Biliopleural fistula (BF) is an uncommon complication that may occur after hepatobiliary surgery or traumatic injury to both the diaphragm and the liver [1]. Few cases of BF have been reported in living donor liver recipients. Pleural biliary effusion poses a high risk of severe complications, particularly in immunocompromised patients, such as liver recipients who are on immunosuppressants. If unrecognized, BF may lead to pneumonia, respiratory insufficiency, empyema, and the life-threatening condition known as acute respiratory distress syndrome [2]. Here, we report a clinical case of successful treatment of a BF following adult-to-adult left lobe living donor liver transplantation (LDLT).

This case study was conducted according to the Declaration of Helsinki. The case report was approved by the Local Bioethical Committee of National Research Oncology Center (No. 03-006/01). Written informed consent was obtained from the patient for publication of the details of their medical case and any accompanying images at the time of admission to the National Research Oncology Center.

A 61-year-old female with a clinical history of liver cirrhosis due to antimitochondrial antibody (AMA-2) positive primary sclerosing cholangitis (PSC) was referred to our hospital in December 2023 for LT. The initial clinical manifestations, which included pruritus and jaundice, appeared in 2021. Acute surgical conditions were ruled out, and the patient was referred to a gastroenterology specialist. Further workup confirmed AMA-2 positivity, and a liver biopsy revealed histopathological features consistent with PSC. Despite treatment with ursodeoxycholic acid and steroids, the disease progressed to liver cirrhosis by 2022. Liver function continued to deteriorate, leading to a referral to a transplant center in 2023. The patient had multiple comorbidities, including diabetes mellitus and its complications. She underwent a LDLT from her 36-year-old son. The transplant included the left lobe of the donor’s liver, featuring a single duct, single left hepatic artery, and single portal vein; biliary reconstruction was performed using a single duct-to-duct anastomosis (DDA) from the donor's left duct to the right lobe duct of the recipient. On postoperative day (POD) 5, bile leakage was detected, as the ascitic fluid from the abdominal drainage was mixed with bile. The drainage volume gradually increased, and the patient developed symptoms of clay-colored stools. Computed tomography (CT) scans of the abdomen and chest on POD 10 showed no pleural effusion and a normal radiological appearance of the lungs (Fig. 1A), while CT scans of the abdomen revealed a subdiaphragmatic bile accumulation, or biloma (Fig. 1B). The patient was diagnosed with DDA anastomotic leakage. Due to the absence of an experienced endoscopy specialist, it was impossible to perform endoscopic retrograde cholangiopancreatography (ERCP) with intrabiliary stenting. Instead, surgery was chosen to evacuate the biloma and close the biliary anastomosis defect. On POD 11, a relaparotomy was performed, and the biloma was evacuated. During the surgery, leakage at the DDA site was detected, although the DDA was not completely disrupted. An attempt to close the defect with two interrupted sutures was made; however, due to inflammation, repeated leakage was suspected, and the common bile duct was drained with an external stent. In the postoperative period, the bile output through the external bile duct showed no signs of decreasing (700 mL per day). The patient's condition worsened due to a lack of bile acids and subsequent nutritional deficiency. Blood glucose levels continued to rise despite insulin therapy, and tacrolimus was switched to cyclosporine. On POD 2 after the relaparotomy, lymphorrhea was detected in the abdominal drain. It was decided to start total parenteral nutrition (TPN); after its initiation, blood glucose levels were difficult to control, necessitating the start of a continuous intravenous insulin infusion. Lymphorrhea was treated after 1 week of TPN. Although hyperglycemia became manageable and lymphorrhea ceased, bile output through the external biliary drain (EBD) remained unchanged. DDA anastomotic stricture was suspected, and ultrasonography (US) showed dilated biliary ducts of the transplanted liver. On POD 28 after the second operation, endoscopic retrograde cholangiography was performed, revealing bile leakage and anastomotic stricture (Fig. 2). Endoscopic sphincterotomy was performed, and a control cholangiogram confirmed the presence of contrast in the duodenum. In the postoperative period, the patient's stool became colored, and the bile output through the EBD significantly decreased. On POD 2 after the ERCP, the patient experienced tachypnea and pain in the right side of the chest cavity. CT revealed a pleural effusion in the right pleural cavity (Fig. 3), for which a pleural pigtail was inserted, and fluid was aspirated; the fluid's bilirubin level was over 200 mmol/L. The patient was diagnosed with a BF (Fig. 4). Antibacterial therapy was prescribed according to bacterial culture results. Daily thoracostomy drain output gradually decreased and declined to zero on day 15 after pigtail insertion, along with normalization of the biliary ducts' diameter on US. Fistulography through the EBD showed a small remnant low-volume biloma, persistent radiological picture of biliary stricture, and nondilated graft bile ducts; the pleural effusion resolved (Fig. 5). The patient was discharged on POD 70 after LT with normal functional tests of the liver graft; the EBD was not removed due to the high risk of repeated development of anastomotic stricture. Laboratory test results, performed procedures, and operations during inpatient admission are shown in Supplementary Table 1. The EBD was removed 1 month after discharge. Two months after discharge, the patient was admitted due to hyperbilirubinemia. Percutaneous transhepatic biliary drainage (PTBD) was performed (Fig. 6). Three months later, the PTBD was removed. The biliary stricture resolved without intrabiliary stenting. During 9 months of follow-up, the patient showed no signs of repeated biliary leakage or any other complications.

Bilothorax was first described in 1897 by Graham [3], and since then, only a limited number of papers have addressed this rare complication. Biliary pleural effusion is often caused by iatrogenic damage following major hepatectomy, thoracic and abdominal trauma, or percutaneous transhepatic or gallbladder drainage. BF has been reported in only a few cases following LT. According to a systematic literature review by Acharya et al. [4], the most common cause of bilothorax was PTBD, followed by traumatic injuries such as gunshot or stab wounds to the right chest and abdomen. Among the 123 studies reviewed, bilothorax was attributed to LT in six cases, all of which involved patients who received a liver graft from a deceased donor [4]. In 2023, Kazemi et al. [5] reported a case of BF following an adult-to-child LDLT. In all the case reports mentioned above, either bile leakage alone or in association with early biliary stricture preceded the development of bilothorax.

LDLT is regarded as an effective tool in addressing the disparity between organ demand and the available organ pool. LDLT offers several advantages over LT from deceased donors, including a shorter cold ischemic time, the ability to schedule surgery electively, and the opportunity to obtain a liver before severe decompensation occurs. However, the complication rate following LDLT is higher and may present more challenges in management compared to deceased donor LT [6].

In particular, biliary complications are notably more frequent in patients who have received a partial graft, with occurrence rates ranging from 15% to 25% [7]. From 2014 to 2024, our center performed 105 LTs. The overall incidence of bile leakage among liver recipients at our center was 32.5%. Consequently, patients who received a liver graft with complex biliary anatomy are at a significantly higher risk of early bile leakage. Despite our experience in treating and diagnosing biliary complications, BF was observed in only one recipient in our case series.

CT is the most common diagnostic tool for detecting bilothorax. However, bilothorax is often underdiagnosed due to the high incidence of right-sided or bilateral pleural effusion in the early postoperative period following LT, particularly after partial graft transplantation [8]. Pleural effusions are typically caused by low total protein and albumin levels, as well as impaired renal function. These conditions are common after LDLT, especially if the graft-to-recipient-weight ratio is less than 0.8 or if the recipient's preoperative state is complicated by hepatorenal syndrome and a high Model for End-Stage Liver Disease score. Pleural effusions occur in 39% to 95% of cases post-LT [9]. Most pleural effusions resolve within the first weeks after LT and do not require intervention. At our center, liver recipients in the early postoperative period with low-volume pleuritis (under 400 mL) rarely undergo chest tube drainage or active pleural effusion aspiration due to coagulopathy and the subsequent risk of intrathoracic bleeding and infection. In most cases, pleuritis resolves once the liver graft's synthetic function normalizes. In the case presented, right-sided pleuritis was also observed; however, drainage was not indicated due to the absence of respiratory dysfunction. Retrospectively, it is difficult to determine whether the pleuritis was initially caused by bile leakage detected on POD 5 and only clinically manifested when BF was diagnosed on POD 30.

Kazemi et al. [5] proposed several theories regarding the formation of BF, including biliary hypertension that leads to bile flowing into low-pressure spaces. Another theory suggests the accumulation of bile followed by inflammation and subsequent erosion through the diaphragm. Less plausible are the theories that describe the transmission of bile through pleuroperitoneal pores in the diaphragm, and the theory of diaphragmatic injury during surgery, which is considered less likely due to the absence of pneumothorax, easily detectable in the early postoperative period [10,11]. In our case, we suggest that the most likely cause of the biliopleural fistula (BF) was early and persistent bile leakage, influenced by several patient-associated risk factors, particularly preexisting diabetes mellitus, older age, and a long history of cirrhosis. The aforementioned patient characteristics were risk factors for bile leakage due to compromised synthetic and recovery capabilities, which are crucial for anastomosis sufficiency. Bile leakage and the accumulation of bile in the right subdiaphragmatic region led to inflammation of the nearby tissue and eventually the rupture of the collected bile into the right pleural cavity.

As BF is a rare complication in liver transplant recipients, there are no widely accepted treatment protocols. BF is more likely to occur following thoracoabdominal trauma. The majority of the reviewed papers recommend conservative treatment for patients with posttraumatic bilothorax. Surgical repair of the fistula was necessary for only 10% of patients who did not respond to conservative therapy. If the conservative approach fails, surgical treatment is indicated. The surgical procedures should be carefully tailored to each individual case. In our case, the bile leakage and biloma evacuation, as well as biliary hypertension due to early biliary stricture, were managed with open surgery (biloma evacuation and external biliary drainage) followed by ERCP with sphincterotomy. Right chest drainage was performed in response to respiratory symptoms and pain.

If nonsurgical treatment of BF fails, the preferred surgical approach in our case would be an abdominal one, incorporating radical management of bile leakage (T-tube placement, external cholangiostomy, or hepaticojejunostomy) along with closure of the fistula. In cases of severe lung involvement (trapped lungs), a combined abdominal and thoracic approach may be necessary, which would include lung decortication.

Although rare, BF can significantly impact liver transplant outcomes due to the immunocompromised state of the recipient and the high risk of infectious complications. Therefore, pleural effusions that do not resolve more than 4 weeks after surgery should be considered as potential BFs, especially in patients with a history of early bile leakage and biliary stricture. Draining the chest and measuring the bilirubin level in the pleural fluid may be the best diagnostic approach.