Advanced Robotic Angiography Systems for Image Guidance During Conventional Transarterial Chemoembolization: Impact on Radiation Dose and Image Quality

Image Quality and Radiation Exposure in Abdominal Angiography: A Head-to-Head Comparison of Conventional Detector-Dose-Driven Versus Contrast-to-Noise Ratio-Driven Exposure Control at Various Source-to-Image Receptor Distances and Collimations in a Pilot Phantom and Animal Study

OBJECTIVES

This phantom and animal pilot study aimed to compare image quality and radiation exposure between detector-dose-driven exposure control (DEC) and contrast-to-noise ratio (CNR)-driven exposure control (CEC) as functions of source-to-image receptor distance (SID) and collimation.

MATERIALS AND METHODS

First, an iron foil simulated a guide wire in a stack of polymethyl methacrylate and aluminum plates representing patient thicknesses of 15, 25, and 35 cm. Fluoroscopic images were acquired using 5 SIDs ranging from 100 to 130 cm and 2 collimations (full field of view, collimated field of view: 6 × 6 cm). The iron foil CNRs were calculated, and radiation doses in terms of air kerma rate were obtained and assessed using a multivariate regression. Second, 5 angiographic scenarios were created in 2 anesthetized pigs. Fluoroscopic images were acquired at 2 SIDs (110 and 130 cm) and both collimations. Two blinded experienced readers compared image quality to the reference image using full field of view at an SID of 110 cm. Air kerma rate was obtained and compared using t tests.

RESULTS

Using DEC, both CNR and air kerma rate increased significantly at longer SID and collimation below the air kerma rate limit. When using CEC, CNR was significantly less dependent of SID, collimation, and patient thickness. Air kerma rate decreased at longer SID and tighter collimation. After reaching the air kerma rate limit, CEC behaved similarly to DEC. In the animal study using DEC, image quality and air kerma rate increased with longer SID and collimation ( P < 0.005). Using CEC, image quality was not significantly different than using longer SID or tighter collimation. Air kerma rate was not significantly different at longer SID but lower using collimation ( P = 0.012).

CONCLUSIONS

CEC maintains the image quality with varying SID and collimation stricter than DEC, does not increase the air kerma rate at longer SID and reduces it with tighter collimation. After reaching the air kerma rate limit, CEC and DEC perform similarly.

Radiation dose during transarterial chemoembolization and associated factors

PURPOSE

To provide detailed reports on radiation doses during transarterial chemoembolization (TACE) in the cone-beam computed tomography (CBCT) era and to identify the associated factors.

METHODS

This retrospective study included 385 consecutive patients who underwent initial conventional TACE for hepatocellular carcinoma (HCC) between January 2016 and December 2017. In most cases, CBCT was performed at the common hepatic artery or celiac axis to confirm the location of the tumor and the three-dimensional hepatic artery anatomy. Superselective TACE was performed for all technically feasible cases. Information on total dose area product (DAP), total cumulative air kerma (CAK), fluoroscopy time, and DAP and CAK of each digital subtraction angiography (DSA) and CBCT scan was recorded. Multiple linear regression analysis was performed to identify the factors associated with increased DAP during TACE.

RESULTS

The mean values of total DAP and CAK were 165.2 ± 81.2 (Gy·cm²) and 837.1 ± 571.0 (mGy), respectively. The mean fluoroscopy time was 19.1 ± 10.3 min. The mean DAP caused by fluoroscopy, DSA, and CBCT was 51.8 ± 43.9, 28.0 ± 24.1, and 83.9 ± 42.1 Gy·cm², respectively. Male sex, a high body mass index, largest tumor size > 3 cm, presence of aberrant right and left hepatic arteries, and superselective TACE were identified as independent predictors of increased total DAP during TACE.

CONCLUSION

We were able to provide detailed reports on radiation doses during TACE and associated factors.

Intraoperative Cone-Beam Computed Tomography Assessment of Spinal Pedicle Screws Placement Precision Is in Full Agreement with Postoperative Computed Tomography Assessment

OBJECTIVE

To assess agreement between pedicle screw placement evaluated on postoperative computed tomography (CT) and on intraoperative cone-beam CT (CBCT) and compare procedure characteristics when using first-generation and second-generation robotic C-arm systems in the hybrid operating room.

METHODS

All patients who received pedicle screws for spinal fusion at our institution between June 2009 and September 2019 and underwent intraoperative CBCT and postoperative CT were included. The CBCT and CT images were reviewed by 2 surgeons to assess the screw placement using the Gertzbein-Robbins and the Heary classifications. Intermethod agreement of screw placement classifications as well as interrater agreement were assessed using Brennan-Prediger and Gwet agreement coefficients. Procedure characteristics using first-generation and second-generation generation robotic C-arm systems were compared.

RESULTS

Fifty-seven patients were treated with 315 pedicle screws at thoracic, lumbar, and sacral levels. No screw had to be repositioned. On CBCT, accurate placement was found for 309 screws (98.1%) using the Gertzbein-Robbins classification and 289 (91.7%) using the Heary classification and on CT, these were 307 (97.4%) and 293 (93.0%), respectively. Intermethod between CBCT and CT and interrater agreements between the 2 raters were almost perfect (>0.90) for all assessment. There were no significant differences in mean radiation dose (P = 0.83) and fluoroscopy time (P = 0.82), but length of surgery using the second-generation system was estimated at 107.7 minutes (95% confidence interval, 31.9-183.5 minutes; P = 0.006) shorter.

CONCLUSIONS

Intraoperative CBCT provides accurate assessment of pedicle screw placement and enables intraoperative repositioning of misplaced screws.

Radiation Exposure Reduction by Digital Variance Angiography in Lower Limb Angiography: A Randomized Controlled Trial

BACKGROUND

digital variance angiography (DVA) provides higher image quality than digital subtraction angiography (DSA). This study investigates whether the quality reserve of DVA allows for radiation dose reduction during lower limb angiography (LLA), and compares the performance of two DVA algorithms.

METHODS

this prospective block-randomized controlled study enrolled 114 peripheral arterial disease patients undergoing LLA into normal dose (ND, 1.2 µGy/frame, n = 57) or low-dose (LD, 0.36 µGy/frame, n = 57) groups. DSA images were generated in both groups, DVA1 and DVA2 images were generated in the LD group. Total and DSA-related radiation dose area product (DAP) were analyzed. Image quality was assessed on a 5-grade Likert scale by six readers.

RESULTS

the total and DSA-related DAP were reduced by 38% and 61% in the LD group. The overall visual evaluation scores (median (IQR)) of LD-DSA (3.50 (1.17)) were significantly lower than the ND-DSA scores (3.83 (1.00), p < 0.001). There was no difference between ND-DSA and LD-DVA1 (3.83 (1.17)), but the LD-DVA2 scores were significantly higher (4.00 (0.83), p < 0.01). The difference between LD-DVA2 and LD-DVA1 was also significant (p < 0.001).

CONCLUSIONS

DVA significantly reduced the total and DSA-related radiation dose in LLA, without affecting the image quality. LD-DVA2 images outperformed LD-DVA1, therefore DVA2 might be especially beneficial in lower limb interventions.

Value of Latest-generation Cone-beam Computed Tomography for Post Lipiodol-embolization Imaging in Hepatic Transarterial Chemoembolization in Comparison with Multi-detector Computed Tomography

RATIONALE AND OBJECTIVES

To evaluate image quality, radiation dose (phantom study) and tumor volumetry of intraprocedural cone-beam computed tomography (CBCT) compared to postprocedural multidetector computed tomography (MDCT) in patients undergoing hepatic conventional transarterial chemoembolization (cTACE).

MATERIALS AND METHODS

One hundred fourteen patients (64/50 female/male; mean age, 57 ± 14 years) who had undergone cTACE including intraprocedural-CBCT and postprocedural-MDCT were retrospectively enrolled. Subjective image quality (IQ) and suitability for assessing Lipiodol distribution were compared using 4-point Likert scales; additionally, lesion to liver contrast (LLC) and contrast-to-noise-ratio (CNR) were compared. Tumor volumes were measured semi-automatically and compared to magnetic resonance imaging (MRI). Effective doses were measured using an anthropomorphic phantom.

RESULTS

The suitability of CBCT for assessing Lipiodol distribution during cTACE was comparable to MDCT (mean score, 3.2 ± 0.6) and CBCT (3.4 ± 1.0, p = 0.29). Subjective overall IQ was rated with a mean score of 3.2 ± 0.7 (κ = 0.66) in CBCT and 3.1 ± 0.4 (κ = 0.57, p = 0.15) in MDCT. Evaluation of LLC showed significant differences between CBCT and MDCT (mean scores 3.6 ± 1.2 and 2.6 ± 1.5, respectively). CNR analysis demonstrated comparable mean values for CBCT and MDCT (3.5 ± 1.3 vs. 3.4 ± 1.8, p = 0.31). No significant differences were found regarding tumor volumetry (mean volumes: CBCT, 27.0 ± 17.4 mm³; MDCT: 26.8 ± 16.0 mm³; p = 0.66) in comparison to T2-weighted MRI (25.9 ± 17.6 mm³). Effective doses were 3.2 ± 0.6 mSv (CBCT) and 2.5 ± 0.3 mSv (MDCT) (p < 0.001). No cTACE-related complications (bleeding, non-target embolization) were missed on intraprocedural CBCT in comparison to postprocedural MDCT.

CONCLUSION

Latest-generation intraprocedural CBCT provides suitable assessment of Lipiodol distribution and similar image quality compared to MDCT while allowing for robust volumetric tumor measurements and immediate complication control by visualizing non-target embolization and hematoma. Therefore, it may improve patient safety and outcome as well as clinical workflow compared to postprocedural MDCT in hepatic cTACE in certain cases.

Tantalum-specific contrast-to-noise ratio or conventional detector dose-driven exposure control in angiography: radiation dose and image quality evaluation in a porcine model

Background

The aim of this animal study was to compare the fluoroscopic image quality (IQ) and radiation dose between a tantalum (Ta)-specific contrast-to-noise ratio-driven exposure control (Ta-CEC) and a detector dose-driven exposure control (DEC) in abdominal angiography.

Methods

Nine angiography scenarios were created in seven anaesthetised pigs using Ta-based embolisation material during percutaneous liver and kidney intervention. Fluoroscopic images were acquired using three DEC protocols with different dose levels and Ta-CEC protocols with different IQ levels, sampled in small steps. Polymethyl-methacrylate and aluminium plates were used to simulate attenuation of three water equivalent thicknesses (WET). Three blinded readers evaluated the IQ of DEC and dose equivalent Ta images and selected the Ta-IQ equivalent image corresponding to the DEC image.

Results

Interobserver agreement for the IQ assessment was 0.43 for DEC, 0.56 for Ta-CEC and for the assessment of incident air kerma at the interventional reference point (K_a,r) for the Ta-IQ equivalent image 0.73. The average IQ of the dose equivalent Ta images was superior compared to the DEC images (p < 0.001) and also for every WET (26, 31, or 36 cm) and dose level (p ≤ 0.022). The average K_a,r for the Ta-IQ equivalent images was 59 ± 16% (mean ± standard deviation) lower compared to the DEC images (p < 0.001).

Conclusions

Compared to DEC, Ta-CEC significantly improved the fluoroscopic depiction of Ta, while maintaining the K_a,r. Alternatively, the K_a,r can be significantly reduced by using Ta-CEC instead of DEC, while maintaining equivalent IQ.

UTERINE ARTERY EMBOLISATION: CONTINUOUS QUALITY IMPROVEMENT REDUCES RADIATION DOSE WHILE MAINTAINING IMAGE QUALITY

The purpose of this study was to introduce a continuous quality improvement (CQI) program for radiation dose optimisation during uterine artery embolisation (UAE) and assess its impact on dose reduction and image quality. The CQI program investigated the effects of optimising radiation dose parameters on the kerma-area product (KAP) and image quality when comparing a 'CQI intervention' group (n = 50) and 'Control' group (n = 50). Visual grading characteristics (VGC) analysis was used to assess image quality, using the 'Control' group as a reference. A significant reduction in KAP by 17% (P = 0.041, d = 0.2) and reference air kerma (Ka, r) by 20% (P = 0.027, d = 0.2) was shown between the two groups. The VGC analysis resulted in an area under the VGC curve (AUCVGC) of 0.54, indicating no significant difference in image quality between the two groups (P = 0.670). The implementation of the CQI program and optimisation of radiation dose parameters improved the UAE radiation dose practices at our centre. The dose reduction demonstrated no detrimental effects on image quality.

Digital variance angiography allows about 70% decrease of DSA-related radiation exposure in lower limb X-ray angiography

Our aim was to investigate whether the previously observed higher contrast-to-noise ratio (CNR) and better image quality of Digital Variance Angiography (DVA) - compared to Digital Subtraction Angiography (DSA) - can be used to reduce radiation exposure in lower limb X-ray angiography. This prospective study enrolled 30 peripheral artery disease patients (mean ± SD age 70 ± 8 years) undergoing diagnostic angiography. In all patients, both normal (1.2 µGy/frame; 100%) and low-dose (0.36 µGy/frame; 30%) protocols were used for the acquisition of images in three anatomical regions (abdominal, femoral, crural). The CNR of DSA and DVA images were calculated, and the visual quality was evaluated by seven specialists using a 5-grade Likert scale. For investigating non-inferiority, the difference of low-dose DVA and normal dose DSA scores (DVA30-DSA100) was analyzed. DVA produced two- to three-fold CNR and significantly higher visual score than DSA. DVA30 proved to be superior to DSA100 in the crural region (difference 0.25 ± 0.07, p < 0.001), and there was no significant difference in the femoral (− 0.08 ± 0.06, p = 0.435) and abdominal (− 0.10 ± 0.09, p = 0.350) regions. Our data show that DVA allows about 70% reduction of DSA-related radiation exposure in lower limb X-ray angiography, providing a potential new radiation protection tool for the patients and the medical staff.

Impact of a contrast-to-noise ratio driven and material specific exposure control on image quality and radiation exposure in angiography

Conventional detector-dose driven exposure controls (DEC) do not consider the contrasting material of interest in angiography. Considering the latter when choosing the acquisition parameters should allow for optimization of x-ray quality and consecutively lead to a substantial reduction of radiation exposure. Therefore, the impact of a material-specific, contrast-to-noise ratio (CNR) driven exposure control (CEC) compared to DEC on radiation exposure was investigated. A 3D-printed phantom containing iron, tantalum, and platinum foils and cavities, filled with iodine, barium, and gas (carbon dioxide), was developed to measure the CNR. This phantom was placed within a stack of polymethylmethacrylate and aluminum plates simulating a patient equivalent thickness (PET) of 2.5-40 cm. Fluoroscopy and digital radiography (DR) were conducted applying either CEC or three, regular DEC protocols with parameter settings used in abdominal interventions. CEC protocols where chosen to achieve material-specific CNR values similar to those of DEC. Incident air kerma at the reference point(K_a,r), using either CEC or DEC, was assessed and possible K_a,r reduction for similar CNR was estimated. We show that CEC provided similar CNR as DEC at the same or lower K_a,r. When imaging barium, iron, and iodine K_a,r was substantially reduced below a PET of 20 cm and between 25 cm and 30 cm for fluoroscopy and Dr When imaging platinum and tantalum using fluoroscopy and DR and gas using DR, the K_a,r reduction was substantially higher. We estimate the K_a,r reduction for these materials between 15% and 84% for fluoroscopy and DR between 15% and 93% depending on the PET. The results of this study demonstrate a high potential for skin dose reduction in abdominal radiology when using a material-specific CEC compared to DEC. This effect is substantial in imaging materials with higher energy K-edges, which is beneficial, for example, in long-lasting embolization procedures with tantalum-based embolization material in young patients with arterio-venous malformations.