SS

AT

Ab

stra

cts

analysis the CHILD classification (mortality: 6% Child A; 11% Child B, 45% Child C;p<0.001), higher/increasing MELD score (p<0.001), higher/increasing ASA score (p<0.001),emergency procedures (35% vs 5% elective; p<0.001), major procedures (p<0.02), need fortransfusions (36% vs 4% in patients without transfusions; p<0.001) and various preoperativelaboratory values (anemia, thrombocytopenia, hyponatremia; all p<0.05) were associatedwith increased mortality. In multivariate risk factor analyses blood transfusions (p<0.001;RR 7), ASA score (p<0.01), Child class (p<0.02) and a thrombocytopenia (p<0.02) wereindependent predictors for mortality. The MELD score, emergent procedures and extent ofsurgery showed a trend but did not significantly predict mortality in the multivariate model.Conclusions: Patients requiring blood transfusions have a very high risk for mortality.Preoperative liver function and co-morbidity also predict early mortality after surgery. Inour series the CHILD score was a better predictor for postoperative mortality than theMELD-score.

Su1538

Single Hepatocellular Carcinoma Less Than 3 CM in Cirrhotic Livers: IsResection Better Than Ablation?Marco Massani, Cesare Ruffolo, Luca Bonariol, Ezio Caratozzolo, Marco Scarpa, FrancescoCalia di Pinto, Francesco E. D'Amico, Bruno Pauletti, Giuseppe Battistella, Nicolò Bassi

Background Hepatocellular carcinoma (HCC) is one of the most frequent tumors and it isthe predominant primitive liver cancer and in most cases associated with cirrhosis, regardlessof the etiology. Aim The purpose of this study was to compare the overall survival afterpercutaneous ablation and resection in patients with a single HCC lesion measuring lessthan 3 cm in diameter originating on liver cirrhosis. Patients and methods From January1999 to December 2008 556 consecutive patients were evaluated in our center for HCC.Only patients with cirrhosis and a single nodule < 3cm were taken into consideration: 48underwent liver resection and 51 treated with percutaneous procedures. Results The min-imum follow up was 20 months. The survival rate of resected patients was significantlyhigher compared to the ablated patients (p=0.0006). Child A was a protective factor forboth groups (p=0.0001) andHCVpositive patients hadworse outcomes (p=0.005).Moreover,age and survival were significantly associated to resected patients (p=0.0195). Early recurrenceoccurred in 3 patients after resection and in 7 after ablation. Conclusions This experienceconfirmed that in case of single nodule <3cm in cirrhosis resection must be considered asthe primary choice for all patients with adequate functional reserve. Nevertheless the ablativetreatment can be considered a good second line option since it ensure good results in termsof survival.

Su1539

Safety and Outcomes Following Resection of Colorectal Cancer LiverMetastases in the Era of FOLFOXIlia Gur, Jesse A. Wagner, Brett C. Sheppard, Susan L. Orloff, Gina M. Vaccaro, CharlesD. Lopez, Brian S. Diggs, Kevin G. Billingsley

Objective Report the safety and outcomes following the introduction of routine periopereativemultiagent chemotherapy in the multidisciplinary treatment of patients with colorectal livermetastases (CRLM). Background Increasingly preoperative chemotherapy is integrated intothe management of patients with liver metastases. This strategy has likely expanded thenumber of surgical candidates but postoperative safety and survival have not been clearlydefined. Methods We performed a retrospective review of all patients undergoing liverresections for metastatic colorectal cancer between 2003 and 2011 in a single academiconcology center. Demographic data, tumor characteristics, chemotherapy, details of surgicalprocedure, complications and survival were analyzed. Results The study population consistedof 158 patients that underwent 169 liver operations. 11(6.9%) patients had repeat resections.Median length of follow up from a first liver resection was 22.3 months. 87 (55%) patientspresented with synchronous lesions. 114 patients (72%) underwent chemotherapy prior tothe liver resection (of them 68% FOLFOX, 12% FOLFIRI). Mean size of the lesions 3.97cm (0.2 -18 cm) .Preoperative Portal Vein Embolization was utilized in 16 (10.1%) patientsOverall survival was 89%, 57% and 17% at 1, 3 and 5 years respectively. Median survivalwas 42.8 months. Perioperative mortality (30, 60 and 90 days respectively) was 1.26%,1.89 % and 2.53 %. Overall complication rate was 24% (5% - liver related) The complicationrate was not significantly different if patients had preoperative chemotherapy (27% nochemotherapy, 24 % with chemotherapy). Mean length of stay was 8.68 days. On univariateanalysis negative predictors of survival included positive margins, > 3 lesions, patient age>70 years. On a multivariate analysis only the presence of >3 lesions predicted poor survival.Conclusions In recent years preoperative oxaliplatin-based chemotherapy has become com-monplace in the management of patients with resectable CRLM. Our results suggest thateven with chemotherapy and resection only a subset of patients remain disease free after 5years. However, even in high risk patient with multiple lesions, preoperative chemotherapymay be administered safely without increase in postoperative complications. These resultssupport the use of perioperative chemotherapy particularly in patients with multifocal ( >3 lesions) metastatic disease in the liver.

Su1542

Underuse of Surgical Therapy of Gastrointestinal Cancer in the United StatesAttila Dubecz, Norbert Solymosi, Michael Schweigert, Rudolf J. Stadlhuber, Jeffrey H.Peters, Hubert J. Stein

BACKGROUND Surgery is the mainstay of curative therapy for most localized gastrointestinal(GI) malignancies. Our objective was to evaluate the utilization of surgery in non-metastaticGI cancer and identify factors predicting failure to undergo surgery. METHODS Using theNational Cancer Institute's Surveillance Epidemiology and End Results-Database (1998-2008), a total of 331,911 patients (esophagus: 20,475; stomach: 18,585; small bowel: 2,647;colon: 184,675; rectum: 45,599; liver: 24,318; pancreas: 35,612) were identified with non-metastatic cancer. The rate of surgical therapy in each type was calculated. Multivariate

S-1058SSAT Abstracts

logistic regression was employed to identify factors predicting failure to undergo surgicaltherapy. Reason for no surgery and the impact of surgery on survival were also assessed.RESULTS Surgical resection for locoregional cancer was surprisingly low for cancers of theliver (27%), pancreas (32%), and esophagus (56%). Cancers of the colon (91%) rectum(72%) stomach (78%) and small intestine (74%) had higher rates although as many as onequarter of patients did not undergo surgical resection. The primary reason for not undergoingsurgery was classified as "not recommended" in from 1-49% of the patients again highestin pancreas (49%), liver (47%) and esophagus (26%). Men, non-white race, patients >80yrs, or those undergoing surgical therapy later in the study period and living in areas withhigh poverty rates were significantly less likely to receive surgical treatment (all p<.0001).Median survival in patients who did not undergo surgical resection was significantly betterthan those with metastatic disease (9 vs 6, p<0.0001) but far worse than patients whounderwent surgery for locoregional disease (96 vs 9, p<0.0001). CONCLUSIONS Whenviewed from a national perspective the rates of surgical resection for locoregional GI cancervary considerably. These data suggest that operative therapy in esophageal, liver and pancre-atic cancer is particularly underutilized.

Su1543

Serum Lipid Levels are Associated With the Severity of Acute PancreatitisJahangir Khan, Isto Nordback, Juhani Sand

Aims. Serum lipid concentrations are known to react during acute disease. In this study,we sought to measure changes in the serum lipid profile during acute pancreatitis andwhether these changes were associated with the severity of the disease. Methods. We analyzed233 patients hospitalized for acute pancreatitis between 1995-1995. All etiologies of acutepancreatitis were included, as were patients with their first acute pancreatitis or recurrencies.Serum samples were obtained during the first days after admission and further follow-upsamples were obtained later during the course of the disease. In most cases (n=203, 87%),samples were available from the first two days of hospitalization. The serum total cholesterol,HDL-cholesterol and triglyceride levels were measured enzymatically and the concentrationsof serum LDL-cholesterol were calculated using the Friedewald formula. Results. The mostcommon etiology for acute pancreatitis was alcohol use (n=131, 56%), followed by biliary(n=48, 21%) and idiopathic (n=36, 16%) pancreatitis. 64 (28%) patients had a severepancreatitis, with 13 (6%) mortalities. Serum total cholesterol, HDL-cholesterol and LDL-cholesterol measured within 2 days of admission were significantly lower in patients withsevere pancreatitis and associated with in-hospital mortalities and longer hospitalization(p<0.05). In subgroup analysis, the findings remained statistically significant in patientswith alcohol induced acute pancreatitis, though were similar with all etiologies. Furthermore,these findings were evident even later during the course of the disease. Conclusions. Serumlipid concentrations react during acute pancreatitis. The levels of serum total cholesterol,HDL-cholesterol and LDL-cholesterol are significantly lower in patients with severe acutepancreatitis and are associated with in-hospital mortality and longer hospital stay. Thesechanges are already present during the early stages of the disease and are similar in alletiologies of acute pancreatitis. Furthermore, the changes observed are present even laterduring the course of the disease. Further studies are needed to study the mechanisms ofthis association.

Su1544

Intra-Abdominal Pressure in Acute Pancreatitis: Canary in Coal Mine?;ResultAfter a Rigorous Validation ProtocolVimal Bhandar, Sumit Budania, Jiten Jaipuria

INTRODUCTION: Intra-abdominal hypertension [I.A.H.] is increasingly reported in patientswith severe acute pancreatitis [S.A.P.] and is associated with significantly higher mortalityrates. Though a clear causal relationship could not be demonstrated, some reports showexcellent outcomes in pancreatitis patients undergoing abdominal decompression suggestingthat I.A.H. may be a target for early intervention. Many studies however highlight the issueof I.A.H. in patients with severe disease with absence of data in those with mild diseasemaking it confusing to conclusively recommend whether Intra-abdominal Pressure measure-ment should be a routine in all patients. AIMS AND OBJECTIVES: The present study wasundertaken to evaluate Intra-abdominal Pressure as a marker of severity in acute pancreatitisand to ascertain the relationship between I.A.H. and development of complications in patientswith S.A.P. MATERIAL AND METHOD: A total of 40 patients [24 male, 16 female] fulfillingthe inclusion criteria were selected in the study. Selected patients were further enrolled intotwo groups [Group 1: Mild Pancreatitis, n=24 and Group 2: Severe pancreatitis, n=16] basedon the definitions given in the Atlanta Symposium. Group 2 patients were further categorizedinto two sub-groups depending upon the presence and absence of raised intra-abdominalpressure [Group 2a: consistently raised I.A.P. > 12mmHg and Group 2b: not satisfyingabove criteria, no elevations in I.A.P.]. OBSERVATIONS: Development of intra-abdominalhypertension was noted to be an early phenomenon in patients with S.A.P. The positiveand negative predictive value of I.A.H. in developings S.A.P. were 100% and 75% respectively.Sensitivity of I.A.H. in identifying those with severe pancreatitis was 50% while the specificitywas 100%. Patients with S.A.P. and I.A.H. also had significantly higher APACHE-2 Scores,a higher CT severity index and increased incidence of persistent SIRS, organ failure, occur-rence of pleural effusions , intra-abdominal collections and overall mortality. CONCLU-SIONS: Presence of I.A.H. in the setting of S.A.P. is associated with a higher incidence ofcomplications including pancreatic necrosis, persistent SIRS, organ failure, pleural effusions,intra-abdominal collections, longer duration of hospital stay , mortality and thus intra-abdominal pressure measurement may have a definite place in being used as a predictivemarker for severe disease. REFERENCES: 1.DeWaele JJ, Hoste E, Blot SI et al Intra-abdominalhypertension in patients with severe acute pancreatitis. Crit Care 2005;9: R452-57 2.AdishBasu. A low cost technique for measuring the intra-abdominal pressure in non-industrializedcountries. Ann R Coll Engl 2007;89:431-37