Toledo C, Salmerón J M, Rimola A, Navasa M, Arroyo V, Llach J, Ginès A, Ginès P, Rodés J
Liver Unit, Hospital Clinic i Provincial, University of Barcelona, Spain.
Hepatology. 1993 Feb;17(2):251-7.
Cefotaxime is the most commonly used antibiotic for initial therapy of spontaneous bacterial peritonitis in cirrhosis. However, since the introduction of cefotaxime no study has been performed to investigate factors influencing prognosis in cirrhotic patients with this type of infection. In this study, predictive factors for infection resolution and patient survival were investigated in 213 consecutive episodes of spontaneous bacterial peritonitis in 185 cirrhotic patients. All patients were initially treated with cefotaxime. One hundred sixty-five episodes (77%) resolved with cefotaxime alone, and two more episodes (1%), initially unresponsive to cefotaxime, were cured after modification of antibiotic therapy. In a multivariate analysis (stepwise logistic regression), only 4 of 51 clinical and laboratory variables obtained at the time of diagnosis of infection were identified as independent predictors of infection resolution: band neutrophils in white blood cell count, community-acquired vs. hospital-acquired peritonitis, blood urea nitrogen level and serum aspartate aminotransferase level. No patient experienced serious adverse effects related to cefotaxime. Eighty-two patients died during hospitalization (38% mortality rate in relation to the 213 episodes of peritonitis). In the multivariate analysis, six variables were independently correlated with survival: blood urea nitrogen level, serum aspartate aminotransferase level, community-acquired vs. hospital-acquired peritonitis, age, Child-Pugh score and ileus. No microbiological data had predictive value for infection resolution or survival.(ABSTRACT TRUNCATED AT 250 WORDS)
头孢噻肟是肝硬化患者自发性细菌性腹膜炎初始治疗中最常用的抗生素。然而,自头孢噻肟应用以来,尚未有研究调查影响这类感染的肝硬化患者预后的因素。在本研究中,对185例肝硬化患者连续发生的213次自发性细菌性腹膜炎发作的感染缓解和患者生存的预测因素进行了调查。所有患者最初均接受头孢噻肟治疗。165次发作(77%)仅用头孢噻肟即可缓解,另有2次发作(1%)最初对头孢噻肟无反应,在调整抗生素治疗后治愈。在多变量分析(逐步逻辑回归)中,在感染诊断时获得的51个临床和实验室变量中,只有4个被确定为感染缓解的独立预测因素:白细胞计数中的杆状中性粒细胞、社区获得性与医院获得性腹膜炎、血尿素氮水平和血清天冬氨酸转氨酶水平。没有患者出现与头孢噻肟相关的严重不良反应。82例患者在住院期间死亡(相对于213次腹膜炎发作,死亡率为38%)。在多变量分析中,6个变量与生存独立相关:血尿素氮水平、血清天冬氨酸转氨酶水平、社区获得性与医院获得性腹膜炎、年龄、Child-Pugh评分和肠梗阻。没有微生物学数据对感染缓解或生存具有预测价值。(摘要截短于250字)