摘要
目的分析真菌性败血症的发病情况、临床特点、易患因素,并在此基础上探讨本病的死亡危险因素,利用有统计意义的因素建立数学模型,以期对真菌性败血症的预后进行预测,利于指导临床治疗。方法对我院近17年确诊的91例真菌性败血症患者临床资料进行回顾性分析,并将其分为治愈和(或)显效组、死亡组,先后以单因素及Logistic多元回归考察死亡危险因素,进一步建立Fisher判别方程对本病死亡率进行预测。结果真菌性败血症易患因素多、病死率高,91例中治愈和(或)显效60例,死亡31例。单因素分析显示患者高龄(≥60岁)、合并细菌感染、感染性休克、多器官功能衰竭(MODS)、属ICU患者、使用糖皮质激素、外科术后、化疗、增强免疫药物、抗真菌治疗前的住院时间、抗真菌疗程、侵入性操作种类这12项指标在两组间差异有统计意义。Logistic回归显示侵入性操作的种类、MODS、外科手术后、抗真菌治疗前的住院时间这4项指标在两组间差异有统计意义。同时建立了Fisher预测方程。结论侵入性操作的种类、MODS、外科手术后、抗真菌治疗前的住院时间可能是真菌性败血症患者死亡的独立危险因素,可根据相应的Fisher判别方程进行预测。
Objective To analyze the incidence, clinical features and the predisposing factors of fungal septicemia, and investigate the risk factors for death due to fimgal septicemia and the prognosis of the patients. Methods We retrospectively analyzed the clinical data of 91 patients with fungal septicemia diagnosed in the last 17 years, including 60 patients with clinical cure or improvement, and 31 who die of the disease. Based on the results by univariate analysis, the data were analyzed using logistic multiple regression and Fisher＇s discriminant analysis. Results Fungal septicemia had many predisposing factors with high mortality rate. Univariate analysis revealed significant differences between the cured/improved cases and the fatal cases for 12 variables, including advanced age, complication by bacterial infection, septic shock, multiple organ dysfunction syndrome （MODS）, ICU patients, cortical hormone therapy, surgery, chemotherapy, use of immunopotentiating agents, length of hospital stay before antifungal therapy, time of anti-fungus therapy and types of invasive procedures. Logistic multiple regression analysis showed that the types of invasive procedures, MODS, surgery and prolonged hospital stay before antifungal therapy were the independent risk factors for ftmgal septicemia-related death. Fisher＇s linear discriminant equation was established for predicting the prognosis of the disease. Conclusion The types of invasive procedure, MODS, surgery and prolonged hospital stay before antifungal therapy are the independent risk factors for fungal septicemia-related death, and the patients＇ prognosis can be predicted using Fisher＇s linear discriminant equation.
出处
《南方医科大学学报》
CAS
CSCD
北大核心
2008年第11期2018-2021,共4页
Journal of Southern Medical University
