Download PDF
Endoscopy 2025; 57(11): 1317
DOI: 10.1055/a-2669-8239
Letter to the editor

Reply to Wang et al.

Authors

  • Andrea Sorge 

    1   Gastroenterology and Hepatology, University Hospital Ghent, Ghent, Belgium (Ringgold ID: RIN60200)
    2   Pathophysiology and Transplantation, University of Milan, Milan, Italy (Ringgold ID: RIN9304)

  • Michele Montori

    1   Gastroenterology and Hepatology, University Hospital Ghent, Ghent, Belgium (Ringgold ID: RIN60200)
    3   Gastroenterology, Hepatology and Emergency Digestive Endoscopy, Università Politecnica delle Marche, Ancona, Italy (Ringgold ID: RIN9294)

  • Maria Eva Argenziano 

    1   Gastroenterology and Hepatology, University Hospital Ghent, Ghent, Belgium (Ringgold ID: RIN60200)
    4   Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium (Ringgold ID: RIN26656)

  • David James Tate

    1   Gastroenterology and Hepatology, University Hospital Ghent, Ghent, Belgium (Ringgold ID: RIN60200)
    4   Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium (Ringgold ID: RIN26656)
Further Information
Also available at
   Permissions and Reprints

Letter to:

Methodological considerations for recurrence analysisEndoscopy 2025; 57(11): 1315-1316
DOI: 10.1055/a-2669-8182

10.1055/a-2669-8182

We thank Wang et al. for their interest in our meta-analysis comparing cold snare polypectomy (CSP) versus hot endoscopic mucosal resection (H-EMR) for large nonpedunculated colorectal polyps [1]. Their commentary addresses the subgroup analysis of adenoma recurrence.

We acknowledge a minor reporting error in the pooled adenoma recurrence rate following CSP. The correct value is 25.8% (95%CI 20.9%–31.3%), rather than 21.4% (95%CI 17.2%–26.1%), as originally reported. This typographical oversight does not affect the reported risk ratio (RR), confidence intervals, or P values, which were calculated using the correct data (RR 1.97, 95%CI 0.58–6.74; P = 0.14), or the corresponding forest plot in our figure 7s. We have already communicated this correction to the editorial office and an erratum is in process. Importantly, the overall findings and their interpretation remain valid and unchanged.

Regarding the method used for the meta-analysis, we preferred to use the random-effects model with restricted maximum likelihood (REML) estimation, which is an established approach in meta-analyses, especially when few studies are included and heterogeneity is moderate. Alternative variance estimators such as DerSimonian–Laird or Paule–Mandel may yield different results, particularly with sparse data and heterogeneous populations; however, REML has been shown to offer better performance in terms of bias and confidence interval coverage, especially under conditions similar to ours (three included studies). As reported in the Cochrane Handbook [2], “An empirical comparison of different ways to estimate between-study variation in Cochrane meta-analyses has shown that they can lead to substantial differences in estimates of heterogeneity, but seldom have major implications for estimating summary effects” [3].

In conclusion, although we appreciate the methodologic considerations raised regarding the subgroup analysis of adenoma recurrence, the correction to the point estimate for adenoma recurrence does not alter the overall findings or conclusions of our study.



Publication History

Article published online:
28 October 2025

© 2025. Thieme. All rights reserved.

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany