Optimal timing of elective laparoscopic cholecystectomy after acute cholangitis and subsequent clearance of choledocholithiasis

Transcription

1 The American Journal of Surgery (2010) 200, Clinical Science Optimal timing of elective laparoscopic cholecystectomy after acute cholangitis and subsequent clearance of choledocholithiasis Vicky Ka Ming Li, M.B.B.S., F.R.C.S. (Edin.),* Jonathan Lau Kai Yum, M.B.B.S., M.R.C.S. (Edin.), Yuk Pang Yeung, M.B.B.S., F.R.C.S. (Edin.) Department of Surgery, Kwong Wah Hospital, Hong Kong, China KEYWORDS: Timing; Acute cholangitis; Laparoscopic cholecystectomy; Endoscopic sphincterotomy Abstract BACKGROUND: Elective laparoscopic cholecystectomy is recommended after endoscopic clearance of choledocholithiasis for patients with acute cholangitis, according to Toyko guidelines. However, the optimal timing remains uncertain. METHODS: Perioperative outcomes were retrospectively reviewed and compared between patients with early ( 6 weeks) and late ( 6 weeks) surgeries, while risk factors for postoperative complications were assessed using multivariate analysis. RESULTS: One hundred twelve patients (mean age, 64 years; range, years) were analyzed. Rate of conversion and intraoperative and postoperative complications (classified per Dindo et al) were 21.4% (24 of 112), 23.2% (26 of 112), and 34.8% (39 of 112), respectively. The late group had significantly more intraoperative (28.8% vs 9.4%, P.029) and postoperative (42.5% vs 15.6%, P.007) complications compared with the early group. Multivariate analysis showed both late (95% confidence interval, ; P.008) and a history of endoscopic sphincterotomy (95% confidence interval, ; P.038) to be independent risk factors for postoperative complications. CONCLUSIONS: Patients with endoscopic clearance of choledocholithiasis, especially after endoscopic sphincterotomy, should receive elective laparoscopic cholecystectomy within 6 weeks after a cholangitic attack Elsevier Inc. All rights reserved. Patients suspicious for acute cholangitis should initially be treated conservatively with antibiotics, while those with severe cholangitis required immediate endoscopic, percutaneous, or surgical decompression of the biliary system. Following resolution of the acute episode and confirmed * Corresponding author: Tel.: ; fax: address: he3979hk@yahoo.com.hk Manuscript received June 24, 2009; revised manuscript November 17, 2009 successful clearance of choledocholithiasis, elective laparoscopic cholecystectomy is recommended for those who are medically fit, according to Tokyo guidelines (grade B recommendation). 1 However, the optimal timing for cholecystectomy after an acute cholangitic attack remains uncertain. The potential benefit of early laparoscopic cholecystectomy lies in the possible prevention of recurrent biliary complications while waiting for the operation. A study of patients with symptomatic choledocholithiasis and subsequent endoscopic sphincterotomy showed a rate of recurrent /$ - see front matter 2010 Elsevier Inc. All rights reserved. doi: /j.amjsurg

2 484 The American Journal of Surgery, Vol 200, No 4, October 2010 biliary complications of 20% after a median waiting time of 7 weeks, and those who developed recurrence had worse postoperative outcomes than those who did not. 2 However, another more focused study solely on patients with histories of acute cholangitis revealed a rate of 2.4% of recurrent biliary events during a waiting period of 6 to 12 weeks for the operation. 3 Worry regarding early laparoscopic cholecystectomy stems from the incompletely resolved inflammation that may result in a difficult operation and hence potential morbidities. Studies of laparoscopic cholecystectomy soon after acute pancreatitis subsides have shown high conversion rates of 12% to 24%. 4 7 British guidelines suggest a delay of cholecystectomy in case of severe biliary pancreatitis. 8 The effect of inflammation after acute cholangitis is uncertain. In 2 retrospective studies including only patients with acute cholangitis, there was great variation regarding the interval between endoscopic sphincterotomy and elective laparoscopic cholecystectomy, ranging from within 1 to 12 weeks. 3,9 Nonetheless, both reports showed similar conversions and morbidities, and there is no consensus on the optimal timing of cholecystectomy after clearance of choledocholithiasis. Therefore, in an attempt to determine optimal timing, we performed a comparative study of the perioperative outcomes of patients with early versus late laparoscopic cholecystectomy, concentrating only on patients with histories of acute cholangitis. Methods Study design A retrospective review of a prospectively collected database and clinical records in a single general surgical unit was performed for patients who underwent elective laparoscopic cholecystectomy between January 2002 and June In our department, laparoscopic cholecystectomy was attempted for all patients who were indicated for cholecystectomy. Only patients with a diagnosis of acute cholangitis as defined by Tokyo guidelines 10 and subsequent confirmed clearance of choledocholithiasis either by endoscopic retrograde cholangiopancreatography (ERCP) or magnetic resonance cholangiopancreatography were included for analysis. Patients were excluded if they had (1) acute pancreatitis, (2) concomitant or histories of acute cholecystitis, (3) liver abscesses, (4) doubt concerning ductal clearance with preoperative decision for laparoscopic exploration of the common bile duct (CBD) or operative cholangiography, (5) recurrent pancreatobiliary sepsis requiring emergency open or emergency while waiting for elective laparoscopic cholecystectomy, (6) recurrent pyogenic cholangitis, or (7) pancreatobiliary malignancies. All patients with acute cholangitis were initially treated conservatively with medical treatment which included nil per os, intravenous fluid, and antibiotics. Those who did not respond to initial medical treatment or those with clinical deterioration were considered to have severe cholangitis, and emergency decompression using endoscopic, percutaneous, or operative approach was arranged as appropriate. Repeated ERCP for common duct stone removal was arranged 2 weeks after the index admission. Those who responded to medical treatment underwent ERCP for the identification and removal of any common duct stones during the index admission. Data including age, gender, previous upper abdominal, other abdominal, recurrent acute or severe cholangitic attack, and the presence of endoscopic sphincterotomy, together with important preoperative co-morbidities, including diabetes mellitus, ischemic heart disease, cirrhosis, and smoking habit, were recorded. Severe cholangitis was defined if emergency biliary drainage procedure was required at the time of index admission. The time interval between the first day of symptoms of the latest cholangitic attack and the date of elective laparoscopic cholecystectomy was calculated and expressed in terms of weeks. All operations were performed by specialists or residents under supervision by specialists, using the standard 4-port or 3-port techniques. All patients received a single dose of prophylactic antibiotics on induction. Our primary outcome measures were intraoperative complications and postoperative complications, and the secondary outcome measures were the rate of conversion, operative time, and the length of hospital stay. Postoperative complications were recorded according to the classification recommended by Dindo et al. 11 These outcomes were compared between patients with laparoscopic cholecystectomy performed 6 weeks (early group) and 6 weeks (late group) from the point of the latest cholangitic attack. Risk factors, including age, gender, smoking habit, presence of diabetes mellitus, ischemic heart disease, cirrhosis, American Society of Anesthesiologists score, previous upper abdominal, other abdominal, previous acute cholangitic attacks, previous severe cholangitic attack, endoscopic sphincterotomy for stone retrieval, and timing of cholecystectomy (6 weeks after last cholangitic attack), were examined with univariate and multivariate analyses to identify risk factors for postoperative complications. Patients with signs of postoperative sepsis requiring antibiotic treatment as part of treatment were considered to have septic complications for subsequent analysis (Table 1). Statistical analysis Data were analyzed using SPSS version 15.0 (SPSS, Inc, Chicago, IL). Univariate analysis was performed using t tests for continuous variable and 2 tests or Fisher s exact tests for nominal variables as appropriate. Multivariate analysis was performed using logistic regression analysis (stepwise forward Wald). Those variables with P values.20 in

3 V.K.M. Li et al. Timing of elective laparoscopic cholecystectomy and cholangitis 485 Table 1 Comparison of complications of early versus late groups Outcome univariate analysis were selected for multivariate analysis. A P value.05 on multivariate study was considered significant. Results Early (n 32) Late (n 80) P Intraoperative complications 3 (9.4%) 23 (28.8%).029 All postoperative complications 5 (15.6%) 34 (42.5%).007 Grade I 4 (12.5%) 15 (18.7%) Grade II 1 (3.1%) 16 (20.0%) Grade III 0 3 (3.8%) Grade IV 0 0 Grade V 0 0 All postoperative septic complications* 4 (12.5%) 20 (25.0%).203 *Refer to Table 4 for septic complications. Table 2 Reasons for all conversions during laparoscopic cholecystectomy (n 112) Reason n (%) Unclear anatomy 16 (66.7%) Intraoperative complication 4 (16.7%) Slow progress 3 (12.5%) Bleeding 1 (4.2%) Total 24 (21.4%) Table 3 Nature of all intraoperative complications (n 112) Nature n (%) Gall bladder perforation 20 (77.0%) Stone spillage 2 (7.7%) Tear at junction between cystic duct and CBD 2 (7.7%) CBD transaction 1 (3.8%) Serosal tear of transverse colon 1 (3.8%) Total 26 (23.2%) Between January 2002 and June 2008, 112 patients fit the inclusion and exclusion criteria. The mean age was 64 years (range, years). Thirty-three patients (29.5%) had 2 medical comorbidities. The median waiting time for operation was 10.5 weeks (range, weeks). Twenty-four patients (21.4%) had the operation converted to open, and the median operative time was 90 minutes (range, minutes). The rates of intraoperative and postoperative complications were 23.2% (26 of 112) and 34.8% (39 of 112), respectively. The median hospital stay was 4 days (range, 2 25 days). Details of reasons for all conversions and intraoperative and postoperative complications are summarized in Tables 2 to 4. Only 1 patient required conversion for gallbladder perforation, while the 2 patients who had stone spillage were managed laparoscopically. One patient had a tear at the cystic duct near the T-junction and required conversion for proper cystic duct stump ligation, while the other required hepatojejunostomy after conversion because of perforation at the CBD. A serosal tear at transverse colon was noted in another patient during mobilization of the gallbladder. The operation was converted to open for repair, and the patient recovered uneventfully. One patient had common bile transaction, identified only after, further details of which are discussed below. Three patients (2.6%) had grade III postoperative complications (Table 4). All had operations performed 6 weeks after the last cholangitic attack. The first patient was an 82-year-old woman whose operation was converted for unclear anatomy with suspected Mirizzi s syndrome. She developed subhepatic collection, demonstrated on computed tomography on day 6 for persistent fever. She was treated with ultrasound-guided percutaneous drainage together with antibiotics and was discharged on day 15. The second patient was a 50-year-old woman who also had converted to open for unclear anatomy. She developed jaundice after, and subsequent ERCP revealed abrupt contrast cutoff at the lower CBD, and CBD transaction was diagnosed. Emergency laparotomy and hepaticojejunostomy was performed. She recovered uneventfully and was discharged on day 21 after the second. The last patient was a 63-year-old man who developed signs of peritonitis on day 2, and emergency laparotomy revealed leakage of bile from minor ducts over the gallbladder fossa that was controlled with ligature plication. He was then discharged on day 10. Table 4 Details of all postoperative complications according to classification of Dindo et al 11 (n 112) Grade Nature n (%) I Atelectasis requiring physiotherapy 6 (15.4%) II Fever requiring antibiotics* 10 (25.6%) Nausea requiring antiemetics 9 (23.1%) Chest infection* 5 (12.8%) Wound infection* 3 (7.7%) Urinary tract infection* 3 (7.7%) IIIa Collection on computed tomography* 1 (2.6%) IIIb CBD transaction* 1 (2.6%) Bile leakage from minor ducts in 1 (2.6%) liver bed* IV V Total 39 (34.8%) *Considered septic complications.

4 486 The American Journal of Surgery, Vol 200, No 4, October 2010 Table 5 Comparison of baseline characteristics of early ( 6 weeks) and late ( 6 weeks) groups Characteristic Early (n 32) Late (n 80) P Mean age (y) Men Diabetes mellitus Ischemic heart disease Cirrhosis Smokers American Society of Anesthesiologists score Upper abdominal operations Recurrent acute cholangitis Total number of cholangitis attacks Severe cholangitis Endoscopic sphincterotomy Table 5 shows the results of comparisons of baseline characteristics between patients in the early and late groups. History of recurrent cholangitic attack presented in 20 patients (17.9%), while 19 (17%) had severe cholangitic attacks. Endoscopic sphincterotomy was required in 82 patients (73%) for CBD stone clearance. Three patients had only magnetic resonance cholangiopancreatography instead for CBD stone exclusion. When the whole group of patients was included for analysis, patients in the late group had significantly more intraoperative (27.5% vs 9.4%, P.045) and postoperative (42.5% vs, 15.6% P.007) complications compared with those in the early group. However, despite a 2-fold increase in postoperative septic complications in the late group, it did not reach statistical significance. There were also no significant differences in rate of conversion, operative time, and hospital stay (Table 6). Multivariate logistic regression showed that both late (B 1.445; SE.546; odds ratio, 7.117; 95% confidence interval, ; P.008) and history of Table 7 Univariate analysis of risk factors for all postoperative complications Risk factor Postoperative complications Yes Gender.736 Male 19 (48.7%) 38 (52.1%) Female 20 (51.3%) 35 (47.9%) Mean age (y) Diabetes mellitus.324 Yes 6 (15.4%) 17 (23.3%) No 33 (84.6%) 56 (76.7%) Ischemic heart disease.303 Yes 2 (5.1%) 8 (11.0%) No 37 (94.9%) 65 (89.0%) Cirrhosis.463 Yes 0 (0%) 1 (1.4%) No 39 (100%) 72 (98.6%) Smoking.223 Yes 2 (5.1%) 9 (12.3%) No 37 (94.9%) 64 (87.7%) ASA (20.5%) 21 (28.8%) 2 25 (64.1%) 40 (54.8%) 3 6 (15.4%) 12 (16.4%) Previous upper abdominal operations.297 Yes 1 (2.5%) 2 (2.7%) No 39 (97.5%) 72 (97.3%) Other abdominal operations.106 Yes 8 (20.5%) 7 (9.6%) No 31 (79.5%) 66 (90.4%) Previous acute cholangitis.839 Yes 7 (17.9%) 12 (16.4%) No 32 (82.1%) 61 (83.6%) Previous severe cholangitis.745 Yes 6 (15.4%) 13 (17.8%) No 33 (84.6%) 60 (82.2%) Endoscopic sphincterotomy.046 Yes 33 (84.6%) 49 (67.1%) No 6 (15.4%) 24 (32.9%) Operation after 6 wk.007 Yes 34 (87.2%) 46 (63.0%) No 5 (12.8%) 27 (37.0%) endoscopic sphincterotomy (B 1.085; SE.523; odds ratio, 4.302; 95% confidence interval, ; P.038) were independent significant risk factors for postoperative complications (Tables 7 and 8). No P Table 6 Comparison of operative outcomes of early versus late groups Outcome Early (n 32) Late (n 80) P Conversion 7 (21.9%) 17 (21.3%).942 Mean operative time (min) Mean hospital stay (d) Table 8 Risk factor Multivariate analysis Odds ratio 95% confidence interval Operation after 6 wk Endoscopic sphincterotomy P

5 V.K.M. Li et al. Timing of elective laparoscopic cholecystectomy and cholangitis 487 Comments The role of cholecystectomy after endoscopic sphincterotomy for choledocholithiasis has been debated for years, and a recent Cochrane review has further stressed its importance in preventing recurrent biliary events. 12 Consensus for managing patients with acute cholangitis and subsequent clearance of choledocholithiasis was reached in 2007 that elective laparoscopic cholecystectomy is recommended for those who are medically fit (grade B recommendation). 1 However, the optimal timing of is unclear. The timing of laparoscopic cholecystectomy after endoscopic sphincterotomy for patients with different clinical presentations varies from 1 day to 12 weeks. 2,3,9,13 17 However, residual local inflammation in patients after cholangitic attacks may negatively affect operative outcomes. A recent study has shown the presence of inflammation as an independent risk factor for bile duct injury during laparoscopic cholecystectomy. 18 A study by Poon et al, 3 which focused on only 82 patients with previous cholangitic attacks, delayed the operation from 6 to 12 weeks after ductal clearance. Therefore, to examine the effect of timing on operative outcomes, we compared the perioperative outcomes between patients with laparoscopic cholecystectomy performed within and after 6 weeks from the latest cholangitic attack. In our study design, we used the time interval between the first day of the latest cholangitic attack and laparoscopic cholecystectomy, rather than that between the point of confirmation of ductal clearance or endoscopic sphincterotomy and the operation, because it is practically difficult to confirm a definite date of complete ductal clearance, because although some patients might already have spontaneous stone passage before ERCP, other patients after first therapeutic ERCP might have all stones passed well before second ERCP intended to confirm clearance. The present study showed a higher rate of intraoperative and postoperative complications if laparoscopic cholecystectomy was delayed 6 weeks since the last cholangitic attack. Late was an independent risk factor for postoperative complications. All 3 patients with serious postoperative complications (grade 3) had operations 6 weeks after cholangitic attack. Endoscopic sphincterotomy, which was performed in 74% of our patients, was another independent risk factor for postoperative complications. The rate of endoscopic sphincterotomy in the current study was similar to those reported in other studies with similar patient populations. 3,9,19,20 Bactobilia can develop and persist after endoscopic sphincterotomy, and this may explain why patients had more postoperative complications. A study by Sugiyama and Atomi, 21 who collected bile for bacterial culture at intervals after endoscopic sphincterotomy, showed rates of positive culture of 80%, 67%, and 60% at intervals of 7 days, 1 year, and 5 years, respectively. Another study by Mandryka et al 22 showed 100% positive bacterial culture at 12 to 36 months after endoscopic sphincterotomy. Persistent bactobilia may first result in persistent or even increasing inflammation along the biliary tree. 23 Longer waiting times may be associated with more subclinical cholangitic attacks, and recurrent biliary pain may be the only symptom, 12 during which patients may not seek medical attention. Inflammation may hence become chronic and have a negative impact on surgical outcomes. 18 Second, manipulation around the T-junction during cholecystectomy may result in transient cholangiovenous reflux, and the resultant bacteremia may manifest itself as nausea or nonspecific fever and even lead to septic complications. The 2-fold increase in septic complications in the late group, though it did not reach statistical significance, may give clues to our hypothesis. Therefore, we suggest that after ductal clearance, elective laparoscopic cholecystectomy should be performed 6 weeks from the last cholangitic attack, in particular in those with endoscopic sphincterotomy. Further prospective studies may necessary to further support or refute our recommendation. One criticism of the present study was our high postoperative complication rate of 34.8%. Reports on operative outcomes focusing only on patients with histories of cholangitis secondary to choledocholithiasis are scarce. The study by Poon et al 3 showed a morbidity rate of 3.6%, which was lower than that of our series. One possible explanation is that Poon et al s institution is a tertiary referral center, and their patients were a highly selected subgroup, because half of their initial target patients either refused or were considered not fit for. A study by Sarli et al 9 also showed a lower overall postoperative complication rate of 7.4%. Yet when comparing the rate of our complications that required invasive intervention, ours was similar (2.5% vs 2.7%). We also believe that our high morbidity rate was related to the fact that we adopted the classification advocated by Dindo et al, 11 which includes minor complications that might not have been taken into account in previous research. Our comprehensive account of these complications may truly reflect the real figure of morbidity and can be better generalized for future comparison and application. Other outcome measures of our study were comparable with those of others. Our rate of intraoperative complications was 22.3%. A national study conducted in 82 Swiss hospitals, including 10,174 patients with laparoscopic cholecystectomy for all indications, showed an intraoperative complications rate of 34.4%, 24 which was even higher than ours. Our overall conversion rate was higher (9.8% vs 21.4%) than that of Poon et al 3 but similar to that of Lau et al, 13 in whose study only 40% of patients had histories of cholangitis (19.5% vs 21.4%). Drawbacks of this study include its retrospective design, with the inherent problem of incomplete event documentation, and the fact that operations were performed by different surgeons in a general surgical unit. In addition, patients defaulted, and those not undergoing laparoscopic cholecystectomy because of poor risk could not be included for analysis, because of study design limitations. Last, from a

6 488 The American Journal of Surgery, Vol 200, No 4, October 2010 financial point of view, a significant difference in intraoperative and postoperative complications in the late group did not translate into differences in other operative outcomes such as the rate of conversion, operative time, and length of hospital stay. One possible explanation is that most of these complications were mild and the effects were short lasting and thus did not justify a change in usual surgical management. Elective laparoscopic cholecystectomy, in particular in patients who had endoscopic sphincterotomy for choledocholithiasis clearance, should be performed 6 weeks after the last cholangitic attack. References 1. Nagino M, Takada T, Kawarada Y, et al. Methods and timing of biliary drainage for acute cholangitis: Tokyo guidelines. J Hepatobiliary Pancreat Surg 2007;14: Schiphorst AH, Besselink MG, Boerma D, et al. Timing of cholecystectomy after endoscopic sphincterotomy for common bile duct stones. Surg Endosc 2008;22: Poon RT, Liu CL, Lo CM, et al. Management of gallstone cholangitis in the era of laparoscopic cholecystectomy. Arch Surg 2001;136: Taylor EW, Dunham RH, Bloch JH. Laparoscopic management of gallstone pancreatitis. J Laparoendosc Surg 1994;4: Tate JJ, Lau WY, Li AK. Laparoscopic cholecystectomy for biliary pancreatitis. Br J Surg 1994;81: Tang E, Stain SC, Tang G, et al. Timing of laparoscopic cholecystectomy in gallstone pancreatitis. Arch Surg 1995:130: Bulkin AJ, Tebyani N, Dorazio RA. Gallstone pancreatitis in the era of laparoscopic cholecystectomy. Am Surg 1997;63: Johnson CD, Charnley R, Rowlands B, et al; UK Working Party on Acute Pancreatitis. UK guidelines for the management of acute pancreatitis. Gut 2005;54(suppl):iii Sarli L, Iusco D, Sgobba G, et al. Gallstone cholangitis a 10-year experience of combined endoscopic and laparoscopic treatment. Surg Endosc 2002;16: Mayumi T, Takada T, Kawarada Y, et al. Results of the Tokyo consensus meeting Tokyo guidelines. J Hepatobiliary Pancreat Surg 2007;14: Dindo D, Dermartines N, Clavien PA. Classification of surgical complications a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240: McAlister VC, Davenport E, Renouf E. Cholecystectomy deferral in patients with endoscopic sphincterotomy. Cochrane Library 2009; Lau JY, Leow CK, Fung TMS, et al. Cholecystectomy or gallbladder in situ after endoscopic sphincterotomy and bile duct stone removal in Chinese patients. Gastroenterology 2006;130: Boerma D, Rauws EA, Keulemans YC, et al. Wait-and-see policy or laparoscopic cholecystectomy after endoscopic sphincterotomy for bile duct stones: a randomized trial. Lancet 2002;360: Schachter P, Peleg T, Cohen O. Interval laparoscopic cholecystectomy in the management of acute biliary pancreatitis. HPB Surg 2000;11: Cameron DR, Goodman AJ. Delayed cholecystectomy for gallstone pancreatitis: readmissions and outcomes. Ann R Coll Surg Engl 2004; 86: Hamy A, Hennekinne S, Pessaux P, et al. Endoscopic sphinctertomy prior to laparoscopic cholecystectomy in the treatment of cholelithiasis. Surg Endosc 2003;17: Georgiades CP, Mavromatis TN, Kourlaba GC, et al. Is inflammation a significant predictor of bile duct injury during laparoscopic cholecystectomy? Surg Endosc 2008;22: Rijna H, Borgstein PJ, Meuwissen SGM, et al. Selective preoperative endoscopic retrograde cholangiopancreatography in laparoscopic biliary. Br J Surg 1995;82: Lee SH, Jin HH, Yang KY, et al. Does endoscopic sphincterotomy reduce the recurrence rate of cholangitis in patients with cholangitis and suspected of a common bile duct stone not detected by ERCP? Gastrointest Endosc 2008;67: Sugiyama M, Atomi Y. Does endoscopic sphincterotomy causes prolonged pancreatobiliary reflux? Am J Gastroenterol 1999;94: Mandryka Y, Klimczak J, Duszewski M, et al. Bile duct infections as a late complication after endoscopic sphincterotomy [article in Polish]. Polski Merkuriusz Lekarski 2006;21: de Vries A, Donkervoort SC, van Geloven AA, et al. Conversion rate of laparoscopic cholecystectomy after endoscopic retrograde cholangiography in the treatment of choledocholithiasis: does the time interval matter? Surg Endosc 2005;19: Z graggen K, Wehrli H, Metzger A, et al. Complications of laparoscopic cholecystectomy in Switzerland a prospective 3-year study of patients. Surg Endosc 1998;12: