Abstract：
OBJECTIVE: To determine whether mortality is lower in patients with Klebsiella pneumoniae bloodstream infection (BSI) who receive combination antimicrobial therapy than in those who receive monotherapy.
METHODS: Two authors independently searched for relevant articles in the PubMed, Embase, Web of Science, and Cochrane Library databases through to August 10, 2023. Risk of bias was evaluated using the ROBINS-I tool. Possible sources of heterogeneity were evaluated by meta-regression using a mixed-effects model.
RESULTS: Among 8044 articles screened, there were 23 studies (3443 patients) that were eligible for meta-analysis. Meta-regression analysis identified the proportion of patients with carbapenem-resistant Klebsiella pneumoniae (CRKP) BSI to be a potential source of heterogeneity. Subgroup analysis showed that mortality on monotherapy was significantly higher when the proportion of patients with CRKP BSI was ≥50% (OR 1.75, 95% CI 1.33-2.30) and significantly lower when this proportion was <50% (OR 0.55, 95% CI 0.24-1.24). Overall mortality was significantly higher on tigecycline monotherapy (OR 2.86, 95% CI 1.46-5.59) than on combination therapy containing both these agents. There was a trend in favor of colistin/polymyxin B-containing combination therapy (OR 1.37, 95% CI 0.83-2.28).
CONCLUSIONS: Combination antimicrobial therapy can lower mortality in patients with CRKP but may not show a survival advantage over monotherapy when the proportion of patients with CRKP BSI is <50%. High-quality prospective observational studies are needed because of the high risk of bias and limited data in the studies performed to date.
METHODS: Two authors independently searched for relevant articles in the PubMed, Embase, Web of Science, and Cochrane Library databases through to August 10, 2023. Risk of bias was evaluated using the ROBINS-I tool. Possible sources of heterogeneity were evaluated by meta-regression using a mixed-effects model.
RESULTS: Among 8044 articles screened, there were 23 studies (3443 patients) that were eligible for meta-analysis. Meta-regression analysis identified the proportion of patients with carbapenem-resistant Klebsiella pneumoniae (CRKP) BSI to be a potential source of heterogeneity. Subgroup analysis showed that mortality on monotherapy was significantly higher when the proportion of patients with CRKP BSI was ≥50% (OR 1.75, 95% CI 1.33-2.30) and significantly lower when this proportion was <50% (OR 0.55, 95% CI 0.24-1.24). Overall mortality was significantly higher on tigecycline monotherapy (OR 2.86, 95% CI 1.46-5.59) than on combination therapy containing both these agents. There was a trend in favor of colistin/polymyxin B-containing combination therapy (OR 1.37, 95% CI 0.83-2.28).
CONCLUSIONS: Combination antimicrobial therapy can lower mortality in patients with CRKP but may not show a survival advantage over monotherapy when the proportion of patients with CRKP BSI is <50%. High-quality prospective observational studies are needed because of the high risk of bias and limited data in the studies performed to date.
摘要：
目的：确定接受联合抗菌药物治疗的肺炎克雷伯菌血流感染（BSI）患者的死亡率是否低于接受单一药物治疗的患者。
方法：两位作者在PubMed中独立搜索相关文章，Embase,WebofScience,和Cochrane图书馆数据库到2023年8月10日。使用ROBINS-I工具评估偏倚风险。使用混合效应模型通过元回归评估可能的异质性来源。
结果：在筛选的8044篇文章中，有23项研究(3443例患者)符合荟萃分析的条件.荟萃回归分析确定耐碳青霉烯类肺炎克雷伯菌（CRKP）BSI患者的比例是异质性的潜在来源。亚组分析显示，当CRKPBSI患者的比例≥50％时，单药治疗的死亡率显着升高（OR1.75，95％CI1.33-2.30），当该比例<50％时，单药治疗的死亡率显着降低（OR0.55，95％CI0.24-1.24）。替加环素单药治疗的总死亡率（OR2.86，95％CI1.46-5.59）明显高于含这两种药物的联合治疗。有趋势支持含粘菌素/多粘菌素B的联合治疗（OR1.37，95％CI0.83-2.28）。
结论：联合抗菌治疗可以降低CRKP患者的死亡率，但当CRKPBSI患者的比例<50%时，与单药治疗相比可能没有生存优势。由于迄今为止进行的研究存在较高的偏倚风险和有限的数据，因此需要高质量的前瞻性观察研究。
方法：两位作者在PubMed中独立搜索相关文章，Embase,WebofScience,和Cochrane图书馆数据库到2023年8月10日。使用ROBINS-I工具评估偏倚风险。使用混合效应模型通过元回归评估可能的异质性来源。
结果：在筛选的8044篇文章中，有23项研究(3443例患者)符合荟萃分析的条件.荟萃回归分析确定耐碳青霉烯类肺炎克雷伯菌（CRKP）BSI患者的比例是异质性的潜在来源。亚组分析显示，当CRKPBSI患者的比例≥50％时，单药治疗的死亡率显着升高（OR1.75，95％CI1.33-2.30），当该比例<50％时，单药治疗的死亡率显着降低（OR0.55，95％CI0.24-1.24）。替加环素单药治疗的总死亡率（OR2.86，95％CI1.46-5.59）明显高于含这两种药物的联合治疗。有趋势支持含粘菌素/多粘菌素B的联合治疗（OR1.37，95％CI0.83-2.28）。
结论：联合抗菌治疗可以降低CRKP患者的死亡率，但当CRKPBSI患者的比例<50%时，与单药治疗相比可能没有生存优势。由于迄今为止进行的研究存在较高的偏倚风险和有限的数据，因此需要高质量的前瞻性观察研究。
