Management of acute variceal bleeding with standard treatment is effective in most patients. However, treatment of refractory AVB still remains a challenge. According to Baveno Ⅵ conference, e-PTFE covered TIPS must be considered in patients presenting failure to control bleeding despite combined medical and endoscopic therapy, and balloon tamponade should be used temporarily until TIPS can be performed[2]. Thus, balloon tamponade sequentially combined with TIPS is a feasible strategy theoretically. However, there are few data in the literature on the safety and efficacy of this therapeutic strategy. Escorsell et al.[4] have performed TIPS as a definitive treatment to prevent rebleeding after BT in part of their patients, but nearly half of their patients received TIPS as a rescue therapy when BT failed. Also, their research mainly focused on the comparison between esophageal stent versus BT and did not report this strategy in detail.
Our research was intentionally designed to evaluate the strategy of balloon tamponade sequentially combined with TIPS. The success rate of TIPS procedure reached to 100% without developing procedure-related complications, and all patients remained free of recurrent bleeding during follow-up. These results were relatively better than previous studies [5, 18, 19]. This improvement may be attributed to the following reasons. Firstly, compared with performing TIPS only, BT can control gastroesophageal variceal bleeding rapidly and effectively, which allows to correct hemodynamic instability or other concomitant diseases so that patients are in better conditions to receive TIPS. Previous researches have demonstrated that for high-risk patients, failure of initial treatment will lead to further deterioration in liver function, which is the predictor of poor outcomes, and may also preclude the use of TIPS[5, 18–20]. Therefore, this strategy may improve technical success rate and decrease the risk of poor prognosis because of the stable conditions for TIPS placement provided by BT. Secondly, compared with using BT to control hemorrhage only, this strategy can shorten the compression time due to the early use of TIPS, which in turn reduces the incidence of balloon-related complications. Meanwhile, the long-term hemostasis effect of TIPS is superior to BT and even endoscopic therapy [5, 21, 22], so the incidence of rebleeding was lower than that of using balloon only. In a study involving patients with AVB who received BT for 24 h, complications occurred in 30% of patients[11]. Similarly, another study involving patients treated with BT for 48 h, the incidence of complications was 33% and the rebleeding rate was 58%[23]. In our study, except Patient 11 who received BT for a total of 65 h had vomiting occasionally, other patients with an average compression time of 15 h (range:4.5–43 h), which was lower than that of aforementioned studies, did not develop balloon-related complications. Thus, it indicates that balloon tamponade sequentially combined with emergency TIPS can reduce the incidence of variceal rebleeding and balloon-related complications.
Balloon was deflated during TIPS procedure in seven patients in this study. Different from standard TIPS procedure, balloon was deflated once access to the portal vein was achieved, then we performed portal venography to determine whether to embolize varices depending on their numbers and sizes under DSA and the severity of portal hypertension. Afterwards, a standard TIPS procedure was performed. This modified TIPS procedure has two advantages. Firstly, once the portal vein is accessed, embolization can be conducted promptly even if bleeding restarts after balloon deflation, which in turn increases the safety of TIPS procedure and has a better hemostasis effect than deflating balloon before TIPS. Secondly, the compression time can be further shortened on the basis of deflating balloon after finishing TIPS and a lower risk of complications may be achieved. Thus, our findings suggest that balloon deflation during TIPS is superior to balloon deflation before and after TIPS. It should be pointed out that balloon must be deflated before portography, otherwise contrast medium may not be observed in the varices (Fig. 2).
Hepatic encephalopathy is one of the most severe complications after TIPS placement, which is related to the severity of liver insufficiency and shunt diameter[16, 24, 25]. OHE is the decompensated phase of HE and the 1-year cumulative incidence after TIPS ranges from 10–50% [16, 26]. Our incidence of OHE after TIPS was 21%, which was similar to previous researches[16, 26]. According to current guidelines, no definitive prophylactic therapy is recommended to prevent against post-TIPS HE[16]. Therefore, controlling precipitating factors of HE and treating episodes of OHE actively is of vital importance. 3 patients had a total of 7 episodes of OHE in our study, which were mainly precipitated by high-protein diet. This case highlights the importance of education about strict protein nutrition intake for patients (especially for patients with poor compliance) receiving TIPS placement. Shunt dysfunction was defined as the portocaval gradient ≥ 12 mmHg or a stent lumen stenosis ≥ 50%[27] after TIPS placement. No patients experienced shunt dysfunction during follow-up and the rate of maintaining shunt patency was better than that in previous studies[27, 28].
Although several randomized controlled trials (RCTs) and meta-analysis have been performed, there are no universally accepted standards to date for the evaluation of survival after TIPS treatment. In our study, the 6-week survival rate was significantly higher than the strategy of inserting balloon only or performing rescue TIPS only (100% vs 57% and 64%, respectively)[13, 29, 30]. Meanwhile, the 1-year survival rate was improved compared with performing rescue TIPS only (76% vs 58%)[29], and both two deaths were neither caused by operation nor by portal hypertension-related complications. Hence, BT sequentially combined with TIPS demonstrated a beneficial effect on survival compared with using BT or rescue TIPS only, and this effect might be attribute to better control of factors leading to death, including failure to control bleeding, recurrent rebleeding and severe procedure-related complications. However, no firm conclusions can be drawn because of the small number of patients receiving this strategy.
Recent researches report a self-expandable esophageal metal stent might be an alternative to BT in managing refractory AVB because of its higher effectiveness and less adverse events[4, 13]. However, esophageal stent has some drawbacks such as a greater cost and sophisticated procedure of deployment and retrieval, and mortality still remains high despite combining with definitive treatment to prevent rebleeding (non-selective beta-blockers + EVL)[4]. In addition, a higher bleeding control rate and fewer complications was achieved by combining BT and TIPS in our study. Since esophageal stent can be maintained for over a week, it may have a higher value in medical centers which could not provide emergency TIPS. But for centers that are able to provide TIPS readily, we favor use of balloon tamponade combined with TIPS for refractory AVB.
This study has some limitations that should be considered. Firstly, the hemostatic effect of SB tube on esophageal varices is better than on gastric varices (Linton-Nachlas tube or endoscopic therapy with tissue adhesive is preferred)[17], and the type of varices were mainly EV and GOV Ⅰ in our research. Thus, whether our strategy equally benefits GOV Ⅱ and IGV patients has not been confirmed. Secondly, two cases rebled after balloon deflation among the patients deflating balloon before TIPS procedure, and received a second BT until TIPS was conducted. Thereby we believe the strategy of deflating balloon during TIPS may prevent failure to control bleeding and achieve better results in these cases. Finally, our research had a small number of study populations and a relatively short follow-up period, and was conducted as a retrospective study. Therefore, further researches, especially RCTs including a larger study group and a longer follow-up period are needed.