Therapeutic profitability of centralising the evaluation and treatment of complex polyps

Dear Editor,

Colorectal cancer is the most common tumour in western countries and the second leading cause of death due to cancer.¹ Preventive strategies are based on detection and endoscopic resection of the precursor lesion, which may be an adenoma or an advanced serrated lesion.² However, large or complex lesions may require surgical treatment³ which carries an associated risk of over-treatment. At our centre, we considered whether centralised evaluation of said lesions (video recording in diagnostic colonoscopy and subsequent evaluation in a joint session by trained endoscopists) plus the creation of a specific agenda for endoscopic treatment with specialised endoscopists could improve therapeutic yield and reduce the number of surgeries performed. The objective of this scientific letter is to present the results of this initiative.

We collected data from 573 patients with complex polyps in the gastrointestinal tract in joint sessions with specialised endoscopists between October 2012 and September 2016. For each patient, the following were recorded: lesion characteristics (morphology according to the Paris classification and size), lesion location and decision made in the session. For the lesions that were resected, we recorded the resection technique: endoscopic (polypectomy, mucosectomy or submucosal dissection) or surgical (open, laparoscopic or transanal resection). For endoscopic resections, we recorded whether the resection had been complete and in a single block. We also evaluated the indication for surgical procedures and the eventual histological diagnosis. In those patients for whom endoscopic follow-up was available, we recorded whether recurrence was detected, lesion histology and treatment performed.

The mean patient age was 70.3 ± 11 years, and the patients were predominantly men (62.3%). The lesions evaluated were located in the colon (477), stomach and/or duodenum (49) or rectum (47). The mean size was 23.2 ± 14 mm, and the predominant morphologies were sessile (36.4%), pedunculated (31.6%) and flat (31.2%). In the evaluation session, endoscopic resection was proposed in 548 patients, surgery in 22 patients and no procedure in three patients. Of the patients referred for endoscopic resection, endoscopic resection was achieved in 498 patients (65% with complete resection in a single block, 30.5% with complete but fragmented resection and 4.2% with incomplete resection). The most common endoscopic resection technique was conventional polypectomy (80.7%), followed by mucosectomy (18.9%) and submucosal dissection (0.4%). Of the lesions resected, 80.3% were adenomas, 8.6% were serrated lesions and invasive adenocarcinomas were detected in 37 (7.4%). Complications were detected in 41 (8.2%) patients: 33 cases of haemorrhage, three cases of perforation, four cases of post-polypectomy syndrome and one case of splenic haematoma. During the follow-up of 275 patients, 51 (18.5%) cases of recurrence were detected; 70.6% of these cases were treated endoscopically. The histology of the recurrence was adenomatous in 43 cases and invasive adenocarcinoma in two cases.

Finally, 83 patients underwent surgery: 63 patients for lesions that could not be resected endoscopically, 10 patients for invasive adenocarcinomas, nine patients for incomplete resections and one patient for perforation. The most commonly used approaches were open colectomy (47%), laparoscopy (37.3%) and transanal resection (15%). The histology of the resected lesions was benign in 45 patients (adenoma 40, serrated lesion two, no lesion two) and malignant in 37. Mild complications developed in 29.9% of patients and serious complications developed in 12.0% of patients who underwent surgery, and four patients died (4.8%).

The data described above indicate that centralised evaluation of “difficult” lesions along with the creation of agendas for endoscopic treatment with specialised endoscopists may be a strategy for optimising the treatment of complex endoscopic lesions whereby the rate of endoscopic resection is increased and the need for surgical treatment is decreased. This is especially important in the context of colorectal cancer population health screening programmes. In fact, one of the most significant risks associated with the implementation thereof is over-treatment of benign lesions. Continued monitoring of outcomes enables detection of areas for improvement. In this regard, although this strategy implemented in our centre enabled 87% of the lesions evaluated to be treated endoscopically, 55% of the lesions that required surgery were benign. Nevertheless, we found a lower rate of complications than those reported in the literature⁴ and a percentage of recurrence similar to the percentages seen in a recently published meta-analysis.⁵