急性重症卒中患者不同原因发热的危险因素分析

The risk factors of various kinds of fever in critically ill patients with acute stroke

目的探讨急性重症卒中患者早期（住院后1周内）发热的危险因素。方法连续入选137例神经重症监护病房（NICU）中的急性重症卒中患者。2次／d测量腋下体温≥37．5℃，或单次测量体温〉37．8℃定为发热。在Commichau等研究的基础上对发热进行分类。应用多因素Logistic回归分析，寻找任何原因发热、明确的感染性发热和不明原因发热的危险因素。结果任何原因发热者共89例，发生率为65．0％。其中明确的感染性发热者有67例，发生率为48．9％；不明原因发热的有21例，发生率为15．3％；明确的非感染性发热1例，发生率为0．7％。多因素Logistic回归分析结果显示，年龄〉65岁、意识障碍、大面积脑梗死／脑出血量≥30ml和深静脉置管，是任何原因发热的临床高危因素；年龄〉65岁、意识障碍、大面积脑梗死／脑出血量≥30ml、深静脉置管和发热前入住NICU天数，是确定的感染性发热的临床高危因素；脑中线移位、入院时白细胞计数〉12．0×10^9／L，是不明原因发热的临床高危因素。结论不同原因导致急性重症卒中患者的发热危险因素不同，应针对性地采取防治措施。

Objective To identify the risk factors of fever in critically ill patients with acute stroke in a neurological intensive care unit （NICU）. Methods The frequency of risk factors of fever in 137 stroked patients during the first 7 days after admission were retrospectively studied. Fever was defined as a patient＇s axillary temperature ≥37.5℃ in two separate determinations or 〉 37.8℃ in one determination. We categorized the fever into 4 kinds ： （1） any fever, （2） explained infectious fever, （3）explained noninfectious fever and （4）unexplained fever based on Commichau＇s methods. Then logistic regression analysis was used to ascertain the risk factors of the 4 kinds of fever. Results Any fever, explained infectious fever, unexplained fever and explained noninfectious fever occurred in 65%, 48. 9%, 15. 3% and 0. 7% respectively. Muhivariable logistic regression analysis demonstrated that patients older than 65 years, impaired consciousness, large area cerebral infarction/cerebral hemorrhage volume≥30 ml, and deep vein catheterization used were involved into regression equation for any fever. In addition to these variables, length of NICU stay was also involved into regression equation for explained infectious fever. Brain midline shift, initial peripheral leucocytes count more than 12. 0 × 10^9/L were involved into regression equation for unexplained fever. Conclusion The risk factors differ in various kinds of fever in critically ill patients with acute stroke. Effective measures should be taken for different types of febrility.