South Texas Veterans Health Care System, San Antonio, TX, USA.
Long School of Medicine, University of Texas Health San Antonio, San Antonio, TX, USA.
Int J Med Sci. 2023 Feb 13;20(4):437-443. doi: 10.7150/ijms.80621. eCollection 2023.
This study compares treatment failure for patients who received oral beta-lactams (BLs) and fluoroquinolones (FQs) for stepdown treatment of Enterobacterales bloodstream infections (BSIs). We conducted a single-center, retrospective, age- and sex-matched, cohort study, at a Veterans Affairs (VA) hospital in South Texas. Eligible patients were at least 18 years of age with a monomicrobial BSI treated with a single oral BL or FQ antibiotic. Treatment failure was defined as recurrence or all-cause mortality within 90 days of documented BSI. Bivariate (chi-square, Fisher's Exact, and Wilcoxon Rank Sum) and multivariate (logistic regression) statistical tests were used to compare groups. A total of 130 patients were included in this study, with 65 patients per group. Groups were well balanced with respect to exact age, sex assigned at birth, Caucasian race, source control, intensive care unit admission, and Charlson Comorbidity Index. Importantly, 60% of patients in the BL group had cultures that were resistant to FQs and 71% were prescribed cefpodoxime. Patients in the BL group had higher median (interquartile range [IQR]) Pitt bacteremia scores than those in the FQ group: 2 (1-4) vs. 1 (1-2), p=0.04. Patients in the BL group also had a higher median (IQR) duration of intravenous (IV) antibiotics than those in the FQ group: 5 (3-7) vs. 4 (3-5), p=0.02. Treatment failure was statistically comparable for patients in the BL and FQ groups: 15% vs. 12%, p=0.61. This finding was consistent in a multivariate logistic regression model with group (BL vs. FQ) as the independent variable, treatment failure as the dependent variable, and Pitt bacteremia score and duration of IV antibiotics as covariates (OR: 0.76, 95% CI: 0.27-2.18). One patient in the FQ group experienced infection. This study suggests that BLs may be as effective as FQs for oral stepdown treatment of Enterobacterales BSI without the potential associated risks. Furthermore, in the setting of FQ-resistant Enterobacterales BSI secondary to urinary source, third generation oral cephalosporins (i.e., cefpodoxime) may be reasonable alternatives.
本研究比较了接受口服β-内酰胺类药物（BLs）和氟喹诺酮类药物（FQs）进行肠杆菌血流感染（BSI）降阶梯治疗的患者的治疗失败情况。我们在德克萨斯州南部的一家退伍军人事务（VA）医院进行了一项单中心、回顾性、年龄和性别匹配的队列研究。合格的患者年龄至少 18 岁，患有单一微生物的 BSI，接受单一口服 BL 或 FQ 抗生素治疗。治疗失败定义为在有记录的 BSI 后 90 天内复发或全因死亡率。使用双变量（卡方检验、Fisher 精确检验和 Wilcoxon 秩和检验）和多变量（逻辑回归）统计检验来比较组。共有 130 名患者纳入本研究，每组 65 名患者。两组在确切年龄、出生时分配的性别、白种人种族、源控制、重症监护病房入院和 Charlson 合并症指数方面平衡良好。重要的是，BL 组 60%的患者的培养物对 FQs 具有耐药性，71%的患者开了头孢泊肟。BL 组患者的 Pitt 菌血症评分中位数（四分位距[IQR]）高于 FQ 组：2（1-4）比 1（1-2），p=0.04。BL 组患者静脉（IV）抗生素的中位（IQR）持续时间也高于 FQ 组：5（3-7）比 4（3-5），p=0.02。BL 组和 FQ 组的治疗失败率统计学上无差异：15%比 12%，p=0.61。在多变量逻辑回归模型中，以组（BL 与 FQ）为自变量，治疗失败为因变量，Pitt 菌血症评分和 IV 抗生素持续时间为协变量，这一发现是一致的（OR：0.76，95%CI：0.27-2.18）。FQ 组的 1 名患者发生了 感染。本研究表明，BLs 可能与 FQs 一样有效，可用于肠杆菌血流感染的口服降阶梯治疗，而不会有潜在的相关风险。此外，在由于尿源而导致的 FQ 耐药肠杆菌血流感染的情况下，第三代口服头孢菌素（即头孢泊肟）可能是合理的替代药物。
