Keywords
portal vein thrombosis - splenic vein thrombosis - mesenteric vein thrombosis - Budd-Chiarisyndrome - anticoagulant therapy
Introduction
The splanchnic venous system drains the gastrointestinal tract from the lower esophagus to the upper two-thirds of the rectum. Splanchnic vein thrombosis (SVT) is a rare type of venous thromboembolism (VTE) occurring within this system ([Fig. 1]). It includes portal vein thrombosis (PVT), splenic vein thrombosis (SpVT), and mesenteric vein thrombosis (MVT), as well as Budd–Chiari syndrome (BCS). The term BCS refers to a hepatic venous outflow obstruction anywhere between the hepatic venules and the insertion of the inferior vena cava into the right atrium.[1]
[2]
Fig. 1 Possible sites of splanchnic vein thrombosis.
SVT commonly occurs in the context of liver disease, such as cirrhosis, and malignancies.[3] Other causes include intra-abdominal surgery or infections, pancreatitis, inflammatory bowel diseases, thrombophilia, and myeloproliferative neoplasms (MPNs).[3]
[4]
Anticoagulation is the cornerstone of SVT management, intended to prevent thrombus extension and recurrent thrombosis, while carefully balancing the risk of hemorrhage.[5]
[6]
[7]
This narrative review aims to summarize the latest evidence on SVT, with a particular focus on recent updates regarding the anticoagulant treatment.
Epidemiology
SVT is an unusual manifestation of VTE, as compared to the commoner usual site VTE (e.g., pulmonary embolism [PE] and lower extremity deep vein thrombosis [DVT]).[8]
[9] The incidence of SVT is at least 25 times lower than usual site VTE, with considerable variation reported across various studies.[10] SVT is a heterogeneous disorder with differences based on patient populations and sites of thrombosis. While SVT is relatively common among patients with liver cirrhosis and malignant tumours,[11] other types of SVT (primitive PVT, isolated SpVT, isolated MVT, and BCS) are cataloged in Orphanet, the European portal for rare diseases that affect less than 5 per 10,000 individuals.
PVT is the most common presentation, while BCS is the rarest, accounting for 77 to 92% and 2 to 11% of SVT cases, respectively.[1]
[12]
[13]
[14] However, in patients with MPNs, BCS constitutes approximately 17% of all SVT cases.[15] Of note, multisegmental SVT is found in around 40% of SVT cases.[1]
[16]
The incidence of PVT was 1.7 to 3.8 cases per 100,000 inhabitants per year in a study evaluating hospital discharge diagnosis in Northwestern Italy, while a higher incidence rate of 2.8 to 5.9 cases per 10,000 patients was reported in a study evaluating abdominal imaging reports in a large Canadian hospital.[12]
[17] The annual incidence of BCS is around 1 to 2 cases per million inhabitants.[17]
[18]
The prevalence of PVT was 3.7 cases per 100,000 inhabitants in a study evaluating Swedish inpatient and outpatient registers.[19] Higher prevalence rates were reported in certain categories of patients, for instance, 14% in patients with liver cirrhosis and 10% in patients with gastrointestinal malignancies.[20]
[21] The pooled prevalence of BCS in a systematic review and meta-analysis was 11 cases per million inhabitants.[18]
Sex and age distribution in SVT vary with the site of thrombosis. PVT occurs more commonly in males (62–66% of cases), with the mean age at presentation of approximately 54 to 61 years.[16]
[17] BCS occurs more commonly in females (54–69% of cases), with the mean age at presentation of approximately 40 to 50 years.[16]
[17]
[22]
There are no precise estimates on the prevalence of MVT in the general population. A Swedish study reported that the incidence of MVT rose from 2.0 per 100,000 person-years in the period from 1970 to 1982 to 2.7 per 100,000 person-years in the period from 2000 to 2006. Furthermore, the highest incidence rate of 11.3 per 100 person-years was seen in patients among septuagenarians.[23]
Risk Factors
SVTs can be classified into provoked or unprovoked, based on the presence or absence of known risk factors ([Fig. 2]).[24]
[25]
[26]
[27]
[28] This classification follows the categorization of usual site VTE and it is used to direct the anticoagulant treatment in many guidelines.[10]
[29] Unprovoked (i.e., idiopathic) cases account for approximately 15 to 31.7% of SVT cases.[1]
[16]
[28] Provoked SVTs can be due to local or systemic risk factors.
Fig. 2 Risk factors for splanchnic vein thrombosis.
Among local risk factors, liver cirrhosis and solid tumors together account for approximately 50% of SVT cases. Liver cirrhosis is found in approximately 24 to 27% of SVT cases and, in turn, approximately 17 to 26% of cirrhotic patients develop SVT.[11] Cirrhosis can induce SVT through several mechanisms, including reduced portal vein blood flow and endothelial dysfunction (due to portal hypertension), and a local/systemic prothrombotic state.[30]
Solid cancer is found in approximately 22 to 27% of SVT cases. Cancer-associated SVT can be induced by tumor compression or invasion of the splanchnic veins, the tumor-related hypercoagulable state, and cancer-management-related factors (e.g., surgery and chemotherapy).[31]
[32]
[33] Furthermore, studies have suggested that SVT can be a marker for occult cancer. Overall, 8% of SVT patients will receive a diagnosis of cancer within 3 months following presentation, in particular, liver cancer, pancreatic cancer, and MPNs.[32]
[34]
[35]
Less common local risk factors for SVT include recent abdominal surgery (e.g., splenectomy, liver transplantation) and abdominal diseases (e.g., inflammatory bowel diseases, pancreatitis, cholangitis, appendicitis, diverticulitis). Pylephlebitis, a rare form of septic PVT, has been described in association with abdominal infections.[36] SpVT is typically a complication of pancreatic diseases, due to the anatomical proximity of the splenic vein to the pancreas. Approximately 12% of patients with acute pancreatitis develop SVT, and the splenic vein is involved in more than 90% of the cases (isolated SpVT in a third of cases, or part of multivessel involvement in the remaining two-thirds).[37]
Philadelphia-negative MPNs, which include polycythemia vera, essential thrombocythemia, and primary myelofibrosis, are the predominant systemic risk factors for SVT. These conditions are identified in 30 to 50% of BCS patients and in 15 to 30% of those with nonmalignant, noncirrhotic PVT.[38]
[39] The JAK2V617F mutation, which is a marker of MPN, is detected in 76 to 93% of SVT patients with overt MPN.[39]
[40] Thus, JAK2 screening should be performed in SVT patients without liver cirrhosis or malignancy, to identify the possible presence of an MPN.[40] Other driver mutations causing MPNs, such as calreticulin (CALR) and thrombopoietin receptor (MPL) mutations, are rarely found in SVT patients (∼1%). Next-generation sequencing has been utilized in more recent studies to detect nondriver mutations[41]; however, the association of similar mutations with thrombotic risk is still unclear.[42]
[43]
Another rare hematological disorder associated with SVT is paroxysmal nocturnal hemoglobinuria (PNH). Although rarely found in SVT patients (∼1%),[44] PNH screening should be performed if there is evidence of hemolysis or cytopenia.[45]
An association has been described between the presence of thrombophilic abnormalities, both inherited (factor V Leiden mutation; prothrombin G20210A mutation; antithrombin, protein C, or protein S deficiencies) and acquired (antiphospholipid antibodies/lupus anticoagulant), and the development of SVT. However, deficiencies of natural anticoagulants are common in patients with liver diseases, and this might be the effect rather than the cause. Furthermore, the utility of thrombophilia testing in patients with SVT has been questioned by a recent study showing that, with the exception of the JAK2 mutation, other results rarely changed the management of these patients.[46]
Recent guidelines[45]
[47] state that thrombophilia testing should be performed only in situations where it would change the management, that is, the duration or the choice of the anticoagulant drug. The 2023 American Society of Hematology (ASH) guidelines suggest thrombophilia testing in patients with SVT not associated with liver cirrhosis in those settings where anticoagulation would be otherwise discontinued.[47] The 2022 British Society for Haematology (BSH) guidelines recommend testing for antiphospholipid antibodies in SVT patients without clear provoking factors.[45]
Other systemic risk factors for SVT include estrogen-containing oral contraceptives, postmenopausal hormone therapy, pregnancy and puerperium, and certain autoimmune diseases. More recently, several cases of SVT have been described in the 2 weeks following a COVID-19 diagnosis, while rare cases of SVT have been reported after COVID-19 vaccination, as part of vaccine-induced immune thrombotic thrombocytopenia.[48]
[49]
Clinical Manifestations
SVT has nonspecific clinical features. Irrespective of the site of thrombosis, the most common signs and symptoms are abdominal pain, gastrointestinal bleeding, and abdominal distension secondary to ascites. Less common manifestations include fever, nausea, vomiting, diarrhea/constipation, and anorexia.[50] More specific clinical scenarios are found in certain locations (see below).
A significant proportion of patients, however, are asymptomatic and diagnosed incidentally with SVT when undergoing routine imaging for other conditions, most commonly malignancies and cirrhosis.[3]
[50]
[51] Incidentally detected SVTs have been found in 1.7% of abdominal CT scans and, in different cohort studies, they constitute 30 to 59% of all SVTs.[52]
[53]
[54]
Portal Vein Thrombosis
PVT is divided into acute and chronic forms. Acute cases are characterized by sudden onset of abdominal pain without any signs of portal hypertension or portosystemic collateral circulation, while chronic cases are characterized by the presence of portal hypertension (e.g., ascites, gastroesophageal varices, splenomegaly and hypersplenism, thrombocytopenia, hepatic encephalopathy, and portal cholangiopathy) and/or portal cavernoma.[55] Some authors suggest considering the duration of symptoms, with acute PVT symptoms lasting less than 60 days prior to diagnosis and chronic PVT symptoms lasting greater than 60 days, but there is no consensus on this time frame.[56]
[57]
[58]
Most of the cases of cirrhosis-associated PVT are incidental findings. However, the development of PVT in cirrhotic patients can result in a sudden clinical deterioration (e.g., occurrence of diuretic-resistant ascites or bacterial peritonitis) and increases the risk of portal hypertensive bleeding. Defining acute or chronic PVT can be challenging in patients with liver cirrhosis, due to the presence of preexisting portal hypertension and collateral vessels resulting from the chronic liver disease itself.[59]
Mesenteric Veins Thrombosis
MVT is divided into acute, subacute, and chronic forms. Acute cases are characterized by severe, progressive, colicky, mid-abdominal pain. However, the onset of pain is more insidious in patients with acute MVT, compared to acute mesenteric arterial thromboembolism.[60] Associated signs and symptoms include nausea, vomiting, diarrhea, and lower gastrointestinal bleeding. In mesenteric ischemia, the physical findings are initially normal. A third of patients will progress to bowel infarction, which is associated with peritoneal signs (such as guarding and rebound tenderness).[16]
[23]
Patients with subacute thrombosis present with abdominal pain lasting for days/weeks, but without signs of bowel infarction or portal hypertension.[60] Chronic cases are usually characterized by the lack of acute abdominal pain. Abdominal imaging shows the presence of mesenteric thrombosis, together with extensive venous collateral circulation. Extension of the thrombosis into the portal vein is seen in 65 to 85% of cases, and these patients can present with complications of portal hypertension.[23]
Splenic Vein Thrombosis
Isolated SpVT is rare, being more commonly associated with concomitant PVT or MVT. A typical triad has been described for SpVT, which includes splenomegaly, normal liver enzymes, and upper gastrointestinal bleeding. In fact, the presence of splenomegaly and/or isolated gastric varices in a patient without liver cirrhosis or hematological diseases should raise suspicion of an SpVT. The wider availability and improved accuracy of cross-sectional diagnostic imaging, such as computed tomography (CT) and magnetic resonance (MR) scans, has led to an increase in the detection of asymptomatic SpVT, especially when performed for the follow-up of pancreatic diseases.[61]
Budd–Chiari Syndrome
BCS is classified into fulminant, acute, subacute, and chronic forms. Fulminant cases are rare and present with extensive hepatocellular necrosis and manifestations of acute liver failure, such as elevated liver enzymes, coagulopathy, hyperbilirubinemia, and hepatic encephalopathy.[62]
A typical, although uncommon, triad of hepatomegaly, abdominal pain, and ascites has been described.[63] Another specific sign for BCS is the presence of large subcutaneous cavocaval collateral veins on the trunk, suggesting obstruction of the inferior vena cava.[63]
In acute cases, symptoms usually develop within a month, while in subacute cases, there is a more insidious onset. Chronic cases are more likely to present with portal hypertension-related manifestations, such as gastroesophageal varices, splenomegaly, and ascites. However, if extensive intrahepatic collaterals have developed, they can preserve the hepatic venous outflow and patients can be completely asymptomatic.[64]
Diagnosis
D-dimer, a biomarker of fibrinolysis commonly used as a rule-out test for usual site VTE, has very limited utility in the diagnosis of SVT. In fact, several conditions, which are also risk factors for SVT, can raise the D-dimer level. These include liver cirrhosis, malignancies, recent surgery, pancreatitis, and others.[65]
[66]
[67] A recent systematic review and meta-analysis highlighted that, in patients at high risk of SVT (such as those with liver cirrhosis, hepatocellular carcinoma, or in the postoperative period), D-dimer has 96% sensitivity, but only 25% specificity for SVT diagnosis.[68]
Thus, the diagnosis of SVT is generally achieved through abdominal imaging. Color Doppler ultrasound (CDUS) is the first-line imaging for the diagnosis of PVT and BCS. However, the mesenteric and splenic veins are difficult to visualize on CDUS. The advantages of CDUS include the possibility to assess the dynamic blood flow, low cost, wide availability, and lack of ionizing radiation or contrast agents, while the disadvantages include the fact that it is an operator-dependent imaging and that patient's body habitus and presence of meteorism may impair the visualization of the abdominal veins.[69] Contrast-enhanced ultrasound (CEUS) can be used to differentiate between PVT and tumor thrombosis (also known as tumor in vein), with reported 94% sensitivity and 99% specificity in patients with hepatocellular carcinoma.[70]
Cross-sectional imaging techniques, such as CT and MR angiography, are the main diagnostic modalities for SVT, since they can visualize the whole splanchnic venous system. They can also visualize the surrounding structures to identify possible complications, such as bowel ischemia, or provoking condition, such as abdominal malignancies.[71] They are often performed after CDUS to evaluate thrombus extension within the mesenteric and splenic veins, which can be underestimated by CDUS. CT utilizes ionizing radiation and an iodinated contrast agent, which can potentially be nephrotoxic in patients with preexisting chronic kidney disease (contrast-induced nephropathy). MR is less widely available and has a longer acquisition time and lower spatial resolution than CT scan.[71] It does not utilize ionizing radiation and the paramagnetic contrast agent has a better safety profile. However, rare cases of nephrogenic systemic fibrosis have been reported in patients with chronic kidney disease.[50] Recent data suggest that MR non-contrast thrombus imaging (MR-NCTI), a non–contrast-enhanced modality also known as MR direct thrombus imaging, might help differentiate an acute PVT from a chronic PVT.[72] MR-NCTI detects fresh thrombi as high signal intensity due to methemoglobin formation, while the signal intensity is reduced in chronic thrombi due to the lower content of methemoglobin.
Angiography was the gold standard for SVT diagnosis, but it has been replaced by less invasive modalities (such as CDUS, CT, and MR). However, it is still used as a comparison in studies evaluating the diagnostic accuracy of the other imaging tests. For instance, a recent study compared the findings obtained by CDUS and CT versus direct transjugular splenoportography.[73] Overall, 76.6% of CDUS results and 78.4% of CT results were consistent with portography. The remaining 23.4% and 21.6%, respectively, of cases in disagreement were only due to differences in the grading of the thrombosis (partial vs. complete) since the presence of thrombosis was detected by all imaging modalities. The agreement between CDUS and CT in the grading of the thrombosis was 74.4%, suggesting that similar imaging modalities should be used to evaluate thrombosis extension at diagnosis and degree of recanalization during follow-up.[73]
Management: General Principles of Anticoagulant Treatment
Management: General Principles of Anticoagulant Treatment
The aim of the anticoagulant treatment in SVT management is threefold: to prevent thrombus progression (thus reducing the risk of bowel ischemia), to promote vessel recanalization (thus reducing the risk of portal hypertension-related complications), and to prevent VTE recurrence. A recent meta-analysis reported that anticoagulation in SVT patients, compared to no treatment, was associated with higher rates of vessel recanalization (risk ratio [RR]: 2.39; 95% confidence interval [CI]: 1.66–3.44), and lower rates of thrombus progression (RR: 0.24; 95% CI: 0.13–0.42), without increasing the risk of major bleeding events (RR: 0.73; 95% CI: 0.58–0.92).[6] Due to the high risk of major and fatal bleeding, thrombolysis should be limited to selected patients with particularly severe symptoms (e.g., extensive SVT associated with intestinal ischemia) or when medical therapy has failed (e.g., deterioration of clinical conditions despite adequate anticoagulation).[10]
Timing of Anticoagulation
Studies have shown that early anticoagulant treatment is associated with higher recanalization rates, which in turn result in a lower risk of portal hypertension and related complications.[74]
[75]
[76] For instance, Delgado et al reported that partial or complete recanalization was obtained in 71.4% of patients who started anticoagulation within 2 weeks from SVT diagnosis versus 40.0% of those who started after 2 weeks (p = 0.044).[74]
The 2020 Guidance of the International Society on Thrombosis and Haemostasis (ISTH) recommends early anticoagulant treatment for SVT patients without active bleeding or other contraindications, and early esophagogastroduodenoscopy to screen for gastroesophageal varices.[10] The presence of varices does not constitute a contraindication to anticoagulant treatment, but requires appropriate medical prophylaxis. The Baveno VII consensus recommends nonselective beta blockers (in particular carvedilol) for primary prevention of variceal bleeding, while endoscopic variceal band ligation (EVL) is recommended in patients with high-risk varices and contraindications or intolerance to beta blockers.[77] The 2020 Practice Guidance of the American Association for the Study of Liver Diseases (AASLD) also recommends early anticoagulation and suggests that it can be started before complete endoscopic variceal eradication.[78] In fact, studies have shown that anticoagulation does not increase the risk of post-EVL bleeding. For instance, Guillaume et al reported that the rates of post-EVL upper gastrointestinal bleeding were similar between SVT patients receiving anticoagulation (5 of 30 [16.7%] patients with bleeding events corresponding to 9 episodes out of 121 EVL sessions [7.4%]) and not anticoagulated SVT patients (4 of 13 [30.8%] patients with bleeding events corresponding to 6 episodes out of 130 EVL sessions [4.6%]).[79]
Choice of Anticoagulant Drugs
Possible anticoagulant treatment options for SVT patients include unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), vitamin K antagonists (VKAs), and direct oral anticoagulants (DOACs).
UFH is rarely used, due to the very short half-life, the need for intravenous infusion with monitoring of the activated partial thromboplastin time (aPTT target range 1.5–2.5), and high risk of heparin-induced thrombocytopenia.[80] In addition, in cirrhotic patients, it is difficult to find a reliable laboratory assay for monitoring, since the baseline aPTT is frequently prolonged and the anti-Xa assay can also be influenced by the typical reduction of antithrombin levels. However, UFH can be considered in patients with severe renal insufficiency (creatinine clearance < 30 mL/min) or in candidates for invasive procedures.[10]
[77]
LMWH has a short half-life, a subcutaneous administration, and requires no monitoring; however, due to the predominant renal clearance, its use is limited in patients with renal insufficiency. LMWH can be used for the initial phase of treatment of SVT, until clinical stability is reached and patients are transitioned to oral anticoagulants. However, in certain situations (e.g., active cancer, liver disease, or thrombocytopenia), LMWH can also be used for long-term treatment, due to the possibility of reducing to intermediate or prophylactic doses.[29]
VKAs (such as warfarin) are given orally and have several food–drug interactions and high inter- and intraindividual variability, thus requiring frequent monitoring of the international normalized ratio (INR target range 2.0–3.0). They can be used for the long-term treatment of SVT; however, monitoring might be difficult in certain categories of patients, such as cirrhotic patients with baseline prolongation of the INR.[81]
More recently, in line with the progress in the treatment of usual site VTE, there has been a shift toward DOACs in SVT management.[82] They can be given orally at standard doses and do not need monitoring. Apixaban and rivaroxaban can be started immediately after VTE diagnosis, using higher doses for the initial weeks (single-drug approach), while dabigatran and edoxaban can be considered after few days of parenteral heparins (dual–drug approach). However, lacking specific randomized controlled trials in SVT, DOACs are still off-label for this indication.
Several recent studies evaluated DOACs in SVT. Naymagon et al performed a single-center, retrospective study including 330 patients with acute noncirrhotic PVT.[83] The primary outcome of complete resolution of the thrombosis on imaging was highest in the DOAC group (66%), followed by LMWH (57%), VKA (31%), and no anticoagulant treatment (14%). When considering VKA as reference, DOACs and LMWH were associated with higher likelihood of complete resolution (DOACs: hazard ratio [HR], 2.91; 95% CI, 1.87–4.52; LMWH: HR, 2.23; 95% CI, 1.43–3.70). DOACs also resulted in fewer major bleeds compared to warfarin (HR: 0.20; 95% CI: 0.05–0.86).[83] The efficacy of DOACs on the primary outcome was confirmed in the subgroups of patients with PVT secondary to abdominal surgery (n = 107, of whom 35 were treated with DOACs) and PVT secondary to inflammatory bowel disease (n = 63, of whom 23 were treated with DOACs).[84]
[85]
The RIVA-SVT100 study was an interventional, single-arm, multicenter clinical trial that enrolled 100 patients with a first episode of noncirrhotic symptomatic SVT.[86] Patients were treated with rivaroxaban (15 mg twice daily for 3 weeks, then 20 mg once daily up to 3 months) and followed for 6 months. Among the exclusion criteria, there were patients with BCS or liver cirrhosis and patients anticoagulated with therapeutic LMWH for over 7 days prior to enrolment or with ongoing VKA treatment. At the 3-month follow-up, there was SVT recurrence in two patients (2.1%; 95% CI: 0.6–7.2%). There were two major bleeds (2.1%; 95% CI: 0.6–7.2%) and four clinically relevant nonmajor bleeds in three patients (3.1%; 95% CI: 1.1–8.7%). At follow-up imaging, 33 (36.3%) patients had partial recanalization (95% CI: 27.1–46.5%) and 43 (47.3%) had complete resolution (95% CI: 37.3–57.4%).[86]
A recently published meta-analysis included eight studies that compared DOACs with other anticoagulant treatments or no anticoagulation.[82] In noncirrhotic patients (n = 499), DOACs given for a median duration of 4.5 months were more effective than VKAs in achieving complete recanalization (odds ratio [OR]: 4.33; 95% CI: 2.40–7.83). Major bleeding was rarer when using DOACs compared to observation (OR: 0.09; 95% CI: 0.03–0.29), LMWH (OR: 0.13; 95% CI: 0.03–0.62), and VKAs (OR: 0.12; 95% CI: 0.02–0.69).[82]
The 2020 ISTH guidance suggests a full therapeutic dose of DOACs as first-line treatment for nonmalignant noncirrhotic SVT.[10] Due to the risk of accumulation, DOACs are contraindicated in cirrhotic patients with severe liver impairment or Child–Pugh C cirrhosis (for rivaroxaban also Child–Pugh class B), and the Baveno VII consensus suggests caution in patients with impaired liver function or Child–Pugh B cirrhosis.[77] DOACs are also contraindicated in patients with severe renal insufficiency, defined as creatinine clearance less than 15 mL/min (for dabigatran also creatinine clearance <30 mL/min), and in patients with triple-positive antiphospholipid syndrome.[87]
Anticoagulant Treatment Duration
The suggested duration of anticoagulant treatment in patients with SVT follows the general recommendations for the treatment of usual site VTE.[29]
[88] In fact, studies have shown that SVT patients with persistent risk factors or unprovoked SVT have a higher risk of recurrent thrombosis, compared to SVT patients with transient risk factors.[1]
[89] Thus, international guidelines recommend an anticoagulant treatment duration of at least 3 to 6 months for all SVT patients.[10]
[77] This can be sufficient for patients with transient risk factors, while an indefinite duration is recommended for patients with unprovoked SVT, permanent risk factors (e.g., liver cirrhosis, active cancer), recurrent VTE, or BCS. In such cases, regular reassessment of thrombotic and bleeding risk is imperative. Other elements that might influence anticoagulant treatment duration include extension of the thrombosis into the mesenteric vein with bowel ischemia, and the degree of vessel recanalization at follow-up imaging.[77]
[90]
Reduced-dose DOACs (i.e., rivaroxaban 10 mg once daily, apixaban 2.5 mg twice daily) are suggested for the extended treatment of usual site VTE to reduce the risk of bleeding.[88] Despite the lack of specific studies in SVT, reduced doses of DOACs can be considered for extended treatment of SVT.[10]
Anticoagulant Treatment in Special Categories of Patients
Anticoagulant Treatment in Special Categories of Patients
Liver Cirrhosis
Compared to other etiologies, cirrhotic patients have the highest rates of recurrent thrombosis and major bleeding events, yet are also less likely to be anticoagulated (only 41–45% of cirrhotic patients with SVT).[1]
[91] The meta-analysis by Valeriani et al showed that anticoagulation in cirrhotic patients with SVT increases the rates of recanalization (RR: 3.19; 95% CI: 1.42–7.17), while reducing progression of the thrombus (RR: 0.28; 95% CI: 0.15–0.52), major bleeds (RR: 0.52; 95% CI: 0.28–0.97), and mortality (RR: 0.42; 95% CI: 0.24–0.73), compared to no anticoagulation.[92] In one study, anticoagulation expressed as time on treatment was the only factor associated with both reduced SVT recurrence (HR: 0.85; 95% CI: 0.76–0.96) and reduced major hemorrhage (HR: 0.83; 95% CI: 0.69–0.99) at multivariable analyses.[93] The recent IMPORTAL meta-analysis confirmed that the benefit of anticoagulation on overall mortality is independent of PVT severity and degree of recanalization.[91]
The 2020 ISTH guidance recommends anticoagulation in cirrhotic patients with acute SVT, if there is no active bleeding or other contraindications.[10] Other guidelines also consider the degree of occlusion. For instance, the Baveno VII consensus recommends anticoagulation when thrombosis of the portal vein trunk is completely or partially occlusive (>50% of the lumen) and states that anticoagulation should be considered in a minimally occlusive isolated thrombus of the portal vein trunk (<50% of the lumen) if there is progression on serial imaging within a short time frame (1–3 months). Cirrhotic patients with SVT who are candidates for liver transplant should be treated, in order to improve the transplant outcomes.[77]
Parenteral anticoagulation is the usual initial treatment for cirrhotic patients. A randomized controlled trial showed that enoxaparin 1 mg/kg twice daily is associated with less nonvariceal bleeding than 1.5 mg/kg once daily (6.4 vs. 23.5%, p < 0.05).[94] In a retrospective study, dalteparin 200 U/kg once daily for 4 weeks, followed by 150 U/kg once daily was associated with higher recanalization rates than a fixed dose of 150 U/kg once daily (OR: 2.6; 95% CI: 1.06–6.38).[95] Fondaparinux is a possible alternative, after a retrospective study by Senzolo et al concluded that it was associated with higher rates of complete recanalization (HR: 2.38; 95% CI: 1.39–4.09) compared to LMWH. However, there was a nonsignificant trend toward higher bleeding rates in patients treated with fondaparinux (27 vs. 13%, p = 0.06).[96]
Naymagon et al studied 214 cirrhotic subjects (Child–Pugh A to C) with PVT, of whom 86 (39%) were anticoagulated for a median duration of 18.8 months: 42 received LMWH, 26 VKA, and 18 DOACs. Although the number of patients was low, safety and efficacy outcomes occurred at similar rates in the three groups.[97] More recently, dabigatran and rivaroxaban showed comparable outcomes in terms of PVT recanalization, major bleeding, mortality, and improvement of the Child–Pugh score after anticoagulation.[98] The meta-analysis by Koh et al suggested that DOACs may be associated with better recanalization rates compared to VKAs, without increasing the risk of major bleeding.[99]
DOACs were initially absent from many guidelines on cirrhotic SVT.[100] However, since safety and efficacy data emerged, guidelines have started including them. For instance, the Baveno VII guidelines recommend initial anticoagulation with LMWH, and consider VKAs or DOACs for subsequent maintenance.[77] For cirrhotic patients with SVT, the 2020 ISTH guidance suggests starting with therapeutic doses LMWH, then switching to oral anticoagulants (VKAs or DOACs) if not contraindicated by the severity of liver impairment (see above).[10]
Other therapeutic options for patients with liver cirrhosis include transjugular intrahepatic portosystemic shunt (TIPS), which is recommended by the Baveno VII consensus if there is no recanalization after anticoagulant treatment, especially for liver transplant candidates.[77]
Solid Malignancies
Patients with solid cancer have the highest mortality rates, compared to other SVT categories.[1] In the study by Valeriani et al, only 91 of 132 patients (68.9%) with cancer-associated SVT received anticoagulation.[101] However, the rates of clinical outcomes during the 12-month follow-up were similar between patients with cancer-associated SVT and DVT or PE (recurrent thrombosis: 4.7 vs. 5.5%; major bleeding: 2.3 vs. 4.7%, respectively).
The recent retrospective study by García-Villa et al[102] enrolled 203 patients with solid cancer-associated SVT, of whom 95 (46.8%) received anticoagulant treatment.[102] Of note, 8.9% of patients were already anticoagulated when SVT was diagnosed. In the first 30 days of follow-up after SVT diagnosis, major bleeding occurred in 1.0% and recurrent VTE in 1.5%. There was no difference between symptomatic and asymptomatic patients (major bleeding: 1.3 vs. 0.5%; recurrent VTE: 1.3 vs. 1.6%, respectively).
For patients with cancer-associated SVT, the 2020 ISTH guidance recommends LMWH or DOACs.[10] LMWH is suggested in cases of luminal gastrointestinal cancer, other gastrointestinal mucosal diseases, and genitourinary cancer at high bleeding risk, and in cases of concomitant chemotherapy interfering with DOACs. The recent Clinical Practice Guidelines of the European Society for Medical Oncology (ESMO) highlighted that there is limited experience with DOACs in patients with cancer-associated unusual site VTE, including SVT.[103]
Approximately a third of patients with hepatocellular carcinoma have tumor extension within the portal vein (tumor thrombus).[104] Tumor thrombus should be differentiated from bland thrombus since the therapeutic approach will be different (antineoplastic treatments in the former vs. anticoagulation in the latter).
Myeloproliferative Neoplasms
Alvarez-Larrán et al reported that MPN patients with SVT have a 3.4-fold higher risk of recurrent VTE, a 3.2-fold higher risk of major bleeding, and a 2.5-fold higher risk of mortality compared to MPN patients without SVT, after age and sex adjustment.[105]
MPN is a persistent risk factor and hence these patients require long-term anticoagulant treatment. Studies have shown that VKAs halved the risk of recurrent VTE.[39]
[40] However, De Stefano et al demonstrated that despite anticoagulation, the risk of recurrent VTE remained significantly high (3.7 events per 100 patient-years).[15] It is debatable whether a combination of antiplatelets and anticoagulants should be prescribed for MPN-associated SVT, because it might increase the already relevant risk of gastrointestinal bleeding.[41] In the study by Sant'Antonio et al, approximately 14% of MPN patients with SVT received antiplatelet treatment, either alone or in association with anticoagulants. However, at multivariable analysis, major bleeding was mainly related to the presence of gastroesophageal varices and not to the type of treatment.[39]
Pancreatitis
Pancreatitis is a transient, yet important cause of SVT. Management is complicated, and there is paucity of up-to-date, focused, officially endorsed guidelines; hence, it has recently garnered much research attention.[37]
There are two recent systemic review and meta-analysis on the subject. Anis et al found that recanalization was more likely in anticoagulated patients (OR: 0.51; 95% CI: 0.31–0.83). However, there was no difference in mortality (OR: 2.37; 95% CI: 0.86–6.52).[106] A year later, Yin et al concurred with these results by concluding that there is no correlation between anticoagulation and mortality (RR: 1.42; 95% CI: 0.62–3.25), despite a higher rate of recanalization (RR: 1.69; 95% CI: 1.29–2.19).[107] In the study by Oyón et al, 39 of 60 (65.0%) patients with pancreatitis-associated SVT had an SpVT. Of note, 44% of untreated SpVT showed thrombosis resolution versus 0% of untreated PVT or MVT, suggesting that the rate of spontaneous recanalization may be higher in SpVT and that, in turn, PVT and MVT may obtain more benefit from anticoagulant treatment.[108]
Incidentally Detected Splanchnic Vein Thrombosis
Studies have shown that incidentally detected SVT carries similar prognosis than symptomatic SVT.[53]
[54] An individual patient data meta-analysis including 493 patients with incidental SVT highlighted that they carry a similar risk of major bleeding (incidence rate ratio [IRR]: 1.3; 95% CI: 0.8–2.2), a higher risk of recurrent VTE (IRR: 2.0; 95% CI: 1.2–3.3), and lower mortality rates (IRR: 0.5; 95% CI: 0.4–0.7), compared to symptomatic SVT.[109] Incidental SVT was less likely to be anticoagulated (72.4 vs. 83.6%); however, at multivariable analyses, anticoagulation was associated with lower risk of major bleeding, recurrent VTE, and mortality.[109]
The 2020 Guidelines of the American Society of Clinical Oncology (ASCO) suggest considering the treatment of incidental cancer-associated SVT on a case-by-case basis, while the 2020 ISTH guidance suggests providing the same treatment for symptomatic and incidental SVTs.[10]
[110]
Budd–Chiari Syndrome
BCS requires separate consideration due to the severity of this condition. Patients with BCS used to have a 3-year mortality rate greater than 80%; however, recent cohort studies have shown that the 5-year survival is greater than 80% if appropriately treated.[63]
Management of BCS is dual: treatment of the underlying thrombotic condition and decompression of the hepatic venous outflow.[78] A stepwise approach is used, which consists of medical treatment (anticoagulation), followed by endovascular procedures (angioplasty, stenting, thrombectomy, or thrombolysis), then TIPS, and, in most severe cases, liver transplantation.[77]
[100]
Anticoagulation is typically started with LMWH until clinical stability, and can subsequently be continued with oral anticoagulants, such as VKAs (INR target range: 2.0–3.0) or DOACs, if not contraindicated by severe liver dysfunction.[10]
[63] Anticoagulation in BCS patients should be administered long term.[10]
[77]
[100]
Two recent studies evaluated the use of DOACs in BCS patients. Sharma et al performed a retrospective analysis of 36 BCS patients treated with dabigatran after percutaneous endovascular intervention, and compared them with 62 BCS patients treated with VKAs, matched by age and site of thrombosis.[111] They found that dabigatran had similar rates of stent patency (91 vs. 93%) and major bleeding (3.5 vs. 6.5%) at 12 months of follow-up, compared to VKA. In an Austrian multicenter study, Semmler et al described 47 BCS patients, of whom 22 received DOACs for a median duration of 24.4 months.[112] The rate of major bleeding in the DOAC group was significant: four spontaneous major bleeding (18.2%) and one postprocedural major bleeding (4.5%). Of note, half of the patients in the DOAC group had Child–Pugh B/C cirrhosis; however, DOACs are contraindicated in Child–Pugh C and should be used with caution in Child–Pugh B.[77] The 5-year transplant-free survival was 91.6%.
Chronic Splanchnic Vein Thrombosis
Chronic SVT is defined by the presence of extensive venous collaterals or portal cavernoma. The term “chronic” is also applied to PVT persistent for more than 6 months.[77]
[78] The anticoagulant treatment of patients with chronic SVT should be evaluated on a case-by-case basis, balancing the risk of recurrent thrombosis and the risk of bleeding complications.[10] Factors that might favor anticoagulation include the presence of an underlying prothrombotic state, thrombosis involving the mesenteric veins, and current or prior intestinal ischemia.[77]
[113]
Ai et al evaluated 40 patients with chronic cirrhotic PVT treated with DOACs (26 with rivaroxaban 20 mg once daily and 14 with dabigatran 150 mg twice daily) and a control group of 40 patients with chronic cirrhotic PVT receiving no anticoagulant treatment, identified by propensity score matching.[114] At the 6-month follow-up, the rates of partial or complete recanalization were 28.2% in the DOAC group and 2.6% in the control group.
Recently, the RIPORT study enrolled 111 patients with noncirrhotic chronic PVT, without major thrombotic risk factors.[115] They were randomly assigned to rivaroxaban 15 mg once daily or no anticoagulant treatment. An interim analysis (median follow-up of 11.8 months) highlighted that none of the patients in the DOAC group developed VTE, while the incidence was 19.7 per 100 person-years in the control group. Thus, most of the untreated patients were switched to anticoagulation. Major bleeding throughout the entire study (median follow-up of 30.3 months) occurred in two patients in the DOAC group (1.0 per 100 person-years) versus one patient in the untreated group (1.2 per 100 person-years), thus suggesting that rivaroxaban was safe and effective.
Conclusions
SVT is a rare disorder that carries several therapeutic challenges. While it is well known that anticoagulant treatment should be started early to increase the chances of recanalization, there are situations that can delay it, such as active variceal bleeding.
Numerous therapeutic options are available, including UFH, LMWH, VKAs, and DOACs, which may be applicable to different categories of patients ([Table 1]). Treatment is often started with parenteral anticoagulants and continued with oral anticoagulants, if not contraindicated. DOACs are slowly becoming a more common treatment for SVT (sometimes as a single-drug approach); however, their use remains off-label for this indication. It is also uncertain whether lower doses could be used for secondary prevention of SVT.
Table 1
Summary of the anticoagulant treatment in special categories of patients with splanchnic vein thrombosis (SVT)
|
SVT in particular conditions
|
Suggested anticoagulant treatment
|
|
Liver cirrhosis
|
• Start with LMWH[10]
[77]
• Continue with LMWH[77] or switch to oral anticoagulants (VKAs or DOACs), if no contraindications[10]
[77]
• DOACs are contraindicated in Child–Pugh C cirrhosis (rivaroxaban is also contraindicated in Child–Pugh B); should be used with caution in Child–Pugh B cirrhosis[77]
• Indefinite anticoagulant treatment duration, liver cirrhosis being a persistent risk factor[10]
• Anticoagulation should be maintained until portal vein recanalization or for a minimum of 6 mo; continued in patients on the waiting list for a liver transplant; considered after recanalization in all patients[77]
|
|
Solid malignancies
|
• Possible options are LMWH or DOACs[10]
• Preference for LMWH if luminal gastrointestinal cancer, other active gastrointestinal mucosal diseases, genitourinary cancer at high bleeding risk, or concomitant chemotherapy interfering with DOACs[10]
• Indefinite anticoagulant treatment duration, solid cancer being a persistent risk factor[10]
|
|
Myeloproliferative neoplasms
|
• No specific recommendations on the type of anticoagulant
• Indefinite anticoagulant treatment duration, MPN being a persistent risk factor[10]
|
|
Pancreatitis
|
• No specific recommendations on the type of anticoagulant
• Definite anticoagulant treatment duration (3–6 mo), pancreatitis being a transient risk factor[10]
|
|
Incidentally detected SVT
|
• Same treatment as symptomatic acute SVT[10]
[103]
|
|
Budd–Chiari syndrome
|
• Stepwise approach: anticoagulation → endovascular procedures (angioplasty, stenting, thrombectomy, thrombolysis) → TIPS → liver transplantation[77]
• Possible options are LMWH, DOACs, or VKAs[10]
• Indefinite anticoagulant treatment duration[10]
[77]
|
|
Chronic SVT
|
• Case-by-case evaluation.[10]
• Factors that might support anticoagulation include an underlying prothrombotic state, mesenteric vein thrombosis, current or prior intestinal ischemia[113]
|
Abbreviations: DOACS, direct oral anticoagulants; LMWH, low-molecular-weight heparin; MPN, myeloproliferative neoplasms; TIPS, transjugular intrahepatic portosystemic shunt; VKAs, vitamin K antagonists.
Generally, treatment duration is based on the presence (or absence) and the type (transient vs. permanent) of risk factors. Patients presenting with unprovoked SVT or persistent risk factors are candidates for long-term treatment, with periodical reassessment of their thrombotic and bleeding risk. However, the role of thrombophilia screening in SVT patients is still unclear, particularly who should be screened and which thrombophilic abnormalities would result in a change of management, apart from triple-positive antiphospholipid syndrome. Furthermore, it is also uncertain whether the degree of vessel recanalization should influence the duration of anticoagulant treatment. More evidence is also required to assess the benefit and duration of anticoagulation in certain categories of patients, in particular those with MPNs or chronic SVT.
