LAUSR

We read the article with great interest, reported by Evensen et al. [1] aimed to evaluate the clinical outcome of peroral endoscopic myotomy (POEM) in treatment-naive patients. We appreciate the hard work of the authors. However, we have several concerns about their study. The first concern is that the inclusion of the study conducted by Tang et al. [2] is arguable. In Tang et al.’ study, a total of 67 patients were included, divided into hybrid knife subgroup (n1⁄4 31) and conventional method group (n1⁄4 36), of which 2 patients in each group had previous Botox injection and 7 patients and 11 patients had previous balloon dilation, respectively. There were less than 90% of treatmentnaive patients in this study, which does not meet the predefined inclusion criteria (n 20 with 90% treatment-naive patients or studies with the higher number included with separate data presentation on treatment-naive patients). Therefore, the inclusion of this study by Tang et al. is debatable. In addition, the authors defined one exclusion criterion as ‘age< 18 years’, but three of their included studies involved patients younger than 18 years old [2–4]. In addition, another study by Tang et al. [5] with separate data presentation of 39 treatment-naive patients meets the inclusion and exclusion criteria, but it was missed in this systematic review. As a result, we suggest that the inclusion and exclusion criteria should be redefined by the authors. Furthermore, although the author declared they performed this study according to the guidelines of the preferred reported items for systematic review and meta-analyses (PRISMA-P) [6], some of the important items had been missed, such as defining the primary and secondary outcomes, assessing risk of bias of individual and across studies and missing data synthesis. And the roles and contributions of the authors were not clearly addressed in the review. Last but not the least, most included studies about POEM were carried out in China, therefore, Chinese electronic databases (such as Weipu, Wanfang, CNKI and CBM databases) should be searched. We thank the authors for their valuable contribution. However, based on above, the quality of evidence generated by this review is questionable.