Long-term cumulative incidence of metachronous advanced colorectal neoplasia after colonoscopy and a novel risk factor: a cohort study.

BACKGROUND AND AIM: Long-term cumulative incidence of and risk factors for metachronous advanced colorectal neoplasia, including both advanced colorectal adenoma (≥10 mm, or with villous or high-grade dysplasia) and colorectal cancer, are critical for surveillance strategies. The aim of this study was to determine the cumulative incidence of metachronous advanced colorectal neoplasia and its risk factors. METHODS: A retrospective cohort study was conducted on 6720 consecutive individuals who underwent general health check-ups and colonoscopy. Colorectal adenomas at initial colonoscopy were categorized as low-risk (1-2 small [<10 mm] tubular adenomas) or high-risk adenoma (≥3 tubular adenomas of any size; at least one adenoma ≥10 mm; or villous adenoma or adenoma with high-grade dysplasia). Kaplan-Meier estimates and hazard ratio by Cox-proportional hazard regression were calculated. RESULTS: The cumulative incidence (95% confidence interval [CI]) of metachronous advanced colorectal neoplasia at 5 and 10 years was 5.7% [4.6-7.1], and 11% [8.9-14] in the low-risk adenoma group, and 10% [8.6-13], and 17% [14-21] in high-risk adenoma group, respectively. Adjusted hazard ratio [95% CI] of low-risk adenoma (vs. no colorectal adenoma), high-risk adenoma (vs. no colorectal adenoma), current smoking and positive fecal immunochemical test were 1.34 [1.04-1.74], 1.94 [1.48-2.55], 1.55 [1.2-2.02] and 1.69 [1.35-2.1], respectively. Adjusted hazard ratio [95% CI] of positive fecal immunochemical test was 1.88 [1.29-2.74] in those with normal colonoscopy. CONCLUSIONS: Both low-risk and high-risk adenomas confer substantial risk for metachronous advanced colorectal neoplasia at 10 years. Positive fecal immunochemical test was a significant risk factor for metachronous advanced colorectal neoplasia despite normal colonoscopy.

Noneffectiveness of scopolamine for facilitating detection of upper gastrointestinal neoplasia during screening esophagogastroduodenoscopy: propensity score-matched study.

BACKGROUND: Antispasmodics, such as scopolamine, are widely used in several countries prior to diagnostic and screening esophagogastroduodenoscopy (EGD), with the goal of optimizing the detection of minute lesions, typically early gastric cancer (T1 lesions). The aim of this study was to determine whether scopolamine facilitates detection of gastric cancer in the screening setting. METHODS: A propensity score-matched retrospective study was conducted in a tertiary referral medical center in Tokyo, Japan. Consecutive individuals (n = 40 776) underwent screening EGD between January 2011 and May 2016. All outcome lesions were diagnosed with histopathological confirmation. Detection of esophageal cancer, gastric adenoma, duodenal adenoma, and upper gastrointestinal neoplasia (UGIN) were investigated as secondary outcomes. RESULTS: Scopolamine was used in 31 130 patients (76.3 %) and propensity score matching yielded 6625 pairs. Bivariate analysis revealed no significant association between possible confounders (age, sex, overweight, atrophic gastritis, alcohol history, smoking history, midazolam use, endoscopist biopsy rate grade, and gastric cancer in first-degree relatives) and scopolamine use. Lesions detected were 18 gastric cancers, 11 esophageal cancers, 19 gastric adenomas, 6 duodenal adenomas, and 54 UGINs, with no significant association between scopolamine use and lesion detection. CONCLUSIONS: Scopolamine use did not appear to effectively facilitate detection of gastric or esophageal cancer, gastric or duodenal adenoma, and UGIN during screening EGD. Scopolamine should be avoided until its efficacy is confirmed by a randomized controlled trial.