Accuracy of a CBCT-based virtual injection software for vessel detection during hepatic arterial embolization.

PURPOSE: To assess the accuracy, sensitivity, positive predictive value (PPV) and interobserver agreement of a virtual injection (VI) software that simulates selective arterial injection from nonselective cone-beam CT (CBCT) arteriography. METHODS: From March 2019 to May 2020, 20 consecutive patients in whom a nonselective injected CBCT and a selective CT angiography (CTA) were completed in the same procedure, were retrospectively included. The position of the microcatheter tip used for selective CTA injection was identified. The VI was simulated from the exact same point on the nonselective CBCT and the two volumes were merged. VI was compared to the real injection on the selective CTA. Three interventional radiologists evaluated the accuracy using a 6-point scale (Perfect; Good; Fair; Incorrect Origin; False Negative; Non existing). Sensitivity, PPV, and Fleiss' kappa were calculated. Numerical variables were presented as means ± standard deviations. RESULTS: Twenty procedures and 195 vessel segments were analyzed. Most vessels were 4th order (57/195; 29%) and 5th order (96/195; 49%). VI was classified as perfect to good in 96.8% ± 1.4 of 1st-3rd order arteries and in 83.4% ± 0.4 of 4th-5th order arteries. Interobserver agreement was substantial (Fleiss' kappa = 0.79; 95% confidence interval = 0.73-0.84, P < 0.01). False negatives were reported with a mean of 9.4% ± 0.3. Average sensitivity was 90.6% ± 0.3 and average PPV was 92.7% ± 0.02. Fourteen false positives were noted. CONCLUSION: CBCT-based VI software accurately simulated distal injections in the liver with high sensitivity and a substantial interobserver agreement.

Augmented fluoroscopy guided transbronchial pulmonary microwave ablation using a steerable sheath.

Background: Transbronchial microwave ablation (MWA) is a promising novel therapy. Despite advances in bronchoscopy and virtual navigation, real time image guidance of probe delivery is lacking, and distal maneuverability is limited. Cone-beam computed tomography (CBCT) based augmented fluoroscopy guidance using steerable sheaths may help overcome these shortcomings. The aim of this study was to evaluate feasibility and accuracy of augmented fluoroscopy guided transbronchial MWA with a steerable sheath and without a bronchoscope. Methods: In this prospective study, procedures were performed under general anesthesia. Extra-bronchial lung synthetic targets were placed percutaneously. Target and airways extracted from CBCT, with planned bronchial parking point close to the target were overlaid on live fluoroscopy. Endobronchial navigation was solely performed under augmented fluoroscopy guidance. A 6.5 Fr steerable sheath was parked in the bronchus per plan, and a flexible MWA probe was inserted coaxially then advanced through the bronchus wall towards the target. Final in-target position was confirmed by CBCT. Only one ablation of 100 W-5 min was performed per target. Animals were euthanized and pathology analysis of the lungs was performed. Results: Eighteen targets with a median largest diameter of 9 mm (interquartile range, 7-11 mm) were ablated in 9 pigs. Median needle-target center distance was 2 mm (interquartile range, 0-4 mm), and was higher for lower/middle than for upper lobes [0 mm (interquartile range, 0-4 mm) vs. 4 mm (interquartile range, 3-8 mm), P=0.04]. No severe complications or pneumothorax occurred. Two cases of rib fractures in the ablation zone resolved after medical treatment. Median longest axis of the ablation zone on post-ablation computed tomography was 38 mm (interquartile range, 30-40 mm). Histology showed coagulation necrosis of ablated tissue. Conclusions: Transbronchial MWA under augmented fluoroscopy guidance using a steerable sheath is feasible and accurate.

Conventional Hepatic Volumetry May Lead to Inaccurate Segmental Yttrium-90 Radiation Dosimetry.

OBJECTIVE: To compare radioembolization treatment zone volumes from mapping cone beam CT (CBCT) versus planning CT/MRI and to model their impact on dosimetry. METHODS: Y90 cases were retrospectively identified in which intra-procedural CBCT angiograms were performed. Segmental and lobar treatment zone volumes were calculated with semi-automated contouring using Couinaud venous anatomy (planning CT/MRI) or tumor angiosome enhancement (CBCT). Differences were compared with a Wilcoxon signed-rank test. Treatment zone-specific differences in segmental volumes by volumetric method were also calculated and used to model differences in delivered dose using medical internal radiation dosimetry (MIRD) at 200 and 120 Gy targets. Anatomic, pathologic, and technical factors likely affecting segmental volumes by volumetric method were evaluated. RESULTS: Forty segmental and 48 lobar CBCT angiograms and corresponding planning CT/MRI scans were included. Median Couinaud- and CBCT-derived segmental volumes were 281 and 243 mL, respectively (p = 0.005). Differences between Couinaud and CBCT lobar volumes (right, left) were not significant (p = 0.24, p = 0.07). Couinaud overestimated segmental volumes in 28 cases by a median of 98 mL (83%) and underestimated in 12 cases by median 69 mL (20%). At a 200 Gy dose target, Couinaud estimates produced median delivered doses of 367 and 160 Gy in these 28 and 12 cases. At a 120 Gy target, Couinaud produced doses of 220 and 96 Gy. Proximal vs. distal microcatheter positioning, variant arterial anatomy, and tumor location on or near segmental watersheds were leading factors linked to volumetric differences. CONCLUSION: Use of CBCT-based volumetry may allow more accurate, personalized dosimetry for segmental Y90 radioembolization.

Simulation of superselective catheterization for cerebrovascular lesions using a virtual injection software.

BACKGROUND: This report addresses the feasibility of virtual injection software based on contrast-enhanced cone-beam CTs (CBCTs) in the context of cerebrovascular lesion embolization. Intracranial arteriovenous malformation (AVM), dural arteriovenous fistula (AVF) and mycotic aneurysm embolization cases with CBCTs performed between 2013 and 2020 were retrospectively reviewed. Cerebrovascular lesions were reviewed by 2 neurointerventionalists using a dedicated virtual injection software (EmboASSIST, GE Healthcare; Chicago, IL). Points of Interest (POIs) surrounding the vascular lesions were first identified. The software then automatically displayed POI-associated vascular traces from vessel roots to selected POIs. Vascular segments and reason for POI identification were recorded. Using 2D multiplanar reconstructions from CBCTs, the accuracy of vascular traces was assessed. Clinical utility metrics were recorded on a 3-point Likert scale from 1 (no benefit) to 3 (very beneficial). RESULTS: Nine cases (7 AVM, 1 AVF, 1 mycotic aneurysm) were reviewed, with 26 POIs selected. Three POIs were in 2nd order segments, 8 POIs in 3rd order segments and 15 POIs in 4th order segments of their respective arteries. The reviewers rated all 26 POI traces - involving a total of 90 vascular segments - as accurate. The average utility score across the 8 questions were 2.7 and 2.8 respectively from each reviewer, acknowledging the software's potential benefit in cerebrovascular embolization procedural planning. CONCLUSION: The operators considered CBCT-based virtual injection software clinically useful and accurate in guiding and planning cerebrovascular lesion embolization in this retrospective review. Future prospective studies in larger cohorts are warranted for validation of this modality.

Retrospective Use of Breathing Motion Compensation Technology (MCT) Enhances Vessel Detection Software Performance.

PURPOSE: Cone beam CT (CBCT) with planning software is used in intra-arterial liver-directed therapies. Software accuracy relies on high CBCT image quality, which can be impaired by breathing motion. We assessed the impact of a specific MCT on software performance for procedure planning and navigation. MATERIALS AND METHODS: Institutional Review Board (IRB)-approved retrospective evaluation of liver-directed therapies from July 2015 to April 2018 was performed. CBCTs with at least one well-defined tumor and noticeable breathing motion were included. Each CBCT was reconstructed with and without breathing MCT (Motion Freeze, GE Healthcare). Automatic tumor-supplying vessel detection was performed on up to 4 tumors in each CBCT reconstruction (Liver ASSIST V.I., GE Healthcare). Vessel detection sensitivity and positive predictive value (PPV) were measured with and without MCT using Digital Subtracted Angiography (DSA) as reference. Preprocedural contrast-enhanced CT was also utilized in some cases to rule out the possibility of extrahepatic supplying vessels. RESULTS: MCT was applied retrospectively to 18 CBCTs with a total of 30 tumors. At least one supplying vessel was detected for 28/30 (93%) tumors with MCT versus 20/30 (66%) without. On the subgroup of 10 CBCTs (22 tumors, 76 feeders) in which the automatic vessel detection initially worked in both reconstructions, the average sensitivity and PPV increased from 63% (48/76) and 57% (48/84) before MCT to 83% (63/76) and 79% (63/80) after (p = 0.002 and p < 0.001). CONCLUSION: Breathing MCT improves planning software performance in CBCT impaired by breathing motion.

Relationship of radiation dose to efficacy of radioembolization of liver metastasis from breast cancer.

PURPOSE: To determine the relationship of tumoral and nontumoral radiation dose to response and toxicity after transarterial radioembolization (TARE) of breast cancer liver metastasis. METHODS: This retrospective study evaluated all patients with breast cancer liver metastases treated with TARE (2/2011-6/2019). Extent of disease was measured as unilobar or bilobar on baseline PET/CT prior to TARE. Response was assessed for targeted regions with modified PERCIST criteria on first follow-up PET/CT. Tumoral and nontumoral liver dosimetry was evaluated by performing volumetric segmentation on post-TARE Bremsstrahlung SPECT/CT. ≥Grade 3 hepatotoxicity was defined as ≥grade 3 bilirubin/AST/ALT elevation or ascites requiring intervention. Fisher's exact tests, Wilcoxon rank sum tests, and Kaplan-Meier survival analysis were performed. RESULTS: Among 64 women, 60 patients had pre- and post-TARE PET/CT, of whom 46/60 (77 %) achieved objective response (OR). Responders received higher tumoral dose with a median (interquartile range) of 167 (96-217) vs. 54 (45-62) Gy (p < 0.001). ≥Grade 3 hepatotoxicity occurred in 8/64 (12.5 %) and was associated with higher pre-treatment bilirubin levels of 0.9 (0.9-1.1) vs. 0.5 (0.4-0.7) mg/dL (p = 0.013). Median overall survival (OS) was 11 (95 % CI 10-19) months. Bilobar disease (Hazard Ratio [HR]: 2.77, 95 % CI 1.11-6.89, p = 0.028) and elevated pre-TARE AST (HR 1.02, 95 % CI 1.01-1.03, p < 0.001) were independently associated with shorter survival. ≥Grade 3 hepatotoxicity was associated with reduced survival (p < 0.001). OR was associated with longer OS of 17 months, compared with 10 months (p = 0.027). CONCLUSION: In TARE for breast cancer liver metastasis, higher tumoral radiation dose (>79.5 Gy) was associated with OR, which was associated with longer survival. Pre-existing liver dysfunction was associated with hepatotoxicity, which was associated with decreased survival.

Utilization of integrated angiography-CT interventional radiology suites at a tertiary cancer center.

BACKGROUND: Integrated Angiography-Computed Tomography (ACT) suites were initially designed in the 1990's to perform complex procedures requiring high-resolution cross-sectional imaging and fluoroscopy. Since then, there have been technology developments and changes in patient management. The purpose of this study was to review the current usage patterns of a single center's integrated ACT suites. METHODS: All procedures performed in 2017 in 3 ACT suites (InterACT Discovery RT, GE Healthcare) at a tertiary cancer center were reviewed retrospectively. Usage was classified as: Standard, in which the patient underwent a single procedure using either fluoroscopy, CT, or ultrasound (US); Combined, in which the patient underwent a single procedure utilizing both fluoroscopy and CT; or Staged, in which the patient underwent 2 separate but successive procedures using fluoroscopy and CT individually. The most frequently performed Combined and Staged procedures were further reviewed to determine how the different modalities were used. The duration of the most common Staged procedures was compared to analogous procedures' durations in single modality rooms over the period Jan 2016 to Sep 2019. RESULTS: A total of 3591 procedures were performed on 2678 patients in the 3 ACT Suites. 80% of patients underwent a Standard procedure using fluoroscopy (38%), CT (32%) or US (10%) and accounted for 70% of the room occupation time. Fourteen and three percent of the patients underwent Combined or Staged procedures, occupying 19 and 5% of the room time, respectively. The remaining procedures were classified as both Combined and Staged, representing 3% of the patients and 6% of the room occupation time. The most common Combined procedures were drainages, hepatic arterial embolizations or radioembolizations, arterial, and biliary interventions. The most common Staged procedures were multiple drainages and hepatic arterial embolizations followed by biopsies or ablations. The room occupation time for liver tumor embolization and ablation was significantly shorter (p < 0.01) when performed in a Staged fashion versus the analogous procedures in single modality room. CONCLUSION: An integrated ACT system provides the capability to perform complex Combined or Staged procedures as well as scheduling flexibility by allowing any type of case to be performed in the IR suite.

Magnetic resonance venography for 3-dimensional live guidance during venous sinus stenting.

PURPOSE: The purpose of this study was to report the technique for intraprocedural guidance of endovascular Venous Sinus Stenting procedures using 3-Dimensional (3D) Magnetic Resonance Venography (MRV) as an overlay on live biplanar fluoroscopy. MATERIALS AND METHODS: Venous sinus stenting procedures performed between April and December, 2017 with 3D MRV fusion for live guidance were reviewed in this study. A thin-slice, contrast-enhanced MR Venogram was used to create 2 3D models - vessels and skull - for procedural guidance via augmented fluoroscopy (Vessel ASSIST, GE Healthcare, Chicago, IL). The skull model was used in the registration of the 3D overlay on both the frontal and lateral planes, which required 1-2 min of procedural time. The vessel model was used to mark landmarks such as the cortical vein ostia and stenosis on the 3D overlay fused with biplanar fluoroscopy. The retrospective imaging review was conducted by 3 neurointerventionalists and relied on a consensus confidence ranking on a 3-point Likert scale from 1- low confidence to 3- high confidence. The neurointerventionalists first reviewed the conventional 2-dimensional pre-stent deployment fluoroscopy images and then reviewed the corresponding images with the 3D MRV overlay. They ranked their confidence in their understanding of cortical venous anatomy for each group. Statistical analysis was performed using a Paired T Test at a 99% confidence interval. RESULTS: Ten cases were included in the retrospective image review. Operator confidence regarding the location of cortical veins was significantly increased using 3D MRV fusion during venous sinus stenting procedures (1.9 vs 2.9, p = .001). CONCLUSION: 3-Dimensional MRV fusion is feasible and helpful in understanding the venous sinus anatomy and location of important cortical veins during venous sinus stenting procedures.

Intraprocedural guidance for recanalization of post-thrombotic venous lesions using live overlay of center lines from pre-operative cross-sectional imaging: a preliminary experience.

PURPOSE: Pre-procedural contrast-enhanced CT and MRI imaging is typically acquired prior to deep venous recanalization procedures for post-thrombotic syndrome. This technical note reports the utility of live-overlay of augmented centerlines extracted from pre-procedural CT and MRI imaging in facilitating fluoroscopic-guided recanalization of post-thrombotic venous lesions. METHODS AND MATERIALS: Six patients with pre-procedural CT or MR venography data were incorporated into a commercially available 3D overlay software (Vessel Assist, GE Healthcare, Buc, France) during venous disease interventions for post-thrombotic venous lesions. Procedures were performed on the GE Discovery IGS 740 fluoroscopy system. After manual determination of the vasculature from preprocedural CT or MR, centerlines were created representing the location and trajectory of the vessels. Steps showcasing the creation of centerlines and their representation during overlay with real-time fluoroscopic guidance in these cases are outlined. Time required to cross the post-thrombotic and occlusive venous segments were reviewed. RESULTS: All iliocaval recanalization procedures were successfully performed utilizing vessel centerline 3D overlay. In one case where occlusion extended to the femoral vein, mis-registration was identified over the femoral anatomy due to a complex leg rotation compared to pre-procedural imaging. No procedural complications related to utilization of software were noted. Average crossing time for occlusions was 3.4 min (range 1.6-5.2). CONCLUSION: 3D overlay with vessel tracking from pre-procedural CT and MRI imaging is technically feasible and assists in catheter navigation for post-thrombotic venous segments. While results from these preliminary experiences support the continued use of this technology, further prospective and comparative evaluation of this technique is warranted to assess for added value in technical success, reductions in procedure time or reductions in radiation exposure.