Fate of primary determinate and indeterminate target vessel endoleaks after fenestrated-branched endovascular aortic repair.

OBJECTIVE: The aim of this study was to investigate the outcomes of primary determinate and indeterminate target vessel endoleaks (TVELs) after fenestrated-branched endovascular aortic repair (F-BEVAR). METHODS: We conducted a single-center retrospective study (2014-2023) on F-BEVAR for thoracoabdominal (TAAAs) or pararenal aortic aneurysms (PRAAs). TVELs were classified as "primary" if present at the first postoperative computed tomography angiogram. Endoleaks were defined "determinate" (dELs) if the cause (type Ic or IIIc) and implicated target vessel were identifiable and "indeterminate" (iELs) if contrast enhancement was detectable at the level of fenestrations/branches without any evident source. Endoleaks involving multiple inflows (type II and target vessels) were defined as "complex" (cELs). Endpoints were endoleak spontaneous resolution, 1-year aneurysm sac failure to regress (>5 mm diameter decrease), and 4-year endoleak-related secondary interventions. Kaplan-Meier estimates and Cox regression were used for the analysis. RESULTS: There were 142 patients with JRAAs/PRAAs (n = 85; 60%) or TAAAs (n = 57; 40%), with 513 target arteries incorporated through a fenestration (n = 294; 57%) or directional branch (n = 219; 43%). Fifty-nine primary TVELs (12%) were identified in 35 patients (25%), a dEL in 20 patients (14%) and iEL in 15 (11%); 22 (15%) had a determinate or indeterminate cEL. Overall spontaneous resolution rate was 75% (95% confidence interval [CI], 51%-87%) at 4 years. cELs (odds ratio [OR], 5.00; 95% CI, 1.10-49.4; P < .001) and iELs after BEVAR (OR, 9.43; 95% CI, 3.41-56.4; P = .002) were more likely to persist >6 months, and persistent forms were associated with sac failure to regress at 1 year (OR, 1.72; 95% CI, 1.03-12.59; P = .040). Overall freedom from endoleak-related reinterventions was 85% (95% CI, 79%-92%) at 4 years, 92% (95% CI, 87%-97%) for those without primary TVELs and 62% (95% CI, 46%-84%) for those with any primary TVEL (P < .001). In particular, cELs (hazard ratio, 1.94; 95% CI, 1.4-18.81; P = .020) were associated with an increased need for reintervention. In case a secondary intervention was needed, iEL or cEL had an increased risk for multiple secondary procedures (hazard ratio, 2.67; 95% CI, 1.22-10.34; P = .034). CONCLUSIONS: Primary TVELs are frequent after F-BEVAR, and a clear characterization of the endoleak source by computed tomography angiogram is not possible in 40% of patients. Most primary TVELs spontaneously resolve, but during follow-up, patients with any primary TVEL experience a worsened freedom from endoleak-related reinterventions that is mostly driven by persistence of cELs and post-BEVAR iELs. Multiple secondary procedures may be required in case of iELs or cELs.

Effect of narrow paravisceral aorta on target vessel instability after fenestrated and branched endovascular aortic repair.

OBJECTIVE: To investigate the effect of narrow paravisceral aorta (NPA) on target vessel instability (TVI) after fenestrated-branched endovascular aortic repair. METHODS: We conducted a single-center retrospective study (2014-2023) of patients treated by fenestrated-branched endovascular aortic repair for thoracoabdominal aortic aneurysms (TAAA) or pararenal aortic aneurysms. The paravisceral aorta was defined as the aortic segment limited by the diaphragmatic hiatus proximally and the emergence of lower renal artery distally, and was considered "narrow" in case of a minimum inner diameter of <25 mm. The minimum aortic diameter, location, longitudinal extension, angulation, calcification, and thrombus thickness of NPA were evaluated at the preoperative computed tomography angiogram. End points were 30-day technical success and freedom from TVI. RESULTS: There were 142 patients with JRAA/pararenal aortic aneurysm (n = 85 [59%]) and extent IV (n = 24 [17%]) or extent I-III (n = 33 [23%]) TAAA, with 513 target arteries successfully incorporated through a fenestration (n = 294 [57%]) or directional branch (n = 219 [43%]). A NPA was present in 95 patients (70%), 73 (86%) treated by fenestrated endovascular aortic repair (FEVAR) and 22 (39%) by branched endovascular aortic repair (BEVAR). The overall 30-day mortality was 2% and technical success was 99%, without differences between NPA and non-NPA (P = .99). Kaplan-Meier estimated freedom from TVI at 4 years was 82%, 81% (95% CI, 75-95) in patients with a NPA and 80% (95% CI, 68-94) and in those without NPA (P = .220). The result was maintained for both FEVAR (NPA: 81% [95% CI, 62-88]; non-NPA: 76% [95% CI, 60-99]; P = .870) and BEVAR (NPA: 77% [95% CI, 69-99]; non-NPA: 80% [95% confidence interval (CI) 66-99]; P = .100). After multivariate analysis, the concomitant presence of a NPA <20 mm and angulation of >30° was significantly associated with TVI in FEVAR (HR, 3.21; 95% CI, 1.03-48.70; P = .036), being the result mostly driven by target vessel occlusion. In BEVAR, a NPA diameter of <25 mm was not associated with TVI (HR, 2.02; 95% CI, 0.59-5.23; P = .948); after multivariate analysis, the use of outer branches in case of a NPA longitudinal extension of >25 mm (hazard ratio [HR], 3.02; 95% CI, 1.01-36.33; P = .040) and NPA severe calcification (HR, 1.70; 95% CI, 1.00-22.42; P = .048) were associated with a higher chance for TVI. CONCLUSIONS: FEVAR and BEVAR are both feasible in cases of NPA and provide satisfactory target vessels durability. The use of outer branches should be avoided in cases with an inner aortic diameter of <25 mm with a longitudinal extension of >25 mm or moderate to severe NPA calcifications. In FEVAR, bridging stent patency may be negatively influenced by NPA of <20 mm in association with aortic angulation of >30°.

Clinical Impact and Determinants of Fenestration to Target Vessel Misalignment in Fenestrated Endovascular Aortic Repair.

OBJECTIVE: This single centre, retrospective study (2014 - 2022) on juxta-, pararenal, or thoraco-abdominal aortic aneurysms treated by fenestrated endovascular aortic repair (FEVAR) was conducted to investigate the clinical impact and determinants of fenestration to target vessel misalignment in FEVAR. METHODS: Pre-operative supracoeliac, pararenal, and infrarenal aortic angles were measured on three dimensional computed tomography angiography (CTA) reconstructions. Two components of misalignment were measured on the first post-operative CTA: horizontal misalignment (angle between the fenestration and the target vessel ostium on perpendicular CTA cuts) and vertical misalignment (vertical distance between the fenestration and the target vessel at its origin). Endpoints were freedom from target vessel instability (TVI) and alignment change over time. RESULTS: Of 65 patients treated by FEVAR, 60 (202 target arteries) with juxta-, pararenal (80%), or thoraco-abdominal aortic aneurysm (20%) were included. Mean horizontal misalignment was 9 ± 12° (median 5°; IQR 0 - 16) and mean vertical misalignment was 0.7 ± 1 mm (median 0 mm, IQR 0 - 1). Freedom from TVI was 92% (95% CI 88 - 98) at 36 months. Horizontal misalignment > 15° was significantly associated with TVI (HR 5.19; 95% CI 1.54 - 17.48; p = .008); vertical misalignment did not significantly impact TVI (HR 0.99; 95% CI 0.56 - 1.73; p = .97). By multivariable analysis, pararenal aortic angle (OR 1.01 per increased degree of angulation; 95% CI 1.00 - 1.02; p = .044), bridging distance > 5 mm (OR 1.07; 95% CI 1.02 - 1.11; p = .003), and use of higher profile endografts in tortuous iliac access (OR 7.55; 95% CI 4.55 - 1.11; p = .003) were associated with clinically significant misalignment. Bridging distance > 5 mm (OR 2.00; 95% CI 1.02 - 11.29; p = .044), degree of baseline misalignment (OR 1.04; 95% CI 1.01 - 1.08; p = .036), and persistence of any primary endoleak for > 6 months (OR 5.85; 95% CI 1.23 - 29.1; p = .023) were associated with misalignment increase during follow up. CONCLUSION: Horizontal misalignment > 15° is associated with worsened target vessel outcomes. This may occur as a result of excessive iliac access tortuosity, high pararenal aortic angulation, and bridging distance > 5 mm.

Impact of Carotid Stent Design on Embolic Filter Debris Load During Carotid Artery Stenting.

BACKGROUND: The carotid stent design may influence the risk of embolization during carotid artery stenting. The aim of the study was to assess this risk by comparing the quantity of embolized material captured by filters during carotid artery stenting, using different stent designs. METHODS: We conducted a single-center retrospective study of patients undergoing carotid artery stenting for asymptomatic carotid stenosis >70% (2010-2022) in a tertiary academic hospital (Padua University Hospital, Italy). Carotid stents were classified according to their design as open-cell (OCS), closed-cell (CCS), or micromesh stents (MMS). A distal filter protection was used in all patients, and the amount of captured embolized particles was semiautomatically analyzed using a dedicated software (Image-Pro Plus, Media Cybernetics). Primary end point was embolic filter debris (EFD) load, defined as the ratio of the filter area covered by particulate material to the total filter area. Secondary end points were 30 days major stroke and death. RESULTS: Four-hundred-eighty-one carotid artery stentings were included; 171 (35%) using an OCS, 68 (14%) a CCS, and 242 (50%) a MMS. Thirty-days mortality was 0.2% (n=1) and major stroke rate was 0.2% (P=0.987). Filters of patients receiving MMS were more likely to be free from embolized material (OCS, 30%; CCS, 13%; MMS, 41%; P<0.001) and had a lower EFD load (OCS, 9.1±14.5%; CCS, 7.9±14.0%; MMS, 5.0±9.1%; P<0.001) compared with other stent designs. After stratification by plaque characteristics, MMS had a lower EFD load in cases of hypoechogenic plaque (OCS, 13.4±9.9%; CCS, 10.9±8.7%; MMS, 6.5±13.1%; P<0.001), plaque length>15 mm (OC, 10.2±15.3; CC, 8.6±12.4; MM, 8.2±13.6; P<0.001), and preoperative ipsilateral asymptomatic ischemic cerebral lesion (OCS, 12.9±16.8%; CCS, 8.7±19.5%; MMS, 5.4±9.7%; P<0.001). After multivariate linear regression, use of MMS was associated with lower EFD load (P=0.038). CONCLUSIONS: The use of MMS seems to be associated with a lower embolization rate and EFD load, especially in hypoechogenic and long plaques and in patients with a preoperative evidence of asymptomatic ischemic cerebral lesion.

The "safe-line" technique as theoretical additional attempt to mitigate spinal cord ischemia after urgent complete endovascular exclusion of a thoracoabdominal aortic aneurysm.

We describe the feasibility of a technique for temporary aneurysm sac reperfusion after endovascular single-stage thoracoabdominal aortic aneurysm exclusion, to be used in the case of postoperative spinal cord ischemia. Two cases were treated for impending rupture of a thoracoabdominal aortic aneurysm. Before completion of sac exclusion, a supplementary buddy wire (V-18 control guidewire; Boston Scientific) was advanced in parallel fashion from the left percutaneous femoral access into the aneurysmal sac on the posterior aspect of the endograft. Distal aneurysm exclusion was completed using the main superstiff guidewire, and the femoral access was closed with a percutaneous closure device (ProGlide; Abbott) in standard fashion, leaving in place the sole V-18 guidewire, draped in sterile fashion. In the case of spinal cord ischemia, the "safe-line" can be rapidly used for spinal reperfusion after trans-sealing exchange with a 6F, 65-cm-long Destination sheath (Terumo) connected to a 6F introducer on the contralateral femoral artery.

Geometrical determinants of target vessel instability in fenestrated endovascular aortic repair.

OBJECTIVE: To investigate geometrical determinants of target vessels instability in fenestrated endovascular aneurysm repair (FEVAR), using a computed tomography angiogram postimplantation analysis. METHODS: We retrospectively reviewed single-center data on consecutive patients undergoing FEVAR (2014-2021). The geometrical analysis consisted in the assessment of bridging stent lengths and diameters, stent conformation, and graft misalignment. Bridging stent length was categorized in three components: protrusion length (PL) into the main endograft, bridging length (BL) between the fenestration and the origin of the target vessel, and sealing length (SL) of apposition in the target vessel. The conformation was measured as the flare ratio (the ratio of maximum to minimum bridging stent diameter within the PL). Horizontal misalignment was measured as the angle between the fenestration and the target vessel ostium on computed tomography angiography axial cuts. The primary end point was freedom from target vessel instability; secondary end points were target vessels primary patency and freedom from related endoleaks. Time-dependent outcomes were estimated as Kaplan-Meier curves; Cox proportional hazards were used to identify the predictors of target vessel instability. RESULTS: There were 46 patients (juxta/pararenal: n = 34 [74%]; thoracoabdominal: n = 11 [26%]), with 147 target arteries incorporated through a bridging stent. Freedom from target vessel instability was 87% (95% confidence interval [CI], 80-94) at 42 months. Primary patency was 98% (95% CI, 96-100) and freedom from endoleak was 85% (95% CI, 76-93). PL (hazard ratio [HR], 1.08; 95% CI, 0.22-5.28; P = .923), sealing length (HR, 0.95; 95% CI, 0.87-1.03; P = .238), and flare ratio (HR, 4.66; 95% CI, 0.57-37.7; P = .149) were not associated with target vessel instability. By multivariate analysis, a BL of more than 5 mm (HR, 4.98; 95% CI, 1.13-21.85; P = .033) was significantly associated with instability. Patients with a BL 5 mm or more had a significantly greater degree of horizontal misalignment (21 ± 12° vs 9 ± 13°; P = .011). CONCLUSIONS: An optimal geometrical conformation between the bridging stent and the main endograft at the level of target vessels is warranted to improve the midterm outcomes of FEVAR. A BL of more than 5 mm was associated with a greater risk of target vessel instability, likely as a result of a less accurate endograft alignment. The sizing and planning of FEVAR should be performed to maintain a BL of less than 5 mm.

Determination of Optimal and Safest Proximal Sealing Length During Thoracic Endovascular Aortic Repair.

OBJECTIVE: To determine the optimal and safest proximal sealing length (PSL) during thoracic endovascular aortic repair (TEVAR), depending on anatomical aortic arch types and proximal landing zones (LZs). METHODS: This was a single centre retrospective observational study of consecutive TEVAR patients (2008-2020). All aortic pathologies requiring Ishimaru landing zone (LZ) 0 - 3 were included; results were stratified by aortic arch type. The PSL was measured as the length of complete aortic wall to endograft apposition at the level of the proximal neck. The primary endpoint was proximal failure (type 1A endoleak, endograft migration, or re-intervention requiring proximal graft extension). Freedom from proximal failure was estimated with Kaplan-Meier curves. An "optimal" sealing length (PSL cutoff maximising sensitivity + specificity for proximal failure) and "safest length" (PSL cutoff determining ≥ 90% sensitivity) were identified using receiver operating characteristic curve analysis. RESULTS: One hundred and forty patients received TEVAR; mean ± standard deviation PSL was 29 ± 9 mm. Freedom from proximal endograft failure at five years (median 31 months) was 82.4% (95% confidence interval [CI] 72 - 95); the shorter the PSL, the greater was the risk of failure (hazard ratio 0.90, 95% CI 0.84 - 0.97; p = .004). Overall optimal and safest PSL were 25 mm (sensitivity 78%, specificity 66%) and 30 mm (sensitivity 92%, specificity 30%), respectively. In type I arch, the optimal PSL was 22 mm (sensitivity 50%, specificity 87%). In type II, the optimal PSL was 25 mm (sensitivity 89%, specificity 59%) overall and 27 mm for type II/LZ 2 - 3 (sensitivity 31%, specificity 68%). For type III, the optimal PSL was 27 mm (sensitivity 80%, specificity 87%); the safest was 30 mm (sensitivity 100%, specificity 61%) In type III/LZ 2 - 3, the optimal PSL was 27 mm (sensitivity 31%, specificity 68%) and safest was 30 mm (sensitivity 100%, specificity 55%). CONCLUSION: A 20 mm PSL may be acceptable only for type I arches. For types II/III, that represent the majority of cases, a 25 - 30 mm PSL may be required for a safe and durable TEVAR.