Synopsis  

ERCP is the most risky procedure that endoscopists perform on a regular basis. There is potential for technical and clinical failure, and for misdiagnosis, and some small risk to staff, but the main interest is in the risk for adverse clinical events. A consensus definition of complications, and of their severity, and a series of careful prospective studies, has clarified the degree of risk in different circumstances, and the relevant risk factors. This process has allowed a clearer picture to emerge of the risk–benefit ratios in different clinical scenarios, and a greater ability to advise patients about their options. Also, the extensive experience of the last 30 years permits authoritative statements on how to minimize the likelihood of complications, and how to deal with difficult situations when they arise.

 Introduction   

ERCP has become popular worldwide because it can provide significant benefit in many clinical contexts. Sadly, it has also caused considerable harm in a small number of patients. Thus, it is crucial for practitioners and potential patients to understand the predictors of benefit, and of risk. Defining positive and negative outcomes has been a significant challenge [1–4], but much useful information has been gathered from increasingly sophisticated outcomes studies over the last two decades.

This contribution will concentrate on the risks and risk factors, will emphasize ways to reduce them, and will provide guidance about management when adverse events occur.

 The risks of ERCP   

The concept of 'risk' indicates that something can 'go wrong', and is therefore best defined as a deviation from the plan. This assumes that a plan has been clearly formulated. The patient's perspective and understanding of the plan is enshrined in the process of informed consent. Deviations are best described generically as 'unplanned events'[4].

Unplanned events of ERCP are of four types:

• risks to staff;
  
• technical failure;
  
• clinical failure;
  
• unplanned adverse events—complications.
  

Risks for endoscopists and staff   

The endoscopy unit is not a dangerous place, but there are a few risks for the ERCP endoscopist and staff.

The possibility of transmission of infection exists, but should be entirely preventable with standard precautions (gowns, gloves, and eye protection) and assiduous disinfection protocols (Endoscopy Practice and Safety Chapter 7).

Certain immunizations are also appropriate. Rarely, staff may become sensitive to materials used in the ERCP process, such as glutaraldehyde, or latex gloves.

The risks of radiation are minimized by appropriate education, shielding, and exposure monitoring [5].

Many older endoscopists have neck problems caused by looking down fiberscopes, a situation aggravated by ERCP rooms where the video and X-ray monitors are not side by side. Busy ERCP practitioners sometimes complain also of 'elevator thumb'. A Canadian survey found that more than half of 114 endoscopists performing ERCP had some attributable musculo-skeletal problem [6].

Technical failure   

Not all ERCP procedures are successful technically. It may prove impossible to reach the papilla, to gain access to the duct of interest, or to complete the necessary therapeutic maneuvers. The chance of failure depends upon several factors.

Expertise   

An important determinant of the chance of success is the level of expertise (of the endoscopist and team). There are now good data to show that more active ERCP endoscopists have better results [7], as applies in surgery [8].

Complexity   

The risk of technical failure increases with the complexity of the problem. Any procedure can turn out to be technically challenging (e.g. when the papilla is hiding within a diverticulum), but some can be expected to be difficult beforehand (e.g. in patients who have previously undergone Billroth II gastrectomy). The concept of a scale of difficulty was first published by Schutz and Abbot [9]. Modifications led to a scale with three levels [4].

Degree of difficulty scale for ERCP procedures (Fig. 1)   

Level 1   

Standard procedures which any endoscopist providing ERCP services should be able to complete to a reasonable level of competence (say 90%). This includes deep selective cannulation, diagnostic sampling, standard biliary sphincterotomy, removal of stones (up to 10 mm in diameter), and the management of low biliary obstruction and postoperative leaks.

Level 2   

Advanced procedures which require technical expertise beyond standard training, for example cannulation of the minor papilla, diagnostic ERCP after Billroth II gastrectomy, large stones (needing lithotripsy), and the management of benign biliary strictures and hilar tumors.

Level 3   

Tertiary procedures, which are normally offered only in tertiary referral centres, such as Billroth II therapeutics, intrahepatic stones, complex pancreatic treatments, and sphincter manometry. Manometry is included at the tertiary level, not because it is technically challenging, but because the overall management of patients with suspected sphincter dysfunction is particularly difficult (and the risks are greater).

Defining intent   

A confounding issue when trying to assess technical success or failure is how well the goal of the procedure is, or can be, defined beforehand [1,4]. When the intent is obvious, e.g. to remove a known stone, the resulting outcome is unequivocal. However, ERCP is often used to make or confirm a diagnosis, and then to perform treatment 'if appropriate', so that defining intent, and thus 'success' and 'failure', may be more subjective. Also, endoscopists have different thresholds for attempting therapy. Some may back away from a large stone, and count the case as a success for good judgement rather than as a technical failure. Treatment will not even be considered if the diagnosis is not made (e.g. if cannulation fails and a stone is missed), but such a case usually will not be counted as a failure of stone treatment [10,11]. Thus, the success literature should be viewed with some skepticism.

Risk consequences of technical failure   

There are good data showing that failed procedures carry more complications than successful ones. Failure usually necessitates repeat ERCP, or a percutaneous or surgical procedure, which bring additional and significant costs and risks [12]. Strictly speaking, on an 'intention to treat basis', any complications of these subsequent procedures should be attribute to the initial ERCP attempt.

Clinical failure   

Clinical success is dependent upon technical success, but the reverse is not necessarily true. A procedure may be completed technically in an exemplary fashion, but with no resulting benefit. This would be true certainly when the indication is not appropriate.

Our aim is to make patients 'better', but defining precisely what that means can also be a challenge [1,4,13]. In some contexts (e.g. stone extraction, biliary stenting for low tumors) it is reasonable to assume that technical success will almost guarantee clinical success, at least in the short to medium term. However, some of those patients will have recurrent problems (e.g. new stones and stent occlusion), as detailed later, so that the time frame of measurement is relevant to success. It may be helpful to distinguish between initial 'primary' failure, and 'secondary' failure, which means a relapse of the same problem.

It is also difficult to measure the success or failure of interventions in patients who have intermittent problems such as recurrent pancreatitis, or episodes of pain suspected due to sphincter dysfunction. The true outcome in these cases can be measured only after months or years. Furthermore, the clinical response may be incomplete, with a reduction, but not elimination, of attacks of pancreatitis, or some diminution in the overall burden of pain. The question then is how precisely to measure this 'pain burden' (which may fluctuate from day to day, or week to week), and how much of a reduction constitutes 'success'? Progress in this area will come only if we have carefully defined outcome metrics, good baseline evaluation, and structured objective follow-up [13]. Quality of life assessment should feature in these contexts. We are developing a 'Pain-burden' scoring tool. This is used to follow patients sequentially, and incorporates our validated Digestive Quality of life instrument, the DDQ-15 [14].

 Unplanned adverse clinical events—complications   

Unplanned events are deviations from the expectations of the endoscopist, and of the patient (as defined by the process of informed consent). Rarely, the outcome of a procedure may be better than anticipated, for example, finding a treatable benign lesion (such as a stone) in a jaundiced patient with suspected malignancy. However, most unplanned clinical events associated with procedures are unwelcome, and are often called 'adverse events'. Some are significant enough to be called 'complications'[4].

When does an event become a complication?   

Some adverse events are relatively trivial (such as brief hypoxia easily managed with supplemental oxygen, or transient bleeding which stops or is stopped during the procedure). The word 'complication' is not appropriate for these events, not least because of the medico-legal connotations. However, all unexpected and adverse events should be documented and tracked for quality improvement purposes.

The level of severity at which an adverse clinical event becomes a 'complication' is an arbitrary decision, but an important one, since definitions are essential if meaningful data are to be collected and compared. A consensus workshop defined complications of ERCP in 1991 [15]. Whilst the document focused on the complications of sphincterotomy, the principles and definitions apply to all aspects of ERCP.

Complication definition   

• An adverse event.
  
• Attributable to the procedure.
  
• Requiring treatment in hospital.
  

Not all complications are of equal significance, so the workshop also recommended an arbitrary scale of severity, based mainly on the length of hospitalization required, and the need for intensive care and/or surgery (Fig. 2).

Severity criteria   

• Mild: 1–3 nights in hospital.
  
• Moderate: 4–9 nights.
  
• Severe: 10 nights or more, or surgery, or ICU admission.
  
• Fatal: death attributable to the procedure.
  

The workshop also recommended working definitions of the commonest complications (Fig. 3).

These concepts and definitions have been adopted widely, and have been used in many subsequent studies of ERCP outcomes. This has helped considerably in the attempt to better understand the predictors of good and bad outcomes. If surgeons and interventional radiologists used a similar lexicon, it would be easier to compare their outcomes outside the context of formal randomized trials [13].

Types of adverse clinical events   

Unplanned adverse events can be categorized broadly into four groups

• Equipment malfunction.
  
• Medication and sedation issues.
  
• Direct events: those which occur at sites which have been traversed or treated during the endoscopic procedure (e.g. perforation, bleeding or pancreatitis).
  
• Indirect events: those which occur in other organs (e.g. heart, lungs, and kidneys) as a result of the procedure. Indirect events are more difficult to recognize and document because they may not become apparent until several days after the procedure, when the patient has returned home or to other clinical supervision.
  

Timing of events and attribution   

Most adverse events are recognized during or shortly after procedures, but some happen beforehand (e.g. as a result of some aspect of preparation), and some are apparent only later (e.g. delayed bleeding after sphincterotomy).

For adverse events which occur before and during procedures, it is important to note whether the examination had to be terminated early, or could be completed.

The 1991 consensus definition of complications [15] includes the phrase 'attributable to the procedure'. Attribution is not always clear-cut, especially when there is a delay. Is a cardiopulmonary event counted if it occurs a week or two after ERCP, or only if there is some other linking factor (e.g. some important medication was stopped for the procedure)?

To cover this point, the consensus workshop suggested that direct events (as defined above) are always attributable, even if they do not occur or become apparent for several weeks (e.g. delayed bleeding). However, we agreed an arbitrary time limit of 3 days for indirect events, such as cardiopulmonary problems.

As mentioned above, there is also the issue of how to report the complications of other procedures (e.g. percutaneous interventions) which become necessary when ERCP fails [12].

A dataset for unplanned events   

This is shown in Fig. 3.

 Overall complication rates   

Rates of complications published before the 1991 consensus definitions are difficult to interpret due to a lack of consistency in reporting [15–20].

Many reviews and case series have been published subsequently [21–44]. Some of the most recent single- and multicenter data are summarized in Figs 4 and 5.

Overall it appears that complications occur in some 5–10% of ERCP procedures. However, these global figures take no account of severity, and come from a huge variety of procedures performed on a broad spectrum of patients in different contexts. It is now clear that the risks vary considerably with the indication and setting, so that we need more focused data. Patients should be informed about the likely risk in their own precise context.

Accuracy of data collection   

An important issue affecting the accuracy of reported data is the method of collection. Retrospective studies are known to underestimate complication rates, since many delayed events are missed [45–47]. This may apply particularly to the large volume centres (who publish most) since the encounters often are brief and most patients return home, often some distance away, for further care. The most reliable data come from prospective studies which include a routine 30 day follow-up visit or call [44,45], but this is labor intensive and rarely done outside of research studies.

Changes in complications over time   

Bleeding, perforation, and infection were the most common complications of ERCP in the early days of ERCP and biliary sphincterotomy [15–20]; now pancreatitis dominates (Figs 4 and 5). This change appears to be due mainly to a progressive reduction in the risk of bleeding, perforation, and infection as training and techniques have improved, and maybe also to a relative increase in pancreatitis, as ERCP has been used more widely for more speculative (and risky) indications, such as obscure abdominal pain, sphincter dysfunction, and recurrent pancreatitis.

Complication rates at MUSC   

We have used the same definitions and database for the prospective recording of all endoscopic procedures at MUSC for more than 10 years. Delayed complications that we are aware of are reported to the group and added to the database at a weekly pancreatico-biliary service meeting, but there has been no routine follow-up call. From studies performed by ourselves [45], and others [46,47] it is certain that some delayed complications are not recorded, but the system has been consistent, so that trends are probably meaningful.

The overall rate of known complications in almost 10 000 procedures was 4.08%, with severe complications at 0.33%, and 7 deaths (0.07%) (Fig. 6). Pancreatitis has accounted for two-thirds of all recorded complications, occurring at a rate now of around 2.7%. The incidence of severe pancreatitis (more than 10 days in hospital, ICU admission, pseudocyst, or surgery) was 0.14%.

There has been a gradual reduction in the rates (and severity) of complications over the years (Fig. 7), despite an increasing proportion of complex and more risky level 3 procedures.

More details of specific complications and their management are given in the relevant later sections.

 General risk issues   

Endoscopists must be aware of the factors that can increase the risk of ERCP. These are both general and specific. General risks include the skill of the individual endoscopist (and team), the clinical status of the patient, and the precise nature of the procedure.

Details of the specific risks, methods to minimize them, and recommendations for management, are given below. Here we document some important general points.

Operator-related issues   

There are now significant data showing that more experienced endoscopists usually have higher success rates and lower rates of complications than those who are less active, even when dealing with more complex cases [7,24,28,41–43]. This fact has important implications for training, credentialing and informed consent. Lack of experience increases the risk of technical failure. Failures carry risks also of the subsequent needed interventions. In one analysis, failed ERCPs carried three times the complication rate of successful ones (21.5 vs. 7.3%) [12]. The association between inexperience and poor outcomes has been well documented for major surgical procedures [8].

Patient-related issues; clinical status, indications, and comorbidities   

Much attention has been paid to analysing the characteristics of patients which may affect the risk of performing ERCP [30,48].

Age   

Age itself is not a risk factor for ERCP complications [49]. Many studies now testify to the safety of performing diagnostic and therapeutic procedures in infants [50], children [51], and the elderly [52,53].

Illness and associated conditions   

Adverse events are more likely to occur in patients who are already severely ill, for example with acute cholangitis [54], and in those with substantial comorbidities. The most important comorbidities are cardiopulmonary fragility (posing risks for sedation and anesthesia), immunosuppression, and coagulopathies (including therapeutic anticoagulation). It would be helpful if there were an agreed index or score that reflected the degree of risk, but none of the published instruments really fit the ERCP context. The American Society for Anesthesiology (ASA) score is often used in surgical practice as a guide to the risk for sedation and anesthesia, but this appears unhelpful in the context of ERCP [55]. This is because the risk is much more dependant on the indication for the procedure.

ERCP appears to be safe when needed for management of stones in pregnant patients [56].

Indication   

Fortunately it is clear that the risks of ERCP are lowest in those patients with the 'best' indications, i.e. duct stones, biliary leaks, and low tumors. Conversely, the pioneering studies by Freeman and colleagues have revealed the substantial risks involved in performing ERCP in patients with obscure abdominal pain ('suspected sphincter dysfunction') [36]. This was emphasized strongly by the NIH 'State of the Science' conference on ERCP in 2002 [57]. Sadly, it is true that 'ERCP is most dangerous for those who need it least'[58].

Anatomical factors   

In some series, but not in all, the presence of a peripapillary diverticulum appears to be a risk factor [59,60]. With suitable precautions, patients with implanted pacemakers or defibrillators can be treated safely [61].

A normal-sized bile duct was earlier believed to increase the risk of post-ERCP pancreatitis [62,63], but this is a surrogate for sphincter dysfunction, and does not apply to patients with stones [64,65].

Complication-specific risk factors   

Patient risk factors for specific complications are detailed below. For example, Billroth II gastrectomy carries an increased risk for afferent loop perforation, and coagulopathies and certain medications increase the risk of bleeding. Equally, patients with hilar tumors and sclerosing cholangitis are at greater risk for septic complications because it may prove difficult to provide complete drainage.

Procedure performed   

Diagnostic or therapeutic?   

Most people assume that therapeutic ERCP is more dangerous than diagnostic procedures. This is true in several reported series: 5.4 vs. 1.4% [24], 9.1 vs. 1.8% [40], and 4.6 vs. 2.1% [37], but not in another small series (7.4 vs. 17%) [29] (Fig. 5).

Sedation, cardiopulmonary events, and intubation carry the same risks whether the procedure is diagnostic or therapeutic. Therapeutic procedures do carry their own specific risks, e.g. bleeding and perforation after sphincterotomy, or infection after attempted pseudocyst drainage. These complications can be serious, so it would also seem logical that the likelihood of a severe complication would be greater after therapeutic procedures. Remarkably, our own series shows very similar complication rates for diagnostic and therapeutic ERCP, and the risk of severe or fatal complications was actually slightly higher for diagnostic procedures (0.65 vs. 0.37%) (Fig. 8). It is worth noting that diagnostic ERCP in some patients may actually be riskier if not followed immediately by appropriate therapy (e.g. in a patient with malignant obstructive jaundice or proven sphincter dysfunction). Our 'diagnostic' procedures were simply those that involved no therapy, so some of them may have been technical failures. The statistics may well be different on an intention to treat basis.

The implications of the specific therapeutic procedures will be considered further when addressing the individual risks, but some details are given here.

Biliary sphincterotomy   

The commonest therapeutic ERCP procedure, performed in enormous numbers throughout the world. As a result, much of the risk literature refers specifically to biliary sphincterotomy [15,17,19,20,22,31,32,42–44,66], and mainly in the context of stones [47,66,65,67–69]. Representative series indicate an overall morbidity of 5.3–9.8%, with attributable mortality considerably below 1% (Fig. 5). Our overall complication rate for a series of 855 biliary sphincterotomies for stone at MUSC was 2.6%; in all other contexts combined (sphincter dysfunction, pancreatitis, stenting) it was 7.1%. Altogether there were 16 severe complications of biliary sphincterotomy (0.6%), with no deaths.

Pancreatic sphincterotomy   

Pancreatic sphincterotomy (of the major and minor papilla) is used much less frequently than biliary sphincterotomy, but its popularity is increasing. It is performed both with a pull-type sphincterotome and with a needle-knife over a stent [70]. Few studies have analyzed its specific complications [71–73]. The overall complication rate in 1615 pancreatic sphincterotomies at MUSC (many of whom underwent biliary sphincterotomy at the same time) was 6.9%; 80% of these were pancreatitis. There was only one recorded sphincterotomy perforation, and only three (0.2%) severe complications, with no related deaths.

Precut sphincterotomy   

The needle-knife precut technique is useful and safe in treatment of impacted stones [74], and is used by many as the primary method for performing pancreatic sphincterotomy (over a stent), and for biliary sphincterotomy after Billroth 11 gastrectomy. However, precutting used purely as a biliary access technique is contentious [75,76]. Much of the literature suggests that it is valuable and safe when used for good indications by experts [77–87], but there is ample evidence, not least from lawsuits, that it is dangerous when used by inexperienced endoscopists, especially when the indication is not strong. Several studies (including centers with considerable experience) clearly show that precutting increases substantially the risk of pancreatitis and of perforation [22,24,25,36,44]. It has been suggested that precutting has received a bad reputation only because it may used as a last resort, after much other manipulation, and that it may be safe when used early in the cannulation process. However, it seems a poor alternative to good standard methods.

Variants of the precut technique have been described [88,89], including using a standard pull-type sphincterotome in the pancreatic duct. Despite good experience with this method reported from one center [90], this seems to be courting disaster.

The data clearly indicate that precut access techniques should be avoided by inexperienced endoscopists, especially when there is little or no evidence for biliary pathology requiring treatment.

Repeat sphincterotomy   

Biliary and pancreatic sphincterotomy sometimes need to be repeated, for recurrent stones or stenosis. Whether the second procedure carries increased risk clearly depends on the indication, and on the size of the prior procedure [91,92]. One study showed a significant increase in the risk of both bleeding (from 1.7 to 5%) and of perforation (from 1 to 8%), but a reduction in pancreatitis (from 5.5 to 1%) when comparing repeat biliary sphincterotomies with index cases [93]. These factors are discussed further in the specific risk sections.

Balloon sphincter dilation   

As endoscopic stone extraction became more frequently used in relatively fit and young patients (after the advent of laparoscopic cholecystectomy), there was increasing interest in trying to further reduce the (albeit small) risks of sphincterotomy, by using balloon dilation of the papilla instead [94,95]. Since the main concern about this technique is the risk of provoking pancreatitis, it will be discussed further in that section.

Endoscopic papillectomy   

The increasing familiarity and confidence of endoscopists with polypectomy, mucosal resection, and complex ERCP has led many to perform excision of the major papilla for treatment of adenomas. The techniques (including temporary pancreatic stenting) are now fairly well established [96–99], but there is continuing concern about the precise indications [100] and the likelihood of recurrence. The immediate risks are bleeding, pancreatitis, and perforation. One large series of 70 cases reported 10 complications (14%), with bleeding in 4, pancreatitis in 5, and mild perforation in 1, with one-third of the adenomas recurring after a median follow-up of only 7 months [101].

Stenting   

Biliary stenting is widely used for the management of leaks and tumors. It carries the same general risks as any ERCP procedure, although the presence of a pancreatic tumor may protect somewhat against pancreatitis. A small biliary sphincterotomy is not necessary in most cases [102], but wise in hilar tumors, since it may prevent pancreatitis, and also facilitate the placement of two stents [103].

Biliary stenting is most risky when it fails, or when drainage is incomplete. This commonly leads to sepsis, and carries the risks of repeat procedures (whether ERCP or percutaneous). The chance of failure, and of complications, is considerably greater with lesions involving the liver hilum. The specific risks of stenting (and pancreatic stenting) are discussed later.

Pseudocyst drainage   

Endoscopic drainage of pseudocysts through the stomach wall or duodenum carries a significant risk of bleeding, perforation, and infection [104].

 Reducing the risks of ERCP: general issues   

Methods to minimize specific risks are detailed below for each of the main complications. However, there are several general strategies which should be understood. Clearly it is helpful to maximize the clinical and technical expertise of the endoscopist and the assisting team, and to follow accepted standards of practice. Since complications cannot be avoided completely, it is also mandatory to ensure that patients and those close to them are fully informed about the key issues.

The contract with the patient; informed consent   

It is always the responsibility of endoscopists to assure themselves that the potential benefits of the proposed procedure exceed the potential risks, and to convey that information clearly to their patients (Chapter 8) [105]. Truly informed consent means that the patient really does understand the potential risks and benefits, as well as the possible limitations and any available alternative approaches. That is our contract with the patient. Signing an 'informed consent form' is a medico-legal requirement in many institutions, but this is nothing more than confirmation of the education process. It is important to 'tell it as it is'. It is in our nature as physicians to want to reassure nervous patients that 'things will go just fine', but that is neither honest nor wise.

Educational materials   

Nothing can replace a detailed discussion between the endoscopist and the patient (and any accompanying persons), but this process can be enhanced with written, video- or web-based educational materials. Suitable brochures are available from national organizations, and on many websites, and can be adapted for local conditions. The document in routine use at MUSC is attached (Fig. 9). The process of informed consent must be clearly documented, and witnessed. For elective procedures, this process should take place preferably at least 1 day beforehand to give time for review of the materials and unhurried reflection. Whatever the details of the education process, patients must be given the opportunity to ask questions of their endoscopist (and support staff) again before the procedure.

Humanity   

It is appropriate also to emphasize the importance of simple courtesies and common humanity in dealing with our customers. What is familiar and routine to the endoscopist and staff may be viewed by patients as a major ordeal—especially by those unfortunate enough to experience a significant adverse event.

Care after ERCP   

Admission?   

Many patients are kept in hospital under observation overnight after ERCP. The advantage is that nursing staff can ensure adequate fluid intake (mainly intravenously), and can quickly detect and pay appropriate attention to symptoms which may herald important complications. However, overnight observation adds costs, and can add other burdens for patients and their families. Several studies have evaluated factors predicting the need for admission [106–110]. Admission is unnecessary in the majority of standard level 1 procedures (simple biliary stones and stents), but seems wise when the risk is predicted to be higher than average (e.g. sphincter dysfunction management), when the procedure has been difficult in some way, or when the patient is frail, or has no responsible accompanying person. Staying overnight in a local hotel is an appropriate compromise option for patients who live more than an hour or two away. Attempts have been made to use serum tests early in the recovery period to predict subsequent pancreatitis [111,112], but this has not become standard practice.

Early refeeding?   

Patients are often keen to catch up on the meals that they have missed as a result of the procedure, but it has been my practice to recommend taking fluids only until the next morning, when the main risk of pancreatitis has passed. However, a recent randomized trial suggests that early refeeding is not detrimental [113].

 Pancreatitis after ERCP   

Pancreatitis is now by far the most common complication of ERCP and sphincterotomy (Figs 4, 5). Our better understanding of the risk factors in recent years is largely attributable to the seminal studies anchored by Freeman, who has published several comprehensive reviews [114,115]. This contribution will focus on the key facts.

Definitions   

Serum amylase and lipase levels can be shown to rise in almost every patient if measured within a few hours of ERCP, even sometimes when the pancreatic duct has not been entered or opacified. While this indicates some irritation of the pancreas, it does not constitute clinically relevant pancreatitis. The incidence of pancreatitis clearly depends greatly on the criteria used for the diagnosis [116].

The consensus workshop suggested this working definition of post-ERCP pancreatitis [15]. 'Pancreatitis after ERCP is a clinical illness with typical pain, associated with at least a three-fold increase in serum amylase (or lipase) at 24 h, with symptoms impressive enough to require admission to hospital for treatment (or extension of an existing or planned admission).' Severity was graded as mild if hospitalization is needed for less than 3 nights, moderate if 4–9 nights, and severe if more than 10 nights, or if patients require intensive care or surgical treatment.

This definition has been widely used, despite some concern about the relevance of hospitalization, and the fact that the apparent incidence will vary according to the admission policy. It is also sometimes difficult to decide how to deal (statistically as well as clinically) with patients with long-standing pancreatic pain who linger in hospital after their procedures. We do not count this as a complication unless the patient is obviously worse afterwards.

Incidence of pancreatitis after ERCP   

In addition to the major reviews and studies of ERCP complications [21–44], there is extensive literature specific to the risk of pancreatitis after ERCP [114–123]. The reported incidence ranges widely, from less than 1% up to as high as 40%. Much of this huge variation can be attributed to different definitions, incomplete data collection, and differing case mixes. Ranges of 2–9% are representative of more recent prospective series, mostly using consensus definitions [114] (Figs 4, 5). An innovative population-based study of 97 810 ERCPs in Canada reported a pancreatitis rate of 2.2%, with greater risk in younger patients and in women [124].

The overall pancreatitis rate in our series at MUSC is < 3%, with a gradual reduction over the years, despite an increasing number of cases with suspected sphincter dysfunction, where the risk is known to be greater (Figs 6, 7).

More than 75% of all cases of pancreatitis after ERCP were graded as mild (Fig. 6). Mild complications are disappointing and inconvenient, but are not serious or threatening (medically or legally). Severe pancreatitis occurred in 13 cases (0.13%). They are devastating for all involved, and fatalities occur (one in our series). For these reasons, there is great interest in understanding the true risk factors, and ways to minimize them.

Risk factors for pancreatitis   

Any ERCP procedure can cause pancreatitis, but certain factors are well known to increase the risk. A listing, adapted from Freeman [114] is shown in Fig. 10. The factors are both patient- and procedure-related.

Patient factors increasing the risk [114,115,122,123]   

It has become abundantly clear that the risk of developing pancreatitis is greater in younger patients, and in women as compared to men [114,124], and particularly when ERCP is done for 'suspected sphincter dysfunction' in the absence of much objective evidence for biliary or pancreatic pathology. A prior history of recurrent pancreatitis, or of post-ERCP pancreatitis, also increases the risk in most studies. Contrary to earlier reports [62,63], a 'small' or normal-sized bile duct is not an independent risk factor [114,115].

Procedure factors increasing the risk   

Pancreatic manipulation   

Pancreatitis is more likely to occur with aggressive manipulation of the pancreatic orifice [114], and with repeated injections of contrast [117], sometimes evidenced by acinarization or the appearance of a urogram [125]. The importance of increased pressure in the duct is supported by the old observation that postprocedure pancreatitis is less likely in patients who have a patent duct of Santorini. Variations on methods for cannulation have been explored [126–130], without any new consensus.

Sphincter manometry   

For a long time, it was believed that sphincter manometry was a potent cause of pancreatitis [131]. However, it is now clear that manometry is simply a surrogate for sphincter of Oddi dysfunction (SOD), which is the real culprit [115,133]. Several series show that the apparent increase disappears if increased sphincter pressure is treated (by pancreatic sphincterotomy or temporary stenting). It has been shown also that the risk is actually higher when suspected SOD (e.g. postcholecystectomy pain) is treated empirically by sphincterotomy than when manometry is employed [114].

Sphincterotomy   

Several studies now indicate that standard biliary sphincterotomy does not markedly increase the overall risk of pancreatitis (when compared with diagnostic ERCP) (Figs 4, 5, 8). The suggestion that pure cutting current could reduce the risk of pancreatitis has not been proven in most of the published studies [134–139].

In the hands of experts (who publish), access precut sphincterotomy appears to be both useful and safe, at least when used for good (biliary) indications [75–87,140]. However, many series document a significantly increased risk of pancreatitis when precutting is performed [44,114,115,118,123]. In Freeman's large prospective multicenter analysis, the complication rate was 24.3% after precutting, with 3.6% severe pancreatitis [44].

Pancreatic sphincterotomy is being performed increasingly in referral centres for many different indications. The pancreatitis rate for 1615 pancreatic sphincterotomies in our unit (performed mainly for sphincter dysfunction, and with temporary stenting), was 5.6%.

Biliary sphincter dilation   

Balloon dilation of the biliary sphincter has been advocated as an alternative to sphincterotomy for removal of duct stones, in the hope of reducing the (small) short- and long-term risks [94,95]. Early case series gave encouraging results [94,141,142], but the technique can cause pancreatitis [143]. Many randomized studies have been performed to compare the risk with that of standard sphincterotomy [144–150]. Some involved older patients, often with dilated ducts and large stones, and showed that the short-term risks of sphincterotomy and of balloon dilation were similar [142,144]. However, the concept of sphincter preservation is most attractive in younger patients with smaller stones and relatively normal ducts. A major multicenter US study in these types of patients (in the context of laparoscopic cholecystectomy) showed a marked increase in the risk of pancreatitis, with two deaths [150]. This has led to a consensus, at least in the USA, that the balloon technique should be considered now only in special circumstances, such as coagulopathy [151], and maybe Billroth II patients [152,153]. This restrictive recommendation could change with further progress in preventing pancreatitis, e.g. by combining balloon dilation with pharmacological or stenting prophylaxis.

Biliary stenting   

Temporary stenting of the biliary sphincter has been used as a therapeutic trial in patients with suspected sphincter dysfunction. This technique is a potent cause of pancreatitis [154] and should be avoided.

Some endoscopists routinely perform sphincterotomy before placing biliary stents through strictures, to facilitate subsequent stent exchange and to reduce the risk of pancreatitis caused by irritating the pancreatic orifice. The latter hope has been documented only in a few patients with hilar tumors [103], where sphincterotomy is necessary anyway to place more than one stent. Sphincterotomy is not necessary or protective in other circumstances [102].

Pancreatic stenting   

The precise risk of causing immediate pancreatitis by placing pancreatic stents is difficult to measure, since this is done in many ways for many different indications, and often in conjunction with other manipulations such as pancreatic sphincterotomy.

Combining patient- and procedure-related factors   

Many of these risk factors are additive [114]. For instance, precutting in suspected sphincter dysfunction resulted in a complication rate of 35.3%, with no fewer than 23.5% overall graded as severe [44]. In another study from the same group concerning post-ERCP pancreatitis, a woman with a normal serum bilirubin, bile duct stone, and easy cannulation had a 5% risk of pancreatitis. This increased to 16% if cannulation proved difficult, and to 42% if no stone was found (i.e. suspected sphincter of Oddi dysfunction) [115]. These are the unfortunate patients who are still developing severe pancreatitis after ERCP, and who feature in lawsuits.

Prevention of pancreatitis after ERCP   

Avoiding ERCP, especially in high-risk patients   

Post-ERCP pancreatitis cannot (currently) be prevented completely, except by avoiding the procedure, which is a good strategy in many cases; sadly, it is not applicable in retrospect. The availability of sophisticated imaging techniques such as MRCP and EUS means that ERCP should be used nowadays almost exclusively for therapy. There is a less than 10% chance of finding objective pathology by ERCP in a patient with pain, normal or only slightly deranged chemistry, and normal CT/MRCP imaging. Where sphincter dysfunction is suspected, wisdom dictates referral to a center able to perform manometry, and experienced in methods (e.g. pancreatic stenting) known to reduce the risk of pancreatitis.

The consensus panel at the 2002 NIH State of the Science conference on ERCP advised strongly against the 'casual use' of ERCP in investigation of patients with obscure abdominal pain, stating 'Diagnostic ERCP has no role in the assessment of these patients. It is precisely the typical SOD patient profile (young, healthy female) that is at the highest risk for ERCP-induced severe pancreatitis and even death. Indeed the risk of complication exceeds potential benefit in many cases. Therefore, ERCP, if performed, must be coupled with diagnostic SOM (sphincter of Oddi manometry), possible dual sphincterotomy, and possible pancreatic stent placement. ERCP with SOM and sphincterotomy should ideally be performed at specific referral centers, and in randomized controlled trials that examine the impact and timing of therapeutic maneuvers on clinical outcome'[59].

When ERCP is indicated, there are several ways to reduce the risk of the procedure [114,118,126,155].

Mechanical factors   

Attention to the mechanical factors discussed above can reduce the risk. Gentle intelligent probing for the desired duct with minimal injections of contrast will help. The endoscopist who personally injects the contrast has better control of this important variable. Probing with a guidewire rather than contrast may be prudent, but reduced risk has not been proven. It is important to know when to stop. Failure to complete an ERCP may feel bad, but severe pancreatitis feels much worse, to both endoscopist and patient. Persisting, and using more dangerous approaches like precutting, can be justified only when there is a strong indication for the procedure, i.e. good evidence for biliary (or pancreatic) pathology, and a likelihood of needing endoscopic therapy.

When manometry is performed, it is clearly wise to use an aspirating catheter system [156]. In the future, microtransducer technology may be preferable, and safer [157].

The type of current used for sphincterotomy does not appear to be a big factor influencing the pancreatitis rate [134–139], but it is clearly wise to avoid excessive coagulation near the pancreatic orifice.

Contrast agents   

Extensive studies have not shown any consistent benefit for one or other contrast agent for ERCP [158–163].

Pharmacological prophylaxis   

The list of pharmacological agents that have been proposed and tested for prophylaxis of post-ERCP pancreatitis is long and varied [114,126,155,164,165]. It includes antibiotics [166], heparin [167], corticosteroids [168–173], nifedipine [174,175], octreotide and somatostatin derivatives [176–193], trinitrin [194–196], lidocaine spray [197], gabexate [198,199], secretin, cytokine inhibitors [200–203], and a non-steroidal (rectal diclofenac) [204]. Apart from a 12 h infusion of gabexate [198], the study using diclofenac is so far the only one to show some promise. It deserves further evaluation, not least because of its simplicity, and the fact that it can be given selectively after ERCP. Preliminary data on secretin prophylaxis is encouraging [205].

None of the agents tested so far has proven to be sufficiently effective and practicable to find a place in routine practice, at least in Western countries. Reports suggest that octreotide analogs are widely used for this purpose in Japan.

Pancreatic stenting to prevent pancreatitis   

There is now overwhelming evidence that temporary stenting of the pancreatic duct can reduce the risk of pancreatitis after ERCP in high-risk patients, e.g. those with suspected or proven sphincter dysfunction, at least in expert centers [114]. This recognition is one of the most important developments in ERCP in the last 15 years. The first assessment was not convincing [206], but a randomized trial from our group in 1998 showed a dramatic benefit [207]. Eighty patients undergoing biliary sphincterotomy after manometry for suspected SOD were randomized to placement (with extraction next day) of a short 5FG pancreatic stent, or no stent. The pancreatitis rate fell from 26% to 7%.

The need for a second procedure to remove the stent is now obviated by the currently preferred technique of placing small stents that pass spontaneously within 1–2 weeks. We use stents of 3 Fr, with no internal flaps, and 8–12 cm long (so that the internal tip is in a straight part of the duct). Unlike larger and stiffer stents, these stents do not appear to cause any duct damage. Since mid-2000 this has been routine in all of our patients being investigated and treated for suspected sphincter dysfunction. The pancreatitis rate in these patients was 5.8% in 2002–03. We also use 3 Fr stents in other contexts when there has been extensive pancreatic manipulation.

Studies from many centers have amply confirmed the value of this technique [114,208–213]. One important caveat is that additional skills are required to pass small guidewires deeply into the pancreatic duct, so that the safety and value of the method is unproven in less experienced hands.

Feeding and monitoring   

The need for postprocedure observation and possible food restriction to reduce the risks of ERCP have been discussed earlier.

Post-ERCP pancreatitis, recognition, and management   

Many patients experience some epigastric distress and bloating in the hour or two after ERCP. Often this is due to excessive air insufflation, which settles quickly. By contrast, pancreatitis usually becomes evident after a delay of 4–12 h, and is characterized by typical pancreatic-type pain, often associated with nausea and vomiting. Patients have tachycardia, epigastric tenderness, and absent or diminished bowel sounds. Serum levels of amylase and lipase are elevated, but leucocytosis is more predictive of severity than the enzyme levels.

Perforation is the most important alternative diagnosis, which should always be considered early if there is marked distress and abdominal tenderness (and especially if the serum levels of amylase/lipase are not impressive). Abdominal radiographs may be diagnostic in some cases, but CT is more sensitive.

The spectrum of severity and treatment of patients with pancreatitis after ERCP is the same as for pancreatitis occurring spontaneously [214]. Adequate analgesia and aggressive fluid replacement are key. Some experts use octreotide analogs, but proof of benefit is anecdotal. CT scanning is indicated within 24 h if there is suspicion of perforation, and after a few days if clinical progress is slow or if fever develops (Fig. 11). Antibiotics are usually not given unless pancreatic infection is proven by percutaneous aspiration. The rare patient who develops a pseudocyst or pancreatic necrosis may require percutaneous or endoscopic drainage, or surgical debridement, and may require transfer to a tertiary center.

Post-ERCP pancreatitis, conclusion   

Pancreatitis is now the commonest complication of ERCP, and it can be devastating. It cannot yet be prevented completely, even in expert hands. It is most likely to occur when inexperienced endoscopists work on patients with minimal pathology. When the indication is not strong, wise clinicians will exhaust less invasive approaches before recommending or performing ERCP, will make sure that patients fully appreciate their individualized risk/benefit balance, and will include referral to expert centers in their consent process. Skillful technique, and the use of small pancreatic stents, will keep the risk of pancreatitis below 5% in most circumstances, but cannot yet eliminate it.

 Perforation   

Four different types of perforation have been described resulting from ERCP procedures [215]. They are:

• Perforation of ducts or tumors by guidewires and other instruments—perhaps better called 'penetrations'.
  
• Retroduodenal perforation related to sphincterotomy.
  
• Endoscopic perforation of the esophagus, stomach, or duodenum (away from the papilla).
  
• Stent-related perforation.
  

These types have different causes and consequences.

Duct and tumor 'penetrations'   

Guidewires, and occasionally accessories passed over guidewires (such as sphincterotomes, catheters, and dilators) can pass through the wall of the biliary or pancreatic ductal system (or indeed the raw area of a fresh sphincterotomy) [215,216]. This occurs perhaps most often when attempting to cannulate in a patient with a tumor involving the region of the papilla. These incidents are rarely reported, so their frequency is unknown. They are more likely to occur with vigorous probing in difficult cases, especially when there is distortion by tumor, or sharp ductal deviation for other reasons. Rigid guidewires may be more dangerous. Often it is safer to proceed with a 'flipped' tip wire, which tends to find the lumen more easily.

Ducts have also been disrupted occasionally by over-aggressive balloon dilatation of biliary (and pancreatic) strictures. The radiographic appearances may appear somewhat alarming when contrast is injected.

The risk of this event can be reduced by careful insertion of instruments whilst being aware of the potential problem. Recognition is usually straightforward, and the problem is defused satisfactorily by finding the correct lumen, and by completing the procedure (e.g. by stenting). It is very unusual indeed for a patient to have any adverse consequences.

Sphincterotomy-related perforation   

Perforation occurring after sphincterotomy is always retroduodenal. It is defined by the presence of air (and/or contrast) in the retroperitoneum.

A review of more than 12 000 biliary sphincterotomies performed before 1990 showed a sphincterotomy perforation rate of 1.3%; 27% of these patients were operated, and the overall mortality was 0.2%. Since that time most publications show a sphincterotomy perforation rate of < 1% [24,25,44,49,62,215,217]. Three studies have reported higher perforation rates: 1.1% [218], 1.8% [22], and 2.2% [40].

Only 4 perforations have been recorded after 2820 biliary sphincterotomies at MUSC over the last 10 years, a rate of 0.14%: 3 were operated, and none died.

Routine CT scans in asymptomatic patients after uncomplicated sphincterotomy have shown small quantities of periduodenal or retroperitoneal air in up to 10% of patients [219,220], so it may be that there are more asymptomatic 'micro-perforations' than is commonly recognized.

Risk factors for sphincterotomy perforation   

It is assumed that perforation is more likely with larger and repeat biliary sphincterotomies, and that cutting beyond 'one to two o'clock' is more risky. It is not reported more frequently in patients with peripapillary diverticula [59,60]. As discussed above, precut sphincterotomy appears to be relatively safe and useful in expert hands, with restricted indications, but it is clearly more dangerous in routine practice, and when used, for instance, in patients with suspected sphincter dysfunction [44,76]. Perforation rates after precutting as high as 5% have been reported [22,75,77], and precut related perforations feature prominently in medico-legal cases involving ERCP.

Perforation also appears to be more likely in patients with suspected SOD [44]. This may be due simply to the smaller- (often normal-) sized ducts, or because patients with bile duct stones are somehow protected (due to the distorting/fibrotic effect of recurrent stone impaction or passage). It has been reported occasionally after forceful extraction of large stones, and at least once after balloon dilatation of the sphincter to remove stones without sphincterotomy [144].

Perforation after pancreatic sphincterotomy (at the main or minor papilla) is extremely rare [71]. One occurred during 1615 pancreatic sphincterotomies at MUSC over the last 10 years.

Recognition of sphincterotomy perforation   

Perforation may become obvious during the procedure itself, when unusual territory is encountered (Figs 12, 13), or when the radiographs show contrast in non-anatomical shapes around the duodenum. This is best recognized by inflating and then aspirating air to show that the odd radiographic shape does not change (which it does if the contrast is in the duodenum). Occasionally, if sufficient air has been insufflated after the perforation, fluoroscopy may show air around the right kidney and along the lower edge of the liver (Fig. 14)[218].

Most cases of perforation are not recognized until after the procedure, when the patient complains of epigastric pain. The differential diagnosis is pancreatitis, which is far more common. Perforation should always be considered when the pain starts soon after the procedure (pancreatitis may not develop for 4–12 h), when symptoms are more severe than anticipated, and when accompanied by guarding and tachycardia. Rarely, patients may develop subcutaneous emphysema, pneumo-medistinum or pneumo-thoraces after a few hours [221,222]. The white blood count usually rises quickly. Finding a normal or only slightly elevated serum level of amylase or lipase in patients with impressive abdominal pain should raise suspicion of perforation.

A plain abdominal X-ray may show retroduodenal air, but CT scanning is more definitive (Fig. 15)[218], and should be done within 24 h in any patient with severe abdominal symptoms after sphincterotomy.

Reducing risks of sphincterotomy perforation   

Clearly, the best way to reduce the risk of causing perforation at sphincterotomy is to minimize the use of higher risk techniques, such as cutting too far, cutting 'off-line', extending prior sphincterotomies, and precutting.

Management of sphincterotomy perforation   

Perforation is a life-threatening event; prompt recognition and efficient management are very important [215,217,223–227]. Patients should have nothing by mouth and adequate intravenous fluids (and nutrition as necessary), and are usually given antibiotics. Most experts recommend placement of a gastric or duodenal drainage tube. A few endoscopists have suggested placing a biliary stent or nasobiliary drain to reduce contamination of the retroperitoneum, but this is not proven or standard practice, and the additional manipulation may make matters worse.

Surgery?   

Most surgeons equate perforation with immediate operation. However, surgeons exploring cases often are unable to find the site of perforation, and end up simply leaving retroperitoneal drains. Reported experience, including surgical studies [217,226], show that surgery is not usually necessary, and that most (reported) retroduodenal perforations have been managed conservatively. An important caveat is that conservative therapy seems to be effective only when perforation is recognized early [217].

Despite the dominance of non-operative treatment for perforation, it is wise to obtain a surgical opinion at the earliest possible stage. Patients should be managed jointly on a daily basis. I recommend immediate/early surgery only if there is remaining biliary pathology which itself requires operation. Thus, a patient with gallbladder stones can reasonably undergo immediate cholecystectomy and placement of drains. However, most perforations occur in patients with little or no remaining pathology (e.g. cleared retained stones, or SOD), where there is no indication for surgery—other than the known perforation. Conservative management is usually effective if started early, but intervention (percutaneous or surgical) may be required in the ensuing days or weeks if fluid collections/abscesses develop in the right renal or pericolic areas. Operating at a later stage is often difficult because of infection; it may be necessary to perform diversionary procedures as well as multiple drains [217]. A few patients have had a very bad experience, with months in hospital and multiple operations.

One successful case of endoscopic treatment with multiple clips has been reported [228].

Perforation remote from the papilla   

Endoscopic perforation can occur anywhere that endoscopes travel.

The lateral viewing nature of the duodenoscope may perhaps increase the risk of pharyngeal perforation in elderly patients with diverticula. In the absence of pathology, it is difficult to conceive how endoscopic perforation could occur in the esophagus or stomach, but such events have been reported [215,229,230]. It has also happened rarely in the duodenum, when attempting to negotiate a stricture or marked distortion by tumor. The first therapeutic video-duodenoscopes had a long distal tip, which caused perforations during forceful stone extraction maneuvers [231,232].

Perforation of the afferent loop is a definite risk during endoscopy of patients after Billroth II gastrectomy (and more complex bypass procedures) [232–234]. Rates as high as 6% [233] and even 20% [234] have been reported in this context.

Perforation usually occurs as a result of stretching of loops rather than penetration of the endoscopic tip. It can be avoided largely by careful endoscopic technique, especially in patients who have undergone diversionary procedures or who have known stenosing pathology.

One expert suggests that all Billroth II patients (and more complex diversions) should be referred to tertiary centres, because of the added difficulty and significant risk [232].

The incidence of endoscopic (as opposed to sphincterotomy) perforations is really unknown, but should be extremely low. It is less than 1:1000 in our series, all of them in the afferent loop context (Fig. 6).

Recognition and management of endoscopic perforation   

Diagnosis of endoscopic perforation is usually obvious, either during the procedure, or because of obvious patient distress and clinical signs in the chest or abdomen. Radiographs show intraperitoneal (or mediastinal) air.

Endoscopic perforation usually requires surgical intervention, and prompt surgical consultation is mandatory. Rare episodes have been treated conservatively.

Stent migration perforation   

There have been rare reports of penetration and even perforation of the duodenum, small bowel, and colon by stents which have migrated from the bile duct [235,236]. Almost all of these have been 'straight' 10 Fr gauge stents. Those that have migrated down from the bile duct and penetrated the opposite duodenal wall can sometimes be managed simply by endoscopic extraction. Others have required surgical intervention.

 Infection after ERCP   

Generic infection risks of all endoscopic procedures (e.g. endocarditis, viral transmission), are discussed elsewhere (Endoscopy Practice and Safety Chapter 7)[237]. Preprocedure antibiotics are recommended for prophylaxis against endocarditis, by standard guidelines.

ERCP differs from most other endoscopies in that it risks contaminating territory that is usually sterile. Also, when bile is infected (e.g. in patients with stones, or blocked stents), biliary manipulation may disseminate the infection locally or systemically.

By consensus, infection is defined as 'an otherwise unexplained fever of greater than 38°C lasting 24–48 h after ERCP'. It is described as moderately severe if it requires more than 3 days treatment in hospital, or further endoscopic or percutaneous intervention, and as severe if the patient develops septic shock or requires surgery [15].

The reported incidence of clinical infections after ERCP is low, ranging from 0.7 to 1.6% in various modern series [22,27,29,40,44,238–243]. However, bacteremia rates of up to 27% have been reported [237,244].

Nosocomial infection   

In the early days of ERCP, before the importance of disinfection was recognized, there were several unfortunate outbreaks of nosocomial infections (usually due to Pseudomonas) [16,18,237,245]. Sadly, Pseudomonas infections are still being described after ERCP [246–249]. Almost all are due to faulty cleaning and disinfection, and should be preventable.

Cholangitis   

Bacteremia and septicemia occur when the bile is infected and drainage is compromised. This can occur after ERCP in patients with stones or strictures if adequate drainage is not achieved, and when bilary stents become occluded [238–244,250]. The risk of introducing or stirring up infection when the bile is infected can be minimized by adhering to disinfection protocols, by reducing the biliary pressure (by aspirating bile before injecting much contrast), and by ensuring adequate drainage by removing all obstructing stones or placing appropriate stents. Sepsis is a particular risk after ERCP management of hilar tumors, and sclerosing cholangitis, where it may prove impossible to provide complete drainage of all obstructed segments. This is a good reason for obtaining detailed anatomical imaging (by CT and/or MRCP) beforehand to assist therapeutic planning.

Cholecystitis   

Cholecystitis, sometimes with odd characteristics [251,252], has occurred soon after ERCP; presumably this is more likely when there is cystic duct compromise by stone or tumor (or occasionally after stenting). It is managed by standard percutaneous or surgical techniques.

Pancreatic sepsis   

This has occurred as part of severe pancreatitis after ERCP [253], and in patients with pseudocysts [104], due to inadequate disinfection or incomplete drainage.

Prophylactic antibiotics   

The role of prophylactic antibiotics in the prevention of infection after ERCP is still unclear, despite a very substantial literature and much opinion [254–263]. Although one randomized study did appear to show benefit for antibiotic prophylaxis [260], this and others [258] demonstrate most clearly that biliary obstruction is the main hazard, and that effective drainage is the best treatment. The risk of serious infection is now so low that further randomized trials are unlikely to be helpful.

In earlier years, we gave intravenous antibiotics (usually ampicillin and gentamycin) to all patients with clinical or radiological evidence of duct obstruction, and whenever therapy seemed probable (which meant about 90% of all cases). The infection rate was < 1%. We gradually reduced the indications for prophylaxis, and changed to oral ciprofloxacin, without any increase in infection [264]. Our current practice is to give oral ciprofloxacin (two doses) before ERCP when failure of complete drainage is predictable (e.g. complex hilar tumors, sclerosing cholangitis, and pseudocysts), and to give it intravenously immediately after any procedure in which we fail to provide drainage. With this policy, the incidence of clinical infection in our unit remains well below 1% (Fig. 7). Some have advocated mixing antibiotics with the contrast media, but this practice has never been validated.

Delayed infection   

The commonest cause of delayed biliary sepsis is a blocked stent. Patients can become seriously ill quickly with septic cholangitis. For this reason, patients and their caregivers must be fully informed about this risk and instructed to make contact as soon as symptoms develop. For the same reason, it is common practice to change plastic stents routinely (at 3–4 months), especially in patients with benign biliary strictures. The need to do so in patients with malignant disease (as opposed to waiting for obstructive symptoms) has not been validated in controlled studies, but is still common practice.

 Bleeding after ERCP   

Clinically significant bleeding has occurred rarely after diagnostic ERCP (due to retching, or after biopsy in patients with tumors or coagulopathy, or after cannulation in patients with biliary varices). However, the main cause of bleeding is sphincterotomy (or other cutting procedure such as papillectomy, and pseudocyst drainage). It is common to see a small amount of 'endoscopic' bleeding (Fig. 16) (i.e. oozing immediately after sphincterotomy), but clinically relevant bleeding is much rarer.

Bleeding can occur immediately, but is often delayed for up to 2 weeks.

Whilst significant bleeding is usually manifest by hematemesis and/or melena, occasional patients can present with biliary pain and cholangitis if bleeding fills the bile duct.

Definition of bleeding, and incidence   

The consensus conference on complications defined bleeding in clinical terms. Even impressive immediate bleeding is not counted as a complication if it can be stopped by endoscopic manipulation during the procedure. The severity of the bleeding is stratified as follows:

• mild: clinical (not just endoscopic) evidence of bleeding, with a hemoglobin drop less than 3 g and no transfusion;
  
• moderate: transfusion (4 units or less), but no angiographic interventional surgery;
  
• severe: transfusion of 5 units of more, or intervention (angiographic or surgical);
  
• fatal: death attributable to bleeding.
  

Memory and some early publications [15,17,19,20,265] indicate that bleeding was the commonest complication of sphincterotomy in the first 10–15 years after its introduction, with an average rate of 2.5% in over 20 000 reported sphincterotomies [15], with a high of 11% [265]. More recent series, using the consensus definitions, report a lower incidence of 0.8–2% [24,25,44]. One series suggested that the true incidence is higher if hematological parameters are followed routinely [266]. More than half the reported episodes of bleeding are delayed for up to 2 weeks.

At MUSC, the rate of bleeding after biliary sphincterotomy was 0.7%; all were delayed.

Risk factors for bleeding, and avoidance   

Bleeding is certainly more likely to occur in patients with coagulopathy and/or portal hypertension [267,268], renal failure [268], and apparently also when a sphincterotomy is repeated [93]. There is no evidence that the risk is greater in patients taking aspirin and other agents affecting platelet function [268–270], although it is still common practice to ask patients to discontinue their use. Delayed clinical bleeding may [268] or may not [271] be more common when there has been some immediate oozing.

Prevention   

Sphincterotomy should always be performed in a controlled manner, with blended current, avoiding the 'zipper' cut. Coagulopathies should be corrected wherever possible. Anticoagulants should be discontinued, but the need for temporary heparin coverage, and the duration, are controversial. The effect of newer antiplatelet agents has yet to be clearly established, but most endoscopists prefer these to be stopped for 10 days if possible. The type of current used may be relevant. One study showed that the ERBE generator reduced bleeding visible at the time of endoscopy, but not the risk of clinically defined bleeding [272]. Another study utilizing initial cutting current (to reduce the risk of pancreatitis) did show a slightly increased risk of bleeding [273].

Balloon dilatation of the sphincter can be used instead of sphincterotomy for extracting some stones in patients with irreversible coagulopathy, or severe portal hypertension [151,274,275].

Management of sphincterotomy bleeding   

Bleeding immediately after sphincterotomy usually stops spontaneously, and (unless there is a pumping vessel), it is usually not necessary to take any dramatic action. There are varying opinions about management when treatment is needed [271,276–286]. Some have advocated monopolar cautery [283], local injection of contrast agent [284], even hemoclips [285]. However, it appears that epinephrine injection is the most popular and effective technique [271,286].

My practice (with unimpressive bleeding) is first to spray the site with about 10 ml of a dilute (1: 100 000) solution of epinephrine. This often stops oozing temporarily, at least enough to see exactly where the bleeding is coming from. If bleeding is impressive, or if oozing persists, balloon tamponade is the next step. A retrieval balloon is overinflated in the bile duct, and then pulled down forcefully to compress the bleeding site between the balloon and the endoscope tip, for 5 min. If that fails, we inject epinephrine (diluted 1:10 000) using a standard sclerotherapy needle. Up to 5 ml can be injected in aliquots of 1 ml, taking care not to compromise the pancreatic orifice. For this reason it is my practice to inject just outside the top edges of the sphincterotomy, rather than within it. If there has been much manipulation, it may be wise to place a small protective pancreatic stent (if possible).

Very rarely, bleeding is profuse, and endoscopic vision is quickly lost. Expert angiographic management can be effective [287]. Surgical oversewing would seem logical when all else fails, but re-bleeding may occur [265].

Delayed bleeding   

This can occur up to 2 weeks after sphincterotomy, and should be treated like any other episode of bleeding. It is important to confirm the source of bleeding, since patients occasionally bleed from other lesions.

 Complications of stents   

Biliary and pancreatic stents can cause problems through local trauma, blockage, and migration. Much depends on their size, nature, and position. Sphincterotomy (when placing biliary stents) appears not to affect the rate of blockage [102,288], but may reduce the risk of pancreatitis, at least with hilar tumors [103].

Blockage of (plastic) biliary stents   

This is inevitable after a few months, and can cause serious cholangitis. A host of ingenious attempts to prevent this phenomenon over two decades has so far been unavailing (Chapter 4) [289]. It is common practice to reduce the risk by recommending the routine exchange of biliary stents at about 3 months. This is mandatory in patients with benign strictures. It is perhaps legitimate to await events in patients with malignant disease if they (and their caregivers) are well informed about the first symptoms (usually shaking chills), and the need for urgent action. Expandable metal stents usually last much longer, but the consequences of blockage are equally serious.

Stent migration   

Stents which migrate outwards may cause damage to the duodenum [235] or distal intestine [236]. Stents which migrate inwards can be difficult to retrieve, especially in the pancreatic duct [290]. Most migrated stents can be teased out of the papilla with a retrieval balloon, or grasped with foreign body forceps, snare, or basket. Rarely, surgery is needed to rectify these situations.

Duct damage due to stents   

The presence of a stent in the bile duct for many months may cause some wall irregularity and thickening. This can be seen radiologically (and can cause diagnostic difficulty at EUS), but has no clinical relevance. However, stent-induced duct damage is a serious problem in the pancreas [291–295], especially when the duct initially is normal. Irritation by the tip of the stent (especially at a duct bend), or by internal flaps, often causes wall irregularity, and clinically significant narrowing. Some early descriptions suggested that most of these lesions resolved after stent removal, but we have seen many tight fibrotic strictures, which are very difficult to manage. Relatively stiff pancreatic stents of 7 and even 10 Fr can be used legitimately in some patients with established chronic pancreatitis, for the management of stones or strictures. However, when stenting seems indicated in relatively normal ducts, it seems wise to use smaller (3 or 5 Fr) and softer stents, and for only a few weeks [295]. The length of a pancreatic stent should be chosen so that the inner tip is in a straight part of the duct.

Cholecystitis   

This has been reported after biliary stenting for malignancy [296–298].

 Basket impaction   

Baskets may become impacted during attempts to remove large stones from the bile duct [299]. Usually this situation can be rectified quickly by disengaging the stone, or by crushing it with a 'rescue' lithotripsy sleeve (Chapter 2). To prevent this problem, it is wise to use a mechanical lithotripsy system initially when approaching stones > 1 cm in diameter. Baskets should be used sparingly and with great caution in the pancreatic duct. They are effective for removal of soft stones (protein plugs), and mucus, but calcified pancreatic stones are very resistant to mechanical lithotripsy. There is a risk that the basket will break inside the duct and remain impacted.

 Cardiopulmonary complications and sedation issues   

Adverse cardiopulmonary events can occur during any endoscopic procedure [300,301], and myocardial ischemia has been studied specifically during ERCP [302,303].

Transient hypoxia and cardiac dysrhythmias occur occasionally during ERCP procedures, but are usually recognized and managed appropriately without clinical consequence. Very rarely they may result in severe decompensation during or after procedures, and are a significant cause of the rare fatalities attributable to ERCP.

Risk factors for cardiopulmonary complications include known or unsuspected premorbid conditions, and problems related to sedation and analgesia. Oversedation can be a serious problem, especially in the elderly and frail, and particularly if monitoring is inadequate (in a darkened room).

Cardiopulmonary complications can be largely avoided by careful preprocedure evaluation, appropriate collaboration with anesthesiologists (and cardiologists) when dealing with high-risk patients, formal training of endoscopists and nurses in sedation and resuscitation, and careful monitoring (Chapter 3) [304].

Aspiration pneumonia has been described after all types of endoscopic procedures; the incidence is unknown, but it is probably more common than recognized, since the onset may be delayed.

 Rare complications   

Many other untoward events have followed ERCP. These include:

• Gallstone ileus after removing large stones [305,306].
  
• Musculoskeletal injuries (e.g. dislocation of the temporomandibular joint [307] or shoulder, dental trauma).
  
• Opacification of blood vessels. The portal venous system and lymphatics have been seen [308,309] whilst injecting contrast through tapered tip catheters. The contrast moves rapidly on fluoroscopy. If air is injected as well the appearances on CT scan are alarming [310], but no sequelae have been reported.
  
• Antral sinus infection after prolonged nasobiliary drainage.
  
• Renal dysfunction[311] with the use of nephrotoxic medications (such as gentamycin).
  
• Impaction or fracturing of nasobiliary and nasopancreatic drains.
  
• Allergic reactions to iodine-containing contrast agents. Allergic reactions have happened, even with the very small doses which enter the blood stream during ERCP. Endoscopy units should have policies in place to deal with patients who claim to be allergic [312].
  
• Increased cholestasis in patients with sclerosing cholangitis [313].
  
• Splenic injury has been reported several times during ERCP [314–316].
  
• Distant abscesses have occurred in the spleen and kidney [314,317], and no doubt elsewhere.
  
• Hemolysis due to G 6 PD deficiency and hemolytic–uremic syndrome has been reported [318,319].
  
• Dissemination of pancreatic cancer was reported after sphincterotomy [320].
  
• A false aneurysm of a branch of the pancreatico-duodenal artery developed after needle-knife sphincterotomy [321].
  

 Deaths after ERCP   

The literature reporting deaths after ERCP is difficult to analyze as the series contain different spectra of patients and procedures, and some do not distinguish between 30 day mortality and events attributable to the procedure itself. One paper illustrates the difficulty in attributing mortality between concurrent illness, active complications, and complications due to other procedures required after ERCP failure [26]. Data collected for the consensus conference in 1991 reported 103 deaths after 7729 sphincterotomies (1.3%). Most subsequent series report mortality figures of less than 0.5% [24,27,37,44,65,322], with two higher figures of 0.8% [29] and 1% [323].

The causes of death in all of the reported series cover the spectrum of commonest complications, with approximately equal numbers resulting from pancreatitis, bleeding, perforation, infection, and cardiopulmonary events. Delay in diagnosis of perforation is mentioned as a contributing cause in several publications [217,224,324]. Of nine fatalities resulting in claims to insurance in Denmark, seven were attributable to pancreatitis (two of which had undergone precutting) [325].

 Late complications   

There are a number of adverse events attributable to ERCP that may not be apparent for months or even years afterwards.

Diagnostic error   

Failure to make the correct diagnosis is an under-reported and greatly under-appreciated complication of ERCP. It can be due to poor technique (both endoscopic and radiological), as well as incorrect interpretation of adequate images, or both. Bile duct stones are missed with inadequate duct filling, especially in less obvious sites such as the cystic duct stump, and the dependent right intrahepatic duct, or when over-dense contrast is used in a dilated system. Conversely, air bubbles introduced into the system may be misinterpreted as stones (with the potential serious consequences of an unnecessary sphincterotomy). Poor opacification and ignorance of anatomy may lead to missed or erroneous diagnoses in patients with bile duct injuries. Congenital variations of biliopancreatic drainage are under-recognized. Early stages of chronic pancreatitis, and intraductal mucinous tumors are easily missed with inadequate filling. Pancreas divisum may be missed when the ventral duct is rudimentary, and the pancreatic pathology unassessed if dorsal cannulation is not achieved.

Few endoscopists have a radiologist on hand to help with fluoroscopy, film recording, or the immediate interpretation which is needed to formulate therapeutic tactics. It is common practice for radiologists to report the available films after the event, and major discrepancies have been noted [326], a fact which raises complex issues. Providing the reporting radiologist with a detailed copy of the endoscopic report is helpful, and allows radiologists to communicate any differences of opinion.

Late infection   

There is a possibility of transmitting non-bacterial infections at ERCP, with an incubation period long enough to hide the relationship, but there are no proven and reported cases. There is a definite risk of sepsis developing when biliary stents become occluded. Patients present with fevers and shaking chills, and can deteriorate rapidly. Any stented patient (and caregivers) must be warned about the possibility, and the need for speedy medical contact and resolution. Patients receiving plastic stents for benign biliary strictures should be advised to undergo a routine stent service at 3–4 months; practice varies with malignant strictures Chapter 4. Endoscopists placing stents have a continuing responsibility to contact patients with reminders. Occasionally, patients may willfully or accidentally avoid the repeat procedure, with considerable potential for serious complications. The concept of long-term stenting for 'difficult' stones has been discredited because of the risk of delayed cholangitis [327].

Late effects of sphincterotomy   

There has been much interest in the possible long-term adverse consequences of biliary sphincterotomy [328–339]. When performed for 'papillary stenosis', there is a significant risk of further biliary-type symptoms, whether due to restenosis or an incorrect diagnosis (Chapter 6).

Sphincterotomy leads almost inevitably to bacterial contamination of the bile [340–344], which may be a potent promoter of pigment stone formation. One study showed a significant increase in the incidence of cholangiocarcinoma after surgical sphincteroplasty [345], but a cohort study in Scandinavia found no such association after endoscopic sphincterotomy [346]. Many patients have been followed for periods of 10 years or more after sphincterotomy for stone [332,334–336,338–340]. The chance of further biliary problems in these studies ranges from 5 to 24%, with an average of about 10% [347]. The Amsterdam study had the highest figure (24%) and all but one of the patients had recurrent stones [330]. In other series, some patients had episodes of cholangitis without stones, even cholangitis without stenosis of the sphincterotomy [332].

Most of these long-term complications of sphincterotomy are easily managed endoscopically, remembering that repeat incisions do carry a slightly greater risk. A few patients continue to reform stones every 6–12 months despite apparently adequate drainage, and may need to be scheduled for repeated endoscopic 'biliary laundry'[348].

Sphincterotomy with the gallbladder in place   

Most patients having their ducts cleared of stones endoscopically have undergone cholecystectomy soon afterwards. However, some have not, usually because the risk has been judged too great (and especially before the days of laparoscopic cholecystectomy). Several series have examined the long-term risks of leaving the gallbladder in place [349–354]. The reported need for cholecystectomy has ranged from 5 to 33% [337], but most of the follow-up periods are short. Two trials have addressed this issue recently. Thirty-four patients treated endoscopically for acute biliary pancreatitis (and without cholecystectomy) were followed for a mean of 34 months; only 11.6% developed further biliary complications [354]. However, the Amsterdam group performed a randomized trial of 120 patients with the gallbladder in place, after biliary sphincterotomy. No fewer than 47% of those treated expectantly developed further biliary symptoms, compared with 2% of those who underwent early cholecystectomy [353]. The suggestion that non-filling of the gallbladder at the index ERCP (indicating cystic duct obstruction) was a predictor of future trouble has not been substantiated [352]. However it seems clear that the risk is negligible in patients who have no stones remaining in the gallbladder, which is sometimes the case in the context of gallstone pancreatitis [350].

Pancreatic sphincterotomy   

The main risk of pancreatic sphincterotomy appears to be restenosis, which occurs in at least 20% of reported cases Chapter 6 and (Chapter 4).

It is usually treated endoscopically, but strictures that occur beneath the papilla can be challenging even for surgical repair. Hopefully, better techniques (and new stents) may reduce this risk in the future.

Stenosis of the pancreatic orifice causing recurrent pancreatitis has been reported as a late complication of biliary sphincterotomy [355].

 Managing adverse events   

All ERCP endoscopists experience complications. Each event requires specific skillful recognition and management (as detailed above), but there are several very important general guidelines.

Prompt recognition and action   

The keys to effective management of all complications are early recognition and prompt focused action. Delay is dangerous both medically and legally. Patients in pain and distress after procedures should always be examined carefully, and never simply 'reassured' without careful evaluation. If you are not personally on call on the night after your ERCP procedures, it is helpful to make sure that the person covering is aware of what you have done. Get appropriate laboratory studies and radiographs, consult the extensive literature, and do not hesitate to seek advice from other experts in the relevant fields. It is wise to consult an (informed) surgeon early on for anything that might remotely require surgical intervention. Sometimes it may be appropriate to offer transfer care of the patient to a specialty colleague, or to a larger medical center, but, if this happens, try to keep in touch, and to show continuing interest and concern. Apparent abandonment alienates patients and their relatives, and may lead to initiation of legal action.

Professionalism and communication   

Endoscopists often feel devastated when serious complications occur. Your distress is understandable and worthy, and it is important to be sympathetic, but it is equally important to be composed and matter of fact. Excessive apologies may give an unfortunate impression. Never, never, attempt to cover up the facts. Poor communication is the basis for much unhappiness, and many lawsuits. Remember that the truly informed patient, and any accompanying persons, have been told already that complications can happen. This is an integral important part of the consent process. So it is appropriate and correct to address suspected complications in that spirit. 'It looks as if we have a perforation here. We discussed that as a remote possibility beforehand, and I am sorry that it has occurred. Here is what I think we should do.' It is also wise to contact and inform other interested relatives, referring physicians, supervisors and your Risk Management advisors.

Documentation   

Document what has happened, carefully and honestly in real time. Don't even think of adding notes retrospectively. The results of many lawsuits hang on the quality of the documentation, or lack of it.

 Learning from lawsuits   

Fortunately, most complications do not result in legal action. Despite the fact that ERCP is the most dangerous of the routine endoscopic procedures, there are far more claims after colonoscopy and upper endoscopy [356]. There are several reasons why patients (or their survivors) may initiate a claim.

Communication   

Communication, or lack of it, is often a major complaint. Too often we hear that 'we would never have consented to the procedure if we had known that this might happen'. Sometimes this is simply because patients don't want to hear, but often the consent process is quite inadequate. A hurried conversation immediately before the procedure is not sufficient. Taking time to provide the information (face to face, and in writing), making sure that it has been understood, and writing down that you have done it, is simply good medical practice [105].

Good communication after an adverse event is equally important. Show that you care. Litigants are sometimes simply (and justifiably) angry if they get the impression that you do not.

Financial concerns   

These are also often prominent, even if not stated. Hospital bills and loss of earnings can be crippling.

Standard of care practice   

Once a lawsuit has been filed, the key issue is whether the endoscopist (and others involved) practiced within the 'standard of care'. This is defined in various ways, but comes down to what reasonable colleagues would do (and is expressed in court by what expert witnesses opine). The report from the NIH consensus conference is a crucial resource [57], and is particularly forceful in recommending caution when considering ERCP in patients with little or no objective evidence for pathology (i.e. 'suspected sphincter dysfunction').

The key standard of care issues are:

Indications   

Was the ERCP procedure really indicated in the first place? The task clearly is to balance the possible benefits against the potential risks [357]. Although professional societies publish guidelines for the use of ERCP [358], the devil is in the details, e.g. how much elevation of liver tests or increased duct size constitutes 'objective evidence of pathology'. In practice, the validity of the decision to proceed will be judged by the severity of the symptoms, by the thoroughness of prior treatment and investigations, and the process of communication. Were the symptoms (or other signs of pathology) really that pressing? Had less invasive approaches (nowadays including MRCP) been exhausted, or at least considered and discussed [359]? There are some circumstances (such as postcholecystectomy pain with some abnormality of liver tests) which may justify ERCP even if imaging is negative, but where it may be unwise to strive too hard (e.g. by prolonged attempts, or precutting) when cannulation proves difficult.

For less experienced endoscopists, consideration of alternatives (especially for higher risk procedures) should include possible referral to an expert center.

The procedure   

Was there an obvious deviation from customary practice, like placing a 10 Fr stent in a normal pancreatic duct, or trying to extract a stone from the bile duct without sphincterotomy (or papillary balloon dilatation)? Did the level of suspicion of pathology really justify a precut? Was there radiological evidence for over-manipulation of the pancreas, or of over-injection (e.g. acinarization), or injection into a branch duct? The notes of the procedure nurse may contain important evidence, like excessive sedation or contrast, or documentation of patient distress. Pretty endoscopic photographs may also be incriminating, e.g. if they show sphincterotomy in an unusual direction.

Postprocedure care   

Was the patient appropriately monitored, discharged in good condition, and properly advised? Was action taken promptly when unexpected symptoms developed? Was the endoscopist available to advise? Among the most common errors are delay in action (particularly in considering and managing perforation), and inadequate fluid resuscitation in patients with pancreatitis.

 Conclusion   

After more than 30 years, the risks of ERCP and its therapeutic procedures are now well documented. Pancreatitis and sedation-related events are the commonest, but bleeding and perforation still occur. There are a host of rare complications. Understanding and managing the main risk factors can keep these events to a minimum, but cannot eliminate them. For this reason, making sure that patients understand what they are accepting is of crucial importance. Inexperience and over-confidence are dangerous partners.

 Outstanding issues and future trends   

The two biggest issues for ERCP at the present time are the quality of practice, and how to minimize or eliminate postprocedure pancreatitis. These are not unrelated, for we know that experts have lower complication rates, even while dealing with higher risk clientele. Thus, we are forced to focus on how to maximize expertise.

Many experts for a long time have been advocating that fewer endoscopists should be trained in ERCP, so that their skills can be maximized before and after entering practice. This trend is perhaps evident at long last, driven by several forces. Firstly, diagnostic ERCP is obsolescent as non-invasive methods (especially MRCP) improve. This means that would-be ERCP practitioners can often now see the suspected therapeutic issue beforehand. They must be prepared for the challenge, but have also the option of referring problematic cases (e.g. hilar tumors and 'suspected sphincter dysfunction'). Secondly, the seminal studies of Freeman and colleagues, and a few others, have made endoscopists (and lawyers) much more aware of certain high-risk behaviors, such as casual precutting. Thirdly, most gastroenterologists have no shortage of other activities (not least screening colonoscopy) to keep them interested and busy. The final driver is the increasing sophistication of our patients, who are learning that not all interventionists are equal—as is well documented in surgery [8]—and are demanding the data with which to make informed choices [360].

All interventions carry some risks, which are acceptable if the indications are appropriate, i.e. when there are substantial potential benefits. To do a better job of predicting benefit will require many more major prospective outcome studies. We need careful objective and structured cohort studies of ERCP in various clinical contexts, and some randomized studies in comparison with other approaches, such as surgery.

Thus, in the future, we hope that there will be fewer but very well trained and experienced ERCP practitioners, and that both they and their patients will have a better understanding of the risk/benefit ratio in each case.

 References  

  1 Cotton, PB. Outcomes of endoscopic procedures: struggling towards definitions. Gastrointest Endosc 1994; 40: 514–18. PubMed

  2 Fleischer, DE. Better definition of endoscopic complications and other negative outcomes. Gastrointest Endosc 1994; 40 (4): 511–13. PubMed

  3 Fleischer, DE, Van de Mierop, F, Eisen, GM, Al-Kawas, FH, Benjamin, SB & Lewis, JH et al. A new system for defining endoscopic complications emphasizing the measure of importance. Gastrointest Endosc 1997; 45 (2): 128–33. PubMed

  4 Cotton, PB. Income and outcome metrics for the objective evaluation of ERCP and alternative methods. Gastrointest Endosc 2002; 56 (6): S283–90. PubMed

  5 Campbell, N, Sparrow, K, Fortier, M & Ponich, T. Practical radiation safety and protection for the endoscopist during ERCP. Gastrointest Endosc 2002; 55 (4): 552–7. PubMed

  6 O'Sullivan, S, Bridge, G & Ponich, T. Musculoskeletal injuries among ERCP endoscopists in Canada. Can J Gastroenterol 2002; 16 (6): 369–74. PubMed

  7 Petersen, BT. ERCP outcomes: defining the operators, experience, and environments. Gastrointest Endosc 2002; 55 (7): 953–8. PubMed

  8 Birkmeyer, JD, Stukel, TA, Siewers, AE, Goodney, PP, Wennberg, DE & Lucas, FL. Surgeon volume and operative mortality in the United States. N Engl J Med 2003; 349: 2117–27. PubMed

  9 Schutz, SM & Abbott, RM. Grading ERCPs by degree of difficulty: a new concept to produce more meaningful outcome data. Gastrointest Endosc 2000; 51 (5): 535–9. PubMed

 10 Lambert, ME, Betts, CD, Hill, J, Faragher, EB, Martin, DF & Tweedle, DE. Endoscopic sphincterotomy: the whole truth. Br J Surg 1991; 78 (4): 473–6. PubMed

 11 Cotton, PB. Endoscopic management of bile duct stones (apples and oranges). Gut 1984; 25: 587–97. PubMed

 12 Perdue, DG & Freeman, ML, ERCOST Study Group. Failed biliary ERCP: a prospective multicenter study of risk factors, complications and resource utilization. Gastrointest Endosc 2004; 59 (5): AB192.

 13 Cotton, PB. Randomization is not the (only) answer: a plea for structured objective evaluation of endoscopic therapy. Endoscopy 2000; 32 (5): 402–5. PubMed

 14 Hebert, RL, Palesch, YY, Tarnasky, PR, Aabakken, I, Mauldin, PD & Cotton, PB. DDQ-15 health-related quality of life instrument for patients with digestive disorders. Health Services Outcomes Res Methodology 2001; 2: 137–56. CrossRef

 15 Cotton, PB, Lehman, G, Vennes, J, Geenen, JE, Russell, RC & Meyers, WC et al. Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 1991; 37: 383–93. PubMed

 16 Bilbao, MK, Dotter, CT, Lee, TG & Katon, RM. Complications of endoscopic retrograde cholangiography (ERCP). a study of 10,000 cases. Gastroenterology 1976; 70: 314–20. PubMed

 17 Geenen, JE, Vennes, JA & Silvis, SE. Resume of a seminar on endoscopic retrograde sphincterotomy (ERS). Gastrointest Endosc 1981; 27: 31–8. PubMed

 18 Cotton, PB, Progress Report ERCP. Gut 1977; 18: 316–41. PubMed

 19 Neuhaus, B & Safrany, L. Complications of endoscopic sphincterotomy and their treatment. Endoscopy 1981; 13: 197–9. PubMed

 20 Vaira, D, D'Anna, L, Ainley, C, Dowsett, J, Williams, S & Baillie, J et al. Endoscopic sphincterotomy in 1000 consecutive patients. Lancet 1989; 2: 431–4. PubMed CrossRef

 21 American Society for Gastrointestinal Endoscopy. Standards of Practice Committee. Complications of ERCP. Gastrointest Endosc 2003; 57 (6): 633–8. PubMed

 22 Barthet, M, Lesavre, N, Desjeux, A, Gasml, M, Berthezene, P & Berdah, S et al. Complications of endoscopic sphincterotomy: results from a single tertiary referral center. Endoscopy 2002; 34 (12): 991–7. PubMed

 23 Freeman, ML. Adverse outcomes of endoscopic retrograde cholangiopancreatography. Rev Gastroenterol Disord 2002; 2 (4): 147–67. PubMed

 24 Loperfido, S, Angelini, G, Benedetti, G, Chilovi, F, Costan, F & De Berardinis, F et al. Major early complications from diagnostic and therapeutic ERCP: a prospective multicenter study. Gastrointest Endosc 1998; 48: 1–10. PubMed

 25 Masci, E, Toti, G, Mariani, A, Curioni, S, Lomazzi, A & Dinelli, M et al. Complications of diagnostic and therapeutic ERCP: a prospective multicenter study. Am J Gastroenterol 2001; 96: 417–23. PubMed CrossRef

 26 Tzovaras, G, Shukla, P, Kow, L, Mounkley, D, Wilson, T & Toouli, J. What are the risks of diagnostic and therapeutic endoscopic retrograde cholangiopancreatography? Aust N Z J Surg 2000; 70: 778–82. PubMed CrossRef

 27 Vandervoort, J, Soetikno, RM, Tham, TC, Wong, RC, Ferrari, AP Jr & Montes, H et al. Risk factors for complications after performance of ERCP. Gastrointest Endosc 2002; 56: 652–6. PubMed CrossRef

 28 Freeman, ML. Adverse outcomes of endoscopic retrograde cholangiopancreatography: avoidance and management. Gastrointest Endosc Clin N Am 2003; 13 (4): 775–98. PubMed

 29 Garcia-Cano Lizcano, J, Conzalez Martin, JA, Morillas Arino, J & Perez Sola, A. Complications of endoscopic retrograde cholangiopancreatography: a study in a small ERCP unit. Rev Esp Enferm Dig 2004; 96 (3): 155–62. PubMed

 30 Freeman, ML. Understanding risk factors and avoiding complications with endoscopic retrograde cholangiopancreatography. Curr Gastroenterol Rep 2003; 5 (2): 145–53. PubMed

 31 Landoni, N, Chopita, N & Jmelnitzky, A. Endoscopic sphincterotomy: it's complications and their followup. Acta Gastroenterol Latinoam 1992; 22 (3): 155–9. PubMed

 32 Sherman, S, Ruffolo, TA, Hawes, RH & Glehman, GA. Complications of endoscopic sphincterotomy. Gastroenterology 1991; 101: 1068–75. PubMed

 33 Tanner, A. ERCP: present practice in a single region. Eur J Gastroenterol Hepatol 1996; 8: 145–8. PubMed

 34 Mallery, JS, Baron, TH, Dominitz, JA, Goldstein, JL, Hirota, WK & Jacobson, BC et al. Complications of ERCP. Gastrointest Endosc 2003; 57: 633–8. PubMed

 35 Aliperti, G. Complications related to diagnostic and therapeutic endoscopic retrograde cholangiopancreatography. Gastrointest Endosc Clin N Am 1996; 6: 379–40. PubMed

 36 Freeman, ML. Adverse outcomes of ERCP. Gastrointest Endosc 2002; 56 (6): S273–82. PubMed CrossRef

 37 Farrell, RJ, Mahmud, N, Noonan, N, Kellcher, D & Keeling, PW. Diagnostic and therapeutic ERCP: a large single centre's experience. Ir J Med Sci 2001; 170 (3): 176–80. PubMed

 38 Munoz, SR. Towards safer ERCP: selection, experience and prophylaxis. Rev Esp Enferm Dig (Madrid) 2004; 96 (3): 155–62.

 39 Misra, SP & Dwivedi, M. Complications of endoscopic retrograde cholangiopancreatography and endoscopic sphincterotomy: diagnosis, management and prevention. Natl Med J India 2002; 15: 27–31. PubMed

 40 Halme, L, Doepel, M, von Numers, H, Edgren, J & Ahonen, J. Complications of diagnostic and therapeutic ERCP. Ann Chir Gynaecol 1999; 88: 127–31. PubMed

 41 Rabenstein, T, Schneider, HT, Nicklas, M, Ruppert, T, Katalinic, A & Hahn, EG et al. Impact of skill and experience of the endoscopist on the outcome of endoscopic sphincterotomy techniques. Gastrointest Endosc 1999; 50: 628–36. PubMed

 42 Davis, WZ, Cotton, PB, Arias, R, Williams, D & Onken, JE. ERCP and sphincterotomy in the context of laparoscopic cholecystectomy: academic and community practice patients and results. Am J Gastroenterol 1997; 92: 597–601. PubMed

 43 Escourrou, J, Delvaux, M & Busail, L et al. Clinical results of endoscopic sphincterotomy: comparison of two activity periods in the same endoscopy units. Gastrointest Endosc 1990; 36: 205–6.

 44 Freeman, ML, Nelson, DB, Sherman, S, Haber, GB, Herman, ME & Dorsher, PJ et al. Complications of endoscopic biliary sphincterotomy. N Engl J Med 1996; 335: 909–18. PubMed CrossRef

 45 Newcomer, MK, Jowell, PS & Cotton, PB. Underestimation of adverse events following ERCP: a prospective 30 day follow up study [Abstract]. Gastrointest Endosc 1995; 41: 408.

 46 Zubarik, R, Fleischer, DE, Mastropietro, C & Lopez, J et al. Prospective analysis of complications 30 days after outpatient colonoscopy. Gastrointest Endosc 1999; 50 (3): 322–8. PubMed

 47 Zubarik, R, Eisen, G, Mastropietro, C, Lopez, J, Carroll, J & Benjamin, S et al. Prospective analysis of complications 30 days after outpatient upper endoscopy. Am J Gastroenterol 1999; 94 (6): 1539–45. PubMed CrossRef

 48 Arenson, N, Flamm, CR, Bohn, RI, Mark, DH & Speroff, T. Evidence-based assessment: patient procedure, or operator factors associated with ERCP complications. Gastrointest Endosc 2002; 56: s294–s301. PubMed CrossRef

 49 Deans, GT, Sedman, P, Martin, DF, Royston, CMS, Leow, CK, Thomas, WEG & Brough, WA. Are complications of endoscopic sphincterotomy age related? Gut 1997; 41: 545–8. PubMed

 50 Derkx, HHF, Huibregtse, K & Taminiau, JAJM. The role of endoscopic retrograde cholangiopancreatography in cholestatic infants. Endoscopy 1994; 26: 724–8. PubMed

 51 Guelrud, M, Mujica, C, Jaen, D, Plaz, J & Arias, J. The role of ERCP in the diagnosis and treatment of idiopathic recurrent pancreatitis in children and adolescents. Gastrointest Endosc 1994; 40 (4): 428–33. PubMed

 52 Mitchell, RM, O'Connor, F & Dickey, W. Endoscopic retrograde cholangiopancreatography is safe and effective in patients 90 years of age and older. J Clin Gastroenterol 2003; 36 (1): 72–4. PubMed CrossRef

 53 Hui, CK, Liu, CL, Lai, KC, Chan, SC, Hu, WH & Wong, WM et al. Outcome of emergency ERCP for acute cholangitis in patients 90 years of age and older. Aliment Pharmacol Ther 2004; 19 (11): 1153–8. PubMed CrossRef

 54 Leung, JW, Chung, SC, Sung, JJ, Banez, VP & Li, AK. Urgent endoscopic drainage for acute suppurative cholangitis. Lancet 1989; 1 (8650): 1307–9. PubMed CrossRef

 55 Cotton, PB, Jowell, PS, Baillie, J, Leung, J, Affronti, J & Branch, MS et al. Spectrum of complications after diagnostic ERCP and effect of comorbidities. Gastrointest Endosc 1994; 40 (2): P18.

 56 Jamidar, PA, Beck, GJ, Hoffman, BJ, Lehman, GA, Hawes, RH & Agrawal, RM et al. Endoscopic retrograde cholangiopancreatography in pregnancy. Am J Gastroenterol 1995; 98 (8): 1263–7.

 57 Cohen, S, Bacon, BR, Berlin, JA, Fleischer, D, Hecht, GA & Loehrer, PJ Sr. et al. National Institutes of Health State-of-the-Science Conference Statement: ERCP for diagnosis and therapy, January 14–16, 2002. Gastrointest Endosc 2002; 56: 803–9. PubMed

 58 Cotton, PB. ERCP is most dangerous for people who need it least. Gastrointest Endosc 2001; 54 (4): 535–6. PubMed

 59 Shemesh, E, Klein, E, Czerniak, A, Coret, A & Bat, L. Endoscopic sphincterotomy in patients with gallbladder in situ: the influence of periampullary duodenal diverticula. Surgery 1990; 107: 163–6. PubMed

 60 Vaira, D, Dowsett, JF & Hatfield, ARW et al. Is duodenal diverticulum a risk factor for sphincterotomy? Gut 1989; 30: 939–42. PubMed

 61 Veitch, A & Fairclough, P. Endoscopic diathermy in patients with cardiac pacemakers. Endoscopy 1998; 30 (6): 544–7. PubMed

 62 Sherman, S, Ruffolo, TA, Hawes, RH & Lehman, GA. Complications of endoscopic sphincterotomy: a prospective series with emphasis on the increased risk associated with sphincter of Oddi dysfunction and nondilated bile duct. Gastroenterology 1991; 101: 1068–75. PubMed

 63 Chen, YK, Foliente, RL, Santoro, MJ, Walter, MH & Collen, MJ. Endoscopic sphincterotomy-induced pancreatitis: increased risk associated with nondilated bile ducts and sphincter of Oddi dysfunction. Am J Gastroenterol 1994; 89 (3): 327–33. PubMed

 64 Cotton, PB, Geenen, JE, Sherman, S, Cunningham, JT, Howell, DA & Carr-Locke, DL et al. Endoscopic sphincterotomy for stones by experts is safe, even in younger patients with normal ducts. Ann Surg 1998; 227: 201–4. PubMed

 65 Wilson, MS, Tweedle, DEF & Martin, DF. Common bile duct diameter and complications of endoscopic sphincterotomy. Br J Surg 1992; 79: 1345–7.

 66 Huibregtse, K. Complications of endoscopic sphincterotomy and their prevention. N Engl J Med 1996; 335: 961–2. PubMed CrossRef

67 Boender, J, Nix, GA, de Ridder, MA, van Blankenstein, M, Schutte, HE & Dees, J et al. Endoscopic papillotomy for common bile duct stones: factors influencing the complication rate. Endoscopy 1994; 26: 209–16. PubMed

 68 Mehta, SN, Pavone, E, Barkun, JS, Bouchard, S & Barkun, AN. Predictors of post-ERCP complications in patients with suspected choledocholithiasis. Endoscopy 1998; 30: 457–63. PubMed

 69 Elfant, AB, Bourke, MJ, Alhalel, R, Kortan, PP & Haber, GB. A prospective study of the safety of endoscopic therapy for choledocholithiasis in an outpatient population. Am J Gastroenterol 1996; 91 (8): 1499–502. PubMed

 70 Alsolaiman, M, Cotton, P, Hawes, R, Aliperti, G, Carr-Locke, DL & Fogel, EL et al. Techniques for pancreatic sphincterotomy: lack of expert consensus. Gastrointest Endosc 2004; 59 (5): AB210.

 71 Elton, E, Howell, DA, Parsons, WG, Qaseem, T & Hanson, BL. Endoscopic pancreatic sphincterotomy: indications, outcome, and a safe stentless technique. Gastrointest Endosc 1998; 47: 240–9. PubMed

 72 Berkes, J, Bernklau, S, Halline, A, Venu, R & Brown, R. Minor papillotomy in pancreas divisum: do complications and restenosis rates differ between use of the needle knife papillotome (NKS) vs. ultratapered traction sphicterotome (UTS)? Gastrointest Endosc 2004; 59 (5): AB207.

 73 Delhaye, M, Matos, C & Deviere, J. Endoscopic technique for the management of pancreatitis and its complications. Best Pract Res Clin Gastroenterol 2004; 18 (1): 155–81. PubMed CrossRef

 74 Leung, JW, Banez, VP & Chung, SC. Precut (needle knife) papillotomy for impacted common bile duct stone at the ampulla. Am J Gastroenterol 1990; 85: 991–3. PubMed

 75 Cotton, PB. Precut papillotomy: a risky technique for experts only. Gastrointest Endosc 1989; 35: 578. PubMed

 76 Cotton, PB. Needleknife precut sphincterotomy: the devil is in the indications. Endoscopy 1997; 29: 888. PubMed

 77 Dowsett, JF, Polydorou, AA & Vaira, D et al. Needle knife papillotomy: how safe and how effective? Gastrointest Endosc 1990; 36 (6): 645–6.

 78 Vandervoort, J & Carr-Locke, DL. Needle-knife access papillotomy: an unfairly maligned technique? Endoscopy 1996; 28: 365–6. PubMed

 79 Rabenstein, T, Ruppert, T, Schneider, HT, Hahn, EG & Ell, C. Benefits and risks of needle-knife papillotomy. Gastrointest Endosc 1997; 46: 207–11. PubMed

 80 Baillie, J. Needle-knife papillotomy revisited [editorial; comment]. Gastrointest Endosc 1997; 46: 282. PubMed

 81 Bruins, SW, Schoeman, MN, DiSario, JA, Wolters, F, Tytgat, GN & Huibregtse, K. Needle-knife sphincterotomy as a precut procedure: a retrospective evaluation of efficacy and complications. Endoscopy 1996; 28: 334–9. PubMed

 82 Dhir, V, Swaroop, VS, Mohandas, KM, Jagannath, P & Desouza, LJ. Precut papillotomy using a needle knife: experience in 100 patients with malignant obstructive jaundice. Indian J Gastroenterol 1997; 16: 52–3. PubMed

 83 Foutch, PG. A prospective assessment of results for needle-knife papillotomy and standard endoscopic sphincterotomy. Gastrointest Endosc 1995; 41: 25–32. PubMed

 84 Gholson, CF & Favrot, D. Needle knife papillotomy in a University referral practice: safety and efficacy of a modified technique. J Clin Gastroenterol 1996; 23: 177–80. PubMed CrossRef

 85 Kasmin, FE, Cohen, D, Batra, S, Cohen, SA & Siegel, JH. Needle-knife sphincterotomy in a tertiary referral center: efficacy and complications. Gastrointest Endosc 1996; 44: 48–53. PubMed

 86 Rollhauser, C, Johnson, M & Al Kawas, FH. Needle-knife papillotomy: a helpful and safe adjunct to endoscopic retrograde cholangiopancreatography in a selected population. Endoscopy 1998; 30: 691–6. PubMed

 87 Harewood, GC & Baron, TH. An assessment of the learning curve for precut biliary sphincterotomy. Am J Gastroenterol 2002; 97: 1708–12. PubMed CrossRef

 88 Freeman, ML. Precut (access) sphincterotomy: Techniques Gastrointest Endosc 1999; 1: 40–8.

 89 Binmoeller, KF, Seifert, H, Gerke, H, Seitz, U, Portis, M & Soehendra, N. Papillary roof incision using the Erlangen-type pre-cut papillotome to achieve selective bile duct cannulation. Gastrointest Endosc 1996; 44: 689–95. PubMed

 90 Goff, JS. Long-term experience with the transpancreatic sphincter pre-cut approach to biliary sphincterotomy. Gastrointest Endosc 1999; 50: 642–5. PubMed

 91 Mavrogiannis, C, Liatsos, C, Papanikolaou, IS, Psilopoulos, DI & Goulas, SS et al. Safety of extension of a previous endoscopic sphincterotomy: a prospective study. Am J Gastroenterol 2003; 98 (1): 72–6. PubMed CrossRef

 92 Choudari, CP, Sherman, S, Fogel, EL, Phillips, S, Kochell, A, Flueckiger, J & Lehman, GA. Success of ERCP at a referral center after a previously unsuccessful attempt. Gastrointest Endosc 2000; 52 (4): 478–83. PubMed

 93 Raijman, I & Escalante-Glorsky, S. Is the complication rate the same for index versus repeat biliary sphincterotomy? Gastrointest Endosc 2004; 59 (5): AB193.

 94 May, GR, Cotton, PB, Edmunds, EJ & Chong, W. Removal of stones from the bile duct at ERCP without sphincterotomy. Gastrointest Endosc 1993; 39 (6): 749–54. PubMed

 95 MacMathuna, P. Endoscopic treatment of bile duct stones: should we cut or dilate the sphincter? Am J Gastroenterol 1997; 92 (9): 1411–12. PubMed

 96 Norton, ID, Gostout, CJ, Baron, TH, Geller, A, Petersen, BT & Wiersema, MJ. Safety and outcome of endoscopic snare excision of the major duodenal papilla. Gastrointest Endosc 2002; 56: 239–43. PubMed CrossRef

 97 Desilets, DJ, Dy, RM, Ku, PM, Hanson, BL, Elton, E & Mattia, A et al. Endoscopic management of tumors of the major duodenal papilla: refined techniques to improve outcome and avoid complications. Gastrointest Endosc 2001; 54: 202–8. PubMed CrossRef

 98 Zadorova, Z, Dvofak, M & Hajer, J. Endoscopic therapy of benign tumors of the papilla of Vater. Endoscopy 2001; 33: 345–7. PubMed CrossRef

 99 Catalano, MF, Linder, JD, Chak, A, Sivak, MV Jr, Raijman, I & Geenen, JE et al. Endoscopic management of adenoma of the major duodenal papilla. Gastrointest Endosc 2004; 59: 225–32. PubMed

100 Fujita, N, Noda, Y, Kobayashi, G, Kimura, K & Ito, K. Endoscopic papillectomy: is there room for this procedure in clinical practice? Digestive Endoscopy 2003; 15: 253–5. CrossRef

101 Cheng, C, Sherman, S, Fogel, EL, McHenry, L & Watkins, JL et al. Endoscopic snare papillectomy of ampullary tumors: 10-year review of 55 cases at Indiana University Medical Center. Gastrointest Endosc 2004; 59 (5): AB193.

102 Giorgio, PD & Luca, LD. Comparison of treatment outcomes between biliary plastic stent placements with and without endoscopic sphincterotomy for inoperable malignant common bile duct obstruction. World J Gastroenterol 2004; 10 (8): 1212–14. PubMed

103 Tarnasky, PR, Cunningham, JT, Hawes, RH & Hoffman, BJ et al. Transpapillary stenting of roximal biliary strictures: does biliary sphincterotomy reduce the risk of post-procedure pancreatitis? Gastrointest Endosc 1997; 45: 46–51. PubMed

104 Baron, TH. Endoscopic drainage of pancreatic fluid collections and pancreatic necrosis. In: Gastrointest Endosc Clin N Amer (ed. Lightdale, C) W. B. Saunders Co 2003, 13 (4): 743–64.

105 Plumeri, PA. Informed consent for upper gastrointestinal endoscopy. In: Gastroenterol Clin North America, Sivak, MV W.B. Saunders Co, 1994, ; 4 (2.: 1997, 455–61.

106 Duncan, HD, Hodgkinson, L, Deakin, M & Green, JR. The safety of diagnostic and therapeutic ERCP as a daycase procedure with a selective admission policy. Eur J Gastroenterol Hepatol 1997; 9 (9): 905–8. PubMed

107 Ho, KY, Montes, H, Sossenheimer, MJ, Tham, TC, Ruymann, F & Van Dam, J et al. Features that may predict hospital admission following outpatient therapeutic ERCP. Gastrointest Endosc 1999; 49: 587–92. PubMed

108 Freeman, ML, Nelson, DB, Sherman, S, Haber, GB, Fennerty, MB & DiSario, JA et al. Same-day discharge after endoscopic biliary sphincterotomy: observations from a prospective multicenter complication study. Gastrointest Endosc 1999; 49: 580–6. PubMed

109 Cvetkovski, B, Gerdes, H & Kurtz, RC. Outpatient therapeutic ERCP with endobiliary stent placement for malignant common bile duct obstruction. Gastrointest Endosc 1999; 50: 63–6. PubMed

110 Linder, JD & Tarnasky, P. There are benefits of overnight observation after outpatient ERCP. Gastrointest Endosc 2004; 59 (5): AB208.

111 Testoni, PA, Bagnolo, F, Caporuscio, S & Lella, F. Serum amylase measured four hours after endoscopic sphincterotomy is a reliable predictor of postprocedure pancreatitis. Am J Gastroenterol 1999; 94: 1235–41. PubMed CrossRef

112 Friedland, S, Soetikno, RM, Vandervoort, J, Montes, H, Tham, T & Carr-Locke, DL. Bedside scoring system to predict the risk of developing pancreatitis following ERCP. Endoscopy 2002; 34: 483–8. PubMed CrossRef

113 Barthet, M, Desjeux, A, Gasmi, M, Bellon, P, Hoi, MT & Salducci, J et al. Early refeeding after endoscopic biliary or pancreatic sphincterotomy: a randomized prospective study. Endoscopy 2002; 34 (7): 546–50. PubMed

114 Freeman, ML & Guda, NM. Prevention of post-ERCP pancreatitis: a comprehensive review. Gastrointest Endosc 2004; 59 (7): 845–64. PubMed

115 Freeman, ML, DiSario, JA, Nelson, DB, Fennerty, MB, Lee, JG & Bjorkman, DJ et al. Risk factors for post-ERCP pancreatitis: a prospective, multicenter study. Gastrointest Endosc 2001; 54 (4): 535–6. PubMed

116 Testoni, PA. Why the incidence of post-ERCP pancreatitis varies considerably? Factors affecting the diagnosis and the incidence of this complication. JOP 2002; 3 (6): 195–201. PubMed

117 Johnson, GK, Geenen, JE, Johanson, JF, Sherman, S, Hogan, WJ & Cass, O et al. Evaluation of post-ERCP pancreatitis: potential causes noted during controlled study of differing contrast media. Gastrointest Endosc 1997; 46 (3): 217–22. PubMed

118 Rabenstein, T & Hahn, EG. Post-ERCP pancreatitis: new momentum. Endoscopy 2002; 34 (4): 325–9. PubMed CrossRef

119 Sherman, S & Lehman, GA. ERCP- and endoscopic sphincterotomy-induced pancreatitis. Pancreas 1991; 6 (3): 350–67. PubMed

120 Gottlieb, K & Sherman, S. ERCP and biliary endoscopic sphincterotomy-induced pancreatitis. Gastrointest Endosc Clin N Am 1998; 8: 87–114. PubMed

121 Cotton, PB, Baillie, J, Leung, J, Jowell, PS, Affronti, J & Branch, MS et al. Correlations with post-ERCP pancreatitis [Abstract]. Gastrointest Endosc 1994; 40: P29.

122 Christoforidis, E, Goulimaris, I, Kanellos, I, Tsalis, K, Demetriades, C & Betsis, D. Post-ERCP pancreatitis and hyperamylasemia: patient-related and operative risk factors. Endoscopy 2002; 34: 286–92. PubMed

123 Masci, E, Mariani, A, Curioni, S & Testoni, PA. Risk factors for pancreatitis following endoscopic retrograde cholangiopancreatography: a meta-analysis. Endoscopy 2003; 35: 830–4. PubMed CrossRef

124 Urbach, DR & Rabeneck, L. Population-based study of the risk of acute pancreatitis following ERCP. Gastrointest Endosc 2003; 57 (5): AB116.

125 Roszler, MH & Campbell, WL. Post-ERCP pancreatitis: association with urographic visualization during ERCP. Radiology 1985; 157: 595–8. PubMed

126 Haber, GB. Prevention of post ERCP pancreatitis. Gastrointest Endosc 2000; 51: 100–3. PubMed

127 Cortas, GA, Mehta, SN, Abraham, NS & Barkun, AN. Selective cannulation of the common bile duct: a prospective randomized trial comparing standard catheters with sphincterotomes. Gastrointest Endosc 1999; 50: 775–9. PubMed

128 Schwacha, H, Allgaier, HP, Deibert, P, Olschewski, M, Allgaier, U & Blum, HE. A sphincterotome-based technique for selective transpapillary common bile duct cannulation. Gastrointest Endosc 2000; 5: 387–91. CrossRef

129 Laasch, HU, Tringali, A, Wilbraham, L, Marriott, A, England, RE & Mutignani, M et al. Comparison of standard and steerable catheters for bile duct cannulation in ERCP. Endoscopy 2003; 35: 669–74. PubMed CrossRef

130 Maeda, S, Hayashi, H, Hosokawa, O, Dohden, K, Hattori, M & Morita, M et al. Prospective randomized pilot trial of selective biliary cannulation using pancreatic guide-wire placement. Endoscopy 2003; 35: 721–4. PubMed CrossRef

131 Maldonado, ME, Brady, PG, Mamel, JJ & Robinson, B. Incidence of pancreatitis in patients undergoing sphincter of Oddi manometry (SOM). Am J Gastroenterol 1999; 94: 387–90. PubMed CrossRef

132 Singh, P, Gurudu, SR, Davidoff, S, Sivak, MV Jr, Indaram, A & Kasmin, FE et al. Sphincter of Oddi manometry does not predispose to post-ERCP acute pancreatitis. Gastrointest Endosc 2004; 59 (4): 499–505. PubMed

133 Tarnasky, P, Cunningham, J, Cotton, P, Hoffman, B, Palesch, Y & Freeman, J et al. Pancreatic sphincter hypertension increases the risk of post-ERCP pancreatitis. Endoscopy 1997; 29: 252–7. PubMed

134 Lee, SJ, Song, KS, Chung, JP, Lee, DY, Jeong, YS & Ji, SW et al. Type of electric currents used for standard endoscopic sphincterotomy does not determine the type of complications. Korean J Gastroenterol 2004; 43 (3): 204–10. PubMed

135 MacIntosh, D, Love, J & Abraham, N. Endoscopic sphincterotomy using pure-cut current does not reduce the risk of post-ERCP pancreatitis: a prospective randomized trial. Gastrointest Endosc 2003; 57: AB189.

136 Elta, GH, Barnett, JL, Wille, RT, Brown, KA, Chey, WD & Scheiman, JM. Pure cut electrocautery current for sphincterotomy causes less post-procedure pancreatitis than blended current. Gastrointest Endosc 1998; 47: 149–53. PubMed

137 Stefanidis, G, Karamanolis, G, Viazis, N, Sgouros, S & Papadopoulou, E et al. A comparative study of postendoscopic sphincterotomy complications with various types of electrosurgical current in patients with choledocholithiasis. Gastrointest Endosc 2003; 57 (2): 192–7. PubMed

138 Kohler, A, Maier, M, Benz, C, Martin, WR, Farin, G & Riemann, JF. A new HF current generator with automatically controlled system (Endocut mode) for endoscopic sphincterotomy-preliminary experience. Endoscopy 1998; 30: 351–5. PubMed

139 Norton, I, Bosco, J, Meier, P, Baron, T, Lange, S & Nelson, D et al. A randomized trial of endoscopic sphincterotomy using pure cut versus Endocut electrical waveforms [Abstract]. Gastrointest Endosc 2002; 55: AB175.

140 Katsinelos, P, Mimidis, K, Paroutoglou, G, Christodoulou, K, Pilpilidis, I & Katsiba, D et al. Needle-knife papillotomy: a safe and effective technique in experienced hands. Hepatogastroenterology 2004; 51 (56): 349–52. PubMed

141 Komatsu, Y, Kawabe, T, Toda, N, Ohashi, M, Isayama, M & Tateishi, K et al. Endoscopic papillary balloon dilation for the management of common bile duct stones: experience of 226 cases. Endoscopy 1998; 30: 12–7. PubMed

142 MacMathuna, PM, White, P, Clarke, E, Merriman, R, Lennon, JR & Crowe, J. Endoscopic balloon sphincteroplasty (papillary dilation) for bile duct stones: efficacy, safety, and follow-up in 100 patients. Gastrointest Endosc 1995; 42: 468–74. PubMed

143 Ueno, N & Ozawa, Y. Pancreatitis induced by endoscopic balloon sphincter dilation and changes in serum amylase levels after the procedure. Gastrointest Endosc 1999; 49: 472–6. PubMed

144 Bergman, JJ, Rauws, EA, Fockens, P, van Berkel, AM, Bossuyt, PM & Tijssen, JG et al. Randomised trial of endoscopic balloon dilation versus endoscopic sphincterotomy for removal of bile duct stones. Lancet 1997; 349: 1124–9. PubMed

145 Fujita, N, Maguchi, H, Komatsu, Y, Yasuda, I, Hasebe, O & Igarashi, Y et al. Endoscopic sphincterotomy and endoscopic papillary balloon dilatation for bile duct stones: a prospective randomized controlled multicenter trial. Gastrointest Endosc 2003; 57: 151–5. PubMed

146 Minami, A, Nakatsu, T, Uchida, N, Hirabayashi, S, Fukuma, H & Morshed, SA et al. Papillary dilation vs sphincterotomy in endoscopic removal of bile duct stones: a randomized trial with manometric function. Dig Dis Sci 1995; 40: 2550–4. PubMed

147 Ochi, Y, Mukawa, K, Kiyosawa, K & Akamatsu, T. Comparing the treatment outcomes of endoscopic papillary dilation and endoscopic sphincterotomy for removal of bile duct stones. J Gastroenterol Hepatol 1999; 14: 90–6. PubMed CrossRef

148 Arnold, JC, Benz, C, Martin, WR, Adamek, HE & Riemann, JF. Endoscopic papillary balloon dilation vs. sphincterotomy for removal of common bile duct stones: a prospective randomized pilot study. Endoscopy 2001; 33: 563–7. PubMed CrossRef

149 Vlavianos, P, Chopra, K, Mandalia, S, Anderson, M, Thompson, J & Westaby, D. Endoscopic balloon dilatation versus endoscopic sphincterotomy for the removal of bile duct stones: a prospective randomised trial. Gut 2003; 52: 1165–9. PubMed CrossRef

150 DiSario, JA, Freeman, ML, Bjorkman, DJ, MacMathuna, P, Petersen, B & Sherman, S et al. Endoscopic balloon dilation compared to sphincterotomy (EDES) for extraction of bile duct stones: preliminary results. Gastrointest Endosc in press.

151 Kawabe, T, Komatsu, Y, Tada, M, Toda, N, Ohashi, M, Shiratori, Y & Omata, M. Endoscopic papillary balloon dilation in cirrhotic patients: removal of common bile duct stones without sphincterotomy. Endoscopy 1996; 28: 694–8. PubMed

152 Prat, F, Fritsch, J, Choury, AD, Meduri, B, Pelletier, G & Buffet, C. Endoscopic sphincteroclasy: a useful therapeutic tool for biliary endoscopy in Billroth II gastrectomy patients. Endoscopy 1997; 29: 79–81. PubMed

153 Bergman, JJGHM, van Berkel, A-M, Bruno, MJ, Fockens, P & Rauws, EAJ et al. A randomized trial of endoscopic balloon dilation and endoscopic sphincterotomy for removal of bile duct stones in patients with a prior Billroth II gastrectomy. Gastrointest Endosc 2001; 53 (1): 19–26. PubMed

154 Goff, JS. Common bile duct sphincter of Oddi stenting in patients with suspected sphincter dysfunction. Am J Gastroenterol 1995; 90: 586–9. PubMed

155 Freeman, ML. Prevention of post-ERCP pancreatitis: pharmacologic solution or patient selection and pancreatic stents. Gastroenterology 2003; 124 (7): 1977–80. PubMed CrossRef

156 Sherman, S, Hawes, RH, Troiano, FP & Lehman, GA. Pancreatitis following bile duct sphincter of Oddi manometry: utility of the aspirating catheter. Gastrointest Endosc 1992; 38: 347–50. PubMed

157 Wehrmann, T, Stergiou, N, Schmitt, T, Dietrich, CF & Seifert, H. Reduced risk for pancreatitis after endoscopic microtransducer manometry of the sphincter of Oddi: a randomized comparison with the perfusion manometry technique. Endoscopy 2003; 35: 472–7. PubMed CrossRef

158 Cunliffe, WJ, Cobden, I, Lavelle, MI, Lendrum, R, Tait, NP & Venables, CW. A randomised, prospective study comparing two contrast media in ERCP. Endoscopy 1987; 19: 201–2. PubMed

159 Hannigan, BF, Keeling, PW, Slavin, B & Thompson, RP. Hyperamylasemia after ERCP with ionic and non-ionic contrast media. Gastrointest Endosc 1985; 31: 109–10. PubMed

160 Johnson, GK, Geenen, JE, Bedford, RA, Johanson, J, Cass, O & Sherman, S et al. A comparison of nonionic versus ionic contrast media: results of a prospective, multicenter study. Gastrointest Endosc 1995; 42: 312–16. PubMed

161 O'Connor, HJ, Ellis, WR, Manning, AP, Lintott, DJ, McMahon, MJ & Axon, AT. Iopamidol as contrast medium in endoscopic retrograde pancreatography: a prospective randomised comparison with diatrizoate. Endoscopy 1988; 20: 244–7. PubMed

162 Sherman, S, Hawes, RH, Rathgaber, SW, Uzer, MF, Smith, MT & Khusro, QE et al. Post-ERCP pancreatitis: randomized, prospective study comparing a low- and high-osmolality contrast agent. Gastrointest Endosc 1994; 40 (4): 422–7. PubMed

163 Goebel, C, Hardt, P, Doppl, W, Temme, H, Hackstein, N & Klor, HU. Frequency of pancreatitis after endoscopic retrograde cholangiopancreatography with iopromid or iotrolan: a randomized trial. Eur Radiol 2000; 10 (4): 677–80. PubMed

164 Andriulli, A, Caruso, N, Quitadamo, M, Forlano, R, Leandro, G & Spirito, F et al. Antisecretory vs. antiproteasic drugs in the prevention of post-ERCP pancreatitis: the evidence-based medicine derived from a meta-analysis study. JOP 2003; 4: 41–8. PubMed

165 Andriulli, A, Leandro, G, Niro, G, Mangia, A, Festa, V & Gambassi, G et al. Pharmacologic treatment can prevent pancreatic injury after ERCP. a meta-analysis. Gastrointest Endosc 2000; 51: 1–7. PubMed

166 Raty, S, Sand, J, Pulkkinen, M, Matikainen, M & Nordback, I. Post-ERCP pancreatitis: reduction by routine antibiotics. J Gastrointest Surg 2001; 5: 339–45. PubMed

167 Rabenstein, T, Roggenbuck, S, Framke, B, Martus, P, Fischer, B & Nusko, G et al. Complications of endoscopic sphincterotomy: can heparin prevent acute pancreatitis after ERCP? Gastrointest Endosc 2002; 55 (4): 476–83. PubMed

168 Weiner, GR, Geenen, JE, Hogan, WJ & Catalano, MF. Use of corticosteroids in the prevention of post-ERCP pancreatitis. Gastrointest Endosc 1995; 42: 579–83. PubMed

169 Dumot, JA, Conwell, DL, O'Connor, JB, Ferguson, DR, Vargo, JJ & Barnes, DS et al. Pretreatment with methylprednisolone to prevent ERCP-induced pancreatitis: a randomized, multicenter, placebo-controlled clinical trial. Am J Gastroenterol 1998; 93: 61–5. PubMed

170 Sherman, S, Blaut, U, Watkins, JL, Barnett, J, Freeman, M & Geenen, J et al. Does prophylactic steroid administration reduce the risk and severity of post-ERCP pancreatitis: a randomized prospective multicenter study. Gastrointest Endosc 2003; 58: 23–9. PubMed

171 Manolakopoulos, S, Avgerinos, A, Vlachogiannakos, J, Armonis, A, Viazis, N & Papadimitriou, N et al. Octreotide versus hydrocortisone versus placebo in the prevention of post-ERCP pancreatitis: a multicenter randomized controlled trial. Gastrointest Endosc 2002; 55: 470–5. PubMed CrossRef

172 Budzynska, A, Marek, T, Nowak, A, Kaczor, R & Nowakowska-Dulawa, E. A prospective, randomized, placebo-controlled trial of prednisone and allopurinol in the prevention of ERCP-induced pancreatitis. Endoscopy 2001; 33: 766–72. PubMed

173 De Palma, GD & Catanzano, C. Use of corticosteroids in the prevention of post-ERCP pancreatitis: results of a controlled prospective study. Am J Gastroenterol 1999; 94: 982–5. PubMed CrossRef

174 Prat, F, Amaris, J, Ducot, B, Bocquentin, M, Fritsch, J & Choury, AD et al. Nifedipine for prevention of post-ERCP pancreatitis: a prospective, double-blind randomized study. Gastrointest Endosc 2002; 56: 202–8. PubMed CrossRef

175 Sand, J & Nordback, I. Prospective randomized trial of the effect of nifedipine on pancreatic irritation after endoscopic retrograde cholangiopancreatography. Digestion 1993; 54: 105–11. PubMed

176 Binmoeller, KF, Harris, AG, Dumas, R, Grimaldi, C & Delmont, JP. Does the somatostatin analog octreotide protect against ERCP induced pancreatitis? Gut 1992; 33: 1129–33. PubMed

177 Arvanitidis, D, Anagnostopoulos, GK, Giannopoulos, D, Pantes, A, Agaritsi, R & Margantinis, G et al. Can somatostatin prevent post-ERCP pancreatitis? Results of a randomized controlled trial. J Gastroenterol Hepatol 2004; 19: 278–82. PubMed CrossRef

178 Bordas, JM, Toledo, V, Mondelo, F & Rodes, J. Prevention of pancreatic reactions by bolus somatostatin administration in patients undergoing endoscopic retrograde cholangio-pancreatography and endoscopic sphincterotomy. Horm Res 1988; 29: 106–8. PubMed

179 Bordas, JM, Toledo-Pimentel, V, Llach, J, Elena, M, Mondelo, F & Gines, A et al. Effects of bolus somatostatin in preventing pancreatitis after endoscopic pancreatography: results of a randomized study. Gastrointest Endosc 1998; 47: 230–4. PubMed

180 Guelrud, M, Mendoza, S, Viera, L & Gelrud, D. Somatostatin prevents acute pancreatitis after pancreatic duct sphincter hydrostatic balloon dilation in patients with idiopathic recurrent pancreatitis. Gastrointest Endosc 1991; 37: 44–7. PubMed

181 Persson, B, Slezak, P, Efendic, S & Haggmark, A. Can somatostatin prevent injection pancreatitis after ERCP? Hepatogastroenterology 1992; 39: 259–61. PubMed

182 Poon, RT, Yeung, C, Lo, CM, Yuen, WK, Liu, CL & Fan, ST. Prophylactic effect of somatostatin on post-ERCP pancreatitis: a randomized controlled trial. Gastrointest Endosc 1999; 49: 593–8. PubMed

183 Saari, A, Kivilaakso, E & Schroeder, P. The influence of somatostatin on pancreatic irritation after pancreatography: an experimental and clinical study. Surg Res Comm 1988; 2: 271–8.

184 Testoni, PA, Masci, E, Bagnolo, F & Tittobello, A. Endoscopic papillo-sphincterotomy: prevention of pancreatic reaction by somatostatin. Ital J Gastroenterol 1988; 20: 70–3.

185 Arcidiacono, R, Gambitta, P, Rossi, A, Grosso, C, Bini, M & Zanasi, G. The use of a long-acting somatostatin analog (octreotide) for prophylaxis of acute pancreatitis after endoscopic sphincterotomy. Endoscopy 1994; 26: 715–18. PubMed

186 Arvanitidis, D, Hatzipanayiotis, J, Koutsounopoulos, G & Frangou, E. The effect of octreotide on the prevention of acute pancreatitis and hyperamylasemia after diagnostic and therapeutic ERCP. Hepatogastroenterology 1998; 45: 248–52. PubMed

187 Sternlieb, JM, Aronchick, CA, Retig, JN, Dabezies, M, Saunders, F & Goosenberg, E et al. A multicenter, randomized, controlled trial to evaluate the effect of prophylactic octreotide on ERCP-induced pancreatitis. Am J Gastroenterol 1992; 87: 1561–6. PubMed

188 Testoni, PA, Lella, F, Bagnolo, F, Caporuscio, S, Cattani, L & Colombo, E et al. Long-term prophylactic administration of octreotide reduces the rise in serum amylase after endoscopic procedures on Vater's papilla. Pancreas 1996; 13: 61–5. PubMed

189 Testoni, PA, Bagnolo, F, Andriulli, A, Bernasconi, G, Crotta, S & Lella, F et al. Octreotide 24-h prophylaxis in patients at high risk for post-ERCP pancreatitis: results of a multicenter, randomized, controlled trial. Aliment Pharmacol Ther 2001; 15: 965–72. PubMed CrossRef

190 Tulassay, Z & Papp, J. The effect of long-acting somatostatin analog on enzyme changes after endoscopic pancreatography. Gastrointest Endosc 1991; 37: 48–50. PubMed

191 Tulassay, Z, Dobronte, Z, Pronai, L, Zagoni, T & Juhasz, L. Octreotide in the prevention of pancreatic injury associated with endoscopic cholangiopancreatography. Aliment Pharmacol Ther 1998; 12: 1109–12. PubMed CrossRef

192 Testoni, PA, Lella, F, Bagnolo, F, Buizza, M & Colombo, E. Controlled trial of different dosages of octreotide in the prevention of hyperamylasemia induced by endoscopic papillosphincterotomy. Ital J Gastroenterol 1994; 26: 431–6. PubMed

193 Binmoeller, KF, Dumas, R, Harris, AG & Delmont, JP. Effect of somatostatin analog octreotide on human sphincter of Oddi. Dig Dis Sci 1992; 37: 773–7. PubMed

194 Sudhindran, S, Bromwich, E & Edwards, PR. Prospective randomized double-blind placebo-controlled trial of glyceryl trinitrate in endoscopic retrograde cholangiopancreatography-induced pancreatitis. Br J Surg 2001; 88: 1178–82. PubMed CrossRef

195 Kaffes, A, Alrubaie, A & Ding, S et al. A prospective, randomized, double-blind, placebo-controlled trial of transdermal glyceryl trinitrate in technical success of ERCP and the prevention of post-ERCP pancreatitis: preliminary results. Gastrointest Endosc2003; 57: AB191.

196 Moreto, M, Zaballa, M, Casado, I, Merino, O, Rueda, M & Ramirez, K et al. Transdermal glyceryl trinitrate for prevention of post-ERCP pancreatitis: a randomized double-blind trial. Gastrointest Endosc 2003; 57: 1–7. PubMed

197 Schwartz, JJ, Lew, RJ, Ahmad, NA, Shah, JN, Ginsberg, GG & Kochman, ML et al. The effect of lidocaine sprayed on the major duodenal papilla on the frequency of post-ERCP pancreatitis. Gastrointest Endosc 2004; 59: 179–84. PubMed

198 Cavallini, G, Tittobello, A, Frulloni, L, Masci, E, Mariana, A & Di Francesco, V. Gabexate for the prevention of pancreatic damage related to endoscopic retrograde cholangiopancreatography: gabexate in digestive endoscopy – Italian Group. N Engl J Med 1996; 335: 919–23. PubMed

199 Masci, E, Cavallini, G, Mariani, A, Frulloni, L, Testoni, PA & Curioni, S et al. Comparison of two dosing regimens of gabexate in the prophylaxis of post-ERCP pancreatitis. Am J Gastroenterol 2003; 98: 2182–6. PubMed CrossRef

200 Van Laethem, JL, Marchant, A, Delvaux, A, Goldman, M, Robberecht, P & Velu, T et al. Interleukin 10 prevents necrosis in murine experimental acute pancreatitis. Gastroenterology 1995; 108: 1917–22. PubMed

201 Deviere, J, Le Moine, O, Van Laethem, JL, Eisendrath, P, Ghilain, A & Severs, N et al. Interleukin 10 reduces the incidence of pancreatitis after therapeutic endoscopic retrograde cholangiopancreatography. Gastroenterology 2001; 120: 498–505. PubMed

202 Dumot, JA, Conwell, DL, Zuccaro, G Jr, Vargo, JJ, Shay, SS & Easley, KA et al. A randomized, double blind study of interleukin 10 for the prevention of ERCP-induced pancreatitis. Am J Gastroenterol 2001; 96: 2098–102. PubMed CrossRef

203 Singh, P, Lee, T, Davidoff, S & Bank, S. Efficacy of Interleukin 10 (IL10) in the prevention of post-ERCP pancreatitis: a meta-analysis [Abstract]. Gastrointest Endosc 2002; 55: AB150.

204 Murray, B, Carter, R, Imrie, C, Evans, S & O'Suilleabhain, C. Diclofenac reduces the incidence of acute pancreatitis after endoscopic retrograde cholangiopancreatography. Gastroenterology 2003; 124: 1786–91. PubMed CrossRef

205 Jowell, PS, Branch, S, Robuck-Mangum, G, Fein, S, Purich, ED & Stiffler, H et al. Synthetic secretin administered at the start of the procedure significantly reduces the risk of post-ERCP pancreatitis: a randomized, double-blind, placebo controlled trial (presented at. DDW 2003. in press.

206 Smithline, A, Silverman, W, Rogers, D, Nisi, R, Wiersema, M & Jamidar, P et al. Effect of prophylactic main pancreatic duct stenting on the incidence of biliary endoscopic sphincterotomy-induced pancreatitis in high-risk patients. Gastrointest Endosc 1993; 39: 652–7. PubMed

207 Tarnasky, PR, Palesch, YY, Cunningham, JT, Mauldin, PD, Cotton, PB & Hawes, RH. Pancreatic stenting prevents pancreatitis after biliary sphincterotomy in patients with sphincter of Oddi dysfunction. Gastroenterology 1998; 115: 1518–24. PubMed

208 Fogel, EL, Eversman, D, Jamidar, P, Sherman, S & Lehman, GA. Sphincter of Oddi dysfunction: pancreaticobiliary sphincterotomy with pancreatic stent placement has a lower rate of pancreatitis than biliary sphincterotomy alone. Endoscopy 2002; 34: 280–5. PubMed CrossRef

209 Aizawa, T & Ueno, N. Stent placement in the pancreatic duct prevents pancreatitis after endoscopic sphincter dilation for removal of bile duct stones. Gastrointest Endosc 2001; 54: 209–13. PubMed CrossRef

210 Fazel, A, Quadri, A, Catalano, MF, Meyerson, SM & Geenen, JE. Does a pancreatic duct stent prevent post-ERCP pancreatitis? A prospective randomized study. Gastrointest Endosc 2003; 57: 291–4. PubMed

211 Sherman, S, Earle, DT, Bucksot, L, Baute, P, Gottlieb, K & Lehman, G. Does leaving a main pancreatic duct stent in place reduce the incidence of precut biliary sphincterotomy (ES) -induced pancreatitis? A final analysis of a randomized prospective study. (Abstract). Gastrointest Endosc 1996; 43: A489.

212 Freeman, ML, Overby, C & Qi, D. Pancreatic stent insertion: consequences of failure and results of a modified technique to maximize success. Gastrointest Endosc 2004; 5: 8–14.

213 Tarnasky, PR. Mechanical prevention of post-ERCP pancreatitis by pancreatic stents: results, techniques, and indications. JOP 2003; 4 (1): 58–67. PubMed

214 Vege, SS, Chari, ST, Petersen, BT, Baron, TH, Jones, JR & Munukuti, PN et al. Morbidity and mortality of ERCP-induced severe acute pancreatitis. Gastrointest Endosc 2004; 49 (5): AB207.

215 Enns, R, Eloubeidi, MA, Mergener, K, Jowell, PS, Branch, MS, Pappas, TM & Baillie, J. ERCP-related perforations. risk factors and management. Endoscopy 2002; 34 (4): 293–8. PubMed CrossRef

216 Jayaprakash, B & Wright, R. Common bile duct perforation. an unusual complication of ERCP. Gastrointest Endosc 1986; 32: 246–7. PubMed

217 Howard, TJ, Tan, T, Lehman, GA, Sherman, S, Madura, JA, Fogel, E, Swack, ML & Kopecky, KK. Classification and management of perforations complicating endoscopic sphincterotomy. Surgery 1999; 126 (4): 658–65. PubMed CrossRef

218 Zissin, R, Shapiro-Feinberg, M, Oscadchy, A, Pomeraz, I, Leichtmann, G & Novis, B. Retroperitoneal perforation during endoscopic sphincterotomy: imaging findings. Abdom Imaging 2000; 25 (3): 279–82. PubMed CrossRef

219 Genzlinger, JL, McPhee, MS, Fisher, JK, Jacob, KM & Helzberg, JH. Significance of retroperitoneal air after endoscopic retrograde cholangiopancreatography with sphincterotomy. Am J Gastroenterol 1999; 94 (5): 1267–70. PubMed CrossRef

220 Sezgin, O, Tezel, A & Sahin, B. Limited duodenal pneumatosis during needle-knife sphincterotomy. Endoscopy 1999; 31: 554. PubMed CrossRef

221 Savides, T, Sherman, S & Kadell, B et al. Bilateral pneumothoraces and subcutaneous emphysema after endoscopic sphincterotomy. Gastrointest Endosc 1993; 39: 814. PubMed

222 Ciaccia, D, Branch, MS & Baillie, J. Pneumomediastinum after endoscopic sphincterotomy. Am J Gastroenterol 1995; 90: 475–7. PubMed

223 Byrne, P, Leung, JWC & Cotton, PB. Retroperitoneal perforation during duodenoscopic sphincterotomy. Radiol 1984; 150: 383–4.

224 Martin, DF & Tweedle, DE. Retroperitoneal perforation during ERCP and endoscopic sphincterotomy: causes, clinical features and management. Endoscopy 1990; 22 (4): 174–5. PubMed

225 Dunham, F, Bourgeois, N & Gelin, M et al. Retroperitoneal perforations following endoscopic sphincterotomy; clinical course and management. Endoscopy 1982; 14: 92–6. PubMed

226 Stapfer, M, Selby, RR, Stain, SC, Katkhouda, N & Parekh, D et al. Management of duodenal perforation after endoscopic retrograde cholangiopancreatography and sphincterotomy. Ann Surg 2000; 232 (2): 191–8. PubMed CrossRef

227 Scarlett, PY & Falk, GL. The management of perforation of the duodenum following endoscopic sphincterotomy: a proposal for selective therapy [Review]. Aust N Z J Surg 1994; 64: 843–6. PubMed

228 Baron, TH, Gostout, CJ & Herman, L. Hemoclip repair of a sphincterotomy-induced duodenal perforation. Gastrointest Endosc 2000; 52 (4): 566–8. PubMed

229 Tezel, A, Sahin, T, Kosar, Y, Oguz, D, Sahin, B & Cumhur, T. Esophageal perforation due to endoscopic retrograde cholangiopancreatography. Endoscopy 1998; 30: 52.

230 Lee, DW & Chan, AC. Visualization of the peritoneum during endoscopic retrograde cholangiopancreatography. Hong Kong Med J 2001; 7 (4): 445–6. PubMed

231 Cotton, PB. Take care with the tip of your video duodenoscope. Gastrointest Endosc 1989; 35: 582–3. PubMed

232 Costamagna, G. ERCP and endoscopic sphincterotomy in Billroth II patients: a demanding technique for experts only? Ital J Gastroenterol Hepatol 1998; 30: 306–9. PubMed

233 Faylona, JM, Qadir, A, Chan, AC, Lau, JY & Chung, SC. Small-bowel perforations related to endoscopic retrograde cholangiopancreatography (ERCP) in patients with Billroth II gastrectomy. Endoscopy 1999; 31 (7): 546–9. PubMed CrossRef

234 Kim, MH, Lee, SK, Lee, MH, Myung, SJ, Yoo, BM & Seo, DW et al. Endoscopic retrograde cholangiopancreatography and needle knife sphincterotomy in patients with Billroth II gastrectomy: a comparative study of the forward-viewing endoscope and the side-viewing duodenoscope. Endoscopy 1997; 29: 82–5. PubMed

235 Gould, J, Train, JS & Dan, SL et al. Duodenal perforations as a delayed complication of placement of a biliary endoprosthesis. Radiology 1988; 157: 467–90.

236 Ruffolo, TA, Lehman, GA & Sherman, S et al. Biliary stent migration with colonic diverticular impaction. Gastrointest Endosc 1991; 38: 81–3.

237 Nelson, DB. Infectious disease complications of GI Endoscopy: Part I, endogenous infections. Gastrointest Endosc 2003; 57 (4): 546–56. PubMed

238 Benchimol, D, Bernard, JL, Mouroux, J, Dumas, R, Elkaim, D, Chazal, M, Bourgeon, A & Richelme, H. Infectious complications of endoscopic retrograde cholangio-pancreatography managed in a surgical unit. Int Surg 1992; 77: 270–3. PubMed

239 Motte, S, Deviere, J, Dumonceau, J-M, Serruys, E, Thus, J-P & Cremer, M. Risk factors for septicemia following endoscopic biliary stenting. Gastroenterology 1991; 101: 1274–81. PubMed

240 , YuJ-L & Luungh, A. Review. Infections associated with biliary drains. Scand J Gastroenterol 1996; 31: 625–30. PubMed

241 Deviere, J, Motte, S & Dumonceau, JM et al. Septicemia after endoscopic retrograde cholangiopancreatography. Endoscopy 1990; 22: 72–5. PubMed

242 Mollison, LC, Desmond, PV & Stockman, KA et al. A prospective study of septic complications of endoscopic retrograde cholangiopancreatography. J Gastroenterol Hepatol 1994; 9: 55–9. PubMed

243 Novello, P, Hagege, H & Ducreux, M et al. Septicemias after endoscopic retrograde cholangiopancreatography. Risk factors and antibiotic prophylaxis. Gastroenterol Clin Biol 1993; 17: 897–902. PubMed

244 Kullman, E, Borch, K, Lindstrom, E, Ansehn, S, Ihse, I & Anderberg, B. Bacteremia following diagnostic and therapeutic ERCP. Gastrointest Endosc 1992; 38 (4): 444–9. PubMed

245 Axon, ATR, Cotton, PB, Phillips, I & Avery, SA. Disinfection of gastrointestinal fiber endoscopes. Lancet 1974; 1: 656–8. PubMed CrossRef

246 Katsinelos, P, Dimiropoulos, S, Katsiba, D, Arvaniti, M & Tsolkas, P et al. Pseudomonas aeruginosa liver abscesses after diagnostic endoscopic retrograde cholangiography in two patients with sphincter of Oddi dysfunction type 2. Surg Endosc 2002; 16 (11): 1638. PubMed

247 Bass, DH, Oliver, S & Bornman, PC. Pseudomonas septicaemia after endoscopic retrograde cholangiopancreatography – an unresolved problem [Review]. S Afr Med J 1990; 77: 509–11. PubMed

248 Struelens, MJ, Rost, F & Deplano, A et al. Pseudomonas aeruginosa and enterobacteriaceae bacteremia after biliary endoscopy: an outbreak investigation using DNA macrorestriction analysis. Am J Med 1993; 95: 489–98. PubMed CrossRef

249 Allen, JI, Allen, MO & Olson, MM et al. Pseudomonas infection of the biliary system resulting from the use of a contaminated endoscope. Gastroenterology 1987; 192: 759–63.

250 Novello, P, Hage.g. e, H, Buffet, C, Fritsch, J, Choury, A & Etienne, JP. Septicemia after endoscopic retrograde cholangiopancreatography. Gastroenterology 1992; 103: 1367. PubMed

251 Baker, JP, Haber, GB, Gray, RR & Handy, S. Emphysematous cholecystitis complicating endoscopic retrograde cholangiography. Gastrointest Endosc 1982; 28 (3): 184–6. PubMed

252 Alvarez, C, Hunt, K, Ashley, SW & Reber, HA. Emphysematous cholecystitis fater ERCP. Dig Dis Sci 1994; 39: 1719–23. PubMed

253 Tseng, A, Sales, DJ, Simonowitz, DA & Enker, WE. Pancreas abscess: a fatal complication of endoscopic cholangiopancreatography (ERCP). Endoscopy 1977; 9: 250–3. PubMed

254 Finkelstein, R, Yassin, K, Suissa, A, Lavy, A & Eidelman, S. Failure of cefonicid prophylaxis for infectious complications related to endoscopic retrograde cholangiopancreatography. Clin Infect Dis 1996; 23: 378–9. PubMed

255 Byl, B & Deviere, J. Antibiotic prophylaxis before endoscopic retrograde cholangiopancreatography. Ann Int Med 1997; 126 (12): 1001. PubMed

256 Sauter, G, Grabein, B, Huber, G, Mannes, GA, Ruckdeschel, G & Sauerbruch, T. Antibiotic prophylaxis of infectious complications with endoscopic retrograde cholangiopancreatography: a randomized controlled study. Endoscopy 1990; 22: 164–7. PubMed

257 Subbani, JM, Kibbler, C & Dooley, JS. Review article: antibiotic prophylaxis for endoscopic retrograde cholangiopancreatography (ERCP). Aliment Pharmacol Ther 1999; 13 (2): 103–16. PubMed CrossRef

258 Niederau, C, Pohlmann, U, Lubke, H & Thomas, L. Prophylactic antibiotic treatment in therapeutic or complicated diagnostic ERCP. results of a randomized controlled clinical study. Gastrointest Endosc 1994; 40 (5): 533–7. PubMed

259 Van den Hazel, SJ, Speelman, P, Tytgat, GNJ, Dankert, J & van Leeuwen, DJ. Role of antibiotics in the treatment and prevention of acute and recurrent cholangitis. Clin Infect Dis 1994; 19: 279–86. PubMed

260 Byl, B, Deviere, J, Struelens, MJ, Roucloux, I, De Conick, A, Thys, J-P & Cremer, M. Antibiotic prophylaxis for infectious complications after therapeutic endoscopic retrograde cholangiopancreatography: a randomized, double-blind, placebo-controlled study. Clin Infect Dis 1995; 20: 1236–40. PubMed

261 Alveyn, CG, Robertson, DAF, Wright, R, Lowes, JA & Tillotson, G. Prevention of sepsis following endoscopic retrograde cholangiopancreatography. J Hosp Infect 1991; 19 (Suppl. C): 65–70. PubMed CrossRef

262 Van de Meeberg, PC & van Berge Henegouwen, GP. No routine antibiotic prophylaxis necessary in endoscopic retrograde cholangiopancreatography. Ned Tijdschr Geneeskd 1997; 141 (9): 412–3. PubMed

263 Alveyn, CG. Antimicrobial prophylaxis during biliary endoscopic procedures. Antimicrob Chemother1993, 31 (Supple B), 101–5.

264 Connor, P, Hawes, RH, Cunningham, JT & Cotton, PB. Antibiotics before ERCP; a sequential quality improvement approach. Gastrointest Endosc2002, 55: AB97.

265 Goodall, RJR. Bleeding after endoscopic sphincterotomy. Ann R Coll Surg 1985; 67: 87–8.

266 Mellinger, JD & Ponsky, JL. Bleeding after endoscopic sphincterotomy as an underestimated entity. Surgery 1991; 172: 465–9.

267 Moreira, VF, Arribas, R & Sanroman, AL et al. Choledocholithiasis in cirrhotic patients: is endoscopic sphincterotomy the safest choice? Am J Gastroenterol 1991; 86: 1006–9. PubMed

268 Nelson, DB & Freeman, ML. Major hemorrhage from endoscopic sphincterotomy: risk factor analysis. J Clin Gastroenterol 1994; 19 (4): 283–7. PubMed

269 http://www.asge.org/

270 Hussain, N, Toubouti, Y & Jean-Francois, B. Are medications that affect platelet function associated with bleeding following therapeutic Endoscopy: a case-control study. Am J Gastroenterol 2003; 98 (9): S220. CrossRef

271 Wilcox, CM, Canakis, J, Monkemuller, KE, Bondora, AW & Geels, W. Patterns of bleeding after endoscopic sphincterotomy, the subsequent risk of bleeding, and the role of epinephrine injection. Am J Gastroenterol 2004; 99: 244–8. PubMed CrossRef

272 Rafael, RF, Sadurski, R, Cotton, PB, Patel, RS, Hawes, RH & Cunningham, JT. Post-sphincterotomy bleeding after the introduction of a microprocessor-controlled electrocautery. Does the new technology make the difference? GI Endosc2004.

273 Gorelick, A, Cannon, M, Barnett, J, Chey, W, Scheiman, J & Elta, G. First cut, then blend: an electrocautery technique affecting bleeding at sphincterotomy. Endoscopy 2001; 33: 976–80. PubMed CrossRef

274 Park, DH, Kim, M-H, Lee, SK, Lee, SS, Song, MH & Choi, JS et al. Endoscopic sphincterotomy versus endoscopic papillary balloon dilatation for choledocholithiasis in liver cirrhosis with coagulopathy. Gastrointest Endosc 2004; 59 (5): AB192.

275 Moneira, VF, Morono, E & Larraon, JL et al. Choledocholithiasis in cirrhotic patients: is endoscopic sphincterotomy the safest choice? Am J Gastroenterol 1991; 86: 1006–9. PubMed

276 Mosca, S & Galasso, G. Immediate and late bleeding after endoscopic sphincterotomy. Endoscopy 1999; 31 (3): 278–9. PubMed

277 Ooujaoude, J, Pelletier, G, Fritsch, J, Choury, A, Lefebvre, JF, Roche, A, Frouge, C, Liguory, C & Etienne, JP. Management of clinically relevant bleeding following endoscopic sphincterotomy. Endoscopy 1994; 26: 217–21. PubMed

278 Matsushita, M, Hajiro, K & Takakuwa, H et al. Effective hemostatic injection above the bleeding site for uncontrolled bleeding after endoscopic sphincterotomy. Gastrointest Endosc 2000; 51: 221–3. PubMed

279 Vasconez, C, Llach, J & Bordas, L et al. Injection treatment of hemorrhage induced by endoscopic sphincterotomy. Endoscopy 1998; 1: 37–9.

280 Petersen, S, Henke, G, Freitag, M & Ludwig, K. Management of hemorrhage and perforation following endoscopic sphincterotomy. Zentralbl Chir 2001; 126 (10): 805–9. PubMed CrossRef

281 Costamagna, G. What to do when the papilla bleeds after endoscopic sphincterotomy. Endoscopy 1998; 30: 40–2. PubMed

282 Petersen, S, Henke, G, Freitag, M & Ludwig, K. Management of hemorrhage and perforation following endoscopic sphincterotomy. Zentralbl Chir 2001; 125 (10): 805–9. CrossRef

283 Sherman, S, Hawes, RH, Nisi, R & Lehman, GA. Endoscopic sphincterotomy-induced hemorrhage: treatment with multipolar electrocoagulation. Gastrointest Endosc 1992; 38: 123–6. PubMed

284 Lo, SK & Patel, A. Treatment of endoscopic sphincterotomy-induced hemorrhage: injection of bleeding site with ERCP contrast solution using a minor papilla diagnostic catheter. Gastrointest Endosc 1993; 39: 346–9.

285 Baron, TH, Norton, ID & Herman, L. Endoscopic hemoclip for post-sphincterotomy bleeding. Gastrointest Endosc 2000; 52: 662. PubMed

286 Leung, JWC, Chan, FK, Sung, JJ & Chung, S. Endoscopic sphincterotomy-induced hemorrhage: a study of risk factors and the role of epinephrine injection. Gastrointest Endosc 1995; 52: 550–4.

287 Saeed, M, Kadir, S, Kaufman, SL & Murray, RR et al. Bleeding following endoscopic sphincterotomy: angiographic management by transcatheter embolization. Gastrointest Endosc 1989; 35: 300–3. PubMed

288 Margulies, C, Siqueira, ES, Silverman, WB, Lin, XS, Martin, JA & Rabinovitz, M et al. The effect of endoscopic sphincterotomy on acute and chronic complications of biliary endoprostheses. Gastrointest Endosc 1999; 49: 716–9. PubMed

289 Speer, AG, Cotton, PB & Rode, J et al. Biliary stent blockage with bacterial biofilm: a light and electron microscopy study. Ann Intern Med 1988; 108: 546–33. PubMed

290 Johanson, JF, Schmalz, MJ & Geenen, JE. Incidence and risk factors for biliary and pancreatic stent migration. Gastrointest Endosc 1992; 38: 341–6. PubMed

291 Kozarek, RA. Pancreatic stents can induce ductal changes consistent with chronic pancreatitis. Gastrointest Endosc 1990; 36: 93–5. PubMed

292 Sherman, S, Hawes, RH, Savides, TJ, Gress, FG, Ikenberry, SO & Smith, MT et al. Stent-induced pancreatic ductal and parenchymal changes: correlation of endoscopic ultrasound with ERCP. Gastrointest Endosc 1996; 44: 276–82. PubMed

293 Siegel, J & Veerappan, A. Endoscopic management of pancreatic disorders: potential risks of pancreatic prostheses. Endoscopy 1991; 23: 177–80. PubMed

294 Smith, MT, Sherman, S, Ikenberry, SO, Hawes, RH & Lehman, GA. Alterations in pancreatic ductal morphology following polyethylene pancreatic stent therapy. Gastrointest Endosc 1996; 44: 268–75. PubMed

295 Rashdan, A, Fogel, E, McHenry, L, Sherman, S, Schmidt, S & Lazzell, L et al. Pancreatic ductal changes following small diameter long length unflanged pancreatic stent placement [Abstract]. Gastrointest Endosc 2003, 57: AB213.

296 Dolan, R, Pinkas, H & Brady, PG. Acute cholecystitis after palliative stenting for malignant obstruction of the biliary tree. Gastrointest Endosc 1993; 39: 447–9. PubMed

297 Ainly, CC, Williams, SJ, Smith, AC, Hatfield, ARW, Russel, RCG & Lees, WR. Gallbladder sepsis after stent insertion for bile duct obstruction: management by percutaneous cholecystostomy. Br J Surg 1991; 78: 961–3. PubMed

298 Leung, JWC, Chung, SCS, Sung, YJ & Li, MKW. Acute cholecystitis after stenting of the common bile duct for obstruction secondary to pancreatic cancer. Gastrointest Endosc 1989; 35: 109–10. PubMed

299 Payne, WG, Norman, JG & Pinkas, H. Endoscopic basket impaction [Review]. Ann Surg 1995; 61: 464–7.

300 Lee, JF, Leung, JWC & Cotton, PB. Acute cardiovascular complications of endoscopy: prevalence and clinical characteristics. Dig Dis 1995; 13 (2): 130–5. PubMed

301 Lieberman, DA, Wuerker, CK & Katon, RM. Cardiopulmonary risk of esophagogastroduodenoscopy: role of endoscope diameter and systemic sedation. Gastroenterology 1985; 88: 468–72. PubMed

302 Rosenberg, J, Jorgensen, LN & Rasmussen, V et al. Hypoxaemia and myocardial ischaemia during and after endoscopic cholangiopancreatography: Call for further studies. Scand J Gastroenterol 1992; 27: 717–20. PubMed

303 Johnston, SD, McKenna, A & Tham, TC. Silent myocardial ischaemia during endoscopic retrograde cholangiopancreatography. Endoscopy 2003; 35 (12): 1039–42. PubMed CrossRef

304 Freeman, ML. Sedation and monitoring for gastrointestinal endoscopy. In: Gastrointest Endosc Clin N Amer. Sivak, MV W.B. Saunders Co 1994, 4 (3): 475–98.

305 Despland, M, Calvein, PA, Mentha, G & Rohner, A. Gallstone ileus and bowel perforation after endoscopic sphincterotomy. Am J Gastroenterol 1992; 87: 886–7. PubMed

306 Prackup, GM, Baborjee, B, Piorkowski, RJ & Rossan, RS. Gallstone ileus following endoscopic sphincterotomy. J Clin Gastroenterol 1990; 12: 230–2. PubMed

307 To, EW, Pang, PC & Lee, DW. Temporomandibular joint dislocation during endoscopic retrograde cholangiopancreatography examination. Endoscopy 2000; 32 (6): S36–7. PubMed

308 Blind, PJ, Oberg, L & Hedberg, B. Hepatic portal vein gas following ERCP with sphincterotomy. Eur J Surg 1991; 157: 299–300. PubMed

309 Dickey, W, Huibregtse, K & Rauws, EAJ et al. Direct pancreatic lymphangiography during ERCP. Gastrointest Endosc 1995; 41: 528. PubMed

310 Herman, JB, Levine, MS & Long, WB. Portal venous gas as a complication of ERCP and endoscopic sphincterotomy. Am J Gastroenterol 1995; 90: 828–9. PubMed

311 Seibert, DG, Al-Kawas, FH, Graves, J & Gaskins, RD. Prospective evaluation of renal function following ERCP. Endoscopy 1991; 23: 355–6. PubMed

312 Draganov, P & Cotton, PB. Iodinated contrast sensitivity in ERCP. Am J Gastroenterol 2000; 95 (6): 1398–401. PubMed

313 Beuers, U, Spengler, U, Sackmann, M, Paumgartner, G & Sauerbruch, T. Deterioration of cholestasis after endoscopic retrograde cholangiography in advanced primary sclerosing cholangitis. J Hepatol 1992; 15: 140–3. PubMed

314 Furman, G & Morgenstern, L. Splenic injury and abscess complicating endoscopic retrograde cholangiopancreatography. Surg Endosc 1993; 39: 343–4.

315 Lewis, FW, Moloo, N, Steigmann, GV & Goff, JS. Splenic injury complicating therapeutic upper gastrointestinal endoscopy and ERCP. Gastrointest Endosc 1991; 37: 632–3. PubMed

316 Tronsden, E, Rosseland, AR, Moer, A & Solheim, K. Rupture of the spleen following endoscopic retrograde cholangiopancreatography. Acta Chir Scand 1989; 155: 75–6. PubMed

317 Gilad, J, David, A, Hertzanu, Y, Flusser, D, Wolak, T, Levi, I & Krugliak, P. Endoscopic biliary sphincterotomy complicated by abscess formation in a simple renal cyst. Endoscopy 1999; 31 (2): S5–6. PubMed

318 Katsinelos, P, Eugenidis, N, Vasilliadis, T, Tsoukalas, I, Xiarchos, P & Triantopoulos, I. Hemolysis due to G-6-PD deficiency induced by endoscopic sphincterotomy. Endoscopy 1998; 30 (6): 581. PubMed

319 Nguyen, NQ, Maddern, GJ & Berry, DP. Endoscopic retrograde cholangiopancreatography-induced hemolytic uraemic syndrome. Aust N Z J Surg 2000; 70: 235–6. PubMed CrossRef

320 Studley, JGN, Sami, AR & Williamson, RCN. Dissemination of pancreatic carcinoma following endoscopic sphincterotomy. Endoscopy 1993; 25: 301–2. PubMed

321 Al-Jeroudi, A, Belli, AM & Shorvon, PJ. False aneurysm of the pancreaticoduodenal artery complicating theapeutic endoscopic retrograde cholangiopancreatography. Br J Radiol 2001; 74 (880): 375–7. PubMed

322 Male, R, Lehman, G, Sherman, S, Cotton, P, Hawes, R, Gottlieb, K & Rahman, S et al. Severe and fatal complications from diagnostic and therapeutic ERCPs. Gastrointest Endosc 1994; 40 (2): P29.

323 Klimczak, J & Markert, R. Retrospective analysis of post-endoscopic retrograde cholangiopancreatography mortalities. Pol Merkuriusz Lek 2003; 15 (86): 199–201. PubMed

324 Thompson, AM, Wright, DJ, Murray, W, Ritchie, GL, Burton, HD & Stonebridge, PA. Analysis of 153 deaths after upper gastrointestinal endoscopy: room for improvement? Surg Endosc 2004; 18: 22–5. PubMed CrossRef

325 Trap, R, Adamsen, S, Hart-Hansen, O & Henriksen, M. Severe and fatal complications after diagnostic and therapeutic ERCP: a prospective series of claims to insurance covering public hospitals. Endoscopy 1999; 31 (2): 125–30. PubMed CrossRef

326 Khanna, N, May, G, Cole, M, Bass, S & Romagnuolo, J. Post-ERCP radiology interpretation of cholangio-pancreatograms appears to be of limited benefit and may be inaccurate. Gastrointest Endosc 2004; 59 (5): AB186.

327 Cotton, PB. Stents for stones: short-term good, long-term uncertain. Gastrointest Endosc 1995; 42: 272–3. PubMed

328 Prat, F. The long term consequences of endoscopic sphincterotomy. Acta Gastro Enterologica Belgica 2000; LXIII: 395–6.

329 Sheth, SG & Howell, DA. What are really the true late complications of endoscopic biliary sphincterotomy? Am J Gastroenterol 2002; 97 (11): 2699–701. PubMed CrossRef

330 Bergman JJGHM, van der Mey, S, Rauws, EAJ, Tijssen, JGP, Gouma, D-J, Tytgat, GNJ & Huibregtse, K. Long-term follow-up after endoscopic sphincterotomy for bile duct stones in patients younger than 60 years of age. Gastrointest Endosc 1996; 44 (6): 643–9. PubMed

331 Prat, F, Malak, NA, Pelletier, G, Buffet, C, Fritsch, J & Choury, AD et al. Biliary symptoms and complications more than 8 years after endoscopic sphincterotomy for choledocholithiasis. Gastroenterology 1996; 110: 894–9. PubMed

332 Hawes, RH, Cotton, PB & Vallon, AG. Follow-up at 6–11 years after duodenoscopic sphincterotomy for stones in patients with prior cholecystectomy. Gastroenterology 1990; 98: 1008–12. PubMed

333 Sugiyama, M & Atomi, Y. Risk factors predictive of late complications after endoscopic sphincterotomy for bile duct stones: long-term (more than 10 years) follow-up study. Am J Gastroenterol 2002; 97 (11): 2763–7. PubMed CrossRef

334 Costamagna, G, Tringali, A, Shah, SK, Mutignani, M, Zuccala, G & Perri, V. Long-term follow-up of patients after endoscopic sphincterotomy for choledocholithiasis, and risk factors for recurrence. Endoscopy 2002; 34 (4): 273–9. PubMed

335 Pereira-Lima, JC, Jakobs, R, Winter, UH, Benz, C & Martin, WR et al. Long-term results (7–10 years) of endoscopic papillotomy for choledocholithiasis: multivariate analysis of prognostic factors for the recurrence of biliary symptoms. Gastrointest Endosc 1998; 48: 457–64. PubMed

336 Tanaka, M, Takahata, S, Konomi, H, Matsunaga, H & Yokohata, K et al. Long-term consequences of endoscopic sphincterotomy for bile duct stones. Gastrointest Endosc 1996; 44: 465–9. PubMed

337 Frimberger, E. Long-term sequelae of endoscopic papillotomy. Endoscopy 1998; 30 (9): A221–7. PubMed

338 Kullman, E, Borch, K & Liedberg, G. Long-term follow-up after endoscopic management of retained and recurrent common duct stones. Acta Chir Scand 1989; 155: 395–9. PubMed

339 Park, SH, Watkins, JL, Fogel, EL, Sherman, S, Lazzell, L, Bucksot, L & Lehman, GA. Long-term outcome of endoscopic dual pancreatobiliary sphincterotomy in patients with manometry-documented sphincter of Oddi dysfunction and normal pancreatogram. Gastrointest Endosc 2003; 57 (4): 483–91. PubMed

340 Sugiyama, M & Atomi, Y. Does endoscopic sphincterotomy cause prolonged pancreatobiliary reflux? Am J Gastroenterol 1999; 94 (3): 795–8. PubMed CrossRef

341 Cetta, F. The possible role of sphincteroplasty and surgical sphincterotomy in the pathogenesis of recurrent common duct brown stones. HPB Surg 1991; 4: 261–70. PubMed

342 Sand, J, Airo, I, Hiltunen, K-M, Mattila, J & Nordback, I. Changes in biliary bacteria after endoscopic cholangiography and sphincterotomy. Am Surg 1992; 5: 324–8.

343 Bordas, JM, Elizalde, I, Llach, J, Mondelo, F, Bataller, R & Teres, J. Biliary reflux due to sphincter of Oddi ablation: a new pathogenetic explanation for long-term major biliary symptoms after endoscopic-sphincterotomy. Endoscopy 1996; 28 (7): 642. PubMed

344 Johnston, GW. Iatrogenic chymobilia—A disease of the nineties? HPB Surg 1991; 4: 187–90. PubMed

345 Hakamada, K, Sasaki, M, Endoh, M, Itoh, T, Morita, T & Konn, M. Does sphincteroplasty predispose to bile duct cancer? Surgery 1997; 121: 488–92. PubMed CrossRef

346 Karlson, BM, Ekbom, A, Arvidsson, D, Yuen, J & Krusemo, UB. Population-based study of cancer risk and relative survival following sphincterotomy for stones in the common bile ducts. Br J Surg 1997; 84: 1235–8. PubMed CrossRef

347 Tham, TCK, Carr-Locke, DL & Collins, JSA. Endoscopic sphincterotomy in the young patient: is there cause for concern? Gut 1997; 40: 697–700. PubMed

348 Geenen, DJ, Geenen, JE, Jafri, FM, Hogan, WJ, Catalano, MF & Johnson, GK et al. The role of surveillance endoscopic retrograde cholangiopancreatography in preventing episodic cholangitis in patients with recurrent common bile duct stones. Endoscopy 1998; 30: 18–20. PubMed

349 Escourrou, J, Cordova, JA, Lazorthes, F & Frexinos, J. Early and late complications after endoscopic sphincterotomy for biliary lithiasis with and without the gallbladder 'in situ'. Gut 1984; 25: 598–602. PubMed

350 Tanaka, M, Ikeda, S & Yoshimoto, H et al. The long-term fate of the gallbladder after endoscopic sphincterotomy: complete follow-up study of 122 patients. Am J Surg 1987; 154: 505. PubMed CrossRef

351 Ingoldby, CJH, El-Saadi, J & Hall, RI et al. Late results of endoscopic sphincterotomy for bile duct stones in the elderly patient with gallbladders in situ. Gut 1989; 30: 1129. PubMed

352 Hill, J, Martin, DE & Tweedle, DEF. Risks of leaving the gallbladder in situ after endoscopic sphincterotomy for bile duct stones. Br J Surg 1991; 78: 554. PubMed

353 Boerma, D, Rauws, EA, Keulemans, YC, Janssen, IM, Bolwerk, CJ & Timmer, R et al. Wait-and-see policy or laparoscopic cholecystectomy after endoscopic sphincterotomy for bile-duct stones: a randomized trial. Lancet 2002; 360 (9335): 761–5. PubMed

354 Kaw, M, Al-Antably, Y & Kaw, P. Management of gallstone pancreatitis: cholecystectomy or ERCP and endoscopic sphincterotomy. Gastrointest Endosc 2002; 56 (1): 61–5. PubMed

355 Asbun, HJ, Rossi, RL, Heiss, FW & Shea, JA. Acute relapsing pancreatitis as a complication of papillary stenosis after endoscopic sphincterotomy. Gastroenterology 1993; 104: 1814–17. PubMed

356 Gerstenberger, PD & Plumeri, PA. Malpractice claims in gastrointestinal endoscopy: analysis of an insurance industry data base. Gastrointest Endosc 1993; 39 (2): 132–8. PubMed

357 Cotton, PB. Is your sphincterotomy really safe—and necessary? Gastrointest Endosc 1996; 44 (6): 752–5. PubMed

358 http://www.asge.org/

359 Martin, DF. Endoscopic retrograde cholangiopancreatography should no longer be used as a diagnostic test: an independent verdict. Digest Liv Dis 2002; 34: 381–2. CrossRef

360 Cotton, PB. How many times have you done this procedure, Doctor? Am J Gastroenterol 2002; 97: 522–3. PubMed

Copyright © Blackwell Publishing, 2004