Discrimination of High-Risk Type-2 Endoleak after Endovascular Aneurysm Repair through CT Perfusion: A Feasibility Study

Use of the Endoleak-to-Aortic Density Ratio to Distinguish Direct Endoleaks from Indirect Endoleaks after Endovascular Aortic Aneurysm Repair

PURPOSE

To define criteria to distinguish direct (type 1 or 3) from indirect endoleaks (type 2) in the arterial phase of contrast-enhanced computed tomography (CT) scans in patients with abdominal aortic aneurysms treated with endovascular aortic repair.

MATERIALS AND METHODS

This retrospective study was conducted from January 2009 to October 2020 and included consecutive patients treated endovascularly for a direct endoleak or an indirect endoleak associated with an enlarging aneurysm. The following characteristics were evaluated using contrast-enhanced CT: location, size, contact with the endograft, density, morphologic criteria, collateral artery enhancement, and endoleak-to-aortic density ratio. Statistical analysis included the Mann-Whitney U test, Pearson χ2 test, Fisher exact test, receiver operating characteristic curve analysis, and multivariable logistic regression.

RESULTS

Contrast-enhanced CT scans from 71 patients (87% men), who presented with 87 endoleaks (44 indirect and 43 direct endoleaks), treated by endovascular techniques were analyzed. Using visual criteria, 56% of the endoleaks were not characterizable as direct or indirect. An endoleak-to-aortic density ratio of >0.77 could properly distinguish direct from indirect endoleaks, with a theoretical accuracy of 98% (area under the receiver operating characteristic curve, 0.99), sensitivity of 95%, specificity of 100%, positive predictive value of 100%, and negative predictive value of 96%.

CONCLUSION

An endoleak-to-aortic density ratio of >0.77 in the arterial phase of contrast-enhanced CT could be a strong discriminant of a direct-type endoleak.

Dynamic CT angiography is more accurate in diagnosing endoleaks than standard triphasic CT angiography and enables targeted embolization

PURPOSE

The primary objective was to compare the accuracy of dynamic CT angiography (d-CTA) with standardized triphasic contrast enhanced CT angiography (t-CTA) in diagnosing endoleak type after endovascular aortic repair (EVAR) using digital subtraction angiography (DSA) as reference standard. The secondary objective was to study the impact of d-CTA on image-fusion guided endoleak embolization.

MATERIALS AND METHODS

Retrospective review of patients who underwent d-CTA imaging after EVAR between March 2019 and July 2021 was performed. De-identified images were independently reviewed by two-two blinded readers to document endoleak type and target vessels. Impact of d-CTA-guided embolization was evaluated by number of planning angiograms, radiation exposure and accuracy of target vessel overlay.

RESULTS

During the study period, 52 patients underwent d-CTA, 19 had all three modalities available for analysis. DSA imaging confirmed 4(21.0%) type-I, 14(73.7%) type-II and 1(5.3%) type-III endoleak. Findings from d-CTA matched with DSA in 19/19 cases (100%), whereas t-CTA matched in 14/19 cases (73.7%). In type-II endoleaks, number of target vessels identified by d-CTA, t-CTA and DSA were 23, 17 and 16 respectively. Mean dose-length product from d-CTA and t-CTA was 1445±551 and 1612±530 mGy*cm (p=0.26). Nine patients underwent d-CTA-guided type-II endoleak embolization, using a median of 1(range:1-4) planning angiogram before embolization utilizing 21.6(±8.7)% of total procedural radiation dose. Target vessel overlay was accurate in 9/9(100%) cases.

CONCLUSION

Dynamic, time-resolved CTA is more accurate compared to standardized triphasic contrast enhanced CTA in diagnosing endoleak type after EVAR. In type-II endoleak, d-CTA better identified target vessels and enabled safe, targeted embolization.