Unusual complications following endoscopic ultrasound-guided hepaticogastrostomy: A call for refined techniques and dedicated devices

Abstract

We read with great interest the case report by Zhang et al describing unusual complications after endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS) in a patient with pancreatic cancer. The patient developed dark green urine, bile-colored pleural effusion, and ascites, suggesting bile leak and possible biliary-vascular fistula. We believe the complication profile is more consistent with transdiaphragmatic bile leakage during segment 2 puncture, as microperforation near the diaphragm can explain bilothorax. Device selection for fistula dilation and stent type may also have contributed. High pleural fluid-to-serum bilirubin ratios further support the diagnosis of bile leak, with urinary discoloration likely reflecting systemic absorption. This rare case highlights the need for refined techniques, careful puncture site selection, and dedicated devices to enhance safety in EUS-HGS.

Key Words: Adverse event; Biliary drainage; Bile leak; Endoscopic ultrasound; Bilothorax

Core Tip: Unusual complications following endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS) underscore the need for careful attention to technique and device choice. Although plastic stents are still used, fully covered self-expandable metal stents may better prevent bile leakage. Additionally, the choice of puncture site, such as segment 2 vs segment 3, and the method of fistula dilation critically influence outcomes. This editorial highlights the importance of refining EUS-HGS techniques and developing dedicated devices to enhance safety and reduce unexpected complications.

TO THE EDITOR

I read with great interest the case report by Zhang et al[1], which described the unusual complications of endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS) in a patient with pancreatic cancer presenting with recurrent dark green urine following a circadian pattern accompanied with green pleural effusion, and subsequent ascites—findings that raised suspicion of a biliary-vascular fistula and bile leak. This case highlights the unpredictable nature of complications following EUS-HGS, emphasizing the importance of meticulous procedural technique and close postoperative monitoring.

The presence of bile-colored pleural effusion and ascites strongly indicates a significant bile leak during or after the procedure. Although a biliary-vascular fistula might explain the dark green urine, bile leak into the peritoneal or pleural cavity during fistula dilation or stent placement appears to be the more plausible explanation[2,3].

Providing the specifics of the device used for fistula dilation would have been informative, given that the dilation approach can impact the rate of complications in EUS-guided biliary drainage. The needle-knife dilation generates an axial force that can lead to unintended incisions or tissue injury, increasing the risk of bile leaks and pneumoperitoneum[2]. On the contrary, electrocautery dilators transmit energy along a linear vector, thereby minimizing the risk of unintended incisions and collateral tissue injury. Balloon dilators generate a uniform radial force, enabling stable tract dilation while minimizing tissue separation, and can provide adequate dilation to accommodate the delivery system[3]. Spiral screw-type dilators, although offering benefits comparable to electrocautery, frequently require subsequent balloon dilation, thereby extending the procedure and increasing the risk of bile leak[4]. Altogether, these aspects highlight the importance of selecting the appropriate dilation device to minimize complications and optimize favorable outcomes in EUS-HGS.

Although a plastic stent was placed for EUS-HGS in the patient reported by Zhang et al[1], fully covered self-expandable metal stents (FCSEMS) are preferred in EUS-HGS, due to their potential in sealing the tract immediately, thereby reducing the risk of bile peritonitis and pneumoperitoneum[5]. Moreover, revision failure is encountered more frequently with plastic stents than with FCSEMS[2]. Conversely, although FCSEMS prevent bile leakage and maintain fistula patency, they also require additional fistula dilation because of their thicker delivery system and a more precise and cautious deployment of the stent, owing to the risk of intraperitoneal migration[5]. A recent study demonstrated that plastic stents have a good safety profile and comparable patency for up to 100 days, suggesting they remain a viable option in selected cases[6].

The intrahepatic puncture was performed in segment 2 (B2), as shown in figure 2 of the report[1]. Puncture of segment 3 is preferred because this approach improves the visualization of the stent tip and reduces the risk of serious complications such as mediastinitis and diaphragm injury[2]. Herein, although the stent was positioned within the gastric lumen near the high body–cardia, the B2 approach still carries the risk of traversing the diaphragm, because the trajectory from the gastric cardia to B2 runs in close proximity to the left hemidiaphragm. Even without mediastinal stent mispositioning, microperforation of the diaphragm during B2 access could explain the post-procedural bilious pleural effusion. Thus, the complication profile of this case has more resemblance with that observed in transdiaphragmatic bile leak than with primary vascular shunting of bile.

In addition to the choice of the dilating device and stent type, several technical factors impact the success of EUS-HGS. The puncture site is crucial, as targeting the intrahepatic duct running from the upper left to the lower right on EUS imaging facilitates successful advancement of the guidewire into the hepatic hilum. Furthermore, ensuring the intervention to an adequate volume of hepatic parenchyma is essential to reduce the risk of stent migration. The needle type and guidewire characteristics are equally crucial. A 19 gauge fine-needle aspiration needle with a Menghini-type tip and guidewires with sufficient stiffness and torque control are preferred for EUS-HGS[7]. In centers with limited EUS-HGS experience, EUS-guided choledochoduodenostomy using dedicated devices and the EUS-guided rendezvous technique may serve as effective and safe alternatives for managing biliary obstructions[8,9].

Although the serum bilirubin levels dramatically declined after EUS-HGS, bile leakage into the pleural and peritoneal cavities resulted in third-space accumulation of bilirubin-rich fluid, mimicking spontaneous biliary peritonitis characterized by dark yellow–green ascites[10]. Bile that accumulates in extrabiliary spaces can enter the systemic circulation via lymphatic and venous absorption. This condition can be diagnosed by a peritoneal fluid-to-serum bilirubin ratio greater than 3.25 in cases of bile leak[11]. In the present case, the bilirubin level was 326.6 μmol/L in the pleural effusion, yielding fluid-to-serum ratios of 2.36 and 4.16, consistent with bilothorax and bile leak. This mechanism may also partially explain the paradoxical urinary discoloration observed despite declining serum bilirubin levels. The circadian variation is likely attributable to hydration-dependent changes in urinary concentration, independent of the presence of a vascular fistula.

A step-up approach is recommended in patients who develop bile leakage or bilothorax following EUS-HGS. Initial management includes pleural or peritoneal drainage with antibiotics, followed by stent optimization, such as replacing the plastic stent with an FCSEMS—or adding biliary decompression via percutaneous transhepatic biliary drainage or endoscopic retrograde cholangiopancreatography. In cases where rescue is necessary after plastic stent placement, caution is warranted during the exchange of the original stent with a large-diameter FCSEMS, as excessive tract dilation can paradoxically aggravate bile leakage. In such cases, a slim 6-mm FCSEMS with a smaller delivery system can be advantageous, effectively sealing the tract while minimizing further tissue disruption. If these measures fail, surgical repair may be necessary to address a large bilopleural fistula or diaphragmatic defect; however, over 90% of patients improve with drainage and biliary decompression alone, without requiring surgical interventions. Overall, the prognosis of bilothorax is favorable in most cases managed successfully with drainage and biliary decompression; however, outcomes may be worse in patients with advanced hepatobiliary malignancy[12].

This rare case highlights the ongoing challenges and unexpected risks of EUS-HGS, despite the high technical and clinical success rates of this procedure as an alternative to percutaneous biliary drainage. Future development of dedicated, one-step EUS-HGS devices can aid in standardizing the procedure, shortening procedure time, and reducing complications, ultimately improving patient safety and outcomes.