Search

Background: The Memorial Sloan Kettering Cancer Center (MSK) nomogram combined both gastroesophageal junction (GEJ) and gastric cancer patients and was created in an era from patients who generally did not receive neoadjuvant chemotherapy. We sought to reevaluate the MSK nomogram in the era of multidisciplinary treatment for GEJ and gastric cancer., Study Design: Using data on patients who underwent R0 resection for GEJ or gastric cancer between 2002 and 2016, the C-index of prediction for disease-specific survival (DSS) was compared between the MSK nomogram and the American Joint Committee on Cancer (AJCC) 8th edition staging system after segregating patients by tumor location (GEJ or gastric cancer) and neoadjuvant treatment. A new nomogram was created for the group for which both systems poorly predicted prognosis., Results: During the study period, 886 patients (645 gastric and 241 GEJ cancer) underwent up-front surgery, and 999 patients (323 gastric and 676 GEJ) received neoadjuvant treatment. Compared with the AJCC staging system, the MSK nomogram demonstrated a comparable C-index in gastric cancer patients undergoing up-front surgery (0.786 vs 0.753) and a better C-index in gastric cancer patients receiving neoadjuvant treatment (0.796 vs 0.698). In GEJ cancer patients receiving neoadjuvant chemotherapy, neither the MSK nomogram nor the AJCC staging system performed well (C-indices 0.647 and 0.646). A new GEJ nomogram was created based on multivariable Cox regression analysis and was validated with a C-index of 0.718., Conclusions: The MSK gastric cancer nomogram's predictive accuracy remains high. We developed a new GEJ nomogram that can effectively predict DSS in patients receiving neoadjuvant treatment., (Copyright © 2022 by the American College of Surgeons. Published by Wolters Kluwer Health, Inc. All rights reserved.)

Background: Training of foreign medical graduates in surgical oncology is an undervalued intervention for improving global cancer care. The aim of this work was to describe the design and results of a clinical training program for international surgeons from a single comprehensive cancer center., Study Design: Of 39 international fellows trained during 20 years, 34 were surveyed about education, research, and current context of surgical practice. A citation and H-index calculation (ie h number of publications that each has at least h citations) was performed to assess scientific productivity of each graduated fellow., Results: Twenty-one of 39 (54%) fellows came from countries in which English is not the primary language. Europe was the continent with the most graduates (17 of 39 [43%]), and only 5 of 39 (13%) were from Latin America. Three of 39 (8%) were women. Thirty-one of 39 graduated fellows (80%) returned to their countries of origin. The survey response rate was 73% (25 of 34). Seventeen of twenty-five (68%) work in an academic setting and 13 (52%) reported surgical oncology as their main clinical practice. Total number of citations and H-index are homogeneous among the different regions from which the fellows originated, with a median of 165 citations and median H-index of 5., Conclusions: The International General Surgical Oncology Fellowship has successfully trained foreign surgeons for academic practice in surgical oncology. Most of the graduates have returned to their country of origin and contributed to education and research there., (Copyright © 2016 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: Surgical quality improvement requires well-defined benchmarks and accurate reporting of postoperative adverse events, which have not been well defined for total gastrectomy., Study Design: Detailed postoperative outcomes on 238 patients who underwent total gastrectomy with curative intent, from 2003 to 2012, were reviewed by a dedicated surgeon chart reviewer to establish 90-day patterns of adverse events., Results: Of the 238 patients with stage I to III gastric adenocarcinoma who underwent curative-intent total gastrectomy, the median age was 66 years, and 68% were male. Median body mass index was 28 kg/m(2), and 68% of patients had at least 1 medical comorbidity. Forty-three percent of our patients received neoadjuvant chemotherapy, and 34% received postoperative adjuvant chemotherapy. Over the 90-day study period, 30-day mortality was 2.5% (6 of 238), and 90-day mortality was 2.9% (7 of 238). At least 1 postoperative adverse event was documented in 62% of patients, with 28% of patients experiencing a major adverse event requiring invasive intervention. The readmission rate was 20%. Anemia was the most common adverse event (20%), followed by wound complications (18%). The most common major adverse event was esophageal anastomotic leak, which required invasive intervention in 10% of patients., Conclusions: This analysis has defined comprehensive 90-day patterns in postoperative adverse events after total gastrectomy with curative intent in a Western population. This benchmark allows surgeons to measure, compare, and improve outcomes and informed consent for this surgical procedure., (Copyright © 2015 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: Postpancreatectomy hemorrhage (PPH) is a life-threatening complication of pancreatic resection. Most published series span decades and do not reflect contemporary practice. This study analyzes the rate, management, and outcomes of PPH during a recent 5-year period., Study Design: Patients in whom PPH developed between 2006 and 2011 were identified from a prospective database. Postpancreatectomy hemorrhage was defined as evidence of bleeding associated with a drop in hemoglobin (≥ 3 g/dL) and/or clinical signs of hemodynamic compromise, and categorized as early or late (<24 hours or >24 hours from operation). Demographics and operative and perioperative outcomes were analyzed using standard descriptive statistics., Results: Overall incidence of PPH was 3% (33 of 1,122 pancreatectomies) and was similar for pancreaticoduodenectomy (25 of 739 [3%]), distal (6 of 350 [2%]), and central pancreatectomy (2 of 31 [6%]) (p = 0.26). Early hemorrhage was seen in 21% (7 of 33) and was always extraluminal; these patients underwent reoperation and recovered fully. Late hemorrhage (26 of 33 [79%]) was predominantly intraluminal (18 of 26 [69%]), occurring at a median of 12 days postoperatively (4 to 23 days), and was treated endoscopically (13 of 26 [50%]), angiographically (10 of 26 [38%]), or surgically (3 of 26 [10%]). Postpancreatectomy hemorrhage was associated with longer hospitalization (10 [range 8 to 17] days vs 7 [range 6 to 9] days; p < 0.01); mortality, however, was not increased (1 of 33 [3%] vs 17 of 1,089 [2%]; p = 0.95). Hemorrhage began after discharge in 39% of patients (13 of 33), with the only death occurring in a patient from this group., Conclusions: Postpancreatectomy hemorrhage can be managed successfully with low mortality (3%). Early hemorrhage requires urgent reoperation, and management of delayed hemorrhage should be guided by location (intra- vs extraluminal). Greater pressure to reduce length of hospital stay appears to have increased the likelihood of PPH occurring after discharge; patients and physicians should be aware of this possibility., (Copyright © 2012 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: Little is known about the outcomes of patients with microscopically positive (R1) resections for primary gastrointestinal stromal tumors (GIST) because existing retrospective series contain small numbers of patients. The objective of this study was to analyze factors associated with R1 resection and assess the risk of recurrence with and without imatinib., Study Design: We reviewed operative and pathology reports for 819 patients undergoing resection of primary GIST from the North American branch of the American College of Surgeons Oncology Group (ACOSOG) Z9000 and Z9001 clinical trials at 230 institutions testing adjuvant imatinib after resection of primary GIST. Patient, tumor, operative characteristics, factors associated with R1 resections, and disease status were analyzed., Results: Seventy-two (8.8%) patients had an R1 resection and were followed for a median of 49 months. Factors associated with R1 resection included tumor size (≥ 10 cm), location (rectum), and tumor rupture. The risk of disease recurrence in R1 patients was driven largely by the presence of tumor rupture. There was no significant difference in recurrence-free survival for patients undergoing an R1 vs R0 resection of GIST with (hazard ratio [HR] 1.095, 95% CI 0.66, 1.82, p = 0.73) or without (HR 1.51, 95% CI 0.76, 2.99, p = 0.24) adjuvant imatinib., Conclusions: Approximately 9% of 819 GIST patients had an R1 resection. Significant factors associated with R1 resection include tumor size ≥ 10 cm, location, and rupture. The difference in recurrence-free survival with or without imatinib therapy in those undergoing an R1 vs R0 resection was not statistically significant at a median follow-up of 4 years., (Copyright © 2012 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: Cystic lesions of the pancreas are being identified more frequently, and a selective approach to resection is now recommended. The aim of this study was to assess the change in presentation and management of pancreatic cystic lesions evaluated at a single institution over 15 years., Study Design: A prospectively maintained registry of patients evaluated between 1995 and 2010 for the ICD-9 diagnosis of pancreatic cyst was reviewed. The 539 patients managed from 1995 to 2005 were compared with the 885 patients managed from 2005 to 2010., Results: A total of 1,424 patients were evaluated, including 1,141 with follow-up >6 months. Initial management (within 6 months of first assessment) was operative in 422 patients (37%) and nonoperative in 719 patients (63%). Operative mortality in patients initially submitted to resection was 0.7% (n = 3). Median radiographic follow-up in patients initially managed nonoperatively was 28 months (range 6 to 175 months). Patients followed radiographically were more likely to have cysts that were asymptomatic (72% versus 49%, p < 0.001), smaller (1.5 versus 3 cm, p < 0.001), without solid component (94% versus 68%, p < 0.001), and without main pancreatic duct dilation (88% versus 61%, p < 0.001). Changes prompting subsequent operative treatment occurred in 47 patients (6.5%), with adenocarcinoma identified in 8 (17%) and pancreatic endocrine neoplasm in 4 (8.5%). Thus, of the 719 patients initially managed nonoperatively, invasive malignancy was identified in 12 (1.7%), with adenocarcinoma seen in 1.1%., Conclusion: Cystic lesions of the pancreas are being identified more frequently, yet are less likely to present with concerning features of malignancy. Carefully selected patients managed nonoperatively had a risk of malignancy that was equivalent to the risk of operative mortality in those patients who initially underwent resection., (Copyright © 2011 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: The pancreas remains an organ for which routine laparoscopic resection is uncommon., Study Design: This is a review of all distal pancreatectomies performed between January 2003 and December 2009 at Memorial Sloan-Kettering Cancer Center. Variables were compared between laparoscopic and open groups in unmatched and matched analyses., Results: During the 7-year study period, 343 distal pancreatectomies were performed; 107 (31%) were attempted laparoscopically and 236 (69%) were performed open. The conversion rate was 30%. Laparoscopic patients were younger (median 60 vs 64 years, p < 0.0001), experienced less blood loss (median 150 vs 350 mL, p < 0.0001), longer operative times (median 163 vs 194 minutes, p < 0.0001), shorter hospital stay (median 5 vs 7 days, p < 0.0001), and had fewer postoperative complications (27% vs 40%, p = 0.03) than open patients. The rates of complications of grade 3 or greater (20% vs 20%, p = NS) and pancreatic leak (15% vs 13%, p = NS) were similar between laparoscopic and open groups. Patients having procedures that were converted had a higher body mass index (BMI) than patients who did not (28 vs 25, p = 0.035). Patients with converted resections experienced higher rates of complications of grade 3 or greater (36% vs 20%, p = 0.008) and pancreatic leaks (27% vs 13%, p = 0.03) than open patients. Compared with matched open patients, laparoscopic patients had longer operative times (195 minutes vs 160 minutes, p < 0.0001), less blood loss (175 mL vs 300 mL, p < 0.0001), and shorter hospital stay (5 days vs 6 days, p < 0.001)., Conclusions: Patients who had laparoscopic distal pancreatectomy experienced decreased blood loss and a shorter hospital stay compared with matched patients undergoing open resection. Careful patient selection is important because patients who required conversion experienced higher rates of complications and pancreatic leak., (Copyright © 2010 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: Electronic synoptic operative reports (E-SORs) have replaced dictated reports at many institutions, but whether E-SORs adequately document the components and findings of an operation has received limited study. This study assessed the reliability and completeness of E-SORs for pancreatic surgery developed at our institution., Study Design: An attending surgeon and surgical fellow prospectively and independently completed an E-SOR after each of 112 major pancreatic resections (78 proximal, 29 distal, and 5 central) over a 10-month period (September 2008 to June 2009). Reliability was assessed by calculating the interobserver agreement between attending physician and fellow reports. Completeness was assessed by comparing E-SORs to a case-matched (surgeon and procedure) historical control of dictated reports, using a 39-item checklist developed through an internal and external query of 13 high-volume pancreatic surgeons., Results: Interobserver agreement between attending and fellow was moderate to very good for individual categorical E-SOR items (kappa = 0.65 to 1.00, p < 0.001 for all items). Compared with dictated reports, E-SORs had significantly higher completeness checklist scores (mean 88.8 +/- 5.4 vs 59.6 +/- 9.2 [maximum possible score, 100], p < 0.01) and were available in patients' electronic records in a significantly shorter interval of time (median 0.5 vs 5.8 days from case end, p < 0.01). The mean time taken to complete E-SORs was 4.0 +/- 1.6 minutes per case., Conclusions: E-SORs for pancreatic surgery are reliable, complete in data collected, and rapidly available, all of which support their clinical implementation. The inherent strengths of E-SORs offer real promise of a new standard for operative reporting and health communication., (Copyright 2010 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: This study analyzes factors associated with differences in long-term outcomes after hepatic resection for metastatic colorectal cancer over time., Study Design: Sixteen-hundred consecutive patients undergoing hepatic resection for metastatic colorectal cancer between 1985 and 2004 were analyzed retrospectively. Patients were grouped into 2 eras according to changes in availability of systemic chemotherapy: era I, 1985 to 1998; era II, 1999 to 2004., Results: There were 1,037 patients in era I and 563 in era II. Operative mortality decreased from 2.5% in era I to 1% in era II (p = 0.04). There were no differences in age, Clinical Risk Score, or number of hepatic metastases between the 2 groups; however, more recently treated patients (era II) had more lymph node-positive primary tumors, shorter disease-free intervals, more extrahepatic disease, and smaller tumors. Median follow-up was 36 months for all patients and 63 months for survivors. Median and 5-year disease-specific survival (DSS) were better in era II (64 months and 51% versus 43 months and 37%, respectively; p < 0.001); but median and 5-year recurrence-free survival (RFS) for all patients were not different (23 months and 33% era II versus 22 months and 27% era I; p = 0.16). There was no difference in RFS or DSS for high-risk (Clinical Risk Score >2, n = 506) patients in either era. There was a marked improvement in both RFS and DSS for low risk (Clinical Risk Score < or =2, n = 1,094) patients., Conclusions: Despite worse clinical and pathologic characteristics, survival but not recurrence rates after hepatic resection for colorectal metastases have improved over time and might be attributable to improvements in patient selection, operative management, and chemotherapy. The improvement in survival over time is largely accounted for by low-risk patients., (Copyright 2010 American College of Surgeons. Published by Elsevier Inc. All rights reserved.)

Background: Anastomotic fistula, leak, and abscess are common complications of pancreatectomy. The goal of this study was to describe our current management and outcomes of clinically significant postpancreatectomy fistula, leak, and abscess., Study Design: Review of a prospectively maintained database identified 908 patients who underwent pancreatectomy between January 2000 and August 2005. Complication data were prospectively entered into a validated postoperative complication database. Patients were included if they were identified as having a clinically significant (>/=grade 2) pancreatic fistula, leak, or abscess. Multivariate analyses were performed to identify factors predictive of prolonged drainage (> 30 days)., Results: Clinically significant postoperative fistula, leak, or abscess occurred in 158 of 908 resected patients (17%) and included 63 culture-positive pancreatic fistulas, 29 noninfected pancreatic fistulas, 42 abscesses, and 24 other collections (biliary fistula, culture-negative collection). Surgical drains were placed at the time of initial resection in 88 of these 158 patients (56%). Adequate drainage was obtained by prolonged use of surgical drains in 16 patients (16 of 88 [18%]). Reoperation was required in 26 of the 158 patients (16%). ICU admission was required in 22%. Within this group of 158 patients the mortality rate was 5% (8 of 158; 90 days). At the time of discharge a home health aide was required in 56% of patients, 8% were discharged to a rehabilitation facility, and readmission was required in 50% of patients. Mean drainage time was 38 days (range 3 to 228). Predictors of prolonged drainage included drain output > 200 mL during the first 48 hours (odds ratio = 2.88; p = 0.02) and distal (versus proximal) pancreatectomy (odds ratio = 4.29; p = 0.01)., Conclusions: Although mortality after pancreatectomy has decreased to approximately 2%, the morbidity associated with pancreatic fistula, leak, and abscess remains substantial.

Background: Endoscopic ultrasound (EUS) is the most accurate locoregional staging tool for gastroesophageal junction (GEJ) adenocarcinoma, and it may allow pretreatment risk stratification. The purpose of this study was to compare preoperative EUS staging with postoperative pathologic staging and to assess the ability of EUS to predict survival after resection for GEJ adenocarcinoma., Study Design: Patients with GEJ adenocarcinoma, who had preoperative staging with EUS followed by resection, were identified from a prospectively maintained database. Patients receiving neoadjuvant therapy were excluded. EUS stage was compared with pathologic stage. Survival analyses were performed in patients who underwent complete gross resection., Results: From 1985 through 2003, 209 patients underwent preoperative EUS followed by surgery without neoadjuvant therapy for GEJ adenocarcinoma. EUS correlated with pathologic T stage in 128 of 209 (61%) patients and with pathologic nodal stage in 154 of 206 (75%) patients. EUS accurately stratified patients into "early" (T0-2 N0) or "advanced" (T3-4 or N1) disease categories in 173 (83%) patients. Curative (R0) resection was performed in 184 patients: EUS "early" (n=84) and "advanced" (n=122) stages were associated with R0 rates of 100% and 82%, respectively (p=0.001). EUS "early" versus "advanced" stage was highly predictive of outcomes (p < 0.0001). The 5-year disease-specific survival for EUS "early" patients was 65% compared with 34% for EUS "advanced" stage., Conclusions: EUS accurately predicts pathologic stage. In addition, EUS is predictive of outcomes after complete gross resection without neoadjuvant treatment for GEJ adenocarcinoma and identifies a high-risk population that might benefit from preoperative therapy.

Background: Improving surgical quality of care requires accurate reporting of postoperative complications., Study Design: Accuracy of a prospective surgical complication grading database was assessed by performing a retrospective review of 204 pancreaticoduodenectomies (PDs) entered into the database from January 1, 2001, to December 31, 2003. This updated database was then used to characterize 30-day morbidity and mortality after PD., Results: On review, 13% of patients had a complication not identified in the prospective complication database, 8% of patients had a complication reclassified, and 4% of patients had a complication removed. At least 1 postoperative complication was experienced by 47% of patients. After PD, 45 different complications occurred. Postoperative mortality at 30 days was 1%, and 30-day readmission rate was 11%. The 30-day reoperation rate was 9%, and 14% of patients required a percutaneous drainage procedure. Pancreatic anastomotic leak (12%), wound infection (11%), and delayed gastric emptying (7%) were the 3 most common postoperative complications, and all were associated with an increased length of stay., Conclusions: Our prospective surgical complication database accurately characterized outcomes after PD and facilitated information gathering and analysis. The accuracy, efficiency, and reproducibility of a prospective surgical complication database favor its widespread use in postoperative complication reporting.

Background: It is not clear if more intense surveillance is associated with improved survival after curative resection for cancer. In the context of a followup program after curative gastrectomy, recurrence and survival were investigated for patients presenting with either symptomatic or asymptomatic recurrence., Study Design: A prospectively maintained gastric cancer database was used to identify all patients who underwent a curative (R0) gastrectomy from July 1985 to June 2000. Survival curves were generated for patients with either symptomatic or asymptomatic recurrence, and the prognostic variables associated with outcomes were identified., Results: Of 1,172 patients who underwent a curative (R0) gastrectomy, 561 patients (48%) had documented recurrence and 382 patients had complete data about symptoms. Median time to recurrence was 10.8months for asymptomatic patients and 12.4months for symptomatic patients (p = NS). Median postrecurrence survival was 13.5months for asymptomatic patients and 4.8months for symptomatic patients (p < 0.01). Median disease-specific survival was 29.4months for asymptomatic patients and 21.6months for symptomatic patients (p < 0.05). Variables predictive of poor postrecurrence survival included symptomatic recurrence, advanced stage (III/IV), poor differentiation, short disease-free interval (<12months), and multiple sites of recurrence., Conclusions: Followup did not identify asymptomatic recurrence earlier than symptomatic recurrence. Patients with symptomatic recurrence have more aggressive disease with a shorter postrecurrence survival. The impact of detecting asymptomatic recurrence in the course of followup after curative gastrectomy could not be distinguished from the effects of four powerful biologic variables that also interact to govern outcomes.

Background: Effective palliation rather than cure is often the most appropriate goal in the management of patients with advanced gastric cancer. The literature to date is limited by the imprecise use of the term palliative and subsequent variable designation of patients into evaluable groups., Study Design: Between July 1985 and July 2001, 1,595 patients were entered into a prospective database after undergoing a resection for gastric adenocarcinoma. Patients who received a noncurative (R1/R2) resection were identified. A procedure was defined as palliative if it was performed explicitly to palliate symptoms or improve quality of life., Results: Three hundred seven patients received a noncurative gastric resection. The operation was palliative in 48% (147/307) and nonpalliative in 52% (160/307). Palliative operations included an esophageal anastomosis less frequently (46% versus 69%, p < 0.001) and had a less extensive lymphadenectomy performed compared with nonpalliative operations. Surgical intent did not alter operative morbidity (54%) or mortality (6%) significantly. The overall median survival after a noncurative gastric resection was 10.6 months and was independently associated with operations performed with explicit palliative indications (8.3 months [palliative] versus 13.5 months [nonpalliative], p < 0.001) and patient age > 65 years., Conclusions: There are important differences among patients undergoing noncurative operations for gastric cancer. Studies designed to measure palliative interventions would benefit from precise designations of palliative intent in patients receiving noncurative operations.

Background: Efforts to improve surgical outcomes have traditionally focused on improving preoperative patient selection and reducing the risk of postoperative medical complications. Strategies to optimize surgical technique have been less well studied. We sought to assess the relation between complications related to surgical technique and outcomes after esophagogastrectomy for cancer., Study Design: Medical records of 510 consecutive patients undergoing esophagogastrectomy for invasive squamous cell carcinoma or adenocarcinoma at Memorial Sloan-Kettering Cancer Center from 1996 to 2001 were reviewed. Data on diagnosis, stage of disease, therapies received, surgical approach, patient comorbidities, technical complications, and postoperative medical complications and outcomes including length of stay and overall survival were determined by one reviewer of the medical records. The primary predictor was surgical complications and the primary outcome was survival., Results: Of the 150 patients studied 138 (27%) had complications directly attributable to surgical technique, such as an anastomotic leak, a paralyzed vocal cord, or chylothorax. At 3 years 43 of 138 patients (31%) with technical complications were alive, whereas 179 of 372 patients (48%) without technical complications were alive. Technical complications were associated with increased length of stay (median 23 days versus 11 days, p < 0.001), increased in-hospital mortality (12.3% versus 3.8%, p < 0.001), and a higher rate of medical complications (77.5% versus 47.3%, p < 0.001). After controlling for age, medical comorbidities, use of induction therapy, tumor stage, histology, and location, and completeness of resection the presence of a technical complication was highly predictive of poorer overall survival; the multivariable hazard ratio was 1.41 (1.22 to 1.63, p = 0.008)., Conclusions: Technical complications have a large negative impact on survival after esophagogastrectomy for cancer. Strategies to optimize surgical technique and minimize complications should improve outcomes in this cancer operation.

Background: The inferior vena cava (IVC) is a rare site for primary soft tissue sarcoma. There are limited data in the literature regarding surgical management of the IVC and longterm survival of these patients., Study Design: From 1982 to 2002, a total of 25 patients with primary IVC leiomyosarcoma was treated as inpatients and followed in a prospective database at Memorial Sloan-Kettering. Presenting symptoms, tumor characteristics, operative management, postoperative morbidity, and disease-specific survival were assessed for each patient., Results: The 25 patients with primary IVC leiomyosarcoma accounted for 0.5% of all adult patients with soft tissue sarcoma treated during this time. The median patient age was 56 years (range 41 to 79 years). The three most common presenting symptoms were abdominal pain (52%), distention (20%), and deep venous thrombosis (12%). Of the patients, 21 (84%) underwent complete resection of the tumor. The IVC was managed in one of three ways: ligation (n = 11), primary/patch repair (n = 8), and expanded polytetrafluoroethylene tube grafting (n = 2). Among patients undergoing IVC ligation and primary/patch repair (n = 19), 11% had severe postoperative edema and none had worsening renal function. Local recurrence occurred in 33% of patients and distant recurrence occurred in 48% of patients. Patients undergoing complete resection had 3-year and 5-year disease-specific survival rates of 76% and 33%, respectively. There were no 3-year survivors among patients with incomplete resections., Conclusions: Complete resection of primary IVC leiomyosarcomas is feasible and associated with improved survival. The IVC can be managed by primary repair or ligation with a low risk of severe postoperative edema.

Background: Distant metastasis is uncommon in differentiated thyroid cancer and the prognosis is unclear. This study aims to evaluate outcomes and to define independent variables that are associated with tumor-related mortality in patients with distant metastasis from thyroid carcinoma., Study Design: A retrospective review of the thyroid cancer research database identified 336 patients with distant metastasis from differentiated thyroid carcinoma treated at a single institution between 1941 and 2000. After excluding patients with local or regional recurrence, distant disease was either the first site of recurrence or was detected at the time of diagnosis of the primary tumor in 242 patients (72%). Patient, tumor, and treatment-related factors were analyzed for their relation to disease-specific survival (DSS) using multivariate Cox regression and the log-rank test., Results: Median survival was 4.1 years and 10-year DSS was 26%. Distant disease was synchronous with the primary diagnosis in 97 of 242 (40%) patients. The site of metastasis was lung only in 103 (43%) patients, bone only in 80 (33%), other sites in 14 (6%), and more than one organ system in 45 (19%). Multivariate analysis identified age 45 years or more, symptoms, site other than lung only or bone only, and no radioactive iodine treatment for the metastasis as predictors of poor outcome with 13%, 11%, 16%, and 12% 10-year DSS, respectively. This compares with age less than 45 years, asymptomatic presentation, metastasis only in the lung or bone, and radioactive iodine treatment with 10-year DSS rates of 58%, 45%, 32%, and 33%, respectively (all p < 0.0001). Radioactive iodine treatment was more often given in patients who were less than 45 years of age, asymptomatic, and with metastasis only in the lung or bone only (p = 0.03, 0.11, 0.01)., Conclusions: Longterm survival is possible in patients with distant metastasis from differentiated thyroid cancer. This retrospective study found that age of 45 years or more, site other than lung only or bone only, and symptoms at the time of diagnosis are associated with poorer outcomes.

Background: Randomized prospective trials have addressed various treatment approaches to pancreatic adenocarcinoma in order to improve on the dismal prognosis associated with this disease. We conducted a comprehensive review of prospective randomized clinical trials and summarized the contemporary treatment of resectable pancreatic carcinoma., Study Design: A literature search strategy identified prospective randomized clinical trials for pancreatic carcinoma using standard medical subject heading terms. The articles were critically reviewed and ranked according to a standardized three-tiered system (Ia, Ib, Ic) by a panel of experts., Results: Surgical studies have demonstrated that morbidity and mortality are similar for pylorus-preserving and classic pancreaticoduodenectomy. Extended retroperitoneal lymphadenectomy can be performed with similar mortality but increased morbidity compared with standard pancreaticoduodenectomy but does not prolong survival. Pancreaticogastrostomy and pancreaticojejunostomy appear to be comparable techniques for pancreatic duct reconstruction. Pancreatic-enteric anastomosis is associated with lower rates of pancreatic fistula and endocrine insufficiency than duct occlusion without anastomosis. Intraperitoneal drainage after pancreatic resection is unwarranted and may contribute to intraabdominal complications. Routine use of prophylactic octreotide does not lower the rate of pancreatic fistula; it should be considered for reoperative pancreatic resection or for a soft gland. Early trials found that adjuvant chemoradiation therapy prolongs survival. But in more recent studies chemoradiation after resection has failed to show a survival advantage over surgery alone., Conclusions: Surgical resection remains the only potentially curative therapy for adenocarcinoma of the pancreas. There is no clear indication as to a single preferable resection approach.

Background: In gastric adenocarcinoma, only complete resection (R0) translates into survival benefit. Given the potential for increased morbidity and mortality from multiple organ resection we asked the question as to whether extended (multiple organ) resection was justified for advanced gastric cancer., Study Design: From July 1985 to July 2000, 1,283 patients underwent gastric resection for adenocarcinoma at Memorial Sloan-Kettering Cancer Center, and were entered and followed in a prospectively recorded database. Four hundred eighteen patients (33%) underwent primary resection and had one or more organs resected in addition to the stomach. Eight hundred twenty-six patients (64%) underwent gastrectomy alone, with 39 patients (3%) not undergoing gastrectomy. Clinicopathologic, operative, and morbidity data were evaluated in this group. Complications were categorized by severity on a scale from 0 to 5, 0 being no complication to 5 being death. Chi-square analysis and the logistic regression method were used to compare and estimate factors significantly associated with having a complication., Results: Three hundred thirty-seven patients had a single additional organ resected, 63 had two organs, and 18 had three organs. Five hundred eighty complications occurred in 33% of patients (404 of 1,283). The perioperative mortality was 4% (48 patients). Logistic regression identified the number of organs resected, two or greater, to be predictive of complications (RR 2.0), as well as age greater than 70 years old (RR 1.57). When excluding minor complications (values 1 and 2), only the number of organs resected (RR 3.8) was a major factor for severe complications (values 3, 4, and 5)., Conclusions: Resection of two or more adjacent organs in advanced gastric adenocarcinoma is associated with a greater risk of developing a complication. The use of a graded surgical complication scale is needed for better reporting and comparison of complications. Achieving an R0 resection should still be considered the goal, even in locally advanced gastric cancer, but resection of additional organs should be performed judiciously.

Background: We define the natural history and influence of primary anatomic site for completely resected locally recurrent soft tissue sarcoma (STS) without distant metastasis., Study Design: We selected 290 patients having at least one local recurrence (LR) after complete resection of primary STS (all sites) between 1982 and 1999. Of these, 239 patients had complete resection of their first LR: 161 extremity-trunk and 78 retroperitoneal-head and neck-visceral-thoracic sarcomas. Study end points included second local recurrence-free survival, distant recurrence-free survival, and disease-specific survival, estimated by the Kaplan-Meier method. Univariate and multivariate analyses were performed using a log-rank test and Cox's proportional hazards model for extremity-trunk and retroperitoneal (RP; n 39) tumor sites only., Results: Median followup was 82 months. Complete gross resection rates declined with each subsequent recurrence. Primary tumor site (extremity-trunk: relative risk ERR] 0.74, confidence interval [CI] 0.57 to 0.96, p = 0.03) and microscopic resection margin (negative: RR 0.80, CI 0.62 to 0.99, p 0.04) independently predicted subsequent local control. Extremity-trunk sire and high tumor grade were significant independent predictors of distant disease relapse after complete resection of LR. Site also had a notable influence on disease-specific survival: RP recurrence was associated with 1.4 times increased risk of tumor-related mortality (p = 0.02; 5-year disease-specific survival: extremity-trunk 70% versus RP 57%). High tumor grade was the most marked predictor of tumor-related mortality (RR 1.66, CI 1.28 to 2.18, p<0.001. Nine percent of extremity-trunk and 77% of RP STS-related deaths were caused by advanced LR without synchronous metastasis., Conclusions: Site governs local control, distant recurrence-free and disease-specific survival for completely resected locally recurrent sarcoma without metastasis. Distant disease relapse determines outcomes for recurrent extremity-trunk STS, and local recurrence is the determinant of tumor-related death in the RP.

Complete surgical resection is the most effective modality for the treatment of retroperitoneal sarcomas. Previous studies of all types of retroperitoneal sarcomas have not shown a survival benefit of incomplete resection over no resection. Because death often occurs as a result of local progression in retroperitoneal liposarcomas (RPLS), it is possible that incomplete resection may be beneficial in this histologic type. In this study we have sought to determine the clinical outcomes in patients with incompletely resected and unresected RPLS with the aim of defining patients who may benefit from palliative resection. From a prospective clinical database 55 patients with incompletely resected (n = 43) or unresected (n = 12) RPLS were identified between 1982 and 1999. Statistical analyses were performed using the log-rank test and Kaplan-Meier estimates with disease-specific survival as the primary end point. Variables studied included age, gender, recurrent versus primary disease, tumor grade, and tumor size. The patient population consisted of 34 men and 21 women with a mean age of 61 +/- 14 (SD) years. The median time to death was 10 months (range 1 to 83 months) with a median followup of 12 months (range 1 to 60 months) for survivors. Partial resection was an independent factor for increased survival as compared with exploration or biopsy only (median survival 26 versus 4 months, p < 0.0001). Of patients who received incomplete resections, locally recurrent presentation (n = 19) versus primary disease (n = 24) was a negative prognostic variable (median survival 17 versus 46 months, p = 0.009). Successful palliation of symptoms was achieved in 24 of 32 patients (75%) with preoperative symptoms. In select patients with unresectable RPLS, incomplete surgical resection can provide prolongation in survival and successful symptom palliation. Most likely to benefit are those patients presenting with primary tumors, suggesting that surgical resection should be attempted in the majority of patients.

Background: Sott tissue sarcomas (STS) of the groin may present a difficult problem because or misdiagnosis as groin hernia and proximity to major neurovascular structures. We evaluated our management and survival in a large cohort of patients., Study Design: Patients treated between July 1, 1982 and July 1, 1998 with primary or recurrent STS of the groin were included. Groin sarcomas were defined as those tumors within 5 cm of the inguinal crease. Patient, tumor, clinical, and survival data were analyzed using a log rank test and Cox regression., Results: We treated and followed 88 patients with STS of the groin. The median age was 52 years (range 16 to 86 years) and 55 patients (63%) were male. Disease-specific survival was 72% at 5 years. Tumors tended to be larger than 5 cm (52%), deep (72%), and high-grade (60%). Unfavorable prognostic factors for disease-specific survival were high grade (p < 0.001), neurovascular invasion (p < 0.001), positive margin (p < 0.01), deep depth (p < 0.01), and selection for adjuvant therapy (p < 0.005). Multivariate analysis indicated age greater than 50 years (p < 0.05), high grade (p < 0.001), neurovascular invasion (p < 0.001), and positive microscopic margins (p < 0.001). Fourteen patients (16%) were diagnosed with STS at hernia operation then went on to a definitive operation with no impact on survival. Seventeen patients (19%) had involvement of a major vessel or nerve, and 5 of these ultimately required amputations, 3 for local recurrence., Conclusions: High grade, neurovascular invasion, and positive microscopic margins are associated with poor outcomes. The biology of these tumors is similar to other extremity STS, and similar principles of management apply. Even with neurovascular involvement, most patients with primary groin STS do not require amputation.

Background: Even after an apparent complete resection of sarcomatous pulmonary metastases, 40% to 80% of patients will re-recur in the lung. The benefit of subsequent re-resection is poorly defined. This study examines patient survival after repeat pulmonary exploration for re-recurrent metastatic sarcoma at a single institution., Study Design: Between July 1982 and December 1997, data on 3,149 adult in-patients with soft tissue sarcoma were prospectively gathered. Of these, pulmonary metastases were present or developed in 719 patients and 248 underwent at least one resection. Of the patients relapsing in the lung after an apparently complete resection, 86 underwent reexploration. Disease-specific survival (DSS) after re-resection was the end point of the study. Time to death was modeled using the method of Kaplan and Meier. The association of factors to time-to-event end points was analyzed using the log-rank test for univariate analysis and the Cox proportional hazards model for multivariate analysis. Clinicopathologic factors were analyzed with the Pearson chi-square or Fisher's exact test when appropriate., Results: The median DSS after re-resection for all patients undergoing at least two pulmonary resections was 42.8 months with an estimated 5-year survival of 36%. The median DSS in patients with complete reresection was 51 months (n = 68) compared with 6 months in patients with an incomplete re-resection (n = 16, p<0.0001). Patients with one or two nodules at re-resection (n = 39) had a median DSS of 51 months compared with 20 months in patients with three or more nodules (n = 40, p = 0.003). Patients in whom the largest metastasis re-resected was less than or equal to 2 cm (n = 33) had a median DSS of 44 months compared with 20 months in patients with metastasis greater than 2 cm (n = 43, p = 0.033). Patients with primary tumor high-grade histology (n = 75) had a median DSS of 32 months and patients with low-grade histology (n = 11) had a median DSS that was not reached (p = 0.041). Three independent prognostic factors associated with poor outcomes may be determined preoperatively: > or =3 nodules, largest metastases > 2 cm, and high-grade primary tumor histology. Patients with either zero or one poor prognostic factor had a median DSS > 65 months and patients with three poor prognostic factors had a median DSS of 10 months., Conclusions: Reexploration for recurrent sarcomatous pulmonary metastases appears beneficial for patients who can be completely re-resected. Outcomes are described by factors that may be determined preoperatively, including metastasis size, metastasis number, and primary tumor histologic grade. Patients who cannot be completely re-resected or those with numerous, large metastasis and high-grade primary tumor pathology have poor outcomes and should be considered for investigational therapy.

Background: Laparoscopic staging is an effective and accurate means of staging pancreatic cancer. But, the frequency of subsequent surgical bypass to treat biliary or gastric obstruction in laparoscopically staged patients with unresectable adenocarcinoma is unknown. The development of biliary and gastric obstruction in patients with unresectable pancreatic adenocarcinoma has been reported to occur in as many as 70% and 25% of patients, respectively. Previously, staging for patients with pancreatic cancer was achieved by laparotomy and the anticipated high rate for these patients to develop obstruction led to prophylactic bypass procedures. As laparoscopic staging for pancreatic cancer becomes a standard modality, the need for prophylactic bypass procedures in these patients needs to be examined., Study Design: Analyses of laparoscopically staged patients (n = 155) with unresectable, histologically proved pancreatic adenocarcinoma, from a single institution treated between 1993-1997 were performed. The frequency of surgical bypass in a prospective cohort of patients with unresectable pancreatic adenocarcinoma who did not undergo open enteric or biliary bypass at the time of laparoscopic staging was determined., Results: Laparoscopic staging revealed that 40 patients had locally advanced disease and 115 had metastatic disease. Median survival for patients with locally advanced and metastatic disease was 6.2 and 7.8 months, respectively. Postlaparoscopy followup revealed that 98% (152 of 155) of these patients did not require a subsequent open surgical procedure to treat biliary or gastric obstruction., Conclusions: These results do not support the practice of routine prophylactic bypass procedures. As such, we propose that surgical biliary bypass can be advocated only for those patients with obstructive jaundice who fail endoscopic stent placement, and gastroenterostomy should be reserved for patients with confirmed gastric outlet obstruction.

Background: Splenectomy at the time of resection of esophageal, gastric, or colon cancer has been correlated with inferior longterm survival. No such effect has yet been demonstrated for pancreatic cancer., Study Design: Patients undergoing resection of pancreatic adenocarcinoma with curative intent at Memorial Sloan-Kettering Cancer Center between October 1983 and October 1995 were identified from a prospective clinical database. The impact of splenectomy on hospital stay and survival was calculated with univariate and multivariate nonparametric methods., Results: Of 332 patients undergoing pancreatectomy, 326 with confirmed local or regional disease only formed the study cohort. Of these, 37 underwent concomitant splenectomy (11.4%). Splenectomy was significantly correlated with distal or total pancreatectomy, primary location in tail or body, portal vein invasion or resection, a larger maximal tumor diameter, and an operative blood loss of greater than 2,000 mL. Death or need for reoperation was not affected by splenectomy. Patients undergoing splenectomy had a higher median transfusion requirement (3 versus 1; p = 0.002). The median postoperative length of stay was 15 days regardless of splenectomy. At a median followup of 16.3 months (36.4 months for surviving patients), the median actuarial survival was 12.2 months with splenectomy versus 17.8 months without splenectomy (p<0.005). On multivariate analysis, splenectomy emerged as an independent factor predictive of decreased postoperative survival (p = 0.02), in addition to pathologic lymph node status (p = 0.0002), tumor diameter (p = 0.0004), and tumor differentiation (p = 0.007). Tumor location within the pancreas and the type of pancreatectomy were not independent prognostic factors influencing survival., Conclusions: After pancreatectomy for pancreatic cancer, splenectomy has no significant measurable impact on postoperative recovery, but has a negative influence on longterm survival independent of disease-related factors. Unless required because of tumor proximity or invasion, splenectomy should be avoided in the operative treatment of exocrine pancreatic cancer at any location.

Background: Peritoneal cytology is clinically useful in gastric and gynecologic malignancies. Its role in pancreatic adenocarcinoma remains less well defined. Controversy exists as to the relationship between percutaneous fine needle aspiration (FNA) of the pancreas and shedding of malignant cells with the peritoneum. The aim of this study was to determine whether positive peritoneal cytology (PPC) predicts unresectability of pancreatic adenocarcinoma and impacts on overall survival. In addition, the study aimed to determine whether antecedent FNA increases the incidence of PPC., Study Design: Between January 1993 and June 1996, 228 patients with radiographically resectable pancreatic adenocarcinoma underwent laparoscopic staging. Specimens were taken from right and left upper quadrants at the beginning of laparoscopy. Various prognostic factors were analyzed., Results: PPC was identified in 34 patients (15%). Of patients that had an antecedent FNA, 20% had PPC, and 13% of those without an antecedent FNA had PPC (p = 0.22). The majority of patients with PPC had stage IV disease (26 of 34 [76%]) and only 8 (24%) had no evidence of metastases. Overall survival was significantly higher in patients with negative peritoneal cytology (NPC) compared with PPC (p<0.0006). PPC had a positive predictive value of 94.1%, specificity of 98.1%, and a sensitivity of 25.6% for determining unresectability of pancreatic adenocarcinoma. PPC was not an independent prognostic variable for survival on multivariate analysis., Conclusions: PPC is associated with advanced disease and is highly specific in predicting unresectability of pancreatic adenocarcinoma, resulting in decreased survival. Antecedent FNA is not associated with an increased the incidence of PPC, nor does it significantly impact on overall survival.

Background: Studies have consistently confirmed the benefit of liver resection for metastatic colorectal cancer. Few reports, however, have a long enough followup or sufficient 5-year survivors to study the clinical course of patients beyond 5 years., Study Design: From July 1985 through December 1991, 456 patients underwent liver resection for colorectal metastases. Ninety-six actual 5-year survivors (21%) were identified and their clinical course retrospectively reviewed., Results: Five-year survivors (n = 96) were more likely to have a Duke's B primary colorectal carcinoma, fewer than four metastatic lesions, unilobar disease, and a negative histologic margin when compared with patients not surviving 5 years (n = 298). Forty-four (46%) of the 96 five-year survivors had a recurrence after hepatectomy. Of these 44, 19 (43%) were rendered disease free after further treatment. Overall, 71 of the 96 five-year survivors were free of disease at last followup. The actuarial 10-year survival of this group was 78%., Conclusions: Patients that are disease free 5 years after liver resection are likely to have been cured by liver resection. Patients should be aggressively followed for recurrence because of the potential for further treatment and longterm survival.

Background: Variable reports exist about the early and long-term outcome after operative treatment for gastric carcinoma in the elderly. This study was designed to describe perioperative morbidity and factors influencing the length of hospitalization in patients older than 70 years of age in a tertiary care cancer center., Methods: Patient information for a 10-year period, between July 1985 and July 1995, was obtained through a prospective database and chart review. Complications and length of stay were tabulated. Patient-, disease-, and treatment-related factors and their influence on outcome were compared by univariate and multivariate analysis using nonparametric product-limit models., Results: Of 385 patients aged 71 years or older with gastric adenocarcinoma, 310 underwent resection. Postoperative complications occurred in 47.1 percent. Infectious complications predominated, most frequently involving intra-abdominal and pulmonary sites. Perioperative mortality was 7.1 percent. The median postoperative length of stay was 13 days (95 percent confidence interval 12 to 14 days; 25th percentile, 10 days; 75th percentile, 20 days). Factors independently predicting an increased duration of stay were presence of any complications, the type of resection, site of the primary carcinoma, and presence of postoperative infection. Complications added 30.4 percent of total patient days, or an average of 11.5 hospital days per patient with a complication. Although patients with postoperative complications had inferior overall and disease-specific survival, this was not an independent prognostic variable. Compared with patients younger than 70 years of age, elderly patients had a significantly increased hospital stay., Conclusions: The single most important factor leading to increased length of hospitalization is the occurrence of any complication. Although complicated by higher morbidity and mortality, the resection of gastric carcinoma in elderly patients can be performed relatively safely and leads to survival that is comparable to younger patients. The findings support careful patient selection and optimal preparation of elderly patients undergoing resection for gastric carcinoma.

Background: Duodenal adenocarcinoma is a rare malignancy with a poorly defined natural history and outcome. The factors that affect management and survival of patients with this disease remain controversial. This study analyzed the ten-year experience at one institution with primary duodenal adenocarcinoma to define factors that have an impact on patient survival. In addition, the outcome of patients with resected duodenal adenocarcinoma was compared with that of patients with gastric and pancreatic adenocarcinoma., Study Design: A retrospective review of the prospective database for patients with peripancreatic lesions treated at Memorial Sloan-Kettering Cancer Center between 1983 and 1994 identified 79 patients with a primary duodenal adenocarcinoma. Demographics, presenting symptoms, operative variables, pathologic findings, and survival data were analyzed. Multivariate comparisons and actuarial survival were calculated using these variables., Results: A curative resection was performed in 42 (53 percent) of the 79 patients, including 38 pancreaticoduodenectomies and four duodenal resections. The overall projected five-year survival rate was 31 percent, with resected and nonresected patient survival rates of 60 and zero percent, respectively (p < 0.0001). Nodal metastases, regardless of location, did not have an impact on survival. While stage was a significant factor in survival on univariate analysis, no survival difference was noted between stages I, II, and III. Only resectability and presence of non-nodal metastases predicted outcome on multivariate analysis., Conclusions: Resectability and presence of distant metastatic disease are the strongest determinants of outcome for patients with duodenal adenocarcinoma. Staging and nodal status offer little prognostic information and nodal positivity should not preclude resection. As patients have symptoms similar to those of pancreatic adenocarcinoma but have an outlook more comparable to gastric adenocarcinoma, a vigorous approach to resection is justified.