Search

Surgical site infection is a source of significant morbidity after colorectal surgery. Previous efforts to develop models that predict surgical site infection have had limited accuracy. Machine learning has shown promise in predicting post-operative outcomes by identifying non-linear patterns within large datasets.We sought to use machine learning to develop a more accurate predictive model for colorectal surgical site infections.Patients who underwent colorectal surgery were identified in the American College of Surgeons National Quality Improvement Program database from years 2012-2019 and were split into training, validation, and test sets. Machine learning techniques included random forest, gradient boosting, and artificial neural network. A logistic regression model was also created. Model performance was assessed using area under the receiver operating characteristic curve.A national, multicenter dataset.Patients who underwent colorectal surgery.The primary outcome (surgical site infection) included patients who experienced superficial, deep, or organ-space surgical site infections.The dataset included 275,152 patients after application of exclusion criteria. 10.7% of patients experienced a surgical site infection. Artificial neural network showed the best performance with area under the receiver operating characteristic curve of 0.769 (95% CI 0.762 - 0.777), compared with 0.766 (95% CI 0.759 - 0.774) for gradient boosting, 0.764 (95% CI 0.756 - 0.772) for random forest, and 0.677 (95% CI 0.669 - 0.685) for logistic regression. For the artificial neural network model, the strongest predictors of surgical site infection were organ-space surgical site infection present at time of surgery, operative time, oral antibiotic bowel prep, and surgical approach.Local institutional validation was not performed.Machine learning techniques predict colorectal surgical site infections with higher accuracy than logistic regression. These techniques may be used to identify patients at increased risk and to target preventative interventions for surgical site infection. See Video Abstract at http://links.lww.com/DCR/C88.

Background For rectal cancer with unresectable metastases, current practice favors omitting interventions directed at the primary tumor in asymptomatic patients. Objective This study aimed to determine the proportion of patients with primary tumor-related complications, characterize salvage outcomes, and measure survival in patients with metastatic rectal cancer who did not undergo upfront intervention for their primary tumor. Design This is a retrospective analysis. Setting This study was conducted at a comprehensive cancer center. Patients Patients who presented between January 1, 2008, and December 31, 2015, with synchronous stage IV rectal cancer, an unresected primary tumor, and no prior primary tumor-directed intervention were selected. Main outcome measures The main outcome measured was the rate of primary tumor-related complications in the cohort that did not receive any primary tumor-directed intervention. The Kaplan-Meier method and Cox regression analysis were used to determine whether complications are associated with survival. Results The cohort comprised 358 patients with a median age of 56 years (22-92). Median follow-up was 26 months (range, 1-93 months). Among the 168 patients (46.9%) who eventually underwent elective resection of the primary tumor, the surgery was performed with curative intent in 66 patients (18.4%) and preemptive intent in 102 patients (28.5%). Of the 190 patients who did not undergo an upfront or elective intervention for the primary tumor, 68 (35.8%) experienced complications. Nonsurgical intervention for complications was attempted in 34 patients with an overall success rate of 61.8% (21/34). Surgical intervention was performed in 47 patients (including 13 patients for whom nonsurgical intervention failed): diversion in 26 patients and resection in 21 patients. Of those 47 patients, 42 (89.4%) ended up with a colostomy or ileostomy. Limitations This study was conducted at a single center. Conclusion A significant proportion of patients with metastatic rectal cancer and untreated primary tumor experience primary tumor-related complications. These patients should be followed closely, and preemptive intervention (resection, diversion, or radiation) should be considered if the primary tumor progresses despite systemic therapy. See Video Abstract at http://links.lww.com/DCR/B400. COMPLICACIONES RELACIONADAS CON EL TUMOR PRIMARIO Y RESULTADOS DE RESCATE EN PACIENTES CON CANCER DE RECTO METASTASICO Y UN TUMOR PRIMARIO NO TRATADO: Para el cancer de recto con metastasis no resecables, la practica actual favorece la omision de las intervenciones dirigidas al tumor primario en pacientes asintomaticos.Determinar la proporcion de pacientes con complicaciones relacionadas con el tumor primario, caracterizar los resultados de rescate y medir la supervivencia en pacientes con cancer rectal metastasico que no se sometieron a una intervencion inicial para su tumor primario.Analisis retrospectivo.Centro oncologico integral.Pacientes que se presentaron entre el 1 de enero de 2008 y el 31 de diciembre de 2015 con cancer de recto en estadio IV sincronico, un tumor primario no resecado y sin intervencion previa dirigida al tumor primario.Tasa de complicaciones relacionadas con el tumor primario en la cohorte que no recibio ninguna intervencion dirigida al tumor primario. Se utilizo el metodo de Kaplan-Meier y el analisis de regresion de Cox para determinar si las complicaciones estan asociadas con la supervivencia.La cohorte estuvo compuesta por 358 pacientes con una mediana de edad de 56 anos (22-92). La mediana de seguimiento fue de 26 meses (rango, 1 a 93 meses). Entre los 168 pacientes (46,9%) que finalmente se sometieron a reseccion electiva del tumor primario, la cirugia se realizo con intencion curativa en 66 pacientes (18,4%) y con intencion preventiva en 102 pacientes (28,5%). De los 190 pacientes que no se sometieron a una intervencion inicial o electiva para el tumor primario, 68 (35,8%) experimentaron complicaciones. Se intento una intervencion no quirurgica para las complicaciones en 34 pacientes con una tasa de exito global del 61,8% (21 de 34). La intervencion quirurgica se realizo en 47 pacientes (incluidos 13 pacientes en los que fallo la intervencion no quirurgica): derivacion en 26 pacientes y reseccion en 21 pacientes. De esos 47 pacientes, 42 (89,4%) terminaron con una colostomia o ileostomia.Unico centro.Una proporcion significativa de pacientes con cancer de recto metastasico y primario no tratado experimentan complicaciones relacionadas con el tumor primario. Se debe hacer un seguimiento estrecho de estos pacientes y considerar la posibilidad de una intervencion preventiva (reseccion, derivacion o radiacion) si el tumor primario progresa a pesar de la terapia sistemica. Consulte Video Resumen en http://links.lww.com/DCR/B400.

BACKGROUND A watch-and-wait strategy is a nonoperative alternative to sphincter-preserving surgery for patients with locally advanced rectal cancer who achieve a clinical complete response after neoadjuvant therapy. There are limited data about bowel function for patients undergoing this organ-preservation approach. OBJECTIVE The purpose of this study was to compare bowel function in patients with rectal cancer managed with a watch-and-wait approach with bowel function in patients who underwent sphincter-preserving surgery (total mesorectal excision). DESIGN This was a retrospective case-control study using patient-reported outcomes. SETTINGS The study was conducted at a comprehensive cancer center. PATIENTS Twenty-one patients underwent a watch-and-wait approach and were matched 1:1 with 21 patients from a pool of 190 patients who underwent sphincter-preserving surgery, based on age, sex, and tumor distance from the anal verge. MAIN OUTCOME MEASURES Bowel function was measured using the Memorial Sloan Kettering Cancer Center Bowel Function Instrument. RESULTS Patients in the watch-and-wait arm had better bowel function on the overall scale (median total score, 76 vs 55; p < 0.001) and on all of the subscales, with the greatest difference on the urgency/soilage subscale (median score, 20 vs 12; p < 0.001). LIMITATIONS The study was limited by its retrospective design, small sample size, and temporal variability between surgery and time of questionnaire completion. CONCLUSIONS A watch-and-wait strategy correlated with overall better bowel function when compared with sphincter-preserving surgery using a comprehensive validated bowel dysfunction tool. See Video Abstract at http://links.lww.com/DCR/B218. FUNCION EVACUATORIA INFORMADA POR PACIENTES EN CANCER RECTAL MANEJADO CON UNA ESTRATEGIA DE OBSERVAR Y ESPERAR DESPUES DE LA TERAPIA NEOADYUVANTE: UN ESTUDIO DE CASOS Y CONTROLES: Observar y esperar es una alternativa no operativa a la cirugia de preservacion del esfinter para pacientes con cancer rectal localmente avanzado que logran una respuesta clinica completa despues de la terapia neoadyuvante. Hay datos limitados sobre la funcion evacuatoria en pacientes sometidos a este abordaje para preservacion de organos.Evaluar la funcion evacuatoria en pacientes con cancer rectal manejados con observar y esperar comparado a pacientes sometidos a cirugia de preservacion de esfinteres (escision mesorrectal total).Estudio retrospectivo de casos y controles utilizando resultados reportados por pacientes.Centro especializado oncologico.21 pacientes se sometieron a observar y esperar y se compararon con 21 pacientes de un grupo de 190 pacientes que se sometieron a cirugia de preservacion de esfinteres controlando por edad, sexo y la distancia del tumor al borde anal.Funcion evacuatoria utilizando un instrumento de valoracion del Centro de Cancer Memorial Sloan Kettering.Los pacientes de observar y esperar demostraron mejor funcion evacuatoria en la escala general (puntuacion total media, 76 versus 55; p

Evaluating patients for recurrent anal cancer after primary treatment can be difficult owing to distorted anatomy and scarring. Many institutions incorporate endoscopic ultrasound to improve detection, but the effectiveness is unknown.The aim of this study is to compare the effectiveness of digital rectal examination and endoscopic ultrasound in detecting locally recurrent disease during routine follow-up of patients with anal cancer.This study is a retrospective, single-institution review.This study was conducted at an oncologic tertiary referral center.Included were 175 patients with nonmetastatic anal squamous-cell cancer, without persistent disease after primary chemoradiotherapy, who had at least 1 posttreatment ultrasound and examination by a colorectal surgeon.The primary outcomes measured were the first modality to detect local recurrence, concordance, crude cancer detection rate, sensitivity, specificity, and predictive value.Eight hundred fifty-five endoscopic ultrasounds and 873 digital rectal examinations were performed during 35 months median follow-up. Overall, ultrasound detected 7 (0.8%) mesorectal and 32 (3.7%) anal canal abnormalities; digital examination detected 69 (7.9%) anal canal abnormalities. Locally recurrent disease was found on biopsy in 8 patients, all detected first or only with digital examination. Four patients did not have an ultrasound at the time of diagnosis of recurrence. The concordance of ultrasound and digital examination in detecting recurrent disease was fair at 0.37 (SE, 0.08; 95% CI, 0.21-0.54), and there was no difference in crude cancer detection rate, sensitivity, specificity, and negative or positive predictive values.The heterogeneity of follow-up timing and examinations is not standardized in this study but is reflective of general practice.Endoscopic ultrasound did not provide any advantage over digital rectal examination in identifying locally recurrent anal cancer, and should not be recommended for routine surveillance.

BACKGROUND Serrated polyposis syndrome is a rare syndrome associated with an increased risk for colorectal cancer. The World Health Organization criteria were established to standardize the diagnosis and management of patients afflicted with serrated polyposis. Although useful, the criteria may not be ideal for the initial screening of at-risk populations. OBJECTIVE The aim of this study was to examine the use of a minimal cutoff point of serrated lesions to increase the yield of serrated polyposis cases. DESIGN This was a retrospective review of colonoscopy and pathology reports to identify patients who met the World Health Organization criteria for serrated polyposis. SETTING This study was conducted at a tertiary cancer care referral center. PATIENTS Five hundred patients who had at least 2 pathologically confirmed hyperplastic polyps and/or sessile serrated adenomas/polyps diagnosed between 1999 and 2009 were assessed. MAIN OUTCOME MEASURES The primary outcome measure was the number of serrated polyposis cases. RESULTS Forty of the 500 (8%) patients met the World Health Organization criteria for serrated polyposis syndrome. Patients underwent a median of 4 colonoscopies (range, 1-23) before satisfying the criteria, and only 1 (3%) patient met the criteria for diagnosis during the initial colonoscopy. All 16 patients with a history of colorectal cancer were only diagnosed with serrated polyposis either at the time of their cancer diagnosis or during postoperative colonoscopies. Only 5 of the 40 (13%) patients were enrolled in our institutional Hereditary Colorectal Cancer Family Registry before our study for prospective serrated lesion tracking and colorectal cancer screening. LIMITATIONS This tool requires validation in a prospective setting. CONCLUSIONS The cutoff point of at least 2 pathologically confirmed serrated lesions can serve as a screening tool for identifying patients meeting the World Health Organization criteria for serrated polyposis syndrome who would otherwise go undetected.

Perianal Paget's disease (intraepithelial adenocarcinoma) is rare and sometimes difficult to diagnose because symptoms are nonspecific. It is often noninvasive but frequently recurs locally. Invasive disease can metastasize to distant sites.The purpose of this work was to review the diagnosis, management, and outcomes of patients with perianal Paget's disease.Institutional databases were queried for all of the cases of perianal Paget's disease at Memorial Sloan-Kettering Cancer Center between 1950 and 2011. Clinicopathologic factors were investigated for association with recurrence and survival.The study was conducted at a tertiary care center.Sixty-five patients with perianal Paget's disease were included in the study (35 women [54%]; median age at diagnosis, 66 years [range, 60-72 years]; and 41 with invasive disease/24 with noninvasive disease). A total of 56% with invasive disease were men.Measures included median follow-up, disease status, local and distant recurrence, sites of recurrence, disease-specific survival, overall survival, and treatment modality.A total of 95% with invasive disease and 87% with noninvasive disease were symptomatic at presentation. The most common symptoms were pruritus and perianal bleeding. The duration of symptoms was longer in patients with invasive (12.0 months; range, 4.0-18.0 months) versus noninvasive (3.5 months; range, 1.0-10.0 months) disease. Synchronous malignancies unrelated to the primary disease were noted in 5 patients with invasive disease and 3 with noninvasive disease. Noninvasive disease was treated with a wide local excision and invasive disease with a wide local excision (n = 32, 78%) or abdominoperineal resection (n = 9, 22%). Forty-one patients (27 invasive and 14 noninvasive) required multiple operations for tumor clearance. In those with invasive disease, the median time to recurrence was 5 years, and the median tumor-specific survival rate was 10 years.This was a retrospective study, limited by selection bias.Perianal Paget's disease is associated with nonspecific symptoms, frequently delaying diagnosis. Wide local excision is the treatment of choice if negative margins can be obtained. Abdominoperineal resection should be considered for invasive disease. Local recurrence is common; follow-up includes periodic proctoscopy and digital examination. Invasive disease can metastasize to distant sites; follow-up should include the examination of inguinal lymph nodes and the imaging of liver and lungs.

Locoregionally recurrent squamous-cell carcinoma of the anal canal is managed with salvage surgery. High-dose-rate intraoperative radiation therapy has been used in selected patients with this disease to reduce the risk of local recurrence.The aim of this article is to present our institutional experience with this technique.Medical records of 14 patients with locoregionally recurrent squamous-cell carcinoma of the anal canal who underwent this technique between 1992 and 2007 were reviewed.The study was conducted at an academic cancer center.The median age was 45 years (range, 36-77), and 13 of the patients were women. All had prior radiation with or without chemotherapy.The surgical procedures included abdominoperineal resection with or without sacrectomy (n = 8), low anterior resection (n = 2), and pelvic exenteration (n = 4). The median radiation dose was 1500 cGy (range, 1500-1750). All cases of radiographic invasion of adjacent structures correctly predicted pathologic invasion. There was pathologic invasion into adjacent structures in 11 cases (79%), and adherence to the sacrum without invasion in 2 cases (14%). Surgical margins were positive (n = 6), close (1 mm) (n = 3), and negative (n = 5).The median follow-up from our technique was 17 months (range, 5-145). Subsequent recurrence occurred in 11 cases, at a median of 8 months from treatment. Two-year actuarial control was 7.1%, and the overall survival was 21.4%. Acute toxicities included wound-healing complications (n = 6); gastrointestinal obstruction (n = 5); neurogenic bladder (n = 1); ureteral stricture (n = 3); and peripheral neuropathy (n = 2).This is a small retrospective series in which the meaningful analysis of associations between clinical variables and outcomes was not possible.Salvage surgery with high-dose-rate intraoperative radiation therapy did not appear to be associated with a locoregional control or survival benefit in this series. The addition of high-dose-rate intraoperative radiation therapy to salvage surgery is insufficient to compensate for positive surgical margins. Preoperative imaging should be used to aid in patient selection to identify those patients in whom negative margins can be obtained and to aid in the determination of appropriate salvage surgery.

Purpose Several series report higher recurrence after transanal excision of T1 rectal cancer than after radical resection. However, the impact of transanal excision on cancer mortality has not been adequately studied. The purpose of this study was to compare oncologic outcomes of transanal excision with those of radical resection. Methods Patients with transanal excision or radical resection for T1 rectal cancer treated between 1985 and 2004 were identified from a prospective database. Patients receiving preoperative chemotherapy or radiation or with tumors >12 cm from the anal verge were excluded. Results The final cohort comprised 145 radical resections and 137 transanal excisions. The transanal excision group was notable for older mean age (64 vs. 59 years), shorter mean distance from anal verge (5.9 vs. 7.8 cm), and smaller tumor size (2.3 vs. 3.1 cm). Lymphovascular invasion and poor differentiation were similar in both groups. Twenty percent of radical resection specimens had lymph node metastasis. Median follow-up was 5.6 years. Local recurrence was noted in a higher proportion of transanal excision patients (13.2 vs. 2.7 percent, P = 0.001). After transanal excision the hazard ratio for local recurrence was 11.3 (95 percent confidence interval, 2.6-49.2), and disease-specific survival was inferior (87 vs. 96 percent at 5 years, P = 0.03, hazard ratio 2.8 [range, 1.04-7.3]). Conclusions Transanal excision offers inferior oncologic results, including greater risk of cancer-related death. This procedure should be restricted to patients who have prohibitive medical contraindications to major surgery or have made an informed decision to accept the oncologic risks of local excision and avoid the functional consequences of rectal resection.

Familial adenomatous polyposis, caused by mutations in the adenomatous polyposis coli gene located at chromosome 5q21, is an autosomal dominant syndrome characterized by polyposis of the colon and rectum and nearly 100 percent progression to colorectal cancer. We report a case of familial adenomatous polyposis and mental retardation caused by a chromosomal deletion at 5q15-q22. Chromosomal analysis is considered part of the evaluation of children with mental retardation and developmental delay. The resulting karyotypes from high-resolution chromosomal analysis can help characterize large deletions, some of which involve known tumor suppressor genes. Because familial adenomatous polyposis may arise from de novo chromosomal deletions involving the adenomatous polyposis coli gene locus, individuals with chromosomal deletions involving 5q21 should be considered at-risk for familial adenomatous polyposis and offered standard screening with flexible sigmoidoscopy by 10 to 12 years of age.

Sphincter-preserving surgery is technically feasible for many rectal cancers, but functional results are not well understood. Therefore, the purpose of this study was to develop an instrument to evaluate bowel function after sphincter-preserving surgery.A 41-item bowel function survey was developed from a literature review, expert opinions, and 59 patient interviews. An additional 184 patients who underwent sphincter-preserving surgery between 1997 and 2001 were asked to complete the survey and quality-of-life instruments (Fecal Incontinence Quality of Life, European Organization for Research and Treatment of Cancer QLQ 30/Colorectal Cancer 38). A factor analysis of variance was performed. Test-retest reliability was evaluated, with 20 patients completing two surveys within a mean of 11 days. Validity testing was done with clinical variables (gender, age, radiation, length of time from surgery), surgical variables (procedure: local excision, low anterior resection, coloanal anastomosis), reconstruction (J-pouch, straight), anastomosis (handsewn, stapled), and quality-of-life instruments.The survey response rate was 70.1 percent (129/184). Among the 127 patients with usable data, 67 percent were male, the median age was 64 (range, 38-87) years, and the mean time for restoration of bowel continuity after sphincter-preserving surgery was 22.9 months. Patients had a median of 3.5 stools/day (range, 0-30), and 37 percent were dissatisfied with their bowel function. Patients experienced a median of 22 symptoms (range, 7-32), with 27 percent reported as severe, 37 percent as moderate, and 36 percent as mild. The five most common symptoms were incomplete evacuation (96.8 percent), clustering (94.4 percent), food affecting frequency (93.2 percent), unformed stool (92.8 percent), and gas incontinence (91.8 percent). The factor analysis identified 14 items that collapsed into three subscales: FREQUENCY (alpha = 0.75), DIETARY (alpha = 0.78), and SOILAGE (alpha = 0.79), with acceptable test-retest reliability for the three subscales and total score (0.62-0.87). The instrument detected differences between patients with preoperative radiation (n = 67) vs. postoperative radiation (n = 15) vs. no radiation (n = 45) (P = 0.02); local excision (n = 10) vs. low anterior resection (n = 55) vs. coloanal anastomosis (n = 62) (P = 0.002); and handsewn (n = 18) vs. stapled anastomosis (n = 99) (P = 0.006). The total score correlated with 4 of 4 Fecal Incontinence Quality of Life (P0.01) and 9 of 17 European Organization for Research and Treatment of Cancer subscales (all P0.01).Patients undergoing sphincter-preserving surgery for rectal cancer have impaired bowel function, and those treated with radiation, coloanal anastomoses, or handsewn anastomoses have significantly worse function. This reliable and valid instrument should be used to prospectively evaluate bowel function after sphincter-preserving surgery in patients undergoing rectal cancer therapy.

Locoregional recurrence after resection of colon carcinoma is an uncommon and difficult clinical problem. Outcome data to guide surgical management are limited. This investigation was undertaken to review our experience with surgical resection for patients with locoregional recurrence colon cancer, determine predictors of respectability, and define prognostic factors associated with survival. A prospective database was queried for patients who had recurrent colon cancer between January 1991 and October 2002. Patients were selected for analysis if they had either isolated resectable locoregional recurrence or concomitant resectable distant disease. Disease-specific survival analysis was performed with the Kaplan-Meier actuarial method, and factors associated with outcome were determined by the log-rank test and Cox regression. During this period of time, 744 patients with recurrent colon cancer were identified and 100 (13.4 percent) underwent exploration with curative intent for potentially resectable locoregional recurrence: 75 with isolated locoregional recurrence, and 25 with locoregional recurrence and resectable distant disease. The median follow-up for survivors was 27 months. Locoregional recurrence was classified into four categories: anastomotic; mesenteric/nodal; retroperitoneal; and peritoneal. Median survival for all patients was 30 months. Fifty-six patients had an R0 resection (including distant sites). Factors associated with prolonged disease-specific survival included R0 resection (P < 0.001); age

Primary chemoradiation failure for epidermoid carcinoma of the anus is treated by surgical resection. This study evaluates the outcome of salvage surgery at one institution. All patients (n = 177) with a diagnosis of epidermoid carcinoma of the anus undergoing surgery since 1980 were reviewed. After criteria-based exclusion (n = 115), the remaining patients (n = 62) were analyzed. Kaplan-Meier survival analysis was performed on abdominoperineal resection/low anterior resection patients. Variable comparisons were made using log-rank and Cox regression analyses. Inguinal lymph node dissection patients (n = 5) were analyzed separately. Median follow-up was 24.2 months. Actuarial five-year survival in all abdominoperineal resection/low anterior resection patients (n = 57) was 33 percent (median, 34.1 months). Univariate predictors of decreased survival were tumor size > 5 cm or adjacent organ involvement at salvage, positive nodal disease at salvage, and positive margins. Independent predictors of decreased survival were the same except for tumor size or adjacent organ involvement at salvage (not significant). Patients undergoing potentially curative resections (n = 47) had an actuarial five-year survival of 40 percent (median, 49 months). The univariate and multivariate predictors of both decreased survival and recurrence in this subgroup included: disease persistence after chemoradiation and nodal disease at salvage. Tumor size > 5 cm or adjacent organ involvement at salvage predicted recurrence with only univariate analysis. Interestingly, actuarial five-year survival after potentially curative resection for recurrence after chemoradiation was 51 percent (as opposed to 31 percent for persistence). After potentially curative resections, most documented recurrences (79 percent) occurred within two years and were locoregional (74 percent). Actuarial five-year recurrence-free survival was 46 percent. Three of five inguinal lymph node dissection patients were alive without disease at 21.2, 81.7, and 84.3 months. Salvage surgery after failed chemoradiation therapy has a reasonable chance of cure. Favorable independent prognostic factors include recurrence (vs. persistence) after chemoradiation (when salvage is potentially curative), absence of nodal disease at salvage, and negative margins. Salvage inguinal lymph node dissection after failed chemoradiation therapy also is potentially curative.

Recent data suggest a favorable prognosis for rectal cancer patients with a pathologic complete response to preoperative combined modality therapy. Prolongation of the interval between preoperative combined modality therapy and surgery (RT-surgery interval) as a means of increasing pathologic complete response rate has not been fully examined. One hundred and fifty-five rectal cancer patients undergoing preoperative pelvic external beam radiation therapy and 5-fluorouracil-based chemotherapy followed by rectal resection were identified. All patients had endorectal ultrasound prior to combined modality therapy. Final pathology reports were reviewed for ypT and ypN stage and margin status. Medical records were reviewed for sphincter preservation, operative time, estimated blood loss, hospital stay, and morbidity (overall, anastomotic, and perineal). A pathologic complete response (ypT0N0) occurred in 24 patients (15 percent). Median RT-surgery interval was 44 (range, 15-206) days. A pathologic complete response occurred in 19 percent of patients with an interval >44 days, vs. 12 percent in those with an interval ≤44 days (P = 0.27). Downstaging by three stages occurred more frequently in the long-interval group (15 percent vs. 6 percent, P = 0.11). The rates of sphincter preservation, positive margins, estimated blood loss, and operative time were not significantly different. Overall morbidity was similar between groups. Our results demonstrate a trend toward increased pathologic complete response rate and downstaging with increased RT-surgery interval. However, sphincter preservation is not increased. Until prospective analyses are conducted assessing the impact of prolonged RT-surgery interval on long-term outcome, the benefit of a prolonged interval between the completion of preoperative combined modality therapy and surgery remains unclear.

PURPOSE: The purpose of this study was to determine the local control and survival of patients with adenocarcinoma of the anus who received combined modality therapy as a component of their treatment. METHODS: Thirteen patients with primary anal adenocarcinoma (T1: 1, T2: 4, T3: 3, T4: 5; and N0: 9, N1: 2, N2: 2) were treated between 1989 and 2001 in the Department of Radiation Oncology at Memorial Sloan Kettering Cancer Center. Three general treatment approaches were used that were based on physician and patient preference as well as tumor stage. These included preoperative combined modality therapy followed by abdominoperineal resection (n = 5), with four of the five receiving postoperative chemotherapy; local excision followed by postoperative radiation alone or combined modality therapy (n = 5); and abdominoperineal resection followed by postoperative combined modality therapy (n = 3). Two patients received brachytherapy. RESULTS: With a median follow-up of 19 months, the median survival was 26 months, the local failure rate was 37 percent, and the two-year actuarial survival was 62 percent. In the subset of eight patients treated with abdominoperineal resection and preoperative or postoperative radiation or combined modality therapy, local control was 63 percent, and three of eight are without evidence of disease. Of the five patients who underwent a local excision followed by postoperative radiation or combined modality therapy, the local control rate was 60 percent, with one of the local failures salvaged by abdominoperineal resection and one of five patients without evidence of disease. CONCLUSION: Although the experience is limited, our data suggest that the combination of abdominoperineal resection and combined modality therapy is a reasonable approach for this rare tumor.

PURPOSE: The balance between local levels of matrix metalloproteinases and tissue inhibitor of metalloproteinases is believed to play a key role in tumor invasion and metastases. Because tissue inhibitor of metalloproteinase-3 suppresses tumorigenicity and tumor invasionin vitro, the aim of this study was to determine its expression in human colorectal cancer. METHODS: Thirty-nine human colorectal cancer specimens, three adenomas, and matched normal adjacent mucosa from 39 colorectal cancer patients were analyzed. Tissue inhibitor of metalloproteinase-3 ribonucleic acid and protein expression were analyzed by Northern blot hybridization and Western blot analysis, respectively. The cellular localizations of tissue inhibitor of metalloproteinase-3 ribonucleic acid and protein were determined byin situ hybridization and immunolocalization. RESULTS: Tissue inhibitor of metalloproteinase-3 ribonucleic acid expression was increased in colorectal cancer compared with paired normal mucosa. In contrast, tissue inhibitor of metalloproteinase-3 protein level was higher in normal mucosa than in the corresponding colorectal cancer. In addition, tissue inhibitor of metalloproteinase-3 protein levels progressively decreased with advancing colorectal cancer stages. Tissue inhibitor of metalloproteinase-3 protein tumor to normal mucosa ratio was 0.74 ± 0.12, 0.51 ± 0.18, 0.48 ± 0.12, and 0.45 ± 0.2 for Dukes A (n=8), B (n=9), C (n=9), and D (n=13) stages, respectively. Both tissue inhibitor of metalloproteinase-3 messenger ribonucleic acid and protein were located predominantly within spindle-shaped and round stromal cells. Furthermore, in colonic epithelium, tissue inhibitor of metalloproteinase-3 and type IV collagen protein were similarly concentrated in the basal region. CONCLUSIONS: These data provide the first detailed description of the cellular expression of tissue inhibitor of metalloproteinase-3 in colorectal cancer and identify it as a basement membrane-associated protein. This is an important observation, because the presence of tissue inhibitor of metalloproteinase-3 protein near the basement membrane supports its role in preventing proteolytic degradation, angiogenesis, and apoptosis.

PURPOSE: Although important for the diagnosis of familial clustering of colorectal cancer and hereditary nonpolyposis colorectal cancer, the accuracy of familial cancer history assessment in the office setting has been questioned. Furthermore, there are few publications describing the optimal method for accurately capturing a family cancer history. The purpose of this study was to determine how well family cancer history is assessed in patients with early age-of-onset colorectal cancer at initial surgical consultation compared with a telephone interview and mailed questionnaire. METHODS: Medical records of patients 40 years old or younger at the time of colorectal cancer surgery were reviewed for documentation of family cancer history at initial surgical consultation. In addition, family cancer history was solicited from surviving patients or their next of kin by telephone and a mailed questionnaire. The kappa coefficient was used to measure degree of correlation between family cancer history obtained at initial surgical consultation and subsequent telephone interview and questionnaire. RESULTS: One hundred twenty-five patients were available for analysis. Family cancer history was documented on the initial surgical consultation report in 78 percent of cases. Although 31.2 percent were identified as having no family cancer history at initial surgical consultation, this proportion decreased to 13.5 percent after telephone interviews and questionnaires. Family history assessment at initial surgical consultation also failed to identify 7 of 11 individuals meeting Amsterdam criteria for hereditary nonpolyposis colorectal cancer and 10 of 16 individuals meeting modified clinical criteria for hereditary nonpolyposis colorectal cancer. CONCLUSIONS: Although family cancer history was commonly obtained during the initial surgical consultation of patients with colorectal cancer, there was a tendency to underestimate the extent of familial cancer. A telephone interview and questionnaire conducted at a later date may reveal a more comprehensive family cancer history. This is an important observation, because individuals identified as high-risk for hereditary nonpolyposis colorectal cancer or familial clustering of colorectal cancer require special consideration with respect to screening, surveillance, and surgical management.

PURPOSE: Although sphincter-preservation procedures have replaced abdominoperineal resection as the treatment of choice for rectal cancer, a subset of patients with rectal cancer will still require abdominoperineal resection. The use of adjuvant radiotherapy has been shown to reduce local recurrence, and combined modality therapy (chemoradiation) improves survival. Sharp mesorectal excision compared with the classic teaching of blunt retrorectal dissection is also an important component of local control. The primary aim of the present study was to evaluate the postoperative complications associated with neoadjuvant therapy in patients requiring complete rectal excision. Oncologic outcomes for all patients with abdominoperineal resection are also provided. METHODS: A prospective database of 5,634 patients who underwent surgery for colorectal cancer at Memorial Sloan-Kettering Cancer Center between the years 1987 and 1997 was reviewed. Patients with primary adenocarcinoma of the rectum who underwent abdominoperineal resection were identified. In 1,622 patients who were operated on for primary rectal cancer, 292 patients (18 percent) underwent abdominoperineal resection and the rest had a sphincter-preserving procedure. Ten patients were excluded from the study because of prior pelvic irradiation for other cancer (8 patients) and insufficient radiation dose (

PURPOSE: Because the elderly population in Western countries is rapidly increasing, as is their life expectancy, studies aimed at determining the impact of major surgery for primary rectal cancer in this group are warranted. The purpose of this study was to compare perioperative morbidity and mortality and long-term disease-specific and overall survival in primary rectal cancer patients, older and younger than 75 years of age, subject to major pelvic surgery. METHODS: From September 1986 to December 1996, the Prospective Colorectal Service Database identified 1,120 consecutive patients who underwent major pelvic surgery for primary rectal cancer. Of these, 157 (15 percent) were 75 years of age or older and comprise the elderly group. From the remaining 963 patients younger than 75 years of age, a representative random sample of 174 was selected and constitutes the younger group. Data were obtained from computerized databases and confirmed via chart review and telephone interviews. RESULTS: Perioperative complications were observed in 53 (34 percent) elderly and 63 (36 percent;P=not significant) younger patients. Perioperative deaths occurred in two (1.3 percent) elderly and one (0.6 percent;P=not significant) younger patient. The median follow-up time was 48 months. Although the overall survival was lower in the elderly group (P=0.02; the 5-year overall survival rates were 51 and 66 percent), the disease-specific survival rate was similar in the two groups (P=0.75; the 5-year disease-specific survival rates were 69 and 71 percent). CONCLUSION: In select individuals 75 years of age or older, major pelvic surgery for primary rectal cancer can be done with perioperative morbidity and mortality rates comparable to those obtained in younger individuals, while achieving excellent disease-specific and overall long-term survival.

PURPOSE: The role of extended resections for locally advanced carcinomas of the right colon infiltrating duodenum and pancreas remains unclear. This investigation was undertaken to review our experience with pancreatic head or duodenal resections for advanced right-sided colon cancer. METHODS: The clinical, pathologic, and follow-up details of eight patients with bulky primary carcinomas of the right colon infiltrating the duodenum (n=4) or pancreatic head (n=4) surgically managed at Memorial Sloan-Kettering Cancer Center between 1986 and 1998 were reviewed. RESULTS: Six patients presented with anemia, and one patient each with epigastric pain and an abdominal mass. All patients had T4 lesions, whereas five had lymph node metastases at presentation. All patients were resected with clear pathologic margins either by right colectomy anden bloc duodenectomy (n=4), oren bloc pancreaticoduodenectomy (n=4). The 30-day mortality rate was zero. Six patients remained alive and free of disease at a median follow-up of 26 months, and there was one long-term survivor who was alive and free of disease at 84 months after resection. CONCLUSION: Extended resection for localized primary colonic carcinoma invading pancreas or duodenum can be undertaken safely and is associated with prolonged survival time.

Multiple endocrine neoplasia type 2B is best known for its endocrine manifestations and typical phenotype. The gastrointestinal manifestations, however, are also an important and commonly unrecognized component of the syndrome. We present four cases that demonstrate the varied presentation of patients with colonic manifestations of multiple endocrine neoplasia type 2B. We discuss the cause, diagnostic significance, and management of the colonic disease that is a component of multiple endocrine neoplasia type 2B.

PURPOSE: There is little information comparing signet-ring cell carcinoma to common non-signet-ring cell colon and rectal cancers. The aim of this study was to better define the clinicopathologic differences between these two distinct entities. METHODS: Using a prospective database of 5,350 surgical patients with rectal cancers operated on at Memorial Sloan-Kettering Cancer Center between 1986 and 1997, 46 patients with signet-ring cell carcinoma were identified. Signet-ring cell carcinoma lesions were those in which signet-ring cells constituted more then 50 percent of the tumor. Six patients who presented with recurrent disease were excluded from the study. Control patients were matched for age, gender, TNM stage, primary site, procedure, and adjuvant therapy. Age, primary site of the tumor, stage at presentation, and survival times of patients with signet-ring cell carcinoma were also compared with 3,371 patients with primary non-signet-ring cell rectal cancers. Survival was calculated using Kaplan-Meier survival estimates. RESULTS: Mean age of the signet-ring cell carcinoma group was 59±12 years and median age was 61 (range, 20–91) years. Male-to-female ratio was 1.1:1. Lymphatic and peritoneal spread was more common among the signet-ring cell carcinoma group. Approximately one-third of signetring cell carcinoma patients presented with metastatic disease. Mean survival time of the signet-ring cell carcinoma group was 45.4 months (95 percent confidence interval, 26.9–63.8) compared with 78.5 months (95 percent confidence interval, 62.0–94.9) for the control patients group;P=0.02 by the log-rank test. The cumulative survival curve of patients with signet-ring cell carcinoma resembles that of patients with poorly differentiated rectal cancers. CONCLUSIONS: Patients with signet-ring cell carcinoma of the colon and rectum have a worse prognosis compared with matched controls with the same stage of disease.

PURPOSE: Although local excision can be curative in patients with early-stage rectal cancer, approximately 20 percent of patients will develop local recurrence, many as a result of unrecognized and unresected regional lymph node metastases. Our objective was to determine if standard pathologic factors can predict lymph node metastases in small intramural rectal cancers and provide a basis for patient selection for nonradical surgery. METHODS: Between June 1986 and September 1996, 318 patients with T1 or T2 rectal cancers underwent radical resection at our institution. Of these, 159 patients (48 T1 and 111 T2) were potentially eligible for curative local excision (≤4 cm in size, ≤10 cm from the anal verge, no synchronous metastases), and the prevalence of lymph node metastases based on T stage and other pathologic factors was analyzed in this group. RESULTS: The overall frequency of lymph node metastasis was 15 percent (24/159 patients). T stage (T1, 10 percent; T2, 17 percent), differentiation (well-differentiated or moderately differentiated, 14 percent and poorly differentiated, 30 percent), and lymphatic vessel invasion (lymphatic vessel invasion-negative, 14 percent and lymphatic vessel invasion-positive, 33 percent) influenced the risk of lymph node metastasis. However, only blood vessel invasion (blood vessel invasion-negative, 13 percent and blood vessel invasion-positive, 33 percent) reached statistical significance as a single predictive factor (P=0.04). Tumors with no adverse pathologic features (low-risk group) had a lower overall frequency of lymph node metastasis (11 percent) compared with the remaining tumors (high-risk group, 31 percent;P=0.008). However, even in the most favorable group (T1 cancers with no adverse pathologic features) lymph node metastases were present in 7 percent of patients. CONCLUSION: In rectal cancer patients potentially eligible for local excision, the overall risk of undetected and untreated lymph node metastases is considerable (15 percent). The use of pathologic factors alone after local excision does not reliably assure the absence of lymph node metastases.

PURPOSE: Although the criteria for clinical diagnosis of hereditary nonpolyposis colorectal cancer are not fully agreed on, young age seems to be a common trait. The purpose of this study is to identify clinicopathologic features of hereditary nonpolyposis colorectal cancer in early age-of-onset colorectal cancer patients stratified as a function of family cancer history. METHODS: Two hundred thirty consecutive colorectal cancer patients 40 years or older at time of diagnosis were registered into an ongoing database during a ten-year period. Accurate family history was obtainedvia medical records, telephone calls, and questionnaires on 146 patients. According to extent of family history of cancer, patients were stratified into seven groups: 1) fulfilling Amsterdam criteria, 2) fulfilling less strict criteria, 3) having at least one first-degree relative with colorectal cancer, 4) having at least one distant relative with colorectal cancer, 5) having at least one first-degree relative with any cancer, 6) having at least one distant relative with any cancer, 7) having no family history of cancer. RESULTS: Twenty-two of 146 patients fulfilled Amsterdam and less strict hereditary nonpolyposis colorectal cancer criteria (15 percent). These hereditary nonpolyposis colorectal cancer patients were significantly younger (31vs. 35 years;P=0.0003) and had more metachronous colorectal cancer (27 percentvs. 2 percent;P=0.007) and less colorectal cancer with nodal or metastatic spread than the non-hereditary nonpolyposis colorectal cancer patients (35 percentvs. 65 percent;P=0.01). CONCLUSION: Precise familial cancer assessment in early age-of-onset colorectal cancer increases the yield of hereditary nonpolyposis colorectal cancer diagnosis. Because of the frequent development of metachronous colorectal cancer and favorable prognosis, extensive rather than segmental surgery should be considered in early age-of-onset colorectal cancer patients belonging to hereditary nonpolyposis colorectal cancer families.

PURPOSE: We report the downstaging, sphincter preservation, acute toxicity, and preliminary local control and survival results of preoperative 5-fiuorouracil (5-FU), low-dose leucovorin (LV), and concurrent radiation therapy followed by postoperative LV/5-FU for treatment of patients with clinically resectable T3 rectal cancer. MATERIALS AND METHODS: A total of 32 patients received two monthly cycles of preoperative LV/5-FU (bolus daily×5). Radiation therapy (5,040 cGy) began concurrently on day 1. Postoperatively, patients received a median of two monthly cycles of LV/5-FU (range, 0–10). RESULTS: The complete response rate was 9 percent pathologic and 13 percent clinical, for a total of 22 percent. Total Grade 3+ acute toxicity during the preoperative combined modality segment was 25 percent (8/32). Of the 20 patients who were thought to initially require an abdominoperineal resection and for whom the intent of treatment was sphincter preservation, 17 (85 percent) were able to undergo sphincter-preserving surgery. With a median follow-up of 22 (3–59) months, none have developed local failure, and the three-year actuarial diseasefree survival rate was 60 percent. CONCLUSION: Our data reveal encouraging downstaging, sphincter preservation, and acute toxicity with this regimen. Additional follow-up is needed to assess the long-term local control and survival rates.

This study was designed to evaluate the spectrum, clinical presentation, management, and outcome of anorectal disease in neutropenic leukemic patients and to compare operative and nonoperative management in neutropenic leukemic patients.A retrospective review of hospital records was performed.One hundred fifty-one of 2,618 (5.8 percent) patients hospitalized with leukemia had concomitant symptomatic anorectal disease. Data from 81 patients were available for analysis. Fifty-two (64 percent) were treated nonoperatively and 29 (36 percent) underwent operative treatment. Fifty-seven (70.4 percent) had absolute neutrophil counts1,000/mm3, and 54 (66.7 percent) were severely neutropenic (absolute neutrophil count500/mm3). Management and outcomes of 54 severely neutropenic patients were analyzed. In 20 patients who underwent surgery there were 4 deaths (20 percent) and 4 recurrences (20 percent), whereas in 34 patients managed nonoperatively there were 6 deaths (18 percent) and 4 recurrences (12 percent) (P0.05).Symptomatic anorectal disease afflicted 5.8 percent of hospitalized leukemic patients. In these patients, anorectal sepsis was a major source of mortality. Our data suggest that anorectal abscesses in neutropenic leukemic patients may be safely drained. Because we did not observe excessive morbidity or mortality (20 percent vs. 18 percent) in the operated neutropenic leukemics as compared with the nonoperated patients, selected neutropenic leukemic patients should not be denied anorectal surgery when otherwise indicated.

The fate of the excluded rectal segment after surgery for Crohn's colitis remains poorly defined. To determine prognostic factors relating to the fate of the rectal segment, records of 47 patients who underwent creation of an excluded rectal segment were studied. Disease developed in 33 patients (70 percent) in the excluded rectal segment by five years; 24 patients (51 percent) had completion proctectomy by 2.4 years; and 9 patients (19 percent) retained a rectum with disease at a median follow-up period of five years (range, 2-13 years). At a median follow-up time of six years (range, 2-21 years), 14 patients were without clinical disease. The three groups were equivalent with respect to sex, duration of preoperative disease, indication for operation, distribution of disease, and histologic involvement of the proximal rectal margin. The median age of patients in the proctectomy group at diagnosis tended to be younger than that of patients with a retained excluded rectal segment (22, 30, and 31 years for patients having proctectomy, patients with a diseased excluded rectal segment, and patients with a normal excluded rectal segment, respectively). Neither initial involvement of the terminal ileum nor endoscopic inflammatory changes seen in the rectum predicted eventual disease of the excluded rectal segment. However, initial perianal disease complicating Crohn's colitis was predictive of persistent excluded rectal segment disease and often required proctectomy. Therefore, because the presence of perianal disease and Crohn's colitis predicts persistent or recurrent excluded rectal segment disease, primary total proctocolectomy or early completion proctectomy may be indicated in this subgroup of patients.

Individuals with a family history of colorectal cancer are believed to be at an increased risk of developing colorectal neoplasia. To estimate this risk and the potential yield of screening colonoscopy in this population, we recruited and prospectively colonoscoped 181 asymptomatic first-degree relatives (FDR) of colorectal cancer patients and 83 asymptomatic controls (without a family history of colorectal cancer). The mean ages for the FDR and control groups were 48.2 +/- 12.5 and 54.8 +/- 11.0, respectively. Adenomatous polyps were detected in 14.4 percent of FDRs and 8.4 percent of controls. Although 92 percent of our FDRs had only one FDR afflicted with colon cancer, those subjects with two or more afflicted FDRs had an even higher risk of developing colonic adenomas (23.8 percent) than those with only one afflicted FDR (13.1 percent). A greater proportion of adenomas was found to be beyond the reach of flexible sigmoidoscopy in the FDR group than in the controls (48 percent vs. 25 percent, respectively). Logistic regression analysis revealed that age, male sex, and FDR status were independent risk factors for the presence of colonic adenomatous polyps (RR = 2.32, 2.86, and 3.49, respectively; P less than 0.001). Those at greatest risk for harboring an asymptomatic colonic adenoma are male FDRs over the age of 50 (40 percent vs. 20 percent for age-matched male controls). Based on probability curves, males with one FDR afflicted with colon cancer appear to have an increased risk of developing a colonic adenoma beginning at 40 years of age. Our results document, for the first time, an increased prevalence of colonoscopically detectable adenomas in asymptomatic first-degree relatives of colon cancer patients, as compared with asymptomatic controls, and support the use of colonoscopy as a routine screening tool in this high-risk group.

The detection of primary or recurrent colorectal cancer during pregnancy is a rare event. Early diagnosis is difficult, because symptoms are often attributed to pregnancy. Here we describe a case of recurrent adenocarcinoma of the colon in a 23-year-old pregnant patient. The relationship between various hormonal and immunological changes of pregnancy and colorectal cancer is discussed.