Determination of the Length of Pancreatic Ductotomy by Pancreaticoscopy During Frey’s Procedure for Chronic Pancreatitis

Abstract

Objective:

To study the impact of pancreaticoscopy during Frey’s procedure for treating chronic pancreatitis (CP).

Background:

Excision of the central part of the head of the pancreas along with longitudinal pancreaticotomy (Frey’s procedure) is widely performed for the treatment of CP. However, there is no reliable method to determine the necessary length of longitudinal pancreaticotomy during surgery for CP.

Methods:

Thirty-five consecutive patients with CP were scheduled for Frey’s procedure with intraoperative pancreaticoscopy. The length of the longitudinal pancreaticotomy was tailored by pancreaticoscopy in the following manner: (1) it did not extend beyond the neck in case of a uniformly dilated main duct with patent branch duct confluences and a clear lumen; (2) in case of an obstructed main duct or branch duct confluence of any cause, the main duct was opened to include the most distal obstruction.

Results:

All patients underwent Frey’s procedure and intraoperative pancreaticoscopy. Based on the pancreaticoscopy findings, pancreaticotomy over the body of the gland was not necessary in 34% of the patients. A short (4–6 cm) ductotomy extension over the pancreatic body was required in 14% of the patients. Full-length pancreaticotomy was required in 52% of the patients. The median operative time was 145 minutes, and the median blood loss was 70 mL. Four patients (11.4%) experienced postoperative complications. There were no 90-day postoperative mortality or hospital readmission rates. At the median follow-up of 19 months, 31 patients (88.5%) had no pain attacks requiring medication.

Conclusions:

Intraoperative pancreaticoscopy helps to determine the length of longitudinal pancreaticotomy and reduce pancreatic trauma during Frey’s procedure for treating CP.

INTRODUCTION

Decompression of the pancreatic ductal system is one of the goals of surgery for chronic pancreatitis (CP), for which the excision of the central part of the head of the gland along with longitudinal pancreaticotomy was proposed by Frey and Smith in 1987.¹ However, the pioneers of pancreatic surgery, such as Partington and Rochelle² and Frey and Amikura,³ have emphasized that full-length pancreatic ductotomy is not always necessary to effectively drain the main duct. Furthermore, there is no reliable method to determine the necessary length of pancreaticotomy during surgery for CP. Intraoperative pancreaticoscopy seems useful for this purpose, but its value in surgery for CP remains undetermined. Therefore, we aimed to study the use of pancreaticoscopy during Frey’s procedure for treating CP.

METHODS

Thirty-five consecutive patients with CP, with pancreatic duct dilatation >4 mm and no extrapancreatic pseudocyst or suspected pancreatic malignancy, were scheduled for Frey’s procedure between 2015 and 2021. Transabdominal ultrasound, contrast-enhanced computed tomography, and upper gastrointestinal endoscopy were performed as preoperative diagnostic examinations for all patients. Patients with biliary obstruction underwent endoscopic retrograde cholangiopancreatography with biliary stent placement.

The initial operative exploration included midline laparotomy, the Kocher maneuver, opening of the lesser sac through the gastrocolic omentum, and exposure of the entire anterior surface of the pancreas. Intraoperative ultrasonography (IOUS) of the pancreas was performed to map the ductal system, as well as to determine the location and patency of the intra- and peripancreatic blood vessels. A small (<15 mm) ductotomy was performed at the level of the neck under IOUS guidance. Pancreaticoscopy was then performed through the ductotomy in both directions—toward the head and toward the tail, using 1 of 2 commercially available flexible endoscopes (MAF-GM2 fiber bronchoscope and CHF-V video-choledochoscope, Olympus, Japan). Saline was constantly flushed through the instrument channel of the endoscope to ensure a clear visual field and to clear the ducts of debris and nonimpacted calculi. Pancreaticoscopy aimed to evaluate the patency of the main duct and branch duct confluences, as well as the presence of inflammation, intraductal debris, and calculi. Partial resection of the head was then performed as described elsewhere.^1,3 The length of the pancreatic ductotomy toward the body and tail was tailored by pancreaticoscopy in the following manner: (1) the ductotomy was not extended beyond the neck in case of a uniformly dilated main duct with patent branch duct confluences and a clear lumen; (2) in case of an obstructed main duct or branch duct confluence of any cause, the main duct was opened to include the most distal obstruction. Subsequently, transmesocolic Roux-en-Y pancreaticojejunostomy was performed. Additional extrapancreatic procedures were performed when necessary. The study design was approved by the institutional ethics review board, and all patients provided written informed consent.

RESULTS

Intraoperative pancreaticoscopy during Frey’s procedure was performed in 35 patients with CP at a median age of 55.6 years, who were predominantly male (female:male, 1:2.5). Pancreaticoscopy revealed obstruction of the Wirsung’s duct at the head of the pancreas in all patients. Moreover, no patient had identifiable patent duct confluence at the head of the gland. In 12 patients (34.3%), pancreaticoscopy toward the tail revealed a dilated main duct with patent branch duct confluences and a clear lumen (Fig. 1A); therefore, ductotomy was not extended. In the remaining 23 patients, pancreaticoscopy toward the tail showed inflammatory changes in the duct and main/branch duct obstruction of various degrees and extents (Fig. 1B–D), requiring extension (4–6 cm) of the ductotomy over the pancreatic body (n = 5, 14.3%) or full-length ductotomy (n = 18, 51.4%). Although IOUS can always detect intraductal calculi, we were unable to evaluate the patency of the branch duct confluence and inflammatory changes in the duct with IOUS (Fig. 1E, F).

FIGURE 1.

Intraoperative photographs. A, Pancreaticoscopy showing dilated main duct with normal appearing branch duct confluence (arrow); (B) fibrotic main duct with stenosis, obstructed by fibrin web; (C) fibrotic branch duct confluence, partially obstructed by fibrin web (arrow); (D) completely obstructed branch duct confluence (arrow); (E) intraoperative ultrasonography—dilated main duct with multiple intraductal stones (arrows); (F) pancreaticoscopy image corresponding to (E).

Additional procedures were performed in 10 patients (28.5%), including choledochoduodenostomy (n = 6), gastroenterostomy (n = 3), and left nephrectomy (n = 1; renal cancer). The median operative time was 145 minutes, and the median blood loss was 70 mL. Four patients (11.4%) had postoperative complications (wound infection managed at the bedside, n = 3; cholangitis requiring antibiotics, n = 1). There were no 90-day postoperative mortality or hospital readmission rates. The diagnosis of CP was confirmed histologically. At a median follow-up of 19 months, 31 patients (88.5%) had no pain attacks requiring medication and had gained a body weight (at median of 6 kg).

DISCUSSION

The accumulated clinical evidence for the treatment of CP recognizes surgery as the most effective therapeutic modality.^4,5 Among the existing surgical options, Frey’s procedure permits, conceptually and technically, the most tailored approach for both resection of the head and longitudinal pancreaticotomy, causing less trauma and similar or better results compared to other procedures.⁶

In a patient with CP, information from preoperative imaging studies does not meet the surgeon’s need for real-time, fast, and reliable identification of important anatomical structures and morphological changes during surgery. The introduction of pancreatic IOUS in 1980⁷ resolved this problem partially. The limited accuracy of both the preoperative imaging studies and IOUS to detect discrete changes in the pancreatic ductal system led to the introduction of intraoperative pancreaticoscopy, initially used for intraductal papillary mucinous neoplasms.⁸ Our findings demonstrate that pancreaticoscopy might also be useful during surgery for CP, as it provides information that cannot be obtained by other diagnostic modalities. Furthermore, our experience suggests that intraoperative pancreaticoscopy can be performed by the surgeon safely, without the need for specific training, and with no procedure-related complications, similar to widely used for common bile duct exploration intraoperative cholangioscopy.

The main finding of this study was that with intraoperative pancreaticoscopy, modified Frey’s procedure can be performed in about half of the patients, helping to reduce pancreatic trauma, the length of pancreaticojejunostomy, and subsequently the operative time. Based on the pancreaticoscopy findings, ductotomy extension over the body of the gland was not necessary in 34% of the patients in this study. A short (4–6 cm) ductotomy extension over the pancreatic body was required in 14% of the patients to resolve segmental main duct obstructions or to include partially obstructed branch duct confluences. It should be noted that no conclusive statements can be made based on the aforementioned limited experience. However, the observations of this report justify further prospective evaluation of the value of intraoperative pancreaticoscopy in the treatment of CP.