To the Editor—We commend Yahav et al on their recent publication of the results of their individually randomized noninferiority trial of 7 vs 14 days of antibiotic therapy for stable patients with Gram-negative bacteremia [1]. As emphasized by Daneman et al [2], this study has the potential to significantly change hospital practice. With this correspondence, we wish to highlight an important consideration for future noninferiority trials of reduced durations of antibiotics—namely, that sequential noninferiority studies may fail to identify progressive reductions in efficacy with decreasing antibiotic durations and could produce misleading results. We also discuss some approaches to prevent this from occurring.

中文翻译：

---

降低抗生素治疗持续时间的非劣效性试验中疗效下降的潜力。

致编辑-我们赞扬Yahav等人最近发表的有关稳定革兰氏阴性菌血症患者的7到14天抗生素治疗的个体随机非劣效性试验结果的报道[1]。正如Daneman等[2]所强调的那样，这项研究具有显着改变医院实践的潜力。有了这种对应关系，我们希望着重指出未来减少抗生素使用时间的非劣效性试验的重要考虑因素，即，连续的非劣效性研究可能无法确定随着抗生素使用期限的降低药效的逐步降低，并可能产生误导性的结果。我们还将讨论一些防止这种情况发生的方法。