Factors predicting conversion from colon capsule endoscopy to conventional optical endoscopy: findings from the CESCAIL Study

Le résumé fourni par la source

Background Colon capsule endoscopy (CCE) is an established non-invasive alternative to colonoscopy for selected low-risk patients. However, its diagnostic limitations, including incomplete transit, and an inability to biopsy or perform polypectomy, often lead to conversion to conventional colonoscopy (CCC). Understanding the factors contributing to CCC is critical to optimise patient selection, procedural efficiency and cost-effectiveness. Objectives This study aimed to identify and characterise the pre- and intra-procedural factors predicting CCC, including issues with bowel preparation, capsule excretion, pathology detection (polyp number and size) and other clinical or technical variables that influence the need for follow-up investigations. Methods This prospective analysis is nested within the CESCAIL study, involving patients who underwent CCE between November 2021 and June 2024. Data were analysed from both symptomatic and post-polypectomy surveillance cohorts. Variables included demographics, comorbidities, medications and laboratory values. Statistical modelling employed LASSO selection, followed by multivariate logistic and linear regression. Sensitivity analyses were conducted using both complete case analysis (CCA) and multiple imputation by chained equations (MICE). Results A total of 603 participants were included. The CCC rate was 54.1% (326/603), with colonoscopy accounting for 32.9% and flexible sigmoidoscopy for 21.2% of follow-up procedures. Among intra-procedural contributors, pathology detection during CCE alone led to 145 CCC cases (24%), making it the leading cause. Inadequate bowel preparation and capsule battery depletion also contributed to CCC. Pre-procedural predictors of CCC included elevated log-transformed faecal haemoglobin (f-Hb) levels (OR=1.48, 95% CI: 1.18–1.86, p<0.001) and current smoking status (OR=1.44, 95% CI: 1.01–2.11, p=0.047). The adjusted predictive accuracy of f-Hb for CCC was modest (AuROC=0.62), reflecting the low-risk nature of this cohort, selected using an f-Hb threshold of ≤100 µg/g. Men had significantly higher capsule excretion rates (OR=2.22, 95% CI: 1.10–4.58, p=0.024), but this did not translate to lower CCC, likely because of their higher rate of advanced polyp detection. While male sex was associated with better bowel cleansing on univariate analysis (OR=1.46, p=0.023), this was not significant in multivariable analysis (p=0.084). Bowel preparation quality was significantly poorer in patients with type 2 diabetes mellitus (OR=0.40, 95% CI: 0.18–0.87, p=0.022), consistent with previous colonoscopy literature. Factors associated with pathology detection included alcohol consumption (p=0.023), smoking (p=0.025), psychological conditions (p=0.013) and haemoglobin level (p=0.046) for polyp number; and antidepressant (p=0.028) and beta-blocker use (p=0.008) for polyp size. Larger polyps were more commonly observed in antidepressant users, although the underlying mechanism remains unclear. Conversely, beta-blocker use was associated with smaller polyps and lower CCC rates, aligning with some prior findings suggesting a potential tumour-suppressive role. Conclusion Non-smokers with lower f-Hb levels are less likely to need CCC. Patient selection criteria are key to minimising the colonoscopy conversion rate. Our findings would benefit from validation in different populations to develop a robust CCE Conversion Scoring System (CECS) and ultimately improve the cost-effectiveness of treatment.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.