Partial splenic embolization combined with endoscopic therapies and NSBB decreases the variceal rebleeding rate in cirrhosis patients with hypersplenism: a multicenter randomized controlled trial

Abstract

Background: Global research on endoscopic therapies in combination with partial splenic embolization (PSE) for variceal hemorrhage (VH) is limited. Therefore, we aimed to evaluate the efficacy and safety of endoscopy plus PSE (EP) treatment in comparison to endoscopic (E) treatment for the secondary prophylaxis of VH in cirrhosis patients with hypersplenism.

Methods: Cirrhosis patients with hypersplenism (platelet count < 100, 000/µL) and those who had recovered from an episode of VH were enrolled in a multicenter randomized controlled trial. The participants were randomly assigned into EP and E groups in a 1:1 ratio. The primary endpoint was variceal rebleeding, and the secondary endpoints were severe variceal recurrence and mortality during the 2-year follow-up. Hematological indices, serum biochemical parameters, and the Child-Pugh score were measured at each time point.

Results: From June 2016 to December 2019, 108 patients were enrolled in the study, among which 102 patients completed the protocol (51 in EP and 51 in E group). The rebleeding rate of the varices was significantly reduced in the EP group compared to that in the E group during the 2 years (16% vs. 31%, p < 0.001). The EP group showed a significantly lower variceal recurrence rate than the E group (22% vs. 67%, p < 0.001). The COX proportional hazard models revealed that grouping was an independent predictor for variceal rebleeding (H = 0.122, 95% CI 0.055-0.270, p < 0.001) and variceal recurrence (hazard ratio, H = 0.160, 95% CI 0.077-0.332, p < 0.001). The peripheral blood cell count, Child-Pugh class/score, albumin concentration, and coagulation function in the EP group improved significantly compared to the values observed in the E group at any time point (p < 0.05).

Conclusions: The EP treatment was more effective in preventing variceal rebleeding and variceal recurrence than the conventional E treatment during the secondary prophylaxis of VH in cirrhosis patients with hypersplenism. Furthermore, the EP treatment could significantly increase the peripheral blood cell count and albumin concentration and also improved the coagulation function and the Child-Pugh score.

Clinical trials registration: Trial registration number ClincialTrials.gov: NCT02778425. The URL of the clinical trial: https://clinicaltrials.gov/.

Keywords: Cyanoacrylate injection; Endoscopic variceal ligation; Gastroesophageal varices; Hypersplenism; Liver cirrhosis; Non-selective beta-blocker; Partial splenic embolization; Portal hypertension; Secondary prophylaxis; Thrombocytopenia.

Fig. 1

Flow diagram of the study. PSE Partial splenic embolization

Fig.2

Kaplan–Meier curves of variceal rebleeding (a), variceal recurrence (b). The log-rank p values of which were 0.000 and 0.008, respectively; PSE partial splenic embolization

Fig.3

Follow-up results of peripheral blood cell count. a White blood cell (WBC), b red blood cell (RBC), c platelet (PLT), d hemoglobin (HGB) in EP group and E group. PSE partial splenic embolization. *There was a significant difference between the EP group and E group (p < 0.05). **It was significantly increased at this point than the pre-EP treatment (p < 0.05). ***It was significantly increased/decreased at this point than the pre-E therapy(p < 0.05)

Fig.4

Follow-up results of albumin concentration. PSE, partial splenic embolization. *There was a significant difference between the EP group and E group(p < 0.05). **It was significantly increased at this point than the pre-EP treatment (p < 0.05). ***It was significantly increased at this point than the pre-E therapy (p < 0.05)

Fig.5

Follow-up results of coagulation function parameters. a prothrombin time (PT), b international normalized ratio (INR), c prothrombin activity (PTA) in EP group and E group. PSE partial splenic embolization. *There was a significant difference between the EP group and E group (p < 0.05). **It was significantly increased/decreased at this point than the pre-EP treatment (p < 0.05)