Introduction

Post-ERCP pancreatitis (PEP) is the most frequent adverse event following ERCP with an incidence ranging from 5 to 10% [1, 2]. The course of PEP can range from mild with a brief hospital admission to a more severe fulminant course with multi-organ failure and rarely death. The estimated financial burden of PEP is $200 million per year [2]. In this article, we review risk factors associated with PEP and discuss strategies for prevention and management.

Definitions

The consensus definition of PEP requires abdominal pain with an amylase level > 3 × normal for > 24 h post-ERCP [1, 3••]. Modifications include utilizing lipase and defining clinical pancreatitis as “new or worsened abdominal pain” [4••, 5]. The 2012 Atlanta classification does not differentiate PEP from acute pancreatitis characterized by abdominal pain, serum lipase, or amylase 3 × the upper limit of normal (ULN), and associated imaging findings. The European Society of Gastrointestinal Endoscopy (ESGE) utilizes the modified consensus criteria and requires new or prolonged hospital admission. Severity is determined by the revised Atlanta classification based on organ failure duration [6]. A serum amylase or lipase should be checked 2–6 h post-ERCP for patients with abdominal pain. Levels < 1.5 × and < 4 × the ULN, respectively, are low risk for PEP and acceptable for discharge [5, 7••].

Mechanism of Injury

Mechanical injuries to the sphincter of Oddi, ampulla, and biliary or pancreatic ducts during instrumentation with guidewires and catheters can result in edema or perforation of the pancreatic duct, outflow obstruction, and leakage of pancreatic enzymes caustic to surrounding tissue. Hydrostatic, chemical, and allergic reactions can occur with contrast injection and thermal injuries with electrocautery. Superimposed infections may develop from bacterial introduction via contaminated instruments or intestinal flora [1]. These injuries stimulate an inflammatory, cytokine-driven cascade with the degree of injury depending on whether the inflammation is localized to the pancreatic parenchyma or initiates a systemic inflammatory response syndrome (SIRS).

Risk Factors

PEP risk factors are categorized into three groups related to patient characteristics, provider characteristics, and procedural techniques. The first major multicenter study to describe independent risk factors for PEP prospectively followed 2347 patients for 30 days following endoscopic sphincterotomy (ES). Higher PEP rates were seen in cases with sphincter of Oddi dysfunction (SOD), difficulty with bile duct cannulation, use of precut sphincterotomy, and when combined with percutaneous intervention [4••]. Multiple additional risk factors have since been described.

Patient-Related Risk Factors

Independent patient-related risk factors include SOD, female gender, younger age, normal serum bilirubin, and history of previous pancreatitis or PEP [1, 3••, 4••, 5, 8, 9]. Other potential risk factors include primary sclerosing cholangitis (PSC), hilar tumors, nondilated extrahepatic bile ducts, pancreas divisum, and absence of chronic pancreatitis (Table 1) [1, 3••, 4••, 5, 8,9,10,11]. These factors when combined further increase PEP risk.

Table 1 Independent risk factors for post-ERCP pancreatitis
Full size table

Provider and Facility-Related Risk Factors

Endoscopist experience and procedural volume do not significantly impact PEP rates though improve successful bile duct cannulation [4••, 8]. Multi-center international studies show similar PEP rates among expert and non-expert endoscopists and low and high–volume centers [9, 12, 13]. However, higher-risk populations are more frequently treated at high-volume centers and may confound reports.

Procedure-Related Risk Factors

Guidewire passage into the pancreatic duct (PD) > 1 × and deep PD cannulation are predominant risk factors followed by PD contrast injection and prolonged biliary cannulation (> 10 min or 5 attempts) [14••, 15, 16]. Others include precut ES, pancreatic sphincterotomy, ampullectomy, minor papillectomy, failure to clear bile duct stones, and intraductal ultrasound (IDUS) (Table 1) [1, 3••, 4••, 5, 8,9,10, 14••, 15, 17]. Fully covered and uncovered self-expanding metal stents (SEMS) increase PEP risk compared to polyethylene stents placed for biliary obstruction [16, 18, 19, 20•].

Management

The principles of PEP management are similar as for other causes of acute pancreatitis with aggressive and judicious intravenous fluid resuscitation with Ringer’s lactate (LR) at 5–10 ml/kg/h in the initial 24 h, bowel rest and pain management [21]. A single-center retrospective study evaluated the role of repeat ERCP for salvage pancreatic stent (PS) placement to attenuate PEP progression. All 14 of 64 patients with severe PEP who underwent salvage ERCP (5 without and 9 with prior PS) within 10 h had a significant improvement in pain, amylase/lipase level, and SIRS resolution by 24 h post treatment (P = 0.003). No delayed adverse events occurred [22]. Larger prospective studies are needed before salvage ERCP can be incorporated into clinical practice.

Prevention

It is important for endoscopists to know factors proven to reduce PEP and appropriate instances when these can be applied. Methods of cannulation, papillary balloon dilation, type of electrocautery used, prophylactic pancreatic stenting, rectal NSAIDs, and aggressive hydration (AH) have all been examined in this regard and are further discussed below.

Cannulation Techniques

Selective biliary cannulation failure rates are as high as 18% in some reports [23•]. Wire-guided cannulation (WGC) has gained traction over traditional contrast-assisted due to improved cannulation success rates, reduced papillary trauma, avoidance of PD contrast injection, decreased need for precut ES, and overall reduced risk for PEP [24,25,26]. However, even experts endure complications with WGC such as creation of a false tract, intramural dissection, perforation, and PD injury, predisposing to PEP. PEP rates have been reported as high as 9% with WGC. The risk is greater after inadvertent PD guidewire insertion versus PD opacification (OR 2.64 vs 1.89) [15, 27, 28]. More centers are adopting the short-wire system with the endoscopist controlling the wire during cannulation as it was shown to decrease PEP due to reduced trauma related to tactile feedback. A recent trial randomized 216 patients with native papilla cannulation for standard biliary indications to endoscopist versus assistant-controlled attempt. This study was halted at interim analysis due to overwhelming risk of PEP in the assistant-controlled arm (OR 2.8 vs 9.3; P = 0.049) [29]. In cases of difficult selective biliary cannulation (i.e., > 5 contacts with the papilla, > 5 min of attempts, or > 1 unintended PD cannulation), early utilization of advanced techniques may help to improve access and reduce risk of PEP. Advanced techniques include double-guidewire technique (DGWT), wire-guided cannulation over a pancreatic stent (WGC-PS), and early precut ES.

Double Guidewire Technique

The DGWT utilizes a guidewire inserted into the PD to straighten the common channel and facilitate deep biliary access using a second wire. It initially showed success in cases of a tortuous major papilla, or one located in a duodenal diverticulum [30, 31]. Later studies found no improvement in successful cannulation and increased PEP rates compared to other methods [24, 32,33,34,35]. Notably, pancreatic stenting was not performed in most of these studies. A recent RCT randomized 100 patients with intact papilla and inadvertent pancreatic wire placement during ERCP for biliary therapy to DGWT or WGC-PS. The primary outcome assessed was successful selective biliary cannulation within 5 min without need for needle knife precut (NK-precut) ES. All participants had a prophylactic PS placed and no rectal NSAIDs were used. DGWT showed superior initial selective biliary cannulation compared to WGC-PS (90% vs 54%; P < 0.01) and decreased need for NK-precut (10% vs 46%; P < 0.01). After inclusion of NK-precut, the overall successful cannulation rate between both intention-to-treat groups was remarkably high at 98%. Also noteworthy, the overall PEP rate (2%) was low in this otherwise high-risk cohort [23•].

Discordances in reported selective biliary cannulation success rates with DGWT may be explained by differences in study inclusion criteria (i.e., if only patients with difficult biliary cannulation are included or in combination with cases of inadvertent PD wire passage). Similarly, differences in reported PEP rates can be attributed to whether or not a PS is placed [36]. Regardless, the above results are encouraging and support early DGWT as a useful strategy to gain deep biliary access particularly in cases of inadvertent pancreatic wire placement. When employed, simultaneous placement of a prophylactic PS is needed which has shown to significantly reduce PEP rates with minimal added risk [7••, 23•, 36, 37•].

Wire-Guided Cannulation over a Pancreatic Stent

WGC-PS uses a PS in place of a wire to straighten the common channel and guide wire cannulation of the bile duct. It has proven to be an effective technique in facilitating difficult biliary cannulation and decreasing the need for precut ES [38]. Additionally, it offers protective measures in cases where repeated inadvertent pancreatic cannulation occurs. Initial studies showed similar rates of successful biliary cannulation and PEP compared to DGWT [38, 39]. In the most recent comparative trial, it was inferior to the DGWT for initial selective biliary cannulation; however overall successful cannulation rates were the same after NK-precut was included [23•]. A multicenter retrospective study found that patients with native papilla requiring ERCP for biliary intervention who underwent WGC-PS immediately after incidental pancreatic guidewire insertion versus repeated WGC had a decreased PEP (8.7% vs 19%; OR: 0.31; P = 0.001) and PEP severity (moderate and severe PEP; 2.2% vs 6.4%; P = 0.04) although rates of difficult cannulation and overall successful biliary cannulation were not changed (66% vs 70%; P = 0.39 and 98% vs 96%; P = 0.21) [40]. Overall, WGC-PS may be less effective than DGWT for initial biliary cannulation. However, after including NK-precut, overall cannulation success rates are similar and both techniques significantly reduce the risk for PEP provided a PS is placed. While the ESGE recommends the use of DGWT in cases of difficult selective biliary cannulation where inadvertent pancreatic wire insertion occurs, WGC-PS may be considered an alternative based on endoscopist preference, particularly in cases where multiple PD cannulations are made.

Precut Sphincterotomy

Precut ES is considered to be a second-line salvage for selective biliary access. It previously was identified to be a significant independent risk factor for PEP. However, PEP rates related to precut technique do not differ from those of standard sphincterotomy suggesting the risk may be operator-dependent [5, 10, 12, 41, 42]. Alternatively, papillary trauma from preceding unsuccessful conventional cannulation may be responsible for PEP. A single-center RCT randomized 333 patients to very early precut (after two failed attempts of WGC) or primary precut (n = 151) to identify the true incidence of PEP. Cannulation success rates were the same (92.7%) for both groups but the rates of PEP were significantly lower with primary precut (0.67% vs 5.2%; P = 0.04) [43•]. While this suggests that precut ES may be less of an independent risk factor for PEP than once thought, these results should not be extrapolated to support the use of primary precut as a first-line technique for deep biliary access. This study primarily included patients with a dilated common bile duct (CBD) from obstructive biliary disease (ideal for precut), with ERCPs performed by an expert endoscopist, excluded cases of papillary distortion, and did not include rectal indomethacin administration which may have negated the difference observed [44]. Also, biliary cannulation success rates by primary precut were lower than those reported in prior studies using DGWT or WGC-PS and salvage NK-precut [23•]. Nonetheless, studies continue to show that early precut results in improved primary cannulation success and reduced PEP rates compared to persistent standard cannulation [45,46,47]. The ESGE recommends using early NK-precut ES in cases of difficult biliary cannulation [7••]. This technique should be reserved for expert endoscopists, and indications should be evaluated on a case-by-case basis. Transpancreatic biliary sphincterotomy is preferable over NK when dealing with small, flat, or intradiverticular papilla, or when anatomy is highly distorted as in malignancy. The efficacy of NK papillotomy tends to diminish as the CBD diameter decreases, particularly under 4 mm [48,49,50].

Papillary Balloon Dilation

Endoscopic papillary balloon dilation (EPBD) as an adjunct to biliary ES for large biliary stone extraction is standard of care. Both the American Society for Gastrointestinal Endoscopy (ASGE) and ESGE recommend ES with sequential EPBD over ES alone for the removal of large CBD stones due to multiple studies showing improved complete clearance, decreased need for mechanical lithotripsy (ML), and an improved safety profile including reduced risk for PEP [5, 51•, 52,53,54]. EPBD alone without ES has the advantages of anatomic and functional preservation of the sphincter of Oddi (ideal for younger patients), reduced bleeding risk (useful in patients on anticoagulants or low platelets), and fewer late stone recurrences. Prior RCTs found a higher incidence of PEP with EPBD alone compared to biliary ES and increased need for ML [55, 56]. However, the recent MARVELOUS trial (RCT including 19 Japanese institutions involving 171 patients with large CBD stones) showed that EPBD alone compared to ES resulted in significantly higher single-session stone extraction rates (90.7% vs 78.8%; P = 0.04) and decreased the need for ML (30.2% vs 48.2%; P = 0.02) with comparable overall rates of early adverse events including PEP (4.7% vs 5.9%) [57•]. More RCTs comparing these techniques are needed in the western population. While PEP incidence with EPBD is thought to be impacted by the duration of inflation, a new meta-analysis showed no difference between short (15–20 s) and long (1–5 min) dilation groups in the complete stone removal, the need for ML, and rates of PEP, bleeding, biliary infection, or perforation [58, 59].

In cases of very large (> 10 mm) and difficult CBD stones, the use of a limited ES followed by endoscopic papillary large-balloon dilation (EPLBD) with a balloon diameter of 12–20 mm remains first-line treatment [54]. A large systematic review showed overall adverse events (PEP, bleeding, and perforation) to be lower with ES and EPLBD compared to ES monotherapy (8.3% vs 12.7%, OR 1.60; P < 0.001) [60]. Large ES cuts independently increase the risk for bleeding as opposed to limited ES and do not improve therapeutic effect [61]. Balloon dilation diameter should not exceed that of the distal CBD to avoid increased risk for perforation [62]. Studies are now showing that EPLBD alone is comparable to EPLBD following ES regarding the initial stone extraction, the need for ML, the risk of PEP, and the total procedure time [63, 64]. Additional comparison studies are needed before this becomes common practice.

Electrocautery

The type of electrocautery current used during ES influences the degree of thermal tissue injuries and by extension risk for PEP. Pure cutting improves cutting ability and produces less edema while low voltage coagulating current is better for hemostasis. A mixed current combines both types either simultaneously in a blended cut or in a pulsed alternating fashion (i.e., endocut or pulsecut mode). A meta-analysis of four RCTs showed increased bleeding with pure cut compared to mixed current with similar rates of pancreatitis [65]. Compared to blended current, endocut and pulsecut mode produces fewer uncontrolled cuts and bleeding though rates of perforation and PEP do not differ [66,67,68,69].

Pancreatic Stents

Pancreatic stents have long been studied for PEP prevention. They maintain anterograde flow of enzymes and reduce intraductal pressure following pancreatic duct trauma. They reduce the incidence of PEP and nearly eliminate the development of severe PEP (OR 0.22–0.26) [70,71,72,73,74,75]. Yet, only 4% of average to high-risk ERCPs includes a prophylactic PS [76]. Practitioner reluctance is attributed to inexperience and concern for induced PEP [77, 78]. Failed placement (~ 5–10%), stent migration, and ductal perforation are other concerns [79]. Most published success rates are from high-volume tertiary centers with expert endoscopists.

Which patients are likely to benefit from prophylactic pancreatic stents is of ongoing discussion. Their efficacy has been thoroughly demonstrated in high-risk cases (i.e., inadvertent PD cannulation/wire insertion or precut ES). The data is less clear for low- to average-risk individuals [72, 75, 80]. Most of the meta-analyses published over the last decade highlighting their effectiveness excluded non-high-risk patients [70,71,72,73,74,75]. However, three such studies did show benefit in average-risk individuals (OR 0.21 and 0.25) [70, 71, 73]. One also demonstrated a significant risk reduction in unselected mixed-risk groups (RR: 0.23; 95% CI: 0.08–0.66) [73]. A more recent RCT involving 4 European tertiary referral centers and 167 unselected patients who had inadvertent pancreatic cannulation during first-time ERCP also found that those randomized to pancreatic stenting had significantly reduced PEP rates compared to those observed (OR: 0.43; 95% CI: 0.19–0.98; P = 0.04; NNT: 8.1) [81•]. The ESGE recommends prophylactic PS placement in selected patients at high risk for PEP (inadvertent guide-wire insertion of the PD and with DGWT) [7••].

A 5-French (Fr) PS appears more efficacious than a 3-Fr (96.9% vs 3.1%) [82]. This is attributed to easier placement, better decompression, lower rates of early dislodgement, and decreased need for endoscopic retrieval [83, 84]. Stents with a duodenal flange or pigtail that are also devoid of an internal flange should be used to reduce intraductal migration and facilitate spontaneous passage, respectively [7••, 85]. Stent length has less of a clear impact [86, 87]. Stents should be left for a minimum of 12–24 h though most are kept for 5–10 days before re-evaluation for spontaneous passage. Earlier removal negates their protective effects [88, 89]. While pancreatic stents are effective in high-risk cases, certain questions still remain. Ideal stent characteristics, the risk of attempting to stent a difficult PD, and the consequences of failed placement should be weighed carefully.

Rectal NSAIDs

Rectal nonsteroidal anti-inflammatory drugs (NSAIDs) have become a first-line modality for PEP prevention after a landmark trial in 2012 demonstrated lower PEP rates and decreased severity of pancreatitis in high-risk patients who received post-procedural indomethacin compared to placebo. While a majority of these patients simultaneously received a PS, post-hoc analysis suggested NSAIDs to be beneficial regardless of concurrent PS placement [90••]. Numerous meta-analyses have since evaluated their efficacy with an overwhelming majority showing reduction in PEP incidence (OR 0.24 to 0.63) [91,92,93,94,95,96]. This effect has been demonstrated with both average-risk and high-risk procedures as well as unselected patients. Their use primarily reduces the incidence of mild PEP (NNT 8 to 21) and to a lesser extent moderate-to-severe PEP (NNT 33–39) [91,92,93,94,95,96]. Diclofenac and indomethacin are both effective. The most frequently studied dose is 100 mg [91]. Rectal administration is superior to alternative routes [96,97,98,99,100,101,102]. In a recent prospective analysis of a RCT, 409 moderate-to-high risk patients undergoing ERCP were randomized to 100 mg of rectal diclofenac monotherapy 30 min before (n = 346) or after (n = 63) based on endoscopist preference. Those who received pre-procedural treatment had lower PEP rates (8% vs 18%; RR: 2.32; 95% CI: 1.21–4.46; P = 0.02) shortened hospital stays (1 day; interquartile range [IQR] 1–2 days vs 1 day; IQR 1–4 days; P = 0.02) and were less likely to be admitted to the ICU (0.3% [1 patient] vs 6% [4 patients]; P = 0.002) [103•]. Overall, the efficacy of rectal NSAIDs along with their excellent safety profile has made them an attractive prophylactic measure for most field experts. ESGE guidelines recommend routine use of 100 mg pre-procedural rectal indomethacin or diclofenac in all patients unless a contraindication exists. The ASGE also recommends their use in all high-risk patients and consideration in average-risk patients [7••, 104].

PS vs. Rectal NSAIDS

The emergence of rectal NSAIDs has led to question the ongoing need for pancreatic stents. Comparison trials are now emerging (Table 2). In a recent abstract, 321 patients with naïve papilla were randomized to receive a prophylactic PS, 50 mg of rectal diclofenac, or both. Five patients (PS 2/101, NSAID 1/106, and combination 2/102) developed mild PEP with an overall occurrence of 1.6% and no significant difference among groups [105]. NSAIDs were concluded to be noninferior though more information is needed regarding study inclusion/exclusion criteria, biliary access technique, and frequency of PD cannulation. Conclusions based on naïve papilla alone cannot be extrapolated to patients of moderate or high risk. A noninferiority RCT compared pancreatic stenting plus pharmacological prophylaxis against pharmacological monotherapy for PEP prevention in high-risk individuals (n = 414). PEP incidence was similar (12.6% vs 15.9%) and NSAID monotherapy again deemed noninferior [106•]. However, the difference in PEP rates was 3.3% with an upper boundary 95% CI of 10.2% which is greater than the prespecified noninferiority margin of 5%. Thus, the study actually failed to show noninferiority of pharmacological monotherapy prophylaxis. In contrast, a large network meta-analysis of 29 trials (n = 7862) comparing four preventative strategies against PEP in high-risk patients found that pancreatic stenting had the highest SUCRA probability (0.81; 95% CI: 0.80–0.83) of being ranked the best prophylactic treatment [107•]. A second meta-analysis showed that only pancreatic stents reduced the risk of moderate and severe PEP in both average and high-risk patients compared to NSAIDs and placebo [108•]. In one of the largest exploratory meta-analyses to date involving 55 RCTs (n = 7062), 20 PEP interventions including combination therapies were analyzed and efficaciously graded for PEP prophylaxis. Pancreatic stents, rectal NSAIDs, and rectal NSAIDs plus standard hydration were all associated with reduced odds for PEP compared to placebo. The GRADE confidence rating was low to moderate for 98.3% of the pairwise comparisons [109•].

Table 2 Comparison trials of pharmacotherapy, fluids, and pancreatic stents for PEP prevention
Full size table

Over the last decade, the use of rectal NSAIDs in the USA has steadily increased (though remained < 50% as of 2018) with a simultaneous abrupt decline in the use of pancreatic stents from 40.7% in 2013 to a nadir of 3% in 2017 [110•]. With this, PEP rates have increased from 4.3% in 2011 to 5.2% in 2017 (OR: 1.23; 95% CI: 1.04–0.46; P = 0.016). Mortality rates from PEP have nearly doubled from 2.8 to 4.4% (OR: 1.62; 95% CI: 1.10–2.38; P = 0.014) [111•]. Whether this is related to inadequate use of rectal NSAIDs, a decline in use of pancreatic stents or both is not entirely clear. The SVI (stent vs indomethacin) trial is an ongoing multicenter (nine academic medical centers in the USA), double-blinded, non-inferiority study analyzing PEP prevention rates in 1430 high-risk patients receiving indomethacin and PD stenting versus indomethacin monotherapy. These results are eagerly expected to reach a verdict on this quintessential debate [112••].

Intravenous Aggressive Hydration

Intravenous (IV) aggressive hydration (AH) improves the hemodynamics and microcirculation of the pancreas and is increasingly gaining traction for PEP prevention. A pilot study demonstrated that in patients undergoing first time ERCP, AH with LR at 3 ml/kg/h intraoperatively followed by a post-operative 20 ml/kg bolus and fixed infusion at 3 ml/kg/h for 8 h versus standard hydration (SH) resulted in significantly lower PEP rates (0% vs 17%; P = 0.016) [113•]. Several meta-analyses have since confirmed AH to be a protective factor in PEP reduction compared to SH [114,115,116]. It has also been associated with reduced PEP severity and length of hospital stay [117, 118]. However, there appears to be no added benefit in patients already receiving rectal NSAIDs. A large multicenter RCT involving 22 Dutch hospitals randomly assigned moderate-to-high risk patients (n = 826) to a combination of AH and rectal NSAIDs (100 mg diclofenac or indomethacin) or rectal NSAID and SH. The incidence of PEP was similar (8% vs 9%, respectively; RR: 0.84; P = 0.53) and there was no difference in serious adverse events, number of ICU admissions, or 30-day mortality [119•]. Of note, AH varied slightly from what was originally defined and SH included normal saline (NS) rather than LR. The major limitation to AH is that it is too time- and resource-intensive to be adopted in the outpatient setting. Additionally, fluid overload has been a reported side effect even after excluding patients at risk for this complication [120].

Alternative Therapies

Alternative pharmacotherapies for PEP prevention have been evaluated as well (Table 3).

Table 3 Alternative medical therapies for PEP prevention
Full size table

Nitrates

Multiple meta-analyses have supported the use of sublingual nitrate to prevent PEP [121,122,123]. Efficacy appears limited to high-incidence groups and does not reduce PEP severity.

Somatostatin

A recent meta-analysis showed that a long-term somatostatin infusion of 10–24 h reduced PEP rates in high-risk individuals. No benefit was observed in low-risk patients or with short infusions or bolus dosing [124].

Protease Inhibitors

While initial studies showed promising use of Nafamostat, a recent multicenter RCT showed no statistically significant change in PEP frequency in Nafamostat groups compared to placebo [125,126,127].

Tacrolimus

The use of tacrolimus for PEP prevention is a new concept. In a study of 337 liver transplant patients who had undergone 937 ERCPs for biliary complications related to liver transplantation, a tacrolimus trough level > 2.5 ng/mL was associated with up to 79% reduced rates of PEP (OR 0.21; 95% CI 0.06–0.72; P = 0.01) [128].

Epinephrine

Although prior RCTs suggested that the use of topical epinephrine may be preventative for PEP, the results of the three recent RCTs have shown topical epinephrine to have no protective benefit and in fact may even increase PEP risk when combined with rectal NSAIDs [129,130,131].

Other Pharmacological Agents

Corticosteroids, calcium-channel blockers, antibiotics, heparin, and allopurinol among other drugs have proven ineffective for PEP prevention in prospective clinical trials and meta-analyses [132,133,134,135,136,137,138,139].

How We Approach Prevention of PEP?

As research continues to progress so do the techniques and strategies for PEP prevention. In our high-volume tertiary care referral center for pancreaticobiliary diseases, we perform over 1500 ERCPs a year. For diagnostic studies, we prioritize the use of advanced imaging techniques including MRCP, endoscopic ultrasound, CT imaging, and abdominal ultrasound whenever possible. Prior to any ERCP, we perform a thorough assessment to determine all possible risk factors (patient and procedural) for PEP. Rectal indomethacin is administered pre-ERCP for patients of all risk levels unless a clear contraindication exists. Patients with contraindications receive AH with LR (at 3 ml/kg/h intraoperatively followed by a post-operative 20 ml/kg bolus and infusion at 3 ml/kg/h for 8 h). In cases of inadvertent PD cannulation, we use a combination of DGWT and WGC-PS depending on a case-by-case basis. We place a PS for PEP prophylaxis whenever inadvertent pancreatic duct cannulation takes place. We recommend the use of 5-Fr polyethylene stents to be kept in place for 1–2 weeks prior to assessment of spontaneous passage. In patients who are undergoing ERCP for primary biliary indications, pancreatic duct injection is minimized.

Conclusions

Though our knowledge behind the risk factors and mechanisms of PEP continues to expand, its incidence remains high, and it is still considered the most common adverse event of ERCP. A multifactorial preventative approach is essential to minimizing this risk. This includes careful patient selection, comprehensive assessment of PEP risk factors, tailored procedural techniques, and evidence-based prophylactic measures. Preventative procedural techniques include WGC, minimized cannulation attempts and contrast injections, and placement of a temporary PS in high-risk patients. Proven effective medical measures include the routine use of pre-procedural rectal NSAIDs and AH with LR when possible. Numerous other pharmacological measures are of ongoing research and discovery. Large-scale prospective RCTs are still needed to further evaluate additional treatment modalities.