Quality Issues and Measures in ERCP

Appropriate Indication

One of the most important quality indicators for ERCP is an appropriate indication (see Chapter 7 ). In the United States, lack of indication for ERCP is the most common reason for legal allegation (see Chapter 13 ). Indications for ERCP vary and include:

• 1.

  Obstructive jaundice

• 2.

  Clinical, biochemical, or imaging data suggestive of pancreatic or biliary tract disease

• 3.

  Clinical suspicion of pancreatic malignancy when direct imaging studies are normal or equivocal

• 4.

  Evaluation and treatment of idiopathic recurrent pancreatitis

• 5.

  Preoperative evaluation and treatment of chronic pancreatitis

• 6.

  Preoperative evaluation of pancreatic pseudocyst

• 7.

  Sphincter of Oddi manometry

• 8.

  Endoscopic sphincterotomy:

  • a.

    Choledocholithiasis

  • b.

    Papillary stenosis or sphincter of Oddi dysfunction causing disability

  • c.

    Facilitation of biliary stent placement

  • d.

    Facilitation of balloon dilatation

  • e.

    Treatment of sump syndrome

  • f.

    Treatment of symptomatic choledochocele

  • g.

    Palliation of obstructive jaundice in poor surgical candidates with ampullary carcinoma

  • h.

    To provide access to the main pancreatic duct

• 9.

  Stent placement for treatment of:

  • a.

    Benign or malignant strictures

  • b.

    Fistulas

  • c.

    Postoperative leak

  • d.

    Irretrievable large common bile duct stone(s)

• 10.

  Balloon dilatation of ductal stricture

• 11.

  Nasobiliary drain placement

• 12.

  Drainage of symptomatic and infected pancreatic fluid collections

• 13.

  Tissue sampling of pancreatic or bile ducts

• 14.

  Pancreatic therapeutics

• 15.

  Endoscopic papillectomy

• 16.

  Facilitation of cholangioscopy/pancreatoscopy

ERCP is generally not indicated in the following clinical scenarios :

• 1.

  Abdominal pain without objective evidence of pancreaticobiliary disease. Objective evidence includes abnormal laboratory or imaging studies suggestive of pancreaticobiliary disease. The risk-benefit ratio is high in the absence of these objective findings. If ERCP is pursued for abdominal pain without objective evidence of pancreaticobiliary disease, it has been suggested that sphincter of Oddi manometry be performed (see Chapters 16 and 47 ). A recent randomized controlled trial in type III sphincter of Oddi dysfunction found no benefit for sphincterotomy regardless of manometry findings.

• 2.

  Routine ERCP before cholecystectomy. Preoperative ERCP before cholecystectomy should be performed in the setting of acute cholangitis or if the preprocedure likelihood for choledocholithiasis is high (e.g., abnormal imaging showing stones, persistently abnormal liver tests, biliary duct dilation).

• 3.

  Routine ERCP for the relief of a malignant biliary obstruction in patients with resectable pancreaticobiliary malignancy. ERCP in this setting has been associated with higher preoperative and postoperative adverse events compared with no ERCP, as confirmed in a recent randomized prospective trial. Nevertheless, ERCP should be considered in the setting of a malignant biliary obstruction in patients with intense pruritus, particularly if there is a delay in surgical resection, and to treat acute cholangitis.

The indication for the procedure should be documented in the medical record. If the ERCP is being performed for a nonstandard indication, this should be discussed with the patient in detail, well documented, and justified.

Informed Consent

Given the high adverse event rate inherent with ERCP, informed consent must be obtained from the patient or legal guardian before the procedure, except in the setting of a life-threatening medical emergency. The components of the informed consent process include the following: (1) voluntary consent, (2) rational decision-making capability of the patient or legal guardian, and (3) conveyance of “adequate information.” It is the physician's responsibility to disclose as much information as a reasonable patient would wish to know when making a decision. Determining “reasonable” is not a precise science, and the physician must simultaneously balance the need to avoid overwhelming the patient with providing pertinent risk information. The consent should address the most commonly encountered adverse events associated with ERCP and their expected rates. These include pancreatitis, infectious adverse events (cholangitis and cholecystitis, infection of pancreatic fluid collection), postsphincterotomy bleeding, perforation, and sedation-induced cardiopulmonary adverse reactions (see Chapter 8 ). Although there are differing opinions as to whether patients must be informed of the potential need for surgery, a prolonged hospital stay, or death, we cannot overemphasize the value of extensive patient education about ERCP and its potential adverse events. Disputes about the extent of the education and consent process are common in ERCP lawsuits (see Chapter 13 ). Although state laws vary on who can legally obtain informed consent, most experts recommend that the endoscopist performing the procedure obtain the consent. In a large prospective multicenter study describing the ERCP consent process in England, the majority of endoscopists (84%) obtained the consent themselves and 14% delegated this responsibility to another member of the team.

The incidence of post-ERCP pancreatitis (PEP) typically ranges from 1% to 7% but can be higher in certain clinical situations. Multiple studies have identified risk factors for PEP. These factors can be classified into patient-related and procedure-related factors based on prospective studies. Patient-related factors include a history of PEP and known or suspected sphincter of Oddi dysfunction, age <60 years, female gender, absence of chronic pancreatitis, and normal serum bilirubin. Procedure-related factors include more than one to two pancreatic contrast injections, moderate to difficult cannulation (defined variably but usually ≥10 attempts at cannulation), pancreatic sphincterotomy, precut sphincterotomy, minor papilla sphincterotomy, balloon sphincteroplasty without sphincterotomy, and trainee involvement. Although such factors should be addressed with the patient during the informed consent process, the optimal degree of explanation (detail) of these factors is unknown and not standardized. However, patients should be informed that pancreatitis can be severe in a small percentage of cases and may require prolonged hospitalization.

Infectious complications are uncommon after ERCP. Acute cholangitis occurs in up to 1% of ERCPs. Post-ERCP acute cholecystitis is seen in 0.2% to 0.5% of cases. Recently, carbapenem-resistant Enterobacteriaceae (CRE) has been transmitted person-to-person through exposure to duodenoscopes (see Chapter 5 ). The complex design of the duodenoscope with the elevator assembly has been postulated to impede high-level disinfection, leading to transmission in several cases. One hospital changed its reprocessing to include ethylene oxide gas sterilization, without subsequent additional infections. Other strategies for enhanced reprocessing have included a culture and sequester protocol (duodenoscopes are not returned to service until cultures return negative, usually 2 days later), and double reprocessing. There have been and will likely continue to be updates to recommendations on duodenoscope reprocessing from endoscope manufacturers, with at least one manufacturer issuing a recall of duodenoscopes for a redesign of the elevator assembly.

Post-ERCP bleeding can occur after sphincterotomy, ampullectomy, or transmural drainage of pancreatic fluid collection. Postsphincterotomy bleeding occurs in 0.8% to 2% of cases, and the possible need for transfusion, surgery, or radiologic embolization should be discussed. The patient's willingness to receive blood transfusion, if needed, should be discussed and documented.

ERCP-induced perforation can be related to the guidewire (periductal or ductal perforation), sphincterotomy (duodenal perforation), or endoscope induced at a site remote from the papilla. Post-ERCP perforation occurs in 0.35% to 0.6% of cases, and the potential need for surgery and prolonged hospital stay for this adverse event should be discussed. Death occurs after ERCP in 0.07% of cases.