The natural history of portal venous system aneurysms.

backgroundPortal venous system aneurysms (PVAs) are increasingly diagnosed on cross-sectional computed tomography imaging. However, the natural history of these aneurysms is poorly understood, and reports are limited to small case series.

methodsTerms relevant to PVAs were searched in radiology reports (2010-2022), with PVA presence confirmed by manual review. PVA were defined as a diameter greater than 1.5 cm in patients without cirrhosis and 1.9 cm in those with cirrhosis. Aneurysm growth was defined as greater than 20% increase in size, whereas aneurysm regression was defined as greater than 20% decrease in size. Patient demographics, comorbid conditions, and PVA outcomes were abstracted. Univariate statistics were used to compare groups.

resultsThirty-eight aneurysms with radiographic follow up were identified in 35 patients, involving the portal vein (n = 18; 47.4%), splenic vein (n = 10; 26.3%), superior mesenteric vein (n = 3; 7.9%), and portal confluence (n = 7; 18.4%). Although 12 (31.6%) were idiopathic, the remaining 26 (68.4%) were associated with portal hypertension (n = 20; 52.6%) and prior liver transplant (n = 4; 10.5%). The median growth was 0.2 cm (range, -2.6 to 2.4 cm) over median follow up over 5.0 years (range, 0.3-16.6 years). Five PVAs (13.2%) regressed and were largely idiopathic (80.0%; P = .03). Thirteen PVAs (34.2%) grew and were associated with portal hypertension (n = 11; 84.6%; P = .003) and thrombosis (n = 6; 46.2%; P = .05). Nine PVAs (23.7%) thrombosed, predominantly in males (n =7; 77.8%). The median growth was 1.0 cm (range, -0.7 to 1.9 cm). Three patients (33.3%) were symptomatic from PVA thrombosis including abdominal pain (n = 2; 22.2%), intestinal ischemia (n = 1; 11.1%), and variceal bleeding (n = 2; 22.2%). Four patients (44.4%) were treated with anticoagulation. No aneurysms ruptured. Of the 58 PVAs initially identified with and without radiographic follow up, five (8.6%) underwent intervention with a median diameter of 4.0 cm (range, 3.4-5 cm). Intervention included vein ligation (n = 1; 20.0%), aneurysmorrhaphy (n = 1; 20.0%), and aneurysmectomy (n = 3; 60.0%). There was one case of aneurysm recurrence 20 years following resection and one postoperative mortality.

conclusionsTwo-thirds of PVAs, including those with size greater than 3 cm, remain stable on surveillance. Although annual surveillance is initially recommended to confirm PVA stability, interval imaging can be subsequently extended given low growth rates. Over 20% of PVAs thrombosed, but none ruptured. Although we did not observe any cases of rupture, the devastating consequences of rupture necessitate consideration of surgical intervention for large symptomatic PVAs.