ERCP (Ensuring Really Competent Practitioners)
30 October 2014 (online)
Endoscopic retrograde cholangiopancreatography (ERCP) can provide great clinical benefits, but can also cause serious complications. We should all be interested in maximizing the benefits and minimizing the risks . In this issue of Endoscopy, Ekkelenkamp and colleagues from the Rotterdam group extend their earlier work  on measuring the performance of their trainees , adding to the literature showing that the procedure is difficult to learn. Using a basic quality metric (deep biliary cannulation with a virgin papilla), they showed that their trainees achieved only around 60 % success after 150 procedures. Based largely on the earlier seminal study by Jowell and colleagues , the American Society for Gastrointestinal Endoscopy (ASGE) and other societies have suggested that 80 % – 85 % should be the target, but it is still far from clear how many procedures are required to achieve this degree of proficiency. The Rotterdam data on 15 trainees shows substantial variability. The average cannulation rate was 85 % after 200 ERCPs, but the 95 % confidence interval spanned 70 % and 90 %. Some individuals may be ready for “prime time” sooner than others, but to assure competency in nearly all trainees the volume threshold may have to exceed 300. It is worth remembering that the number of cases required will depend also on the quality of the teaching, and how much a trainee has to be involved in a procedure before he or she can count it as a “case.”
It is also important to understand what the figures include (and exclude). Not counting cases with prior sphincter intervention is obvious, but should we also exclude patients with surgical biliary diversion (an increasingly common challenge), and maybe some with disease in and around the papilla, as highlighted in a recent abstract ? And, when we get to other criteria, such as success in stone removal, do we count all-comers, or only those with successful cannulation – a point we made just 30 years ago ?
It gives us no pleasure to state that the situation in the USA is currently unsatisfactory, which is not a new observation. Only 3 years ago we published data showing that 77 % of trainees had been involved (but the extent of participation was not recorded) in fewer than 180 cases , and made several suggestions for improving the situation , which seem to have been ignored. The number of specialized “4th year” advanced training positions (focusing on ERCP and EUS) has increased in USA in recent years , the quality and effects of which are unknown. Nevertheless, many trainees are still getting limited hands-on training in the standard 3-year fellowship program, and take marginal skills into practice .
Concern about the quality of ERCP has been raised in publications from many other countries, including Austria , Spain , Finland , Singapore , Norway , Israel , and Australia . No doubt there is much attention also in other countries, and we encourage readers to add data and comments in letters to the editor of this esteemed journal.
The key question is whether these well-rehearsed problems are being addressed. The paper by Ekkelenkamp and colleagues  indicates a welcome shift in the Netherlands, with a planned reduction in the number of ERCP trainees, and there is evidence for considerable progress in at least two other countries. Earlier proposals for changes in the structure of training in practice in Britain   did not gain momentum until a survey by the British Society of Gastroenterology (BSG) showed rather poor results of ERCP in practice . The various stakeholders then formed the Joint Advisory Group (JAG) , which has been driving the quality agenda for ERCP . Training places for ERCP are strictly limited and the progress of all trainees is documented with an e-Portfolio in the JETS (JAG Endoscopy Training System) . These data, along with DOPS (direct observation of procedural skills) have provided the basis for certification (mirroring the “driving test” for colonoscopy). These important initiatives are being updated, as proposed recently by a multidisciplinary working party, in “ERCP – The Way Forward, A Standards Framework” .
The situation in Australia is similar, in that endoscopy training and practice are overseen by the multidisciplinary Conjoint Committee . This group has made many useful changes, and promulgated a very strict test of trainee competence, demanding completion of no fewer than 200 procedures without assistance.
Measuring performance in training is important, but what we really need to know is how neophyte ERCPists perform when they enter and maintain an independent practice, away from the mother ship. Their results may be different when working under new pressures in unfamiliar surroundings, with less experienced assistants and different equipment, and without an expert nearby to advise or take over when needed. The British working party recognize this risk by recommending that newly appointed specialists are mentored for their first 100 cases in practice , with the assumption that learning continues long into practice.
That point raises the contentious issue about how many cases it is necessary to do each year to maintain skills and hopefully to improve them . We have little relevant data. Most experts would probably suggest at least 50 cases/year, but that number must depend on the prior volume and expertise. We do know that ERCPists doing fewer than 50 cases (or sphincterotomies) each year have worse results than those who do more  , but those doing fewer cases do not share their data. The Joint Advisory Group in Britain has previously recommended a minimum of 75 cases/year, and now wants to raise this to 100 . That number would exclude most of the current ERCPists in USA, where it has been reported that 40 % do fewer than 50/year . Another recent study in USA suggests that outcomes continue to improve as annual volumes increase to 115 cases/year, before plateauing above this level .
Since ERCP is unique amongst common endoscopy procedures in that it is done only in hospitals, those authorities have the responsibility to ensure that ERCP is done only by people who are competent and can prove it. Are those institutions doing so?
One of the reasons that many ERCPists in USA continue to do small volumes is the need to have an on-call roster for emergency cases. It is tough for the chosen ones if only two of the group do ERCP. The obvious answer is for groups to collaborate and for hospitals to mandate this .
In addition to the skills of the practitioners, it has become clear that the volume of cases in a center is also important, to ensure that there are adequate facilities, equipment, and trained staff. There is evidence from surgery that low volume surgeons perform better in high volume centers . JAG recommends a minimum of 200 cases/center/year , whereas a study in USA showed that half of the hospitals were doing fewer than 49/year, and 5 % were doing more than 200 .
All of these proposed guideline numbers for ERCPists and hospitals are well-meaning but arbitrary. Clearly the real answers lie in the actual outcomes that are being achieved. A few motivated (and obviously competent) individuals and groups have reported their post-training performance . Capturing data from multiple endoscopists is needed to allow benchmarking. However, systems such as the ERCP Quality Network  are voluntary, and will never capture data from those nervous about providing it. The only logical (if contentious) way forward is to require quality reporting as an essential part of permission to enter and to continue practice  .
We could go further and suggest that it is time for all countries, including USA, to follow the lead of Britain and Australia and to initiate certification for advanced procedures such as ERCP. We need to take (and pass) tests to do other dangerous things such as driving a car or a plane. How is endoscopy different? The risks of an accident are certainly higher.
Climbing down from that podium, it may be helpful to consider why ERCP is so difficult technically, and whether ERCP (and training for it) could be made easier. Getting deep into the bile duct is the key task in most cases. The channel is somewhat tight and tortuous, and probably constricts with edema with each unsuccessful poke. Some claim quicker success with a guidewire, and others espouse early precut, which seems acceptable in the hands of experts (who need it less) but is dangerous when used by the inexperienced for speculative indications. Have we exhausted medical methods for facilitating cannulation? Relaxing agents such as nitrates, sildenafil, cholecystokinin, and secretin appear to be somewhat helpful, but have not become popular. Since we have largely abandoned deep venous access without ultrasound visualization, perhaps we should push for image-guided techniques instead of the current “probe and feel” or “dunk and squirt” approaches.
Would a small ultrasound detector in the tip of the duodenoscope be helpful? Maybe we could learn something from the Ascaris worm, that seems to cannulate the bile duct easily and often in certain countries.
Whatever the approach, cannulation of a tiny and collapsed orifice through a flexible endoscope will remain difficult to learn and always have some potential for catastrophe. Expert trainers should be embraced and elevated on a national scale, as not all of the current mentors enjoy teaching – and few have been trained to do so. Again, Britain is a shining exception, since all those teaching endoscopy must attend a “train the trainers” course. Additionally, simulation tools obviously have the potential to assist the learning process (and possibly to weed out the no-hopers), but they are little used, largely because the focus has been on sophisticated electronic devices that are expensive. Simple math might show that keeping trainees away from patients until they have achieved some skills on simulators might actually be cost-effective in speeding the cases, but there is no mechanism for sharing/swapping the resources. There is a strong case for exploring simpler and cheaper mechanical training aids  . Since maintaining high volumes of ERCP is more challenging than for other endoscopic procedures, simulators may be helpful for moderate volume providers to maintain competency. However, it is unrealistic to think that an endoscopist performing one ERCP/week or less could ever achieve outcomes, with regard to technical success and complication rates, comparable to those who perform several ERCPs on most days. This cannot be corrected by choosing only “low risk” patients; any ERCP performed for any indication can become a high risk or high cost intervention if the procedure is not executed well.
Training methods may improve in the future, and new techniques may be described, but the bottom line will remain unchanged. It has been brutally obvious for a long time, at least in USA, that too many people are being trained (or half-trained) and that too many are not doing enough cases to maintain or enhance their skills. ERCP used to be a fun (mainly diagnostic) procedure for lots of gastroenterologists. With the widespread availability of magnetic resonance cholangiopancreatography (MRCP) and endoscopic ultrasound (EUS), it should now be restricted to a small number of well-trained endoscopists with good volumes, who can document good outcomes in standard cases and explore the more complex ones with integrity. This is scarcely a unique observation: it is a broken record that has been spinning for decades . Let’s get things back on track.
It is time for all stakeholders (patients, trainees, trainers, payers, professional societies, and hospitals) to address the issue. Let’s follow the lead of colleagues in Britain and Australia, who have attempted to enhance the quality of ERCP in their countries by bringing all interested parties to the table. This cannot be done by individual professional societies, no matter how well-intentioned.
We owe it to our patients. Someone in your family, or you, may need an ERCP one day.
- 1 Cotton PB. Endoscopic retrograde cholangiopancreatography; maximizing benefits and minimizing risks. Gastrointest Endosc Clin N Am 2012; 22: 587-599
- 2 Ekkelenkamp VE, Koch AD, Haringsma J et al. Quality evaluation through self-assessment: a novel method to gain insight into ERCP performance. Frontline Gastroenterol 2014; 5: 10-16
- 3 Ekkelenkamp VE, Koch AD, Rauws EAJ et al. Competence development in ERCP: the learning curve of novice trainees. Endoscopy 2014; 46 Epub 2014 Sep 10
- 4 Jowell PS, Baillie J, Branch MS et al. Quantitative assessment of procedural competence. A prospective study of training in endoscopic retrograde cholangiopancreatography. Ann Intern Med 1996; 125: 983-989
- 5 Sheppard D, Craddock S, Warner B et al. ERCP cannulation success benchmarking: implications for certification and validation. Gut 2014; 63: A210-A211
- 6 Cotton PB. Endoscopic management of bile duct stones (apples and oranges). Gut 1984; 25: 587-97
- 7 Cote GA, Keswani RN, Jackson T et al. Individual and practice differences among physicians who perform ERCP at varying frequency: a national survey. Gastrointest Endosc 2011; 74: 65-73
- 8 Cotton PB. Are low-volume ERCPists a problem in the United States? A plea to examine and improve ERCP practice – now. Gastrointest Endosc 2011; 74: 161-166
- 9 Rosenthal LS. Is a fourth year of training necessary to become competent in EUS and ERCP? Notes from the 2008 class of advanced endoscopy fellows. Gastrointest Endosc 2008; 68: 1150-1152
- 10 Kowalski T, Kanchana T, Pungpagong S. Perceptions of gastroenterology fellows regarding ERCP competency and training. Gastrointest Endosc 2003; 58: 345-349
- 11 Kapral C, Duller C, Wewalka F et al. Case volume and outcome of endoscopic retrograde cholangiopancreatography: results of a nationwide Austrian benchmarking project. Endoscopy 2008; 40: 625-630
- 12 Riesco-Lopez JM, Vazquez-Romero M, Rizo-Pascual JM et al. Efficacy and safety of ERCP in a low-volume hospital. Rev Esp Enferm Dig 2013; 105: 68-73
- 13 Siiki A, Tamminen A, Tomminen T et al. ERCP procedures in a Finnish community hospital: a retrospective analysis of 1207 cases. Scand J Surg 2012; 101: 45-50
- 14 Ang TL, Cheng J, Khor JL et al. Guideline on training and credentialing in ERCP. Singapore Med J 2011; 52: 654-657
- 15 Glomsaker T, Soreide K, Aabakken L et al. A national audit of temporal trends in ERCP in Norway. Scand J Gastroenterol 2011; 46: 116-121
- 16 Suissa A, Yassin K, Lavy A et al. Outcome and early complications of ERCP; a prospective single center study. Hepatogastroenterology 2005; 52: 352-325
- 17 Dundee PE, Chin-Lenn L, Syme DB et al. Outcomes of ERCP; prospective series from a rural centre. ANZ J Surg 2007; 77: 1013-1017
- 18 Wicks AC, Robertson GS, Veitch PS. Structured training and assessment in ERCP has become essential in the Calman era. Gut 1999; 45: 154-156
- 19 Murray A, Pathmakanthan S, Heeley R et al. The future of ERCP training in the UK. Gut 2001; 49: 739-740
- 20 Williams EJ, Taylor S, Fairclough P et al. Are we meeting the standards set for endoscopy? Results of a large-scale prospective survey of endoscopic retrograde cholangio-pancreatograph practice. Gut 2007; 56: 821-829 Epub 2006 Dec 4
- 21 JAG Joint Advisory Group on GI Endoscopy. www.thejag.org.uk [Accessed: June 29 2014]
- 22 Isaacs P. ERCP training in the United Kingdom: a critical review. World J Gastrointest Endosc 2011; 3: 30-33
- 23 JETS JAG Endoscopy Training System. www.jets.nhs.uk [Accessed: July 2 2014]
- 24 British Society of Gastroenterology. ERCP – the way forward A standards framework. Available at: http://www.bsg.org.uk/images/stories/docs/clinical/guidance/bsge_ercp_statement_14.pdf [Accessed: June 29 2014]
- 25 Conjoint Committee for Recognition of Training in Gastrointestinal Endoscopy. www.conjoint.org.au [Accessed: June 29 2014]
- 26 Guda NM, Freeman ML. Are you safe for your patients – how many ERCPS should you be doing?. Endoscopy 2008; 40: 675-676
- 27 Coté GA, Imler TD, Xu H et al. Lower provider volume is associated with higher failure rates for endoscopic retrograde cholangiopancreatography. Med Care 2013; 51: 1040-1047
- 28 Birkmeyer JD, Stukel TA, Siewers AE et al. Surgeon volume and operative mortality in the United States. N Engl J Med 2003; 349: 2117-2127
- 29 Varadarajulu S, Kilgore ML, Wilcox CM et al. Relationship among hospital ERCP volume, length of stay, and technical outcomes. Gastrointest Endosc 2006; 64: 338-347
- 30 Colton JB, Curran CC. Quality indicators, including complications, of ERCP in a community setting: a prospective study. Gastrointest Endosc 2009; 70: 457-467
- 31 Enochsson L, Swahn F, Arnelo U et al. Nationwide, population-based data from 11,074 ERCP procedures from the Swedish Registry for Gallstone Surgery and ERCP. Gastrointest Endosc 2010; 72: 1175-1184
- 32 Cotton PB, Romagnuolo J, Faigel DO et al. The ERCP quality network: a pilot study of benchmarking practice and performance. Am J Med Qual 2013; 28: 256-260
- 33 Cotton PB. How many times have you done this procedure, doctor?. Am J Gastroenterol 2002; 97: 522-523
- 34 Laio WC, Leung JW, Wang HP et al. Coached practice using ERCP mechanical simulator improves trainees’ ERCP performance: a randomized controlled trial. Endoscopy 2013; 45: 799-805
- 35 Schneider AR, Schepp W. Do it yourself: building an ERCP training system within 30 minutes. Gastrointest Endosc 2014; 79: 828-832
- 36 Sivak MV. Trained in ERCP. Gastrointest Endosc 2003; 58: 412-414