The objectives of this study were to report those variables which are readily identifiable at the bedside and that are able to predict mortality in patients with bacteraemia caused by anaerobes. Patients with clinically significant anaerobic bacteraemias detected between January 2016 and December 2019 in a tertiary hospital in Granada (Spain) were retrospectively included. Species identification was performed by MALDI-TOF MS and/or molecular methods. Finally, 136 cases of anaerobic bacteraemia were included, being the most frequent anaerobes Bacteroides (45.5%; n = 62), Clostridium (24.2%, n = 33), and Gram-positive anaerobic cocci (16.1%, n = 22). Crude mortality was 25.7%, corresponding to 35 patients who died, with 82.8% of deaths directly attributable to bacteraemia. A multivariable logistic regression model with non-parametric bootstrap estimation identified three variables that were independently and significantly associated with an increased risk of death: 1) hospitalization in the intensive care unit; 2) septic shock; and 3) presence of any kind of cancer. These variables were as recorded at the time that the first positive blood culture was obtained. An index score, obtained from these variables, was calculated and divided patients into two groups with increasing likelihood of mortality resulting from anaerobic bacteraemia. The sensitivity and specificity of a prediction of death based on this model were 65.2% and 97%, respectively.

这项研究的目的是报告那些易于在床旁识别并能够预测由厌氧菌引起的菌血症患者死亡率的变量。回顾性分析了2016年1月至2019年12月在西班牙格拉纳达的一家三级医院发现的具有临床意义的无氧菌血症患者。通过MALDI-TOF MS和/或分子方法进行物种鉴定。最后，136案件厌氧菌血症都包括在内，是最常见的厌氧菌杆菌（45.5％; N = 62），梭状芽胞杆菌（24.2％，n = 33）和革兰氏阳性厌氧球菌（16.1％，n = 22）。粗死亡率为25.7％，相当于35例死亡，其中82.8％的死亡直接归因于菌血症。具有非参数bootstrap估计的多变量logistic回归模型确定了三个独立且显着与死亡风险增加相关的变量：1）重症监护病房住院；2）败血性休克；3）任何类型的癌症。这些变量记录为获得首次阳性血液培养时的记录。计算从这些变量获得的指数得分，并将患者分为两组，由于厌氧菌血症导致死亡的可能性增加。基于该模型的死亡预测的敏感性和特异性为65。