Impact of inadequate bowel cleansing in sigmoidoscopy screening

Figures & data

Table 1. Characteristics of study population and endoscopy setting and association with quality of bowel cleansing at sigmoidoscopy.

Category	Total No. (col%)	Quality of bowel cleansing at sigmoidoscopy				p^a
		Excellent No. (row %)	Good No. (row %)	Partly poor No. (row %)	Poor No. (row %)
Overall	35,710	20,934 (58.6)	6580 (18.4)	7097 (19.9)	1099 (3.1)
Age (years)
50-55	4646 (13.0)	3082 (66.3)	689 (14.8)	774 (16.7)	101 (2.2)	<0.001
56-60	8561 (24.0)	5423 (63.3)	1406 (16.4)	1536 (17.9)	196 (2.3)
61-65	8199 (23.0)	4818 (58.8)	1500 (18.3)	1642 (20.0)	239 (2.9)
66-70	8014 (22.4)	4410 (55.0)	1617 (20.2)	1729 (21.6)	258 (3.2)
> 70	6290 (17.6)	3201 (50.9)	1368 (21.7)	1416 (22.5)	305 (4.8)
Mean (IQR)	64 (58–69)	63 (58–69)	65 (60–70)	65 (59–70)	66 (60–72)
Sex
Men	17,658 (49.4)	9550 (54.1)	3604 (20.4)	3944 (22.3)	560 (3.2)	<0.001
Women	18,052 (50.6)	11,384 (63.1)	2976 (16.5)	3153 (17.5)	539 (3.0)
Symptoms
Yes	7265 (20.3)	4260 (58.6)	1215 (16.7)	1529 (21.1)	261 (3.6)	0.020
No	28,445 (79.7)	16,674 (58.6)	5365 (18.9)	5568 (19.6)	838 (3.0)
Center
Center A	18,406 (51.5)	10,100 (54.9)	3352 (18.2)	4473 (24.3)	481 (2.6)	<0.001
Center B	17,304 (48.5)	10,834 (62.6)	3228 (18.7)	2624 (15.2)	618 (3.6)
Endoscopist^b
Trainee	32,194 (90.4)	19,066 (59.2)	5874 (18.3)	6305 (19.6)	949 (3.0)	<0.001
Consultant	3402 (9.6)	1806 (53.1)	688 (20.2)	760 (22.3)	148 (4.4)

^ap-value from the Chi-square for trend to test the association between each variable and the quality of bowel cleansing.

^b Data was missing for 114 sigmoidoscopies; IQR: Interquartile range.

Table 2. Univariate analysis: endoscopy quality metrics and association with bowel cleansing at sigmoidoscopy.

Endoscopy quality metrics		Quality of bowel cleansing at sigmoidoscopy
Sigmoidoscopy	Total No. 35,710	Excellent No. 20,934	Good No. 6580	Partly poor No. 7097	Poor No. 1099	p^e
Mean duration in minutes	7.5	7.6	7.4	7.3	6.3	<0.001
Complete examination ^a	31,580 (88.4)	18,899 (90.3)	5884 (89.4)	6248 (88.0)	549 (50.0)	<0.001
Referral for colonoscopy	3344 (9.4)	1739 (8.3)	628 (9.5)	779 (11.0)	198 (18.0)	<0.001
At least one adenoma detected^b	5797 (16.2)	3546 (17.0)	1085 (16.6)	1024 (14.5)	142 (13.0)	<0.001
Non-advanced adenoma^b^,c	4522 (12.7)	2805 (13.5)	846 (12.9)	780 (11.0)	91 (8.3)	<0.001
Advanced adenoma^b^,d	1275 (3.6)	741 (3.6)	239 (3.7)	244 (3.5)	51 (4.7)	0.510
Adenocarcinoma	165 (0.5)	96 (0.5)	31 (0.5)	31 (0.4)	7 (0.6)	0.809
Sigmoidoscopy and follow-up colonoscopy	Total No. 3267	Excellent No. 1692	Good No. 620	Partly poor No. 760	Poor No. 195	p^e
At least one adenoma detected^b^,e	2690 (87.5)	1382 (87.1)	523 (89.7)	626 (86.9)	159 (85.0)	0.669
Non-advanced adenoma ^b,c,e	1026 (33.4)	474 (29.9)	208 (35.7)	259 (36.0)	85 (45.4)	<0.001
Advanced adenoma ^b,d,e	1664 (54.1)	908 (57.3)	315 (54.0)	367 (51.0)	74 (39.6)	<0.001
Adenocarcinoma^e	191 (5.9)	106 (6.3)	37 (6.0)	40 (5.3)	8 (4.1)	0.164

^a 35 cm from anus with a straightened scope or descending-sigmoid junction judged by use of a magnetic endoscopic imaging system.

^b Individuals with adenocarcinomas were excluded from the analysis;

^c Adenoma with diameter < 10 mm, no high-grade dysplasia, and no villous component;

^d Adenoma with diameter ≥ 10 mm and/or with high-grade dysplasia and/or villous component

^e Combined detection rate at sigmoidoscopy and colonoscopy;

^f p-value for trend. Column percentages are reported in parentheses.

Table 3. Multivariable analysis: referral for colonoscopy and adenoma detection at sigmoidoscopy and follow-up colonoscopy.

Variables		Sigmoidoscopy (n = 35,710)		Sigmoidoscopy and follow-up colonoscopy (n = 3076) ^b
		Referral to colonoscopy OR (95% CI)	Adenoma detection ^a OR (95% CI)	Adenoma detection OR (95% CI)	Advanced adenoma detection OR (95% CI)
Age	5-year increase	1.23 (1.19–1.26)	1.13 (1.11–1.16)	1.13 (1.04–1.22)	1.06 (1.00–1.12)
Sex	Women	Ref.	Ref.	Ref.	Ref.
	Men	1.80 (1.67–1.94)	1.72 (1.62–1.82)	1.71 (1.38–2.13)	1.06 (0.91–1.23)
Symptoms	No	Ref.	Ref.	Ref.	Ref.
	Yes	1.37 (1.26–1.50)	1.07 (1.00–1.16)	0.82 (0.641.05)	1.07 (0.90–1.28)
Endoscopist	Trainee	Ref.	Ref.	Ref.	Ref.
	Consultant	1.10 (0.97–1.24)	0.99 (0.90–1.10)	0.80 (0.57–1.13)	1.36 (1.07–1.74)
Quality of bowel cleansing at sigmoidoscopy	Excellent	Ref.	Ref.	Ref.	Ref.
	Good	1.05 (0.96–1.16)	0.89 (0.83–0.96)	1.21 (0.89–1.65)	0.87 (0.72–1.06)
	Partly poor	1.24 (1.14–1.36)	0.76 (0.71–0.82)	0.92 (0.71–1.20)	0.74 (0.62–0.89)
	Poor	2.10 (1.78–2.47)	0.66 (0.55–0.79)	0.80 (0.51–1.23)	0.47 (0.34–0.64)

Models included age, sex, presence of symptoms, screening centre, endoscopist’s experience and quality of bowel cleansing. Odds ratios (OR) and 95% confidence intervals (CI) were reported.

^a Individuals with adenocarcinomas detected at sigmoidoscopy were excluded from the analysis;

^b Individuals with adenocarcinomas were excluded from the analysis; includes all adenomas at the initial sigmoidoscopy screening examination and subsequent follow-up colonoscopy until all polyps were removed; additionally adjusted for bowel cleansing quality at colonoscopy.

Figure 1. Quality of bowel cleansing and adenoma detection rate at sigmoidoscopy by individual endoscopists.

Footnote: Twenty endoscopists (E1 to E20) who performed at least 500 flexible sigmoidoscopies were selected. Individuals with adenocarcinomas detected at sigmoidoscopy were excluded from the calculation of ADR (mean ADR 16.3%, dotted horizontal line).

Data availability statement

Access to research data for external investigators, or use outside of the current protocol, will require approval from the Norwegian Regional Committee for Medical and Health Research Ethic and the Bowel cancer screening in Norway steering committee (information available on the project website https://www.kreftregisteret.no/screening/Tarmscreeningpiloten/). Research data are not openly available because of the principles and conditions set out in articles 6[1] (e) and 9 [2] (j) of the General Data Protection Regulation (GDPR). Questions regarding data access should be directed to the corresponding author, Anna Lisa Schult, [email protected].