Endoscopy 2012; 44(04): 389-393
DOI: 10.1055/s-0031-1291872
Endoscopy Essentials
© Georg Thieme Verlag KG Stuttgart · New York

Pancreatic endoscopic retrograde cholangiopancreatography (ERCP)

S. I. Gan, R. A. Kozarek
  • Virginia Mason Medical Center, Seattle, Washington, USA
Further Information

Publication History

Publication Date:
21 March 2012 (online)

Pancreatic duct stents for prophylaxis against post-ERCP pancreatitis

Several important papers published recently investigated the prophylactic use of pancreatic stents against the development of pancreatitis following endoscopic retrograde cholangiopancreatography (ERCP).

  • Pancreatic stents for prophylaxis against post-ERCP pancreatitis: a meta-analysis and systematic review (Choudhary et al., Gastrointest Endosc 2011 [1])

  • Prophylactic pancreatic stent placement and post-ERCP pancreatitis: a systematic review and meta-analysis (Mazaki et al., Endoscopy 2010 [2])

  • Acute pancreatitis after removal of retained prophylactic pancreatic stents (Moffatt et al., Gastrointest Endosc 2011 [3])

  • Prophylactic 5-Fr pancreatic duct stents are superior to 3-Fr stents: a randomized controlled trial (Zolotarevsky et al., Endoscopy 2011 [4])

  • Risk factors for post-ERCP pancreatitis in high risk patients who have undergone prophylactic pancreatic duct stenting: a multicenter retrospective study (Ito et al., Intern Med 2011 [5]).

Pancreatitis remains the most common complication of ERCP, with large studies showing a rate of between 1.6 % and 6.7 % [6] [7] [8]. Risk factors for post-ERCP pancreatitis (PEP) are additive and include operator characteristics (experience and case volume), patients’ underlying diagnoses and indications (e. g. recurrent pancreatitis, previous PEP, Sphincter of Oddi dysfunction [SOD]), and procedure-related factors (e. g. pancreatic duct manipulation, manometry, difficult cannulation) [6]. Several mechanisms have been suggested, including papillary edema, and main and side-branch ductal injury secondary to mechanical and contrast-induced trauma.

Prophylactic pancreatic duct stenting with small-diameter stents is believed to reduce elevated intraductal pancreatic duct pressures secondary to papillary edema and to maintain pancreatic duct outflow after papillary injury. Pancreatic duct stenting has been shown to decrease the risk of PEP in several randomized controlled trials (RCTs) [9] [10] [11] but there is heterogeneity between studies in populations, methods, and outcomes. Two meta-analyses were published in 2004 and 2007, respectively, both of which showed a benefit of stent placement [12] [13]. More studies have been published over the past 4 years and two meta-analyses were published within the past year [1] [2].

Choudhary et al. systematically reviewed both randomized and non-randomized studies [1]. A total of 87 articles were initially chosen and 69 were excluded. Of the remaining 18 studies, there were 8 RCTs and 10 non-randomized studies. Data from RCTs were statistically pooled and analyzed on both a per-protocol and an intention-to-treat basis. Pooled estimates for PEP, hyperamylasemia, and severity of pancreatitis were calculated.

In the meta-analysis of RCTs, results from 656 patients were pooled. Indications for stenting were all patients undergoing ERCP in two studies, but in the other six studies, high risk inclusion criteria included sphincterotomy, ampullectomy, and SOD. There was some variation in the types of stents used, but all studies used either 5-Fr or 7-Fr stents between 2 and 4 cm in length, some flanged, some not. All but one of the studies reported using the consensus definition of PEP. No significant heterogeneity between the studies was detected. In the 322 patients who underwent placement of pancreatic duct stents, the rate of PEP was 4.96 % compared with 19.76 % in the group who did not undergo stent placement (n = 334). The odds ratio reduction was 0.22, with an absolute risk reduction of 13.3 %. No significant difference was found between the length of the stent or the presence or absence of flanges. In the analysis of the severity of PEP, stents appeared to reduce the incidence of mild and moderate PEP but only a trend towards benefit was noted in cases of severe PEP. For the intention-to-treat analysis, five studies included information on the success of pancreatic duct stent placement. If a case of PEP was considered as unsuccessful stent placement, prophylactic stent placement still showed an absolute risk reduction of 10.6 %.

A systematic review of the 10 non-RCTs also supported placement of prophylactic pancreatic duct stents. In five of the studies, a statistically significant reduction in the incidence of PEP was found, and in the other five studies there was a trend towards benefit.

In the meta-analysis by Mazaki et al., the same eight RCTs were analyzed with almost duplicate results, although no intention-to-treat analysis was conducted [2]. Pooled data showed a relative risk of 0.36 for development of mild-to-moderate PEP and a statistically significant 0.23 relative risk reduction of severe pancreatitis, which differed slightly from the Choudhary results [1]. In addition, the authors performed a systematic review of both RCT and observational studies to determine immediate complication rates of pancreatic duct stent placement. From 17 studies (two RCTs were included), data from 4115 patients were collected, and pooled weighted estimates of complications were infection 3 %, bleeding 2.5 %, stent migration 4.9 %, stent occlusion 7.9 %, duct perforation 0.4 %, pseudocyst 3 %, and retroperitoneal perforation 1.2 %. Stent fracture and pancreatic duct stricture were not reported.

These two meta-analyses provide a clear signal that prophylactic stenting does appear to be beneficial in the prevention of PEP. However, questions remain concerning patient selection, stent selection, and whether all practicing endoscopists should be attempting pancreatic duct stent placement. Complications from attempted and successful pancreatic duct stent placement are evident from the Mazaki study [2], and the benefits of pancreatic duct stenting must be weighed accordingly; it is likely that pancreatic duct stenting is not necessary for every patient undergoing ERCP.

Another complication reported recently involves pancreatitis after endoscopic stent retrieval of retained stents. Moffatt et al. reported that in 3 % of cases of endoscopic stent retrieval, mild and moderate cases of acute pancreatitis occurred [3]. Risk factors for this included use of a 5-Fr stent, internal flanged-stents, and PEP that occurred after the initial ERCP. It is partly for this reason that the choice of stent size for PEP prophylaxis is of great interest to ERCP practitioners, not only for possible differences in rates of PEP but also in the rates of spontaneous passage. Current studies appear to favor shorter length, larger diameter stents. In 2009, Chahal et al. published an RCT of 5-Fr 3-cm unflanged stents vs. 3-Fr 8-cm single pigtail stents and found a trend towards higher rates of PEP in the 3-Fr group [14]. In addition, stent placement failures were significantly higher in the 3-Fr group and the rates of spontaneous dislodgement were lower, necessitating more endoscopic removal. Zolotarevsky et al. published a similar study using mostly 3-Fr 6-cm stents and 5-Fr 5-cm stents [4]. The study was stopped prematurely after 78 patients were enrolled because no statistically significant difference was found in spontaneous passage (approximately 90 %) and no difference was detected in the rates of PEP. Furthermore, 5-Fr stents were deemed easier to place and fewer guide wires were used as there was no need for small-caliber guide wires.

Regardless of placement of pancreatic duct stents, the risks of PEP persist. Ito et al. published a multicenter retrospective study that attempted to determine the risk factors for the development of PEP after prophylactic stent placement in a high risk population [5]. Over 400 patients were included, all of whom received a 5-Fr single pigtail stent. In total, 64 % of patients developed hyperamylasemia and pancreatitis developed in 9.9 % (95 % of which was mild-or-moderate pancreatitis). Numerous possible risk factors were analyzed (e. g. pancreatic duct contrast injection, sphincterotomy, biliary stent placement, difficult or failed cannulation). Both univariate and multivariate analyses showed that the sole risk factor for development of PEP in this high risk group was the presence of intraductal papillary mucinous neoplasm (IPMN), with an odds ratio of 2.9 (P = 0.040). This echoed a retrospective study by Harada et al., which showed a higher rate of PEP in patients with IPMN who received a prophylactic stent compared with those who did not [15]. The role of prophylactic stent placement in patients with IPMN should thus be considered questionable.

In conclusion, prophylactic stents are clearly effective in reducing the risk of PEP in high risk patients. The choice of when to place a stent must be determined on an individual basis, and given the identified risks of stent placement, practitioner skill and experience may also play a role in the decision to attempt stent placement. Failed stent placement can lead to even higher rates of PEP [16]. Shorter, wider stents appear to be easier to place than long 3-Fr stents, with equivalent or better spontaneous passage rates and equivalent efficacy in reduction of PEP. Follow-up studies of stent-related ductitis/strictures and sphincterotomy stenosis have not been studied but these complications could be higher with higher-caliber stents. Special care should be taken in patients with IPMN, who appear to have a higher rate of PEP when pancreatic duct stents are placed. Additional studies are necessary to determine optimal stent characteristics and to define which patient populations have the highest benefit-to-risk ratio.