Endoscopy 2018; 50(04): S49-S50
DOI: 10.1055/s-0038-1637175
ESGE Days 2018 oral presentations
20.04.2018 – Colonoscopy quality parameters
Georg Thieme Verlag KG Stuttgart · New York

ESTIMATING DETECTION RATES FOR ADENOMAS, SERRATED POLYPS AND CLINICALLY RELEVANT SERRATED POLYPS BY USING INDIVIDUAL DETECTION RATE RATIOS IN AVERAGE-RISK SCREENING COLONOSCOPIES

C Schramm
1   University Hospital of Cologne, Department of Gastroenterology and Hepatology, Cologne, Germany
,
A Mert Senel
1   University Hospital of Cologne, Department of Gastroenterology and Hepatology, Cologne, Germany
,
JH Hofer
2   Magen Darm Zentrum Wiener Platz, Cologne, Germany
,
H Toermer
3   GastroPraxis Köln Nord, Cologne, Germany
,
F Stenschke
4   Gastroenterologie Köln West, Cologne, Germany
,
M Stollenwerk
5   Praxis Stollenwerk, Cologne, Germany
,
I Scheller
6   Gastroenterologie Remscheid, Remscheid, Germany
,
P Kasper
1   University Hospital of Cologne, Department of Gastroenterology and Hepatology, Cologne, Germany
,
T Goeser
1   University Hospital of Cologne, Department of Gastroenterology and Hepatology, Cologne, Germany
,
HM Steffen
1   University Hospital of Cologne, Department of Gastroenterology and Hepatology, Cologne, Germany
› Author Affiliations
Further Information

Publication History

Publication Date:
27 March 2018 (online)

Also available at
 

Aims:

Adenoma detection rate (ADR) is cumbersome to obtain in routine practice. We aimed to evaluate the previously introduced adenoma-to-polyp-detection-rate-ratio (APDRR) to estimate ADR from polyp detection rate (PDR) and its applicability for estimating detection rates (DRs) of serrated polyps (SPs) and clinically relevant SPs (crSPs).

Methods:

We retrospectively analysed average-risk screening colonoscopies performed by 12 experienced gastroenterologists at a tertiary academic hospital and 5 community-based private practices in Germany between 01/01/2012 and 14/12/2016. Exclusion criteria were age < 50 years, chronic inflammatory bowel disease, history of colorectal cancer, hereditary cancer syndromes, and incomplete procedures, e.g. failed cecal intubation.

APDRR was calculated as ADR/PDR individually for each gastroenterologist on the basis of his/her first 50 and 100 colonoscopies, respectively. Estimated ADR (ADRest) was calculated as the product of PDR and APDRR for his/her subsequent procedures in portions of 50 and 100 procedures, respectively. Detection rate ratios for SP, including hyperpastic polyps, sessile and traditionally serrated adenomas, and crSP, defined as SPs ≥10 mm and/or SPs > 5 mm located proximal to the splenic flexure were calculated accordingly. Pearson's correlation coefficient (r) was used to describe the strength of association between actual and estimated DRs.

Results:

3944 patients were analyzed (48.1% men; median age 62 years, interquartile-rage 56 – 69). We observed strong and significant correlations between actual and estimated DRs of adenomas (r = 0.865, and r = 0.870, respectively, p < 0.001), SP (r = 0.884 and r = 0.929, respectively, p < 0.001) and crSP (r = 0.820 and r = 0.846, respectively, p < 0.001). Mean differences (± standard deviation) between actual and estimated DRs of adenomas, SPs and crSPs were 5.2% (± 4.5%), 4.5% (± 3.3%), and 2.9% (± 2.8%) (portions of 50 procedures), and 5.0% (± 5.3%), 3.7% (± 2.8%), and 2.6% (± 2.5%) (portions of 100 procedures), respectively.

Conclusions:

Detection rate ratios may offer a useful tool to estimate DRs for adenomas, SPs, and crSP by simply calculating PDR in daily routine.