Editorial

 

Towards safe ERCP: selection, experience and prophylaxis

 

Since its conception (1), endoscopic retrograde cholangio-pancreatography (ERCP) has been notorious due to the presence of complications. All steps in this technique entail a rate of undesired outcomes above that of the remaining endoscopy types (sedation, diagnosis, and therapeutic procedures), and every individual action is associated with a host of specific adverse effects, which seldom coincide with those in other steps and are often linked to individual patient characteristics. Even though ERCP represents a serious mechanical, chemical, hydrostatic, enzymatic, microbiological, and thermal insult in itself for the papilla, it is considered a relatively safe endoscopic procedure, once its potential to induce serious, life-threatening complications is assumed (2).

In addition to training, ERCP requires a lot of cognitive effort on the performer's part: clinical judgement and decision making are essential items to confront the successive crossroads one is daily brought at by this technique. Factors such as the explorer's personality, restfulness, pressure level, and stress influence results obtained, as does the total amount of procedures performed (lifetime accumulation), current training level (daily practice), or simply time elapsed since the last similar case (3).

On the other hand, ERCP is marked by an evaluation that is almost exclusively based on its complication rate, and many other axes exist that may influence a technique's results, such as the extent of technical success, extent and duration of beneficial effects, patient satisfaction, and cost.

ERCP has evolved along the same lines as the rest of diagnostic and therapeutic modalities, mainly cholangioresonance and laparoscopic cholecystectomy, and its appropriate use has thus constantly changed. ERCP equipment and tools have also changed and improved, and retrospective studies face a mixture of cases, which hinders the assessment of risk factors for complication development.

Such risk factors are specific for each complication and seldom overlap. This is in addition true for both patient- and explorer-dependent factors, as well as for post-ERCP care (4).

A huge number of factors has been involved in the development of complications. Risk factors common to all complications include precut, combined or "rendezvous" access, difficult cannulation, and failed stone removal. All these are surrogate markers of simply a difficult ERCP and of the way in which each endoscopist or center faces it (4).

Pancreatitis is the most common complication following ERCP. It is defined as the development or worsening of abdominal pain, with amylasemia rising above 3 times its normal value, within 24 hours after the procedure and requiring at least a 2-day stay in hospital, and represents a true scourge on a daily basis.

Appropriate patient selection is the most effective method to reduce the rate of unnecessary pancreatitis, as García-Cano et al. suggest in this issue (5), and patients unlikely to undergo a therapeutic procedure shoud be referred to other diagnostic modalities. Both cholangioresonance and echoendoscopy are free of the risk for pancreatitis, have similar sensitivities for the detection of choledocholithiasis, and may identify patients who will benefit from surgery without the need to have recourse to ERCP. As García-Cano states, the greatest rate of complications is that of patients where ERCPderived benefits are less likely, and therefore the most effective way of reducing post- ERCP pancreatitis rates is by "simply" avoiding unnecessary ERCPs.

Now, how should this be done? ¿How many amongst us have been forced by environmental pressures such as patients or their doctors (those same subjects who will later complain on your complication rates) to perform an ERCP in a younger woman with recurrent abdominal pain, non-dilated main bile duct, normal bilirubin, and suspected Oddi's sphincter dysfunction (OSD). My slate is not clean, and I will not throw any stones, not even against those who facing a difficult access in our hypothetical patient, and after repeated cannulation attempts where only Wirsung's duct is reached, went on to consider precut sphincterotomy.

At least, this fleeing-forward situation allows us to select patients for whom a prophylaxis of post-ERCP pancreatitis will be cost-effective. Unfortunately, most prophylactic measures must be implemented before ERCP is initiated. The placement of a pancreatic stent in patients with suspected OSD may be effective (6,7), as is the use of 24-hour somatostatin or gablexate (the latter is not available in our country) perfusion (8). Nitroglycerin (9) or heparin (10) may also be tried, whereas the use of non-ionic contrast media, corticosteroids, allopurinol, and nifedipine have been excluded from prophylaxis. When interleukin 10 eventually becomes available, its use to manipulate the proinflammatory cytokine cascade may represent a new contribution, even though results from preliminary controlled studies are dispar-ate (it was only effective when used at higher doses –30 mg/kg– 30 minutes prior to the procedure); nevertheless, IL-10 may be effective immediately after ERCP as a result of its mechanism of action, once all risk factors have been detected. Excellent results have been reported of late with the administration of a single diclofenac suppository immediately after ERCP (12). This prophylaxis system meets all of the criteria to be considered ideal, as it is cheap and effective, and may be used in all centers following ERCP and risk factor identification. However, one should be not very hopeful, since the history of post-ERCP pancreatitis prophylaxis is full of great promises that end up in disappointment.

Technical variables are not negligible either, and the use of pure-cutting monopolar current seems beneficial. In contrast, Oddi's sphincter manometry increases risk if a continuous aspiration catheter is not used to prevent excessive pressure in Wirsung's duct. However, the most controversial of all technical factors is the use of precut. Some consider it a significant risk factor, whereas other authors believe it unrelated to increased complications. A solution may lie in experience, the way this accessory is used, and its timing: so many cannulation attempts have failed –and how– prior to precut! In expert hands, precut is probably an acceptable modality to gain access to the papilla, provided it is used in the right place at the right time (2).

Placing an elastic prosthesis within Wirsung's duct may reduce the risk of post- ERCP pancreatitis from 26 to 7% in patients with OSD, but entails added technical difficulties and a delayed second procedure for stent withdrawal.

Bleeding is the second most common complication following ERCP. It is always related to sphincterotomy, although severe bleeding (requiring the transfusion of more than two blood units or surgery) develops in only 0.1-0.5% of cases. Risk factors for this complication primarily include the presence of prior coagulopathy and anticoagulant ingestion within 72 hours after papillotomy. These are also associated with post-sphincterotomy hemorrhage and, to a lesser extent, bleeding during sphincterotomy itself, precut, presence of cirrhosis, cholangitis, and papillary stenosis. In contrast, neither incision length nor NSAID use seem to be in association with hemorrhage development (13-15).

Perforation rates approach 0.3% and may be divided into three types: guide-induced, sphincterotomy-induced in the papillary region, and those developing away from the papilla. Early recognition of the former two types, and aggressive therapy using antibiotics and both gastric and nasobiliary drainage may suffice in 86% of patients. The remaining patients will need surgery. Risk factors for perforation include Billroth II gastrectomy, sphincterotomy or biliary dilatation, intramural injection, and of course suspected OSD (13-15).

A final characteristic complication of ERCP is the development of cholangitis, which is usually related to incomplete drainage, combined percutaneous-endoscopic procedure, presence of fever within 72 hours of the procedure, and stent placement for malignant obstructive jaundice. From a technical viewpoint, leaving a plastic stent in place is recommended for patients with incomplete choledochal stone clearance, as is attempting not to fill intrahepatic radicals that will bear no adequate drainage, and placing just one prosthesis, since unilateral drainage has an effectiveness similar to that of bilateral drainage, yet a lower incidence of cholangitis. Even though prophylactic antibiotics are again controversial, their use in patients with known bile obstruction and patients with heart valves, prior endocarditis history, systemic-pulmonary shunt is still recommended, as well as within one year after synthetic vascular grafting. In 0.2-0.5% of cases post-ERCP cholecystitis may develop, and its risk is associated with contrast accessing the inner bladder; the only way to prevent this complication is obviously by performing a cholecystectomy (13-15).

Experience influences the two aspects conditioning a favorable post-ERCP outcome: technical success and reflective skill. The experience of a biliary endoscopist is defined by initial training, number of procedures performed in his/her career, current (weekly or monthly) number of procedures, and practice setting (small hospital, tertiary hospital, university hospital) (3).

Basic training is essential, as success possibilities increase from 38% for endoscopists with fewer than 25 ERCP procedures to 85% when the number of procedures performed is in excess of 200. Technical competence develops from the second hundred of explorations on (3). Lifetime experience is also relevant and its effects are always revealed when historic series are compared to prospective ones. Lifetime experience not only results in a decreased number of complications, but also in improved success rates, which although partly a reflection of better tools, accessories and clinical patterns, emphasizes the presence of a learning curve that may be spotted even in expert endoscopists. Last, current workload also seems to influence ERCP outcomes, and endoscopists performing less than 1 sphincterotomy per week, or fewer than 200 ERCP procedures per year, run a slightly increased risk of severe complications and exhibit a decreased technical success rate (4), even though results match those in other centers for García-Cano (5).

The frequency of pancreatitis is similar in all centers, regardless of size, and no differences exist between indications and findings in comparative series. Complication rates are even higher in some series from greater centers (16), and the observation of an increased number (up to 46%) of advanced cannulation maneuvers in reference centers has been accordingly dubbed "tertiary effect". However, this sort of advanced maneuvers is associated with an increased number of successful ERCPs (up to 96%), with a lower number of difficult papillas, unnoticed pancreatic injections, and severe complications. Another aspect of this tertiary effect is resident training. Although this does not appearingly influence outcome, trainer-induced complications will undoubtedly be affected by his/her residents' successive learning curves, particularly early during the teaching.

Fortunately though, factors that decrease the risk of complications such as bilirubin elevation, presence of chronic pancreatitis, and presence of choledocholithiasis also exist. Again, these are individual patient-related factors that escape control by endoscopists.

Anyway, a biliary endoscopist must face mortality rates approaching 0.5% (1 death per 500 patients); although such mortality should be assessed within the context of expected mortality in the absence of ERCP, this is hard to face on a daily basis, particularly when considering that most serious pancreatitides are almost exclusively concentrated in younger patients without severe disease or an absolute indication (17). In contrast, many risk factors for the development of complications are considered indications by ERCP prescriptors; thus, in the presence of a doubtful indication or cumulative risk factors a biliary endoscopist has a part of the solution not in his/her hands but in his/her head: learning to say "NO!" is all it takes.

S. Rodríguez Muñoz

Service of Digestive Diseases. Hospital 12 de Octubre. Madrid, Spain

 

References

1. McCune WS, Shorb PE, Moscovitz H. Endoscopic cannulation of the ampulla of Vater: a preliminary report. Ann Surg 1968; 167: 752-6.

2. Mallery JS, Baron TH, Dominitz JA, Goldstein JL, Hirota WK, Jacobson BC, et al. Coplications of ERCP. Gastrointest Endosc 2003; 57: 633-8.

3. Petersen BT. ERCP outcomes: defining the operators, experience, and eviroments. Gastrointest Endosc 2002: 55: 953-8.

4. Arenson N, Flamm CR, Bohn RL, Mark DH, Speroff T. Evidence-based assessment. Patient procedure, or operator factors associated with ERCP complications. Gastrointest Endosc 2002; 56: s294-s301.

5. García-Cano Lizcano J, González Martín JA, Morillas Ariño J, Pérez Sola A. Complicaciones de la colangiopancreatografía retrógrada endoscópica. Estudio en una unidad pequeña de CPRE. Rev Esp Enferm Dig 2004; 96 (3): 163-73.

6. Tarnasky PR, Palesch YY, Cunningham JT, Mauldin PP, Cotton PB, Hawes RH. Pancreatic stenting prevents pancreatitis after biliarysphincterotomy in patients with sphincter of Oddi dysfunction. Gastroenterology 1998; 115: 1518–24.

7. Hogan W. Stenting the Pancreas: Is this the solution to post-ERCP pancreatitis? Gastroenterology 1998; 115: 1591-4.

8. Andriulli A, Leandro G, Niro G, Manglia A, et al. Pharmacologic treatment can prevent pancreatic injury after ERCP: a meta-analysis. Gastrointest Endsoc 2000; 51: 1-7.

9. Moretó M, Zaballa M, Casado I, Merino O, et al. Transdermal glyceryl trinitrate for prevention of post-ERCP pancreatitis: a randomized double-blind trial. Gastrointest Endosc 2003; 57: 1-7.

10. 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: 476-83.

11. Devière J, Le Moine O, Van Lathem JL, Eisendrath P, et al. Interleukin º10 reduces the incidence of pancreatitis after therapeutic endoscopic retrograde cholangiopancreatography. Gastroenterology 2001; 120: 498-505.

12. Murray B, Carter R, Imrie C, Evans S, O'Suillleabhain C. Diclofenac reduces the incidence of acute pancreatitis after endoscopic retrograde cholangiopancreatography. Gastroenterology 2003; 124: 1786-91.

13. Freeman ML, Nelson DB, Sherman S, Haber GB, Herman, ME, Dorsher PT, et al. Complicationsof endoscopic biliary sphincterotomy. N Engl J Med 1996; 335: 909-18.

14. 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.

15. 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.

16. 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: 425-34.

17. Trap R, Adamsen S, Hart-Hansen O, Henricksen M. Severe and fatal complications after diagnostic and therapeutic ERCP: a prospective series of claims to insurance covering public hospitals. Endoscopy 1999; 31: 125-30.