- Email: [email protected]
Clinical outcomes of ERCP-related retroperitoneal perforations
Review Article
Hepatobiliary & Pancreatic Diseases International
Clinical outcomes of ERCP-related retroperitoneal perforations Francesco Guerra, Giuseppe Giuliani, Diego Coletta, Stefano Amore Bonapasta and Giovanni Battista Levi Sandri Rome, Italy
ABSTRACT: Endoscopic retrograde cholangiopancreatography (ERCP)-related perforations represent rare but often severe conditions. While lesions with intraperitoneal perforation have an almost imperative indication to surgery, whether or not to manage retroperitoneal perforations surgically is still an area of debate. The aim of the present work was to review the available clinical evidence on the operatively and medically treated ERCP-related retroperitoneal perforations. From MEDLINE/PubMed databases 137 patients with retroperitoneal perforation were included from 12 studies that met the selection criteria for data investigation and analysis. Twentyfour patients were treated by prompt surgery; 113 were primarily managed conservatively and about 20% of these patients required surgery subsequently. Overall, the morbidity and mortality were 15.4% and 6.6%, respectively. Although most patients with retroperitoneal perforation may benefit from a non-operative management, a considerable number of patients fail to respond to medical treatment and require surgery afterwards. Identifying those patients who are at highest risk of poor outcome after conservative treatment should be considered a research priority.
Introduction
T
he rates of adverse events sustained during endoscopic retrograde cholangiopancreatography (ERCP) vary widely in the scientific literature (approximately from 4% to 10%), and the true incidence in clinical practice is still unclear.[1] As a major event predicting increased morbidity and mortality, perforating iatrogenic lesions have an incidence of 0.1% to 2.1%.[2] This specific risk is substantively higher in therapeutic procedures than in diagnostic ones, and operative ERCPs are increasingly performed.[2, 3] The key point of ERCP-related perforations is to decide which patient can be managed conservatively and who should be promptly operated on. Nevertheless, there is no consensus on the treatment for patients with no free abdominal air.[2-5] Despite many classifications have been proposed to predict the prognosis and to address adequate treatment,[6, 7] neither the optimal management (Hepatobiliary Pancreat Dis Int 2017;16:160-163) nor clinical or radiological findings that could guide the decision-making process have yet been defined. KEY WORDS: endoscopic retrograde cholangiopancreatography; Over the last years, several studies[6-9] have suggested complications; that patients with iatrogenic ERCP-related retroperitoretroperitoneum neal perforation can recover without surgery. Nevertheless, surgical management is still advocated by a number of authors[5] and it is widely known that delayed surgery can be associated with dismal outcomes.[8, 9] Disagreement regarding how patients with such condition should be managed therefore exists. Accordingly, this study was undertaken to review the available evidence from the literature to analyze the clinical outcomes of patients manAuthor Affiliations: Department of Surgery, Umberto I University Hospital, Rome, Italy (Guerra F, Giuliani G, Coletta D, Amore Bonapasta S aged surgically or conservatively after a retroperitoneal perforating lesion occurred during an ERCP procedure. and Levi Sandri GB) Corresponding Author: Francesco Guerra, MD, Department of Surgery, Umberto I University Hospital-viale del Policlinico 155 00161, Rome, Italy (Tel: +39-333-3555-235; Email: [email protected]) © 2017, Hepatobiliary Pancreat Dis Int. All rights reserved. doi: 10.1016/S1499-3872(16)60106-6 Published online June 21, 2016.
Methods An initial electronic search was performed to identify all studies dealing with ERCP-related perforations, without
160 • Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com
ERCP-related retroperforations
discrimination between intra- and retroperitoneal lesions. The PubMed/MEDLINE database was searched in December 2014. The search strategy was (“perforation”) AND (“ERCP” OR “cholangiography” OR “cholangiopancreatography”). When inclusion was unclear based on abstracts, full papers were obtained. Hand searches of relevant articles’ references were also performed to find possible additional works. Inclusion criteria accepted all articles in English language, from 2004 onwards, which presented data concerning treatment of ERCP-related perforations. Abstracts and unpublished data as well as small series involving less than five patients and case reports were excluded from the analysis. Were included: 1. All patients, regardless of age or gender, in whom an ERCP-related retroperitoneal perforation was documented and whose management was described in detail; 2. Patients with altered anatomy, such as those with Billroth II or Rouxen-Y reconstruction only if a retroperitoneal perforation was diagnosed. Two reviewers (GF and GG) carried out the investigation independently. From the studies that were included, data for meta-analysis were extrapolated following a pre-established pattern. Results from extrapolation were then submitted to other two authors (CD and ABS) to be verified. All disagreements were resolved by consensus, involving all authors. Patients were assorted into an operative treatment group (OT) when they were primarily treated by surgery (within 24 hours from the diagnosis). In contrast, patients were assorted in a non-operative treatment (including the need for percutaneous drainage) group (NOT) when a conservative therapy was initially adopted, regardless of the need for subsequent surgical intervention. Treatment modalities and clinical outcomes such as overall morbidity, mortality and length of hospital stay of both operative and non-operative treatments were assessed. Extemporaneous attempts of mucosal closures when the perforation was diagnosed at the time of endoscopy were not considered in the analysis. Statistical analysis was performed using the Statistical Package for the Social Sciences, v20.0 (SPSS Inc., Chicago, IL, USA). Results were presented in descriptive statistics. To determine if homogeneous samples were different between medically and surgically treated patients, univariate analysis was performed using Student’s t test or Fisher’s exact test for continuous and categorical variables, respectively. A two-tailed P value of 0.05 or less was deemed statistically significant.
Table 1. Summary characteristics of the included studies Studies Rabie et al[4] Alfieri et al[5] Wu et al[10] Assalia et al[11] Avgerinos et al[12 ] Li et al[13] Morgan et al[14] Salminen et al[15] Neri et al[16] Ercan et al[17] Krishna et al[18] Kim et al[19] Total
Year of publi- Country cation 2013 2013 2006 2007 2009 2012 2009 2008 2006 2012 2011 2009
Saudi Arabia Italy Canada Israel USA China USA Finlandia Italy Turkey India Korea
ERCPs
Perfora- Retropertions forations
597 14 618 6 620 3 104 4 358 8 504 12 817 2 788 101 9 209 2 247 64 963
10 30 30 22 15 16 24 5 5 52 14 18 241
9 27 28 19 1 9 12 2 5 8 8 9 137
After the evaluation of abstracts, full-texts and references, 12 studies[4, 5, 10-19] with a total of 137 retro-perforated patients (out of 241 perforation from a total of 64 963 ERCPs included) met all the inclusion criteria. General characteristics of studies and patients are summarized in Table 1. One of the major reasons for exclusion of studies has been the inability to extrapolate data specific to single patients. The percentages were often referred to the total of patients treated, with impossibility to extrapolate data fractionated according to the type of perforation, therapeutic approach, incidence and type of complications. Overall, 24 patients were treated operatively, whereas 113 were primarily managed by non-operative treatment. Twenty-one (18.6%) patients failed to non-operative treatment and required surgery subsequently. These patients proceeded to surgery because of one or more of the following features: persistent fever despite adequate antibiotic therapy, abscess or collection development believed to require direct drainage or ongoing or worsening symptoms. Overall, non-operative management consisted of nihil per os, nasoduodenal/nasobiliary drainage, intravenous fluids, and antibiotic therapy with or without percutaneous drainage. In the OT group, primary surgery consisted of a combination of the following: direct repair/debridement,[4, 5, 11, 15-17] common bile duct exploration and T-tube placement,[4, 11, 15-17] debridement via a retroperitoneal laparostomy,[5] and duodenostomy/jejunostomy[4, 16] or gastrojejunostomy.[16] Information on incidence of complications was reported in detail in 11 out of 12 studies. Sepsis (5 patients, 3 in the OT group), acute pancreatitis (3 patients, all in Results the NOT group), retroperitoneal abscess (4 patients, 3 A total of 151 potentially relevant articles were identified. in the NOT group) and duodenal leaks (3 patients, all in Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com • 161
Hepatobiliary & Pancreatic Diseases International
Table 2. Clinical outcomes of ERCP-related retroperitoneal perforations Morbidity (n, %)
Mortality (n, %)
Groups n
Hospital stay (d)
OT
24
20.3 (data from 21 patients)
5 (22.7) (data from 22 patients)
5 (20.8)
NOT
113
21.1 (data from 94 patients)
12 (13.6) (data from 88 patients)
4 (3.5)
Total
137
20.9 (data from 115 17 (15.4%) (data from 9 (6.6) patients) 110 patients)
Values are reported as weighted means (percentages in parentheses). OT: operative treatment; NOT: non-operative treatment.
the NOT group, who subsequently underwent surgery) were the most frequently observed morbidities. Overall, 17 patients experienced complications. By analyzing only those studies whose specific data were provided appropriately, the total morbidity rate was 15.4%. In the OT group, the overall incidence of complications was 22.7% whereas it was 13.6% in the NOT group. This difference did not elicit statistical significance (P=0.43). Mean postoperative hospital stays were 21.1±11.6 days for the NOT group (weighted data from 10 studies including 94 patients) and 20.3±6.5 days in the OT group (6 studies, 21 patients) with no statistically significant difference between the two groups (P=0.76). The overall mortality of the two groups was 6.6%. Particularly, in the OT group, the overall mortality was 20.8%, whereas in the NOT group it was 3.5% (P=0.01). Data concerning clinical outcomes are shown in Table 2. According to those studies that provided detailed data, 80% of the patients who eventually died had delayed diagnosis of perforation. Notably, of the 5 patients who received more than one surgical procedure, 4 succumbed.
Discussion Iatrogenic perforation resulting from ERCP is a rare but serious complication with high morbidity and mortality.[1-4, 6] The type of perforation can be different during therapeutic ERCPs and many scholars have been investigating specific features in terms of localization and mechanism of injury advocating possible correlation with therapeutic modalities.[2, 6, 7] Despite this, the problem of choosing the appropriate treatment for patients with perforations occurred within an ERCP procedure is often encountered in front of clinical or radiological evidences, rather than a given endoscopic occurrence. This is particularly the case when a radiological evidence of a retroperitoneal perforation is observed on CT that is mostly required due to clinical suspicions after an endo-
scopic procedure. Actually, there is a relatively imperative indication for surgical exploration in the presence of a recognized intraperitoneal perforation.[7, 8, 20] Conversely, decision-making regarding non-operative versus operative treatment of patients with retroperitoneal perforations can be contentious. Indeed, while several authors[8, 11, 20] intimate that clinical condition should be the main factor determining operative versus conservative management, others[6, 9, 21] have reported that clinical findings are unreliable in predicting the need for surgery. Similarly, other factors such as CT evidence of retroperitoneal fluid collection[6, 10] or contrast extravasation[4] have been considered dependable indications for immediate surgery by some, even though several authors agree on the fact that such features alone should not guide the surgeon’s decision-making process.[7, 11, 12] In fact, the early and accurate detection of patients with ERCP-related retroperitoneal perforation who are at the highest risk for a poor outcome remains a clinical challenge.[6, 8-10] Accordingly, we aimed to identify, summarize, and combine the published literature about the operative and non-operative treatment of such conditions. To the best of our knowledge, this is the first summary specifically focusing on clinical outcomes and associated clinical practice in the management of ERCP-related retroperforations and to establish current incidence and relative morbidity and mortality. In the 64 963 procedures analyzed, 137 (0.2%) patients experienced clinically relevant retroperitoneal perforation. Although the methods used to define an “early recognition” of ERCP-related retroperforation were not homogeneous among studies, delayed diagnosis was almost invariably associated with longer hospitalization, increased morbidity and mortality. These data are consistent with most findings from the literature concerning all types of perforation.[2, 21, 22] Overall, 17.5% of patients received prompt surgical treatment whereas for most patients (82.5%) medical treatment alone was primarily judged to be adequate.[6, 21] Nevertheless, about 20% of patients receiving medical treatment required surgery subsequently, resulting in increased morbidity and mortality. Interestingly, the need for surgery after failure of initial medical management was found in about 21% of patients with sphincterotomy related perforations (type II according to Stapfer et al[6]) in a recent review including 11 studies and more than 500 perforations. In total, nine fatalities were registered in our review, delineating an overall mortality rate of 6.6%. This value is not so far from that reported for all ERCP perforations (7.8%-9.9%), regardless of the type of lesion and mechanism of injury.[22, 23] The limitations of this analysis include certain heterogeneity among studies, especially about indications
162 • Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com
ERCP-related retroperforations
for surgery and type of procedures. Further, the lack of detailed data from most studies precluded the possibility to reliably identify possible factors affecting outcomes of both medical and surgical treatments.[6-8, 12] Based upon the available evidence from the literature, although less dreaded as compared to intraperitoneal perforations, retroperitoneal perforations should not be considered exceptional and are associated with significant rates of morbidity and mortality. Nonetheless, currently there is still no consensus on management guidelines[4,12] given that all available data come from anecdotal reports or small retrospective case series. In this respect, despite a relative difficulty of conducting such investigations in a prospective manner, more detailed analyses are warranted. Identifying high-risk factors (clinical or radiological) that may predict poor outcomes after medical management should be considered crucial in limiting the number of patients subjected to the substantial risks associated with delayed surgery.[8-10, 12, 21] Contributors: GF proposed the study. GF, GG, CD and ABS performed the research and wrote the first draft. GF and LSGB collected and analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. GF is the guarantor. Funding: None. Ethical approval: Not needed. Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
References 1 Glomsaker T, Hoff G, Kvaløy JT, Søreide K, Aabakken L, Søreide JA, et al. Patterns and predictive factors of complications after endoscopic retrograde cholangiopancreatography. Br J Surg 2013;100:373-380. 2 Jin YJ, Jeong S, Kim JH, Hwang JC, Yoo BM, Moon JH, et al. Clinical course and proposed treatment strategy for ERCP-related duodenal perforation: a multicenter analysis. Endoscopy 2013;45:806-812. 3 ASGE Standards of Practice Committee, Maple JT, Ben-Menachem T, Anderson MA, Appalaneni V, Banerjee S, et al. The role of endoscopy in the evaluation of suspected choledocholithiasis. Gastrointest Endosc 2010;71:1-9. 4 Rabie ME, Mir NH, Al Skaini MS, El Hakeem I, Hadad A, Ageely H, et al. Operative and non-operative management of endoscopic retrograde cholangiopancreatography-associated duodenal injuries. Ann R Coll Surg Engl 2013;95:285-290. 5 Alfieri S, Rosa F, Cina C, Tortorelli AP, Tringali A, Perri V, et al. Management of duodeno-pancreato-biliary perforations after ERCP: outcomes from an Italian tertiary referral center. Surg Endosc 2013;27:2005-2012. 6 Stapfer M, Selby RR, Stain SC, Katkhouda N, Parekh D, Jabbour N, et al. Management of duodenal perforation after endoscopic retrograde cholangiopancreatography and sphincterotomy. Ann Surg 2000;232:191-198.
7 Howard TJ, Tan T, Lehman GA, Sherman S, Madura JA, Fogel E, et al. Classification and management of perforations complicating endoscopic sphincterotomy. Surgery 1999;126:658-665. 8 Enns R, Eloubeidi MA, Mergener K, Jowell PS, Branch MS, Pappas TM, et al. ERCP-related perforations: risk factors and management. Endoscopy 2002;34:293-298. 9 Bell RC, Van Stiegmann G, Goff J, Reveille M, Norton L, Pearlman NW. Decision for surgical management of perforation following endoscopic sphincterotomy. Am Surg 1991;57:237-240. 10 Wu HM, Dixon E, May GR, Sutherland FR. Management of perforation after endoscopic retrograde cholangiopancreatography (ERCP): a population-based review. HPB (Oxford) 2006;8:393-399. 11 Assalia A, Suissa A, Ilivitzki A, Mahajna A, Yassin K, Hashmonai M, et al. Validity of clinical criteria in the management of endoscopic retrograde cholangiopancreatography related duodenal perforations. Arch Surg 2007;142:1059-1064. 12 Avgerinos DV, Llaguna OH, Lo AY, Voli J, Leitman IM. Management of endoscopic retrograde cholangiopancreatography: related duodenal perforations. Surg Endosc 2009;23:833-838. 13 Li G, Chen Y, Zhou X, Lv N. Early management experience of perforation after ERCP. Gastroenterol Res Pract 2012;2012: 657418. 14 Morgan KA, Fontenot BB, Ruddy JM, Mickey S, Adams DB. Endoscopic retrograde cholangiopancreatography gut perforations: when to wait! When to operate! Am Surg 2009;75:477-484. 15 Salminen P, Laine S, Gullichsen R. Severe and fatal complications after ERCP: analysis of 2555 procedures in a single experienced center. Surg Endosc 2008;22:1965-1970. 16 Neri V, Ambrosi A, Fersini A, Valentino TP. Duodenal perforation in course of endoscopic retrograde cholangiopancreatography-endoscopic sphincterotomy. Therapeutic considerations. Ann Ital Chir 2006;77:161-164. 17 Ercan M, Bostanci EB, Dalgic T, Karaman K, Ozogul YB, Ozer I, et al. Surgical outcome of patients with perforation after endoscopic retrograde cholangiopancreatography. J Laparoendosc Adv Surg Tech A 2012;22:371-377. 18 Krishna RP, Singh RK, Behari A, Kumar A, Saxena R, Kapoor VK. Post-endoscopic retrograde cholangiopancreatography perforation managed by surgery or percutaneous drainage. Surg Today 2011;41:660-666. 19 Kim JH, Yoo BM, Kim JH, Kim MW, Kim WH. Management of ERCP-related perforations: outcomes of single institution in Korea. J Gastrointest Surg 2009;13:728-734. 20 Silviera ML, Seamon MJ, Porshinsky B, Prosciak MP, Doraiswamy VA, Wang CF, et al. Complications related to endoscopic retrograde cholangiopancreatography: a comprehensive clinical review. J Gastrointestin Liver Dis 2009;18:73-82. 21 Kumbhari V, Sinha A, Reddy A, Afghani E, Cotsalas D, Patel YA, et al. Algorithm for the management of ERCP-related perforations. Gastrointest Endosc 2016;83:934-943. 22 Vezakis A, Fragulidis G, Polydorou A. Endoscopic retrograde cholangiopancreatography-related perforations: diagnosis and management. World J Gastrointest Endosc 2015;7:1135-1141. 23 Andriulli A, Loperfido S, Napolitano G, Niro G, Valvano MR, Spirito F, et al. Incidence rates of post-ERCP complications: a systematic survey of prospective studies. Am J Gastroenterol 2007;102:1781-1788. Received October 9, 2015 Accepted after revision March 3, 2016
Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com • 163
Hepatobiliary & Pancreatic Diseases International
Clinical outcomes of ERCP-related retroperitoneal perforations Francesco Guerra, Giuseppe Giuliani, Diego Coletta, Stefano Amore Bonapasta and Giovanni Battista Levi Sandri Rome, Italy
ABSTRACT: Endoscopic retrograde cholangiopancreatography (ERCP)-related perforations represent rare but often severe conditions. While lesions with intraperitoneal perforation have an almost imperative indication to surgery, whether or not to manage retroperitoneal perforations surgically is still an area of debate. The aim of the present work was to review the available clinical evidence on the operatively and medically treated ERCP-related retroperitoneal perforations. From MEDLINE/PubMed databases 137 patients with retroperitoneal perforation were included from 12 studies that met the selection criteria for data investigation and analysis. Twentyfour patients were treated by prompt surgery; 113 were primarily managed conservatively and about 20% of these patients required surgery subsequently. Overall, the morbidity and mortality were 15.4% and 6.6%, respectively. Although most patients with retroperitoneal perforation may benefit from a non-operative management, a considerable number of patients fail to respond to medical treatment and require surgery afterwards. Identifying those patients who are at highest risk of poor outcome after conservative treatment should be considered a research priority.
Introduction
T
he rates of adverse events sustained during endoscopic retrograde cholangiopancreatography (ERCP) vary widely in the scientific literature (approximately from 4% to 10%), and the true incidence in clinical practice is still unclear.[1] As a major event predicting increased morbidity and mortality, perforating iatrogenic lesions have an incidence of 0.1% to 2.1%.[2] This specific risk is substantively higher in therapeutic procedures than in diagnostic ones, and operative ERCPs are increasingly performed.[2, 3] The key point of ERCP-related perforations is to decide which patient can be managed conservatively and who should be promptly operated on. Nevertheless, there is no consensus on the treatment for patients with no free abdominal air.[2-5] Despite many classifications have been proposed to predict the prognosis and to address adequate treatment,[6, 7] neither the optimal management (Hepatobiliary Pancreat Dis Int 2017;16:160-163) nor clinical or radiological findings that could guide the decision-making process have yet been defined. KEY WORDS: endoscopic retrograde cholangiopancreatography; Over the last years, several studies[6-9] have suggested complications; that patients with iatrogenic ERCP-related retroperitoretroperitoneum neal perforation can recover without surgery. Nevertheless, surgical management is still advocated by a number of authors[5] and it is widely known that delayed surgery can be associated with dismal outcomes.[8, 9] Disagreement regarding how patients with such condition should be managed therefore exists. Accordingly, this study was undertaken to review the available evidence from the literature to analyze the clinical outcomes of patients manAuthor Affiliations: Department of Surgery, Umberto I University Hospital, Rome, Italy (Guerra F, Giuliani G, Coletta D, Amore Bonapasta S aged surgically or conservatively after a retroperitoneal perforating lesion occurred during an ERCP procedure. and Levi Sandri GB) Corresponding Author: Francesco Guerra, MD, Department of Surgery, Umberto I University Hospital-viale del Policlinico 155 00161, Rome, Italy (Tel: +39-333-3555-235; Email: [email protected]) © 2017, Hepatobiliary Pancreat Dis Int. All rights reserved. doi: 10.1016/S1499-3872(16)60106-6 Published online June 21, 2016.
Methods An initial electronic search was performed to identify all studies dealing with ERCP-related perforations, without
160 • Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com
ERCP-related retroperforations
discrimination between intra- and retroperitoneal lesions. The PubMed/MEDLINE database was searched in December 2014. The search strategy was (“perforation”) AND (“ERCP” OR “cholangiography” OR “cholangiopancreatography”). When inclusion was unclear based on abstracts, full papers were obtained. Hand searches of relevant articles’ references were also performed to find possible additional works. Inclusion criteria accepted all articles in English language, from 2004 onwards, which presented data concerning treatment of ERCP-related perforations. Abstracts and unpublished data as well as small series involving less than five patients and case reports were excluded from the analysis. Were included: 1. All patients, regardless of age or gender, in whom an ERCP-related retroperitoneal perforation was documented and whose management was described in detail; 2. Patients with altered anatomy, such as those with Billroth II or Rouxen-Y reconstruction only if a retroperitoneal perforation was diagnosed. Two reviewers (GF and GG) carried out the investigation independently. From the studies that were included, data for meta-analysis were extrapolated following a pre-established pattern. Results from extrapolation were then submitted to other two authors (CD and ABS) to be verified. All disagreements were resolved by consensus, involving all authors. Patients were assorted into an operative treatment group (OT) when they were primarily treated by surgery (within 24 hours from the diagnosis). In contrast, patients were assorted in a non-operative treatment (including the need for percutaneous drainage) group (NOT) when a conservative therapy was initially adopted, regardless of the need for subsequent surgical intervention. Treatment modalities and clinical outcomes such as overall morbidity, mortality and length of hospital stay of both operative and non-operative treatments were assessed. Extemporaneous attempts of mucosal closures when the perforation was diagnosed at the time of endoscopy were not considered in the analysis. Statistical analysis was performed using the Statistical Package for the Social Sciences, v20.0 (SPSS Inc., Chicago, IL, USA). Results were presented in descriptive statistics. To determine if homogeneous samples were different between medically and surgically treated patients, univariate analysis was performed using Student’s t test or Fisher’s exact test for continuous and categorical variables, respectively. A two-tailed P value of 0.05 or less was deemed statistically significant.
Table 1. Summary characteristics of the included studies Studies Rabie et al[4] Alfieri et al[5] Wu et al[10] Assalia et al[11] Avgerinos et al[12 ] Li et al[13] Morgan et al[14] Salminen et al[15] Neri et al[16] Ercan et al[17] Krishna et al[18] Kim et al[19] Total
Year of publi- Country cation 2013 2013 2006 2007 2009 2012 2009 2008 2006 2012 2011 2009
Saudi Arabia Italy Canada Israel USA China USA Finlandia Italy Turkey India Korea
ERCPs
Perfora- Retropertions forations
597 14 618 6 620 3 104 4 358 8 504 12 817 2 788 101 9 209 2 247 64 963
10 30 30 22 15 16 24 5 5 52 14 18 241
9 27 28 19 1 9 12 2 5 8 8 9 137
After the evaluation of abstracts, full-texts and references, 12 studies[4, 5, 10-19] with a total of 137 retro-perforated patients (out of 241 perforation from a total of 64 963 ERCPs included) met all the inclusion criteria. General characteristics of studies and patients are summarized in Table 1. One of the major reasons for exclusion of studies has been the inability to extrapolate data specific to single patients. The percentages were often referred to the total of patients treated, with impossibility to extrapolate data fractionated according to the type of perforation, therapeutic approach, incidence and type of complications. Overall, 24 patients were treated operatively, whereas 113 were primarily managed by non-operative treatment. Twenty-one (18.6%) patients failed to non-operative treatment and required surgery subsequently. These patients proceeded to surgery because of one or more of the following features: persistent fever despite adequate antibiotic therapy, abscess or collection development believed to require direct drainage or ongoing or worsening symptoms. Overall, non-operative management consisted of nihil per os, nasoduodenal/nasobiliary drainage, intravenous fluids, and antibiotic therapy with or without percutaneous drainage. In the OT group, primary surgery consisted of a combination of the following: direct repair/debridement,[4, 5, 11, 15-17] common bile duct exploration and T-tube placement,[4, 11, 15-17] debridement via a retroperitoneal laparostomy,[5] and duodenostomy/jejunostomy[4, 16] or gastrojejunostomy.[16] Information on incidence of complications was reported in detail in 11 out of 12 studies. Sepsis (5 patients, 3 in the OT group), acute pancreatitis (3 patients, all in Results the NOT group), retroperitoneal abscess (4 patients, 3 A total of 151 potentially relevant articles were identified. in the NOT group) and duodenal leaks (3 patients, all in Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com • 161
Hepatobiliary & Pancreatic Diseases International
Table 2. Clinical outcomes of ERCP-related retroperitoneal perforations Morbidity (n, %)
Mortality (n, %)
Groups n
Hospital stay (d)
OT
24
20.3 (data from 21 patients)
5 (22.7) (data from 22 patients)
5 (20.8)
NOT
113
21.1 (data from 94 patients)
12 (13.6) (data from 88 patients)
4 (3.5)
Total
137
20.9 (data from 115 17 (15.4%) (data from 9 (6.6) patients) 110 patients)
Values are reported as weighted means (percentages in parentheses). OT: operative treatment; NOT: non-operative treatment.
the NOT group, who subsequently underwent surgery) were the most frequently observed morbidities. Overall, 17 patients experienced complications. By analyzing only those studies whose specific data were provided appropriately, the total morbidity rate was 15.4%. In the OT group, the overall incidence of complications was 22.7% whereas it was 13.6% in the NOT group. This difference did not elicit statistical significance (P=0.43). Mean postoperative hospital stays were 21.1±11.6 days for the NOT group (weighted data from 10 studies including 94 patients) and 20.3±6.5 days in the OT group (6 studies, 21 patients) with no statistically significant difference between the two groups (P=0.76). The overall mortality of the two groups was 6.6%. Particularly, in the OT group, the overall mortality was 20.8%, whereas in the NOT group it was 3.5% (P=0.01). Data concerning clinical outcomes are shown in Table 2. According to those studies that provided detailed data, 80% of the patients who eventually died had delayed diagnosis of perforation. Notably, of the 5 patients who received more than one surgical procedure, 4 succumbed.
Discussion Iatrogenic perforation resulting from ERCP is a rare but serious complication with high morbidity and mortality.[1-4, 6] The type of perforation can be different during therapeutic ERCPs and many scholars have been investigating specific features in terms of localization and mechanism of injury advocating possible correlation with therapeutic modalities.[2, 6, 7] Despite this, the problem of choosing the appropriate treatment for patients with perforations occurred within an ERCP procedure is often encountered in front of clinical or radiological evidences, rather than a given endoscopic occurrence. This is particularly the case when a radiological evidence of a retroperitoneal perforation is observed on CT that is mostly required due to clinical suspicions after an endo-
scopic procedure. Actually, there is a relatively imperative indication for surgical exploration in the presence of a recognized intraperitoneal perforation.[7, 8, 20] Conversely, decision-making regarding non-operative versus operative treatment of patients with retroperitoneal perforations can be contentious. Indeed, while several authors[8, 11, 20] intimate that clinical condition should be the main factor determining operative versus conservative management, others[6, 9, 21] have reported that clinical findings are unreliable in predicting the need for surgery. Similarly, other factors such as CT evidence of retroperitoneal fluid collection[6, 10] or contrast extravasation[4] have been considered dependable indications for immediate surgery by some, even though several authors agree on the fact that such features alone should not guide the surgeon’s decision-making process.[7, 11, 12] In fact, the early and accurate detection of patients with ERCP-related retroperitoneal perforation who are at the highest risk for a poor outcome remains a clinical challenge.[6, 8-10] Accordingly, we aimed to identify, summarize, and combine the published literature about the operative and non-operative treatment of such conditions. To the best of our knowledge, this is the first summary specifically focusing on clinical outcomes and associated clinical practice in the management of ERCP-related retroperforations and to establish current incidence and relative morbidity and mortality. In the 64 963 procedures analyzed, 137 (0.2%) patients experienced clinically relevant retroperitoneal perforation. Although the methods used to define an “early recognition” of ERCP-related retroperforation were not homogeneous among studies, delayed diagnosis was almost invariably associated with longer hospitalization, increased morbidity and mortality. These data are consistent with most findings from the literature concerning all types of perforation.[2, 21, 22] Overall, 17.5% of patients received prompt surgical treatment whereas for most patients (82.5%) medical treatment alone was primarily judged to be adequate.[6, 21] Nevertheless, about 20% of patients receiving medical treatment required surgery subsequently, resulting in increased morbidity and mortality. Interestingly, the need for surgery after failure of initial medical management was found in about 21% of patients with sphincterotomy related perforations (type II according to Stapfer et al[6]) in a recent review including 11 studies and more than 500 perforations. In total, nine fatalities were registered in our review, delineating an overall mortality rate of 6.6%. This value is not so far from that reported for all ERCP perforations (7.8%-9.9%), regardless of the type of lesion and mechanism of injury.[22, 23] The limitations of this analysis include certain heterogeneity among studies, especially about indications
162 • Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com
ERCP-related retroperforations
for surgery and type of procedures. Further, the lack of detailed data from most studies precluded the possibility to reliably identify possible factors affecting outcomes of both medical and surgical treatments.[6-8, 12] Based upon the available evidence from the literature, although less dreaded as compared to intraperitoneal perforations, retroperitoneal perforations should not be considered exceptional and are associated with significant rates of morbidity and mortality. Nonetheless, currently there is still no consensus on management guidelines[4,12] given that all available data come from anecdotal reports or small retrospective case series. In this respect, despite a relative difficulty of conducting such investigations in a prospective manner, more detailed analyses are warranted. Identifying high-risk factors (clinical or radiological) that may predict poor outcomes after medical management should be considered crucial in limiting the number of patients subjected to the substantial risks associated with delayed surgery.[8-10, 12, 21] Contributors: GF proposed the study. GF, GG, CD and ABS performed the research and wrote the first draft. GF and LSGB collected and analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. GF is the guarantor. Funding: None. Ethical approval: Not needed. Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
References 1 Glomsaker T, Hoff G, Kvaløy JT, Søreide K, Aabakken L, Søreide JA, et al. Patterns and predictive factors of complications after endoscopic retrograde cholangiopancreatography. Br J Surg 2013;100:373-380. 2 Jin YJ, Jeong S, Kim JH, Hwang JC, Yoo BM, Moon JH, et al. Clinical course and proposed treatment strategy for ERCP-related duodenal perforation: a multicenter analysis. Endoscopy 2013;45:806-812. 3 ASGE Standards of Practice Committee, Maple JT, Ben-Menachem T, Anderson MA, Appalaneni V, Banerjee S, et al. The role of endoscopy in the evaluation of suspected choledocholithiasis. Gastrointest Endosc 2010;71:1-9. 4 Rabie ME, Mir NH, Al Skaini MS, El Hakeem I, Hadad A, Ageely H, et al. Operative and non-operative management of endoscopic retrograde cholangiopancreatography-associated duodenal injuries. Ann R Coll Surg Engl 2013;95:285-290. 5 Alfieri S, Rosa F, Cina C, Tortorelli AP, Tringali A, Perri V, et al. Management of duodeno-pancreato-biliary perforations after ERCP: outcomes from an Italian tertiary referral center. Surg Endosc 2013;27:2005-2012. 6 Stapfer M, Selby RR, Stain SC, Katkhouda N, Parekh D, Jabbour N, et al. Management of duodenal perforation after endoscopic retrograde cholangiopancreatography and sphincterotomy. Ann Surg 2000;232:191-198.
7 Howard TJ, Tan T, Lehman GA, Sherman S, Madura JA, Fogel E, et al. Classification and management of perforations complicating endoscopic sphincterotomy. Surgery 1999;126:658-665. 8 Enns R, Eloubeidi MA, Mergener K, Jowell PS, Branch MS, Pappas TM, et al. ERCP-related perforations: risk factors and management. Endoscopy 2002;34:293-298. 9 Bell RC, Van Stiegmann G, Goff J, Reveille M, Norton L, Pearlman NW. Decision for surgical management of perforation following endoscopic sphincterotomy. Am Surg 1991;57:237-240. 10 Wu HM, Dixon E, May GR, Sutherland FR. Management of perforation after endoscopic retrograde cholangiopancreatography (ERCP): a population-based review. HPB (Oxford) 2006;8:393-399. 11 Assalia A, Suissa A, Ilivitzki A, Mahajna A, Yassin K, Hashmonai M, et al. Validity of clinical criteria in the management of endoscopic retrograde cholangiopancreatography related duodenal perforations. Arch Surg 2007;142:1059-1064. 12 Avgerinos DV, Llaguna OH, Lo AY, Voli J, Leitman IM. Management of endoscopic retrograde cholangiopancreatography: related duodenal perforations. Surg Endosc 2009;23:833-838. 13 Li G, Chen Y, Zhou X, Lv N. Early management experience of perforation after ERCP. Gastroenterol Res Pract 2012;2012: 657418. 14 Morgan KA, Fontenot BB, Ruddy JM, Mickey S, Adams DB. Endoscopic retrograde cholangiopancreatography gut perforations: when to wait! When to operate! Am Surg 2009;75:477-484. 15 Salminen P, Laine S, Gullichsen R. Severe and fatal complications after ERCP: analysis of 2555 procedures in a single experienced center. Surg Endosc 2008;22:1965-1970. 16 Neri V, Ambrosi A, Fersini A, Valentino TP. Duodenal perforation in course of endoscopic retrograde cholangiopancreatography-endoscopic sphincterotomy. Therapeutic considerations. Ann Ital Chir 2006;77:161-164. 17 Ercan M, Bostanci EB, Dalgic T, Karaman K, Ozogul YB, Ozer I, et al. Surgical outcome of patients with perforation after endoscopic retrograde cholangiopancreatography. J Laparoendosc Adv Surg Tech A 2012;22:371-377. 18 Krishna RP, Singh RK, Behari A, Kumar A, Saxena R, Kapoor VK. Post-endoscopic retrograde cholangiopancreatography perforation managed by surgery or percutaneous drainage. Surg Today 2011;41:660-666. 19 Kim JH, Yoo BM, Kim JH, Kim MW, Kim WH. Management of ERCP-related perforations: outcomes of single institution in Korea. J Gastrointest Surg 2009;13:728-734. 20 Silviera ML, Seamon MJ, Porshinsky B, Prosciak MP, Doraiswamy VA, Wang CF, et al. Complications related to endoscopic retrograde cholangiopancreatography: a comprehensive clinical review. J Gastrointestin Liver Dis 2009;18:73-82. 21 Kumbhari V, Sinha A, Reddy A, Afghani E, Cotsalas D, Patel YA, et al. Algorithm for the management of ERCP-related perforations. Gastrointest Endosc 2016;83:934-943. 22 Vezakis A, Fragulidis G, Polydorou A. Endoscopic retrograde cholangiopancreatography-related perforations: diagnosis and management. World J Gastrointest Endosc 2015;7:1135-1141. 23 Andriulli A, Loperfido S, Napolitano G, Niro G, Valvano MR, Spirito F, et al. Incidence rates of post-ERCP complications: a systematic survey of prospective studies. Am J Gastroenterol 2007;102:1781-1788. Received October 9, 2015 Accepted after revision March 3, 2016
Hepatobiliary Pancreat Dis Int，Vol 16，No 2 • April 15，2017 • www.hbpdint.com • 163