Systematic review of percutaneous catheter drainage as primary treatment for necrotizing pancreatitis

Transcription

1 Systematic review Systematic review of percutaneous catheter drainage as primary treatment for necrotizing pancreatitis M. C. van Baal 1,H.C.vanSantvoort 1,T.L.Bollen 2,O.J.Bakker 1,M.G.Besselink 1 and H. G. Gooszen 3 for the Dutch Pancreatitis Study Group 1 Department of Surgery, University Medical Centre Utrecht, Utrecht, 2 Department of Radiology, St Antonius Hospital, Nieuwegein, and 3 Department of Operation Room/Evidence Based Surgery, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands Correspondence to: Professor H. G. Gooszen, Department of Operation Room/Evidence Based Surgery, Radboud University Nijmegen Medical Centre, HP 630, PO Box 9101, 6500 MB Nijmegen, The Netherlands ( h.gooszen@ok.umcn.nl) Background: The role of percutaneous catheter drainage (PCD) in patients with (infected) necrotizing pancreatitis was evaluated. Methods: A systematic literature search was performed. Inclusion criteria were: consecutive cohort of patients with necrotizing pancreatitis undergoing PCD as primary treatment for peripancreatic collections; indication for PCD either (suspected) infected necrosis or symptomatic sterile pancreatic necrosis; and outcomes reported to include percentage of infected peripancreatic collections, need for additional surgical necrosectomy, complications and deaths. Exclusion criteria were: cohort of fewer than five patients; cohort included patients with chronic pancreatitis; selected subgroup of patients with acute pancreatitis studied, such as those with pseudocysts, pancreatic abscesses and/or exclusively sterile pancreatic necrosis; and cohort in which PCD was combined with another minimally invasive strategy and results for PCD alone not reported separately. Results: Eleven studies, including 384 patients, fulfilled the inclusion criteria. Only one study was a randomized controlled trial; most others were retrospective case series. Four studies reported on the presence of organ failure before PCD; this occurred in 67 2 per cent of 116 patients. Infected necrosis was proven in 271 (70 6 per cent) of 384 patients. No additional surgical necrosectomy was required after PCD in 214 (55 7 per cent) of 384 patients. Complications consisted mostly of internal and external pancreatic fistulas. The overall mortality rate was 17 4 per cent (67 of 384 patients). Nine of 11 studies reported mortality separately for patients with infected necrosis undergoing PCD; the mortality rate in this group was 15 4 per cent (27 of 175). Conclusion: A considerable number of patients can be treated with PCD without the need for surgical necrosectomy. Presented to the Annual Meeting of the American Pancreatic Association, Honolulu, Hawaii, USA, November 2009 Paper accepted 26 August 2010 Published online in Wiley Online Library ( DOI: /bjs.7304 Introduction Around 20 per cent of patients with acute pancreatitis develop necrosis of the pancreas or peripancreatic fat tissue with associated peripancreatic collections 1,2. Sterile necrosis can generally be managed conservatively and the mortality rate is relatively low (12 per cent) 1,3. Approximately 30 (range 14 62) per cent of patients with necrotizing pancreatitis, however, develop secondary infection of peripancreatic collections, which is associated with sepsis and organ failure and is an indication for intervention 1. Until recently, the first-choice intervention in infected necrotizing pancreatitis has been surgical necrosectomy by laparotomy with the aim of removing all infected necrosis 4 6. This approach is associated with considerable morbidity (34 95 per cent) and mortality (11 39 per cent) 1,7 9. Some patients with sterile necrosis also eventually undergo surgical necrosectomy in the event of clinical deterioration (multiple organ failure) 2010 British Journal of Surgery Society Ltd British Journal of Surgery 2011; 98: 18 27

2 Percutaneous drainage in necrotizing pancreatitis 19 despite maximal supportive therapy when infection is suspected. Others undergo necrosectomy because of persistent symptomatic external hepatobiliary or duodenal compression by peripancreatic collections 5. In 1998, Freeny and colleagues 10 first described a consecutive series of patients exclusively with infected pancreatic necrosis who were treated primarily with imaging-guided percutaneous catheter drainage (PCD), as an alternative to primary surgical necrosectomy. The rationale for PCD was to temporize sepsis and thereby postpone the need for surgical necrosectomy. In their retrospective cohort study, PCD was successful in postponing surgical intervention for a median of 4 weeks and even obviated the need for surgical necrosectomy in almost half of the patients 10. In addition, PCD seems technically feasible in the vast majority of patients with necrotizing pancreatitis 11. In the past decade, several cohort studies on PCD and minimally invasive necrosectomy, such as percutaneous necrosectomy 12, video-assisted retroperitoneal debridement (VARD) 13,14 and transluminal endoscopic necrosectomy 15, have been published. Recently, a minimally invasive step-up approach, consisting of PCD as first step followed by VARD if necessary, has proven to be more effective than primary open necrosectomy in a randomized trial (PAncreatitis, Necrosectomy versus step up approach, PANTER) 16. More than a third of patients with infected necrotizing pancreatitis were treated successfully with PCD alone. This systematic review focused on PCD as primary treatment for (infected) necrotizing pancreatitis. The primary aim was to determine the proportion of patients that can be treated with PCD without the need for additional necrosectomy. Methods The MEDLINE and Embase search terms were (radiologic OR percutaneous OR drainage) AND pancreatitis. Search terms for the Cochrane Library were: pancreatitis AND (radiologic OR percutaneous OR drainage), restricted to title, abstract, keywords and the English language. All titles and abstracts of studies identified by the initial search were screened to select those reporting on patients undergoing PCD of peripancreatic collections associated with pancreatitis. Subsequently, full-text papers of the selected studies were screened independently by three authors to assess eligibility. Inclusion criteria were: a consecutive cohort of patients with acute necrotizing pancreatitis undergoing PCD as a primary treatment for peripancreatic collections; indication for PCD (suspected) either infected necrosis or symptomatic sterile pancreatic necrosis (such as clinical deterioration or significant mechanical obstruction); and essential outcomes reported, including percentage of infected peripancreatic collections, need for additional surgical necrosectomy, complications and mortality. Exclusion criteria were: very small cohort (fewer than 5 patients); cohort included chronic pancreatitis (and results for acute pancreatitis not reported separately); studies on a selected subgroup of patients with acute pancreatitis, classified as pseudocysts or pancreatic abscesses (as defined by the Atlanta classification) or sterile pancreatic necrosis exclusively; and cohorts of patients undergoing minimally invasive surgical necrosectomy including previous PCD and results for PCD alone not reported separately. Potentially relevant articles identified n = 4670 Embase n = 3225 MEDLINE n = 1386 The Cochrane Library n = 59 Study selection A systematic literature search from 1 January 1992 to 31 May 2010 was performed in Embase, MEDLINE and the Cochrane Library. The search was limited to this interval because no universally accepted definitions for acute pancreatitis and pancreatic collections were available before 1992, confounding the comparison of studies on PCD. In 1992, the Atlanta symposium provided definitions for acute pancreatitis and local complications such as pancreatic necrosis, pseudocysts and pancreatic abscesses 17. Although it is now recognized that the Atlanta classification has considerable shortcomings and is currently under revision 18,19, the definitions have been used widely in the literature since Studies that met inclusion criteria based on title and abstract n = 34 Studies included in this systematic review n = 11 Fig. 1 Study selection Excluded after reviewing title and abstract and removing duplicates n = 4636 Studies excluded after reviewing full-text articles n = 23

3 20 M. C. van Baal, H. C. van Santvoort, T. L. Bollen, O. J. Bakker, M. G. Besselink, H. G. Gooszen All cross-references were screened for potentially relevant studies not identified by the initial literature search. The final decision on eligibility was reached by consensus. Data extraction The following variables were extracted, where available, from the included articles: number of patients with acute necrotizing pancreatitis undergoing PCD, aetiology, predictive severity scores before PCD (such as Ranson score 20, Acute Physiology And Chronic Health Evaluation II score 21 ), percentage of patients with organ failure, computed tomography (CT) severity scores (CT Severity Index (CTSI) 22, modified CTSI 23, Balthazar grade 24 ), percentage of patients with infected necrosis, indication for PCD, time between hospital admission and PCD, drain size, total number of drains placed, time to removal Table 1 Characteristics of included studies Reference Year Country Study design Study interval Inclusion criteria Technique used Freeny et al USA Retrospective non-controlled case series Gambiez 1998 France Retrospective non-controlled et al. 48 case series Medically uncontrolled sepsis CT-guided drainage, Fr drains Secondary outbreak or persistence (> 7 days) of signs of sepsis unexplained by a source of infection other than abdominal abscess Organ failure despite medical treatment Bacterial proof of infection CT-guided drainage Fotoohi et al USA Retrospective Sepsis CT-guided drainage, non-controlled case Pain Seldinger/trocar series Biliary obstruction technique, 8 24-Fr drains Baril et al USA Retrospective non-controlled case series (Suspected) pancreatic or peripancreatic sepsis CT-guided drainage, Seldinger technique, Fr drains Cheung et al Hong Kong Retrospective Deteriorated clinical condition CT-guided drainage, 20-Fr non-controlled case Symptomatic pancreatic drains series collections Navalho 2006 Portugal Retrospective non-controlled et al. 52 case series Lee et al Korea Prospective non-controlled case series Bruennler 2008 Germany Retrospective non-controlled et al. 54 case series Infected pancreatic necrosis Ultrasound/CT-guided drainage, trocar technique, Fr drains Infected pancreatic necrosis CT-guided drainage, 14-Fr drains dilated stepwise to 20 Fr Infected pancreatic necrosis CT-guided drainage, Seldinger technique Mortelé et al USA Retrospective NR Infected pancreatic necrosis CT-guided drainage, non-controlled case Suspicion of pancreatic necrosis, Seldinger/trocar series raised by fever, raised WBC technique, up to 14-Fr count, or general clinical drains deterioration refractory to standard medical care Rocha et al USA Retrospective non-controlled case series NR NR Van Santvoort 2010 The Netherlands RCT Infected pancreatic necrosis Ultrasound/CT-guided, et al. 16 Persistent sepsis or progressive Seldinger technique, clinical deterioration despite minimum drain size maximal support in the ICU, 12 Fr without documentation of infected necrosis CT, computed tomography; NR, not reported; WBC, white blood cell; RCT, randomized controlled trial; ICU, intensive care unit.

4 Percutaneous drainage in necrotizing pancreatitis 21 of drains, success of PCD (defined as survival without the need for additional surgical necrosectomy), total hospital stay, complications and mortality. Results The results of the literature search are summarized in Fig. 1. Of 34 papers reporting on PCD of peripancreatic collections associated with pancreatitis, 23 were excluded for the following reasons: cohort of fewer than five patients (1 study) 25, included patients with pseudocysts (1) 26, pancreatic abscesses (3) or both (1) 30, included patients with sterile pancreatic necrosis only (2) 31,32, cohorts of mixed chronic and acute pancreatitis with outcomes not reported separately (1) 33, one or more essential outcomes not reported (6) 34 39, PCD in combination with transgastric drainage (2) 40,41, and primary PCD was part of a step-up approach in which PCD was always followed by minimally invasive surgical necrosectomy (6) The remaining 11 studies were included in the present systematic review 10,16, Nine studies were retrospective non-controlled case series 10,48,50 56 (Oxford level 4 evidence 57 ), one was a prospective non-controlled case series 53 (Oxford level 4) and one was a multicentre RCT 16 (Oxford level 1b). Characteristics of the 11 included studies are summarized in Table 1. Patient characteristics The 11 studies included a total of 384 patients undergoing PCD as primary treatment for (suspected) infected necrosis or symptomatic sterile pancreatic necrosis; the number of patients per study ranged from 8 to 80. Patient characteristics are summarized in Table 2. Table 2 Characteristics of patients in included studies Reference No. of patients Aetiology Predictive severity scores * Organ failure CT severity score * Infected (peri)pancreatic collections Freeny et al Alcoholic 7 NR NR CTSI: mean (100) Biliary 12 Other 15 Gambiez et al NR Ranson: mean (50) Balthazar: 3 (30) Grade D 16 Grade E 37 Fotoohi et al Alcoholic 20 NR NR NR 44 (73) Biliary 6 Other 34 Baril et al NR NR NR NR 25 (66) Cheung et al Alcoholic 1 Ranson: mean 5 9 (3 9) NR NR 4 (50) Biliary 5 Other 2 Navalho et al Alcoholic 9 NR NR NR 30 (100) Biliary 17 Other 4 Lee et al NR NR NR NR 18 (100) Bruennler et al Alcoholic 32 Ranson: median 2 (0 4) 65 (81) in ICU CTSI: median 6 (4 10) 52 (65) Biliary 26 APACHE: median 18 (1 38) Other 22 Mortelé et al Alcoholic 8 Atlanta: mean 0 92 (0 3) 16 (46) Modified CTSI: mean 13 (37) Biliary (8 10) Other 15 Rocha et al NR NR 21 (75) NR 9 (32) Van Santvoort et al Alcoholic 3 APACHE: mean (84) CTSI: median 8 (4 10) 39 (91) Biliary 26 MODS: median 2 (0 9) Other 14 SOFA: median 3 (0 11) Values in parentheses are *ranges and percentages. CT, computed tomography; NR, not reported; CTSI, CT Severity Index 22 ; Ranson, severity score for acute pancreatitis 20 ; Balthazar, CT severity score for acute pancreatitis 24 ; APACHE, Acute Physiology And Chronic Health Evaluation 21 ;ICU, intensive care unit; Atlanta, severity score for acute pancreatitis 17 ; modified CTSI, modified CT Severity Index 23 ; MODS, Multiple Organ Dysfunction Score 58 ; SOFA, Sequential Organ Failure Assessment 59.

5 22 M. C. van Baal, H. C. van Santvoort, T. L. Bollen, O. J. Bakker, M. G. Besselink, H. G. Gooszen The aetiology of acute necrotizing pancreatitis was reported in seven studies, and five reported clinical details and CT severity (Table 2). Various predictive severity scores and CT severity scores were used in the 11 studies. Four studies (116 patients) 16,48,55,56 reported the percentage of patients suffering from organ failure before PCD. Seventy-eight (67 2 per cent) of 116 patients had organ failure (34 patients with single and 44 with multiple organ failure) before drainage. Of the total of 384 patients, 271 (70 6 per cent) had infected peripancreatic necrosis, defined as the presence of gas in the peripancreatic collection on CT or as a positive culture at fine-needle aspiration in all studies. For PCD both Seldinger and tandem trocar techniques were used, and most of the radiological interventions were CT guided. The size of the drains used ranged from 8 to 28 Fr (2 7 to 9 3 mm). Outcome Table 3 shows outcomes as reported in the studies. Whenever possible, outcomes for infected and sterile necrosis are presented separately. The success rate of PCD, defined as the percentage of patients surviving without additional surgical necrosectomy, was 55 7 per cent (214 of 384). Eight studies 10,16,48,50 53,55 reported specific data on infected necrosis: 87 (52 4 per cent) of 166 patients recovered after PCD alone. Five series 10,48,49,52,54 reported on the time between insertion and removal of drains, which ranged from 16 to 98 days. Additional surgical necrosectomy was needed in 133 (34 6 per cent) of 384 patients. The remaining 37 patients (9 6 per cent) were considered unfit for surgery or died before necrosectomy could be performed. The average (mean or median) time interval between first PCD and surgery was reported in six series 10,16,51,52,55,56 and ranged from 18 to 109 days 10,48,49,52,54. In seven studies 10,16,48,49,52,54,55 Table 3 Outcome of percutaneous catheter drainage as primary treatment for necrotizing pancreatitis in the included studies Reference No. of patients Time from admission until PCD (days)* Successful PCD Need for additional surgery Time between PCD and necrosectomy (days)* Patients with one or more complications Deaths Freeny et al Mean 9 (1 48) 16 (47) 18 (53) Mean 32 (6 78) 9 (26) 4 (12) All IPN All IPN All IPN Gambiez et al Mean 17 (10 25) 3 (30) 7 (70) NR 6 (60) 2 (20) IPN 0 of 3 IPN 3 of 3 IPN 2 of 3 SPN 3 of 7 SPN 4 of 7 SPN 0 of 7 Fotoohi et al NR 54 (90) 3 (5) NR 6 (10) 3 (5) Baril et al NR 30 (79) 7 (18) NR 1 (3) 2 (5) IPN 18 of 25 IPN 6 of 25 IPN 2 of 25 SPN 12 of 13 SPN 1 of 13 SPN 0 of 13 Cheung et al Mean 55 (21 154) 3 (38) 5 (63) Mean 70 (1 161) 4 (50) 1 (13) IPN 30 IPN 1 of 4 IPN 3 of 4 IPN 59 IPN 1 of 4 SPN 81 SPN 2 of 4 SPN 2 of 4 SPN 88 SPN 0 of 4 Navalho et al Mean (63) 10 (33) Mean 18 NR 5 (17) All IPN All IPN All IPN Lee et al Median 10 (1 58) 14 (78) 3 (17) NR 2 (11) 1 (6) All IPN All IPN All IPN Bruennler et al Median 3 5 (1 40) 38 (48) 24 (30) NR 23 (29) 27 (34) Mortelé et al Mean 11 (2 33) 17 (49) 13 (37) Mean 69 (3 445) 4 (11) 6 (17) IPN 12 IPN 6 of 13 IPN 7 of 13 IPN 42 IPN 1 of 13 SPN 10 SPN 11 of 22 SPN 6 of 22 SPN 101 SPN 5 of 22 Rocha et al NR 5 (18) 17 (61) Median 109 (1 600) 3 (11) 8 (29) IPN 4 of 9 SPN 4 of 19 Van Santvoort et al Median 30 (11 71) 15 (35) 26 (60) Median 10 (1 52) 17 (40) 8 (19) IPN 13 of 39 IPN 25 of 39 IPN 7 of 39 SPN 2 of 4 SPN 1 of 4 SPN 1 of 4 Values in parentheses are *ranges and percentages. PCD, percutaneous catheter drainage; IPN, infected pancreatic necrosis; SPN, sterile pancreatic necrosis; NR, not reported.

6 Percutaneous drainage in necrotizing pancreatitis 23 reporting on the number of catheters placed, two or more were usually used per patient, with a maximum of 14 catheters 54. No accurate mean number of procedures could be calculated because most reports did not provide these data. Generally, drains were flushed with saline every 8 h and were replaced when occluded. In the single RCT on this subject, a median of one catheter was placed with a median size of 14 Fr. Ultimately, drains were upgraded or replaced in 11 of 43 patients 16. The complication rate was described in all but one series 52. One hundred and three complications occurred in 75 (21 2 per cent) of 354 patients. The majority of complications were pancreaticocutaneous and pancreaticoenteric fistulas (53; 51 5 per cent of all 103 complications). In total, nine procedure-related complications were described. This included four bleeding complications, of which two were self-limiting; the other two comprised massive bleeding due to puncture of the splenic artery, and both patients died from subsequent haemorrhagic shock. One colonic perforation, which required surgical intervention, was reported. The remaining four procedure-related complications were considered minor and consisted of transient abdominal pain (1), self-limiting pneumothorax (1) and catheter dislodgement (2). The overall mortality rate was 17 4 per cent (67 of 384). Nine studies 10,16,48,50 53,55,56 reported the mortality rate for PCD in patients with infected necrosis; the combined rate was 15 4 per cent (27 of 175 patients). The mortality rate for PCD in patients with sterile necrosis was 15 per cent (10 of 69 patients) 16,48,50,51,55,56. Discussion This study reviewed the outcome of PCD in a mixed group of patients with both sterile and infected pancreatic necrosis, by and large in retrospective studies. The combined data, including the PCD step-up arm in the only RCT on the subject, supports the conclusion that a considerable number of patients can be treated with PCD without the need for surgical necrosectomy. Surgical necrosectomy was not needed in approximately half of the patients reviewed here. In patients who did require surgical intervention, PCD allowed additional intervention to be postponed for several weeks. The overall mortality rate was 17 4 per cent (reported in 11 studies), and 15 4 per cent in patients with infected necrosis (reported in 9 studies). Although not all studies provided data on deaths among patients with infected necrosis, the mortality rate was similar to that reported for open conventional and minimally invasive necrosectomy. A recent review ( ) of 11 series with more than 100 patients undergoing open necrosectomy reported a mean mortality rate of 19 per cent 60. The same study reported a mortality rate of 19 per cent in a series of 137 patients undergoing minimally invasive necrosectomy. Furthermore, in the one RCT on this subject, the mortality rate in the step-up arm was 19 per cent and multiorgan failure was present in 35 per cent of patients before intervention 16. Over the past two decades, PCD has been used increasingly as primary minimally invasive treatment for necrotizing pancreatitis with proven or suspected infection. The rationale for such treatment is to improve the clinical condition of these usually seriously ill patients by drainage of infected fluid (pus) under pressure either to postpone surgical intervention or even to obviate the need for surgical necrosectomy. Postponing surgical intervention has been shown to improve outcome in patients needing necrosectomy for infected necrosis 61. In 2000, Büchler and colleagues 4 proposed that (surgical) intervention should be restricted to patients with infected peripancreatic or pancreatic necrosis. This has resulted in a much more critically ill group of patients undergoing intervention. The indication for PCD, however, differed between the 11 series included in this review. Although all 384 patients suffered from necrotizing pancreatitis, only 70 6 per cent had infected peripancreatic collections proven by bacterial culture. Other indications for intervention were symptomatic organized necrosis and severe clinical deterioration despite maximal conservative treatment. These last two indications are not very well defined and one may question whether these patients could not have been successfully treated conservatively. In the pooled data of this systematic review, 21 2 per cent of patients had one or more complications, with only nine reported procedure-related complications. Series on surgical necrosectomy report a considerably higher complication rate, ranging from 34 to 68 per cent 62,63.Furthermore, in the present study, only 15 0 per cent of patients developed a pancreatic fistula, compared with per cent in the studies on surgical necrosectomy However, mostly only early complications were described in the included studies. Follow-up was not reported in the included studies, except for the RCT 16, and as a result late complications (such as pseudocysts, pancreatic duct amputations, pancreatic insufficiency and chronic pancreatitis) were probably missed. In the PANTER trial, a 6-month complication rate of 30 per cent was reported in the step-up arm, consisting of incisional hernia (7 per cent),

7 24 M. C. van Baal, H. C. van Santvoort, T. L. Bollen, O. J. Bakker, M. G. Besselink, H. G. Gooszen endocrine insufficiency (16 per cent) and need for pancreatic enzyme supplementation (7 per cent) 16. These results indicate that late complications do occur, but only shortterm complications were reported in ten of the 11 included studies. It is conceivable that drain placement into a sterile peripancreatic collection can introduce bacteria, resulting in secondary infection. Walser and co-workers 31 showed that initially culture-negative collections more frequently become infected after percutaneous drainage than after simple fine-needle aspiration (13 of 22 versus 3 of 15 respectively; P < 0 03). Not all such infections can be classified as iatrogenic, as sterile necrosis could become infected as a result of bacterial translocation through the bowel wall or by systemic infection 67. A recent randomized study found an increased risk of developing infected necrosis by routine prolonged percutaneous catheter drainage of sterile peripancreatic collections compared with conservative treatment (11 of 20 versus 4 of 20 respectively; P = 0 048) 32,68. None of the included studies reported on the rate of iatrogenic infection, but underreporting is likely to have occurred. In the present review, approximately half of the patients still needed surgical necrosectomy after PCD. Although such drainage may have ameliorated the patient s condition and improved outcome, it may also have caused unnecessary delay in a subgroup that is better off with primary surgical intervention. This review, however, does not allow this specific subgroup to be identified. The size of the drains used varied from 8 to 28 Fr and in only one study was the drainage tract dilated stepwise 53. Occluded or dislocated drains were replaced by the interventional radiologist in all series. There has been a tendency over time to use large-bore catheters with a lower rate of catheter occlusion and need for replacement. In the future, it would be interesting to evaluate whether use of large-size catheters reduces the need for catheter replacement. In most series, catheters were flushed daily with saline, in general every 8 h. The daily flushing in combination with the frequent need for catheter replacement makes PCD a relatively intensive and time-consuming therapy for the patient, surgeon and interventional radiologist 69. However, in the RCT, 56 per cent of patients in the stepup arm needed only one catheter placement with a median size of 14 Fr 16. In this review, studies reporting on PCD as part of a stepup approach forminimally invasive necrosectomy were not included to avoid mixing the effects and complications of drainage alone with those of minimally invasive necrosectomy (with or without videoscopic assistance) alone With minimally invasive necrosectomy, not only is the fluid compartment drained by the catheter, but remaining necrotic debris is also removed. This technique has become increasingly popular over the past decade. A limitation of the present systematic review is that many of the included studies were small and retrospective. Moreover, essential data were not presented in some reports (for example total number of interventions, outcome related to infection, status of collections, percentage of patients with organ failure at the time of PCD). A formal assessment of methodological quality could not be performed because the papers did not provide enough detailed information for such an assessment 70. This makes comparison with studies reporting on open necrosectomy even more difficult. It is likely that the success of PCD without the need for further surgery is overestimated in the overall results of this review owing to selection bias, as the overall success rate was 55 7 per cent compared with 35 per cent in the single RCT 16. Nevertheless, many patients with (suspected) infected necrotizing pancreatitis can recover with PCD as the first and only intervention. Acknowledgements The authors declare no conflict of interest. References 1 Banks PA, Freeman ML; Practice Parameters Committee of the American College of Gastroenterology. Practice guidelines in acute pancreatitis. Am J Gastroenterol 2006; 101: British Society of Gastroenterology. United Kingdom guidelines for the management of acute pancreatitis. Gut 1998; 42(Suppl 2): S1 S13. 3 Nieuwenhuijs VB, Besselink MG, van Minnen LP, Gooszen HG. Surgical management of acute necrotizing pancreatitis: a 13-year experience and a systematic review. Scand J Gastroenterol Suppl 2003; 239: Büchler MW, Gloor B, Müller CA, Friess H, Seiler CA, Uhl W. Acute necrotizing pancreatitis: treatment strategy according to the status of infection. Ann Surg 2000; 232: Uhl W, Warshaw A, Imrie C, Bassi C, McKay CJ, Lankisch PG et al.; International Association Pancreatology. IAP Guidelines for the Surgical Management of Acute Pancreatitis. Pancreatology 2002; 2: Werner J, Feuerbach S, Uhl W, Büchler MW. Management of acute pancreatitis: from surgery to interventional intensive care. Gut 2005; 54:

8 Percutaneous drainage in necrotizing pancreatitis 25 7 Rau B, Bothe A, Beger HG. Surgical treatment of necrotizing pancreatitis by necrosectomy and closed lavage: changing patient characteristics and outcome in a 19-year, single-center series. Surgery 2005; 138: Rodriguez JR, Razo AO, Targarona J, Thayer SP, Rattner DW, Warshaw AL et al. Debridementandclosed packing for sterile or infected necrotizing pancreatitis: insights into indications and outcomes in 167 patients. Ann Surg 2008; 247: Tsiotos GG, Luque-de León E, Sarr MG. Long-term outcome of necrotizing pancreatitis treated by necrosectomy. Br J Surg 1998; 85: Freeny PC, Hauptmann E, Althaus SJ, Traverso LW, Sinanan M. Percutaneous CT-guided catheter drainage of infected acute necrotizing pancreatitis: techniques and results. AJR Am J Roentgenol 1998; 170: Besselink MG, van Santvoort HC, Schaapherder AF, van Ramshorst B, van Goor H, Gooszen HG et al. Feasibility of minimally invasive approaches in patients with infected necrotizing pancreatitis. Br J Surg 2007; 94: Carter CR, McKay CJ, Imrie CW. Percutaneous necrosectomy and sinus tract endoscopy in the management of infected pancreatic necrosis: an initial experience. Ann Surg 2000; 232: Horvath KD, Kao LS, Ali A, Wherry KL, Pellegrini CA, Sinanan MN. Laparoscopic assisted percutaneous drainage of infected pancreatic necrosis. Surg Endosc 2001; 15: van Santvoort HC, Besselink MG, Horvath KD, Sinanan MN, Bollen TL, van Ramshorst B et al. Videoscopic assisted retroperitoneal debridement in infected necrotizing pancreatitis. HPB (Oxford) 2007; 9: Seifert H, Biermer M, Schmitt W, Jürgensen C, Will U, Gerlach R et al. Transluminal endoscopic necrosectomy after acute pancreatitis: a multicentre study with long-term follow-up (the GEPARD Study). Gut 2009; 58: van Santvoort HC, Besselink MG, Bakker OJ, Hofker HS, Boermeester MA, Dejong CH et al. A step-up approach or open necrosectomy for necrotizing pancreatitis. NEnglJ Med 2010; 362: Bradley EL III. A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, Ga, September 11 through 13, Arch Surg 1993; 128: Bollen TL, van Santvoort HC, Besselink MG, van Es WH, Gooszen HG, van Leeuwen MS. Update on acute pancreatitis: ultrasound, computed tomography, and magnetic resonance imaging features. Semin Ultrasound CT MR 2007; 28: Bollen TL, van Santvoort HC, Besselink MG, van Leeuwen MS, Horvath KD, Freeny PC et al. The Atlanta Classification of acute pancreatitis revisited. Br J Surg 2008; 95: Ranson JH, Rifkind KM, Roses DF, Fink SD, Eng K, Spencer FC. Prognostic signs and the role of operative management in acute pancreatitis. Surg Gynecol Obstet 1974; 139: Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13: Balthazar EJ, Robinson DL, Megibow AJ, Ranson JH. Acute pancreatitis: value of CT in establishing prognosis. Radiology 1990; 174: Mortele KJ, Wiesner W, Intriere L, Shankar S, Zou KH, Kalantari BN et al. A modified CT severity index for evaluating acute pancreatitis: improved correlation with patient outcome. AJR Am J Roentgenol 2004; 183: Balthazar EJ, Ranson JH, Naidich DP, Megibow AJ, Caccavale R, Cooper MM. Acute pancreatitis: prognostic value of CT. Radiology 1985; 156: Ashley SW, Perez A, Pierce EA, Brooks DC, Moore FD Jr, Whang EE et al. Necrotizing pancreatitis: contemporary analysis of 99 consecutive cases. Ann Surg 2001; 234: Cantasdemir M, Kara B, Kantarci F, Mihmanli I, Numan F, Erguney S. Percutaneous drainage for treatment of infected pancreatic pseudocysts. South Med J 2003; 96: Mithöfer K, Mueller PR, Warshaw AL. Interventional and surgical treatment of pancreatic abscess. World J Surg 1997; 21: Srikanth G, Sikora SS, Baijal SS, Ayyagiri A, Kumar A, Saxena R et al. Pancreatic abscess: 10 years experience. ANZ JSurg2002; 72: van Sonnenberg E, Wittich GR, Chon KS, D Agostino HB, Casola G, Easter D et al. Percutaneous radiologic drainage of pancreatic abscesses. AJR Am J Roentgenol 1997; 168: Malecka-Panas E, Juszynski A, Chrzastek J, Nowacka B, Jarkowska J, Studniarek M. Pancreatic fluid collections: diagnostic and therapeutic implications of percutaneous drainage guided by ultrasound. Hepatogastroenterology 1998; 45: Walser EM, Nealon WH, Marroquin S, Raza S, Hernandez JA, Vasek J. Sterile fluid collections in acute pancreatitis: catheter drainage versus simple aspiration. Cardiovasc Intervent Radiol 2006; 29: Zerem E, Imamovic G, Omerović S,Imširović B. Randomized controlled trial on sterile fluid collections management in acute pancreatitis: should they be removed? Surg Endosc 2009; 23: Heider R, Meyer AA, Galanko JA, Behrns KE. Percutaneous drainage of pancreatic pseudocysts is associated with a higher failure rate than surgical treatment in unselected patients. Ann Surg 1999; 229: Berzin TM, Banks PA, Maurer R, Mortele KJ. CT-guided percutaneous catheter drainage in necrotizing pancreatitis: outcomes among patients discharged with drains in place. J Vasc Interv Radiol 2008; 19: Kumar P, Mukhopadhyay S, Sandhu M, Berry M. Ultrasonography, computed tomography and percutaneous intervention in acute pancreatitis: a serial study. Australas Radiol 1995; 39:

9 26 M. C. van Baal, H. C. van Santvoort, T. L. Bollen, O. J. Bakker, M. G. Besselink, H. G. Gooszen 36 Lee MJ, Rattner DW, Legemate DA, Saini S, Dawson SL, Hahn PF et al. Acute complicated pancreatitis: redefining the role of interventional radiology. Radiology 1992; 183: Szentkereszty Z, Kerekes L, Hallay J, Czako D, Sápy P. CT-guided percutaneous peripancreatic drainage: a possible therapy in acute necrotizing pancreatitis. Hepatogastroenterology 2002; 49: Szentkereszty Z, Kotán R, Pósán J, Arkossy P, Sápy P. Therapeutic tactics in the treatment of acute necrotizing pancreatitis. Hepatogastroenterology 2008; 55: Ai X, Qian X, Pan W, Xu J, Hu W, Terai T et al. Ultrasound-guided percutaneous drainage may decrease the mortality of severe acute pancreatitis. J Gastroenterol 2010; 45: Becker V, Huber W, Meining A, Prinz C, Umgelter A, Ludwig L et al. Infected necrosis in severe pancreatitis combined nonsurgical multi-drainage with directed transabdominal high-volume lavage in critically ill patients. Pancreatology 2009; 9: Ross A, Gluck M, Irani S, Hauptmann E, Fotoohi M, Siegal J et al. Combined endoscopic and percutaneous drainage of organized pancreatic necrosis. Gastrointest Endosc 2010; 71: Bruennler T, Langgartner J, Lang S, Zorger N, Herold T, Salzberger B et al. Percutaneous necrosectomy in patients with acute, necrotizing pancreatitis. Eur Radiol 2008; 18: Bucher P, Pugin F, Morel P. Minimally invasive necrosectomy for infected necrotizing pancreatitis. Pancreas 2008; 36: Chang YC, Tsai HM, Lin XZ, Chang CH, Chuang JP. No debridement is necessary for symptomatic or infected acute necrotizing pancreatitis: delayed, mini-retroperitoneal drainage for acute necrotizing pancreatitis without debridement and irrigation. Dig Dis Sci 2006; 51: Echenique AM, Sleeman D, Yrizarry J, Scagnelli T, Guerra JJ Jr, Casillas VJ et al. Percutaneous catheter-directed debridement of infected pancreatic necrosis: results in 20 patients. J Vasc Interv Radiol 1998; 9: Endlicher E, VölkM,FeuerbachS,Schölmerich J, Schäffler A, Messmann H. Long-term follow-up of patients with necrotizing pancreatitis treated by percutaneous necrosectomy. Hepatogastroenterology 2003; 50: Bala M, Almogy G, Klimov A, Rivkind AI, Verstandig A. Percutaneous stepped drainage technique for infected pancreatic necrosis. Surg Laparosc Endosc Percutan Tech 2009; 19: e113 e Gambiez LP, Denimal FA, Porte HL, Saudemont A, Chambon JP, Quandalle PA. Retroperitoneal approach and endoscopic management of peripancreatic necrosis collections. Arch Surg 1998; 133: Fotoohi M, D Agostino HB, Wollman B, Chon K, Shahrokni S, van Sonnenberg E. Persistent pancreatocutaneous fistula after percutaneous drainage of pancreatic fluid collections: role of cause and severity of pancreatitis. Radiology 1999; 213: Baril NB, Ralls PW, Wren SM, Selby RR, Radin R, Parekh D et al. Does an infected peripancreatic fluid collection or abscess mandate operation? Ann Surg 2000; 231: Cheung MT, Ho CN, Siu KW, Kwok PC. Percutaneous drainage and necrosectomy in the management of pancreatic necrosis. ANZ J Surg 2005; 75: Navalho M, Pires F, Duarte A, Gonçalves A, Alexandrino P, Távora I. Percutaneous drainage of infected pancreatic fluid collections in critically ill patients: correlation with C-reactive protein values. Clin Imaging 2006; 30: Lee JK, Kwak KK, Park JK, Yoon WJ, Lee SH, Ryu JK et al. The efficacy of nonsurgical treatment of infected pancreatic necrosis. Pancreas 2007; 34: Bruennler T, Langgartner J, Lang S, Wrede CE, Klebl F, Zierhut S et al. Outcome of patients with acute, necrotizing pancreatitis requiring drainage does drainage size matter? World J Gastroenterol 2008; 14: Mortelé KJ, Girshman J, Szejnfeld D, Ashley SW, Erturk SM, Banks PA et al. CT-guided percutaneous catheter drainage of acute necrotizing pancreatitis: clinical experience and observations in patients with sterile and infected necrosis. AJR Am J Roentgenol 2009; 192: Rocha FG, Benoit E, Zinner MJ, Whang EE, Banks PA, Ashley SW et al. Impact of radiologic intervention on mortality in necrotizing pancreatitis: the role of organ failure. Arch Surg 2009; 144: Phillips B, Ball C, Sackett D, Badenoch D, Strauss S, Haynes B et al. Levels of Evidence (May 2001). Oxford Centre for Evidence-based Medicine: Oxford, Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald WJ. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995; 23: Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996; 22: Raraty MG, Halloran CM, Dodd S, Ghaneh P, Connor S, Evans J et al. Minimal access retroperitoneal pancreatic necrosectomy: improvement in morbidity and mortality with a less invasive approach. Ann Surg 2010; 251: Besselink MG, Verwer TJ, Schoenmaeckers EJ, Buskens E, Ridwan BU, Visser MR et al. Timing of surgical intervention in necrotizing pancreatitis. Arch Surg 2007; 142: Howard TJ, Patel JB, Zyromski N, Sandrasegaran K, Yu J, Nakeeb A et al. Declining morbidity and mortality rates in the surgical management of pancreatic necrosis. JGastrointest Surg 2007; 11:

10 Percutaneous drainage in necrotizing pancreatitis Connor S, Alexakis N, Raraty MG, Ghaneh P, Evans J, Hughes M et al. Early and late complications after pancreatic necrosectomy. Surgery 2005; 137: Tzovaras G, Parks RW, Diamond T, Rowlands BJ. Early and long-term results of surgery for severe necrotising pancreatitis. Dig Surg 2004; 21: Tsiotos GG, Smith CD, Sarr MG. Incidence and management of pancreatic and enteric fistulas after surgical management of severe necrotizing pancreatitis. Arch Surg 1995; 130: Harris HW, Barcia A, Schell MT, Thoeni RF, Schecter WP. Necrotizing pancreatitis: a surgical approach independent of documented infection. HPB (Oxford) 2004; 6: Beger HG, Bittner R, Block S, Büchler M. Bacterial contamination of pancreatic necrosis. A prospective clinical study. Gastroenterology 1986; 91: Besselink MG, van Santvoort HC, Bakker OJ, Bollen TL, Gooszen HG. Draining sterile fluid collections in acute pancreatitis? Primum non nocere! Surg Endosc 2010; (in press). 69 Mueller PR. Percutaneous drainage of pancreatic necrosis: is it ecstasy or agony? AJR Am J Roentgenol 1998; 170: Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet 1999; 354: Commentary Systematic review of percutaneous catheter drainage as primary treatment for necrotizing pancreatitis (Br J Surg 2011; 98: 18 27) With the formation of pus in the omental cavity comes the opportunity for the surgeon wrote Fitz in 1889 and thus defined indications for surgery in acute pancreatitis 1. Recently, the significant progress in imaging and endoscopy have allowed minimally invasive removal of infected pancreatic necrotic tissue. Can we abandon open surgery for severe acute pancreatitis (SAP) in the future? The authors tried to find an answer to this question in the literature. The result was expectedly negative the success rate of percutaneous catheter drainage (PCD) was 55 7 per cent. Moreover, one-third of the patients did not suffer infected, but sterile pancreatic necrosis. The PAncreatitis, Necrosectomy versus step up approach (PANTER) trial, the single randomized controlled trial included in this review, reported a much more realistic 35 per cent success rate of PCD alone in the early phase of the disease 2. Late sequelae of SAP still requiring open surgery, such as abscesses or infected pseudocysts persisting after PCD, or pancreatic duct amputation with the consequence of chronic pancreatitis, were not taken into account. Thus, the actual success rate of PCD alone should be even lower. The procedure was even associated with morbidity and mortality rates of 21 2 and17 4 per cent respectively, results not significantly superior to those of open surgery, as confirmed by the PANTER trial 2,3. The lack of efficient necrosectomy due to incomplete or ineffective drainage and persistent pancreatic fistula are the major limitations of PCD. It certainly represents an important minimally invasive tool in the treatment of localized and well organized pancreatic necrosis, but remains inferior to open surgery in complicated SAP 3. The results of this systematic review should be interpreted with caution as it analysed a small number of studies with low-level evidence and a limited total number of patients. Minimally invasive techniques (PCD and more recently endoscopic and transgastric necrosectomy) certainly have their place in the armamentarium for the treatment of SAP. However, we do not know the exact place, and an interdisciplinary team approach to every patient, including surgeons, gastroenterologists and interventional radiologists, is of utmost importance. W. Uhl Department of Surgery, St Josef Hospital, Ruhr University of Bochum Gudrunstrasse 56, Bochum, Germany ( w.uhl@klinikum-bochum.de) DOI: /bjs British Journal of Surgery Society Ltd British Journal of Surgery 2011; 98: 27 28

11 28 M. C. van Baal, H. C. van Santvoort, T. L. Bollen, O. J. Bakker, M. G. Besselink, H. G. Gooszen References 1 Fitz RH. Acute pancreatitis; a consideration of pancreatic hemorrhage, hemorrhagic suppurative, and gangrenous pancreatitis, and of disseminated fat-necrosis. Boston Medical and Surgical Journal 1889; 120: , , van Santvoort HC, Besselink MG, Bakker OJ, Hofker HS, Boermeester MA, Dejong CH et al. A step-up approach or open necrosectomy for necrotizing pancreatitis. NEnglJMed2010; 362: Uhl W, Warshaw A, Imrie C, Bassi C, McKay CJ, Lankisch PG et al.; International Association of Pancreatology. IAP Guidelines for the Surgical Management of Acute Pancreatitis. Pancreatology 2002; 2: If you wish to comment on this, or any other article published in the BJS, please visit the on-line correspondence section of the website ( Electronic communications will be reviewed by the Correspondence Editor and a selection will appear in the correspondence section of the Journal. Time taken to produce a thoughtful and well written letter will improve the chances of publication in the Journal British Journal of Surgery Society Ltd British Journal of Surgery 2011; 98: 27 28