HIGH-QUALITY CLEANSING IMPROVES LESION DETECTION DURING COLONOSCOPY COMPARED TO ADEQUATE CLEANSING: POST HOC ANALYSIS OF 1170 CENTRAL-READER ASSESSED PATIENTS IN THREE RANDOMISED PHASE 3 TRIALS

Aims:

Effective colonoscopy requires successful bowel preparation. ‘Adequate’ preparation allows detection of lesions > 5 mm and has been defined as a Boston Bowel Preparation Scale (BBPS) score of ≥2 per segment. This post hoc analysis determined the relationship between BBPS scores and adenoma and polyp detection rates in patients who had identical scores in each of the three colon segments, using pooled data from three identically designed Phase 3, multicentre, randomised trials.

Methods:

Analysed patients had: treatment-blinded central reader assessed BBPS scores where the BBPS score was identical in each segment; and recorded polyp and adenoma counts. Patients were stratified to assess the relationship between low-quality cleansing (BBPS of 1+1+1), adequate cleansing (BBPS of 2+2+2) or high-quality cleansing (BBPS of 3+3+3), and ADR and PDR. A logistic regression (LR) model was fitted to assess whether ≥1 polyps and, separately, ≥1 adenomas, were detected including the BBPS score group and study as fixed effects.

Results:

1170 patients were analysed (Table 1). For low, adequate and high-quality cleansing, the PDRs were: 40.7%, 41.7%, 54.8%, and the ADRs were: 27.8%, 26.0% and 42.8%. The odds ratio (OR) for detection were significantly higher with high-quality versus adequate cleansing (PDR OR: 1.60, 95% confidence interval [CI] 1.14 – 2.24, P = 0.0067; ADR OR: 1.97, 95% CI 1.39 – 2.80, P = 0.0001). LR analysis revealed a significant correlation between BBPS score and PDR (P = 0.0239) and ADR (P = 0.0006).

Tab. 1:

PDR and ADR by uniform BBPS score

Overall colon lesion detection rates	High-Quality (N = 166)	Adequate (N = 950)	Low-Quality (N = 54)	Odds ratio High-Quality: Adequate (95% CI) [P-value]	Odds ratio Adequate: Low-Quality (95% CI) [P-value]
PDR, n (%)	91 (54.8)	396 (41.7)	22 (40.7)	1.60 (1.14 – 2.24) [0.0067]	0.90 (0.50 – 1.60) [0.7104]
ADR, n (%)	71 (42.8)	247 (26.0)	15 (27.8)	1.97 (1.39 – 2.80) [0.0001]	0.75 (0.39 – 1.43) [0.3829]
LR Analysis of Association between Uniform BBPS Score Group and PDR, P-value	0.0239
LR Analysis of Association between Uniform BBPS Score Group and ADR, P-value	0.0006

Conclusions:

In patients who had uniform low, adequate or high-quality cleansing in each respective colon segment, there was a strong correlation between higher cleansing scores and increased PDR and ADR.