Factors Associated With Recurrence of Advanced Colorectal Adenoma After Endoscopic Resection

• Published in: Clinical gastroenterology and hepatology Vol. 14; no. 8; pp. 1148 - 1154.e4
• Main Authors: Facciorusso, Antonio, Di Maso, Marianna, Serviddio, Gaetano, Vendemiale, Gianluigi, Spada, Cristiano, Costamagna, Guido, Muscatiello, Nicola
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.08.2016
• Subjects: ACA; Adenoma - surgery; Adult; Aged; Aged, 80 and over; Colon Cancer; Colorectal Neoplasms - surgery; CRC; Endoscopic Mucosal Resection - methods; Female; Gastroenterology and Hepatology; Hospitals, University; Humans; Italy; Machine Learning; Male; Middle Aged; Recurrence; Retrospective Studies; Risk Assessment; RPA; HGD; RF; CRC; Colon Cancer; recursive partitioning analysis; colorectal carcinoma; random forest; high-grade dysplasia; Machine Learning; advanced colorectal adenoma; ACA
• ISSN: 1542-3565; 1542-7714
• DOI: 10.1016/j.cgh.2016.03.017

Background & Aims Studies have identified risk factors for recurrence of advanced colorectal adenoma (ACA) after polypectomy, but the relative importance and interaction of these risk factors, and their potential impact on surveillance recommendations, are unclear. We aimed to develop a model to identify ACA features associated with risk of recurrence after polypectomy. Methods In a retrospective study, we collected data from 3360 patients who underwent colonoscopy with polypectomy at University of Foggia from 2004 through 2008 and identified 746 patients with 1017 ACAs. We performed recursive partitioning analysis to identify factors associated with recurrence of ACA within 3 years after polypectomy. Results Median ACA size was 16 mm (range, 8–34 mm) and median number was 1.5 (range, 1–2). Pedunculated, sessile, and nonpolypoid lesions accounted for 41.3%, 39.4%, and 19.3% of ACAs detected, respectively. Factors independently associated with local recurrence of ACA and metachronous distant polyps within 3 years after polypectomy included size and number of ACAs and grade of dysplasia. The recurrence rate was 4.2% in patients with a single ACA ≤15 mm without high-grade dysplasia (HGD), 21.3% in patients with HGD ≤15 mm, ACA without HGD >15 mm, or multiple ACAs without HGD ≤15 mm, and 57.9% in patients with HGD >15 mm. Conclusions In this retrospective analysis of 746 patients with ACA who underwent polypectomy and surveillance colonoscopy within 3 years, the recurrence rate was highest in those with HGD ≥15 mm. These patients might benefit from more intensive surveillance, whereas patients with a single ACA without HGD ≤15 mm are at lower risk for and could be considered for longer follow-up intervals.