Accuracy and Utility of Intraoperative Contrast-Enhanced Ultrasound (CEUS) After Complex Endovascular Aortic Aneurysm Repair
Carla K Scott, Jesus Porras Colon, Marilisa Soto Gonzalez, Felipe L Pavarino, Mirza S Baig, Melissa L Kirkwood, Carlos H Timaran
UT Southwestern Medical Center, Dallas, TX
Background: Endovascular repairs of complex aortic aneurysms generate high radiation exposure for patients and operating staff. Large doses of iodinated contrast media used during these procedures may also result in contrast-induced nephropathy. Exclusion of the aneurysm sac after the repair is usually assessed with angiography, which requires the use of significant additional doses of contrast and radiation. New imaging strategies that limit the exposure to radiation and iodinated contrast are necessary. Contrast agents for ultrasound have been introduced to provide improved blood ﬂow echogenicity for vascular imaging without the risk of nephropathy or radiation exposure. In EVAR surveillance, contrast-enhanced ultrasound (CEUS) has the advantage of being more speciﬁc than angiography, provides hemodynamic information and frequently detects the source of endoleaks. The utility of intraoperative CEUS during complex aortic aneurysm repair has not yet been established. The aim of this study was to assess the role of intraoperative CEUS examination and its accuracy for endoleak detection after Fenestrated/Branched Endovascular Aortic Aneurysm Repair (F-BEVAR) in comparison with standard completion and CT angiography (CTA).Methods: A single-center prospective cohort study of 84 patients who underwent F-BEVAR over a 23-month period using investigational devices for complex aortic aneurysms was performed. Complex aortic aneurysms included juxtarenal, suprarenal and thoracoabdominal aortic aneurysms. After F-BEVAR, completion digital subtraction angiography (DSA) was performed to confirm vessel patency and exclusion of the aneurysm sac. Completion cone-beam CT was also used to assess stent position and integrity, followed by intraoperative Color-Doppler ultrasound (CDUS) and CEUS performed under sterile conditions on the surgical ﬁeld before guidewire removal. The CDUS and CEUS were conducted by a sonography trained vascular surgeon. The CDUS was adjusted for optimum sensitivity to slow flows. The entire endograft and abdominal portion of the aneurysm sac were scanned to detect any endoleak and peak systolic velocities (PSV) of endoleaks were determined. The contrast enhancer was prepared and administered by the anesthesiologist according to the instructions for use. Subsequently, CEUS was performed using a 1-5MHz probe with a low mechanical index (0.06 – 0.10). A 2.4 ml bolus of a second-generation contrast agent was administered through a peripheral venous access (SonoVue™, BR1, Bracco), followed by 10 ml bolus of saline flush, which normally allows enhancer detection in the aorta after approximately 20 seconds. The sac scanning was performed for at least 3 minutes to detect both early and late endoleaks. CEUS images were recorded and analyzed for the presence of contrast enhancement within the aneurysm sac, with monitoring of the time of appearance (synchronous or delayed with respect to graft lumen enhancement) and identification of inflow and outflow sources. Afterwards, CDUS and CEUS were performed again, using the same protocol, after the administration of protamine for heparin reversal and the images were correlated with the prior exams. Presence and type of endoleaks were evaluated intraoperatively with DSA, CDUS and CEUS. CTA before discharge was obtained to assess visceral or branched vessel stenosis or occlusion and presence of endoleaks. The primary end point was the ability of intraoperative CEUS to detect the presence and source of endoleaks. The secondary end points included the utility of CEUS as an intraoperative adjunctive examination for endoleak management after F-BEVAR and the level of agreement between CEUS, CDUS, DSA, and CTA. Sensitivity, specificity, accuracy and predictive values of ultrasound examinations, DSA and CTA were assessed with the two-by-two table method analysis using the
presence of endoleaks on completion DSA and/or pre-discharge CTA as the reference standard. Agreement between diagnostic methods was calculated using kappa statistic. Receiver operating characteristic (ROC) analysis was used to assess the association between endoleak PSV and the presence of endoleak on the 30-day CTA. Results: Endoleaks were present in 56 patients (66.6%) after F-BEVAR according to completion DSA and/or pre-discharge CTA. Sensitivity and specificity of CEUS to detect endoleaks were 98.5% and 50%, respectively. In comparison, sensitivity to detect endoleaks for CDUS, DSA and CTA were 91.2%, 80.8% and 82.2%, with specificity of 86.6%, 93.7% and 75%, respectively (Table). The accuracy of CEUS to determine endoleak type was 89.3%, compared to 90.3% for CDUS, 65.3% for DSA and 81% for CTA. CEUS positive predictive value (PPV) and negative predictive value (NPV) were 89.3% and 88.8%, respectively. Compared to DSA, CEUS detected a greater number of type II endoleaks (sensitivity, 100% vs. 66.6%; P= .032). Similarly, more instances of residual false lumen perfusion were identified with CEUS after F-BEVAR for dissecting aortic aneurysms (sensitivity, 100% vs. 85.7%; P= .016). Other types of endoleaks were rare and the sample size was too small for valid comparisons. The level of agreement among imaging modalities was moderate between CEUS and CDUS (kappa=0.600), moderate between CEUS and DSA (kappa=0.323) and poor between CEUS and pre-discharge CTA (kappa=.095). ROC curve analysis revealed that an endoleak PSV ≥ 36 cm/s was significantly associated with the presence of a persistent endoleak on the 30-day CTA (area under the curve, 0.7; P=.027). The operative plan was altered in 10 procedures (12%) based on CEUS findings. In three procedures, large endoleaks of unknown origin were demonstrated on DSA. CEUS revealed that the source of these endoleaks were the residual constraining suture holes on the graft (Fig 1), which did not require a specific intervention. In six patients, CEUS allowed detection of the endoleak origin, which was key for management. In two procedures, endoleaks of unknown origin on DSA proved to be type IIIa with CEUS; sources of endoleak included the overlap between the thoracic endograft and the fenestrated device (Fig 2) and the overlap between the branched device and the bifurcated device. In both instances, the overlap was ballooned and the endoleak was resolved. In one patient, CEUS identified a type IIIc endoleak from incomplete renal stent apposition that was successfully treated with further ballooning. In two patients, CEUS identified endoleaks from the distal attachment sites, a type Ic endoleak from a renal stent and a Ib endoleak distal to an iliac branch device (IBD), which were successfully treated with stent extensions. In one patient, what appeared to be a type IIIa endoleak on DSA proved to be a type Ia on CEUS, which resolved after proximal ballooning of the fenestrated device. Conclusion: Intraoperative CEUS during F-BEVAR is an effective imaging study and demonstrates a high sensitivity and accuracy in detecting endoleaks when compared to DSA. Additionally, it provides complete hemodynamic real-time imaging with high temporal and spatial resolution for better classification of the endoleaks, preventing or assisting in further interventions before completion of the procedure. CEUS may represent an additional, useful tool for intraoperative complex aneurysm repair. Combined CEUS and CDUS findings provide further optimal assessment of the significance of endoleaks detected after F-BEVAR and the possibility of endoleak persistence on 30-day CTA. Future studies with larger patient numbers are required to further validate the utility and role of CEUS examination during F-BEVAR.
Back to 2022 Abstracts