Venous mesenteric ischemia carries high procedural burden and elevated mortality in patients with severe presentation

Objective: Venous mesenteric ischemia (VMI) presents with variable severity resulting in a spectrum of outcomes. This study sought to characterize the natural history of VMI and identify drivers of intervention and adverse outcomes.

Methods: All patients who presented to our institution with acute and subacute VMI between 1993 and 2016 were identified. Images were reviewed to determine thrombosis location and charts were reviewed to identify clinical factors and outcomes. Univariate analysis was performed for demographics, comorbidities, and presenting characteristics, with primary outcomes of intervention, readmission, and 30-day mortality. A survival analysis was performed with log-rank difference testing for demographics, comorbidities, and presenting characteristics.

Results: We identified 103 patients with acute and subacute VMI. The locations of the thrombosis included the superior mesenteric vein (SMV) (31.1%); SMV and portal vein (35.9%); SMV, portal, and splenic veins (15.5%); and other combinations of portomesenteric veins (17.4%), without correlation between the location and outcomes. Most patients were male (60.6%), 22.3% were actively smoking, and the median Charlson comorbidity score was 4 (interquartile range, 2-7). The mean patient age was 61.3 years. More than one-half had a known hypercoagulability (52.4%), 22.3% had prior bowel resection, and 8.7% had prior mesenteric venous intervention, including transjugular intrahepatic portosystemic shunt procedures. Thirty-five patients underwent 83 procedures during their hospitalization, and 23 patients underwent surgical intervention specifically. Prior mesenteric venous procedure, abdominal tenderness, and lactatemia of more than 1.5 mmol/L were associated with an increased need for surgical intervention (P < .05). Patients with leukocytosis of greater than 10K/μL had increased surgical intervention (P = .10), although without statistical significance. However, symptoms for less than 2 weeks (P < .05) were associated with decreased surgical intervention. The 30-day mortality was low in this cohort (6.8%), but was increased in patients requiring intervention (11.4%). For those undergoing procedures, a shorter time to intervention was associated with an improved 30-day mortality (8.7% for procedures on hospital days 0-1 vs 16.7% for hospital day 2 or later; P = .01). Abdominal tenderness and lactatemia were associated with increased 30-day mortality (6.8% vs 3.6% [P < .01] and 16.0% vs 3.8% [P = .03], respectively). A Kaplan-Meier survival analysis revealed a median survival of 7.1 years, with a 1-year survival rate of 74.9%, a 3-year survival rate of 67.1%, and a 5-year survival rate of 57.9%. Negative predictors of survival included a higher Charlson comorbidity index (hazard ratio, 3.7; P < .01) and malignancy (hazard ratio, 3.1; P < .01).

Conclusions: The 30-day mortality of VMT is low, but more than one-third of patients required an intervention beyond anticoagulation. Comorbidity, a prior mesenteric vessel or intestinal operation, and presentation with tenderness or relevant laboratory abnormalities portend worse outcomes. Early intervention is associated with improved outcomes.