Rex shunt is an optimal surgery for the treatment of extra-hepatic portal venous obstruction (EHPVO) in children. Anticoagulant therapy has been used to keep the patency of the bypass vein in the Rex shunt. This study was to investigate the effectiveness of anticoagulant therapy using heparin combined with Plavix in improving the prognosis and shunt patency of Rex shunt. From January 2010 to September 2019, 51 children with EHPVO underwent a portal cavernoma- Rex shunt. Based on whether using the anticoagulant therapy after the Rex shunt, all patients were divided into two groups: the anticoagulant group and the non-anticoagulant group. The diameter and flow velocity of the bypass vein were measured by the post-operative ultrasound, which was used to calculate the flow volume of the bypass vein (FV) and standard portal venous flow (SPVF). The bypass venous flow index (BVFI) was used to evaluate the ability of portal blood into the liver through the bypass vein after the Rex shunt, which was a ratio of FV to SPVF. The incidence of post-operative re-bleeding, the postoperative patency rate of the bypass vein, the remission rate of postoperative hypersplenism, the remission rate of postoperative esophagogastric varices and the BVFI were compared between the two groups. Of the 51 patients, 12 patients in the anticoagulant group were treated with heparin combined with Plavix after Rex shunt; 39 patients in the non-anticoagulant group were not treated with any anticoagulant therapy. 8 of 51 patients suffered from postoperative re-bleeding, of whom 6 patients with thrombosis of the bypass vein and 2 patients with anastomotic stenosis of the bypass vein. All 8 patients with re-bleeding belonged to the non-anticoagulant group. The remission rate of hypersplenism was no significant difference between the two groups after surgery (91% vs. 58%, P = 0.100). However, 3 patients without hypersplenism before surgery suffered from hypersplenism after surgery, who belonged to the non-anticoagulant group. There was no significant difference in the remission rate of esophagogastric varices (33% vs. 46%, P = 1.000). The BVFI of the anticoagulant group was significantly higher than that of the non-anticoagulant group (5.71 ± 5.89 vs. 1.1 ± 1.52, P = 0.003). Anticoagulant therapy using heparin combined with Plavix plays an important role in maintaining the patency of the bypass vein, which improved the portal blood flow into the liver through the bypass vein after the Rex shunt.