QS35. Colectomy for Colon Cancer in Patients With a Prior Ventriculoperitoneal Shunt the Department of Veterans Affairs Experience

ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS 283 pancreatic adenocarcinoma. However, a more detailed analysis considering outcomes without surgery for each primary tumor site is needed before the value of this aggressive surgical approach can be completely assessed. QS33. EXTENDED LEFT-SIDED PANCREATECTOMY WITH SPLEEN PRESERVATION. Arash Mohebati1, Roderich E. Schwarz2; 1UMDNJ - Robert Wood Johnson Medical School, New Brunswick, NJ; 2UT Southwestern Medical Center, Dallas, TX Background: Left-sided or distal pancreatectomy (DP) is frequently performed in conjunction with splenectomy, although splenectomy can be linked to various untoward effects, and sparing of the main splenic vessels (SV) is not necessary for successful spleen preservation (SP). Methods: We reviewed clinical records of all patients undergoing DP in a single-surgeon practice to assess feasibility and outcomes of SP. Results: Between 1997 and 2007, 41 of 177 pancreatic resections involved a DP (23%). There were 14 men (34%) and 27 women (66%), with a median age of 60 years (range: 34-86). Resection indications included 26 solid masses, 10 cysts, 4 combinations thereof, and one diffuse process. Four procedures were en bloc resections, 2 total pancreatectomies, while 8 of the remaining 35 DPs were performed laparoscopically. SP was accomplished in 33 of 34 possible cases (spleen preservation rate 97%), despite SV resection in 27 of these (82%). The postoperative complication rate was 24%, and there was no postoperative death. The median length of stay was 6 days (4-24). Pancreatic fistulae occurred in 2 patients (5% of patients at risk), and one SPDP led to splenic infarct. At a median follow-up of 16 months (2.5-89), no other clinically relevant problems specific to SP have become apparent. One patient after DP with splenectomy expired from postsplenectomy sepsis. Conclusions: Pancreatic fistula rate and other outcomes in this small DP experience compare favorably to many other DP series. Few spleen-specific complications and the radicality of resection support the liberal use of SP with SV resection, irrespective of an open or laparoscopic approach. QS34. PREOPERATIVE DIAGNOSIS OF TUMOR EXTENT OF BILE DUCT CANCER BY INTRADUCTAL ULTRASONOGRAPHY. Jun Ienaga, Yoshihiko Sadakari, Reiko Tanabe, Norihiro Sato, Shunichi Takahata, Hiroki Toma, Toshinaga Nabae, Masafumi Nakamura, Koji Yamaguchi, Masao Tanaka; Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Backgrounds: For bile duct cancer, it is important to evaluate the vertical and horizontal extent correctly. Aim: To determine the usefulness of intraductal ultrasonography (IDUS) for diagnosis of tumor extent of bile duct cancer. Patients and Methods: Between July 2001 and September 2007, 54 patients with bile duct cancer were studied by IDUS. Among them, 25 patients underwent surgical resection. IDUS findings with respect to vertical invasion to the subserosal layer, pancreatic parenchyma, right hepatic artery and portal vein, and horizontal spread along the bile duct wall were compared with pathological findings in these 25 patients. All pathological descriptions were based on the General Rules for Surgical and Pathological Studies on Cancer of the Biliary Tract (5th edition) published by the Japanese Society of Biliary Surgery. Results: Subserosal invasion was correctly diagnosed by IDUS in 20 patients (80%). Pancreatic parenchymal invasion was accurately determined in 20 patients (80%). Although there were no patients who had histological invasion to the right hepatic artery in this study population, IDUS diagnosis was correlated well with surgical diagnosis in this regard (accuracy: 88%). One patient who had histological invasion to the portal vein could be diagnosed as having the invasion preoperatively by IDUS; all other patients without portal vein invasion were diagnosed correctly. Horizontal spread to the hepatic side was accurately determined in 22 patients (88%). Those who had biliary drainage performed either by the percutaneous transhepatic method or

endoscopic retrograde method tended to show low accuracy compared with those who did not received biliary drainage (71% (5/7), 94% (17/18), respectively, p⫽0.11). The tumor of bile duct and the surrounding structure could be visualized in detail by IDUS. IDUS could accurately evaluate the vertical invasion to the subserosal layer and other organs, and the horizontal extent of mural invasion to the hepatic side. However, once biliary drainage had been introduced, the bile duct wall became thickened and this made it difficult to distinguish between the tumor extent and benign wall thickening in response to a biliary drainage catheter. Therefore, it seemed to be better to perform IDUS before biliary drainage. Conclusions: IDUS is useful to precisely assess the vertical and horizontal tumor extent of bile duct cancer when performed before biliary drainage. QS35. COLECTOMY FOR COLON CANCER IN PATIENTS WITH A PRIOR VENTRICULOPERITONEAL SHUNT THE DEPARTMENT OF VETERANS AFFAIRS EXPERIENCE. Edel M. Doorley1, Andrew R. Barina2, Katherine S. Virgo2, Anil M. Bahadursingh2, Frank E. Johnson2; 1University of Liverpool, Liverpool, United Kingdom; 2Saint Louis University, St. Louis, MO Objective: Many patients have VP shunts implanted for congenital conditions. Subsequent abdominal operations in these patients are reportedly hazardous. We aimed to determine the clinical course of adults with ventriculoperitoneal (VP) shunts for acquired conditions who later required colectomy for colon cancer. Introduction: Approximately 18,000 cerebrospinal fluid shunts, the majority of which are VP, are implanted each year in the USA. These patients may subsequently require colectomy for colon cancer. Whether the risk of adverse events, particularly infection, is increased in such patients is not known. Methods: A search of national Department of Veterans Affairs (DVA) databases was conducted to identify all veterans with a VP shunt from 1989-2003 who later underwent colectomy as curative-intent treatment for colon cancer between 1994-2003. All were healthy upon entry into military service and required shunts for conditions acquired later. Patient medical records were analyzed to determine if the presence of a VP shunt affected the colectomy procedure or the postoperative course. Results: There were 4,219 unique inpatients and 795 unique outpatients with ICD-9 codes for VP shunt and 16,514 with codes for colectomy for colon cancer in the DVA system for the years specified. Fourteen had codes for both colectomy for colon cancer and pre-existing VP shunt. Four met our inclusion criteria and had sufficient data for analysis; all had uncomplicated postoperative courses with no instances of extensive adhesions encountered during colectomy, postoperative infection, or postoperative shunt malfunction. Discussion and Conclusions: This is the only English-language report on this topic, to our knowledge. Patients who receive VP shunts for acquired hydrocephalus as adults and later receive colectomy as curative treatment for colon cancer in the DVA system appear to experience a postoperative course similar to that of patients without VP shunts. QS36. NITRIC OXIDE-MEDIATED RADIOSENSITIZATION TO IONIZING RADIATION-INDUCED APOPTOSIS OF COLON CANCER CELLS. Derrick Chen, Laura Ortega, David Chen, Edward H. Livingston, Sergio Huerta; University of Texas Southwestern, Dallas, TX Background: Metastatic colon cancer cells SW620 are resistant to ionizing radiation (IR)-mediated apoptosis. DETA/NONOate (DETA) is an NO donor, which mimics NO sustained release for over 20 hours. DETA causes mitochondrial permeability resulting in the release of pro-apoptotic mediators, while inhibiting the NF␬B antiapoptotic pathway. Methods: SW620 pre-treated with DETA (1000 ␮g X 24 h) and untreated cells were subjected to IR treatment at 0, 1, 2, 3, and 5 Gy. Apoptosis was measured by flow cytometry