A1220 SSAT ABSTRACTS GASTROENTEROLOGY, Vol. 108, No. 4

UNSUSPECTED COMMON BILE DUCT STONE ENCOUNTERED AT LAPAROSCOPIC CHOLECYSTECTOMY

C.A. Flores~,uerra. l Ponce. S.A. Cot~eland. I.W. Brow4e,', Department of Surgery, East Tennessee State University, Johnson City, Tennessee.

Widespread use of intraoperative cholangiogcaphy during laparoseopic cholerystectomy has identified a small subset of patients with unsuspected common bile duct (CBD) stones. To assess the management of such patients, we reviewed 28 patients treated since 1990. Average age was 56 years and most common symptoms were fight upper quadrant pain and fatty food intolerance. Mean preoperative bilirubin was 0.75 mg % (maximum 1.2 mg %) and mean alkaline phosphatase was 102 units. Diagnosis was established in all patients by ultrasound of the gallbladder. All patients underwent laparoscopic cholecystectomy. Twenty-one had chronic cholecystitis and 7 had acute cholecystitis. Intraoperative cholangingram demonstrated a single common bile duct stone in 23 patients, multiple stones in 4 patients, and no dye into the duodenum in 1 patient. Laparoscopic cholecystectomy was converted into open common bile duct exploration (CBDE) in 17 patients (61%) while 11 patients had the procedure terminated and had a postoperative ERCP. Nine of 11 (82%) had accompanying sphincterotomy. Seven of the ERCP's were performed within 24 hrs. of surgery. One patient required a second ERCP, but all patients had successful results. Morbidity of ERCP (18%) was not significantly different from the CBDE patients (17.6%). Hospital stay was significantly prolonged in the CBDE patients (9.9 days vs 3.7 days) (P < 0.005). We conclude: 1) Unsuspected CBD stones found at laparoscopic cholecystectomy can be successfully managed by early postoperative ERCP a n d sphincterotomy; 2) Such an approach has a decreased hospital stay and no increased morbidity over CBDE.

C O R T I C O S T E R O I D S I M P R O V E R A T H E R T H A N A G G R A V A T E ACUTE PANCREATITIS Th.Foitzik ~. E.Rvschich 2, B.Forgacs 2. H.G.Hotz 1. E.Klar ~. A.L .Warshaw~,HJ .Buhr t Depts.ofSurgery, 1Free Univ.of Berlin, 2Univ.of Heidelberg, and Maz, s.Gen.Hospital, Boston

Despite numerous reports on cortieosteroid-induced pancreatitis corticosteroids (CS) are used for immunosuppression after pancreas transplantation even when posttransplantation pancreatifis is suspect. The present study evaluates possible adverse effects of CS on pancreatic morphology and mortality in a rodent model of acute pancreatitis which, like posttransplantation pancreatitis, is characterized by impaired pancreatic perfusion and a high susceptibility to adverse factors

Acute pancreatitis was induced in 46 rats by a time- and pressure controlled intraductal infusion of 10mM glycodeoxycholic acid followed by intravenous cerulein hyperstimulation for 6 hrs. Thereafter animals were randomly allocated to treatment with saline (group A) or different dosages of methylprednisolone (groups B-D). After two days of therapy surviving animals were sacrificed. The principal outcome measures were mortality, acinar cell injury, production of ascites, and trypsinogen activation peptides (TAP) in plasma and ascites.

Animals treated with high doses of prednisolone had significantly less ascites and lower hematocrit levels, and showed a trend towards decreased early mortality. Acinar cell injury and intrapancreatic protease activation (data not shown) were not affected by corticosteroid treatment.

Dosage N Mort. I Necrosis I Ascites I Hot z Score [ ml ] [ % ]

A NaC10.9% (vol.equi) 12 42% 2 . 1 + 0 . 2 1 . 4 + 0 . 7 46___2 B Prednisolone 2mg/kg/day 12 42% 2 . 1 + 0 . 4 0.9___0.5 4 4 + 2 C Prednisolone 10mg/kg/day 10 20% 2 . 3 + 0 . 3 0 . 7 + 0 . 4 * 4 l + 1 " D Prednisolone 50mg/kg/day 12 33% 2.14-0.4 0.5+__0.1" 42___2*

1) at 48 hrs; 2) at 18 hrs; *) p<0.05 compared to group A CS do not aggravate acute pancreatitis in the rat, but reduce fluid

sequestration. Our results suggest a beneficial rather than a harmful effect of cortieosteroid treatment in the early phase of acute pancreatitis which is consistent with several clinical case reports. We conclude that corticosteroids can safely be used even when pancreatitis is suspect.

• DRAINAGE IS UNNECESSARY AFFER ELECTIVE LIVER RESECTION- RESULTS OF A RANDOMIZED TRIAL Yuman Fong, M.D., Murray F. Brennan, M.D., Nancy Heffernan, R.N., Leslie H. Blum~an, M.D. Department of Surgery, Memorial Sloan-Kettering Cancer Center, NY, NY

A prospective, randomized trial was performed to determine if imra-abdominal drainage catheters are necessary after elective liver resections. Between April 1992 and April 1994, 120 patients subjected to liver resection were randomized to receive or not receive operative closed-suction drainage. Stratification was by extent of resection and by surgeon. Exclusion criteria included biliary-anteric anastomosis, or preoperative biliary stem. Operative blood loss was not an exclusion criteria and no patient who consented to the study was excluded. Eighty-seven patients (73%) had resection of an hepatic lobe or more (27 lobectomies, 54 trisegmentectomies, and 6 bilobar atypical resections) while 33 had less than a lobectomy (8 wedge resections or enncleations, 9 segmemectomies, and 16 bisegmantectomias). Eighty-fourpatiems (70%) had metastatic cancer, while 36 (30%) had primary liver pathology. Mean+_SEM. Comparisons by Chi-square or Mann-Whimey U.

No Drain Drain p (n=60) (n=60)

Lobectomy or more (n) 43 44 NS Hospital Stay (days) 13,44-0.9 13.1 _+0.8 NS 60 day Mortality (n) 2 2 NS Complications (n) 26 29 NS Percutaneous drainage (n) 11 5 NS Biliary Fistula (n) 3 3 NS Infected collection (n) 0 3 NS

Eight patients in the operatively-drained group required prolonged drainage, including 5 requiring post-operative percutaneous drainage for symptomatic collections, and 3 requiring prolonged drainage with the operatively placed drains for biliary fistula. All infected collections occurred in operatively-drained patients. Two other complications were directly related to the operatively placed drains. One patient developed a subcutaneous abscess at the drain site, and a second developed a subcutaneous drain tract tumor recurrence as the only current site of recurrence. No significant difference in any outcome parameter was found between drained and undrained patients. In our service, elective liver resections without biliary anastomosis are no longer operatively drained, In the first forty-five consecutive resections since conclusion of this trial, only three patients have required subsequent percutaneous drainage. We conclude that abdominal drainage is unnecessary after elective liver resection.

• IMPROVED CLINICAL DIAGNOSIS OF INTRA-ABDOM1NAL ABSCESS BASED ON CONSTITUENT ANALYSIS

S.Galandiuk. S,H. Aouel. H.C. Polk. Jr. Dept. of Surgery, University of Louisville, Louisville, KY 40292

Intra-abdominal abscesses ( IAA ) are common and serious complications following gastrointestinal surgery. They are resistant to intravenous antibiotic therapy alone and frequently require operative or CT-guided drainage. Diagnosis in the postoperative patient with other potential sources of infection is often extremely difficult. In order to better characterize and define IAA, we prospectively examined pus and peripheral blood of 15 consecutive patients presenting with IAA, and compared this to similar samples taken from 34 consecutive patients with soft tissue abscesses from other sites. Data regarding serum and pus lysozyme, complement fragments (iC3b, SC5b-9), II-10 and IL-4 are shown below (:~ +_ SEM):

Normal Intra-abdominal Abscess Soft Tissue Abscess Serum Range n = 15 n=34

Blood Blood Pus Blood Pus iC3b ug/ml 4.1_+0.3 156_+ 65 105_+ 37 271_+ 51" 54_+ 11 SC5b-9 ng/ml 195_+34 374_+663'~ 9207+~235 359+_!94 1876_+398 Lysozyme units 32_+ 2 40_+ 2",t, 206_+ 74 ¢~ 44+ 1" 658+- 8i [L-10 pg/ml 81-+ 31'~ f0+ 3 .... [L-4 pg/ml 30_+ 9 • I 42+ 14 * p <0.0l Compared to normal ~I, p <0.05 IAA compared to STA IL-10 levels in IAA pus were 7-fold higher than in soft tissue abscesses, whereas IL-4 was similar in both types of abscesses. The presence of these anti-inflammatory cytokines may indicate a T-Helper 2 lymphocyte response in the etiology of abscess formation and abscess persistence, and provide implications for prevention if not therapy. Elevated serum lysozyme levels in both types of abscess is an inexpensive and reliable aid in diagnosis.