Skip to main navigation Skip to main content
  • KSIR
  • Contact us
  • E-Submission

KJIR : Korean Journal of Interventional Radiology

OPEN ACCESS
Search
Advanced Search
ABOUT
BROWSE ARTICLES
EDITORIAL POLICY
FOR CONTRIBUTORS

Articles

Page Path

Review Article

Clinical Challenges and Transjugular Intrahepatic Portosystemic Shunt Strategies for Pyrrolizidine Alkaloid-Induced Hepatic Sinusoidal Obstruction Syndrome: An Asian Perspective

Tan-Yang Zhou1, Hong-Liang Wang1, Zhi-Cheng Jin2, Bin Xiong1,*, Ji Hoon Shin3,*orcid

Published online: March 9, 2026

1Hepatobiliary and Pancreatic Interventional Treatment Center, Division of Hepatobiliary and Pancreatic Surgery, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China

2Center of Interventional Radiology and Vascular Surgery, Department of Radiology, Zhongda Hospital, Medical School, Southeast University, Nanjing, China

3The Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea

*Corresponding email: jhshin@amc.seoul.kr (J. H. Shin), herr_xiong@126.com (B. Xiong)
• Received: February 8, 2026   • Revised: March 1, 2026   • Accepted: March 2, 2026

© 2026 Korean Society of Interventional Radiology and Korean Journal of Interventional Radiology Institute

This is an open access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 0 Views
  • 0 Download

Full Article

  • Pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome (PA-HSOS) is highly prevalent in Asia, primarily due to the widespread use of traditional herbal medicines containing hepatotoxic pyrrolizidine alkaloids. This condition poses significant clinical challenges, including diagnostic difficulties and limited therapeutic options, frequently leading to severe liver damage and portal hypertension. Transjugular intrahepatic portosystemic shunt (TIPS) treatment has emerged as a key intervention for managing complications associated with PA-HSOS, such as refractory ascites and variceal bleeding, by reducing portal pressure and supporting liver function recovery. However, TIPS has not been widely accepted as a salvage therapy for severe PA-HSOS unresponsive to anticoagulation therapy, mainly due to concerns about post-TIPS complications, particularly hepatic encephalopathy. Consequently, careful patient selection and risk stratification are critical. This review synthesizes the current evidence on PA-HSOS in Asia, evaluates the clinical utility of TIPS, and discusses strategies to optimize outcomes while minimizing adverse effects. Specifically, we review the epidemiology, pathophysiology, and diagnostic advancement of PA-HSOS, with a particular focus on the evolving role of TIPS in its management.
  • Keywords: Hepatic sinusoidal obstruction syndrome; Pyrrolizidine alkaloids; Portal hypertension; Transjugular intrahepatic portosystemic shunt; Treatment
Hepatic sinusoidal obstruction syndrome (HSOS) is a vascular liver disorder characterized by damage to hepatic sinusoidal endothelial cells, resulting in sinusoidal congestion, portal hypertension (PH), and potential liver failure [1-3]. In Western countries, HSOS is predominantly associated with hematopoietic stem cell transplantation (HSCT) or certain chemotherapeutic agents [4,5]. On the other hand, pyrrolizidine alkaloid (PA)–induced HSOS (PA-HSOS) in Asia is primarily induced by the consumption of traditional herbal medicines containing PAs, such as Gynura japonica [1,6]. There are notable clinical and histopathological differences between the two entities. Clinically, PA-HSOS typically presents with abdominal pain, ascites, jaundice, and liver dysfunction, whereas HSCT-related HSOS is more frequently associated with rapid weight gain and renal impairment [7]. Histopathologically, PA-HSOS is characterized by prominent sinusoidal congestion and hepatocyte necrosis, whereas HSCT-related HSOS lacks these specific damage mechanisms inherent to PA exposure [8]. Despite growing clinical recognition, PA-HSOS remains a diagnostic challenge due to overlapping features with other causes of acute liver injury and PH [9]. Histopathologically, PA-HSOS is characterized by sinusoidal dilation, erythrocyte extravasation, and fibrosis, which can be confirmed by liver biopsy [10,11].
PA-HSOS constitutes a substantial health burden in several Asian countries, particularly China, South Korea, and India, where the use of traditional herbal medicines is widespread [12-15]. The market size and diversity of herbal products vary across these countries [16,17]. South Korea shows a clear preference for traditional herbal products, particularly with a growing trend in dietary supplements. A survey of 1,134 Korean respondents revealed that 726 individuals had used herbal products, indicating a high prevalence rate [16]. Notably, Socheongryongtang is the most commonly used herbal formula for treating allergic rhinitis [18]. Similarly, India, often referred to as the 'World's Medicinal Plant Garden,' possesses approximately 8,000 medicinal plant species and provides an estimated 25,000 effective herbal formulations [19,20]. In China, PA-HSOS accounts for a substantial proportion (50.0%–88.6%) of all HSOS cases, frequently linked to herbal products like Gynura japonica [21]. Retrospective studies from China have identified PA-HSOS as a leading cause of drug-induced liver injury (DILI) [11,22]. The condition is closely associated with unregulated herbal products, which underscores the need for greater public awareness and regulatory control [10,23]. Socioeconomic factors, including reliance on folk remedies in rural populations, further increase the risk of PA-HSOS [24]. The condition predominantly affects older individuals and those with chronic exposure to herbal substances, emphasizing the need for targeted preventive strategies [10,25].
Clinical management of PA-HSOS differs notably from Western HSOS, reflecting distinct etiologies and pathophysiology. In PA-HSOS, early anticoagulation or transjugular intrahepatic portosystemic shunt (TIPS) is commonly pursued, especially in severe cases, to reduce portal pressure and improve hepatic perfusion. Liver transplantation is considered for refractory PA-HSOS. By contrast, Western guidelines emphasize defibrotide prophylaxis and broader supportive care as central strategies, with liver transplantation reserved for refractory cases. TIPS appears to show superior outcomes over conservative therapy specifically in PA-HSOS but is rarely used and less favored in Western HSOS. These differences underscore the need for region-specific, tailored management algorithms that account for exposure patterns, resource availability, and clinical trajectories.
This review highlights the key clinical challenges of PA-HSOS, including delayed diagnosis due to nonspecific manifestations such as ascites and hepatomegaly, as well as the lack of reliable biomarkers. We evaluate the current evidence supporting the use of TIPS as a therapeutic option, noting its efficacy in reducing portal pressure despite variable post-procedural outcomes. Management strategies for PA-HSOS are compared with Western HSOS protocols, with a focus on region-specific considerations in Asia. The review also explores emerging treatments, including anticoagulation therapy (AT) with rivaroxaban and gut microbiota modulation. Finally, we advocate for the adoption of standardized diagnostic frameworks, such as the Drum Tower Severity Scoring (DTSS) system, to improve diagnostic accuracy and consistency in PA-HSOS management.
Epidemiology
PA-HSOS is a significant public health concern in Asia, primarily attributed to the widespread use of herbal medicines and dietary supplements containing PAs. In China, PA-HSOS has emerged as the leading cause of DILI, with PA-rich herb products, such as Gynura japonica, playing a central role in its etiology [26]. A systematic review based on 2,156 reported cases of HSOS related to Gynura japonica from 1980 to 2019 indicated that HSOS caused by this herb accounts for over 50% of DILI cases [26]. The cultural reliance on traditional herbal remedies, often consumed without adequate awareness of their hepatotoxic potential, further contributes to the high prevalence of PA-HSOS (Table 1). However, despite its clinical relevance, comprehensive epidemiological data on PA-HSOS remain limited, as the majority of existing studies are retrospective and involve small cohorts. This highlights a critical gap in systematic surveillance and standardized reporting in the region [1,27].
Etiology
The hepatotoxicity of PAs is primarily mediated by metabolic conversion into reactive intermediates, such as dehydro-PAs, which directly damage liver sinusoidal endothelial cells (LSECs) [26,28]. This injury disrupts the sinusoidal architecture, leading to the characteristic features of HSOS, including hepatomegaly, hyperbilirubinemia, and ascites [3,26]. LSECs are particularly vulnerable to PA-induced damage due to their fenestrated structure and high exposure to circulating toxins. The resulting injury to LSECs triggers a cascade of pathological events, including sinusoidal capillarization, perisinusoidal matrix deposition, and impaired vascular exchange, which are key hallmarks of HSOS [29,30]. Additionally, PAs are known to disrupt bile acid homeostasis, aggravating liver injury. The imbalance in bile acid metabolism leads to heightened oxidative stress and hepatocellular damage, contributing to the progression of HSOS [31,32].
Emerging evidence further implicates gut microbiota dysbiosis in PA-HSOS pathogenesis. Experimental models have shown that fecal microbiota transplantation can modulate disease severity, suggesting a role for the gut-liver axis in PA-HSOS development [33,34]. Additionally, macrophage activation and subsequent inflammatory responses amplify sinusoidal injury, creating a pro-fibrogenic microenvironment [29,34]. Over time, PA-HSOS may progress to chronic liver disease, with the development of hepatic fibrosis; however, the precise mechanisms underlying this transition remain incompletely understood [30,35].
High-Risk Populations
Middle-aged and elderly individuals with prolonged exposure to PA-containing herbal products are the highest-risk group for developing PA-HSOS [27,36]. The early mortality rate in this population is approximately 10.3% to 14%, and the 3-year cumulative mortality ranges from 24% to 30.8% [37,38]. Notably, in patients receiving only supportive care, mortality can reach as high as 43.9% [39]. This high mortality rate is concerning, as these individuals often rely on traditional remedies for chronic conditions, typically without awareness of their potential hepatotoxic effects. For instance, Gynura japonica, a popular herb used for treating blood stasis or traumatic injuries, has been strongly associated with numerous cases of HSOS due to its high PAs content [36]. Inadequate regulation of herbal products in many Asian countries exacerbates the risk, as contamination or misidentification of PA-producing plants is not uncommon [32,40]. Reducing the burden of PA-HSOS in these populations requires a multifaceted approach, including enhanced public education, stricter regulatory oversight, and improved strategies for early diagnosis. Additionally, the effects of PAs on individuals are stochastic, characterized by significant variability [41]. Evidence indicates that individual differences in metabolism, species sensitivity, and chemical structure lead to diverse toxic responses, underscoring the non-deterministic nature of these effects rather than a uniform response across all individuals [42].
The diagnosis of PA-HSOS relies on a comprehensive assessment that includes clinical features, imaging findings, and histopathological evaluation [15]. Patients typically present with the hallmark triad of abdominal distension, hepatic region pain, and ascites, often accompanied by jaundice and hepatomegaly (Fig. 1). The widely adopted "Nanjing criteria" for PA-HSOS diagnosis require a confirmed history of exposure to PA-containing plants, such as Gynura segetum, along with at least one of the following: abdominal or hepatic symptoms, elevated serum bilirubin or abnormal liver function tests, imaging evidence of hepatic vascular changes, or compatible pathological findings after excluding other liver injuries [15].
In addition to these criteria, the validated DTSS system has proven useful in stratifying disease severity and guiding therapeutic decisions [27,43]. Imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI) play an important non-invasive role in the diagnostic process, often revealing characteristic signs such as hepatic sinusoidal congestion, hepatic vein narrowing, and a "clover-like" enhancement pattern [3,44] (Fig. 1). Enhanced CT typically shows heterogeneous low-density areas in the liver, which is the most common manifestation, alongside hepatomegaly, ascites, and thickening of the gallbladder wall. During the arterial or portal venous phase, patchy liver enhancement reflects abnormal perfusion [45]. On MRI, gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) enhanced imaging during the hepatobiliary phase consistently shows low signal intensity in PA-HSOS patients, correlating positively with prothrombin time (PT) and international normalized ratio, thus serving as an independent mortality risk factor. Additionally, susceptibility-weighted imaging and T2-weighted imaging reveal low signal areas associated with hemosiderin deposition [46]. Nevertheless, liver biopsy remains the diagnostic gold standard, allowing for the identification of characteristic histopathological lesions. Differentiating PA-HSOS from other vascular liver diseases, including Budd-Chiari syndrome and cirrhotic PH, continues to present a significant clinical challenge (Table 2) [47,48]. Future research should prioritize the validation of non-invasive biomarkers and the refinement of scoring systems such as DTSS to enhance diagnostic accuracy and allow timely intervention, particularly in high-prevalence regions.
Medical Management
The treatment of PA-HSOS remains challenging due to the lack of specific therapeutic options. AT, including low-molecular-weight heparin (LMWH) and rivaroxaban, is often employed as an initial treatment. Specifically, LMWH is dosed at 4,000 IU subcutaneously twice daily [39]. In a study involving five PA-HSOS patients treated with LMWH for 8–21 days, ascites resolved, symptoms improved, and hepatic venous blood flow was restored [49]. However, in severe cases, AT alone appears limited in efficacy. Another study reported a mortality rate of 34.1% with AT versus 0% with TIPS [50]. A multicenter study involving 249 PA-HSOS patients indicated that AT improved survival in some cases; however, the benefits were inconsistent, highlighting the need for better patient stratification [39]. Defibrotide, which has proven effective for managing HSOS after HSCT, has also been studied in PA-HSOS [51]. In HSCT-associated HSOS, defibrotide is dosed at 6.25 mg/kg intravenously every 6 hours (equivalent to 25 mg/kg per day) and continued for at least 3 weeks until total bilirubin (TBIL) levels normalize. However, its efficacy may be constrained by the unique pathogenesis of PA-HSOS, and its limited availability in China raises uncertainties about its effectiveness in treating PA-induced HSOS. A study assessing defibrotide effectiveness in monocrotaline-induced rat HSOS found that it improved outcomes, suggesting that defibrotide may be a preferable option to LMWH in clinical practice [8].
Transjugular Intrahepatic Portosystemic Shunt
TIPS has emerged as an important therapeutic option for PA-HSOS patients with refractory PH or ascites, and although it has no specific contraindications for treatment, guidelines recommend considering it for patients with ineffective medical therapy, highlighting the importance of individualized assessment [15]. Studies have shown that ascites and pleural effusions can markedly improve within a short period after TIPS placement, with previously obstructed hepatic veins often regaining patency in the short term when combined with AT (Figs. 1, 2) [10,52]. Various studies have confirmed the safety and efficacy of TIPS in managing PA-HSOS (Table 3) [50,52-57]. A retrospective study involving 30 PA-HSOS patients demonstrated that TIPS significantly reduced portal pressure, alleviated ascites, and improved clinical symptoms [52]. The median time to complete ascites remission was 52 days, while the median recovery time for liver CT radiological manifestations was 196.5 days. However, the timing of the procedure is critical, with preoperative TBIL levels identified as a key prognostic factor. Early intervention, when TBIL levels are below 10 mg/dL, is associated with improved recovery [53].
In the context of treatment strategy selection, a multicenter study comparing the efficacy between TIPS and AT in 164 PA-HSOS patients revealed that TIPS conferred significantly better mid- to long-term survival rates compared with those of AT, particularly in patients with moderate to severe PA-HSOS [54]. This suggests that TIPS should be considered a viable initial treatment option for this patient cohort [50]. Similarly, another retrospective study involving 69 patients in the TIPS group and 95 patients in the supportive treatment group confirmed that TIPS placement significantly improved survival rates and effectively alleviated PH-related clinical symptoms in PA-HSOS patients [54].
Despite its promising efficacy, the use of TIPS in Asian populations faces several challenges. Research indicates that a significant proportion of PA-HSOS patients present at advanced stages of the disease, complicating the identification of appropriate candidates for TIPS [58]. Additionally, postoperative anticoagulation management protocols vary widely among medical institutions, and there is currently no standardized approach for preventing shunt thrombosis or disease progression [59]. To address these challenges, a research team developed the DTSS system, which incorporates prognostic factors such as liver function and imaging findings to guide treatment strategy selection, including TIPS candidacy [11,27].
Regarding prognostic assessment, several studies have explored predictive factors that may influence the effectiveness of TIPS treatment. One study found that prolonged baseline PT and elevated serum TBIL levels 5 days post-TIPS are independent risk factors predicting mortality in PA-HSOS patients following TIPS placement [55]. Another study suggested that preoperative measurement of the hepatic venous pressure gradient may assist in assessing disease severity and predicting TIPS treatment efficacy [60]. Notably, the response to TIPS may vary depending on the underlying cause of PA-HSOS. For example, a study involving nine patients with HSOS related to Gynura segetum found that TIPS treatment significantly improved clinical symptoms; nevertheless, further research is needed to determine the optimal timing for intervention [61]. Additionally, pathological studies have shown that TIPS placement can lead to significant improvements in pathological features, such as hepatic congestion and hepatocyte swelling, over time [56].
Despite the promising outcomes of TIPS, its long-term efficacy in PA-HSOS requires further validation. A multicenter study involving 117 PA-HSOS patients emphasized the importance of establishing a reliable prognostic assessment system to better guide treatment decisions [38]. Additionally, for some advanced PA-HSOS patients, TIPS may serve primarily as a bridge to liver transplantation [25,62]. In PA-HSOS patients without severe underlying cirrhosis, post-TIPS complications, such as hepatic encephalopathy (HE), are less common than in cirrhotic cohorts [52,63]. A systematic review included 19 studies and 465 HSOS patients, reported an overall HE incidence of 13.2% post-TIPS; however, this data encompasses all types of HSOS and is not limited to the PA-HSOS subgroup [64]. Conversely, a retrospective review of 30 PA-HSOS patients who underwent TIPS found no post-procedure HE cases [52]. By comparison, the pooled HE incidence among cirrhotic patients ranges from 33.2% to 58% [65,66]. The absence of advanced liver disease likely reduces the risks associated with portosystemic shunting, including hyperammonemia and HE, making TIPS a relatively safer option for these patients [10,25]. Overall, existing evidence suggests that TIPS may be an effective means of managing PA-HSOS associated with PH; however, careful consideration of patient-specific factors is essential to determine the appropriate indications and timing for intervention. Future research should focus on establishing standardized preoperative assessment systems and postoperative management protocols to further enhance treatment outcomes [7,45,67].
Liver Transplantation
For end-stage PA-HSOS patients who do not respond to medical and TIPS therapies, liver transplantation is the last resort. However, data on transplantation outcomes in PA-HSOS are limited, as most cases are managed conservatively or with TIPS [25]. Although the high mortality rate and rapid disease progression highlight the importance of early referral to transplant centers, organ availability and patient selection criteria complicate the transplantation process, particularly in Asia [38]. Future studies should aim to optimize the timing of liver transplantation and refine post-transplant management strategies to improve survival outcomes in this critically ill population.
Recent studies have advanced our understanding of PA-HSOS, particularly regarding the underlying toxicity pathways and the role of the gut-liver axis in disease progression. Emerging evidence suggests that the PU.1 signaling pathway plays a vital role in mediating PA-induced hepatotoxicity, primarily by driving damage to sinusoidal endothelial cells [68]. Additionally, the gut microbiota has been implicated in PA-HSOS pathogenesis, with studies on fecal microbiota transplantation revealing that dysbiosis exacerbates liver injury through macrophage activation [33,34]. Experimental models of monocrotaline-induced HSOS have demonstrated the potential therapeutic benefits of modulating the gut microbiota [34,69], further underscoring the importance of investigating specific microbial taxa and their interactions with host immune pathways.
The lack of standardized diagnostic criteria for PA-HSOS in Asia highlights the urgent need for multicenter collaborative efforts to establish region-specific guidelines. Current diagnostic challenges include the overlap of PA-HSOS with other liver diseases and the variable sensitivity of non-invasive diagnostic tools [70]. Furthermore, although TIPS is crucial for managing severe PA-HSOS, its clinical efficacy is often hampered by complications and inconsistent outcomes [55]. The development of enhanced TIPS techniques and public education on the risks associated with the unregulated use of herbal medicines are essential for prevention and improved patient care [71].

Conflict of interest

No potential conflicts of interest relevant to this article were reported.

Funding

None.

Acknowledgments

None.

Author contributions

Conceptualization, investigation, and manuscript writing: TYZ. Conceptualization and investigation: HLW, ZCJ. All aspects of the work: BX, JHS.

Fig. 1.
A 62-year-old man diagnosed with hepatic sinusoidal obstruction syndrome 4 months after ingestion of Gynura segetum. (A) Abdominal computed tomography (CT) during the hepatic venous phase demonstrated patchy parenchymal enhancement with characteristic clover-like enhancement around the hepatic veins. Significant ascites and bilateral pleural effusions were also observed. (B) Coronal CT revealed patchy liver enhancement and narrowing (thinning) of the portal vein, accompanied by abundant ascites. (C, D) Portography before and after transjugular intrahepatic portosystemic shunt (TIPS) creation. A guidewire, introduced via the right hepatic artery, was used to provide real-time guidance during the procedure. (E–H) Post-TIPS abdominal CT showed well-defined opacification of hepatic veins with substantial resolution of both ascites and pleural effusions. Long-term follow-up CT, extending up to 24 months, confirmed homogeneous enhancement of the liver parenchyma and sustained patency of the TIPS shunt.
kjir-2026-00031f1.jpg
Fig. 2.
A 23-year-old woman presented with pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome following ingestion of oral weight-loss medications. (A) Pre–transjugular intrahepatic portosystemic shunt (TIPS) computed tomography (CT) (portal venous phase) demonstrated heterogeneous, patchy hepatic enhancement with multiple regenerative nodules, accompanied by massive ascites and bilateral pleural effusions. (B) Histopathological analysis from a transjugular liver biopsy revealed focal hepatic atrophy, marked sinusoidal dilatation, and hepatocellular cholestasis, consistent with hepatic venous outflow obstruction. Immunohistochemical staining was negative for hepatitis B surface antigen, hepatitis B core antigen, diastase-periodic acid-Schiff, and periodic acid-Schiff, while showing positivity for cytokeratin 7 (biliary epithelium), reticulin, and Masson’s trichrome (indicating underlying fibrosis). (C) Due to the diminished caliber (slenderness) of the portal vein, percutaneous transhepatic balloon-assisted TIPS placement was performed. Initial portal venography confirmed narrowing of the intrahepatic portal branches, stagnant portal flow, and the presence of prominent esophagogastric varices. (D) Post-TIPS venography demonstrated a widely patent shunt with unobstructed portal venous return and successful decompression of the varices. (E, F) Follow-up contrast-enhanced CT (venous phase), extending up to 72 months, revealed persistent heterogeneous enhancement and intrahepatic nodules; however, complete resolution of ascites and pleural effusions was observed. The TIPS shunt remained patent with uninterrupted flow.
kjir-2026-00031f2.jpg
Table 1.
Common plants causing PA-HSOS (Asia region)
Table 1.
Plant family Common genera/species Notes
Asteraceae (Compositae) Gynura japonica (Tusanqi) Leading cause of PA-HSOS in China; roots/herbs used in traditional medicine
Gynura segetum Widely used herbal plant containing PAs; linked to HSOS in China
Senecio spp. PAs (e.g., retrorsine) damage sinusoidal endothelial cells
Tussilago farfara (coltsfoot) Traditional herb with PAs; potential HSOS risk
Boraginaceae Heliotropium spp. Plants containing hepatotoxic PAs; implicated in HSOS cases
Symphytum spp. (comfrey) Traditional medicinal herb; PAs cause HSOS and chronic liver damage
Echium spp. PA-containing plants associated with liver injury
Fabaceae (Leguminosae) Crotalaria spp. Plants containing monocrotaline, a PA that induces sinusoidal obstruction in animal models

PA-HSOS, pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome; PAs, pyrrolizidine alkaloids; HSOS, hepatic sinusoidal obstruction syndrome.

Table 2.
Key features for differential diagnosis of PA-HSOS, Budd-Chiari syndrome, and typical cirrhotic decompensation
Table 2.
Feature PA-HSOS Budd-Chiari syndrome Typical cirrhotic decompensation
Etiology Exposure to PAs Thrombosis of large hepatic veins or inferior vena cava Chronic liver disease (e.g., viral hepatitis, alcohol)
Onset Acute/subacute Acute or subacute Chronic, with acute exacerbations
Portal hypertension Yes, post-sinusoidal portal hypertension due to hepatic outflow obstruction Yes, due to obstruction of major veins Yes, due to cirrhosis
Imaging findings Patchy parenchymal enhancement on CT/MRI Hepatic vein occlusion, collateral circulation Nodular liver surface, ascites, and splenomegaly on imaging
Ascites Common, often refractory to diuretics Common, often with significant fluid accumulation Common, usually associated with other signs of cirrhosis
Clinical presentation Ascites, hepatomegaly, jaundice Acute liver failure, abdominal pain, ascites Varices, jaundice, hepatic encephalopathy
Histopathology Sinusoidal dilatation, necrosis, fibrosis Thrombosis-related changes in the liver Fibrosis, cirrhosis, regenerative nodules
Coagulation status Often coagulopathy present Prothrombotic states common Coagulopathy may be present due to liver dysfunction
Risk factors Herbal medicine use, exposure to specific plants Myeloproliferative disorders, pregnancy, oral contraceptives Alcohol use, viral hepatitis, metabolic disorders
Management Supportive care, TIPS for refractory cases Anticoagulation, TIPS for severe cases Management of underlying liver disease, TIPS for refractory ascites

PA-HSOS, pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome; PAs, pyrrolizidine alkaloids; CT, computed tomography; MRI, magnetic resonance imaging; TIPS, transjugular intrahepatic portosystemic shunt.

Table 3.
Summary of major studies on TIPS for PA-HSOS in Asia
Table 3.
Country Author (Year) Sample size Type of study Aim of study Outcomes
China Dai et al. (2025) [52] 30 Patients Retrospective analysis Evaluated safety/efficacy of TIPS The PPG significantly decreased post-TIPS, with ascites remission occurring in 52 days (median) and liver CT showing recovery in 196.5 days. Disease severity and ALBI grade were key recovery time predictors.
China Huang et al. (2023) [50] 20 Patients (TIPS) and 41 patients (AT) Comparative study Compared mid-long-term outcomes of TIPS vs. AT as initial treatment TIPS demonstrated superior mid- to long-term outcomes compared to anticoagulation therapy, particularly for patients with severe or very severe PA-induced HSOS.
China Wu et al. (2021) [53] 4 in 10 PA-HSOS patients received TIPS Retrospective analysis Evaluated TIPS timing using TBIL as a measure TIPS timing based on TBIL levels affected efficacy.
China Huang et al. (2023) [57] 12 Patients (TIPS) and 10 patients (conservative treatment) Retrospective cohort Investigated efficacy and safety of TIPS in PA-HSOS TIPS may be a safe and effective therapeutic strategy for PA-HSOS patients who do not respond to conservative treatment.
China Zhou et al. (2020) [56] 37 Patients (TIPS) and 17 patients (conservative treatment) Retrospective analysis Evaluated TIPS as treatment for PA-HSOS Better outcomes may be achieved with TIPS compared with conventional symptomatic treatment in patients with PA-HSOS.
China Wang et al. (2023) [54] 164 Patients (69 in TIPS group vs. 95 in supportive group) Retrospective analysis Compared clinical outcomes and liver histology between TIPS and supportive treatment In comparison with supportive treatment, TIPS treatment improved clinical outcomes and liver histology.
China Xiao et al. (2021) [55] 116 Patients Retrospective analysis Evaluated predictors of poor outcomes in PA-HSOS patients receiving TIPS Prolonged prothrombin time at baseline and increased serum TBIL levels 5 days after TIPS were independent risk factors for predicting death after TIPS treatment in PA-HSOS patients.

TIPS, transjugular intrahepatic portosystemic shunt; PA-HSOS, pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome; PPG, portal pressure gradient; CT, computed tomography; ALBI, albumin-bilirubin; AT, anticoagulation therapy; PA, pyrrolizidine alkaloid; HSOS, hepatic sinusoidal obstruction syndrome; TBIL, total bilirubin.

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Clinical Challenges and Transjugular Intrahepatic Portosystemic Shunt Strategies for Pyrrolizidine Alkaloid-Induced Hepatic Sinusoidal Obstruction Syndrome: An Asian Perspective
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.
Format:
Include:
Clinical Challenges and Transjugular Intrahepatic Portosystemic Shunt Strategies for Pyrrolizidine Alkaloid-Induced Hepatic Sinusoidal Obstruction Syndrome: An Asian Perspective
Close

Figure

  • 0
  • 1
Clinical Challenges and Transjugular Intrahepatic Portosystemic Shunt Strategies for Pyrrolizidine Alkaloid-Induced Hepatic Sinusoidal Obstruction Syndrome: An Asian Perspective
Image Image
Fig. 1. A 62-year-old man diagnosed with hepatic sinusoidal obstruction syndrome 4 months after ingestion of Gynura segetum. (A) Abdominal computed tomography (CT) during the hepatic venous phase demonstrated patchy parenchymal enhancement with characteristic clover-like enhancement around the hepatic veins. Significant ascites and bilateral pleural effusions were also observed. (B) Coronal CT revealed patchy liver enhancement and narrowing (thinning) of the portal vein, accompanied by abundant ascites. (C, D) Portography before and after transjugular intrahepatic portosystemic shunt (TIPS) creation. A guidewire, introduced via the right hepatic artery, was used to provide real-time guidance during the procedure. (E–H) Post-TIPS abdominal CT showed well-defined opacification of hepatic veins with substantial resolution of both ascites and pleural effusions. Long-term follow-up CT, extending up to 24 months, confirmed homogeneous enhancement of the liver parenchyma and sustained patency of the TIPS shunt.
Fig. 2. A 23-year-old woman presented with pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome following ingestion of oral weight-loss medications. (A) Pre–transjugular intrahepatic portosystemic shunt (TIPS) computed tomography (CT) (portal venous phase) demonstrated heterogeneous, patchy hepatic enhancement with multiple regenerative nodules, accompanied by massive ascites and bilateral pleural effusions. (B) Histopathological analysis from a transjugular liver biopsy revealed focal hepatic atrophy, marked sinusoidal dilatation, and hepatocellular cholestasis, consistent with hepatic venous outflow obstruction. Immunohistochemical staining was negative for hepatitis B surface antigen, hepatitis B core antigen, diastase-periodic acid-Schiff, and periodic acid-Schiff, while showing positivity for cytokeratin 7 (biliary epithelium), reticulin, and Masson’s trichrome (indicating underlying fibrosis). (C) Due to the diminished caliber (slenderness) of the portal vein, percutaneous transhepatic balloon-assisted TIPS placement was performed. Initial portal venography confirmed narrowing of the intrahepatic portal branches, stagnant portal flow, and the presence of prominent esophagogastric varices. (D) Post-TIPS venography demonstrated a widely patent shunt with unobstructed portal venous return and successful decompression of the varices. (E, F) Follow-up contrast-enhanced CT (venous phase), extending up to 72 months, revealed persistent heterogeneous enhancement and intrahepatic nodules; however, complete resolution of ascites and pleural effusions was observed. The TIPS shunt remained patent with uninterrupted flow.

Fig. 1.

Fig. 2.

Clinical Challenges and Transjugular Intrahepatic Portosystemic Shunt Strategies for Pyrrolizidine Alkaloid-Induced Hepatic Sinusoidal Obstruction Syndrome: An Asian Perspective
Plant family Common genera/species Notes
Asteraceae (Compositae) Gynura japonica (Tusanqi) Leading cause of PA-HSOS in China; roots/herbs used in traditional medicine
Gynura segetum Widely used herbal plant containing PAs; linked to HSOS in China
Senecio spp. PAs (e.g., retrorsine) damage sinusoidal endothelial cells
Tussilago farfara (coltsfoot) Traditional herb with PAs; potential HSOS risk
Boraginaceae Heliotropium spp. Plants containing hepatotoxic PAs; implicated in HSOS cases
Symphytum spp. (comfrey) Traditional medicinal herb; PAs cause HSOS and chronic liver damage
Echium spp. PA-containing plants associated with liver injury
Fabaceae (Leguminosae) Crotalaria spp. Plants containing monocrotaline, a PA that induces sinusoidal obstruction in animal models
Feature PA-HSOS Budd-Chiari syndrome Typical cirrhotic decompensation
Etiology Exposure to PAs Thrombosis of large hepatic veins or inferior vena cava Chronic liver disease (e.g., viral hepatitis, alcohol)
Onset Acute/subacute Acute or subacute Chronic, with acute exacerbations
Portal hypertension Yes, post-sinusoidal portal hypertension due to hepatic outflow obstruction Yes, due to obstruction of major veins Yes, due to cirrhosis
Imaging findings Patchy parenchymal enhancement on CT/MRI Hepatic vein occlusion, collateral circulation Nodular liver surface, ascites, and splenomegaly on imaging
Ascites Common, often refractory to diuretics Common, often with significant fluid accumulation Common, usually associated with other signs of cirrhosis
Clinical presentation Ascites, hepatomegaly, jaundice Acute liver failure, abdominal pain, ascites Varices, jaundice, hepatic encephalopathy
Histopathology Sinusoidal dilatation, necrosis, fibrosis Thrombosis-related changes in the liver Fibrosis, cirrhosis, regenerative nodules
Coagulation status Often coagulopathy present Prothrombotic states common Coagulopathy may be present due to liver dysfunction
Risk factors Herbal medicine use, exposure to specific plants Myeloproliferative disorders, pregnancy, oral contraceptives Alcohol use, viral hepatitis, metabolic disorders
Management Supportive care, TIPS for refractory cases Anticoagulation, TIPS for severe cases Management of underlying liver disease, TIPS for refractory ascites
Country Author (Year) Sample size Type of study Aim of study Outcomes
China Dai et al. (2025) [52] 30 Patients Retrospective analysis Evaluated safety/efficacy of TIPS The PPG significantly decreased post-TIPS, with ascites remission occurring in 52 days (median) and liver CT showing recovery in 196.5 days. Disease severity and ALBI grade were key recovery time predictors.
China Huang et al. (2023) [50] 20 Patients (TIPS) and 41 patients (AT) Comparative study Compared mid-long-term outcomes of TIPS vs. AT as initial treatment TIPS demonstrated superior mid- to long-term outcomes compared to anticoagulation therapy, particularly for patients with severe or very severe PA-induced HSOS.
China Wu et al. (2021) [53] 4 in 10 PA-HSOS patients received TIPS Retrospective analysis Evaluated TIPS timing using TBIL as a measure TIPS timing based on TBIL levels affected efficacy.
China Huang et al. (2023) [57] 12 Patients (TIPS) and 10 patients (conservative treatment) Retrospective cohort Investigated efficacy and safety of TIPS in PA-HSOS TIPS may be a safe and effective therapeutic strategy for PA-HSOS patients who do not respond to conservative treatment.
China Zhou et al. (2020) [56] 37 Patients (TIPS) and 17 patients (conservative treatment) Retrospective analysis Evaluated TIPS as treatment for PA-HSOS Better outcomes may be achieved with TIPS compared with conventional symptomatic treatment in patients with PA-HSOS.
China Wang et al. (2023) [54] 164 Patients (69 in TIPS group vs. 95 in supportive group) Retrospective analysis Compared clinical outcomes and liver histology between TIPS and supportive treatment In comparison with supportive treatment, TIPS treatment improved clinical outcomes and liver histology.
China Xiao et al. (2021) [55] 116 Patients Retrospective analysis Evaluated predictors of poor outcomes in PA-HSOS patients receiving TIPS Prolonged prothrombin time at baseline and increased serum TBIL levels 5 days after TIPS were independent risk factors for predicting death after TIPS treatment in PA-HSOS patients.
Table 1. Common plants causing PA-HSOS (Asia region)

PA-HSOS, pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome; PAs, pyrrolizidine alkaloids; HSOS, hepatic sinusoidal obstruction syndrome.

Table 2. Key features for differential diagnosis of PA-HSOS, Budd-Chiari syndrome, and typical cirrhotic decompensation

PA-HSOS, pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome; PAs, pyrrolizidine alkaloids; CT, computed tomography; MRI, magnetic resonance imaging; TIPS, transjugular intrahepatic portosystemic shunt.

Table 3. Summary of major studies on TIPS for PA-HSOS in Asia

TIPS, transjugular intrahepatic portosystemic shunt; PA-HSOS, pyrrolizidine alkaloid–induced hepatic sinusoidal obstruction syndrome; PPG, portal pressure gradient; CT, computed tomography; ALBI, albumin-bilirubin; AT, anticoagulation therapy; PA, pyrrolizidine alkaloid; HSOS, hepatic sinusoidal obstruction syndrome; TBIL, total bilirubin.

Table 1.

Table 2.

Table 3.

KJIR : Korean Journal of Interventional Radiology

ABOUT

BROWSE ARTICLES

EDITORIAL POLICY

FOR CONTRIBUTORS

© Korean Society of Interventional Radiology.