Editor: Peter B. Cotton
1. ERCP overviewA 30-year perspective
Peter B. Cotton
Endoscopic cannulation of the papilla of Vater was first reported in 1968 . However, it was really put on the map shortly afterwards by several Japanese groups, working with instrument manufacturers
to develop appropriate long side-viewing instruments . The technique (initially called ECPGEndoscopic CholangioPancreatoGraphyin Japan) spread throughout Europe in the early 1970s . Early efforts were much helped by a multinational workshop at the European Congress in Paris in 1972, organized by the Olympus
company. ERCP rapidly became established worldwide as a valuable diagnostic technique, although doubts were expressed in the
USA about its feasibility and role , and the potential for serious complications soon became clear . ERCP was given a tremendous boost by the development of its therapeutic applications, notably biliary sphincterotomy in
1974 , and biliary stenting 5 years later [22,23].
It is difficult for most gastroenterologists today to imagine the diagnostic and therapeutic situation 30 years ago. There
were no scans. Biliary obstruction was diagnosed and treated surgically, with substantial operative mortality. Nonoperative
documentation of biliary pathology by ERCP was a huge step forward. Likewise, ERCP was an amazing development in pancreatic
investigation at a time when the only available test was laparotomy. The ability to 'see into' the pancreas, and to collect pure pancreatic juice  seemed like a miracle. We assumed that ERCP would have a dramatic impact on chronic pancreatitis and pancreatic cancer. Sadly,
these expectations are not yet realized, but endoscopic management of biliary obstruction was clearly a major clinical advance,
especially in the sick and elderly. The period of 15 or so years from the mid 1970s really constituted a 'golden age' for ERCP. Despite significant risks , it was quite obvious to everyone that ERCP management of duct stones and tumours was easier, cheaper, and safer than available
surgical alternatives. Large series were published, including some randomized trials . Percutaneous transhepatic cholangiography (PTC) and its drainage applications were also developed during this time, but
were used (with the exception of a few units) only when ERCP failed or was not available. The 'combined procedure'endoscopic cannulation over a guidewire placed at PTC [32,33]became popular for a while, but was needed less as both endoscopic and interventional techniques improved.
The changing world of pancreaticbiliary medicine
The situation has changed in many ways during the last two decades. ERCP has evolved significantly, but so have many other
The impact of scanning radiology
Imaging modalities for the biliary tree and pancreas have proliferated. High quality ultrasound, computed tomography, endoscopic
ultrasonography, and MR scanning (with MRCP) have greatly facilitated the non-invasive evaluation of patients with known and
suspected biliary and pancreatic disease. The proportion of ERCP examinations now done purely for diagnosis has diminished
Extending the indications for therapeutic ERCP
The second major change has been the attempt of ERCP practitioners to extend their therapeutic territory into more complex
areas like pancreatitis and suspected sphincter of Oddi dysfunction. The value of ERCP in these contexts remains controversial.
Improvements in surgery
The third major change is the substantial and progressive reduction in risk associated with conventional surgery (due to excellent
perioperative and anaesthesia care) and the increasing use of less invasive laparoscopic techniques . It is no longer correct to assume that ERCP is always safer than surgery. Sadly, bad complications of ERCP (especially pancreatitis
and perforation) continue to occur, especially during speculative procedures done by inexperienced practitioners, often using
the needle knife for lack of standard expertise . These facts are forcing the ERCP community to search for ways to reduce the risks. Important examples of this preoccupation
are the focus on refining indications , prospective studies of predictors of adverse outcomes , and attempts to remove stones from the bile duct without sphincterotomy, at least in younger patients with relatively small
stones and normal sized ducts, which we see much more often in the era of laparoscopic cholecystectomy . Equally important is the increasing focus on who should be trained, and to what level of expertise. How many ERCPists are
The fourth major development in this field is the increased participation of patients in decisions about their care. Patients
are rightly demanding the data on the potential benefits, risks, and limitations of ERCP, and the same data about the alternatives
The focus in the early twenty-first century is on careful evaluation of what ERCP can offer (in comparison with available
alternatives), and on attempts to improve the overall quality of ERCP practice. These issues are important in all clinical
contexts, but come into clearest focus where ERCP is still considered somewhat speculative, e.g. in the management of chronic
pancreatitis and of possible sphincter of Oddi dysfunction [36,40].
Benefits and risks
Evaluation of ERCP is a complex topic . Its role is very much dependent on the clinical context [Fig. 1], and colleagues contributing to this resource provide guidance about the current state of practice in their main topic areas.
This discussion focuses on the general difficulties in defining the role and value of ERCP . Figure 2 attempts to illustrate all of the elements of the 'intervention equation'. There is much talk about 'outcomes studies', but 'outcomes' cannot be assessed without detailed knowledge of the precise 'incomes'. Thus, a patient with certain demographics, disease type, size, and severity causing a specific level of symptoms, disability,
and life disruption is offered an ERCP intervention by a certain individual with a particular experience and skill level,
with certain expected, planned, burdens (i.e. pain, distress, disruption, and costs). All of these metrics need clear and
agreed definitions if we are to make any sense of the evaluation . The conjunction of the patient and that intervention result in the 'outcomes'[Fig. 2]. Ultimately, we are most interested in the clinical outcome (reduced burden of symptoms and disease), but there are many
factors along the way, including the technical results (influenced by the 'degree of difficulty') and the occurrence of unplanned events (or complications), which add to the actual burden.
Degree of difficulty and expertise
Some ERCP procedures are more technically challenging than others. Most can be predicted beforehand (e.g. known Billroth 2
resection, hilar tumour). A five-level scoring system for degree of difficulty was developed  and later simplified to three grades (Fig. 3). Grade 1 procedures are those (mainly biliary) interventions which anyone offering ERCP should be able to achieve to a reasonable
level of expertise. Grade 3 procedures are the most difficult, and are performed mainly in tertiary referral centres.
It is clear that some ERCP practitioners are more skilled than others. Some, like most of those just emerging from standard
training programs, will be comfortable only with Grade 1 cases, where they should be technically successful in about 90 per
cent of attempts. Endoscopists with more training (e.g. a dedicated 4th year in the USA), and those who have honed their skills
in practice with the aid of community and academic colleagues, will attempt more complex cases. So-called experts, working
in referral centres, will tackle all comers, but will also have very high success rates in the easier cases. These concepts
of case difficulty and individual expertise can usefully be combined [Fig. 4]. A key and sensitive issue is whether and how widely these variations in expertise should be advertized.
We are becoming accustomed to seeing 'league tables' of hospitals according to outcomes of major procedures, such as cardiovascular surgery and pancreatico-duodenectomy. Increasingly
there is interest in drilling down further to the individual practitioner. The American Society for Gastrointestinal Endoscopy
has recommended the use of 'report cards', i.e. summaries of the practice of individual endoscopists . An example of a report card for one ERCP expert is shown in Fig. 5. Report cards are unlikely to become mandatory. What is the motivation for less expert endoscopists to provide data? Patients
are increasingly advised to ask their potential interventionists about their experience. Some patients will certainly hesitate
if their practitioners are not able to provide benchmark data. Well-trained and skilful practitioners should wear their data
as badges of quality .
The word 'complication' is emotive, raising issues of medical error and legal liability. We prefer to discuss 'unplanned events', since they are best described simply as deviations from the plan which had been agreed with the patient . The phrase 'adverse events' has been used also, but not all unplanned events are negative. A patient with suspected cancer may be delighted to wake up
from a procedure with an unexpected cure (sphincterotomy and stone removal). All unplanned events should be documented in
a standard format , as an aid to efforts at quality improvement. Some events relatively trivial, such as transient hypotension or self-limited
bleeding. At what level of severity do the events become 'complications'? An influential consensus conference  set the threshold at the need for hospital admission and defined levels of severity by the length of stay, as well as the
need for surgery or intensive care [36,44]. Details of complications, their avoidance and management, are addressed in detail elsewhere in this Advanced Endoscopy e-book
Clinical success and value
Clinical success may sometimes be relatively obvious, e.g. removal of a stone, or relief of jaundice with a stent. However,
in many cases (e.g. chronic pancreatitis, sphincter dysfunction), the judgement can be made only after long periods of follow-up.
This greatly complicates evaluation studies in just the clinical circumstances where the knowledge is needed most. Patient
satisfaction is another important parameter. It is determined partly by the clinical results (and how that compares with the
patient's expectation), but also by the patient's perception of the process (accessibility, courtesy, etc.). The cost (burden)
of the intervention is obviously a key consideration. This consists of the planned burden, plus the result of any unplanned
events. The ratio between the clinical impact (benefit) and the burden (cost) determines the 'value' of the procedure in that individual patient . Attempts to provide definitions for all of these metrics are advancing slowly. Their incorporation in endoscopy reporting
databases will allow on-going useful outcomes evaluations to guide further decisions. If the same or similar metrics are used
also by those performing alternative interventions such as surgery, we will get a clearer idea of the relative roles of these
different procedures . In some cases randomization will be necessary to make a final judgement. However, the issue of 'operator dependence' will always exist. A randomized trial of two techniques performed by experts may not be the best guide to the choice of intervention
in everyday community practice.
The trends which we have outlined are likely to continue and to accelerate in the coming years. Quality is the big issue.
That means making sure that we are doing the right things, and doing them right. It has been clear for a long time (but only
now becoming generally accepted) that ERCP is a procedure that should be undertaken only by a minority of gastroenterologists.
The amount of training and continuing dedication in practice needed to attain and maintain high levels of competence, and
to improve, means that the procedures should be focused in relatively few hands. The increasing variety and safety of alternative
procedures, and the vigilance of our customers, will drive that agenda. The other imperative is to pursue the research studies
necessary to improve current methods and to evaluate all of them rigorously. This is best done in collaboration with colleagues
in surgery and radiology to establish the best methods for approaching patients with known or suspected biliary and pancreatic
disease. The dynamics between specialists will change with time, which is one excellent reason for organizing care to be patient-focused,
rather than in traditional technical silos. Multidisciplinary organizations, like our Digestive Disease Center at the Medical
University of South Carolina, attempt to provide that perspective and a platform for the unbiased research and education aimed
at improving the quality of service .
1 McCune, WS, Shorb, PE & Moscovitz, H. Endoscopic cannulation of the ampulla of Vater: a preliminary report. Ann Surg 1968; 167: 7526. PubMed
2 Oi, I, Takemoto, T & Kondo, T. Fiberduodenoscope. Direct observations of the papilla of Vater. Endoscopy 1969; 1: 1013.
3 Ogoshi, K, Tobita, Y & Hara, Y. Endoscopic observation of the duodenum and pancreatocholedochography using duodenofiberscope under direct vision. Gastrointest Endosc 1970; 12: 8396.
4 Takagi, K, Ideda, S, Nakagawa, Y, Sakaguchi, N, Takahashi, T & Kumakura, K et al. Retrograde pancreatography and cholangiography by fiber-duodenoscope. Gastroenterology 1970; 59: 44552. PubMed
5 Kasugai, T, Kuno, N, Aoki, I, Kizu, M & Kobayashi, S. Fiberduodenoscopy. Analysis of 353 examinations. Gastrointest Endosc 1971; 18: 916. PubMed
6 Classen, M, Koch, H, Fruhmorgen, P, Grabner, W & Demling, L. Results of retrograde pancreaticography. Acta Gastroenterologica Japonica 1972; 7: 1316.
7 Cotton, PB. Progress Report. Cannulation of the papilla of Vater by endoscopy and retrograde cholangiopancreatography (ERCP). Gut 1972; 13: 101425. PubMed
8 Cotton, PB, Salmon, PR, Blumgart, LH, Burwood, RJ, Davies, GT & Lawrie, BW et al. Cannulation of papilla of Vater via fiber-duodenoscope. Assessment of retrograde cholangiopancreatography in 60 patients. Lancet 1972; 1: 538. PubMed
9 Gulbis, A, Cremer, M & Engelholm, L. La cholangiographie et la wirsungographic endoscopiques. Acia Endoscopica Radiocinematogr 1972; 2: 7880.
10 Heully, F, Gaucher, P, Laurent, J, Vicari, F, Fays, J, Bigard, M-A & Jenpierre, R. 1972La duodenoscopie et la catheterisme de voies biliares et pancreatiques, Vol 1, pp. 163-8. La Nouvelle Presse Med.
11 Safrany, L, Tari, J, Barna, L & Torok, I. Endoscopic retrograde cholangiography. Experience of 168 examinations. Gastrointestinal Endoscopy 1973; 19: 1638. PubMed
12 Liguory, C, Gouero, H, Chavy, A, Coffin, JC & Huguier, M. Endoscopic retrograde cholangiopancreatography. Br J Surg 1974; 61: 35962. PubMed
13 Cotton, PB. ERCP. Gut 1977; 18: 31641. PubMed
14 Morrissey, JF. To cannulate or not to cannulate (editorial). Gastroenterology 1972; 63: 3512. PubMed
15 Blackwood, WD, Vennes, JA & Silvis, SE. Post-endoscopy pancreatitis and hyperamylasuria. Gastrointest Endosc 1973; 20: 568. PubMed
16 Classen, M & Demling, L. Hazards of endoscopic retrograde cholangio-pancreaticography (ERCP). Acta Hepato-Gastroenterologica 1975; 22: 13. PubMed
17 Nebel, OT, Silvis, SE, Rogers, G, Sugawa, C & Mandelstam, P. Complications associated with endoscopic retrograde cholangio-pancreatography. Results of the ASGE Survey. Gastrointest Endosc 1975; 22: 346. PubMed
18 Bilbao, MK, Dotter, CT, Lee, TG & Katon, RM. Complications of endoscopic retrograde cholangiopancreatography (ERCP). A study of 10,000 cases. Gastroenterology 1976; 70: 31420. PubMed
19 Classen, M & Demling, L. Endoskopische sphinkterotomie der papilla Vateri und steinextraktion aus dem ductus choledochus. Deutsche Med Wochenschrift 1974; 99: 4967.
20 Kawai, K, Akasaka, Y, Murakami, K, Tada, M, Kohill, Y & Nakajima, M. Endoscopic sphincterotomy of the ampulla of Vater. Gastrointest Endosc 1974; 20: 14851. PubMed
21 Cotton, PB, Chapman, M, Whiteside, CG & LeQuesne, LP. Duodenoscopic papillotomy and gallstone removal. Br J Surg 1976; 63: 70914. PubMed
22 Soehendra, N & Reijnders-Frederix, V. Palliative bile duct drainage. A new endoscopic method of introducing a transpapillary drain. Endoscopy 1980; 12: 811. PubMed
23 Laurence, BH & Cotton, PB. Decompression of malignant biliary obstruction by duodenoscope intubation of the bile duct. Br Med J 1980; 280 (6213) 5223. PubMed
24 Robberrecht, P, Cremer, M, Vandermers, A, Vandermers-Piret, M-C, Cotton, PB & de Neef, P et al. Pancreatic secretion of total protein and three hydrolases collected in healthy subjects via duodenoscopic cannulation. Effects
of secretin, pancreozymin and caerulein. Gastroenterology 1975; 69: 3749. PubMed
25 Byrne, P, Leung, JWC & Cotton, PB. Retroperitoneal perforation during duodenoscopic sphincterotomy. Radiology 1984; 150: 3834. PubMed
26 Vaira, D, Ainley, C, Williams, S, Caines, S, Salmon, P & Russell, C et al. Endoscopic sphincterotomy in 1000 consecutive patients. Lancet 1989; 2: 4314. PubMed
27 Cotton, PB. Endoscopic management of bile duct stones (apples and oranges). Gut 1984; 25: 58797. PubMed
28 Leung, JWC, Emery, R, Cotton, PB, Russell, RCG, Vallon, AG & Mason, RR. Management of malignant obstructive jaundice at The Middlesex Hospital. Br J Surg 1983; 70: 5846. PubMed
29 Cotton, PB. Endoscopic methods for relief of malignant obstructive jaundice. World J Surg 1984; 8: 85461. PubMed
30 Speer, AG, Cotton, PB, Russell, RCG, Mason, RR, Hatfield, ARW & Leung, JWC et al. Randomized trial of endoscopic versus percutaneous stent insertion in malignant obstructive jaundice. Lancet 1987; 2: 5762. PubMed
31 Smith, AC, Dowsett, JF, Russell, RCG, Hatfield, ARW & Cotton, PB. Randomised trial of endoscopic stenting versus surgical bypass in malignant low bile duct obstruction. Lancet 1994; 344: 165560. PubMed
32 Shorvon, PJ, Cotton, PB, Mason, RR, Siegel, HJ & Hatfield, ARW. Percutaneous transhepatic assistance for duodenoscopic sphincterotomy. Gut 1985; 26: 13736. PubMed
33 Dowsett, JF, Vaira, D, Hatfield, AR, Cairns, SR, Polydorou, A & Frost, R et al. Endoscopic biliary therapy using the combined percutaneous and endoscopic technique. Gastroenterology 1989; 96: 11806. PubMed
34 Cotton, PB, Chung, SC, Davis, WZ, Gibson, RM, Ransohoff, DF & Strasberg, SM. Issues in cholecystectomy and management of duct stones. Am J Gastroenterol 1994; 89: S16976. PubMed
35 Cotton, PB. ERCP is most dangerous for people who need it least. Gastrointest Endosc, 2001; 54(4) 5356. PubMed
36 Cotton, PB. Income and outcome metrics for objective evaluation of ERCP and alternative methods. Gastrointest Endosc 2002; 56(6 suppl):S28390. PubMed
37 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: 42534. PubMed
38 Huibregtse, K. Endoscopic balloon dilation for removal of bile duct stones: special indications only. Endoscopy 2001; 33 (7): 6202. PubMed
39 Cotton, PB. How many times have you done this procedure, Doctor? Am J Gastroenterol 2002; 97: 5223. PubMed
40 Lehman, GA & Sherman, S. Sphincter of Oddi dysfunction. Int J Pancreatol 1996; 20: 1125. PubMed
41 Cotton, PB. Therapeutic gastrointestinal endoscopy. Problems in proving efficacy. N Engl J Med 1992; 326: 16268. PubMed
42 Schutz, SM & Abbott, RM. Grading ERCPs by degree of difficulty: a new concept to produce more meaningful outcome data. Gastrointest Endosc 2000; 51: 5359. PubMed
43 American Society for Gastrointestinal Endoscopy. Quality and outcome assessment in gastrointestinal endoscopy. Gastrointest Endosc 2000; 52(6): 82730.
44 Cotton, PB, Lehman, G, Vennes, J, Geenen, JE, Russell, RCG & Meyers, WC et al. Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 1991; 37: 38393. PubMed
45 Cotton, PB & Mauldin, P. (2002) Endoscopic research, outcomes and justification . In: Gastrointestinal Endoscopy. (eds. Classen, M, Tytgat, GNJ, Lightdale, C), Thieme.
46 Cotton, PB. Randomization is not the (only) answer: a plea for structured objective evaluation of endoscopic therapy. Endoscopy 2000; 32: 4025. PubMed
47 Cotton, PB. Fading boundary between gastroenterology and surgery. J Gastroenterol Hepatol 2000; 15: G347. PubMed
Copyright © Blackwell Publishing, 2003