SS

AT

Ab

stra

cts

Mo1731

Incomplete Lower Esophageal Sphincter Relaxation on High-ResolutionManometry Is an Independent Predictor of Solid Diet Failure in Post-Roux-en-Y Gastric Bypass PatientsShikha Mangla, Ana C. Tuyama, Robert Burakoff, David B. Lautz, Christopher C.Thompson, Walter W. Chan

Background: Roux-en-Y gastric bypass (RYGB) is an effective surgery for weight loss in obesepatients. Current guidelines recommend advancement to regular diet in 1-2 months post-RYGB. Failure to advance or dietary intolerancemay have clinical and nutritional implications.A prior study suggested that up to 30% of post-RYGB patients may develop dysphagia.RYGB may affect the Vagal innervation to the esophagus, and the resultant esophagealdysmotility may play a role in post-RYGB dietary complications. Understanding esophagealmotor functions by high-resolution manometry (HRM) and their association with dietaryoutcome post-RYGB may allow more effective, targeted therapy for symptoms and dietarycomplications. Aim: To investigate the association between esophageal motor dysfunctionson HRM and intolerance to solid diet among post-RYGB patients. Methods: This was aretrospective cohort study of post-RYGB patients who underwent HRM at a tertiary carecenter in 6/2007-5/2012. Patients with underlying esophageal dysmotility pre-RYGB, HRMperformed less than 2 months after RYGB, or need for parenteral or tube feeding wereexcluded. The primary outcome was diet at the time of HRM (liquid [LD] vs solid [SD]).Esophageal motor characteristics were extracted from HRM. Fisher-exact or chi-squared testfor binary variables and student's t-test for continuous variables were used to assess fordifferences between LD and SD groups. Multivariate analysis was performed using forwardstepwise logistic regression. Results: 63 patients met inclusion criteria (age 51±10.3 yrs,91% F), and 21 subjects (33.3%) could only tolerate LD. Patients on LD were more likelyto have at least one abnormal parameter on HRM than those on SD (61.9% vs 28.6%, p=0.01). Univariate analyses showed that elevated basal lower esophageal sphincter (LES)pressure (9.52% vs 0%, p=0.04), incomplete LES relaxation (22% vs 0%, p= 0.04), increasedesophageal body contraction amplitude (119±56 vs 93±41 mmHg, p=0.05), and dysphagia(52% vs 16%, p= 0.003) were significantly associated with LD. On multivariate analysis,incomplete LES relaxation remained an indepedent predictor for LD (OR 11.73, p=0.02).Conclusions: Post-RYGB patients unable to tolerate SD are more likely to have abnormalfindings on HRM. Incomplete LES relaxation is independently associated with LD use, whileother hypermotility patterns (hypertensive LES and increased esophageal body contraction)are also more prevalent. In addition to pouch or anastomotic abnormalities, esophagealmotor dysfunction should be considered in assessing post-RYGB patients' failure to tolerateSD. HRM should play a role in evaluating post-RYGB dietary complications. Future studiesshould examine the potential causes of this dysfunction and explore the effect of therapiestargeting LES relaxation on clinical and dietary outcome.

Mo1732

Transplantation for HCC Improves Progression Free Survival but Not OverallSurvival When Compared to ResectionRafael Pieretti- Vanmarcke, Hui Zheng, Nahel Elias, David L. Berger, Kenneth Tanabe,Keith D. Lillemoe, Cristina R. Ferrone

Objective: To compare the outcomes of patients with hepatocellular carcinoma (HCC)undergoing either liver transplantation(LT) or resection(LR). Methods: A single institutionretrospective analysis of 327 HCC patients treated between 8/1991-12/2011. Results: A totalof 327 patients with HCC underwent surgical treatment of whom 79% were male, 19%had hepatitis B and 44% hepatitis C. Patients underwent transplantation (n=138) or surgicalresection (n=189). Of the resected patients 126 did not meet the Milan Criteria (MC) while63 patients were within MC. When comparing resected patients within MC to transplantedpatients the median tumor diameter was 3.2 cm and 3.0 cm, respectively. Recurrence rateswere 57% for resected patients within MC and 13% for LT patients (P,0.0001). The modelend-stage liver disease (MELD) score median was 7 for resected patients within MC and 10for LT patients. The median overall survival (OS) was 40 months for both resected andtransplanted patients. The OS at 1, 3, and 5 years was 47%, 40%, and 23% for resectedpatients within MC and 59%, 49%, and 33% for transplanted patients. (p=ns). Significantclinicopathologic factors predicting survival were age, size of lesion, lymphovascular invasion,Patients outside of MC who were resected had a significantly decreased survival comparedto patients within MC and those who were transplanted. Conclusions: For HCC patientswithin Milan criteria, transplantation is associated with a lower recurrence rate, but not asignificantly improved overall survival. Patients outside of Milan criteria had a significantlypoorerOSwhen compared to patients withinMilan criteria whowere resected or transplanted,reflecting a more aggressive disease biology.

Mo1733

Influence of Preoperative Laboratory Values on Perioperative MortalityFollowing Hepatic Resection for MalignancyMashaal Dhir, Lynette M. Smith, George Dittrick, Quan P. Ly, Aaron R. Sasson,Chandrakanth Are

Background: Abnormal preoperative laboratory values have been associated with increasedmortality in patients undergoing hepatic resection for malignancy. However, cutoff valuesfor these preoperative labs have been defined arbitrarily. The aim of the current study wasto identify cut off values for these preoperative laboratory denominators which can helpidentify patients at increased risk of mortality. Methods: Patient undergoing liver resectionfor malignancy (primary and secondary) were extracted from 2005-2010 National SurgicalQuality Improvement Database. We determined the optimal cutoffs for each laboratorydenominator using the classification and regression tree analysis (CART), and the "party"package for conditional inference trees in R. Patients were classified according to the cutoffsdetermined from CART analysis and logistic regression analysis was used to fit a multivariatemodel, with backward variable selection. Results: A total of 4812 patients who underwentliver resections for malignancy were included. Statistically significant association was seen

S-1102SSAT Abstracts

between increased 30 day mortality and preoperative laboratory values including serum Na,= 135 meq/L, BUN .19 mg/dl, serum creatinine .1.68 mg/dl, serum albumin ,=2.6 g/dl, bilirubin .1.8 mg/dl, SGOT .50 IU/L, alkaline phosphatase of . 149 IU/L, WBC.10,790/ul , Hct ,= 28, and INR .1.1. In a multivariate logistic regression model, albumin,=2.6, SGOT.50, INR.1.1, BUN.19, and alkaline phosphatase.149 are independentpredictors of 30 day mortality with an area under the curve of 0.77. Conclusions: Cutofflaboratory values defined in the current study may help identify patients who are at higherrisk of mortality from hepatic resections.

Mo1734

Resected Splenic Masses Discovered on Imaging Are Frequently Malignant: AReview of 148 CasesCiaran T. Bradley, Amudhan Pugalenthi, Vivian E. Strong, William R. Jarnagin, Daniel G.Coit, T. P. Kingham

Background: Solid and cystic splenic masses discovered on imaging studies often posediagnostic and management dilemmas. This study analyses a large series of splenectomiesto identify preoperative factors associated with malignant splenic masses. Methods: Pathologyrecords at a single institution were reviewed for all splenectomies. Those performed as acomponent of a larger resection, for lymphoma staging, or debulking for a surface malignancywere excluded. Demographic and clinicopathologic factors were obtained. Univariate andmultivariate analyses identified factors associated with an increased risk of malignancy.Results: Between 1986 and 2012, 2,745 patients underwent splenectomy. 148 were per-formed for splenic lesions identified on abdominal scans. The indication for resection includedsuspicious imaging characteristics such as heterogeneity or growth over time (120, 81%),a cancer history (113, 76%), and/or symptoms (39, 26%). The majority of patients underwenta preoperative CT (138), although PET (25), MRI (23), and ultrasound (8), were alsoincluded in patient evaluations. Among the resected spleens, the majority had a malignantmass (93, 63%). 90% were parenchymal metastases, including ovarian cancer (39, 42%),followed by melanoma (14, 15%) and colorectal cancer (9, 10%). While the majority of thepatients with malignant splenic lesions had a previous history of cancer (85 of 93; 91%),among those patients without a previous history of cancer (n=35), most had benign lesions(77%). On multivariate analysis of several clinicopathologic factors, a previous history ofcancer was the only independent predictor of malignancy in the splenic lesion (odds ratio6.3; 95% CI, 2.32-16.97; p=0.00). Imaging described as "suspicious for malignancy" by theradiologist (e.g. by virtue of heterogeneity) or lesions that enlarged on interval scans didnot correlate with malignancy in the spleen. Conclusion: While the spleen is an uncommonsite of malignant disease, resected splenic masses are frequently malignant, especially inpatients with a previous history of cancer.Factors Associate with Malignancy in Resected Splenic Masses

Mo1735

Analysis of Clinicopathological Factors Contributing an Actual 5 Year SurvivalAfter Hepatectomy for Intrahepatic CholangiocarcinomaShutaro Hori, Kazuaki Shimada, Satoshi Nara, Minoru Esaki, Yoji Kishi, Tomoo Kosuge,Hidenori Ojima

BACKGROUND: Hepatectomy is the only chance of cure for patient with intrahepaticcholangiocarcinoma (ICC), because there is a lack of other effective treatments for achievingan actual 5 year survival. However clinicopathological feature predicting 5-year survival afterhepatectomy has not been well clarified. METHODS: 113 consecutive ICC patients withmass-forming (MF) macroscopic tumor type and MF plus periductal infiltrating (PI) type,who underwent surgical resection at a single institution between January 1990 and December2006, were retrospectively analyzed. Patients who died of unknown causes, and who waslost to follow up within 5 year were excluded from the study. The clinicopathologic featuresof patients who survived more than 5 years were compared with those died within 5 years.RESULTS: Of all 113 patients underwent surgical resection, 33 patients (29.2%) survivedmore than 5 years. In univariate analysis, MF type (p=0.015), preservation of extra bile ductresection (p=0.014), operation without blood transfusion (p=0.001), absence of intrahepaticmetastasis (p=0.006), absence of vascular invasion (p=0.022), negative lymph node involve-ment (p,0.001), and microscopic curative resection (p=0.001) were significantly related to5 year survival. Multivariate analysis showed that operation without blood transfusion,absence of intrahepatic metastasis and negative lymph node involvement were independentfactors associated with survival for more than 5 years, with odds ratios (95% confidenceinterval) of 6.743 (1.784-25.491; p=0.005), 4.302 (1.391-13.306; p=0.011), 3.886 (1.401-10.664; p=0.009), respectively. CONCLUSION In MF and MF+PI type of ICC, hepatectomywithout blood transfusion, absence of intrahepatic metastasis and negative lymph nodeinvolvement significantly contribute an actual 5 year survival.