Mirizzi syndrome is a rare complication of long-term chronic cholecystitis, characterized by extrinsic compression of the common hepatic duct (CHD) that may progress to the development of cholecystobiliary fistula. It is a challenging clinical entity to manage and given its limited presence in the literature; there are no internationally-accepted guidelines for its management.

A 38-year-old female patient with no significant past medical history presented to the emergency department with acute on chronic, intermittent postprandial right upper quadrant (RUQ) pain, associated with nausea. On exam, the patient had tenderness in the RUQ and midepigastrium. Laboratory studies demonstrated 11,400 white blood cells/mm³ of mild leukocytosis and 1.3 mg/dL of elevated total bilirubin. RUQ ultrasound demonstrated cholelithiasis, wall thickening, positive sonographic Murphy’s sign, common bile duct (CBD) of 0.6 mm, and CHD of 0.4 mm. The patient was admitted for intravenous antibiotics, with a plan to proceed to operating room for operative intervention.

On hospital day 1, the patient was therefore taken to the operating room for laparoscopic cholecystectomy and intraoperative cholangiogram. Upon laparoscopic evaluation of the RUQ, there were dense adhesions between the gallbladder and the duodenum which were carefully taken down with monopolar surgical energy and blunt dissection. The gallbladder was found to be severely inflamed with firm, chronic inflammation involving the hepatocystic triangle as well as several large (>1 cm) stones chronically impacted in the infundibulum. This impaired lateral retraction and limited further progression of infundibular dissection from the densely adherent scar tissue. Therefore, it was decided to remove the stones for improved mobility. A horizontal incision was made high on the infundibulum of the gallbladder revealing impacted stones and frank pus within the gallbladder body. After removing the stones, there was immediate flow of bile appearing to emanate from a single, small lumen along the posterior aspect of the infundibulum, not in the typical location of the cystic duct. There was immediate concern for possible fistula between the chronically impacted infundibulum and the neighboring biliary tree. Laparoscopic ultrasound was used as an adjunct to try and clarify the anatomy of the portal structures but was unable to rule out a fistulous connection to the proximal biliary tree. The back wall of the gallbladder high above the orifice was dissected off the liver bed and divided at this level, allowing the placement of a rigid cholangiogram catheter into the infundibular orifice. A PDS EndoLoop (Ethicon) and Maryland grasper were used to close the edges of gallbladder wall around the catheter allowing for a cholangiogram to be performed (Fig. 1). This demonstrated initial rapid filling of the right hepatic duct followed by slower filling of the distal biliary tree and left hepatic duct, without a clear cystic duct and an apparent defect at the level of the catheter, as well as distal choledocholithiasis.

Glucagon 1 mg was then given and choledochoscopy was performed to achieve clearance of the choledocholithiasis and further characterize the presumed fistula. The choledochoscope was directed inferiorly through the gallbladder infundibulum, which led to a ductal structure that joined the CBD. Two impacted stones at the ampulla were removed with a basket. The choledochoscope was retracted to the first apparent ductal junction and another cholangiogram was performed at this level (Fig. 2), which demonstrated initial rapid filling of the CBD followed by filling of the left hepatic duct, indicating that the ductal structure entered through the inferior infundibulum was likely the distal right hepatic duct. This suggested that the ductal structure initially visualized in the posterior aspect of the infundibulum was the proximal right hepatic duct, and that the gallbladder infundibulum was in luminal continuity with the proximal and distal right hepatic duct (Fig. 3). The proximal ductal structure was also canulated with the choledochoscope as shown in Fig. 4.

As there was an adequate cuff of preserved infundibulum remaining, it became clear that this could be used to reconstitute a lumen and maintain continuity of the right hepatic duct. A wire was passed through the distal right hepatic duct via the infundibular remnant into the duodenum and positioning was confirmed under fluoroscopy. A 6-French (Fr) ureteral stent was then placed over the wire into the duodenum and back fed into the proximal right hepatic duct to ensure patency of our repair (Fig. 5, 6). The infundibular cuff was closed in a running fashion with 3-0 V-loc suture to restore ductal continuity. The remainder of the gallbladder fundus was removed from the liver bed in the usual fashion and hemostasis was achieved. A 10-Fr flat Jackson-Pratt (JP) drain was placed under the liver edge, next to the reconstituted bile duct. Ports were removed under direct visualization and the patient was closed, awoken, and brought to the recovery room in stable condition.

The patient progressed well postoperatively and her JP drain remained serosanguinous without evidence of bile leak. The drain was removed and she was discharged home on postoperative day 2. She was seen for routine surgical follow-up 1 month postoperatively and was doing well. At 10 months out, she was seen for planned esophagogastroduodenoscopy at which time the stent was noted to have migrated out of the biliary tree into the gastrointestinal tract and presumably eliminated through stool. Stent absence was confirmed by abdominal X-ray. At this time, she was noted to be doing well with no evidence of biliary or hepatic disease.

While first described in the early 1900s by Kehr, the constellation of symptoms secondary to hepatic duct obstruction by an impacted, infundibular gallstone gained recognition and its syndromic name following the work of Pablo Luis Mirizzi in 1948 [1]. Mirizzi syndrome is a rare complication of cholelithiasis with a reported incidence of 1.4% in patients undergoing cholecystectomy. Clinical presentation varies but often includes nonspecific symptoms such as RUQ pain (16.7%–100%), fever, nausea, vomiting, diarrhea, or constipation [2]. The syndrome classically involves four components including near parallel cystic and CHD; impaction of a stone in the cystic duct or gallbladder neck; mechanical obstruction of the CHD by the stone itself or by secondary inflammation; and intermittent or constant jaundice [3].

In cases of longstanding compression, Mirizzi syndrome can be complicated by abnormal fistulous connections between the gallbladder and surrounding structures including the biliary tree, bowel, or skin. The development of cholecystobiliary fistula has been explained by two mechanisms. The first mechanism proposes that chronic inflammation, secondary to the impacted gallstone, leads to the development of a pressure ulcer with erosion through the gallbladder wall into the bile duct wall [4]. The second mechanism involves progressive dilation of the cystic duct leading to shortening and contraction fibrosis of the cystic duct, ultimately yielding a large communication between the gallbladder and bile duct [1]. With either mechanism, the result is abnormal biliary anatomy which is delineated using the 1989 Csendes Classification. Csendes type I involves external compression of the bile duct without fistula; type II consists of a cholecystobiliary fistula involving less than 1/3 of the circumference of the bile duct; type III involves up to 2/3 of the circumference; and type IV involves complete destruction of the bile duct wall with fusion of the gallbladder and bile duct, with no recognizable dissection plane between the two structures [5].

At present, there are no internationally-accepted guidelines for the management of Mirizzi syndrome and the paucity of literature makes it difficult to propose an evidence-based standard of care [2]. Preoperative diagnosis, which occurs in less than 30% of cases [6], appears to be an important predictive factor for technical success. In a review, studies reporting a high preoperative diagnosis rate (>80%) had a significantly lower risk of conversion to open (p < 0.05), procedure-related complications (p < 0.05), or need for reoperation (p < 0.05) [7]. Thus, patients with high clinical suspicion of Mirizzi syndrome should undergo an intensive preoperative diagnostic workup including endoscopic retrograde cholangio-pancreatography/endoscopic ultrasonography and consideration of intraoperative cholangiography for further evaluation of patient anatomy.

While preoperative diagnosis of Mirizzi syndrome is preferred, intraoperative identification is imperative to prevent complications. In the urgent setting where optimal preoperative imaging is not feasible, strategies to assist with intraoperative diagnosis are necessary. Initial suspicion for Mirizzi syndrome should be raised when the surgeon is confronted with a contracted gallbladder with dense fibrous adhesions between the gallbladder and CBD. Laparoscopic ultrasound is a useful technique to assist with intraoperative delineation of biliary anatomy [8]. Intraoperative cholangiography and choledochoscopy may also be useful in this regard [2,9]. When adequate identification of biliary anatomy is not feasible, subtotal cholecystectomy with a fundus first approach has been shown to be a safe strategy and conversion to open cholecystectomy should always be considered in complicated Mirizzi syndrome [10].

In the minority of cases in which Mirizzi syndrome is diagnosed preoperatively, surgery is the preferred approach for treatment. In a review, laparoscopic treatment was successful in 59% of cases, and conversion to open was required in 41% with the most common reasons being severe adhesive disease and uncertain/abnormal biliary anatomy [7]. Given this, some authors propose directly proceeding with laparotomy to allow for better visualization and tactile evaluation of anatomy despite its more invasive nature and associated longer postoperative hospital stay.

The case presented here ultimately represents a type IV cholecystobiliary fistula to the right hepatic duct (Fig. 3) as opposed to the CHD, making it a variation on an already exceedingly rare presentation. In a review of 219 cases by Csendes et al. [5], type IV fistulas to the CHD represented only 4% of all cholecystobiliary fistulas, whereas type II and III represented 85% of cases. To our knowledge, very few cases of type IV Mirizzi syndromes have been reported over the last two decades, and no reports of fistulous connection to the right hepatic duct were identified.

The available literature regarding right hepatic duct repair and reconstruction is in the context of iatrogenic injury, often during planned cholecystectomy. No single strategy for management of isolated segmental or sectoral right hepatic bile duct injuries is universally accepted; however, most authors agree that injuries should be immediately repaired if recognized, in order to restore normal anatomy and function. In the event of iatrogenic injury, repair often requires end-to-end anastomosis over a T-tube versus Roux-en-Y hepaticojejunostomy. In contrast, the unique circumstances of this case allowed for the use of an infundibular cuff to repair the fistula and restore luminal continuity of the right hepatic duct.

In conclusion, Mirizzi syndrome is a rare clinical entity that requires careful consideration during preoperative workup and a high suspicion when abnormal anatomy is encountered intraoperatively.

Ethical statements

This study was conducted in compliance with the principles of the Declaration of Helsinki. It was deemed exempt from further review by the Colorado Multiple Institutional Review Board. Written informed consent was waived due to the deidentified, retrospective nature of this study.

Authors’ contributions

Conceptualization: CMS, MGH, CVH

Methodology, Analysis: CMS, MGH, CVH, FMP, MBS

Illustrations: MBS

Writing–original draft: CMS, MGH, CVH

Writing–review & editing: FMP, MBS

All authors read and approved the final manuscript.

Conflict of interest

All authors have no conflicts of interest to declare.

Funding/support

None.