Abstract
-
Background/Aims
- Post-endoscopic variceal ligation ulcer bleeding (PEBU) is a rare but life-threatening complication in cirrhosis, often refractory to standard therapy. While transjugular intrahepatic portosystemic shunt (TIPS) is effective, its complexity and risk of hepatic encephalopathy limit its use. Self-expandable metal stent (SEMS) offers a minimally invasive alternative, but comparative data are limited. This study evaluated SEMS versus TIPS in refractory PEBU.
-
Methods
- In this retrospective cohort study, 30 cirrhotic patients with refractory PEBU at a tertiary Indian center were treated with either SEMS (n=15) or TIPS (n=15). The primary endpoint was hemostasis within 72 hours without rebleeding within 5 days. Secondary outcomes included 12-week rebleeding, mortality, complications, and quality of life (Patient-Reported Outcomes Measurement Information System [PROMIS]-10).
-
Results
- Baseline characteristics were comparable. Immediate hemostasis was achieved in 100% (SEMS) and 93.3% (TIPS) (p=0.31). Rebleeding rates were similar (13.3%). SEMS showed lower mortality (6.7% vs. 20%) and less hepatic encephalopathy (20% vs. 46.7%). SEMS showed superior PROMIS-10 scores (13.93±4.33 vs.10.07±4.57, p<0.001) and shorter intensive care unit stay (2 vs. 5 days, p=0.002). Stent migration occurred in 20%.
-
Conclusions
- SEMS provides outcomes comparable to TIPS with fewer complications and easier applicability, supporting its use in resource-limited settings.
-
Keywords: Endoscopy; Esophageal and gastric varices; Gastrointestinal hemorrhage; Liver cirrhosis; Portal hypertension
Graphical abstract
INTRODUCTION
Portal hypertension (PHTN), a frequent sequela of advanced liver disease, often leads to the development of esophageal varices, which are a major cause of upper gastrointestinal hemorrhage. Esophageal varices, a sequela of elevated portal pressure, bleed in approximately 30% of patients with cirrhosis, with significant recurrence rates within one year of the initial hemorrhagic episode. Despite advancements in therapeutic interventions, esophageal variceal bleeding, a life-threatening complication of PHTN, remains a significant source of morbidity and mortality, with rates ranging from 20 to 50%.1
Endoscopic variceal ligation (EVL) has emerged as the gold standard for first-line management of bleeding esophageal varices, achieving effective hemostasis in 85%–90% of cases and serving as a cornerstone in the treatment of variceal bleeding. However, EVL is not devoid of complications.2 Post-EVL bleeding ulcers (PEBU) present a distinct and formidable challenge, particularly when refractory to conventional therapies. It is estimated that 3.6%–15% of patients undergoing EVL develop PEBU, a potentially life-threatening condition with limited therapeutic options. Prognosis is heavily contingent on the severity of the underlying liver disease, degree of PHTN, and timeliness of adequate intervention.3,4
In scenarios where PEBU is refractory to pharmacologic and endoscopic modalities, current salvage interventions include balloon tamponade (BT) and transjugular intrahepatic portosystemic shunting (TIPS). BT, while efficacious in achieving prompt hemostasis through mechanical compression, is inherently constrained by its propensity for severe complications, including esophageal necrosis, perforation, aspiration pneumonia, and rebound hemorrhage, thereby relegating its role to that of a transient temporizing measure.
TIPS offers a more definitive therapeutic avenue by mitigating portal pressure and substantially reducing the risk of rebleeding. However, its utilization is impeded by prohibitive costs, intricate technical execution, and limited availability, particularly in resource-limited settings. Furthermore, TIPS is accompanied by potential adverse sequelae including hepatic encephalopathy and decompensation, particularly in patients with advanced liver dysfunction. Its deployment also complicates subsequent liver transplantation, thereby curtailing its applicability in select clinical paradigms.5,6
Recently, self-expandable metal stent (SEMS) has garnered significant attention as minimally invasive and promising alternatives for the management of refractory variceal bleeding. Devices such as the SX-ELLA Danis stent have demonstrated remarkable technical success and hemostatic efficacy, with bleeding control reported in 82%–100% of refractory cases across several studies. These stents effectively provide tamponade and facilitate the early initiation of enteral nutrition while mitigating complications typically associated with BT. Furthermore, SEMS deployment is technically straightforward, often feasible at the bedside, and typically obviates the need for airway protection, which is a significant advantage over BT.7-10
Despite these advantages, concerns regarding stent migration, ulceration, and delayed complications persist, underscoring the need for further comparative evaluations. Although SEMS has been successfully employed for refractory variceal bleeding, their role in managing post-EVL ulcer bleeding remains inadequately investigated. The existing literature highlights a paucity of robust comparative data evaluating the efficacy and safety of SEMS placement versus established interventions, such as TIPS, particularly in the context of refractory post-EVL-induced ulcer bleeding. To date, most studies have focused exclusively on either TIPS or SEMS, leaving clinicians without comprehensive evidence to inform optimal therapeutic strategies.7,11-13
This retrospective, single-center, observational study described the short-term clinical outcomes of TIPS and SEMS placement in managing refractory post-EVL ulcer bleeding. By targeting a population with limited therapeutic alternatives, this study aimed to provide valuable insights into optimizing clinical outcomes and guiding decision-making in this high-risk cohort. To the best of our knowledge, this is the first study to report the outcomes of SEMS and TIPS in patients with refractory PEBU.
By conducting this retrospective observational analysis, we aimed to generate evidence that will substantially influence clinical practice, especially as a potential salvage or bridge therapy, in scenarios where immediate access to advanced interventional radiology may be restricted. These findings are expected to elucidate the role of SEMS as a potential bridge to definitive therapy while minimizing the complications associated with TIPS. This hypothesis posits that SEMS offer a safer, more accessible, and more effective alternative in this challenging clinical scenario.
METHODS
Study design and setting
This retrospective cohort study was conducted at an academic tertiary care center in northern India. Patients were identified through a retrospective chart review of endoscopic and interventional radiology logs from January, 2017 to December, 2024. The study was designed to assess short-term outcomes following SEMS or TIPS placement for refractory post-EVL ulcer bleeding; subsequent definitive interventions, including delayed TIPS placement or liver transplantation, were not systematically captured.
Study participants
Adult patients (18–60 years old) with documented refractory PEBU were identified from the electronic medical records. All patients underwent EVL (primary or secondary) and were prescribed proton pump inhibitor (PPI) and oral and oral sucralfate syrup twice daily. To ensure data uniformity, patients who had undergone EVL outside the hospital were excluded to avoid variability in procedural parameters such as the number of bands deployed, procedure duration, and endoscopist expertise.
All patients received standard initial management for post-EVL ulcer bleeding, including vasoactive therapy used for variceal bleeding (octreotide [500 mcg bolus followed by 250 mcg/hour), or terlipressin (2 mg intravenous every 6 hours]) along with high-dose intravenous PPI therapy. Endoscopic management prior to escalation varied according to the clinical presentation and included epinephrine injection (n=23) and cyanoacrylate glue (n=7). Refractory post-EVL ulcer bleeding was defined as the failure to achieve sustained hemostasis despite optimal pharmacological therapy and endoscopic intervention in accordance with the Baveno VII criteria, prompting escalation to salvage therapy with SEMS or TIPS.2
Exclusion criteria included age <18 years, intermediate and advanced hepatocellular carcinoma, simultaneous presence of fundal varices, prior placement of SEMS or TIPS, pregnancy or lactation, significant comorbidities (e.g., advanced cardiac or respiratory disease) precluding safe participation, and an inability to provide informed consent.
Interventions
The exposure variable was the type of salvage intervention for refractory PEBU: either SEMS or TIPS (Fig. 1). The choice of intervention was made by the treating physician based on the clinical presentation, institutional logistics, and patient-specific factors. SEMS placement was typically performed at the bedside or in an endoscopy suite, whereas TIPS procedures were performed in a dedicated interventional radiology suite.
Study outline/flow
Patients who met the inclusion criteria during the study period were identified from their medical records. Baseline Demographic information (age, gender, etiology and duration of cirrhosis, severity of cirrhosis, comorbidities, etc.), clinical characteristics (presenting symptoms and their duration, severity of illness (as per model for end-stage liver disease-sodium (MELD-Na) score and the Child-Turcotte-Pugh (CTP) classification system), grading of varix(small (<5 mm) or large(>5 mm) at the time of endoscopy), vitals, blood transfusion requirement, and laboratory parameters were collected from data. Ulcer morphology was categorized according to the Jamwal and Sarin classification.5
Interventions were conducted using standardized protocols. SEMS placement was performed either at the bedside or in the endoscopy suite, whereas TIPS procedures were conducted under fluoroscopic guidance by experienced interventional radiologists. Immediate outcomes, such as technical success and achievement of hemostasis, have been meticulously documented. The participants were monitored during hospitalization for procedure-related complications, including stent migration, ulcer perforation, and hepatic encephalopathy.
Procedural details
1) Esophageal SEMS placement
After obtaining informed consent, patients received local anesthesia (lidocaine spray) and were placed in the left lateral decubitus position. Endoscopy was performed to assess the bleeding site, with the proximal and distal margins marked under endoscopic and fluoroscopic guidance. A stiff guidewire was passed across the lower esophageal sphincter into the stomach and its position was confirmed fluoroscopically. A preloaded SX-ELLA Danis stent (135×35×25 mm) was advanced over the guidewire and positioned appropriately under fluoroscopic control. A 50 mL gastric balloon was inflated to ensure adequate expansion and the stent was deployed by carefully withdrawing the delivery system. Post-deployment endoscopy confirmed stent patency, hemostasis, and absence of mucosal injury or perforation.
When SEMS placement was performed at bedside, stent positioning was guided by measurements obtained during the preceding diagnostic endoscopy, including the distance from the incisors to the gastroesophageal junction and the bleeding site. This measurement was used to preset the yellow positioning plate on the delivery system, which functioned as an external landmark for determining the appropriate proximal and distal stent margins. The delivery sheath contained graduated scale markers to facilitate the verification of the insertion depth. After insertion into the stomach, the gastric fixation balloon was inflated with 120 mL of air (performed in two 60-mL inflations), followed by gentle traction to secure the balloon against the gastric cardia before stent deployment, in accordance with the manufacturer’s instructions. All SEMS placements were performed under conscious sedation without general anesthesia. Sedation typically consists of intravenous benzodiazepines and/or opioids adjusted according to patient tolerance and hemodynamic status.
The patients were monitored closely for complications such as stent migration, chest pain, or perforation. Dietary instructions included a liquid or soft diet for 6 hours after the procedure. Following SEMS placement, the position of the stent was reassessed approximately 48 hours post-procedure using plain radiographic imaging (chest and abdominal X-rays) to confirm correct stent localization and evaluate early migration. Repeat endoscopy was not routinely performed for positional reassessment, but was reserved for patients with clinical signs suggestive of migration or recurrent bleeding.
If stent migration was confirmed, the endoscopic stent was repositioned. In cases of persistent bleeding, the patients were counselled regarding the possibility of rescue TIPS.
Stent removal was performed after 10 days using a standard endoscope and retrieval was achieved by grasping the removal thread with forceps and retracting the stent under direct visualization.
2) TIPS placement
TIPS placement was performed in a fully equipped surgical unit under fluoroscopic and ultrasound guidance by an experienced interventional radiologist under anesthesia support. The right internal jugular vein was cannulated under direct visualization and aseptic conditions. A catheter was advanced into the right atrium and right hepatic vein, and a CO₂ balloon occlusion venogram was performed to visualize the portal vein.
Real-time ultrasound guidance facilitated the needle puncture of the portal vein, which was confirmed by contrast injection. If direct access failed, a transhepatic guidewire was placed as the fluoroscopic target. A catheter was then advanced into the portal vein for venography, and the portal and right atrial pressures were recorded to calculate the portosystemic gradient. The intrahepatic tract was dilated (8–10 mm), and a partially covered stent (9F*10 mm Solaris stent; Bard Peripheral Vascular Inc.) was deployed. Patients were monitored post-procedure with Doppler ultrasound performed within 48–72 hours to assess shunt patency and function.
Outcomes
The primary outcome was the efficacy of SEMS and TIPS in achieving hemostasis, defined as immediate bleeding control (hemostasis within 72 hours) and the absence of rebleeding within five days, in accordance with the Baveno VII guidelines. Secondary outcomes included the incidence of rebleeding, defined as evidence of recurrent bleeding from portal hypertensive sources per the Baveno V criteria (hematemesis, melena, aspiration of more than 100 mL of fresh blood in patients with a nasogastric tube, or drop in hemoglobin of 3 g/dL without blood transfusion), 30-day and 6 and 12-weeks all-cause mortality rates, and safety profiles of both interventions. Composite Patient-Reported Outcomes Measurement Information System (PROMIS)-10 Global Health scores were retrospectively extracted from medical records where patient-reported symptom assessments were documented as part of routine inpatient and follow-up clinical care. These data were not prospectively collected for research purposes, and the timing of assessment varied according to clinical circumstances.14
1) Follow-up
Data were reviewed and analyzed for every participant during their hospital stay for immediate postintervention outcomes, including control of active bleeding, hemodynamic stability, and early complications such as stent migration, ulcer perforation, or hepatic encephalopathy. Follow-up data at 1, 2, 4, 6, and 12 weeks were retrieved from institutional electronic health records. Each follow-up included clinical assessments, laboratory investigations, and imaging studies when necessary. Mortality, adverse events, rebleeding, and PROMIS-10 scores (if available) were recorded from clinical documentation. Mortality data were categorized as procedure-related or due to liver disease progression. Adverse events such as stent migration in SEMS and hepatic encephalopathy in TIPS were recorded in detail.
Data sources and measurement
Data were collected using a standardized form designed to capture comprehensive clinical information. The form was piloted and revised based on the feedback before full-scale data extraction. Two independent researchers collected the data to ensure accuracy and completeness. Discrepancies were resolved through discussion or consultation with a third reviewer who was blinded to the study.
Statistical methods
In this study, the analysis of continuous variables was performed using either means with standard deviations or medians with interquartile ranges, contingent on the distribution characteristics of the dataset. Categorical variables were summarized as frequencies and percentages. Continuous variables were compared using Student t-test or the Mann-Whitney U-test, depending on whether the data followed a normal distribution. This study followed STROBE reporting principles where applicable. As this was the first study to compare TIPS and SEMS efficacy in the management of post-EVL ulcer bleeding, a formal sample size calculation was not performed, and this was a hypothesis-generating study.
Ethical statement
Institutional ethical clearance was obtained the DMCH (Dayanand Medical College and Hospital) Ludhiana (DMCH/IEC/325) and the study was performed in accordance with the Declaration of Helsinki.
RESULTS
Thirty patients were enrolled; 15 were allocated to the SEMS group and 15 to the TIPS group (Fig. 2).
Baseline characteristics
The baseline characteristics of the study cohorts were comparable. Both groups, consisting of 15 patients each, exhibited similar demographic and clinical characteristics. The mean age of participants in the TIPS group was 52.93±8.1 years, closely paralleling the SEMS group at 50.67±9.85 years (Table 1). Males comprised 80.0% (n=12) and 66.7% (n=10) of the TIPS and SEMS cohorts, respectively. The mean duration of cirrhosis was comparable, at 5.73 ± 3.21 years for TIPS and 6.33±2.52 years for SEMS. Alcohol-related cirrhosis was the predominant etiology in both groups, accounting for 53.33% and 40% of the cases in the TIPS and SEMS cohorts, respectively. MASLD and Met-ALD were present in 4 (26.7%) and 3 (20.0%) patients, respectively, in the TIPS group and in 5 (33.3%) and 2 (13.3%) patients, respectively, in the SEMS group. Laboratory investigations, including hemoglobin, platelet count, international normalized ratio, and serum bilirubin levels, were similar between the groups, indicating comparable baseline hepatic and systemic functions. Disease severity stratification revealed that most patients were Child-Pugh Class C, with a median CTP score of 11 (interquartile range [IQR], 10–12) in both groups. The MELD scores further corroborated the advanced nature of liver disease, with a mean score of 20.4±1.91 in the TIPS group and 19.93±1.22 in the SEMS group.
Outcomes
The primary outcome, defined as immediate bleeding control within 72 hours with no rebleeding within 5 days, was achieved in 14/15 (93.3%) patients in the TIPS group and 15/15 (100%) patients in the SEMS group, demonstrating the robust hemostatic potential of both interventions (Table 2). Two patients in each group experienced rebleeding. However, mortality rates diverged markedly, with the TIPS group exhibiting a higher cumulative mortality rate of 6.67% at one week, 13.33% at six weeks, and 20% at 12 weeks than the SEMS group, which reported a solitary mortality event (6.67%) at 12 weeks attributable to liver failure (Fig. 3). These disparities highlight the effects of hepatic reserve and procedural safety on long-term survival. Quality of life outcomes, as retrospectively derived by the PROMIS-10 score, were significantly superior in the SEMS group (13.93±4.33) compared to the TIPS group (10.07±4.57, p<0.001), reflecting the broader clinical advantages of SEMS in this high-risk population (Fig. 4). Additionally, the SEMS group demonstrated a reduced median intensive care unit (ICU) stay of 2 days (IQR, 2–3 days) versus 5 days (IQR, 3–6 days) in the TIPS group (p=0.002).
Adverse events
Hepatic encephalopathy, a well-documented complication of TIPS, was observed in 7 (46.7%) patients in the TIPS group, exceeding the 3 (20.0%) noted in the SEMS group. Arterial puncture occurred in one patient during TIPS insertion and was recognized intraprocedurally. The event was conservatively managed with manual compression and close hemodynamic monitoring. No patient required surgical or endovascular intervention, and there were no subsequent bleeding-related sequelae. Stent migration was observed in three patients following SEMS placement. Migration was identified via routine radiographic reassessment or clinical suspicion of recurrent symptoms. In all cases, the migrated stent was managed endoscopically by repositioning. No patient experienced esophageal perforation or procedure-related mortality attributable to stent migration.
DISCUSSION
The present single-center retrospective observational study describes the short-term clinical outcomes following SEMS or TIPS placement for refractory post-EVL ulcer bleeding. The study included 30 patients, with 15 in each group, followed up for 12 weeks. The study demonstrated that the primary outcome, that is, immediate hemostasis (within 72 hours), was achieved in 100% and 93.3% of the patients, respectively. The rebleeding rate of 13.33% was similar between the two groups, further emphasizing the comparable clinical efficacy of the two interventions. Importantly, the mortality rate was 6.67% in the SEMS group and 20% in the TIPS group. This difference in mortality between TIPS and SEMS group might stem from difference in underlying diminished hepatic and systemic reserves in both groups. However, given the retrospective design, small sample size, absence of a predefined non-inferiority margin, and limited statistical power, this study was not designed to formally compare SEMS and TIPS or establish equivalence or non-inferiority. Therefore, all the observed differences should be interpreted as descriptive or exploratory.
Several systematic reviews and meta-analyses have evaluated the role of SEMS in refractory variceal bleeding.11,15 A recent randomized controlled trial (RCT) by Singh et al.16 directly compared TIPS and SEMS in patients with advanced cirrhosis and refractory variceal bleeding and concluded higher survival rate in TIPS group compared to SEMS (10% vs. 52.4%, p=0.015) during 6 weeks of intervention. Furthermore, the risks of very early rebleeding (<48 hours) and rebleeding (48 hours to 5 days) were higher in the SEMS group than in the TIPS group (5/21 [23.8%] vs. 1/20 [5.0%], p=0.07; 4/21 [19.0%] vs. 0/20 [0%], p=0.03, respectively). However, our retrospective study specifically addressed the refractory PEBU population and was not intended to replicate a randomized head-to-head comparison.
The hemostasis rate of our study are consistent with the hemostasis rates reported by an indirect comparison meta-analysis by Mohan et al.,17 where SEMS achieved an 84.5% hemostasis rate and TIPS demonstrated a higher rate of 97.9%. This higher success rate of SEMS in present study can stem from a combination of multiple factors such as underlying hepatic and systemic reserve of patients, less duration of time elapsed between presentation to hospital and procedure, promptness and ease of the procedure, to be able to done even bedside in the same setting of doing endoscopy to localize the source of bleed, enhance endoscopic expertise and better deployment techniques with the advancement of science, whereas TIPS might take time of shifting the patient to particular lab with more time elapsed leading to further diminishing of the depleted reserve. This suggests that SEMS has evolved into robust therapeutic options.
Mortality rates in the TIPS group (20% at 12 weeks) mirror findings from prior studies, where TIPS mortality ranged from 15 to 30%, largely attributable to hepatic encephalopathy and liver failure.18-20 By contrast, the lower mortality in the SEMS group (6.67%) highlights its safety, particularly in patients with advanced liver dysfunction. However, this discrepancy may stem from selection bias, as in routine clinical practice, TIPS is often contraindicated in patients with severe hepatic dysfunction, advanced encephalopathy, or multi-organ failure. Consequently, patients selected for SEMS placement may have had relatively better preserved liver function, which could partly account for the favorable outcomes observed in this group. Therefore, multicenter RCTs are required to validate these findings.
The inclusion of quality of life as an endpoint, an unexplored aspect, makes this study the first of its kind. SEMS group had a significantly higher PROMIS-10 quality of life score (13.93±4.33 vs. 10.07±4.57, p<0.001), compared to TIPS group. PROMIS-10 assessments were retrospectively extracted from routine clinical documentation with non-standardized timing and were not collected using a prespecified research protocol. SEMS facilitates early enteral nutrition and mobility, which are critical factors in reducing bacterial translocation from the gut and lowering the risk of hepatic encephalopathy. In our cohort, hepatic encephalopathy occurred in 20% of the SEMS group compared to 46.67% of the TIPS group, which is consistent with earlier findings in which TIPS-associated hepatic encephalopathy rates ranged from 30 to 50%. The shorter ICU stay in the SEMS group (median 2 vs. 5 days, p=0.002) further underscores its potential to reduce healthcare resource utilization, which is an important consideration in resource-limited settings. However, SEMS is not without risks, with 20% of patients experiencing stent migration, comparable to previously reported rates of 15%–25%.
Adverse events in the TIPS group, including arterial puncture (6.67%) and high rates of hepatic encephalopathy (46.67%), emphasize the procedural complexity and potential drawbacks. These complications have been consistently reported in previous studies, where arterial puncture and vascular injury rates for TIPS ranged from 5 to 10%.21-23 While stent migration was noted in the SEMS group (20%), it did not result in severe morbidity or mortality, which is consistent with prior studies that emphasized the overall safety of SEMS when carefully monitored.
The strengths of this study include its retrospective design, evaluation of the efficacy of SEMS and TIPS in patients treated at a single center during the study period for post-EVL ulcer bleeding, comprehensive follow-up, and integration of patient-reported outcomes, marking a significant advancement in the evaluation of therapies for refractory post-EVL ulcer bleeding. However, limitations such as the retrospective design, small sample size, and non-randomized treatment allocation preclude causal inference or equivalence testing between SEMS and TIPS. The sample size in our study was small, which can be attributed to the lower prevalence of refractory post-EVL ulcer bleeding and the limited availability of SEMS and TIPS at specialized centers. Moreover, the quality of life assessment using the PROMIS-10 was performed retrospectively based on routine clinical documentation rather than prospective, standardized administration. As a result, these findings may be subject to documentation bias, incomplete data capture, and variability in the timing of assessment and should therefore be interpreted cautiously. Lastly, data on subsequent definitive therapies, such as delayed TIPS after SEMS placement or liver transplantation, were not available; therefore, the impact of these downstream interventions on long-term outcomes could not be assessed, and further RCTs with longer follow-up periods are required. Despite these constraints, our findings provide compelling evidence that SEMS placement is a rapid, effective, and patient-centered alternative to TIPS.
Although SEMS has been suggested as a pragmatic option in settings where immediate access to TIPS is constrained, the present study was conducted at a tertiary care center and did not directly compare outcomes across resource-limited and resource-rich environments. As such, any implications regarding applicability in resource-limited settings should be regarded as extrapolative and hypothesis generating rather than evidence based.
In conclusion, this retrospective cohort study showed that both SEMS and TIPS were associated with effective short-term control of refractory post-EVL ulcer bleeding. Within the limits of the study design, SEMS was associated with fewer observed complications, while the quality-of-life findings were exploratory and should be interpreted cautiously. Nonetheless, SEMS placement is associated with earlier enteral nutrition and mobilization. However, given the retrospective design, small sample size, absence of a predefined non-inferiority margin, and limited statistical power, this study was not designed to formally compare SEMS and TIPS or establish equivalence or non-inferiority. Therefore, all the observed differences should be interpreted as descriptive or exploratory.
Importantly, this analysis did not establish equivalence or superiority between SEMS and TIPS, nor did it address long-term treatment trajectories or downstream definitive therapies. While SEMS placement is commonly conceptualized as a bridging intervention and TIPS as a definitive portal decompressive strategy, the present study focused on early clinical outcomes following the index intervention. Collectively, our results suggest that SEMS represents a feasible salvage or bridge therapy for selected patients, particularly when immediate TIPS is not feasible. Larger prospective multicenter studies are required to define the optimal positioning of SEMS relative to TIPS and to assess its role in diverse clinical settings.
Conflicts of Interest
The authors have no potential conflicts of interest.
Funding
None.
Author Contributions
Conceptualization: VM, AG, YKG, AS, MKG, ARV; Data curation: VM, AG, YKG; Formal analysis: VM, AG, YKG, MKG; Funding acquisition: VM; Investigation: VM, MK; Methodology: VM; Project administration: VM, MKG; Resources: VM; Software: VM; Supervision: VM, AS; Validation: VM, AS, MKG; Visualization: VM, YKG, AS, MKG; Writing–original draft: VM, AG, YKG, MK, MKG, ARV; Writing–review & editing: all authors.
Fig. 1.Representative endoscopic images illustrating key steps in self-expandable metal stent (SEMS) placement for refractory post-endoscopic variceal ligation (EVL) Ulcer Bleeding. (A) Endoscopic view of esophageal varices prior to any intervention. (B) Endoscopic view of EVL being done (C) Active spurting bleed from a post-banding ulcer in the distal esophagus prior to SEMS deployment. (D) Successfully deployed SEMS with full expansion, tamponading the ulcer bed and adjacent variceal columns.
Fig. 2.Study flow diagram. Flowchart summarizing patient enrollment and study allocation. Of 43 screened patients with post-endoscopic variceal ligation bleeding ulcers, 13 were excluded due to incomplete data (n=6), prior self-expandable metal stent (SEMS)/transjugular intrahepatic portosystemic shunt (TIPS) procedures (n=4), or bleeding from alternative sources (n=3). Thirty patients were included and evenly assigned to the TIPS (n=15) and SEMS (n=15) arms. Primary outcome (immediate hemostasis within 72 hours) was achieved in 14 patients in the TIPS group and 15 in the SEMS group.
Fig. 3.Kaplan-Meier survival curve comparing transjugular intrahepatic portosystemic shunt (TIPS) and self-expandable metal stent (SEMS) groups over 12 weeks. Survival analysis showing cumulative survival over a 12-week follow-up. The SEMS group (group 2) exhibited superior survival probability (0.93 at 12 weeks) compared to the TIPS group (group 1), which showed a decline to 0.80 by week 12. Survival curves illustrate a trend toward better outcomes in the SEMS group, although differences were not statistically significant. CI, confidence interval; std. error, standard error.
Fig. 4.Patient-Reported Outcomes Measurement Information System (PROMIS) physical quality of life scores at week 6 in transjugular intrahepatic portosystemic shunt (TIPS) vs. self-expandable metal stent (SEMS) groups. Boxplot comparison of PROMIS physical health scores between patients receiving TIPS and SEMS at 6 weeks. Median (interquartile range) score in the SEMS group was 15 (14–16) compared to 11 (9–13) in the TIPS group, indicating a significantly higher quality of life in the SEMS cohort. The mean scores are shown in blue colour. QOL, quality of life.
Table 1.Baseline characteristics distribution
|
Baseline characteristic |
TIPS |
SEMS group |
|
Total |
15 |
15 |
|
Age (yr) |
52.93±8.1 |
50.67±9.85 |
|
Sex (female:male) |
3 (20.0):12 (80.0) |
5 (33.3):10 (66.7) |
|
Duration of cirrhosis (yr) |
5.73±3.21 |
6.33±2.52 |
|
Aetiology |
|
|
|
Alcohol |
8 (53.3) |
6 (40.0) |
|
MASLD |
4 (26.7) |
5 (33.3) |
|
MET-ALD |
3 (20.0) |
2 (13.3) |
|
Hepatitis B related |
0 (0) |
1 (6.7) |
|
Hepatitis C related |
0 (0) |
1 (6.7) |
|
Others |
0 (0) |
0 (0) |
|
Comorbidities |
9 (60.0) |
11 (73.3) |
|
Hypertension |
5 (33.3) |
3 (20.0) |
|
Diabetes mellitus |
3 (20.0) |
6 (40.0) |
|
Any other |
1 (6.7) |
2 (13.3) |
|
Past history |
|
|
|
No. of prior hospitalizations |
4 (2–4) |
3 (1–4) |
|
Previous episodes of decompensation |
4 (1–4) |
3 (2–4) |
|
Prior episodes of EVL |
2 (0–1) |
1 (0–1) |
|
At the time of EVL |
|
|
|
Vitals |
|
|
|
Heart rate (beats per minute) |
117.73±4.89 |
106.93±5.94 |
|
Mean arterial pressure (mmHg) |
72.26±4.68 |
75.73±2.15 |
|
Inotropic support |
4 (26.7) |
3 (20.0) |
|
Hepatic encephalopathy |
|
|
|
None |
6 (40.0) |
7 (46.7) |
|
Grade I |
9 (60.0) |
6 (40.0) |
|
Grade II |
0 (0) |
2 (13.3) |
|
Grade III |
0 (0) |
0 (0) |
|
Grade IV |
0 (0) |
0 (0) |
|
Ascites |
|
|
|
None |
3 (20.0) |
3 (20.0) |
|
Mild |
8 (53.3) |
8 (53.3) |
|
Moderate |
3 (20.0) |
4 (26.7) |
|
Severe |
1 (6.7) |
0 (0) |
|
Lab Investigations |
|
|
|
Hemoglobin (g/dL) |
62.2±1.8 |
68.7±2.6 |
|
TLC (×109/L) |
12.1±1.56 |
11.93±1.63 |
|
Platelet count (×109/L) |
66.33±17.31 |
70.6±20.4 |
|
INR |
2.36±0.19 |
2.17±0.14 |
|
Serum bilirubin (μmol/L) |
59.69±12.48 |
61.23±13.85 |
|
Serum albumin (g/L) |
26.13±3.38 |
25.55±3.70 |
|
Serum sodium (mmol/L) |
128.2±4.91 |
131.06±3.89 |
|
Serum creatinine (μmol/L) |
66.95±17.68 |
69.83±20.33 |
|
Serum BUN (mmol/L) |
13.76±1.77 |
11.77±2.01 |
|
Serum lactate (mmol/L) |
2.1±0.45 |
1.97±0.37 |
|
Disease stratification |
|
|
|
CTP score |
11 (10–12) |
11 (10–12) |
|
Child class (A/B/C) |
0 (0)/3 (20.0)/12 (80.0) |
0 (0)/3 (20.0)/12 (80.0) |
|
MELD score |
20.4±1.91 |
19.93±1.22 |
|
Units of blood transfused |
2 (0–3) |
2 (1–3) |
|
Grade of varix |
|
|
|
Small |
0 (0) |
0 (0) |
|
Large |
15 (100.0) |
15 (100.0) |
|
Bands deployed |
3 (3–4) |
3 (3–4) |
|
At the time of post-EVL ulcer bleed |
|
|
|
Vitals |
|
|
|
Heart rate (beats per minute) |
112±7.74 |
111.87±8.50 |
|
Mean arterial pressure (mmHg) |
73.8±2.95 |
74±2.77 |
|
Inotropic support |
4 (26.7) |
3 (20.0) |
|
Hepatic encephalopathy |
|
|
|
None |
6 (40.0) |
7 (46.7) |
|
Grade I |
9 (60.0) |
6 (40.0) |
|
Grade II |
0 (0) |
2 (13.3) |
|
Grade III |
0 (0) |
0 (0) |
|
Grade IV |
0 (0) |
0 (0) |
|
Ascites |
|
|
|
None |
3 (20.0) |
2 (13.3) |
|
Mild |
8 (53.3) |
8 (53.3) |
|
Moderate |
3 (20.0) |
5 (33.3) |
|
Severe |
1 (6.7) |
0 (0) |
|
Lab investigations |
|
|
|
Hemoglobin (gm/L) |
70.6±3.82 |
72.7±4.5 |
|
TLC (×109/L) |
15.18±1.68 |
14.38±2.53 |
|
Platelet count (×109/L) |
68.86±22.53 |
71.4±26.14 |
|
INR |
2.41±0.32 |
2.18±0.42 |
|
Serum bilirubin (μmol/L) |
66.02±17.1 |
69.43±22.4 |
|
Serum albumin (g/L) |
26.7±2.39 |
27.34±2.6 |
|
Serum sodium (mmol/L) |
129.4±3.75 |
131.93±4.3 |
|
Serum creatinine (μmol/L) |
75.02±16.79 |
76.91±18.56 |
|
Serum BUN (mmol/L) |
15.05±2.88 |
12.04±2.22 |
|
Serum lactate (mmol/L) |
2.62±0.71 |
2.03±0.46 |
|
Disease stratification |
|
|
|
CTP score |
12 (11–13) |
11 (11–12) |
|
Child class (A/B/C) |
0 (0)/1 (6.7)/14 (93.3) |
0 (0)/3 (20.0)/12 (80.0) |
|
MELD score |
21.4±1.91 |
20.8±2.37 |
|
Post-EVL presentation time (day) |
9 (8–10) |
9 (8–9) |
|
Units of blood transfused |
4 (3–4) |
4 (3–5) |
|
Ulcer classification |
|
|
|
Type A |
4 (26.7) |
3 (20.0) |
|
Type B |
3 (20.0) |
4 (26.7) |
|
Type C |
5 (33.3) |
5 (33.3) |
|
Type D |
3 (20.0) |
3 (20.0) |
Table 2.Outcome distribution
|
Outcome |
TIPS (n=15) |
SEMS (n=15) |
p-value |
|
Primary outcome |
|
|
|
|
Immediate control of bleeding (within 72 h) |
14 (93.3) |
15 (100.0) |
- |
|
Secondary outcomes |
|
|
|
|
Rebleeding |
2 (13.3) |
2 (13.3) |
1 |
|
Mortality |
|
|
|
|
1 Week |
1 (6.7) |
0 (0) |
- |
|
Bleed related |
1 (6.7) |
0 (0) |
- |
|
Underlying liver failure related |
0 (0) |
0 (0) |
- |
|
6 Weeks |
2 (13.3) |
0 (0) |
0.54 |
|
Bleed related |
1 (6.7) |
0 (0) |
- |
|
Underlying liver failure related |
1 (6.7) |
0 (0) |
- |
|
12 Weeks |
3 (20.0) |
1 (6.7) |
0.28 |
|
Bleed related |
1 (6.7) |
0 (0) |
- |
|
Underlying liver failure related |
2 (13.3) |
1 (6.7) |
- |
|
Technical success (%) |
100 |
100 |
1 |
|
Post-procedure units of blood transfused |
2 (1–2) |
1 (1–2) |
0.03a)
|
|
Physical quality of life (PROMIS-10) |
10.07±4.57 |
13.93±4.33 |
<0.0001c)
|
|
Duration of ICU stay (day) |
5 (3–6) |
2 (2–3) |
0.0022a)
|
|
Duration of hospital stay (day) |
15 (15–15) |
11 (10–11) |
<0.01a)
|
|
Duration of procedure |
82±11.61c)
|
200±27d)
|
<0.001b)
|
|
Post-procedure decompensation |
|
|
|
|
Ascites |
0 (0) |
5 (33.3) |
- |
|
Hepatic encephalopathy |
7 (46.7) |
3 (20.0) |
0.121 |
|
Hepatorenal syndrome |
0 (0) |
0 (0) |
- |
|
Adverse events |
1 (6.7), arterial puncture |
3 (20.0), SEMS migration |
0.28 |
REFERENCES
- 1. European Association for the Study of the Liver. EASL Clinical Practice Guidelines for the management of patients with decompensated cirrhosis. J Hepatol 2018;69:406–460.ArticlePubMed
- 2. de Franchis R, Bosch J, Garcia-Tsao G, et al. Baveno VII: renewing consensus in portal hypertension. J Hepatol 2022;76:959–974.ArticlePubMed
- 3. Lim EJ, Gow PJ, Angus PW. Endoscopic variceal ligation for primary prophylaxis of esophageal variceal hemorrhage in pre-liver transplant patients. Liver Transpl 2009;15:1508–1513.ArticlePubMedPDF
- 4. Song JE, Kim BS. Endoscopic therapy and radiologic intervention of acute gastroesophageal variceal bleeding. Clin Endosc 2019;52:407–415.ArticlePubMedPMCPDF
- 5. Jamwal KD, Maiwall R, Sharma MK, et al. Case control study of post-endoscopic variceal ligation bleeding ulcers in severe liver disease: outcomes and management. J Clin Transl Hepatol 2019;7:32–39.ArticlePubMedPMC
- 6. Rajesh S, George T, Philips CA, et al. Transjugular intrahepatic portosystemic shunt in cirrhosis: an exhaustive critical update. World J Gastroenterol 2020;26:5561–5596.ArticlePubMedPMC
- 7. Elbahr O, Kamal A, Alsebaey A, et al. Self-expandable metallic stents (SEMS) in esophageal varices post-band ulcer refractory bleeding: a retrospective study. Egypt Liver J 2021;11:30.ArticlePDF
- 8. Mohan BP, Chandan S, Khan SR, et al. 596 Efficacy of self expanding metal stent (SEMS) in refractory bleeding esophageal varices, is there a mortality benefit? An indirect-comparison meta-analysis to trans-jugular intra-hepatic porto-systemic shunt (TIPS). Am J Gastroenterol 2019;114:S346–S347.Article
- 9. Müller M, Seufferlein T, Perkhofer L, et al. Self-expandable metal stents for persisting esophageal variceal bleeding after band ligation or injection-therapy: a retrospective study. PLoS One 2015;10:e0126525.ArticlePubMedPMC
- 10. Khan S, Gilhotra R, Di Jiang C, et al. The role of a novel self-expanding metal stent in variceal bleeding: a multicenter Australian and New Zealand experience. Endosc Int Open 2022;10:E238–E245.ArticlePubMedPMC
- 11. McCarty TR, Njei B. Self-expanding metal stents for acute refractory esophageal variceal bleeding: A systematic review and meta-analysis. Dig Endosc 2016;28:539–547.ArticlePubMedPMC
- 12. Maiwall R, Jamwal KD, Bhardwaj A, et al. SX-ella stent danis effectively controls refractory variceal bleed in patients with acute-on-chronic liver failure. Dig Dis Sci 2018;63:493–501.ArticlePubMedPDF
- 13. Choudhary NS, Puri R, Saigal S, et al. Innovative approach of using esophageal stent for refractory post-band ligation esophageal ulcer bleed following living donor liver transplantation. J Clin Exp Hepatol 2016;6:149–150.ArticlePubMedPMC
- 14. Hays RD, Bjorner JB, Revicki DA, et al. Development of physical and mental health summary scores from the patient-reported outcomes measurement information system (PROMIS) global items. Qual Life Res 2009;18:873–880.ArticlePubMedPMCPDF
- 15. Songtanin B, Kahathuduwa C, Nugent K. Esophageal stent in acute refractory variceal bleeding: a systematic review and a meta-analysis. J Clin Med 2024;13:357.ArticlePubMedPMC
- 16. Singh S, Arora V, Maiwall R, et al. TIPS Is Superior to SEMS in the management of refractory VARICEAL bleed in advanced cirrhosis patients (TIPSEMS-VB Trial). Liver Int 2025;45:e70178.ArticlePubMed
- 17. Mohan BP, Chandan S, Khan SR, et al. Self-expanding metal stents versus TIPS in treatment of refractory bleeding esophageal varices: a systematic review and meta-analysis. Endosc Int Open 2020;8:E291–E300.ArticlePubMedPMC
- 18. Dissegna D, Sponza M, Falleti E, et al. Morbidity and mortality after transjugular intrahepatic portosystemic shunt placement in patients with cirrhosis. Eur J Gastroenterol Hepatol 2019;31:626–632.ArticlePubMed
- 19. Adlakha N, Russo MW. Outcomes after transjugular intrahepatic portosystemic shunt in cirrhotic patients 70 years and older. J Clin Med 2020;9:381.ArticlePubMedPMC
- 20. Ascha M, Abuqayyas S, Hanouneh I, et al. Predictors of mortality after transjugular portosystemic shunt. World J Hepatol 2016;8:520–529.ArticlePubMedPMC
- 21. Patel RK, Chandel K, Tripathy TP, et al. Complications of transjugular intrahepatic portosystemic shunt (TIPS) in the era of the stent graft - What the interventionists need to know? Eur J Radiol 2021;144:109986.ArticlePubMed
- 22. Suhocki PV, Lungren MP, Kapoor B, et al. Transjugular intrahepatic portosystemic shunt complications: prevention and management. Semin Intervent Radiol 2015;32:123–132.ArticlePubMedPMC
- 23. Yin X, Gu L, Zhang M, et al. Covered TIPS procedure-related major complications: incidence, management and outcome from a single center. Front Med (Lausanne) 2022;9:834106.ArticlePubMedPMC
Citations
Citations to this article as recorded by
KSGE
, Manjeet Kumar Goyal2
Cite
XML Download
