Predictors of En Bloc, R0 Resection, and Postprocedural Complications After Advanced Endoscopic Resections for Colorectal Neoplasms: Results of 1213 Procedures.

BACKGROUND: Advanced endoscopic resection techniques are used to treat colorectal neoplasms that are not amenable to conventional colonoscopic resection. Literature regarding the predictors of the outcomes of advanced endoscopic resections, especially from a colorectal surgical unit, is limited. OBJECTIVE: To determine the predictors of short-term and long-term outcomes after advanced endoscopic resections. DESIGN: Retrospective case series. SETTINGS: Tertiary care center. PATIENTS: Patients who underwent advanced endoscopic resections for colorectal neoplasms from November 2011 to August 2022. INTERVENTIONS: Endoscopic mucosal resection, endoscopic submucosal dissection, hybrid endoscopic submucosal dissection, and combined endoscopic laparoscopic surgery. MAIN OUTCOME MEASURES: Predictors of en bloc and R0 resection, bleeding, and perforation were determined using univariable and multivariable logistic regression models. Cox regression models were used to determine the predictors of tumor recurrence. RESULTS: A total of 1213 colorectal lesions from 1047 patients were resected (median age 66 [interquartile range, 58-72] years, 484 women [46.2%], median BMI 28.6 [interquartile range, 24.8-32.6]). Most neoplasms were in the proximal colon (898; 74%). The median lesion size was 30 (interquartile range, 20-40; range, 0-120) mm. Nine hundred eleven lesions (75.1%) underwent previous interventions. The most common Paris and Kudo classifications were 0 to IIa flat elevation (444; 36.6%) and IIIs (301; 24.8%), respectively. En bloc and R0 resection rates were 56.6% and 54.3%, respectively. Smaller lesions, rectal location, and procedure type (endoscopic submucosal dissection) were associated with significantly higher en bloc and R0 resection rates. Bleeding and perforation rates were 5% and 6.6%, respectively. Increased age was a predictor for bleeding (OR 1.06; 95% CI, 1.03-1.09; p < 0.0001). Lesion size was a predictor for perforation (OR 1.02; 95% CI, 1.00-1.03; p = 0.03). The tumor recurrence rate was 6.6%. En bloc (HR 1.41; 95% CI, 1.05-1.93; p = 0.02) and R0 resection (HR 1.49; 95% CI, 1.11-2.06; p = 0.008) were associated with decreased recurrence risk. LIMITATIONS: Single-center, retrospective study. CONCLUSIONS: Outcomes of advanced endoscopic resections can be predicted by patient-related and lesion-related characteristics. See Video Abstract . PREDICTORES DE LA RESECCION R, EN BLOQUE Y LAS COMPLICACIONES POR RESECCIONES ENDOSCPICAS AVANZADAS EN CASOS DE NEOPLASIA COLORRECTAL RESULTADOS DE PROCEDIMIENTOS: ANTECEDENTES:Las técnicas avanzadas de resección endoscópica se utilizan para el tratamiento de neoplasias colorrectales que no son susceptibles de resección colonoscópica convencional. La literatura sobre los predictores de los resultados de las resecciones endoscópicas avanzadas, especialmente en una unidad de cirugía colorrectal, es limitada.OBJETIVO:Determinar los predictores de resultados a corto y largo plazo después de resecciones endoscópicas avanzadas.DISEÑO:Serie de casos retrospectivos.LUGAR:Centro de tercer nivel de atención.PACIENTES:Pacientes sometidos a resecciones endoscópicas avanzadas por neoplasias colorrectales desde noviembre de 2011 hasta agosto de 2022.INTERVENCIÓNES:Resección endoscópica de la mucosa, disección endoscópica submucosa (ESD), ESD híbrida, cirugía laparoscópica endoscópica combinada.PRINCIPALES MEDIDAS DE RESULTADO:Los predictores de resección en bloque y R0, sangrado y perforación se determinaron mediante modelos de regresión logística univariables y multivariables. Se utilizaron modelos de regresión de Cox para determinar los predictores de recurrencia del tumor.RESULTADOS:Se resecaron 1.213 lesiones colorrectales en 1.047 pacientes [edad media 66 (58-72) años, 484 (46,2%) mujeres, índice de masa corporal medio 28,6 (24,8-32,6) kg/m 2 ]. La mayoría de las neoplasias se encontraban en el colon proximal (898, 74%). El tamaño medio de la lesión fue de 30mm (RIC: 20-40, rango: 0-120). 911 (75,1%) lesiones tenían intervenciones previas. Las clasificaciones de París y Kudo más comunes fueron 0-IIa elevación plana (444, 36,6%) y III (301, 24,8%), respectivamente. Las tasas de resección en bloque y R0 fueron del 56,6% y 54,3%, respectivamente. Las lesiones más pequeñas, la ubicación rectal y el tipo de procedimiento (ESD) se asociaron con tasas de resección en bloque y R0 significativamente más altas. Las tasas de sangrado y perforación fueron del 5% y 6,6%, respectivamente. La edad avanzada [1,06 (1,03-1,09), p < 0,0001] fue un predictor de sangrado. El tamaño de la lesión [1,02 (1,00-1,03), p = 0,03] fue un predictor de perforación. La tasa de recurrencia del tumor fue del 6,6%. En bloque [HR 1,41 (IC 95% 1,05-1,93), p = 0,02] y la resección R0 [HR 1,49 (IC 95% 1,11-2,06), p = 0,008] se asociaron con un menor riesgo de recurrencia.LIMITACIONES:Estudio unicéntrico, retrospectivo.CONCLUSIONES:Los resultados de las resecciones endoscópicas avanzadas pueden predecirse según las características del paciente y de la lesión. (Traducción-Dr. Xavier Delgadillo ).

Risk Factors for Microscopic Disease Positivity at Ileocolic Resection Margins for Crohn's Disease.

INTRODUCTION: Interest in microscopic margin positivity during surgical resection of medical-refractory Crohn's disease has been renewed with multiple recent studies showing an association between microscopic margin positivity with disease recurrence. Our aim was to determine risk factors for microscopic margin disease positivity following ileocolic resection (ICR). MATERIALS AND METHODS: A prospectively-maintained database of patients with Crohn's disease undergoing ICR at a tertiary-referral center was queried. Margin positivity was defined as the presence of cryptitis, erosion, transmural inflammation with lymphoid aggregates, or architectural distortion at either ileal (proximal) or colonic (distal) margins. RESULTS: Amongst 584 patients, 97 patients had a positive microscopic margin (17%) of which 46% had a positive proximal margin, 17% had a positive distal margin, and 13% had both positive and distal margins. Using multivariable logistic regression analysis, index ICR was associated with less odds of positive margin (odds ratio [OR] 0.46, 95% confidence interval [CI] 0.24-0.89, p=0.02), and granuloma presence was associated with increased odds (OR 2.26, 95% CI 1.23-4.21, p=0.01). CONCLUSION: We found that repeat ileocolic resection and granuloma presence were predictors of microscopic margin disease.

Perioperative outcomes of ileorectal anastomosis - an analysis of 823 patients.

AIM: Ileorectal anastomosis (IRA) following total abdominal colectomy (TAC) allows for resortation of bowel continuity but prior studies have reported rates of anastomotic leak (AL) to be as high as 23%. We aimed to report rates of AL and complications in a large cohort of patients undergoing IRA. We hypothesized that AL rates were lower than previously reported and that selective use of diverting loop ileostomy (DLI) is associated with decreased AL rates. METHOD: Patients undergoing TAC or end-ileostomy reversal with IRA, with or without DLI, between 1980 and 2021 were identified from a prospectively maintained institutional database and retrospectively analysed. Redo IRA cases were excluded. Short-term (30-day) surgical outcomes were collected using our database. AL was defined using a combination of imaging and, in the case of return to the operating room, intraoperative findings. RESULTS: Of 823 patients in the study cohort, DLI was performed in 27% and performed more frequently for constipation and inflammatory bowel disease. The overall AL rate was 3% (1% and 4% in those with and without DLI, respectively) and diversion was found to be protective against leak (OR 0.28, 95% CI 0.08-0.94, p = 0.04). However, patients undergoing diversion had a higher overall rate of postoperative complications (51% vs. 36%, p < 0.001) including superficial wound infection, urinary tract infection, dehydration, blood transfusion and portomesenteric venous thrombosis (all p < 0.04). CONCLUSION: Our study represents the largest series of patients undergoing IRA reported to date and demonstrates an AL rate of 3%. While IRA appears to be a viable surgical option for diverse indications, our study underscores the importance of careful patient selection and thoughtful consideration of staging the anastomosis and temporary faecal diversion when necessary.