The Need for Repeat ERCP After Endoscopic Treatment of Postsurgical Biliary Leaks

Validation of a Clinical Prediction Rule to Determine the Need for Repeat ERCP After Endoscopic Treatment of Postsurgical Biliary Leaks

Post-surgical biliary leaks are relatively common after surgeries associated with hepatobiliary health. Left untreated, biliary leaks can lead to significant morbidity. Biliary leaks are most often successfully managed endoscopically, by way of performing an ERCP (endoscopic retrograde cholangio-pancreatography) procedure. These procedures help manage the bile leak by decreasing pressure at the opening of the common bile duct and promoting bile flow into the small bowel (rather than out the leak) via stent insertion.

Guidelines published by the American Society for Gastrointestinal Endoscopy (ASGE) report that stents are generally placed for 4 to 6 weeks and recommend longer intervals for more complex leaks. However, formal recommendations concerning the modality of biliary stent removal do not exist.

One option is performing a repeat ERCP when removing the stent. While comprehensive, this exposes the patient to additional radiation, and requires additional fluoroscopy resources and/or technicians. Furthermore, ERCPs are less available, especially in smaller centers, and are costly. A second option is a gastroscopy with simple stent removal has, given the low probability of requiring repeat intervention, the relatively low procedural cost, and the relatively favorable adverse event profile, and easier accessibility compared to ERCP procedures.

A simple, safe and reliable prediction rule was developed retrospectively to identify patients in whom biliary stent removal via gastroscopy could be safely performed, as opposed to repeat ERCP. A positive result using the rule requires satisfaction of four non-invasive clinical markers: (1) a normal post-surgical serum alkaline phosphatase value, (2) bile leak 'type C' at initial ERCP (a small or absent leak with no other biliary pathology), (3) a bile leak caused by laparoscopic cholecystectomy, and (4) a time between initial and follow-up endoscopy of 4 to 8 weeks. Validating this prediction rule prospectively could have implications on patient safety by decreasing ERCP-related adverse events, and could also have important implications with regard to health resource utilization.

The previously developed prediction rule will be applied in a prospective fashion to consecutive patients undergoing ERCP for suspected post-surgical biliary leak. The same variables will be measured as in the initial study, but in this study, a research assistant (RA) will prospectively enter data and follow up with patients. For the purposes of this study, a second ERCP will be planned for stent removal and repeat cholangiogram (as per current practice). The sequence of procedures will be as follows:

Initial ERCP Visit:

1. The patient is referred to PLC Endoscopy Unit for consideration of ERCP procedure for indication of post-surgical biliary leak and is booked by the attending advanced/therapeutic endoscopist on call as appropriate, as per the standard of care.

2. Once verbal and written consent to approach has been given to their admitting nurse, patients will be approached by the RA prior to their procedure to discuss the actively enrolling study, and to answer any potential questions.

3. If the patient agrees to participate, the informed consent form(s) is/are signed and witnessed. If the patient decides not to participate, the ERCP proceeds as per the usual standard of care (and the sequence that follows does not apply).

4. The RA collects relevant patient-related and pre-procedural data through combination of direct patient interview and subsequent review of the medical records. Variables to be collected include:

   1. age
   2. gender
   3. presence of imaging evidence showing bile leak
   4. type of surgery performed
   5. time from surgery to ERCP

5. Given that the patient will have already had an intravenous (IV) catheter placed on admission to the Endoscopy Unit (for purposes of receiving sedation or anesthesia during the procedure), a laboratory technician arrives to draw bloodwork from the patient to collect the following variables. If all following bloodwork has already been drawn for the patient within the preceding 24 hours prior to the procedure, no additional bloodwork will be drawn.

   1. complete blood count (CBC), including white blood cell count (WBC)
   2. gamma-glutamyl transferase (GGT)
   3. alkaline phosphatase (ALP)
   4. alanine aminotransferase (ALT)
   5. aspartate aminotransferase (AST)
   6. total and direct bilirubin
   7. international normalized ratio (INR)

6. The patient is then met by the endoscopist performing the procedure to discuss the risks and benefits of the procedure and to have the patient sign informed consent for the procedure. The patient enters the endoscopy/fluoroscopy room and the procedure commences.

7. During the procedure, peri-procedural data are recorded by the RA by direct observation, and/or, if necessary, in consultation with the procedural physician(s) and/or nurse(s). Variables to be collected include:

   1. stent length placed
   2. stent caliber placed
   3. presence or absence of other pathology (stones, strictures, other)
   4. extent of leak

   i. large (defined by the presence of early extravasation of contrast from the leak source prior to opacification of the intrahepatic bile ducts) ii. small (all others) e. location of biliary leak on cholangiography: i. cystic duct remnant ii. accessory bile duct iii. intrahepatic, left or right ducts iv. common hepatic or bile ducts v. no leak found vi. uncertain/ not described f. type of leak i. Type A (large or common bile duct leaks) ii. Type B (no leak observed or small leak, but other pathology observed) iii. Type C (no leak observed or small leak with no other biliary pathology observed)

8. After the procedure is complete, the patient is moved to the recovery room for observation. Depending on their disposition and the course of the procedure, observation times and protocols differ. During this stage, the RA collects any relevant post-procedural data, and any missing pre- or peri-procedural data.

9. Any peri- or immediate post-procedural adverse event is recorded.

10. The patient is examined by the endoscopist prior to discharge and provided with written post-procedure information, including a phone number to call in the event of any issues.

11. The RA meets with the patient prior to their discharge from the unit to answer any final potential questions.

12. The patient is booked for repeat ERCP procedure as per standard current practices, usually between 6-8 weeks, for stent removal and repeat cholangiogram to rule out residual leak.

13. At 30 days post-procedure, the RA reviews the medical record for occurrence of any adverse event(s), in addition to calling the patient to inquire about any unplanned visits to healthcare facilities. Adverse events of interest include:

    1. pancreatitis
    2. bleeding
    3. perforation
    4. cholangitis and/or sepsis (not present prior to the procedure)
    5. pain not meeting the definition of any of the above that requires repeat unplanned presentation to a healthcare facility

14. At 6 months post index procedure, any repeat or recurrent leaks and/or repeat ERCP procedures are captured.

Inclusion Criteria:

• Patients having ERCP for suspected post-surgical biliary leak;
• age 18 years or older
• able to give informed consent to involvement.

Exclusion Criteria:

• Unable or unwilling to provide informed consent;
• age < 18 years;
• non-surgical biliary leak;
• previous ERCP with sphincterotomy;
• previous treatment or attempted treatment of biliary leak;
• unsuccessful initial ERCP.