。我们针对三项分析中的每一项使用了三个预先设定的模型：TBI后的不同时间段、多次TBI和性别。第一个模型对社会人口学因素进行了调整，第二个模型增加了医学和神经科合并症，第三个模型增加了精神科合并症。
我们使用了一个由2794852人组成的队列的数据，这些人共有27632020人年（每位患者平均9.89年）处于痴呆风险中。132093人（4.7%）在1977 - 2013年期间至少有一次TBI，126734人（4.5%）在1999 - 2013年期间发生了新发痴呆。有TBI病史的人全因痴呆的完全调整风险（风险比[HR]1.24，95%CI 1.21 - 1.27）高于无TBI病史的人，阿尔茨海默病的特定风险也是如此（1.16，1.12 - 1.22）。痴呆风险在TBI后的前6个月最高（HR 4.06，3.79 - 4.34），并且随着事件数量的增加而增加（一次TBI时为1.22，1.19 - 1.25；五次或更多次TBI时为2.83，2.14 - 3.75）。此外，与未涉及颅骨或脊柱的非TBI骨折患者相比，TBI患者发生痴呆的风险更高（1.29，1.26 - 1.33）。TBI发生时年龄越小，按TBI后时间分层时痴呆的HR越高。
与无TBI病史的人和有非TBI创伤的人相比，TBI与痴呆风险增加有关。需要做出更大努力来预防TBI，并确定改善后续痴呆风险和影响的策略。
伦贝克基金会。
