Skip to main content
SIR 2026 Event Main banner

SIR 2026

Traditional Poster

Venous Interventions

Imaging

Session: Traditional Poster Presentations

617 - Pulmonary Arterial Perfusion in Acute Pulmonary Embolism: Perfusion Deficit Correlates with Short Term Outcomes

Monday, April 13, 2026
12:10 PM - 1:10 PM ET
Purpose: To evaluate pulmonary arterial perfusion imaging using dual-energy CT (DECT) as a predictor of outcomes and potential clinical decision-making tool for patients with pulmonary embolism (PE).

Materials and Methods: We retrospectively analyzed dual energy CT pulmonary angiography exams in patients with central PE. Patients with segmental/subsegmental PE or parenchymal lung disease were excluded. Median age of the final cohort was 61.5 years (± 11.3), with 59.1% female and 40.9% male. Anatomic imaging was used to identify clot extent, quantified using the Qanadli score. Perfusion data based on dual energy iodine perfusion imaging were extracted and registered to all 5 lung lobes to quantify perfusion to each lobe. Perfusion deficits were compared across clotted vs. non-clotted lobes. Perfusion was also compared to Qanadli scores, discretized into low (< 18), intermediate (18-33), and high ( >33) categories, as well as 30-day mortality. Statistical testing included t-tests, Wilcoxon rank-sum with Benjamini-Hochberg correction, and correlation analyses.

Results: A total of 110 lobes were analyzed (65 clotted, 45 non-clotted), distributed by arterial level as follows: 60 main pulmonary arteries, 30 interlobar arteries, and 20 lobar arteries. The most common cancers were colorectal (11.4%), liver (11.4%), breast (11.4%), sarcoma (11.4%), and pancreas (8.6%). Clotted lobes had greater perfusion deficits than non-clotted lobes (28.8% ± 13.4% vs 16.3% ± 5.3%; p = 1.1×10⁻⁹). Perfusion deficit was also associated with worse 30-day mortality (p = 0.0001) as well as median overall survival (p < 0.001). Perfusion deficit, however, was not associated with Qanadli severity (p = 0.3).

Conclusion: DECT-derived perfusion strongly correlates with clot presence and predicts 30-day mortality, supporting its value as a functional biomarker of PE severity. Lack of association with Qanadli scores suggests perfusion imaging provides complementary prognostic information beyond anatomic measures. For example, small, non-occlusive thrombi yield the same contribution as occlusive clots to the Qanadli score but do not have a significant impact on actual perfusion. On the other hand, lobes without clot also demonstrated low perfusion, likely reflecting underlying lung disease; this information could help inform about overall lung function and facilitate the clinical decision-making for reperfusion interventions to clotted lobes in these patients. These findings highlight the potential of DECT perfusion to supplement serologic, anatomic, and hemodynamic assessments and refine risk stratification.