This is a cross-sectional study of patients who were referred for ERCP, routine and urgent cases, which were systematically evaluated and performed by operators with experience in the applied procedure.
Data collection was performed retrospectively from an electronic database, including exams performed from November 20, 2006 to August 12, 2022 at Endoclinic SP. A total of 2064 patients and 2233 consecutive ERCP exams were included in the initial analysis in this period.
Patients underwent ERCP examination with direct cannulation and CP (conventional papillotomy) or, if the latter failed, they migrated to the EFP group. All ERCP cases, both those submitted to CP and EFP, were singly evaluated, and both groups were analyzed in terms of cannulation success and its complications.
The evaluated patients underwent the examination and the diagnoses were grouped as follows: a) choledocholithiasis; b) ERCP with minimal changes, such as dilatation of the CBD beyond 12 mm in diameter without an obstructive factor, history of jaundice or pancreatitis, tests indicated by imaging or laboratory tests with unconfirmed suspicion of stones or obstruction. Obstructive bile duct neoplasms were divided and grouped as follows: c) pancreatic head neoplasm; d) neoplasms of the hepatic hilum (cholangiocarcinoma, extrinsic compression by metastases); and e) neoplasms of the papilla of Vater. Benign lesions were grouped into: f) benign fibrotic strictures (undetermined strictures, late postoperative sequelae, papillitis or fibrotic thinning of the distal common bile duct); g) Mirizzi syndrome; h) early postoperative complications of the bile ducts, such as partial or total ligations, fistulas, and strictures; i) sclerosing cholangitis; j) chronic pancreatitis; and k) other diagnoses, such as choledochal cyst, and cholelithiasis.
The exam reports and images obtained during this period were recorded and saved in a database (OCRAM® system). The data mined for the composition of the research were extracted from the relational database MySQL Community, version 5.5.40, software, entitled OCRAM Capture of Medical Images and Reports. This system was developed using the Java programming language and was used to capture the photos of the ERCP exams and compose the respective reports during the study period. The ERCP reports were typed using OCRAM software were structured in XML (Extensible Markup Language) format and followed an XML-Schema according to W3C (Worldwide Web Consortium) standards, resulting in a well-formed, valid and standardized structure of the ERCP reports in XML, this allowed the mining of terms referring to diseases to be performed reliably through the declarative search language SQL (Structured Query Language) in conjunction with an XML DOM (Document Object Model) parser [11].
Inclusion criteria
All patients were submitted to an attempt at cannulation through the ostium of the papilla with a sphincterotome and guide wire. When direct cannulation of the common bile duct is not achieved and the guide wire goes into the pancreatic duct, we opt for the double guide wire technique. This technique was used in many of the cases described here and these cases were grouped as successful cannulation through the ostium. We defined cannulation failure after performing the following tactics: a) access to the common bile duct was not achieved after at least 4 attempts with a guide wire; b) attempts at cannulation after injection of a small amount of contrast to identify the common bile duct and the pancreatic duct; c) if the guidewire goes only to the pancreatic duct, we perform the double guidewire technique and, if the second guidewire fails to gain access to the biliary tract, we consider it as a failure. The cases of cannulation failure migrated to EFP, according to the criteria of the fistulotomy technique.
Exclusion criteria
Of the 2064 patients initially evaluated in the electronic system, 105 patients in whom there were anatomical changes that made it impossible to perform ERCP: surgeries such as Roux-en-Y or Billroth II gastrectomy, or esophageal, stomach or duodenal stenosis were excluded. A total of 1959 patients and 2233 exams remained for the final analysis.
Intervention
All patients had their exams previously evaluated and underwent preparation with a 12-hour fast before the procedure. Patients who were using antiplatelet agents and anticoagulants were instructed to discontinue these medications. The use of ciprofloxacin 500 mg every 12 hours was indicated for all patients, starting from 6 hours when there was no increase in bilirubin and 24 hours previously in cases of bile duct obstruction in patients with elevated bilirubin levels. All patients were initially submitted to the standard cannulation technique, using the 3-way sphincterotome and 0.035 or 0.025 guide wire, depending on the availability of the brands Olympus®, Boston®, Cook®, MediGlobe®, Scitech® or GFE®. The WEM® electrosurgical generator, model SS200A, was used in all cases. In the case of cannulation failure, following the criteria for indicating early fistulotomy, the patients underwent EFP. After cannulation failure (defined as failing to introduce the guide wire into the bile duct five times, even after injecting a small amount of contrast into the papilla ostium, or the guide wire inadvertently being directed into the pancreatic duct), our preference was to perform EFP early, avoiding trauma to the papilla and the injection too much contrast (Fig. 1).
Algorithm for access to the common bile duct
After identification and “palpation” with the tip of the fistulotome of the lateral limits of the papilla and exposure of the infundibulum, a wide, shallow incision only of the papillary mucosa was iniciated and, purely cut, with the fistulotome needle adjusted to approximately 2 mm, from top to bottom, just below the transverse crease, avoiding opening the region of the common channel. With the needle-knife, the needle was retracted, and the mucosal edges were pushed aside to expose the submucosa. Lateral incisions were made to expand the exposure and then superficial incisions were made to open the submucosa and, dissect thin layers one at a time, interspersed by blunt dissection with the tip of the needle-knife retracted; these steps were followed by identification of vessels, hemostasis and exposure of the sphincter muscle of the distal common bile duct. If bleeding occured, washing was performed with pressurized water through the fistulotome catheter itself or through the working channel of the duodenoscope, the bleeding point was identified and captured with hot biopsy forceps and sealed by seizure, rapid hemostasis was achieved by coagulation, and underlying thermal tissue damage was avoided. At this stage, transversal accessory incisions were made to remove the mucosa over the papilla and improve exposure (Fig. 2).
EFP with lateral incisions. a papilla of difficult cannulation with the guide wire; b wide longitudinal incision and lateral incisions; c exposure of the submucosa by pushing aside the mucosal edges, sectioning the musculature fibers until opening the common bile duct; d deep cannulation of the bile duct
Subsequently, sectioning of the muscle fibers and of the common bile duct mucosa was performed, with bile outflow in most cases. Once the common bile duct mucosa was identified, the guide wire was introduced, confirmed by radioscopy and bile duct contrast (Fig. 3) (video 1).
EFP with identification of the muscle tissue. a failed cannulation with the guide wire; b repositioning of the duodenoscope to visualize the infundibulum of the papilla; c Wide longitudinal incision from the transverse crease; d wide lateral incisions; e identification of the muscular sphincter of the papilla; f sectioning the muscular fibers to expose the common bile duct mucosa; g cannulation of the bile duct; h deep cannulation and contrast injection
If necessary, we can perform dilation with a 12- or 15-mm balloon catheter to facilitate removal of larger stones (Fig. 4).
Enlarged fistulotomy of the papilla. a failed cannulation with the sphincterotome and guide wire; b failed cannulation with the guide wire and after injection of a small amount of contrast; c EFP with wide incisions; d dissection of the papilla; e identification of the muscular sphincter of the papilla, the muscular fibers, and the common bile duct mucosa; f easy cannulation of the bile duct after dissection; g dilation with 12 mm balloon of the fistula orifice; h common bile duct stone removal
The examinations were performed with the patients in the prone position, with noninvasive heart rate monitoring and pulse oximetry throughout the procedure and the use of routine supplemental oxygen. Intravenous medication was administered for sedation with midazolam, fentanyl and propofol, performed by an anesthesiologist. The devices used were a Fujinon® 4400 series duodenoscope (Fujifilm, Tokyo, Japan), sphincterotome, guide wires and needle-knife brands Olympus®, Boston®, Cook®, MediGlobe®, Scitech® or GFE®. None of the authors have any conflicts of interest.
Postprocedurale care
Patients were contacted post-procedure by a member of the team or by the physicians responsible for the case. Complications such as bleeding, post-ERCP pancreatitis, and cannulation failure were discussed. Patients who presented abdominal pain, nausea and vomiting within 48 hours were examined by a surgeon and amylase and lipase tests were collected. Patients who had bleeding during the procedure or late bleeding were monitored and hematocrit and hemoglobin tests were ordered. Patients or family members were contacted 12, 24, and 48 hours after the procedure and instructed to contact their physician to report any ERCP-related symptoms. When necessary, patients were monitored for days or weeks, according to their clinical condition.