Inclusion criteria: Patients with liver cirrhosis due to any causes; liver function grade: Child-Pugh A-B; splenomegaly and hypersplenism; severe esophageal varicose confirmed by gastroscopy; and previous histories of recurrent upper gastrointestinal bleeding. All patients signed the informed consent and the study was approved by the Ethics Committee of the hospital. These patients’ data within one month after splenectomy with gastroesophageal devascularization were collected and analyzed.

Exclusion criteria: Patients who did not fulfill the inclusion criteria and could not tolerate surgical treatment were excluded.

After fasting for 8 h, all patients lied in the supine or left lateral position, and color Doppler ultrasound examination was performed with ACUSON Sequola 512 SIEMENS Ultrasound system with a probe frequency of 3.5 MHz. The inner diameter of the portal vein and the splenic vein were measured in the sagittal position. Platelets/international normalized ratio (PLT/INR) was also measured before surgery and at day 1, day 3, day 7 and 1 mo after surgery.

Selective gastroesophageal devascularization was performed using intraoperative free portal venous pressure (FPVP) as guidance[24]. A 20G antithrombotic catheter (BD Insyte, Becton Dickinson Medical Devices Co, Ltd. Suzhou, China) was inserted into the right gastro-omental vein to test FPVP after laparotomy and splenectomy, respectively. Gastroesophageal devascularization was performed according to the FPVP.

The patients were classified into two groups according to whether regular anticoagulantion was administered. In group A, the patients received irregular anticoagulant aspirin or warfarin monotherapy for an undesignated time period without LMWH due to poor blood clotting after surgery and perioperative abdominal or GI bleeding.

Patients in group B had nearly normal blood clotting before and after surgery; 24 h after surgery, they received subcutaneous injection of LMWH routinely, 0.3 mL per 12 h for 5 d and then maintained by oral therapy with warfarin for one month to keep the target prothrombin time/international normalized ratio (PT/INR) at a level between 1.25 and 1.5. If the postoperative platelet level was increased to 100 × 10⁹/L or above, aspirin 100 mg daily was added for one month. If the postoperative platelet level was increased to 300 × 10⁹/L or above, ticlopidine was added with the dose of 0.25 g daily for one month.

Color Doppler ultrasound examination and PLT/INR measurement were repeated. Once PSVT was confirmed after splenectomy with gastroesophageal devascularization, the patients would receive a thrombolytic therapy. Urokinase was administered via the peripheral venous route with a bonus dose of 200 000 units within 30 min, followed by continuous infusion of 20 000-50 000 units/h for 3-5 d via a micro-infusion pump. During the thrombolytic treatment, PT/INR and PLT were measured daily. Following the thrombolytic treatment, the patients were administrated with oral warfarin (2.5 mg, 1-2 times daily) and aspirin (100 mg daily). The drug doses were adjusted according to PT/INR and PLT levels. If repeated color Doppler ultrasound examinations showed complete or partial dissolution of target thrombus, the treatment was considered effective and switched to oral warfarin monotherapy (2.5 mg, 1-2 times daily) for one month. If repeated color Doppler ultrasound examinations showed little change or even enlargement of the target thrombus, the thrombolytic therapy was defined as ineffective. The patients would continue to receive oral warfarin and aspirin and followed up regularly.

SPSS version 11.5 software (SPSS Inc., Chicago, IL, United States) was used for statistical analyses. Continuous data were presented as mean ± SD and compared with two-tailed nonpaired Student’s t test. Categorical data were analyzed with Chi-square or Fisher exact test. Chronologic changes in the laboratory data were analyzed by the analysis of variance for repeated measures. P < 0.05 was considered statistically significant.

Currently, a number of studies have suggested possible mechanisms of PSVT formation following splenectomy with gastroesophageal devascularization, including elevated postoperative platelet count, hemodynamic changes in splenic and portal veins, endothelial damage, spleen size, postoperative release of procoagulant factors, the reduction of anticoagulant factors, and the postoperative use of hemostatic drugs[10,25,26]. PSVT may lead to severe clinical adverse events or poor outcomes[3,4]. Therefore, the importance of PSVT prevention has been gradually recognized and emphasized[19-21,27].

Various prevention protocols have been proposed[25,28,29], but the effectiveness of these protocols varied[4,22,30] because the duration and doses of these drugs were variable. Therefore, there was no generally acceptable PSVT prophylactic regimen for all patients[22,23]. There are many blood coagulation factors involved in the formation of thrombosis[31]. Xa is a major factor in the procedure of thrombosis[32,33]. LMWH can suppress factor Xa by combining with ATIII to depress the activation of thrombin and formation of thrombosis[34]. LMWH is safer than heparin because of its lower molecular weight, weaker inhibition of factor IIa and longer impact on coagulation system[35,36].

The increased count and augmented aggregation competence of PLT after operation were important factors related to PSVT[25]. These factors should be taken into account when selecting anticoagulation drugs. Aspirin has been applied in prevention and treatment of thrombotic diseases with satisfied safety because of its anti-PLT aggregation competence[37,38]. Also, warfarin is an important antagonist of vitamin K (VK) with powerful anticoagulation effects by inhibiting VK-depending coagulation factors such as II, VII, IX, X[39,40]. We selected LMWH, warfarin and aspirin as a regular PSVT prevention regimen, which targets the major factors of mechanisms of PSVT. It was reported that the median interval between splenectomy and PSVT was 1-2 wk after surgery[5,22,23,41].

Therefore, we used LMWH for 2-5 d followed by oral warfarin and asprin after surgery in group B. Oral warfarin or aspirin was merely applied to those postoperative patients who were not suitable for LMWH in early postoperative phase as an irregular PSVT prevention regimen in group A.

In this study, we collected and analyzed retrospectively the data about the thrombosis prevention regimens for cirrhotic patients after splenectomy with gastroesophageal devascularization who were treated at our department over the past six years. Our results showed that the early use of LMWH followed by a long-term maintenance therapy with warfarin and aspirin could reduce the incidence of PSVT in group B (20.94%) without causing major bleeding complications, as compared with group A (41.17%) and other studies (51.5%-55%)[5,6]. We also compared the patients with and without PSVT and found that the PT/INR in patients without PSVT was greater than those with PSVT at day 14 after surgery. Similar results were found in patients with and without PSVT in group B, suggesting that our anticoagulant regimen with the PT/INR target value set at 1.25-1.5 was reasonable and effective.

PSVT in patients with portal hypertension due to cirrhosis following splenectomy with gastroesophageal devascularization can have various clinical manifestations, from no symptoms in mild cases to life-threatening recurrence of upper GI bleeding or small bowel necrosis in serious cases[29]. The difference in clinical consequences stems from the location of PSVT, degree of obstruction, and its impact on portal vein (PV) hemodynamics. In fact, grade IV PSVT (Yerdel’s classification[42]), which involved PV and superior mesenteric vein (SMV), would result in significant reduction in venous blood return to the liver and elevation in PVP, leading to esophageal and gastric variceal rupture and bleeding. SMV thrombosis can impede venous blood return from the bowels and cause small bowel necrosis, which could be vital for patients.

In this study, all 94 PSVT cases were grade I, which led to almost no clinical manifestations, such as variceal bleeding recurrence and bowel necrosis, and were easy to treat. Thrombolytic therapy has been proven to be effective for acute PSVT. Previous studies have recommended early and timely thrombolytic anticoagulant therapy[43,44] and commonly used drugs, including urokinase, recombinant tissue plasminogen activator (rt-PA), LMWH, warfarin and dipyridamole. Intervention can be categorized by the administration route, i.e., systemic, portal system, and intravascular interventional treatment[45,46]. Krauth et al[23] reported that immediate thrombolytic anticoagulant therapy in PSVT patients can achieve a complete dissolution rate of 63.3% and a partial dissolution rate of 13.3%. In our study, thrombolytic therapy via the peripheral venous route was administered in all 94 patients with PSVT, among whom 18 patients showed no sign of thrombus dissolution, but 61 patients (64.89%) achieved complete dissolution and 15 patients (15.95%) had partial dissolution. The overall thrombolytic effectiveness in our study was similar to other studies mentioned above. These phenomena were related to the benefit of regular anticoagulant therapy, which confirmed the clinical value of the anticoagulant regimen.

Our findings suggested that regular anticoagulation and early thrombolytic therapy are safe, effective and rational for PSVT patients who had portal hypertension and underwent splenectomy with gastroesophageal devascularization. At present, splenectomy with gastroesophageal devascularization may only be used as a bridge to liver transplantation in cirrhotic patients, timely prevention and thrombolytic treatment for PSVT offers a significant clinical value in terms of facilitating the portal venous reconstruction of the recipients for liver transplantation[47].

Theoretically, preventive use of anticoagulants against PSVT in cirrhotic portal hypertensive patients early after surgery would counter the dilemma of bleeding. But in fact, previous studies have demonstrated that early anticoagulant treatment in these patients is a safe and effective protocol to prevent PSVT[48,49]. Based on our study, the regular monitoring during anticoagulant treatment is necessary, which can guarantee the safety and maintain the PT/INR level between 1.25 and 1.5. Because most of our subjects had end-stage HBV cirrhosis, our suggestion differs from previous studies on non-HBV cirrhotic patients, mostly with hematologic and metabolic disorders and alcoholic cirrhosis, for which, PT/INR value of 2-3 is recommended[50]. In this study, only three patients presented with mild GI bleeding during the anticoagulant treatment. We immediately terminated it and shifted to symptomatic treatment such as hemostatic agents. Bleeding was successfully controlled, and the patients were discharged uneventfully. Therefore, our anticoagulant therapy has been proven safe.

On the other hand, there is a lack of large-scale controlled and long-term studies on the prevention of PSVT in cirrhosis patients receiving splenectomy with gastroesophageal devascularization for splenomegaly and hypersplenism. Since a randomized prospective controlled study is difficult to perform, we choose to conduct this retrospective controlled study. However, randomized controlled trial is necessary in the future.

Overwhelming postsplenectomy infection (OPSI) syndrome is associated with a high mortality, even it is a rare condition. Major risk factors include the age of the patients with splenectomy, the time after splenectomy, the reason for splenectomy, and the overall immune status of the patients. Splenectomy performed for hematological disorders, including thalassemia, hereditary spherocytosis, autoimmune hemolysis, immune thrombocytopenic purpura, or lymphoma, appears to carry a higher OPSI risk than splenectomy performed as a result of other diseases. Treatment of OPSI is generally aggressive due to the serious nature of the condition and associated mortality. The major preventive strategy is the vaccination using the 23-valent pneumococcal polysaccharide vaccine, a 7-valent proteinconjugated pneumococcal vaccine, the hemophilus influenzae type B vaccine, and the meningococcal vaccine[51]. Fortunately, there was no OPSI occurrence in our study. The reasons may be that all the patients are adults and no splenectomy was performed for hematological disorders, the antibiotic agents were administered for 3-5 d after surgery, and the follow-up period was too short for OPSI.

In summary, the early and regular initiation of anticoagulant treatment using LMWH followed by warfarin and aspirin has been proven safe and effective in early prevention of PSVT in patients with cirrhotic portal hypertension, undergoing splenectomy with gastroesophageal devascularization. The treatment can reduce the incidence of PSVT in the early stage after splenectomy with gastroesophageal devascularization, early and timely thrombolytic therapy is imperative and feasible for the prevention and treatment of PSVT. The protocol presented in our study may benefit the patients not only for approximate and long-term clinical outcome, but also for potential liver transplant candidates in the future. But a better designed randomized prospective study with a longer follow-up period is still needed to clarify our conclusions.

Portal or splenic vein thrombosis (PSVT) is a common and potentially life-threatening complication of splenectomy with gastroesophageal devascularization for portal hypertension due to cirrhosis, which may lead to further deterioration of liver function and the recurrence of upper gastrointestinal (GI) bleeding, and significantly affects patient’s life expectancy. In some severe cases, PSVT may be contraindicated for liver transplantation. It is therefore very important to prevent the occurrence of PSVT after splenectomy with gastroesophageal devascularization to achieve better long-term outcomes and following possible liver transplantation if required.

Prophylactic anticoagulation therapy using a combination protocol of low molecular weight heparin (LMWH), vitamin K antagonists, such as warfarin and aspirin, is a prime method to prevent PSVT. But the duration and doses of these drugs were variable according to the literature, there was no generally acceptable PSVT prophylactic regimen for all patients after splenectomy with gastroesophageal devascularization. So, a safe and effective prophylactic anticoagulation protocol is needed.

In order to reduce the incidence of PSVT after splenectomy with gastroesophageal devascularization, a regular prophylactic anticoagulation protocol was established with a confirmed monitoring index. This study used the combined and sequential application of LMWH, warfarin, aspirin and ticlopidine according to the regular coagulating function test and color Doppler flow imaging. The incidence and severity of PSVT caused by anticoagulation therapy were reduced, without causing major bleeding complications.

The study results suggest that prophylactic anticoagulation therapy using LMWH, warfarin and aspirin regularly is a safer and more effective method for PSVT prevention.

PSVT is a sort of clinical disease caused by thrombus development in the portal vein system. The major causes of PSVT included the reduced blood flow, the increased platelet count, the injured vessel endothelium and enhanced coagulation function. PSVT is a common complication of abdominal surgery, especially after splenectomy with gastroesophageal devascularization with a high incidence and morbidity.

This is a good descriptive study in which authors analyze the preventive effect of prophylactic anticoagulation therapy using low molecular weight heparin, warfarin and aspirin for PSVT. The results are interesting and imply that a regular prophylactic anticoagulation protocol is a safer and more effective method that could be used in preventing PSVT after splenectomy with gastroesophageal devascularization.