RISK OF COLORECTAL CARCINOMA IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE AND DYSPLASIA UNDERGOING SURGERY AFTER CHROMOENDOSCOPY
Iktej S. Jabbal*, Luca Stocchi, Amit Merchea, Dorin Colibaseanu, Michael F. Picco, John R. Cangemi, Francis A. Farraye
Mayo Clinic's Campus in Florida, Jacksonville, FL
Background: Current guidelines recommend endoscopic excision of dysplastic lesions in patients with inflammatory bowel disease (IBD) when possible. Chromoendoscopy (CE) can facilitate dysplasia detection and optimize surveillance. However, outcomes of surgery with a preoperative diagnosis of colorectal dysplasia after CE are poorly known. We aimed to investigate disease characteristics, endoscopic findings, surgical variables and rate of invasive cancer among such patients.
Methods: A retrospective review of patients with IBD undergoing CE between 2006 and 2019 with a diagnosis of colorectal dysplasia in a three-site health system was performed. Dysplasia diagnosis was either based on previous white-light endoscopic biopsies or established after initial CE. Patients with a preoperative diagnosis of colorectal cancer or surgery unrelated to IBD were excluded.
Results: A total of 393 patients with dysplasia (68% with ulcerative colitis) undergoing a mean of 7 colonoscopies and 2 CEs were identified. The mean time from IBD diagnosis to first CE was 17.6 years. CE was unremarkable in 170 patients who were recommended continued endoscopic surveillance while 219 patients had dysplasia on CE. Out of these, 158 underwent complete endoscopic excision of dysplasia, 22 had unresectable lesions or incomplete excision and 39 had invisible dysplasia. After discussion of treatment options, 45 patients (11.4%) were offered and underwent surgery. Main indications for surgery included multifocal dysplasia (n=19), endoscopically unresectable lesions (n=13), refractory IBD (n=6, 4 with unremarkable CE), high-grade dysplasia (HGD) (n=5) and symptomatic stricture (n=2). Initial operations performed included total proctocolectomy and end ileostomy (n=17), total proctocolectomy and J-pouch anal anastomosis (n=13), total abdominal colectomy and ileorectal anastomosis (n= 8), total abdominal colectomy and end ileostomy (n=3), segmental colectomy and end ileostomy (n=2), proctectomy/proctosigmoidectomy and colostomy (n=2). Pathologic examination of the specimen revealed no dysplasia or cancer in 18 patients, dysplasia in 23 patients at 35 sites (7 with HGD), and invasive cancer in 5 patients (stage I in one, stage II in 3 and Stage III in one patient, respectively). One surgical specimen contained synchronous carcinoma and low-grade dysplasia (LGD). Cancer rates based on the preoperative dysplasia grade were 9.3% (3/32) for LGD, 9% (1/11) for HGD and 50% (1/2) for indefinite dysplasia. The patient with stage 3 disease underwent postoperative chemotherapy. No patient had evidence of cancer recurrence after a mean follow-up of 21.5 months.
Conclusion: Endoscopic management of dysplasia in IBD including CE is possible in the majority of cases. If endoscopic management of dysplasia is not possible, a cancer with locoregional lymph node involvement is unlikely.
Back to 2021 Abstracts