Objective: According to recent ESGE guideline adequate polypectomy technique should be used in >80% of polypec- tomies. Despite recommendations of snare polypectomy for small polyps (6-9mm) almost half of them are resected by forceps. The aim of this study was to analyse if the publica- tion of clinical data or guideline have more impact on the reduction of the forceps resection rate of polyps ≥5mm.

      Methods: A total of 90,279 screening colonoscopies per- formed by 266 endoscopists between 08/2015 and 10/2018 within a quality assurance program were included in this study. We analysed resection techniques by polyps’ size be- fore (08/2015-02/2017) and after (04/2017-10/2018) pub- lication of the guideline. Further we investigated forceps resection rates after a publication on resection techniques of our present cohort study in 07/2015.

      Results: The polyp detection and resection rate was 38.58% (n= 34,826) and 91.74% (n= 31,948), respectively. Overall, 28.00% (n= 2,521) of polyps sized ≥5mm were re- sected using forceps. Forceps resection rate decreased in private practices after both publications (RR: 0.68 (95%CI: 0.66-0.70)) vs. 0.91 (95%CI: 0.86-0.96)); p<0.001). In con-

      trast, in hospitals a significant decrease of the forceps re- section rate was observed only after the publication of clinical data. (RR: 0.91 (95%CI: 0.89-0.914); vs. 1.19 (95%CI:

      1.13-1.26)); p<0.001). For polyps ≥5mm endoscopists had a mean adequate polypectomy technique rate of 68.31% (95%CI: 64.21-72.41).

      Conclusion: The present study showed that publication of clinical data led to an increase of adequate polypectomy technique for polyps ≥5mm in both, hospitals and private practices. None of the interventions were successful enough to reach the goal of at least 80% adherence to polypectomy guideline.

      Colorectal Cancer (CRC) is the third most common cancer and the second leading cause of cancer-death, worldwide [1]. In accordance to Statistic Austria, CRC incidence and mortality has been decreasing in the last 10 years [2].

      It is well known that higher adenoma detection rates and adequate resection technique are associated with reduction of incidence and mortality of colorectal cancer [3-5]. However, post-polypectomy cancer can occur between the screening and surveillance colonoscopies. Up to 37% of these cancers are as- sociated with incomplete polyp resection [6-9].

      Diminutive (≤5 mm) or small polyps (6-9 mm), which mea- sure less than 10 mm account for the majority (90%) of colonic polyps [3,10,11]. 60% and 70.5% of those polyps were neoplas- tic and in 0.5% and 1.5% advanced histological features were found, respectively [12]. Complete histological resection by forceps was found in only 31-90.7% of small polyps [13-15]. In contrast cold snare polypectomy for small polyps was safe and effective in nearly 100% [16,17].

      According to the European Society of Gastrointestinal Endos- copy (ESGE) polypectomy guideline diminutive (≤ 5 mm) and small polyps (6-9 mm) should be removed by cold snare [18], whereas forceps can be used for polyps between 1 and 3 mm if snare polypectomy is technically difficult to perform. The usage of hot snare is recommended only for polyps between 10 and 19 mm [18].

      The ESGE and United European Gastroenterology (UEG) rec- ommend an appropriate removal technique of >80% [19]. Two surveys have shown that the endoscopic polypectomy tech- nique is associated with the appearance and size of polyps and varies strongly between endoscopists [20,21]. A prior publica- tion of the present cohort study revealed that 46% of all lesions were removed by using forceps, which was incongruous to the EU-Guideline of 2010 [22].

      The aim of this study was to evaluate whether publication of clinical data or the guideline has more impact on improving resection techniques for polyps ≥5 mm.

      The Austrian Certificate of Quality Colonoscopy Screening was launched by the Austrian Society of Gastroenterology and Hepatology (OEGGH) in cooperation with the Association of Austrian Social Security Institutions and Austrian Cancer Aid in 2007.

      All specialists for internal medicine and surgeons can par- ticipate in the project, if they accomplish the conditions and quality standards of the OEGGH. Participating endoscopists had to perform a minimum of 200 complete colonoscopies and 50 polypectomies under supervision to apply. Further, a minimum of 100 ongoing complete colonoscopies and 10 polypectomies per year are required. More information about the project can be found in previous studies [23-25]. Written informed consent for data transmission was obtained by the participating endos- copists. This study was approved by the Ethics Committee of the Medical University of Vienna (EK 2081/2018).

      Screening colonoscopies of all patients between 50 and 100 years and asymptomatic patients without positive history over 30 years, if they obtain screening character (e.g. fear of can- cer) were included. All colonoscopies between 01.03.2017 and 31.03.2017, in which the guideline were sent to the participat- ing endoscopists and all colonoscopies in which both, forceps and snare were used for the same polypectomy, were excluded.

Definition of the periods

      Period 0 (07/2012- 12/2013) is defined as the time before a prior publication of our present cohort study on resection tech- niques.

      Period 1 (08/2015-02/2017) is defined as the time between a prior publication of the present cohort study on resection techniques and the publication of the ESGE-Guideline.

      Period 2 (04/2017-10/2018) is defined as the time after the publication of the ESGE-Guideline.

      For categorical variables absolute and relative frequencies were used and all continuous variables are described by arith- metic mean and Standard Deviation (SD).

      To compare the adherence to the ESGE-Guideline, the for- ceps removal rate (FRR), based on the endoscopists facility type, polyp’s size (<5 mm, ≥5 mm, 5-10 mm, >10 mm) and mor- phology (flat, sessile, pedunculated) were analyzed for both, before (period 1 (P1): 08/2015-02/2017) and after (period 2 (P2): 04/2017-10/2018) the publication of the ESGE-Guideline. Furthermore, we compared the periods before (Period 0 (P0): 07/2012- 12/2013) and after (P1) the prior publication of clini- cal data of our present cohort study. The FRR is defined as the percentage of forceps polypectomies. Chi-Squared test was used to compare the results before and after the publications.

      The confidence interval for all analyses was set at 95% and statistical significance was defined by p values ≤0.01. Statistical analysis was performed using IBM SPSS Statistics version 25.0 and Microsoft Office Excel.

      Between 01.08.2015 and 30.09.2018 107,125 screening colonoscopies were performed. After exclusion, 90,279 (P1: 43.,41; P2: 46,338) screening colonoscopies by 266 (P1: 230; P2:247) endoscopists were analysed. The mean age was 60.35 (SD 9.03) and 47.80% (n= 47,081) were female (Table 1).

      74.44% (n= 198) of endoscopists work in private practices and 25.56% (n= 68) in hospitals. Practitioners in private practic- es have performed 80.15% (n= 72,363) of all colonoscopies and in hospitals 19.85% (n= 17,916). Overall, 80.86% (n= 25,833; P1:12,350; P2:13,483) of all detected polyps were resected by endoscopists in private practices and 19.14% (n= 6,115; P1:2,973; P2:3,142) in hospitals.

      Polyp detection rate was 38.58% (n= 34,826) and the resec- tion rate was 91.74% (n= 31,948). In P1, the detection rate of polyps was 38.08% (n= 16,731) and 91.58% (n= 15,323) of pol- yps were resected. In contrast, in P2 the polyp detection and resection rate were 39.05% (n= 18,095) and 91.88% (n=16,625), respectively. Regarding shape, 63.32% (n= 20,230) of resected polyps were sessile, 28.83% (n= 9,212) flat and 7.85% (n= 2,506) pedunculated. The polyps’ size of all resected lesions are shown in Table 2.

      Forceps was used in 73.04% (n= 23,336) of all polypectomies. In the first period forceps was the preferred polypectomy tech- nique in 73.52% (n= 11,266) and in the second period in 72.60% (n= 12,070). 10.84% (n= 3,462; P1:1,415; P2:2,047) of polyps were resected by cold snare and 16.12% (n= 5,150; P1:2,642; P2:2,508) by hot snare. Forceps resection was used in 28.00% (n= 2,521; P1:1,441; P2: 1,080) of all lesion ≥5 mm, which does not correspond to the ESGE-Guideline. In contrast, cold and hot snare was the preferred polypectomy technique for 24.93% (n= 2,245; P1:975; P2:1,270) and 47.07% (n= 4,238; P1:2,164; P2:2,074) of polyps ≥5 mm, respectively. Furthermore, for- ceps polypectomy was used in 90.72% (n= 20,815; P1:9,825; P2:10,990) of lesions <5 mm and 49.73% (n= 2,398; P1:1,377; P2:1,021) of lesions between 5 and 10 mm. Further, 6.89% (n= 123; P1:64; P2:59) of polyps >10 mm were resected by forceps. 5.30% (n= 1,217; P1:440; P2:777) of polyps <5 mm, 28.60% (n=2,065; P1:900; P2:1,165) of polyps 6-10 mm and 10.09% (n=180; P1:75; P2:105) of polyps >10 mm were resected by cold snare. Hot snare polypectomy was performed in 3.97% (n= 912; P1:478; P2:434) for lesions <5 mm, 28.19% (n= 2,757; P1:1,432; P2:1,325) between 5 and 10 mm and 83.02% (n=1,482; P1:732; P2:749) >10 mm (Table 2).

      Based on the polyp morphology, 76.25% (n= 15,425) of ses- sile, 80.22% (n= 7,390) flat and 30.79% (n= 521) of peduncu- lated polyps were resected by forceps.

Comparison between period 1 and period 2

      Forceps polypectomy for lesions ≥5 mm was used in 30.33% (n= 2,247; P1:1,284; P2:963), 171.8% (n= 274; P1:157; P2:117)

      by endoscopists in private practices and hospitals, respective- ly. Interestingly we only found a significant decrease of 3.12% (P1: 35.73 [95%CI: 30.03-41.42] vs. P2: 32.61% [95%CI: 27.15-

      37.96]) of the FFR in private practices, but a significant increase of 3.96% (P1: 20.62% [95%CI: 12.89-18.35] vs. P2: 24.58%

      [95%CI: 16.49-32.67]) in hospitals (both p<0.001) (Figure 1).

      Regarding polyps sized smaller than 5 mm, a significant decrease of the FRR was observed after the publication of the ESGE-guideline for both, hospitals and private practices (p<0.001). Furthermore, the FRR for polyps between 5 and 10 mm decreased significantly among private practices but in- creased significantly in hospitals (p<0.001). For polyps >10 mm the FRR decreased significantly in hospitals (p<0.001). In con- trast, there was no difference in private practices before and after the publication of the ESGE guideline (p= 0.912) (Table 3).

      Between P1 and P2 we found significant increases in both facilities for cold snare polypectomy for polyps <5 mm, 5-10mm, <10 mm and ≥5 mm. In contrast we observed a significant decrease of the cold snare polypectomy rate for polyps >10 mm in hospitals (p<0.001) and no difference in private practices (p= 0.742) (Table 4).

      For hot snare polypectomy we observed a significant de- crease for polyps <5 mm, 5-10 mm and ≥5 mm in both, hospitals and private practices (all p<0.001). Regarding polyps >10 mm, there was no difference between P1 and P2 in private practices (p=0.756), but a significant increase in hospitals (p<0.001) (Table 5).

Comparison between period 0 and period 1

      Among polyps ≥5 mm we observed a significant decrease in the FRR of 16.85% (P1: 52.58 [95%CI %: 51.49-53.67] vs. P2: 35.73% [95%CI: 30.03-41.42]) and 2.13% (P1: 22.75% [95%CI: 20.97-24.61] vs. P2: 20.62% [95%CI: 12.89-28.35]) after the study publication in private practices and hospitals (both p<0.001) (Figure 1).

      Interestingly, the FRR for polyps <5 mm decreased signifi- cantly in private practices (p<0.001) but showed no differences in hospitals (p= 0.002).When comparing the FRR for polyps be- tween 5 and 10 mm we found significant decreases for both, en- doscopists in hospitals and private practices (both p<0.001). For polyps >10 mm there was no difference in hospitals (p=0.527). In contrast the FRR decreased in private practices (p<0.001) (Table 3).

Figure 1: Forceps resection rates for polyps ≥5 mm by facilities before and after the publications of the study and the guideline.

Table 1

Table 1: Patients characteristics.

Table 2

Table 2: Polypectomy technique according on polyp size.

Table 3

Table 3: Forceps resections rates (FRR) before and after both publications; RR, relative risk; CI, confidence interval.

Table 4

Table 4: Cold snare resection rates before(P1) and after(P2) the publication of the ESGE guideline;

Table 5

Table 5: Hot snare resection rates before(P1) and after(P2) the publication of the ESGE guidelines.

      In this study we evaluated the preferred polypectomy tech- niques based on the polyp size and compared the results of our prior publication such as the results before and after the ESGE guideline for polypectomy. Our study indicates a strong varia- tion in polypectomy techniques and that 73.04% (n= 23,336) of polyps any size, especially 28.00% (n= 2,521) of polyps ≥5 mm were still resected by forceps.

      Although, the adequate polypectomy technique and com- plete polyp resection is important to reduce tissue recurrence and the risk of interval cancer, several studies have shown that forceps polypectomy is not uncommon and inferior to snare pol- ypectomy [26-31]. Our study showed that 73.04% (n=23,336) of polyps any size, especially 90.72% (n= 20,825) of polyps <5mm and 28.00% (n=2,521) of polyps ≥5 mm were resected by for- ceps which does not correspond to the guideline. In contrast only 5.30% (n= 1,217) and 28.60% (n= 2,065) of diminutive and polyps between 5 and 10 mm were removed by cold snare, re- spectively.

      Interestingly, regarding facility type we observed a differ- ence between hospitals (30.63%; n= 2,224) and private prac- tices (17.26%; n= 272) for the usage of forceps in polyps ≥5 mm. After the publication of the ESGE guideline we found an oppo- site trend in the development of the FRR. In private practices the FRR significantly decreased by 3.12% (P1: 35.73% vs. P2: 32.61%) but increased by 3.96% (P1: 20.62% vs. P2: 24.58%) in hospitals. In comparison, the FRR decreased after period 0 in both, hospitals by 16.85% (P0: 52.58% vs. P1: 35.73%) and private practices by 2.13% (P0: 22.75% vs. P1: 20.62%) (Figure 1). Even if we have not observed a further decrease of the FRR in hospitals, we should not disregard that the FRR in hospitals is eventually much lower than in private practices. Several po- tential reasons for the variety between the facilities are that

      e.g. in hospitals there is a better access to education and train- ing. Other important aspects were cost, lack of reimbursement and as often discussed time. Forceps polypectomy is cheaper, quicker and easier to apply than cold snare polypectomy (for- ceps polypectomy: 14.29 seconds vs. cold snare polypectomy

      22.03 seconds, p<0.001) [28,29]. Further reasons for the variety of polypectomy techniques are maybe the lack of knowledge, mis-sizing of polyps and that forceps polypectomy is a simpler technique.

      For polyps any size the usage of cold snare increased by 3.08% (P1: 9.23%, n= 1,415 vs P2: 12.31%, n= 2,047), but decreased for hot snare by 2.15% (P1: 17.24%, n= 2,642; P2:15.09%, n=2,508).

      Regarding to the ESGE guideline, we found an increase of the adequate usage of cold snare polypectomy. Cold snare polypec- tomy increased between P1 and P2 in hospitals by 3.29% (P1: 10.33% vs. P2: 13.62%; p<0.001) and in private practices by 4.03% (P1: 7.75% vs. P2: 11.78%; p<0.001).

      According to the adequate usage of hot snare 83.02% (n= 1.481) of all polyps >10 mm were resected by the recommend- ed technique. Interestingly, we found an opposite trend. In hospitals the usage of hot snare for polyps >10 mm increased by 4.19% (P1: 88.17% vs. P2: 92.36%; p<0.001)) but no differ- ence (P1: 78.68% vs. P2: 78.50%; p= 0.756) was found in private practices. All in all, with 68.31% (95%CI: 64.21-72.41) the mean appropriate polypectomy technique rate is a bit lower than the recommended value of >80% by Kaminski et al. [19].

      The study was limited due to the polyps only being subclas- sified in <5 mm, 5-10 mm and >10 mm, despite the guideline recommending a particular polypectomy technique for polyps sized between 1 and 3 mm. Furthermore, for each colonoscopy, only the polypectomy technique for the most advanced polyp is described. Strengths of our study are that data are available over a long period allowing to observe trends of the different polypectomy techniques. Furthermore, we have a high number of screening colonoscopies, polypectomies and endoscopists from different facilities.

      In conclusion, we have investigated that still 28.00% (n= 2,521) of polyps ≥5 mm were resected by forceps, which does not correspondent to the ESGE guideline. On the opposite al- ready 68.31% (95%CI: 64.21-72.41) of polyps ≥5mm were removed according to guideline, which is a huge improvement in a few years, showing great efficacy of quality improvement programs, publications of papers and the guideline and further need of implementations of such programs in all countries. Further, we observed a rising adherence for cold snare polypec- tomy in both facilities.

      The project “Quality certificate for screening colonoscopy” was established in cooperation with the Main Association of the Austrian Social Security Institution and founded by fund § 447h ASVG (Fund for preventive check-ups and health promotion)

      Role of the Sponsors: The funding organizations played no role in the design and conduction of the study, in the collection, management, analysis, and interpretation of the data, or in the preparation, review, or approval of the manuscript