Complication
Frequency
Pancreatitis
5–25 %
Bleeding
1–2 %
Perforation
<1 %
Cholecystitis
<1 %
Cholangitis
Rare
Death
Rare
Post-ERCP Pancreatitis
Post-ERCP pancreatitis (PEP) is the most common complication of ERCP. It has been reported to occur after 5–30 % of ERCP, depending on patients, procedures, study definitions, and methodology. Most studies report incidence of PEP of about 7 % [3]. Risk factors for PEP have been well defined and are discussed below. When more than one risk factor is present the risk is typically compounded rather than simply additive. Understanding risk factors is a critical piece in reducing incidence of complications and improving outcomes.
Definition
Uniform standards in diagnosis of pancreatitis are essential. The Atlanta criteria-based definition of pancreatitis is widely accepted: [4] two of three criteria are required to diagnose pancreatitis, including (1) abdominal pain in the epigastric region with or without radiation to the back, (2) at least threefold elevation of serum amylase and or lipase, and (3) imaging features suggestive of pancreatitis. These criteria are very similar to the original Cotton criteria with the addition of greater than three times elevation of serum lipase. Some clinicians routinely check serum amylase and/or lipase following ERCP, and elevations of serum enzyme levels may occur in the absence of pain. This does not represent pancreatitis, and is often referred to as “biochemical pancreatitis”; clinically these patients generally do well with no further intervention. In the setting of postprocedure pancreatitis, routine cross-sectional imaging to confirm the diagnosis is not necessary unless imaging is required for other reasons. Severe pain in the absence of significant elevation of serum lipase and/or amylase should be carefully evaluated and prompt a search for other complications such as perforation. It is not uncommon for some patients to complain of postprocedural pain in the absence of any detectable complication. The most problematic to assess are patients with preprocedure pain and equivocal enzyme rises post-ERCP, such as more than three times lipase elevation and less than three times amylase elevation, which is a common scenario.
Post-ERCP pancreatitis can range from mild interstitial to severe necrotizing with multiorgan failure and even death. The severity of post-ERCP pancreatitis is graded according to consensus definitions depending on the duration of hospitalization and need for intervention [4, 5]. Pancreatitis is considered mild if hospitalized for 2–3 days, moderate for 4–10 days, and severe for >10 days hospitalization, development of necrosis or pseudocyst, and/or performance of a drainage procedure or surgery.
What are Potential Mechanisms of Post-ERCP Pancreatitis?
The exact mechanism by which the pathway of inflammation is initiated is unclear. There are several possible explanations and some of the strategies for prevention of postprocedural pancreatitis are based on these postulates. There is some indirect evidence that the following factors play a role:
1.
Mechanical outflow obstruction: It is known from clinical data that instrumentation causes ampullary edema and likely mechanical outflow obstruction of pancreatic ductal drainage. This led to the concept of pancreatic duct stenting for prophylaxis which has resulted in significant reduction in post-ERCP pancreatitis.
2.
Ductal injury/trauma: Pancreatic ductal manipulation including passage of guidewires into the main pancreatic duct or side branches results in increased risk of pancreatitis. Any pancreatic ductal intervention appears to increase the risk. It is likely that ductal injury triggers the inflammatory cascade.
3.
Thermal injury: Pancreatic sphincterotomy increases the risk of pancreatitis. Access/precut sphincterotomy is also a risk factor, suggesting that thermal injury can initiate the inflammatory cascade.
4.
Hydrostatic injury: There appears to be a relationship between the perfusion of the duct and pancreatic inflammation. Pancreatic duct manometry with a perfusion catheter and no aspiration carries a high risk of post-ERCP pancreatitis . Repeated duct injections, depth of pancreatic duct injections, forceful injections/“acinarization” have all been implicated.
Other causes that have been postulated include the introduction of gut flora into the pancreatic duct and hence the “infectious theory,” and chemical injury or allergic response from contrast.
While the exact inflammatory pathway is unclear, there seems to be a significant interplay of mechanisms. Prevention of pancreatitis is aimed at halting one or more of these processes to reduce the severity, if not completely prevent pancreatitis. Pharmacological interventions target the blockage of inflammatory chemokines. One category of drug that has shown promising results is NSAIDs.
How Are Risk Factors Defined?
Several prospective studies have advanced our knowledge of risk factors for PEP. Risk has been stratified as definite, indefinite, or no risk based on the evidence. Definite risk factors have been confirmed by multivariate analysis in studies involving greater than 500 patients and proven statistically significant in more than one study or meta-analyses. Indefinite risk factors are those that were significant on univariate analysis in multiple studies or by multivariate analysis in a single study. If there is no evidence based on multivariate analysis and the data are inconsistent based on univariate analysis, these factors are thought to pose no additional risk (other than the baseline or background risk) [3].
What Are Risk Factors for Post-ERCP Pancreatitis?
1.
Patient-related risk factors
2.
Procedure-related risk factors
3.
Operator-related risk factors
Patient-Related Risk Factors
There are certain groups of patients who are at the highest risk of developing pancreatitis after ERCP, as confirmed by multiple cohort studies. Most clearly at risk are women with abdominal pain in the absence of common duct stones or other identifiable pathology, fitting into the category of “suspected sphincter of Oddi dysfunction (SOD) .” There is no evidence that type III SOD (pain only) are at any more risk than II or I SOD (those with dilated bile ducts or/and abnormal liver chemistries). Those with a prior history of post-ERCP pancreatitis also have a higher risk (Table 3.2).
Table 3.2
Patient-related risk factors for post-ERCP pancreatitis
Definitea | Probable | No risk |
|---|---|---|
Young age | Absence of CBD stone | Normal/small CBD diameter |
Female | Normal serum bilirubin | Periampullary diverticulum |
Suspected SOD | Pancreas divisum | |
Recurrent acute pancreatitis | Allergy to contrast medium | |
Absence of chronic pancreatitis | ||
History of post-ERCP pancreatitis |
Procedure-Related Risk Factors
Difficult cannulation is a known risk factor, probably because of induced papillary edema. Other risk factors include multiple pancreatic duct contrast injections. Data support that the extent of injection corresponds to the incidence of pancreatitis. Deep passage of a guidewire into the pancreatic duct has been shown to be a powerful risk factor [6]. Certain high-risk procedures including precut sphincterotomy or access papillotomy, balloon dilation of the bile duct especially without a biliary sphincterotomy, pancreatic sphincterotomy, and any pancreatic duct interventions are consistently associated with increased risk by multivariable analyses. Although there has remained concern over increased rates of pancreatitis with metal biliary stent placement, several studies including a small randomized trial failed to confirm this with uncovered and partially covered metal stents, [7–9] and biliary sphincterotomy before stent placement did not impact PEP. Despite these data, concern remains over potentially increased pancreatitis with use of fully covered metal stents. A small retrospective series reported that nonpancreatic cancer and injection of the pancreatic duct were risk factors for pancreatitis in patients with partially and fully covered metal stents placed [10]. The strength or osmolarity of the contrast plays no significant role in increasing the risk of pancreatitis. Degree of pancreatic opacification has shown to increase the risk. Despite popular opinion, acinarization of the pancreas did not pose any significant risk by multivariate analyses (Table 3.3) [11, 12].
Table 3.3
Procedure-related risk factors for post-ERCP pancreatitis
Definite | Probable | No risk |
|---|---|---|
Pancreatic duct injection | Pancreatic acinarization | Intramural contrast injection |
Pancreatic sphincterotomy | Pancreatic brush cytology | Diagnostic vs. therapeutic |
Balloon dilation of intact sphincter | Pain during ERCP | Biliary sphincterotomy |
Difficult/failed cannulation | Prior failed ERCP | |
Precut sphincterotomy | Sphincter of Oddi manometry (esp. aspiration catheter) |
Physician (Operator)-Related Risk Factors
Data from various studies suggest that endoscopist case volume and experience is inversely proportional to the risk of complications [13]. One study showed that trainee involvement was associated with increased risk [14]. Presence of multiple risk factors in a single patient has a compounding effect on risk. Thus a young woman with suspected SOD , normal liver functions and normal common bile duct diameter would have the highest risk [15]. The odds ratios for post-ERCP pancreatitis of some common risk factors as calculated based on various prospective studies and meta-analyses are summarized in Table 3.4.
Table 3.4
Common risk factors and odds ratios for pancreatitis based on available data
Risk factor | Odds ratio (95 % CI) |
|---|---|
Female gender | 2.23 (1.75, 2.84) |
Suspected SOD | 4.09 (3.37, 4.96) |
History of recurrent acute pancreatitis | 2.46 (1.93, 3.12) |
Pancreatic duct injection | 2.20 (1.60, 3.01) |
Pancreatic sphincterotomy | 3.10 (1.60, 5.80) |
Precut sphincterotomy | 2.71 (2.02, 3.63) |
Balloon dilation of intact sphincter | 4.50 (1.50, 13.5) |
How to Prevent Post-ERCP Pancreatitis
Careful Patient Selection
Diagnostic ERCP or ERCP for “suspicion” of most diseases is now obsolete and should be avoided. As an example, noninvasive or less invasive techniques including MRCP, endoscopic ultrasound , and intraoperative cholangiography during cholecystectomy provide similar information, which may obviate the need for ERCP. On the other hand, if there is biochemical, radiological, and/or clinical support for choledocholithiasis , then an ERCP first followed by cholecystectomy is a reasonable approach.
Appropriate Physician (Operator) Experience
The endoscopist should be familiar with his or her own limitations and the type of therapeutic procedure that is required. The endoscopist must be capable of recognizing and handling unplanned events. Ability to place prophylactic pancreatic stents is a prerequisite of ERCP [16]. Consistent placement of pancreatic stents often requires use of small diameter wires (0.018″, 0.021″, or 0.025″).
Careful Procedure Techniques
It is recommended to avoid or minimize the extent of pancreatic duct opacification. Any contrast injection should be done under fluoroscopic guidance, and contrast should be gently injected a small amount at a time especially if opacifying the pancreatic duct inadvertently or unintentionally (Table 3.5).
Patient selection | Technical considerations | Pharmacological methods |
|---|---|---|
Avoid ERCP for marginal/weak indication—consider alternatives including EUS/MRCP/IOC | Efficient cannulation (including judicious use of guidewires) | Rectal indomethacin/diclofenaca |
Avoid unintended pancreatic duct cannulation/opacification | ||
Placement of pancreatic stents prophylactically (preferably small bore and soft stents) for high-risk patients |
Guidewire Cannulation
Guidewire cannulation was proposed as a way to minimize contrast injection and reduce the risk of pancreatitis . Cannulation techniques have been described elsewhere in the book. By using the guidewire instead of contrast, one can advance the wire into the desired duct. If the wire crosses over the spine, it is thought to be in the pancreatic duct while if the wire advances up along the spine, it is believed to be in the bile duct . Once the wire is passed in the direction of the bile duct, the cannula is advanced into the duct and contrast injected. There are at least 12 randomized controlled studies comparing guidewire cannulation to the standard technique using contrast. A recent meta-analysis of these published trials suggests that the guidewire cannulation technique reduces the risk of PEP with a risk reduction ratio (RR) of 0.51 [95 % confidence interval (CI) 0.32–0.82]. Cannulation success was also more successful with guidewire cannulation [17]. Guidewire cannulation appears to reduce risk of pancreatitis not only by avoiding contrast injection into the pancreatic duct but also by likely reducing papillary trauma owing to the smaller diameter of the wire compared to the cannula used for cannulation. The problem with the published studies of guidewire cannulation is that the control groups used a technique of cannulating and injecting contrast without use of a guidewire, which is long antiquated and does not represent a realistic alternative. Guidewire cannulation also does not ensure safety. There are concerns for intramural dissection, ductal injury, trauma, or perforation, especially of the side branches. Care should be taken not to push wires, especially if passage is difficult. It is reasonable to inject a small amount of contrast to delineate the duct when in doubt as to the location of the tip of the wire, rather than to cause ductal injury or dissection by forcefully advancing the wire [18, 19]. If biliary access is the goal, but repeated passage of the guidewire occurs into the pancreatic duct, or even perhaps once in a high-risk patient, it is ideal to leave the wire in the pancreatic duct and cannulate the bile duct alongside this wire (dual guidewire cannulation technique). Double wire access should be followed by prophylactic pancreatic duct stent placement, as shown in a randomized trial [20].
Prophylactic Pancreatic Duct Stent Placement
Pancreatic duct stents have been proven effective at reducing risk of post-ERCP pancreatitis . It is thought that papillary edema from ERCP can impede the flow of pancreatic secretions. The hypothesis is that placement of a stent across the pancreatic sphincter would preserve flow of pancreatic secretions and thereby minimize the risk of post-ERCP pancreatitis. Since the initial reports demonstrating the benefits of pancreatic duct stenting in high-risk patients, numerous well designed studies and meta-analyses have assessed the value of prophylactic stenting, and currently stenting has the best evidence as a strategy to reduce risk of PEP [21, 22]. In the latest meta-analysis including 14 studies, pancreatic duct stent placement was associated with a statistically significant reduction of PEP (RR 0.39; 95 % CI 0.29–0.53; p<0.001). Subgroup analysis stratified according to the severity of PEP showed that a stent was beneficial in patients with mild to moderate PEP (RR 0.45; 95 % CI 0.32–0.62; p<0.001) and in patients with severe PEP (RR 0.26; 95 %CI 0.09–0.76; p = 0.01) [23–25].
Patients shown to benefit from pancreatic stents include those with one of the following characteristics: sphincter of Oddi dysfunction (suspected or documented, regardless of manometry findings); difficult cannulation involving pancreatic instrumentation or injection; aggressive instrumentation of the pancreatic duct (e.g., brush cytology) ; pancreatic guidewire placement during biliary cannulation; pancreatic sphincterotomy (major or minor papilla); precut sphincterotomy starting at papillary orifice; balloon dilation of intact biliary sphincter; prior post-ERCP pancreatitis; and endoscopic ampullectomy (Table 3.6). While there is overwhelming evidence for placement of stents in those with patient- and procedure-related risk factors, there are only two studies to assess the utility of pancreatic duct stenting for low-risk patient and procedures: interestingly, both showed a positive effect [22].
Table 3.6
Pancreatic stent placement: when and when not
Indicated (based on evidence) | Not indicated |
|---|---|
Sphincter of Oddi dysfunction (suspected or documented, regardless of manometry findings) | Lower-risk patients (older or with obstructed pancreatic duct) undergoing a low-risk procedure |
Difficult cannulation involving pancreatic instrumentation or injection | Pancreatic duct not injected with contrast material and limited guidewire manipulation in low-risk patient |
Aggressive instrumentation of pancreatic duct (e.g., brush cytology) | Needle-knife precut or fistulotomy starting above the orifice in absence of other risks |
Pancreatic guidewire placement during biliary cannulation | Doubtful feasibility of successful pancreatic wire access and stent placement |
Pancreatic sphincterotomy (major or minor papilla) | Biliary therapy in patients with pancreas divisum |
Precut sphincterotomy starting at papillary orifice | |
Balloon dilation of intact biliary sphincter | |
Prior post-ERCP pancreatitis | |
Endoscopic ampullectomy |
There are clear downsides to pancreatic stent placement (Table 3.7). Not all endoscopists are trained or familiar with pancreatic duct stenting both in terms of indications and techniques. Training/simulation models for practicing placement and removal of pancreatic stents might be useful. Familiarity with specialized guidewires and pancreatic stents and enhanced understanding of pancreatic duct anatomy are required. The endoscopist must be aware of potential complications associated with either stent placement or failed attempts at prophylactic pancreatic stent placement. Failed placement is associated with increased risk. When placing pancreatic stents, guidewire or stent-related ductal perforation is possible. Inward delivery or stent migration may occasionally occur. Finally, stent-induced pancreatic duct or parenchymal injury may occur, even occasionally following short-term pancreatic duct stenting for prophylaxis [26].
Table 3.7
Challenges to pancreatic duct stent placement
Education of endoscopists regarding indications and applications |
Need for training in techniques of pancreatic stent placement |
Familiarity with specialized guidewires and pancreatic stents |
Enhanced understanding of pancreatic duct anatomy |
Appropriate follow-up to ensure stent passage or removal |
Awareness of potential complications |
Failed placement |
Guidewire/stent-related ductal perforation |
Inward delivery or stent migration |
Stent-induced pancreatic duct or parenchyma injury |
Technique of Pancreatic Stent Placement
When placing a pancreatic stent , vigorous manipulation of the wire in the pancreas should be avoided, since it can lead to side branch perforation and thus increase risk of pancreatitis. Although many endoscopists use 0.035-in. wires for general use, many experts use a 0.018–0.021-in. guidewire for pancreatic stent placement, and these are a prerequisite for small-caliber (3F) stent placement. Although a randomized trial failed to demonstrate that 3F stents were superior to 5F stents [27], 0.018-in. wires necessary for 3F stent placement were only passed after randomization. In addition, 0.035-in. wires may not be suitable for tiny or tortuous pancreatic ducts. The authors do not recommend passage of large-bore wires and placement of large-bore stents especially in small-caliber or tortuous pancreatic ducts.
Stents are made of different materials with some stents being softer than others. Intuitively softer stents without inner flanges should conform to the ductal configuration and cause less trauma and ductal injury than rigid flanged stents, although they have never been formally compared. Data are clear that larger stent diameter is associated with a significantly higher risk of ductal injury [28]. For prophylactic stents in high-risk patients, the authors recommend either short (2–3 cm), soft 4–5Fr, inner flanged stents or long (9–11 cm) soft 3F or 4F unflanged stents with a single pigtail. Patients should have an abdominal radiograph within 2–4 weeks, which preferably should be checked by the gastroenterologist since inexperienced radiologists may not readily recognize small pancreatic stents . If a stent remains at follow-up , it should be removed endoscopically. There are rare reports of pancreatitis following removal of pancreatic stents but this occurred mostly with stents having internal flanges.
One special situation occurs when the pancreatic duct takes a 360° loop in the head of the pancreas, the so-called ansa loop. In these situations and similar difficult ductal configurations, it is not possible to pass the wire deep into the duct (Video 3.1). If one can use a small-caliber 0.018-in. guidewire and create a “knuckle” or a “j” shaped intentional hook to the wire which is inserted as little as 2 cm into the duct, a 2 cm long 4 or 5Fr stent, preferably of soft material with an inner flange to avoid immediate outward migration, can be inserted. Immediate removal or passage of a pancreatic stent at the end of the procedure does not protect against post-ERCP pancreatitis compared with a stent that remains within the duct for at least a few days [29].
Can a Pill Prevent Post-ERCP Pancreatitis?
Many pharmacologic agents have been tested that could potentially work at various stages of the inflammatory cascade leading to pancreatitis. To date, at least 48 randomized controlled studies have been reported utilizing 15 different agents with most studies including patients at average or mixed risk for post-ERCP pancreatitis . At least six studies included high-risk patient populations [30]. Drugs that have been evaluated and their efficacy are listed in Table 3.8.
Table 3.8
Pharmacological prevention of post-ERCP pancreatitis
Effective | Ineffective | Possibly effective |
|---|---|---|
Rectal NSAIDs | Calcium channel blockers | Topical nitroglycerine |
Gabexate infusion (> 12 h) | Topical liodcaine | Nafamostat |
Corticosteroids | Antibiotics | |
Allopurinol | Somatostatin (12–24 h infusion) | |
PAF inhibitors | Topical epinephrinea | |
IL-10 | ||
Heparin derivatives | ||
Octreotide
Related posts: Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|
