Acute gallstone pancreatitis

Transcription

1 Surg Endosc (1999) 13: Springer-Verlag New York Inc Acute gallstone pancreatitis Timing of laparoscopic cholecystectomy in mild and severe disease W. Uhl, C. A. Müller, L. Krähenbühl, S.W. Schmid, St. Schölzel, M. W. Büchler Department of Visceral and Transplantation Surgery, University Hospital of Bern, CH-3010 Bern, Switzerland Received: 21 August 1998/Accepted: 12 June 1999 Abstract Background: In acute gallstone pancreatitis, the ideal point in time for laparoscopic cholecystectomy with special reference to the severity of the disease has been prospectively analyzed. Methods: A total of 77 patients with biliary acute pancreatitis were admitted between November 1993 and July 1998 (37 men and 40 women; mean age, 63 years; median Apache II score, 13.3) and staged by contrast-enhanced computed tomography findings as having edematous or necrotizing disease. Results: In 48 patients, laparoscopic cholecystectomy was found to be possible: 35 patients (73%) with mild and 13 patients (27%) with severe acute pancreatitis. The overall success rate was 79% (38 of 48 patients), with 85% (30 of 35 patients) and 62% (8 of 13 patients) having mild and severe disease, respectively. Median duration of time between onset of symptoms and surgery was 10 days (range, 4 19 days) in edematous and 14 days (range, 7 29 days) in necrotizing pancreatitis (p ). Operating time (median, 80 min) and hospital stay (median, 5 days) were almost the same in both groups. Total morbidity was 8%, with no mortality. Conclusions: Laparoscopic cholecystectomy with preoperative endoscopic common bile duct clearance is recommended as a treatment of choice for biliary acute pancreatitis. In mild disease, this is performed safely within 7 days, whereas in severe disease, especially in extended pancreatic necrosis, at least 3 weeks should elapse because of an increased infection risk. Key words: Acute pancreatitis Cholelithiasis Endoscopic retrograde cholangiography Laparoscopic cholecystectomy Surgery Correspondence to: M. W. Büchler Gallstones and alcohol represent the most common etiologic factors in acute pancreatitis, accounting for 70% to 95% of all patients with this disease [10, 50]. Other etiologic factors in acute pancreatitis (e.g., hyperlipidemia, post-ercp [endoscopic retrograde cholangiopancreatography), postoperative and drug-related acute pancreatitis) and idiopathic acute pancreatitis account for 5% to 15% of the cases. The causative pathophysiologic factor of biliary acute pancreatitis is a distal common channel of the biliary and pancreatic ducts, which can be found in up to 80% of gallstone-related acute pancreatitis [1, 23, 30]. That migration of gallstones through the biliary tract system is the underlying cause has been substantiated by the fact that in up to 85% of patients with biliary acute pancreatitis, gallstones can be found by screening of the feces [23]. This suggests that the pathophysiology relates to a temporary impaction of migrating stones at Vater s ampulla. As regards a surgical approach to biliary acute pancreatitis, the removal of the gallbladder was recommended during the same hospital stay because the recurrence rate of acute pancreatitis has been shown to be 29% to 63% if the patient is discharged from the hospital without additional treatment [13, 17, 19, 21, 36]. Therefore, conventional cholecystectomy has been the standard technique in the past after subsidence of the acute symptoms. In 1990, Dubois et al. [14] and Perissat and Vitale [34] introduced laparoscopic cholecystectomy, which quickly became the operation of choice for symptomatic cholelithiasis. The objective of this prospective study was to evaluate the role of laparoscopic cholecystectomy in biliary acute pancreatitis with reference to its feasibility and efficacy in mild and severe disease. Special attention was focused on the literature and history of surgical treatment in biliary acute pancreatitis. Materials and methods Acute pancreatitis was defined as an episode of acute upper abdominal pain in combination with elevated serum pancreatic enzymes (three times the upper normal level). Gallstone-related acute pancreatitis as the underlying etiologic factor was diagnosed by ultrasound and laboratory markers

2 1071 Fig. 1. Prospective algorithm in biliary acute pancreatitis. (ASAT, GT, AP, bilirubin). Alcohol was assumed to be the causative factor when a careful evaluation of the patient s history revealed a daily consumption of more than 40 g of alcohol and biliary tract stones were excluded by sonography. Acute pancreatitis induced by hyperlipidemia, drugs, endoscopic retrograde cholangiopancreatography (ERCP), or an operation and idiopathic acute pancreatitis were classified as other etiologic factors of the disease. Patients with biliary acute pancreatitis were treated according to the algorithm shown in Fig. 1. They were examined by an endoscopic retrograde cholangiography (ERC) as early as possible after hospitalization, and the pancreatic duct was not necessarily opacificied. The severity of the disease in all patients was assessed by contrast-enhanced computed tomography (CT) findings, which were used to stage the patients as having edematous (clinically mild) or necrotizing (severe) acute pancreatitis. Patients with edematous biliary acute pancreatitis were hospitalized on regular wards, whereas patients with the severe or necrotizing course of the disease were treated in the intensive care unit (ICU) with maximum conservative therapy measures and antibiotics over at least 14 days. Surgical management In mild acute pancreatitis, an elective laparoscopic approach for the removal of the stone-filled gallbladder after subsidence of the acute symptoms was the first surgical option, without revision of the pancreas after the fifth day, depending on the availability of operating room time. Patients with necrotizing acute pancreatitis and development of sepsis underwent fine-needle aspiration of pancreatic necrosis with Gram staining and culture of the aspirate [18]. If the aspirate was positive for germs, emergency operation was performed, with necrosectomy and postoperatively closed continuous lavage of the retroperitoneum. Patients with sterile necrotizing pancreatitis were treated by a laparoscopic approach after more than 7 days, depending on patient s clinical condition (response to ICU treatment and complete subsidence of the acute symptoms) and availability of operating room time. Statistics Medians and ranges were calculated after normal distribution was checked. The Mann-Whitney rank sum test for independent random samples and Fisher s exact test for qualitative variables were used. Two-tailed p values less than 0.05 were considered significant. Results From November 1993 to July patients with acute pancreatitis were admitted to the University Hospital of Bern. On the basis of the aforementioned definitions, 77 patients (45%) had biliary acute pancreatitis and 61 patients (35%) had alcohol-related disease. In 35 patients (20%), other causes of the disease were assumed because no association with gallstones or alcohol had been found in these patients (Fig. 2). In the group with acute gallstone pancreatitis (n 77), there were 37 men (48%) and 40 women (52%), with an average age of 63 years (range, years). Mean Apache II score was 13.3 (range, 1 28). Patient characteristics of mild and severe disease according to contrast-enhanced CT findings are given in Table 1. These findings show that 48 patients (62%) had edematous acute pancreatitis and 29 patients (38%) a necrotizing course of the disease. The CT scanning results, especially regarding the extent of pancreatic necrosis and the risk for development of infected pancreatic necrosis are given in Table 2. ERC/ERCP examination In 65 of 77 patients (88%) with biliary acute pancreatitis, ERC/ERCP examinations were performed within a median of 14 hours (range, 5 hours to 7 days) after admission to the hospital. In 48 of 65 patients (74%), stones in the common bile duct were found at the time of investigation, and in 47 patients (98%) the stones were successfully extracted after endoscopic papillotomy. One patient had too high a risk for bleeding caused by anticoagulation, so no papillotomy and clearance of the biliary tract were done. Additionally, the pancreatic duct was opacified in every second patient, showing duct disruption in one-third of the patients with acute necrotizing pancreatitis. Laparoscopic surgery During the same hospital stay, 60 of 77 patients (78%) with biliary acute pancreatitis underwent cholecystectomy. Seventeen patients (21%) were excluded from surgical intervention: 11 patients who had a history of cholecystectomy,

3 1072 Fig. 2. Etiologic factors in acute pancreatitis (n 173). Table 1. Characteristics of patients with acute biliary pancreatitis (n 77) Total (n 77) AIP (n 48) NP (n 29) Age (years) 63 (28 88) 61 (28 88) 65 (37 87) Women Men Apache II 13.3 (1 28) 11 (1 21) 15 (8 28) AIP, acute edematous pancreatitis; NP, acute necrotizing pancreatitis Table 2. Computed tomography (CT) findings and infection rate in patients with biliary acute pancreatitis (n 77) CT findings (n 77) (%) Infection rate of pancreatic necrosis (n 11) (%) Edema 27 (35) Edema with exudation 21 (27) Necrosis ( 30%) 9 (12) 0 (0) Necrosis ( 50%) 7 (9) 1/7 (14) Necrosis (>50%) 13 (17) 10/13 (77) including 4 who developed infected pancreatic necrosis; 4 patients at high cardiopulmonary risk; and 2 patients who were discharged to another hospital after recovery from the acute symptoms (Fig. 3). Furthermore, 12 of the remaining 60 surgically treated patients (20%) had to be excluded from a laparoscopic approach for the following reasons. Seven patients with necrotizing acute pancreatitis developed fineneedle aspiration-proven infected pancreatic necrosis in a mean of 27 days (range, days) after onset of disease. The CT scans in all these 7 patients showed pancreatic necrosis to an extent of 50% and more (Table 2). In these cases, necrosectomy, conventional cholecystectomy, and continuous closed lavage were performed. In five additional patients (four with edematous and one with necrotizing acute pancreatitis) a primary open approach was used. The one patient with necrotizing acute pancreatitis had a severe chronic obstructive pulmonary disease and thus was considered not suitable for a pneumoperitoneum. One patient with edematous acute pancreatitis also had complicated diverticulosis of the colon. For this patient, therefore, open cholecystectomy was combined with subtotal colectomy. Because three patients had undergone upper abdominal open surgery in the past (operation for an aneurysm of the arteria gastroduodenalis in one patient and recurrent gastrointestinal surgical interventions in two patients), severe adhesions were suspected and the decision for primary conventional cholecystectomy was made. However, in 48 of 60 patients (80%) with biliary acute pancreatitis, a laparoscopic management for gallbladder removal was found to be possible, including 35 of 48 patients (73%) with edematous and 13 of 48 patients (27%) with necrotizing acute pancreatitis. In 38 of 48 patients, the laparoscopic approach was performed successfully, giving an overall success rate of 79%. The success rate was higher in patients with mild disease (30/35; 85%) than in patients with severe acute pancreatitis (8/13; 62%), but the difference was not significant (p ). Conversion to open surgery was necessary in 10 patients (5 with edematous and 5 with necrotizing acute pancreatitis): 1 patient because of a Mirizzi syndrome type II, 6 patients because of severe acute cholecystitis and/or empyema, 2 patients because of severe inflammatory adhesions and difficulties in preparation of Calot s triangle, and 1 patient because of a large cystic liver lesion in segment 8 that needed to be treated conventionally by an open procedure during the same intervention. The time between onset of pain and the laparoscopic cholecystectomy in the group with edematous acute pancreatitis was a median of 10 days (range, 4 19 days), and in the group with necrotizing disease a median of 14 days (range, 7 29 days). Median operating times were 80 min (range, min) for mild disease and 90 min (range, min) for severe disease. Hospital stay after laparoscopic cholecystectomy was a median of 5 days (range, 2 9 days) for the patients with edematous acute pancreatitis, which was almost the same as the median of 6 days (range, 4 16 days) for the group of patients with necrotizing disease. The overall mortality was 0% in both groups, and postoperative complications occurred in 3 of 38 patients (7.8%). Postoperative morbidity was 3.3% (1 of 30 patients experienced subhepatic hematoma, which was managed conservatively) in the edematous group and 25% (2 of 8 patients) in the necrotizing group (Table 3). One patient with severe acute pancreatitis experienced fever and a septic state, caused by infected pancreatic necrosis and proved by fine-needle aspiration, 19 days after primary laparoscopic cholecystectomy. The initial CT scan in this case showed an extent of pancreatic necrosis greater than 50%, and laparoscopic removal of the gallbladder was performed 14 days after onset of symptoms. This patient underwent a second operation involving necrosectomy and lavage therapy and had an uneventful further course. The second patient with severe disease had a thrombosis of the portal vein. Discussion In 1901, Opie [30] described impacted gallstones at autopsy in Vater s ampulla of two patients who had severe acute pancreatitis. However, the association between gallstones and acute pancreatitis was first reported by Bernard and Prince in 1852 and 1882, respectively [25, 35]. The incidence of biliary acute pancreatitis in patients with symptomatic gallstones is about 3% to 8% [3, 27, 36]. As a pathophysiologic precondition in these patients, a common distal channel can be found assuming migration through the Vater s papilla, with transient or persistant im-

4 1073 Fig. 3. Flow chart of patients with acute biliary pancreatitis (n 77). Table 3. Results of laparoscopic approach in acute biliary pancreatitis (n 48) Total (n 48) AIP (n 35) NP (n 13) p value Time between onset of pain and surgery (days) 10 (4 29) 10 (4 19) 14 (7 29) P Hospitalization after LCE (days) 5 (2 21) 5 (2 9) 6 (4 21) NS Conversion 10/48 (21%) 5/35 (15%) 5/13 (38%) NS Success rate 38/48 (79%) 30/35 (86%) 8/13 (62%) NS Operation time (min) LCE 80 (30 160) 80 (30 120) 90 (60 160) NS Morbidity in LCE 3/38 (8%) 1/30 (3%) 2/8 (25%) NS Mortality NS LCE, laparoscopic cholecystectomy; AIP, acute edematous pancreatitis; NP, acute necrotizing pancreatitis; NS, not significant paction of the stone [1, 23, 30]. This has been confirmed by screening of the feces for gallstones, in which concrements were found in about 85% of the cases [1]. Regarding the cause and outcome of biliary acute pancreatitis, there are controversal findings in the literature. However, a prospective analysis in 190 patients with acute pancreatitis showed that no differences were caused by the underlying etiologic factor [50]. The policy of operating on patients with biliary acute pancreatitis is based on the high recurrence rate of acute pancreatitis if patients are discharged without further therapy [13, 17, 19, 21, 36]. Because the recurrence rate for acute pancreatitis was reported to be more than 30%, surgical intervention during the same hospital stay has been recommended. Before the laparoscopic approach came into use, a conventional cholecystectomy during the same hospital stay was recommended [31, 48]. From the 1960s to the 1980s, there was much discussion and controversy about early versus delayed surgical intervention in these patients, beginning with a recommendation for early operation, for which reduced mortality rates of 16% to 2% were shown in a retrospective study [2] and 6.9% to 2.8% in a prospective study [44]. Later trials, however, showed a beneficial effect if patients underwent operation only after the acute symptoms had subsided. There are two retrospective analyses of patients by Frei et al. [17] and Runkel et al. [40], and one prospective study published by Kelly et al. [24] in In this prospective investigation, including 165 patients with acute gallstone pancreatitis, both morbidity and mortality rates were reduced, favoring delayed surgical intervention (morbidity, 48% vs 11.3% and mortality 3.3% vs 0% in early versus delayed surgery, respectively). The first to recommend early endoscopic papillotomy in acute gallstone pancreatitis was Rosseland in 1984 [34]. Until then it had been considered dangerous to perform endoscopic procedures in patients with acute pancreatitis due to the risk of aggravating the disease. Rosseland rec-

5 1074 ommended this method as an effective procedure for drainage and relief of acute symptoms. However, the recurrence rate of acute pancreatitis after endoscopic retrograde cholangiography and papillotomy with stone removal is unknown. A first report showed that the recurrence rate of acute pancreatitis was reduced. The same was not the case with further inflammatory complications, which arise in association with biliary acute pancreatitis (such as cholecystitis and gallbladder empyema), and the policy to wait should be restricted only to patients who have a high perioperative risk [7]. Three randomized multicenter trials studying the role of urgent endoscopic retrograde cholangiopancreatography (ERCP) in biliary acute pancreatitis were published in recent years. Two [15, 28] showed a positive effect, but the last trial [16] showed no influence on the course and outcome of the disease in nonjaundiced patients. Although the final answer is open, early ERC is recommended in biliary acute pancreatitis with cholangitis and cholestasis indicating impacted stones [11, 29, 42, 43]. Common bile duct stones were found to be present in 40% to 59% of the patients in these three studies, and the success rate for the stone removal was 86% to 100%. In the current study, ERC examinations were performed within a median of 14 h after hospitalization, and in 48 of 65 patients (74%) biliary tract stones were found. These stones were successfully extracted after endoscopic papillotomy in 47 of 48 patients (98%). In the early 1990s, laparoscopic cholecystectomy was introduced by Dubois et al. [14] and Perrisat and Vitale [34]. It rapidly became the method of choice for elective cholecystectomy. In acute cholecystitis, laparoscopic cholecystectomy is recommended within 24 to 48 h after the onset of this disease as a fast and safe method [20], but in biliary acute pancreatitis, laparoscopic cholecystectomy should be performed, according to the NIH consensus conference, 5 to 6 days after the onset of the acute pancreatitis attack [33]. In this general recommendation, however, no attention was paid to the severity of the disease. The rationale to wait in acute pancreatitis is underscored by the fact that a necrotizing course of the disease, which indicates intensive care therapy, is established within 4 days [22]. There must be a wait during this time before a surgical approach to mild or edematous acute pancreatitis is taken. In the current study, a success rate of 86% was achieved in mild disease with no major postoperative complications. The role of surgery in acute necrotizing pancreatitis has changed tremendously to a more conservative approach in recent years [26, 32, 41, 49]. Currently, the only clear indication for surgical intervention in severe disease is proof of infected pancreatic necrosis and, rarely, nonresponse to intensive care therapy [49]. Therefore, most patients with necrotizing acute pancreatitis can be treated conservatively with suitable antibiotics that can penetrate into the pancreas [8] and cover the germs, which usually are found to reduce septic complications in infected pancreatic necrosis [5, 6, 18, 26, 32, 41]. Patients with sterile pancreatic necrosis also may be managed with laparoscopic cholecystectomy, although there may be some problems intraoperatively with the Calot s triangle because of the inflammatory process. The current study revealed a success rate of 62% in these cases, with no prolongation of operating time or postoperative hospitalization. However, it seems reasonable to wait more than 7 days to treat patients with necrotizing acute pancreatitis because there is an increased risk of late infection of pancreatic necrosis, which was shown to depend on the extent of pancreatic necrosis. Especially in extended necrosis involving more than 50% of the pancreas, at least 3 weeks should elapse before laparoscopic cholecystectomy is performed. In patients with infected pancreatic necrosis, conventional removal of the gallbladder together with necrosectomy and lavage therapy is the method of choice. One recently published study showed that laparoscopic cholecystectomy can be postponed after debridement [51]. However, this second operation should be avoided by combining it with the first. The split combined approach that uses early endoscopic common bile duct clearance via ERC and papillotomy before the performance of laparoscopic cholecystectomy in patients with biliary acute pancreatitis has been studied and recommended by several investigators [12, 37, 38, 43, 45, 47] (Table 4). However, it is also possible for experienced laparoscopic surgeons to examine the biliary tract with an intraoperative cholangiogram [4, 9, 12, 38, 46]. This approach requires more surgical skills because if a common bile duct stone is found or suspected, a laparoscopic revision of the ductus choledochus must be performed, perhaps by an open procedure with the risk of increasing morbidity. Therefore, it is our strategy to perform common bile duct examination and its clearance before the laparoscopic cholecystectomy. In the literature, the reported overall success rate of this laparoscopic approach is between 79% and 100%, with a conversion rate between 0% and 16%, depending on the severity of the condition in the patients included in the respective studies. The time interval until operation differed remarkably among the studies, with a mean of 3.4 days if calculations are based on the date of hospitalization, and 10 days if the onset of symptoms is taken into account. The latter should be used in the future because the onset of symptoms may be assessed exactly, whereas a variable time interval usually elapses before patients are hospitalized. Operating time and postoperative hospital stay were comparable with those in the current study. In conclusion, the recommended principal step in managing biliary acute pancreatitis is the clearance of stones from the common bile duct by ERC and papillotomy as soon as possible. In mild or edematous pancreatitis, 5 to 7 days after the onset of symptoms, laparoscopic cholecystectomy is the surgical procedure of choice for preventing recurrence of acute pancreatitis or other infectious complications during the same hospital stay. Patients with severe or necrotizing acute pancreatitis should be treated in the ICU with maximum intensive care measures and antibiotics. The only indication for surgery in severe acute gallstone pancreatitis is sepsis caused by infected pancreatic necrosis, which can be proved easily by imaging-guided fine-needle puncture and analysis of the aspirate for Gram staining and culture. The operative method is necrosectomy together with cholecystectomy using closed continuous lavage of the retroperitoneum in the postoperative period. In patients improving from necrotizing acute biliary pancreatitis and sterile pancreatic necrosis,

6 1075 Table 4. Literature research: results of laparoscopic treatment in biliary acute pancreatitis Author Study design Patient no. Success rate (%) Conversion rate (%) Time interval until operation (days) Operating time (min) Postoperative hospital stay (days) Morbidity (%) Mortality (%) Common bile duct examination Rhodes et al. [37] p (4 34) a 50 (30 120) a 2 (1 3) a preop ERC a 1IOC a Tate et al. [47] p (3 24) a 76 (NA) b 2 (1 12) preop ERC Canal et al. [9] r (1 16) b NA 3.3 (2 16) b preop ERC 14 IOC Soper et al. [43] r (±0.6) b 110 (70 240) b 2.3 (1 10) b preop ERC De Iorio et al. [12] r NA NA NA preop ERC 31 IOC Tang et al. [46] r (±2.0) b NA NA NA 25 preop ERC 104 IOC Sungler et al. [45] p NA NA 14.3 (8 30) b preop ERC Ballestra-Lopez et al. [4] p /15 b,c 86 (45 210) b 8.4 (NA) b preop ERC 40 IOC Ricci et al. [38] p NA NA 14.9 (7 37) preop ERC 47 IOC Current study p (4 29) a 80 (30 225) a 5 (2 21) a preop ERC a Median (range) b Mean (range or SD) c Mild/severe disease p, prospective; r, retrospective; NA, not assessed; preop, preoperative; ERC, endoscopic retrograde cholangiography; IOC, intraoperative cholangiogram we also recommend a laparoscopic approach for the removal of the gallbladder. However, this surgical procedure should be performed more than 7 days after the onset of disease. Especially in extended pancreatic necrosis, there should be a wait of at least 3 weeks because there is an increased risk of infection. Acknowledgment. We are most grateful to Ms. Sabine Hauri for assisting us in editing this manuscript. References 1. Acosta JM, Pellegrini CA, Skinner DB (1980) Etiology and pathogenesis of acute biliary pancreatitis. Surgery 88: Acosta JM, Rossi R, Galli OM, Pellegrini CA, Skinner DB (1978) Early surgery for acute gallstone pancreatitis: evaluation of a systematic approach. Surgery 83: Armstrong CP, Taylor TV, Torrance HB (1985) Effects of bile, infection, and pressure on pancreatic duct integrity. Br J Surg 72: Ballestra-Lopez C, Bastida-Vila X, Bettonica-Larranaga C, Zaraca F, Catarci M (1997) Laparoscopic management of acute biliary pancreatitis. Surg Endosc 11: Bassi C, Falconi M, Girelli R, Nifosi F, Elio A, Martini N, Pederzoli P (1989) Microbiological findings in severe acute pancreatitis. Surg Res Commun 5: Beger HG, Bittner R, Block S, Buchler M (1986) Bacterial contamination of pancreatic necrosis: a prospective clinical study. Gastroenterology 91: Billi P, Pezzilli R, D Imperio M, Baroncini D, Barakat B, Piemontese A, Dal Monte PR, Miglioli M (1995) Endoscopic sphincteroromy for the prevention of further attacks of acute pancreatitis in patients with gallbladder in situ. Digestion 56: A Büchler M, Malfertheiner P, Friess H, Isenmann R, Vanek E, Grimm H, Schlegel P, Friess T, Beger HG (1992) Human pancreatic tissue concentration of bactericidal antibiotics. Gastroenterology 103: Canal DF, Broadie TA (1994) Results of laparoscopic cholecystectomy for the treatment of gallstone pancreatitis. Am Surg 60: , discussion Carter DC (1993) Special aspects of gallstone pancreatitis. In: Trede M, Carter DC (eds). Surgery of the pancreas, Churchill Livingstone, Edinburgh, London, Madrid, Melbourne, New York, Tokyo, pp Cooperman AM, Siegel J, Neff R, Reddy S, Hammerman H (1991) Gallstone pancreatitis: combined endoscopic and laparoscopic approaches. J Laparoendosc Surg 1: De Iorio AV, Jr, Vitale GC, Reynolds M, Larson GM (1995) Acute biliary pancreatitis: the roles of laparoscopic cholecystectomy and endoscopic retrograde cholangiopancreatography. Surg Endosc 9: Dixon JA, Hillam JD (1970) Surgical treatment of biliary tract disease associated with acute pancreatitis. Am J Surg 120: Dubois F, Icard P, Berthelot G, Levard H (1990) Coelioscopic cholecystectomy: preliminary report of 36 cases. Ann Surg 211: Fan ST, Lai EC, Mok FP, Lo CM, Zheng SS, Wong J (1993) Early treatment of acute biliary pancreatitis by endoscopic papillotomy. N Engl J Med 328: Fölsch UR, Nitsche R, Lüdtke R, Hilger RA, Creutzfeld W (1995) Controlled randomized trial of urgent endoscopic retrograde cholangiopancreatography (ERCP)/papillotomy (EPT) for acute biliary pancreatitis. Digestion 56: A Frei GJ, Frei VT, Thirlby RC, McClelland RN (1986) Biliary pancreatitis: clinical presentation and surgical management. Am J Surg 151: Gerzof SG, Banks PA, Robbins AH, Johnson WC, Spechler SJ, Wetzner SM, Snider JM, Langevin RE, Jay ME (1987) Early diagnosis of pancreatic infection by computed tomography-guided aspiration. Gastroenterology 93: Glenn F, Frey C (1964) Reevaluation of the treatment of pancreatitis associated with biliary tract disease. Ann Surg 160: Hawasli A (1994) Timing of laparoscopic cholecystectomy in acute cholecystitis. J Laparoendosc Surg 4: Howard JM, Ehrlich EW (1980) Gallstone pancreatitis: a clinical entity. Surgery 51: Isenmann R, Buchler M, Uhl W, Malfertheiner P, Martini M, Beger HG (1993) Pancreatic necrosis: an early finding in severe acute pancreatitis. Pancreas 8: Kelly TR (1976) Gallstone pancreatitis: pathophysiology. Surgery 80: Kelly TR, Wagner DS (1988) Gallstone pancreatitis: a prospective randomized trial of the timing of surgery. Surgery 104: Kircher R, Lausen M, Salm R, Schölmerich J (1990) Acute gallstone pancreatitis. Dig Surg 7: Luiten EJT, Hop WCJ, Lange JF, Bruining HA (1995) Controlled

7 1076 clinical trial of selective decontamination of severe acute pancreatitis. Ann Surg 222: Moreau JA, Zinsmeister AR, Melton LJd, Di Magno EP (1988) Gallstone pancreatitis and the effect of cholecystectomy: a populationbased cohort study. Mayo Clin Proc 63: Neoptolemos JP, Carr-Locke DL, London NJ, Bailey IA, James D, Fossard DP (1988) Controlled trial of urgent endoscopic retrograde cholangiopancreatography and endoscopic sphincterotomy versus conservative treatment for acute pancreatitis due to gallstones. Lancet 2: Neoptolemos JP, London NJ, Carr-Locke DL (1993) Assessment of main pancreatic duct integrity by endoscopic retrograde pancreatography in patients with acute pancreatitis. Br J Surg 80: Opie EL (1901) The relation of cholelithiasis to disease of the pancreas and to fat necrosis. Bull Johns Hopkins Hosp 12: Osborne DH, Imrie CW, Carter DC (1981) Biliary surgery in the same admission for gallstone-associated acute pancreatitis. Br J Surg 68: Pederzoli P, Bassi C, Vesentini S, Campedelli A (1993) A randomized multicenter clinical trial of antibiotic prophylaxis of septic complications in acute necrotizing pancreatitis with imipenem. Surg Gynecol Obstet 176: Pellegrini CA (1993) Surgery for gallstone pancreatitis. Am J Surg 165: Perissat J, Vitale GC (1991) Laparoscopic cholecystectomy: gateway to the future. Am J Surg 161: 408 [editorial] 35. Prince M (1882) Pancreatic apoplexy with a report of two cases. Boston Med Surg J 107: Ranson JH (1979) The timing of biliary surgery in acute pancreatitis. Ann Surg 189: Rhodes M, Armstrong CP, Longstaff A, Cawthorn S (1993) Laparoscopic cholecystectomy with endoscopic retrograde cholangiopancreatography for acute gallstone pancreatitis. Br J Surg 80: Ricci F, Castaldini G, de Manzoni G, Borzellino G, Rodella L, Kind R (1997) Minimally invasive treatment of acute biliary pancreatitis. Surg Enosc 11: Rosseland AR, Solhaug JH (1984) Early or delayed endoscopic papillotomy (EPT) in gallstone pancreatitis. Ann Surg 199: Runkel NS, Buhr HJ, Herfarth C (1996) Outcome after surgery for biliary pancreatitis. Eur J Surg 162: Sainio V, Kemppainen E, Poulokainen P, Taavitsainen M, Kivisaari L, Valtonen V. Haapiainen R, Schröder T, Kivilaakso E (1995) Early antibiotic treatment in acute necrotising pancreatitis. Lancet 346: Scheurer U (1994) Acute pancreatitis: ERCP and papillotomy. Dig Surg 11: Soper NJ, Brunt LM, Callery MP, Edmundowicz SA, Aliperti G (1994) Role of laparoscopic cholecystectomy in the management of acute gallstome pancreatitis. Am J Surg 167: 42 50; discussion Stone HH, Fabian TC, Dunlop WE (1981) Gallstone pancreatitis: biliary tract pathology in relation to time of operation. Ann Surg 194: Sungler P, Holzinger J, Waclawiczek HW, Heinerman PM, Boekel O (1997) Urgent ERCP and early-elective laparoscopic cholecystectomy in gallstone pancreatitis. Zentralbl Chir 122: Tang E, Stain SC, Tang G, Froes E, Berne TV (1995) Timing of laparoscopic surgery in gallstone pancreatitis. Arch Surg 130: ; discussion Tate JJT, Lau WY, Li AKC (1994) Laparoscopic cholecystectomy for biliary pancreatitis. Br J Surg 81: Tondelli P, Stutz K, Harder F, Schuppisser JP, Allgower M (1982) Acute gallstone pancreatitis: best timing for biliary surgery. Br J Surg 69: Uhl W. Büchler MW (1996) Approach to the management of necrotizing pancreatitis. Probl Gen Surg 13: Uhl W, Isenmann R, Curti G, Friess H, Vogel R, Büchler MW (1996) Influence of the etiology on the course and outcome of acute pancreatitis. Pancreas 13: Vezakis A, Dexter SPL, Martin IG, Larvin M, Mc Mahon MJM (1998) Laparoscopic cholecystectomy after pancreatic debridement. Surg Endosc 12: