Mesenteric vein thrombosis can be safely treated with anticoagulation but is associated with significant sequelae of portal hypertension.

BACKGROUND

Mesenteric venous thrombosis (MVT) is a relatively uncommon but potentially lethal condition associated with bowel ischemia and infarction. The natural history and long-term outcomes are poorly understood and under-reported.

METHODS

A single-institution retrospective review of noncirrhotic patients diagnosed with MVT from 1999 to 2015 was performed using International Classification of Diseases, Ninth Revision and radiology codes. Patients were excluded if no radiographic imaging was available for review. Eighty patients were identified for analysis. Demographic, clinical, and radiographic data on presentation and at long-term follow-up were collected. Long-term sequelae of portal venous hypertension were defined as esophageal varices, portal vein cavernous transformation, splenomegaly, or hepatic atrophy, as seen on follow-up imaging.

RESULTS

There were 80 patients (57.5% male; mean age, 57.9 ± 15.6 years) identified; 83.3% were symptomatic, and 80% presented with abdominal pain. Median follow-up was 480 days (range, 1-6183 days). Follow-up radiographic and clinical data were available for 50 patients (62.5%). The underlying causes of MVT included cancer (41.5%), an inflammatory process (25.9%), the postoperative state (20.7%), and idiopathic cases (18.8%). Pancreatic cancer was the most common associated malignant neoplasm (53%), followed by colon cancer (15%). Twenty patients (26%) had prior or concurrent lower extremity deep venous thromboses. Most patients (68.4%) were treated with anticoagulation; the rest were treated expectantly. Ten (12.5%) had bleeding complications related to anticoagulation, including one death from intracranial hemorrhage. Four patients underwent intervention (three pharmacomechanical thrombolysis and one thrombectomy). One patient died of intestinal ischemia. Two patients had recurrent MVT, both on discontinuing anticoagulation. Long-term imaging sequelae of portal hypertension were noted in 25 of 50 patients (50%) who had follow-up imaging available. Patients with long-term sequelae had lower recanalization rates (36.8% vs 65%; P = .079) and significantly higher rates of complete as opposed to partial thrombosis at the initial event (73% vs 43.3%; P < .005). Long-term sequelae were unrelated to the initial cause or treatment with anticoagulation (P = NS).

CONCLUSIONS

Most cases of MVT are associated with malignant disease or an inflammatory process, such as pancreatitis. A diagnosis of malignant disease in the setting of MVT has poor prognosis, with a 5-year survival of only 25%. MVT can be effectively treated with anticoagulation in the majority of cases. Operative or endovascular intervention is rarely needed but important to consider in patients with signs of severe ischemia or impending bowel infarction. There is a significant incidence of radiographically noted long-term sequelae from MVT related to portal venous hypertension, especially in cases of initial complete thrombosis of the mesenteric vein.