Objectives  To systematically review the efficacy and safety of de-escalation therapy for severe pneumonia. Methods  We searched PubMed, EMbase, The Cochrane Library, CBM, CNKI, VIP and WanFang Data databases and the Chinese Clinical Trial Registry (www.chictr.org.cn) to collect randomized controlled trials (RCTs) of de-escalation therapy for patients with severe pneumonia from inception to June, 2017. Two reviewers independently screened literature, extracted data and assessed the risk of bias of included studies. Meta-analysis was then performed using RevMan 5.3 software. Results  A total of 13 RCTs involving 1 860 patients were included. The results of meta-analysis showed that: the de-escalation therapy group was superior to the control group on clinical cure rate (RR=1.28, 95%CI 1.20 to 1.35, P<0.000 01), the total hospitalization time (MD=–6.86, 95%CI –9.12 to –4.59,P<0.000 01), remission time of complications (MD=–6.26, 95%CI –8.43 to –4.10,P<0.000 01) and mortality (RR=0.48, 95%CI 0.28 to 0.82,P=0.001). Reported cases of adverse reactions were rare, in which the degree of reactions ranged from mild to moderate. The safety was fairly satisfactory. Conclusions  Current evidence shows that de-escalation therapy for patients with severe pneumonia has improved efficacy compared with conventional treatments, and can significantly shorten the total hospitalization time and reduce mortality. Due to the limited quality and quantity of the included studies, more high quality studies are required to verify above conclusions.

Citation： CHEN Zhaoyan, ZHAN Mei, XU Ting. Efficacy and safety of antibiotic de-escalation therapy for severe pneumonia: a meta-analysis. Chinese Journal of Evidence-Based Medicine, 2018, 18(1): 52-58. doi: 10.7507/1672-2531.201706116 Copy