Validity of Clinical Criteria in the Management of Endoscopic Retrograde Cholangiopancreatography–Related Duodenal Perforations

Home , Omentopexy

ORIGINAL ARTICLE Validity of Clinical Criteria in the Management of Endoscopic Retrograde Cholangiopancreatography–Related Duodenal Perforations

Ahmad Assalia, MD; Alain Suissa, MD; Anat Ilivitzki, MD; Ahmad Mahajna, MD; Kamal Yassin, MD; Moshe Hashmonai, MD; Michael Moshe Krausz, MD

Objective: To assess the validity of predetermined clini- Results: Diagnosis of DP was accomplished early (within cal and radiologic criteria in the management of endo- 6 hours of ERCP) in 20 of 22 patients (91%). Three pa- scopic retrograde cholangiopancreatography (ERCP)– tients with early positive peritoneal signs were treated related duodenal perforations (DPs). surgically; 2 of them sustained injury from the endoscope, and the third by papillotomy. All 3 patients had Design: Prospective case series. significant findings justifying immediate surgery. Nine- teen patients with retroperitoneal DPs due to papil- Setting: Tertiary medical facility. lotomy or guidewire insertion were treated nonoperatively. In patients diagnosed early (n=17), only 1 failure Patients: Twenty-two individuals with ERCP-related DPs. (6%) occurred. In the 2 patients with delayed diagnosis, there was 1 failure, which culminated in death. Interventions: The guidelines advocated operative management for patients with free contrast leak and those with Conclusions: Our results might validate the role of clini- positive peritoneal signs or other indicators suggesting sep- cal criteria for the selective management of ERCP- sis irrespective of the mechanism or location of injury. related DPs. These criteria were found to correlate well Main Outcome Measures: Thirty-day mortality rates, with radiologic findings and the mechanism of injury. success of nonoperative management, and complications related to nonoperative and operative policies. Arch Surg. 2007;142(11):1059-1064

UODENAL PERFORATIONS since the early era of ERCP and endo- (DPs) after endoscopic scopic sphincterotomy,6,7,10 multiple stud- retrograde cholangiopan- ies8,9,11,13,16 advocating primary surgical creatography (ERCP) and management have been published. endoscopic sphincter- Recently, several researchers4,5,17 have in- otomy are rare, with a reported incidence troducedclassificationsofERCP-relatedDPs D 1-3 of 0.3% to 1.3%. However, DPs are well- recognized and potentially hazardous com- See Invited Critique plications. Retroperitoneal DPs, mainly at end of article from papillotomy or precut papillotomy, represent almost 90% of cases, whereas in- based on anatomical location and mecha- traperitoneal injuries are less common and occur mainly as a result of injury from the nisms of injury (Table 1) and accordingly have suggested management guidelines. Therefore, it seems that the current consen- Author Affiliations: CME available online at sus favors selective operative management. Departments of Surgery B www.archsurg.com However, existent guidelines still differ as to (Drs Assalia, Ilivitzki, and the validity of clinical criteria for stratifying Hashmonai), Gastroenterology endoscope itself.3-5 Because of the rarity of patientsintosurgicalandnonsurgicalcohorts (Drs Suissa and Yassin), and this complication, management recom- and the appropriate imaging modality for the Surgery A (Drs Mahajna and mendations have largely been based on diagnosis and monitoring of treatment. Fur- Krausz), Rambam Medical 6-14 1 Center and the Bruce Rappaport small series, expert opinion, and lit- thermore,theroleofendoscopicmanagement 2,15 School of Medicine, erature reviews. Although nonoper- of these complications remains controver- 18 Technion-Israel Institute of ative management has been recognized as sial. In this study, we prospectively evalu- Technology, Haifa. a viable option for retroperitoneal DPs ated the applicability of preset clinical and

(REPRINTED) ARCH SURG/ VOL 142 (NO. 11), NOV 2007 WWW.ARCHSURG.COM 1059

Stapfer et al17 Howard et al4 Enns et al5

Type Definition Recommendation Type Definition Recommendation Type Definition Recommendation I Lateral duodenal Immediate surgery III Duodenal injury Immediate surgery Duodenal Not related to Immediate surgery wall injury remote from sphincterotomy/ papilla guidewire (scope related) II Peri-Vaterian Surgery required if II Periampullary Early diagnosis Sphincterotomy Periampullary Nonoperative injury large free or injury and immediate related management retroperitoneal (duodenal endoscopic with or without fluid collections wall and drainage endoscopic exist or in cases ducts) intervention in of retained most patients; stones or surgery in hardware hemodynamic instability, high fever, and nonresolving abdominal pain III Ductal injury Same as in type II I Guidewire Nonoperative Guidewire Ductal or Same as above (wire/basket) perforations management related periampullary (of ducts) injury IV Retroperitoneal Nonoperative NA NA NA Others Dilatations (bile Immediate air alone management duct injury) stenting

Abbreviations: ERCP, endoscopic retrograde cholangiopancreatography; NA, not applicable.

ment, surgical procedures, subsequent endoscopic or surgical Table 2. Characteristics of 22 Patients With Duodenal interventions, length of hospital stay, and general outcome. Perforations Before conducting the present study, we applied a set of criteria for the selective management of ERCP-related DPs based Characteristic Value on previous experience and pertinent literature. These guide- Age, mean ± SD, y 63.8 ± 6.7 lines advocated immediate operative management of docu- Sex, M/F, No. 10/12 mented DPs for the following criteria: Comorbidities, No. v Clinical criteria: Patients with positive peritoneal signs, Ischemic heart disease 6 abdominal pain of increasing intensity, or other signs sugges- Hypertension 4 tive of sepsis irrespective of the type, mechanism, and loca- Congestive heart failure 3 tion of injury. These signs included any combination of 2 or Cardiac arrhythmia 4 Ͼ Chronic obstructive pulmonary disease 2 more of the following: fever (temperature 100.4°C), tachycardia (heart rate Ͼ100/min), systolic blood pressure less than Peripheral vascular disease 3 Ͼ Liver cirrhosis 1 100 mm Hg, dyspnea (respiratory rate 20/min), and leuko- Ͼ ϫ 9 ϫ 9 Diabetes mellitus 5 cytosis (white blood cell count 12 10 /µL [to convert to 10 Hypothyroidism 1 per liter, multiply by 0.001]). v Radiologic criteria: Any free (intraperitoneal or retroperitoneal) leakage of contrast material documented during ERCP, upper gastrointestinal (UGI) study, or abdominal computed tomography (CT). radiologic criteria for the selective management of ERCP- related DPs. Contained minor leaks were not considered an indication for surgical intervention. All other patients underwent initial nonoperative treatment, which included nasogastric drainage, METHODS intravenous fluids, ceftriaxone disodium (a third-generation cephalosporin), and close clinical monitoring in the surgical This study was approved by the ethics committee, Rambam service for possible deterioration. The nasogastric tube was re- Health Care Campus, Haifa, Israel. Between January 1, 1996, moved when patients started to demonstrate clear improve- and June 30, 2003, 3104 ERCP procedures with (n=1250, 40%) ment, and diet was usually resumed 1 to 2 days later. Patients or without (n=1854, 60%) sphincterotomy were performed at were discharged when they were completely asymptomatic and the Rambam Medical Center. Precut papillotomy was per- tolerating food intake. An immediate UGI sodium amidotri- formed in 306 patients (10%). Twenty-two patients (0.7%) had zoate (Gastrografin; Schering AG, Berlin, Germany) study was documented DPs. The ERCP was diagnostic in 2 patients and used in the first 4 patients; however, because it missed the di- therapeutic in 20. agnosis in 3 of them, it was not used further. Routine early ab- Characteristics of the patients are given in Table 2. Data dominal CT was performed in all patients with no immediate were prospectively collected for ERCP findings, method of de- indication for surgery (including cases in which a UGI study tection of perforation, clinical presentation of patients, imaging was initially used). The purpose was to confirm the suspected findings, time to diagnosis and treatment, type of manage- diagnosis and to assess free leaks, large fluid collections, or any

(REPRINTED) ARCH SURG/ VOL 142 (NO. 11), NOV 2007 WWW.ARCHSURG.COM 1060

Clinical and Laboratory Findings Patients, No. (%) Comments Abdominal/flank pain or discomfort 22 (100) During ERCP in 16 patients (73%) Within4hofERCP in 4 patients 6-8 h after ERCP in 2 patients (delayed diagnosis) Mild to moderate abdominal tenderness 14 (64) Peritoneal signs 4 (18) Intraperitoneal DPs in 2 patients Retroperitoneal DPs in 2 patients (diagnosis: 1 immediate and 1 delayed) Tachycardiaa 14 (74) Mild: heart rate Յ100/min above baseline in 10 patients Overt: heart rate Ͼ100/min, in 4 patients Low-grade fevera 9 (47) Ͻ100.4°C Subcutaneous emphysemaa 3 (16) All managed nonsurgically Mild leukocytosisa 6 (32) White blood cell count of 10 000-12 000/µL Hyperamylasemiaa 7 (37) Amylase level Ͼ150 U/L

Abbreviations: DP, duodenal perforation; ERCP, endoscopic retrograde cholangiopancreatography. SI conversion factors: To convert amylase to µkat/L, multiply by 0.01667; white blood cell count to ϫ109/L, multiply by 0.001. a Data were collected from 19 patients who were treated nonoperatively.

intraperitoneal component of the perforation and the presence of possible pancreatitis. Failure to clearly improve within Table 4. Methods of DP Diagnosis 24 to 48 hours, development of peritonitis, hemodynamic deterioration, and septic signs were considered indications for Method Patients, No. prompt surgical intervention. Early diagnosisa Visualization of peritoneum (during ERCP) 2 RESULTS Contrast media leakage during ERCP 8 (6 Nonequivocal, 2 equivocal) Retroperitoneal air on plain abdominal film 13 (11 Nonequivocal, ERCP DATA 1 equivocal, and 1 negative) b The indications for ERCP were as follows: obstructing Retroperitoneal air on abdominal CT 17 Delayed diagnosisc periampullary tumors (8 patients), choledocholithiasis Retroperitoneal air on abdominal CTb 2 (1 After 8 h, before elective cholecystectomy (8 patients), choledo- 1 after 24 h) cholithiasis associated with cholangitis (2 patients), biliary leaks and retained stones after cholecystectomy (2 Abbreviations: CT, computed tomography; DP, duodenal perforation; patients), dilatation of bile ducts with impaired liver func- ERCP, endoscopic retrograde cholangiopancreatography. a Diagnosis during or within 6 hours of ERCP (20 patients [91%]). tion test results (1 patient), and papillary stenosis (1 pa- b Along with other findings (see Table 5). tient). c Diagnosis more than 6 hours after ERCP (2 patients [9%]). Difficulties in the procedures were noted in 16 of 22 patients (73%). Possible risk factors for ERCP complications included periampullary diverticulum of the duo- MANAGEMENT AND OUTCOMES denum in 8 patients and Billroth II gastrectomy and papillary stenosis in 1 patient each. A biliary stent was Operative Management introduced in 2 patients, one with a bile leak after laparoscopic cholecystectomy and the other with choledo- Patients with lateral DP caused by the endoscope (n=2) cholithiasis and cholangitis. A nasobiliary drain was used underwent surgery immediately (Figure). Suture closure in 1 patient with choledocholithiasis and cholangitis af- with omentopexy was performed. In 1 patient with inop- ter precut papillotomy. The procedure was stopped af- erable carcinoma of the head of the pancreas, a cholecys- ter precut papillotomy and before selective cannulation tostomy was added. In a third patient who had sustained of the bile ducts in 4 patients. an injury from papillotomy performed for retained chole- dochal stones after cholecystectomy, a 3-cm perforation was CLINICAL FEATURES found in the retroperitoneal second part of the duodenum AND METHODS OF DIAGNOSIS along with injury of the intraperitoneal distal common bile duct and subhepatic bile collection. Suture closure of the The clinical presentation of patients who were initially duodenum, bile duct exploration, and retroduodenal area treated nonoperatively is depicted in Table 3, and the and T-tube drainage were performed. All immediately op- methods of diagnosis of DPs are given in Table 4. Ab- erated-on patients recovered and were discharged after a dominal CT was performed within 6 to 8 hours of ERCP mean±SD hospital stay of 8.5±2.2 days. in all patients except 1 who was diagnosed 24 hours after ERCP and 2 with perforation of the lateral duodenal Nonoperative Management wall caused by the endoscope that necessitated immediate surgery. The findings in these 19 patients are given Early Diagnosis (Յ6 Hours After ERCP). Seventeen pain Table 5. tients who were immediately diagnosed after ERCP were

(REPRINTED) ARCH SURG/ VOL 142 (NO. 11), NOV 2007 WWW.ARCHSURG.COM 1061

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/03/2021 foration was diagnosed 24 hours after the procedure. The Table 5. Early Abdominal Computed Tomography Findings a CT showed retroperitoneal and mediastinal air, 1 small in 19 Patients retroperitoneal fluid collection, and a slight amount of subhepatic fluid. Because of advanced age, comorbidi- Finding Patients, No. ties, and borderline presentation, the patient was ini- Retroperitoneal air 19 (massive in 3) tially treated nonoperatively. After clinical deteriora- Contained minimal contrast leakage 4 tion, a salvage operation performed on day 3 showed a Free contrast leakageb 1 Intraperitoneal fluid collectionb 1 retroduodenal abscess and bilious perihepatic fluid, and Free intraperitoneal air 3 (1 fair amountb the site of perforation was not found. Cholecystectomy and 2 minimal) and retroduodenal and T-tube drainage were under- Minimal free intraperitoneal air 2 taken. Subsequently, 1 repeated operation was neces- Minimal subhepatic fluid 1 sary for uncontrolled duodenal leakage, and the patient Minimal periduodenal fluid 2 died of sepsis on day 18. Pneumomediastinum 2 Mild pancreatitis (limited to head) 1 Subcutaneous emphysema 3 COMMENT Thickening of posterior duodenal wall 4

a Eighteen patients underwent computed tomography within 6 hours of Since the introduction of ERCP and endoscopic sphinc- endoscopic retrograde cholangiopancreatography and 1 patient within terotomy, the management of DPs has been controver- 8 hours of endoscopic retrograde cholangiopancreatography. b Seen in 1 patient who was treated surgically. The remaining findings sial. The debate has mainly focused on the management were observed in patients treated conservatively. of retroperitoneal perforations because many intraperitoneal injuries were unquestionably managed surgically. It was difficult to draw clear guidelines owing to initially treated nonoperatively (Figure). These patients the limited experience in individual centers. Although met the predetermined criteria for nonoperative treat- endoscopists have provided early accounts of successful ment. Thirteen patients (76%) improved significantly nonoperative management,1,6,7,10 surgeons were reluc- within 12 to 24 hours. Ten patients (59%) were nearly tant to adopt this policy. Their recommendations varied asymptomatic 2 to 4 days after the perforation. The na- from early surgery in most cases8,9,11,13 to a policy of non- sogastric tube was removed after 48 hours in 13 pa- operative management supported by a few.4 Some stud- tients (76%). In the remaining patients, it was removed ies9,12,17,19 stated nonsignificant findings or inability to iden- 3 to 6 days after the procedure. tify the perforation site during exploration. On the other Three patients were treated endoscopically during hand, a high incidence of morbidity and mortality was ERCP with biliary stents (2 patients) and a nasobiliary claimed to be associated with failed nonoperative man- drain (1 patient). One other patient with inoperable pan- agement.2,9,17 These data, however, should be viewed with creatic malignancy and deep jaundice underwent percu- scrutiny because it was unclear whether these opera- taneous biliary drainage on day 2 after ERCP as a defini- tions were primary or salvage procedures. The indica- tive palliation. Five patients agreed to undergo a second tions for and timing of nonsurgical and surgical treat- therapeutic ERCP, which was performed successfully 7 ment were not reported either. Furthermore, either to 14 days after the perforation. In the remaining pa- diagnosis was late in many cases8,11,13,16 or salvage sur- tients, surgery was performed for disease management. gery was performed with considerable delay.8,9,11,13,16,17 As Three patients demonstrated a slow course of im- the body of data grew, it became clear that many pa- provement. Repeated CT of the abdomen after 3 to 4 days tients could be treated nonsurgically and that the main showed mild pancreatitis in 2 patients and a small ret- clinical challenge was to distinguish patients who could roduodenal fluid collection in the third that was success- be treated nonoperatively from those who would re- fully drained percutaneously. The mean±SD hospital stay quire early surgery. in this successfully managed group was 6.2±2.3 days. One Similar to other perforations of the gastrointestinal patient who was initially treated nonoperatively deterio- tract, the importance of early diagnosis cannot be over- rated within 6 hours of initiation of nonoperative man- emphasized. This allows the immediate institution of treat- agement. Tachycardia (heart rate, 120/min), dyspnea, and ment, with more patients being selected for nonoper- signs of peritonitis in the right upper abdomen man- ative management. This might potentially ensure better dated urgent surgery. A 2-cm retroperitoneal tear in the outcomes.4,6,10,20,21 The present results in patients diag- duodenum was primarily sutured by means of omento- nosed early are comparable with those in other recently pexy. The retroduodenal space (in which air and a mini- published series.4,5 These results contrast those re- mal amount of bile were present) was drained; cholecys- ported in some other series8,11,13,16 describing delayed di- tectomy and bile duct exploration with T-tube drainage agnosis in 70% to 75% of cases, in which most of them were performed as well. The patient recovered unevent- were managed operatively with unfavorable outcomes. fully and was discharged 9 days after surgery. In accordance with other studies,4,6,7,10,17,20 our experience showed that the clear demonstration of free ret- Delayed Diagnosis (Ͼ6 Hours After ERCP). One pa- roperitoneal air or leakage during ERCP allowed imme- tient with retroperitoneal DP after removal of a biliary diate diagnosis in most patients (80%). In the remaining stent was successfully treated nonsurgically and was dis- 4 cases, equivocal findings were confirmed by CT. Al- charged 6 days after ERCP. In the other patient the per- though routine CT was not used in several series,4,5,17 we

(REPRINTED) ARCH SURG/ VOL 142 (NO. 11), NOV 2007 WWW.ARCHSURG.COM 1062

1 Other 2 Scope related 17 Sphincterotomies or 2 Guidewire (stent removal) (lateral duodenal) precut papillotomy related related

Delayed Immediate 16 Immediate 1 Late Immediate diagnosis diagnosis diagnoses diagnoses diagnosis

Nonoperative Nonoperative Nonoperative Nonoperative management Surgery management management management

Recovery Recovery 15 Recoveries 1 Failure Failure Recovery

Salvage surgery Salvage surgery

Recovery Death

Figure 1. Patient management and outcomes according to type of perforation.

suppose that this modality should be used in every symp- ence and that of others4 lends further support to a policy tomatic case or in those with retroperitoneal air demon- of nonoperative management. We tend to disagree with the strated in plain abdominal films as well as in cases with conclusions of Stapfer et al17 and Howard et al4 regarding uncertain or delayed diagnosis and in nonsurgically treated the unreliability of clinical findings in directing the treat- patients displaying an unsatisfactory course. In these in- ment of these patients. In fact, it was surprising to realize stances, it is considered to be the most accurate modal- that in the series by Howard et al,4 all patients who were ity for the diagnosis of these perforations and their com- operatively salvaged had harbored early clinical findings plications.4,9,22-26 If used routinely after ERCP with suggestive of peritonitis. In addition, in the study by Stapfer papillotomy, CT may detect retroperitoneal air in 13% et al,17 at least several reported failures of initially nonop- to 29% of patients.24,25 In the absence of symptoms, the erative cases were, in retrospect, not suitable for this policy. mere presence of retroperitoneal air has no clinical sig- Another major issue is the role of immediate endo- nificance, and these patients do not require any special scopic intervention in periampullary DP. The rationale treatment. The recommendation for the routine use of behind this is to divert the bile and pancreatic juice to UGI studies17 is not supported by our limited experi- ensure adequate drainage into the duodenum instead of ence and does not seem to be substantiated in the litera- the perforation site. Dunham et al10 pioneered this con- ture. Furthermore, UGI studies are considered to be in- cept in 1982, and since then several proponents have ad- ferior to CT regarding sensitivity to detect sealed-off or vocated immediate endoscopic drainage, either by com- minimal leaks, and they do not demonstrate intraperi- pleting the sphincterotomy or by placing a stent or a toneal findings or the presence of other complications, nasobiliary catheter.4,5,7 Healing problems were not ob- such as pancreatitis or fluid collections. served, and the general outcome was successful. No com- This study differs from previously reported series in that parisons with patients who did not undergo drainage were patients were treated according to predetermined guide- performed, thus making this mode of management dif- lines and data were collected prospectively. In agreement ficult to assess. Our policy advocated a trial of immedi- with us, several researchers5,12,15 advocated that the clini- ate drainage only in those with obstructed and infected cal condition of the patient should be the key factor deter- biliary systems (mainly cholangitis due to stones). With mining the mode of treatment. We believe that clinical and obstructing tumors, bile is not infected in most cases, and laboratory variables should be correctly interpreted in the perforations usually occur distal to the tumor, thus ob- overall clinical context. Combined with correlation with viating the immediate need for drainage. Nevertheless, mechanism of injury and radiologic findings, this seems endoscopic management seems to be a logical and use- to guarantee the best guidance for selective management. ful option provided that an experienced endoscopist is Radiologic findings alone should not determine the mode available and there is a real need for immediate drainage of therapy because small amounts of intraperitoneal air or of the biliary system. Delayed therapeutic ERCP consti- fluid in the setting of benign clinical presentation (2 pa- tutes a viable option for the management of disease, as tients in this series) might be successfully treated nonop- was shown by us and recently by others.27 eratively.5,17 Furthermore, the amount of retroperitoneal The recent classification systems introduced by sev- or subcutaneous air correlates more with the degree and eral researchers4,5,17 attempted to improve our under- length of endoscopic manipulation and insufflation after standing of the issue of ERCP-related DPs and to better the injury than with the actual size of the perforation.1,5,23 direct management policy based on mechanisms of in- Provided that these cases are silent clinically, our experi- jury or anatomical locations. These essentially similar sys-

(REPRINTED) ARCH SURG/ VOL 142 (NO. 11), NOV 2007 WWW.ARCHSURG.COM 1063

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/03/2021 tems (Table 1) are further supported by the present re- 2. Scarlett PY, Falk GL. The management of perforation of the duodenum following endoscopic sphincterotomy: a proposal for selective therapy. Aust NZJSurg. sults (Figure). It is our opinion that these classifications 1994;64(12):843-846. should be prudently viewed just as a general predictor 3. Loperfido S, Angelini G, Benedetti G, et al. Major early complications from di- of the mode of treatment that a specific patient will prob- agnostic and therapeutic ERCP: a prospective multicenter study. Gastrointest Endosc. 1998;48(1):1-10. ably need, and they should be cautiously used only in 4. Howard TJ, Tan T, Lehman GA, et al. Classification and management of perfora- conjunction with clinical and radiologic findings for tions complicating endoscopic sphincterotomy. Surgery. 1999;126(4):658-665. proper decision making. 5. Enns R, Eloubeidi MA, Mergener K, et al. ERCP-related perforations: risk factors and management. Endoscopy. 2002;34(4):293-298. We conclude that most ERCP-related DPs in sympto- 6. Safrany L. Duodenal sphincterotomy and gallstone removal. Gastroenterology. matic patients should be diagnosed during or immedi- 1977;72(2):338-343. ately after the procedure by the demonstration of retro- 7. Byrne P, Leung JWC, Cotton PB. Retroperitoneal perforation during duodeno- scopic sphincterotomy. Radiology. 1984;150(2):383-384. peritoneal free air or contrast leakage. A high index of 8. Bell RCW, Steigmann GV, Goff J, et al. Decision for surgical management of per- suspicion and early recognition are key to a successful out- foration following endoscopic sphincterotomy. Am Surg. 1991;57(4):237-240. come. In doubtful cases or those with delayed presenta- 9. Sarr MG, Fishman EK, Milligan FD, et al. Pancreatitis or duodenal perforation after peri-Vaterian therapeutic endoscopic procedures: diagnosis, differentiation tion, CT is the modality of choice for establishing a diag- and management. Surgery. 1986;100(3):461-466. nosis and monitoring conservatively treated patients. The 10. Dunham F, Bourgois N, Gelin M, et al. Retroperitoneal perforations following en- clinical utility of UGI studies is doubtful. The decision for doscopic sphincterotomy: clinical course and management. Endoscopy. 1982; 14(3):92-96. stratifying patients into surgical and nonsurgical cohorts 11. Booth FV, Doerr RJ, Khalafi RS, Luchette FA, Flint LM. Surgical management of should be based mainly on clinical grounds, which were complications of endoscopic sphincterotomy with precut papillotomy. Am J Surg. found to correlate well with radiologic findings. In con- 1990;159(1):132-136. 12. Chung RS, Sivak MV, Ferguson R. Surgical decisions in the management of duo- junction with mechanism of injury, this constitutes the best denal perforation complicating endoscopic sphincterotomy. Am J Surg. 1993; available predictor of the necessity of surgery. In the ab- 165(6):700-703. sence of evidence-based data, endoscopic management may 13. Chaudhary A, Aranya RC. Surgery in perforation after endoscopic sphincterotomy: sooner, later or not at all? Ann R Coll Surg Engl. 1996;78(3, pt 1):206-208. also be considered, especially in cases with an urgent need 14. Mustard R, Mackenzie R, Jamieson C, et al. Surgical complications of endoscopic for biliary decompression. sphincterotomy. Can J Surg. 1984;27(3):215-217. 15. Cohen SA, Siegel JH, Kasmin FE. Complications of diagnostic and therapeutic ERCP. Abdom Imaging. 1996;21(5):385-394. Accepted for Publication: April 25, 2006. 16. Preetha M, Chung YFA, Chan WH, et al. Surgical management of endoscopic ret- Correspondence: Ahmad Assalia, MD, Laparoscopic Ser- rograde cholangiopancreatography-related perforations. ANZ J Surg. 2003; 73(12):1011-1014. vice, Department of Surgery B, Rambam Medical Cen- 17. Stapfer M, Selby R, Stain SC, et al. Management of duodenal perforation after ter, Haifa 31096, Israel ([email protected]). endoscopic retrograde cholangiopancreatography and sphincterotomy. Ann Surg. Author Contributions: Dr Assalia had full access to all 2000;232(2):191-198. 18. Howard TJ. Management of duodenal perforation after endoscopic retrograde of the data in the study and takes responsibility for the cholangiopancreatography and sphincterotomy [letter]. Ann Surg. 2001;234 integrity of the data and the accuracy of the data analy- (1):132-133. sis. Study concept and design: Assalia, Suissa, Ilivitzki, 19. Petersen S, Henke G, Freitag M, et al. Management of hemorrhage and perforation following endoscopic sphincterotomy. Zentralbl Chir. 2001;126(10):805-809. Mahajna, Hashmonai, and Krausz. Acquisition of data: 20. Martin DF, Tweedle DEF. Retroperitoneal perforation during ERCP and endo- Assalia, Suissa, Ilivitzki, Mahajna, and Yassin. Analysis scopic sphincterotomy: causes, clinical features and management. Endoscopy. and interpretation of data: Assalia, Suissa, and Krausz. 1990;22(4):174-175. 21. Gu¨itro´n-Cantu´ A, Adalid-Martinez R, Gutierrez-Bermudez JA. Conservative man- Drafting of the manuscript: Assalia, Suissa, Ilivitzki, and agement of duodenal perforation following endoscopic sphincterotomy of pa- Hashmonai. Critical revision of the manuscript for impor- pilla of Vater. Rev Gastroenterol Mex. 2003;68(1):6-10. tant intellectual content: Assalia, Suissa, Mahajna, Yassin, 22. Pannu HK, Fishman EK. Complications of endoscopic retrograde cholangiopancreatography: spectrum of abnormalities demonstrated with CT. Radiographics. Hashmonai, and Krausz. Statistical analysis: Assalia. Ob- 2001;21(6):1441-1453. tained funding: Assalia and Suissa. Administrative, tech- 23. Kuhlman JE, Fishman EK, Milligan FD, et al. Complications of endoscopic retro- nical, and material support: Assalia, Suissa, Ilivitzki, grade sphincterotomy: computed tomographic evaluation. Gastrointest Radiol. 1989;14(2):127-132. Mahajna, Yassin, and Krausz. Study supervision: Assalia, 24. Hans de Vries J, Duijm LEM, Dekker W, et al. CT before and after ERCP: detec- Suissa, and Hashmonai. tion of pancreatic pseudotumor, asymptomatic retroperitoneal perforation and Financial Disclosure: None reported. duodenal diverticulum. Gastrointest Endosc. 1997;45(3):231-235. 25. Genzlinger JL, McPhee MS, Fisher JK, et al. Significance of retroperitoneal air after endoscopic retrograde cholangiopancreatography with sphincterotomy. Am REFERENCES J Gastroenterol. 1999;94(5):1267-1270. 26. Zissin R, Shapiro-Feinberg M, Oscadchy A, et al. Retroperitoneal perforation during endoscopic sphincterotomy: imaging findings. Abdom Imaging. 2000;25 1. Cotton PB, Lehman G, Vennes J, et al. Endoscopic sphincterotomy complica- (3):279-282. tions and their management: an attempt at consensus. Gastrointest Endosc. 1991; 27. KayhanB,AkdoganM,SahinB.ERCPsubsequenttoretroperitonealperforationcaused 37(3):383-393. by endoscopic sphincterotomy. Gastrointest Endosc. 2004;60(5):833-835.

(REPRINTED) ARCH SURG/ VOL 142 (NO. 11), NOV 2007 WWW.ARCHSURG.COM 1064