THE ROLE OF MAGNETIC RESONANCE CHOLANGIOPANCREATOGRAPHY FOR PREVENTING PANCREATIC FLUID COLLECTION RECURRENCE AFTER EUS-GUIDED DRAINAGE

Background & Aims: Minimal prospective data are available on intraoperative pancreatoscopy (IOP) for the diagnostic work-up of intraductal papillary mucinous neoplasia (IPMN). The aim of the current report is to summarize perioperative impact on patient management from an ongoing 5-year prospective cohort study of IOP in patients with pancreatic surgery for suspected or confirmed IPMN.

Methods: Prospective multicenter study in patients with main pancreatic duct (MPD) diameter >5mm on preoperative imaging and scheduled for surgery for suspected MPD IPMN or mixed-type IPMN in 8 centers from 6 countries. IOP was performed using SpyGlass™ DS Direct Visualization System or SpyGlass™ Discover Digital System (Boston Scientific Corporation, Marlborough, Massachusetts) at the time of the index pancreas resection. This was an observational cohort study in which IOP findings were permitted to be interpreted by and acted upon at the discretion of the local surgical team. Patients had postoperative follow-up before hospital discharge and at 4 weeks. Main endpoint was IOP technical success; secondary endpoints included influence of IOP on the surgical resection plan, and 4-week procedure-related serious adverse events (SAEs).

Results: Overall, 100 patients were included, mean age 68.9±8.8 years and 58% male. On preoperative imaging, 94 (94%) patients had IPMN involving the main duct. MPD dilation was >10 mm in 58%, and 5-9 mm in 42% of patients. The operative approach included 45 pancreaticoduodenectomies, 19 PPPD, 14 distal and 11 total pancreatectomies, and 11 other surgeries. Both open and minimally invasive approaches were utilized: 71 (71%) open, 23 (23%) robotic, and 6 (6%) laparoscopic. IOP was technically successful in 91 (91%) patients. In 25 patients (25%), the initial surgical plan was modified based on IOP findings. Their procedural modifications included one or more of the following: extension (8) or sparing (5) of additional parenchymal resection, avoidance of total pancreatectomy (6), removal of cast, stone, or sludge from the remnant duct (2), conversion to total pancreatectomy (3), or conversion from distal pancreatectomy to a pancreatoduodenectomy. (1). Ninety-eight (98.0%) of patients completed a 4-week follow-up visit, one missed the 4-week visit, and one died. There were 25 reported SAEs related to surgery with Clavien-Dindo classification ≥ III in 15 patients, including one Clavien-Dindo V. None were reported as related to IOP.

Conclusion: In this prospective multicenter study, IOP was technically highly feasible and led to changes in intraoperative decision-making in 25% of surgical patients with IPMN and main duct involvement, especially regarding extension or sparing of pancreatic parenchymal resection. No IOP-related SAEs occurred. (ClinicalTrials.gov number NCT03729453)

Background: Pancreatic fluid collections (PFC) may recur after initial successful endoscopic drainage of walled-off necrosis (WON). The most common reason is due to the presence of a disconnected main pancreatic duct (DPD), which has been noted in up to 40% of patients undergoing EUS-guided drainage. DPD can lead to an increased risk of pancreatic fluid recurrence following transmural stents removal. The primary aim of this study is to assess the role of magnetic resonance cholangiopancreatography (MRCP) for preventing PFC recurrence by evaluating the integrity of the main pancreatic duct.

Methods: This is a single-center retrospective cohort study of patients with WON who underwent EUS-guided drainage with lumen apposing metal stents (LAMS) between August 2014 and May 2022. Patients with pseudocysts or who did not achieve WON resolution were excluded. Patients were divided between two different groups: those with MRCP vs those without MRCP prior to removal of transmural stents. PFC recurrence was defined as a new collection that developed after resolution of WON as documented by cross-sectional imaging and/or endoscopic reports after transmural stent removal. Chi-Square analysis was performed to compare categorical variables. Independent sample t test was performed to compare means of continuous variables. Statistical significance was defined by p-value < 0.05.

Results: A total of 127 patients with diagnosis of WON were identified, of whom 47 had MRCP performed during endoscopic management (Table 1). In patients without MRCP, 13/80 (16.2%) had PFC recurrence vs 0/47 (0%; p=0.004) in those with MRCP performed. MRCP identified DPD in 12 (25.5%) patients, all of whom were managed with indefinite drainage with double-pigtail plastic stents (DPPS) without PFC recurrence. Those without MRCP, DPD was suspected in 17 patients, all of whom were managed with long term DPPS. In the remaining 63 (78.7%) patients without MRCP or mention of ductal integrity, PFC recurrence was noted in 13 (20.6%) (Figure 1). PFC recurrence was noted in all patients who were neither evaluated, nor had pancreatic duct integrity addressed. PFC recurrence was identified at a median interval of 201 days after transmural stents were removed. Among those with PFC recurrence, 11 patients (85%) had undiagnosed DPD, which was later confirmed after PFC recurrence (9 by MRCP, 2 by CT). Of these, 9 patients (82%) were managed with indefinite DPPS and 2 (18%) with percutaneous drains, all without further PFC recurrence.

Conclusion: In this study, patients with WON who underwent EUS-guided drainage had a lower rate of PFC recurrence when MRCP assessment for main pancreatic duct integrity guided removal vs retention of transmural stents. This difference was largely driven by the identification of DPD with MRCP and appropriate endoscopic management.