Off-label use of a WRAPSODY^TM stent-graft in the portal vein for septic gas embolism following pancreatic abscess fistulization: a case report

Introduction

Portal venous gas embolism is an exceedingly rare and life-threatening condition, most commonly associated with bowel ischemia or severe intra-abdominal infection (1). In even more exceptional cases, it may arise secondary to pancreatic abscess fistulization, posing significant management challenges, particularly in patients recovering from complex pancreatic surgery (2). Pancreatic inflammatory collections, including pseudocysts and abscess, are known to exhibit aggressive local behavior. Ongoing enzymatic activity and chronic inflammation may lead to progressive tissue destruction and erosion into adjacent organs or vascular structures. Although uncommon, fistulous communication with major vessels such as the portal or mesenteric veins can occur and may result in life-threatening hemorrhagic or septic complications (3). These scenarios are further complicated by concurrent portal or mesenteric vein thrombosis, which can compromise hepatic perfusion and worsen systemic sepsis (4,5). Awareness of this erosive potential is therefore essential when managing complex postoperative pancreatic conditions.

While surgical intervention may be considered, it carries substantial morbidity and mortality, especially in recently operated or hemodynamically unstable patients (1). Endovascular approaches to the portal vein remain uncommon, largely owing to anatomical challenges and the nascent stage of devicebased interventions in this vascular territory (6).

The WRAPSODY^TM stent-graft, is a self-expanding nitinol endoprosthesis originally developed for the treatment of stenosis and occlusion in hemodialysis access circuits. It incorporates a proprietary trilaminar membrane (PTFE) designed to provide complete luminal coverage, resistance to cellular ingrowth, and reduced permeability to fluids. In dialysis access applications, clinical studies have demonstrated favorable primary patency rates and durability compared with conventional angioplasty, with reduced neointimal hyperplasia attributed to its multilayer PTFE design and uniform radial force distribution. Reported advantages include high flexibility allowing adaptation to tortuous vascular anatomy, effective sealing capability due to full PTFE coverage, and surface characteristics associated with relatively low thrombogenicity.

However, its use in the portal venous system remains off label. No dedicated data exists regarding long-term patency, infection susceptibility, or performance in septic venous environments. Furthermore, portal venous hemodynamics differ substantially from dialysis access circuits, limiting direct extrapolation of existing patency data. A focused literature search was conducted using PubMed and major indexed databases to identify reports of WRAPSODY^TM stent-graft in the portal venous system or other systemic veins. Based on this review of indexed literature, no prior reports of implantation in the human portal circulation were identified.

The objective of this report is to describe the technical considerations and short-term clinical outcome of the portal venous reconstruction using an off-label WRAPSODY^TM stent-graft in the setting of septic portal venous gas embolism. We present this article in accordance with the CARE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2560/rc).

Case presentation

A 58-year-old man with chronic alcohol induced pancreatitis had a long-standing history of pancreatic pseudocysts requiring prior endoscopic interventions, including transgastric cyst drainage and endosonographic puncture with metal stent placement, followed by exchange to double-pigtail plastic stents. Due to progressive ductal stenosis of the pancreatic corpus, recurrent pseudocyst formation, and therapy-refractory pain despite prior endoscopic management, he underwent a subtotal distal pancreatectomy with splenectomy and resection of a pancreatic-gastric fistula with gastric wall closure. Postoperative day (POD) 16, he presented to our emergency department with rising fever, severe upper abdominal discomfort, and elevated inflammatory markers [C-reactive protein (CRP) 112 mg/L; leukocytes 13,000/µL]. Contrast-enhanced multidetector computed tomography (MDCT) of the abdomen imaging revealed a postoperative complex peripancreatic fluid collection with multiple intralesional gas near the resection margin, suggestive of a pancreatic fistula with secondary abscess formation. A high-attenuation, wall-adherent thrombus was detected extending along the portal vein into the proximal superior mesenteric vein (SMV) (see Figure 1A-1D).

Figure 1 Initial contrast-enhanced abdominal CT. (A) Axial image demonstrating a postoperative peripancreatic fluid collection at the resection margin with intralesional gas, consistent with a pancreatic abscess. (B) Axial image showing wall-adherent thrombus within the distal superior mesenteric vein extending into the proximal portal vein. (C) Coronal reconstruction demonstrating partial occlusion of the portal vein confluence. (D) Coronal reconstruction showing inflammatory changes extending into the surrounding mesenteric fat; a residual splenic artery stump is also visible. CT, computed tomography.

In response to these findings and the patient’s worsening clinical condition, the gastroenterology team performed an emergency endoscopic placement of a transgastric lumen-apposing metal stent (LAMS) on POD 18 to decompress the abscess in the bursa omentalis. Broad-spectrum intravenous antibiotics with piperacillin/tazobactam (Tazobac^®, Pfizer Pharma GmbH, Berlin, Germany) was administrated at a dose of 4.5 g every 8 hours for 13 days. Therapeutic anticoagulation with low-molecular-weight heparin (enoxaparin, Clexane^®, Paris, France) was initiated at a full therapeutic dose of 1 mg/kg body weight subcutaneously twice daily (approximately 80 mg twice daily) due to concern for evolving portal and mesenteric vein thrombosis.

However, on POD 21, despite these measures, the patient’s condition remained stagnant. Follow-up esophagogastroduodenoscopy (EGD) revealed copious brown, purulent fluid in the stomach (see Figure 2A-2D). MDCT demonstrated persistent and now increasing intraluminal gas within the central portal vein and intrahepatic branches, consistent with septic air embolism (see Figure 3A-3D).

Figure 2 Endoscopic and fluoroscopic images demonstrating LAMS placement and follow-up. (A) Fluoroscopic image showing transgastric deployment of a LAMS into the peripancreatic abscess cavity within the lesser sac. (B) Endoscopic view confirming correct positioning of the LAMS with visualization of the stent lumen. (C) Endoscopic image demonstrating drainage of purulent fluid through the stent following decompression of the abscess cavity. (D) Endoscopic view showing ulcerated and friable gastric mucosa overlying the collapsed cavity. LAMS, lumen-apposing metal stent.

Figure 3 Contrast-enhanced MDCT demonstrating septic portal venous gas embolism and associated portal vein thrombosis. (A) Axial image demonstrating low-attenuation mural thrombus along the wall of the main portal vein, adjacent to a peripancreatic abscess cavity with imaging features suggestive of fistulous communication with the portal venous system. (B) Axial image demonstrating persistent portal venous gas extending into the portal vein confluence and proximal superior mesenteric vein. (C) Coronal reconstruction illustrating branching gas within the intrahepatic portal venous system, extending toward peripheral hepatic segments. (D) Coronal image demonstrating the previously placed LAMS in situ, with adjacent intrahepatic portal venous gas consistent with ongoing septic portal venous embolism. LAMS, lumen-apposing metal stent; MDCT, multidetector computed tomography.

At this point, progression of thrombus burden and high risk of further gas emboli were considered life-threatening. After interdisciplinary discussion, it was determined that a transhepatic endovascular intervention with stent-graft placement was the most feasible option, given the inaccessibility of the portal confluence surgically and the patient’s critical condition.

All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. Written informed consent was obtained from the patient for the publication of this article and accompanying images. A copy of the written consent is available for review by the editorial office of this journal.

Procedure

The procedure was performed under general anesthesia in a hybrid interventional suite. Under general anesthesia, percutaneous transhepatic access to the portal venous system was achieved under ultrasound and fluoroscopic guidance using a 22G needle (Neff Percutaneous Access Set^TM, Cook Medical, Bloomington, IN, USA). After successful puncture of a left portal vein branch, a 0.018-inch guidewire was advanced to facilitate placement of a 6F sheath. Portography confirmed opacification of the intrahepatic portal vein and visualized multiple wall irregularities of the main portal vein trunk with adherent thrombi and collateral formation.

Using a 0.035-inch hydrophilic guidewire (Radifocus Guide Wire M^TM, Terumo, Tokyo, Japan) and a 4F, 65 cm angled catheter (Glidecath^®, Terumo), selective catheterization of the portal vein trunk and SMV was performed. Given the complex anatomy and thrombotic burden, the hydrophilic guidewire was exchanged for a 0.035-inch super stiff guidewire (Amplatz Super Stiff^TM, Boston Scientific, Marlborough, MA, USA) to support device delivery. After serial dilatation of the parenchymal tract, a 12F sheath was carefully advanced into the portal vein.

Pre-procedural MDCT was carefully reviewed to assess portal vein diameter and landing zones. The main portal vein measured approximately 11–12 mm in diameter proximal and distal to the fistulous segment. A 12 mm diameter stent-graft was therefore selected to achieve adequate wall apposition without excessive oversizing, which could increase the risk of vessel injury in the inflamed and friable venous wall. The stent length (40 mm) was chosen to ensure complete coverages of the fistulous communication while preserving flow in the SMV. A 12 mm × 40 mm self-expanding stent graft (WRAPSODY^TM, Merit Medical, South Jordan, UT, USA) was deployed across the eroded segment of the portal vein trunk, achieving adequate expansion with minimal waist formation. Pre-dilatation with balloon angioplasty was avoided. Given the presence of septic inflammation, mural thrombosis and suspected vessel wall erosion, ballon expansion prior to stent-graft deployment was considered to carry an increased risk of thrombus dislodgement, septic embolization, and potential venous rupture. Direct deployment of the covered stent-graft was therefore preferred to achieve immediate sealing of the fistulous tract. Despite adequate initial expansion, residual mural thrombus and extensive wall irregularities at the proximal landing zone were observed. To optimize proximal sealing and increase radial force in this segment, an additional 14 mm × 60 mm stent graft (BeGraft Aortic^®, Bentley InnoMed, Hechingen, Germany) was placed in an overlapping fashion to extend the coverage (see Figure 4A-4F).

Figure 4 Transhepatic endovascular reconstruction of the portal venous confluence. (A) Fluoroscopic image demonstrating ultrasound-guided puncture of the left portal vein using a Neff^TM access needle. (B) Advancement of a 5F sheath into the portal confluence; contrast injection reveals mural thrombus and wall irregularity within the main portal vein and SMV. (C) Deployment of a self-expanding covered stent-graft (WRAPSODY^TM, Merit Medical, South Jordan, UT, USA) across the affected portal venous segment. (D) Fluoroscopic image showing mild waist formation of the stent at the level of prior surgical alteration. (E) Placement of an overlapping balloon-expandable covered stent (BeGraft^TM, Bentley InnoMed, Hechingen, Germany) to optimize proximal sealing and radial force. (F) Final DSA demonstrating restored portal venous flow and complete exclusion of the fistulous communication without residual filling defect. DSA, digital subtraction angiography; SMV, superior mesenteric vein.

Completion portography demonstrated improved flow through the stent grafts with resolution of venous congestion and satisfactory contrast filling of the intrahepatic and extrahepatic portal branches. Given the transhepatic access route into the high-flow portal system and the need for continued therapeutic anticoagulation, closure of the hepatic puncture tract was considered mandatory to minimize the risk of post-procedural hemorrhage. Embolization of the transhepatic access tract is widely regarded as standard practice following portal vein interventions to prevent intraperitoneal bleeding and hemoperitoneum, particularly when large-caliber sheaths are used.

In this case, staged tract closure was performed. Initially, a 6 mm vascular plug (Amplatzer^TM Vascular Plug IV, Abbott, Abbott Park, IL, USA) was deployed within the hepatic puncture tract. However, persistent contrast pooling prompted insertion of a second, larger 8 mm vascular plug (Amplatzer^TM Vascular Plug IV, Abbott) more proximally. To secure final hemostasis, a 1:2 mixture of Histoacryl^TM (n-butyl cyanoacrylate) and Lipiodol^TM was carefully injected during sheath removal to seal the puncture tract entirely. Manual compression was applied for 20 minutes and sealed with a sterile dressing. Final fluoroscopy confirmed absence of contrast extravasation and no residual intrahepatic gas (see Figure 5).

Figure 5 Embolization of the transhepatic access tract. (A) Fluoroscopic image demonstrating controlled withdrawal of the transhepatic access sheath with deployment of a vascular plug within the puncture tract. (B) Final fluoroscopic image showing complete tract embolization using two Amplatzer^TM Vascular Plugs (6 and 8 mm; Abbott, Abbott Park, IL, USA) followed by injection of a Lipiodol^® (Guerbet, Villepinte, France) and Histoacryl^® (B. Braun, Melsungen, Germany) mixture (1:2), achieving complete tract sealing.

Outcome

The procedure was technically successful, with brisk hepatoportal flow through the portal vein and both endoprostheses well-positioned. The patient remained hemodynamically stable and afebrile. Post-procedurally, therapeutic anticoagulation with low-molecular-weight heparin (LMWH) (enoxaparin, Clexane^®, Paris, France) was resumed and subsequently transitioned to oral anticoagulation with rivaroxaban (Xarelto^®, Bayer AG, Leverkusen, Germany) as a dose of 20 mg once daily. Oral anticoagulation was continued for a planned duration of 3 months to reduce the risk of stent thrombosis and promote portal vein recanalization. Intravenous antibiotic therapy with piperacillin/tazobactam (Tazobac^®, Pfizer Pharma GmbH, Berlin, Germany) was continued for a total duration of 13 days, as documented in the surgical discharge summary. No additional oral antibiotic step-down therapy was administered. The transgastric LAMS was removed on POD 31 (10 days after the initial implantation) following resolution of abscess drainage. Subsequent follow-up imaging at 6 weeks demonstrated complete resolution of intrahepatic gas, patent portal vein stents, and no signs of restenosis or thrombotic progression (Figure 6A,6B). Additionally, a small subcapsular liver hematoma noted post-intervention had regressed spontaneously, with no need for drainage. At 3-month clinical follow-up, the patient remained asymptomatic with normalized inflammatory markers and stable function tests.

Figure 6 Six-week follow-up contrast-enhanced MDCT. (A) Axial image demonstrating preserved patency of the portal vein stent-graft with contrast opacification of the lumen. A small subcapsular hepatic hematoma is noted adjacent to the transhepatic access tract. (B) Coronal reconstruction confirming adequate expansion and patency of the portal vein stent-grafts without evidence of occlusion. MDCT, multidetector computed tomography.

Timeline of clinical course

The patient underwent subtotal distal pancreatectomy with splenectomy for chronic pancreatitis and recurrent pseudocysts. On POD 16, he presented with fever and abdominal pain; imaging demonstrated a pancreatic abscess with portal and SMV thrombosis. Therapeutic anticoagulation and intravenous antibiotics were initiated. Persistent portal venous gas prompted endoscopic drainage followed by percutaneous transhepatic portal vein stent-graft implantation. Six-week follow-up MDCT confirmed stent patency and resolution of portal venous gas. Oral anticoagulation was continued for 3 months with stable clinical recovery.

Discussion

Septic gas within the portal venous system represents an unusual and critical complication, often indicating direct communication between infected or necrotic tissue and the portal circulation. These conditions are associated with a mortality rate that may exceed 75% when linked to bowel necrosis, intra-abdominal sepsis, or mesenteric ischemia (7). Management strategies depend on severity, etiology, and overall clinical status. In stable patients without signs of bowel necrosis or uncontrolled sepsis, conservative management, including antibiotics, supportive care and close monitoring, may be considered (8). Surgical intervention (resection, drainage, fistula repair) in the context of bowel necrosis, perforation or abscess and uncontrolled sepsis carries high morbidity, especially in the immediately postoperative or hemodynamically unstable patient (9). Endovascular approaches, including stenting, thrombectomy, thrombolysis, and embolization, are increasingly used for managing portal vein thrombosis, fistulas, and post-surgical complications, particularly in patients at high surgical risk. These interventions often involve the off-label use of peripheral stent-grafts and catheters not specifically designed for the portal venous system. Although technically feasible, these procedures carry notable risks such as bleeding, stent thrombosis, or access-related complications (10,11).

In our case conservative management was deemed insufficient on POD 21 due to persistent gas embolization and a fistulous tract between the abscess cavity and the portal venous confluence. Although surgical repair typically involves resection of necrotic or perforated bowel segments and correction of the underlying source, it was contraindicated here due to recent pancreatic surgery, ongoing sepsis, and overall postoperative vulnerability. In such settings, alternative minimally invasive therapies including interventional radiology procedures may offer safer and more feasible solutions (10,11).

While endovascular treatment of the portal vein remains technically challenging and infrequently reported, its use has been described in select cases, including post-transplant stenosis, hemorrhage, and malignancy-related obstruction. Devices such as FLUENCY plus^TM, Viabahn^TM, or iCAST^TM covered stents have been used off-label to provide luminal scaffolding and lesion exclusion in venous interventions (12,13). While several covered stents have been used off-labell in venours reconstructions, their mechanical properties differ. Devices such as Viabahn^TM and FLUENCY^TM are primarily designed for arterial applications and may exhibit greater longitudinal stiffness. In the present case, the portal venous confluence demonstrated angulation and irregular vessel contours related to postoperative changes and inflammatory remodeling. A device with high conformability and uniform radical force distribution was therefore considered advantageous to achieve adequate wall apposition without inducing vessel straightening or excessive focal stress. The multilayer PTFE design and flexibility of the WRAPSODY^TM stent-graft were deemed particularly suitable for this anatomical configuration.

As previously described, fistulous communication was the primary source of portal venous gas embolism in our patient. Therefore, the principal therapeutic objective was to seal this abnormal connection, thereby preventing further entry of gas or infected material into the portal circulation. This required a covered stent-graft, as only a nonporous device could provide immediate and durable exclusion of the affected segment.

Implanting a covered stent-graft in the setting of active infection inevitably raises concerns about potential device infection. In our case, however, the septic focus was localized to a pancreatic abscess with a defined fistulous communication to the portal vein. Endoscopic drainage provided initial source control, and subsequent exclusion of the fistula by the covered stent prevented further contamination of the portal circulation. Broad-spectrum intravenous antibiotics were continued during the acute phase. No clinical or radiological signs of graft infection were observed during follow-up.

Anticoagulation after portal vein stent placement was considered necessary because mural thrombus was already present and the septic inflammatory state further increased the risk of thrombosis. Although there are no established guidelines specifically for portal vein stent-grafts, anticoagulation is commonly used in portal vein thrombosis and venous stenting to reduce early occlusion. In this case, anticoagulation was continued for 3 months, balancing the risks of thrombosis and bleeding.

The WRAPSODY^TM stent-graft was selected, despite its off-label status, based on its favorable characteristics. It features a fully covered, multilayer PTFE membrane, impermeable to fluid and cellular ingrowth, and a flexible nitinol frame, allowing for excellent conformability to the angulated anatomy of the portal confluence. Its design also minimizes neointimal hyperplasia and thrombus formation, which is particularly advantageous in a septic and prothrombotic environment.

A focused review of indexed literature did not identify prior reports of WRAPSODY^TM stent-graft implantation in the portal circulation. However, given the limitations of literature indexing and reporting, the absence of published cases cannot definitively exclude unreported or non-indexed experiences. Further limitations are the single-case nature, short follow-up duration, unknown long-term patency and infection risk.

Conclusions

This case demonstrates that endovascular reconstruction of the portal vein using a WRAPSODY^TM stent-graft is technically feasible and potentially lifesaving in selected cases of septic portal venous gas embolism. In situations where surgical or conservative approaches are contraindicated or ineffective, the off-label use of flexible, low-thrombogenicity peripheral stent-grafts may offer a safe and effective minimally invasive alternative.

Acknowledgments

The authors would like to thank Muhammad Haris (King Edward Medical University) for his valuable assistance and support in the preparation of this manuscript. For the publication fee, we acknowledge financial support by Deutsche Forschungsgemein-schaft within the funding program “Open Access Publikationskosten” as well as by Heidelberg University.

Footnote

Reporting Checklist: The authors have completed the CARE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2560/rc

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2560/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. Written informed consent was obtained from the patient for the publication of this article and accompanying images. A copy of the written consent is available for review by the editorial office of this journal.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Abboud B, El Hachem J, Yazbeck T, Doumit C. Hepatic portal venous gas: physiopathology, etiology, prognosis and treatment. World J Gastroenterol 2009;15:3585-90. [Crossref] [PubMed]
2. Duncheskie R, Anjum F. Pancreatic fistula. Treasure Island, FL, USA: StatPearls Publishing; 2025.
3. Muhammad Zohaib Uddin M, Memon WA, Shahid M, Eltaly H, Akram S, Virji SN, Ahmad MN, Nadeem N, Khan F, Zafar U. A Case of Perinephric Pancreatic Pseudocyst Secondary to Pancreatitis. Clin Case Rep 2025;13:e70537. [Crossref] [PubMed]
4. Misbahuddin-Leis M, Ademi B, Ankolvi M, Dubasz K, Mishra M, Riechmann M, Graeb C, Radeleff B. Successful heparin-perfusion therapy for complete thrombosis of the intra- and extrahepatic portal and mesenteric vein. A case report and literature review. Radiol Case Rep 2022;18:850-5. [Crossref] [PubMed]
5. Makary MS, Gayou EL, Kyrouac D, Shah ZK. Pancreaticoportal Fistula Formation as a Consequence of Recurrent Acute Pancreatitis: Clinical and Imaging Considerations. J Clin Imaging Sci 2018;8:4. [Crossref] [PubMed]
6. Ueda T, Saito H, Shirai S, Sugihara F, Fujitsuna R, Matsumoto T, Hayashi H, Kumita SI. Endovascular Treatment for Acute Portal Vein Thrombosis. Interv Radiol (Higashimatsuyama) 2025;10:e20230027. [Crossref] [PubMed]
7. Hou SK, Chern CH, How CK, Chen JD, Wang LM, Lee CH. Hepatic portal venous gas: clinical significance of computed tomography findings. Am J Emerg Med 2004;22:214-8. [Crossref] [PubMed]
8. Moser A, Stauffer A, Wyss A, Schneider C, Essig M, Radke A. Conservative treatment of hepatic portal venous gas consecutive to a complicated diverticulitis: A case report and literature review. Int J Surg Case Rep 2016;23:186-9. [Crossref] [PubMed]
9. Yoo SK, Park JH, Kwon SH. Clinical outcomes in surgical and non-surgical management of hepatic portal venous gas. Korean J Hepatobiliary Pancreat Surg 2015;19:181-7. [Crossref] [PubMed]
10. Ju C, Li X, Gadani S, Kapoor B, Partovi S. Portal Vein Thrombosis: Diagnosis and Endovascular Management. Rofo 2022;194:169-80. [Crossref] [PubMed]
11. Ramalingam V, Yang LM, McCarthy CJ, Ahmed M. Interventional Approach to Portal Vein Thrombosis and Liver Transplantation: State of the Art. Life (Basel) 2023;13:1262. [Crossref] [PubMed]
12. Pescatori LC, Kobeiter H, Zaarour Y, Herin E, Vitellius M, Tacher V. Transhepatic endovascular repair for portal vein haemorrhage. CVIR Endovasc 2020;3:56. [Crossref] [PubMed]
13. Corti R, Quaretti P, Galli F, Moramarco LP, Cionfoli N, Leati G, Corbetta R, Tozzi M. New therapeutic options provided by off-label deployment of stent graft for tailored arteriovenous access salvage: Two cases. SAGE Open Med Case Rep 2017;5:2050313X17741827.