Reply to Ferreira et al.

We appreciate the comments by Ferreira et al. on our paper “Trends in quality of screening colonoscopy in Austria” published in the December issue of Endoscopy [1]. As the authors recognized, an interesting finding of the paper was that all 20 perforations occurred when sedation was used. However, we did not conclude a causal relationship between the use of sedation and perforation because there was no significant association (P = 0.17). This was clearly described in the results section, but not mentioned in the letter.

The impact of sedation on complication rate and perforations is certainly a very important research question. However, it was not the primary aim of this study. As mentioned in the present letter, evidence for the hypothesis that sedation is a relevant factor for perforations is still lacking [2] [3] and this could also not be confirmed by our study. For interpretation of our data, especially of the observed rates of perforation and complications in general, it is important to consider that all included colonoscopies were screening colonoscopies, without special focus on therapeutic indications or colonoscopies without interventions.

One of our observations was that the use of sedation was associated with a higher rate of complications (including bleeding, cardiopulmonary events, perforation, and others). As possible reasons for this, the authors discussed reduced perception of pain by patients and subsequent failure of the endoscopist to notice alert signals, which of course is plausible.

In order to analyze the causal relationship between sedation and complications, particularly to define the impact of “uncontrolled confounders” that might have caused bleeding (rather than sedation itself), studies specifically designed to evaluate this important research question are indispensable. Eleven of the 20 perforations in our study cohort occurred in colonoscopy with polypectomy (the most malignant lesion described was a flat or sessile polyp/adenoma in 10/11 and a pedunculated polyp in 1/10). However, the study aims proposed by Ferreira et al. (i. e. moderate vs. deep sedation, details in differences between sedated and unsedated groups, differentiation between polypectomy-/insertion-related perforations) need to be elaborated in a specifically designed study rather than as a descriptive post hoc analysis of the present screening colonoscopy cohort. This becomes evident when we consider that in our study, sedation was used, at the discretion of the endoscopist and the patient, in almost 90 % of procedures, and perforations and complications were generally rare.

In conclusion, we agree with Ferreira et al. that the effect of sedation on complications in colonoscopy is a very important research question in this field and the proposed considerations should be clarified in future studies. Our study contributed important information on quality aspects of screening colonoscopies in Austria, including the observation that 100 % of perforations occurred in sedated patients. These observations and the issues pointed out in the letter by Ferreira et al. should encourage further research on this important topic.

• Reference

• 1 Waldmann E, Gessl I, Sallinger D. et al. Trends in quality of screening colonoscopy in Austria. Endoscopy 2016; 48: 1102-1109
• 2 Cooper GS, Kou TD, Rex DK. et al. Complications following colonoscopy with anesthesia assistance: a population-based analysis. JAMA Intern Med 2013; 173: 551-556
• 3 Vargo JJ, Niklewski PJ, Williams JL. et al. Patient safety during sedation by anesthesia professionals during routine upper endoscopy and colonoscopy: an analysis of 1.38 million procedures. Gastrointest Endosc 2017; 85: 101-108