Related Experiment Video
Updated: Oct 7, 2026

Multimodality Diagnosis of Mesenteric Ischemia
Published on: July 21, 2023
Failure Modes of Endovascular Therapy for Spontaneous Isolated Superior Mesenteric Artery Dissection
Jianping Xiong1, Shilu Zhao1, Qichen Feng1
1Department of Interventional Radiology and Vascular Surgery, Peking University Third Hospital, 100191, Beijing, China.
Background:
Mechanisms of failure after endovascular stenting for symptomatic spontaneous isolated superior mesenteric artery dissection (SISMAD) remain poorly characterized. We analyzed device- and procedure-related failure modes in a 17-year single-center cohort.
Methods:
We reviewed 117 consecutive patients with SISMAD (2008-2025). Patients stented for persistent or escalating pain despite conservative management formed the cohort; catheter-infusion and open-surgery patients were analyzed separately. Lesions were classified by Yun morphology. Failure modes were classified a priori as Type I (intraprocedural technical failure), Type II (thrombotic events occurring after the procedure), and Type III (in-stent restenosis or intimal hyperplasia at ≥6 months). Patency and reinterventions were recorded.
Results:
Thirty-two patients underwent endovascular therapy (28 stenting, 4 catheter-directed infusion). In the stent cohort, uncomplicated technical success was 82.1% (23/28). Type I failures occurred in 17.9% (5/28), dominated by false-lumen deployment (10.7%) clustered in Yun III occlusions (50.0% vs 9.1%; P = .050). Type II events (3 weeks; 24 months) were rescued by reintervention. Type III events developed in 22.7% (5/22), confined to multi-stent constructs (45.5% vs 0%; P = .035). Primary patency was 81.8% (18/22) at a median 19-month imaging follow-up; secondary patency was 100% after 3 reinterventions. Greater baseline true-lumen stenosis and Yun III morphology predicted composite events. No patient died; the only bowel resection preceded secondary stenting.
Conclusions:
Endovascular therapy for SISMAD fails in characteristic, morphology-dependent patterns. The proposed three-type classification is exploratory and requires validation in larger cohorts.