Evidence Hub

Piperacillin/tazobactam plus erythromycin improves clinical outcomes in AECOPD with bacterial lower respiratory tract infections: a retrospective cohort study

Yemeng Yang, Tao Zhang, Xi Zheng, Yi Lu, Dan C. Qu, Zhijing Zhu, Xinjuan Liu, Jiaman Wang et al.

Scientific Reports, 2026

Abstract

It remains uncertain whether patients with acute exacerbations of chronic obstructive pulmonary disease (AECOPD) and bacterial lower respiratory tract infections (LRTIs) could similarly benefit from β-lactam and macrolide antibiotics therapy as community-acquired pneumonia (CAP) does. In this study, we compared the clinical success rates of piperacillin/tazobactam (TZP) monotherapy versus its combination with erythromycin lactobionate injection (Ery) in patients with AECOPD and bacterial LRTIs and developed a machine learning (ML) model to predict treatment outcomes. The patients with AECOPD and bacterial LRTIs received antimicrobial therapy with either piperacillin-tazobactam (TZP) alone or TZP in combination with Ery. Inverse probability of treatment weighting (IPTW) was performed between the two groups. Subsequently, a stacking ensemble learning (SEL) model was developed and deployed as a web application to simultaneously predict clinical outcomes for both treatment options. The result demonstrated that TZP combined with Ery significantly reduced the incidence of clinical treatment failure compared to TZP monotherapy (14.00% vs. 19.75%; OR, 0.66; 95% CI, 0.49–0.89; P = 0.006). In an independent test set, the SEL model demonstrated strong performance across multiple metrics, including ROC-AUC (0.71), recall (sensitivity) (0.72), and accuracy (0.69). Finally, a web application based on the SEL was developed ( http://106.12.146.54/ ). This study demonstrated that the…

How strong is the evidence?

Evidence Hub reads this as a cohort study, done in people. Its evidence level is 3 of 5 (Moderate · trial / cohort / secondary analysis). Trials without a large randomized design, cohorts and secondary analyses give moderate evidence.

How the five levels of evidence work. The level is assigned automatically from the abstract and can be wrong, so check the paper.

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free