Management of Refractory Esophageal Variceal Bleeding when TIPSS Is Not Possible: Review of a Challenging Case

Esophageal variceal bleeding (EVB) carries significant morbidity and mortality in patients with cirrhosis. Following appropriate resuscitation measures, initiation of splanchnic vasoconstrictors, and antimicrobial prophylaxis, endoscopic variceal ligation (EVL) is the first-line endoscopic therapy. Despite this, up to 18% of patients will experience early rebleeding by 2 weeks [1, 2]. To reduce early rebleeding and improve long-term survival, guidelines recommend a pre-emptive (inserted within 72 h) trans-jugular intrahepatic portosystemic shunt (TIPSS) in patients with Child-Pugh (CP) class C (10–13), CP 8–9 with evidence of active bleeding on endoscopy [3, 4]. Unfortunately, TIPSS is either not feasible or contra-indicated in up to 35% of cases due to anatomical or medical considerations [5]. In these cases, secondary prevention includes a combination of non-selective β-blockers and repeated EVL [3, 4].

We recently encountered an otherwise-well 61-year-old man with a first presentation of HCV cirrhosis with acute EVB (Model for End-Stage Liver Disease-Sodium was 12 (MELD-Na), CP score B-9). The patient’s initial heart rate and blood pressure were stable, and biochemical testing showing hemoglobin of 8.9 g/dL, requiring no blood transfusion. Following standard therapy, the patient was admitted to the intensive care unit and underwent upper endoscopy showing actively bleeding large EVs which were successfully treated by EVL (Fig. 1a). As part of the evaluation, a computed tomography scan confirmed the presence of cirrhosis with large amount of ascites and splenomegaly. Moreover, we discovered a 13.5 cm × 10.5 cm centrally located right liver lobe tumor with complete tumor thrombosis of the main portal vein and right and left branches (Fig. 1b), anticoagulation was discouraged due to EVB. Following discussion with our interventional radiology team, TIPSS placement was deemed not possible. After completion of vasoactive treatment, carvedilol was started at 12.5 mg per day, and the patient was scheduled for repeat EVL. Unfortunately, the patient experienced recurrent EVB on day 15 and retreated with EVL. We discussed with the patient the very high chance of rebleeding, and the very high short-term mortality associated. He shared his wishes to avoid a highly invasive procedure and to favor endoscopic management. Unfortunately, he re-experienced EVB on day 21, treated with combination of EVL and tromboject (sodiumtretradecyl sulfate), achieving hemostasis. At this stage, the patient agreed to pursue an alternative therapeutic approach with embolization of left gastric vein (LGVE) as it was feeding the EVs. On day 22, using a trans-splenic approach under fluoroscopic guidance, LGV was embolized using vascular coils proximally and a mixture of sodium tetradecyl sulfate 2%, Lipiodol and air distally. This resulted in immediate obliteration of the EVs (Fig. 1c, d). The patient was discharged 5 days later with carvedilol with outpatient follow-up.

However, the patient was readmitted 3 weeks later with worsening ascites and weakness with a MELD-Na of 20. Although he was hemodynamically stable, laboratory tests revealed a drop in hemoglobin, prompting a repeat gastroscopy. A large ulcer was noted in the distal esophagus at the site of repeated endoscopic interventions without bleeding. Abdominal imaging confirmed no perforation, and he was managed conservatively. He was discharged home tolerating oral diet with palliative support and comfort care. He unfortunately passed away at home, 60 days following his initial presentation.

While a pre-emptive TIPSS carries excellent long-term outcomes in managing EVB, a significant proportion of patients cannot benefit from this approach due to the existence of certain contraindications. Other treatment options such as balloon tamponade and self-expandable metallic stents have been proposed in managing refractory EVB, but these measures are considered a bridge to more definitive treatment [6]. Salvage surgical interventions, such as selective esophagogastric devascularization or splenectomy, are invasive and require careful patient selection [7].

In terms of pathophysiology, EVs result from an increase in flow and resistance within the portosystemic circulation. These collaterals can be supplied and drained by several pathways. The most common feeding vein for EVs is the LGV in 94% of cases [8]. To date, LGVE has been explored as an adjunct to TIPSS creation. In a multicenter propensity-matched cohort, Calame et al. [9] compared TIPSS alone versus TIPSS with LGVE in pre-emptive or salvage EVB. This study failed to show a difference in 1-year rebleeding rate between the two groups (9% vs 11%, p value = 0.7) [9]. This observation perhaps can be explained by the major reduction in portal pressure gradient following shunt creation alone, driving a reduction in bleeding recurrence.

In our case, anatomical considerations made TIPSS not possible. As LGV was identified as the main supplier for the EVs, we opted for a trans-splenic approach for LGVE. The implementation of this approach proved successful in achieving hemostasis, with no subsequent recurrence of bleeding. As noted in our case, embolization of a major portosystemic shunt such as the LGV can lead to an increase in portal hypertension as manifested by worsening ascites. This approach is justifiable in the emergency setting where other measures have failed or were not possible. We speculate that the targeted LGVE could be contemplated in instances where anatomical complexities or contraindications to TIPSS arise. Our cases raise additional questions that need further studies related to patient selection and timing of such an intervention when TIPSS is not possible.

All procedures followed were in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Written informed consent was obtained from the patient for publication of this case report and any accompanying images. Informed consent was obtained from the patient for publication of the case details. Ethical approval is not required for this study in accordance with local or national guidelines. The CARE Checklist has been completed by the authors for this case report, attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000543537).

Ali Bessissow and Yen-I Chen are consultants for Boston Scientific and are the co-founders of ChessMedical. Other authors have no conflict of interest.

This study was not supported by any sponsor or funder.

Abdulrahman Qatomah analyzed the patient’s clinical course and performed literature review, manuscript writing, and preparation. Ali Bessissow participated in direct patient care and edited the manuscript. Yen-I Chen participated in manuscript writing and preparation. Talat Bessissow and Amine Benmassaoud participated in direct patient care and manuscript writing.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.