摘要
目的探讨晚期肝病患者并发真菌感染的早期预警指标。方法100例合并院内真菌感染的晚期肝病患者被纳入观察组,选择同期入院、肝病基础相当、未合并真菌感染的另80例患者为对照组,采用多因素logistic回归分析,评价与发生真菌感染有关的各种危险因素,达到预警作用。结果多因素分析显示,细菌感染(OR=4.45,95%CI1.86-13.68;P=0.006)、广谱抗菌素应用疗程(11~15天疗程,发生真菌感染的OR值为2.02,95%CI1.54-8.50,P=0.005;16~21天疗程,OR=5.36,95%CI2.68-12.31,P=0.003;22天以上疗程,OR=12.88,95%CI6.07-22.40,P=0.001)、糖皮质激素使用(OR=3.78,95%CI1.07-11.23;P=0.007)、中性粒细胞减少(OR:6.62,95%CI1.68~14.05;P=0.002)、糖尿病(OR=2.23,95%CI1.36~9.81;P=0.026)是晚期肝病患者发生真菌感染的重要危险因素。结论细菌感染、广谱抗菌素疗程、糖皮质激素应用、中性粒细胞减少、糖尿病是晚期肝病患者并发真菌感染的主要危险因素,可作为早期预警指标。
Objective To evaluate the early warming signs of nosocomial fungal infections in patients with end - stage liver disease. Methods One hundred patients with end - stage liver disease were included as cases because of nosocomial fungal infections, and another 80 patients with similar liver disease background without fungal infections were admitted as controis. Patients were matched according to the first episode of fungal infections. Results The risk factors that showed a significant association with fngal infections by multivariate analysis were bacteria infection ( OR = 4. 45,95 percent confidence interval, 1.86 - 13.68;P =0. 006), the period of treatment of broad - spectrum antibiotics ( 11 - 15days, the odds ratio of fungal infections 2.02, 95%CI 1.54 -8.50,P=0.005; 16 - 21days, OR =5.36, 95%CI 2.68 - 12.31,P =0.003; more than 22 days, OR = 12. 88, 95% CI 6.07 - 22. 40, P : 0. 001 ), use of corticosteroids ( OR = 3.78, 95% CI1.07 - 11.23 ; P = 0. 015 ) , neutropenia ( OR = 6. 62, 95% CI 1.68 - 14. 05 ; P = 0. 002 ) , and diabetes history ( OR = 2. 23, 95% CI1.36 - 9. 81 ; P =0. 026). Conclusion Bacteria infection, the period of treatment of broad - spectrum antibiotics, corticosteroids, neutropenia, and diabetes history are the risk factors of fungal infections in patients with end - stage liver diseases, and can be the early warming signs.
出处
《临床肝胆病杂志》
CAS
2007年第5期362-364,共3页
Journal of Clinical Hepatology
关键词
肝病
晚期
真菌
感染
危险因素
预警
多因素分析
liver disease
end- stage
fungal infection
risks factors, premonition
multivariate analysis