The aim of the present study was to describe patient dose indicator levels during intra-arterial catheter (IAC) implantation for liver chemotherapy, and to determine factors affecting the dose indicators. Between January 2017 and January 2019, 61 patients with hepatic metastases from colorectal cancer were retrospectively included. Interventions were carried out in a standardised manner by three experienced radiologists on the same angiographic table without changes in protocol parameters. For each patient, clinical, radiological and dosimetry data were collected, including the air kerma area product (KAP), part of KAP due to the fluoroscopy and fluoroscopy time (FT), total kerma at the reference interventional point and peak skin dose (PSD). Local dose reference levels (RLs) were determined as the third quartile of the patient dose distributions. Univariate and multivariate analysis of factors affecting dose indicators was performed. The mean KAP was 111 Gy cm2, the mean reference point air kerma (Ka,r) was 648 mGy, the mean PSD was 613 mGy, and the mean FT was 3190 s (62% of the KAP). The mean cone beam computed tomography dose was 37.3 ± 11.8 Gy cm2, which accounted for 37% of the KAP. The RL could be proposed taking into account the third quartiles (KAP = 164.6 Gy cm2, Ka,r = 904.5 mGy, FT = 4011 s and standard deviation = 772.7 mGy). The factors affecting dose indicators were related to the patients (sex, cardiovascular risk factors, weight, body mass index), to the vascular anatomies (coeliac trunk angulation) and to the procedures (number of embolised arteries). This study allowed a better understanding of dose indicators and factors affecting these indicators during the implantation of IACs for hepatic chemotherapy, which is a long and difficult procedure. Local dose RLs were determined. Multicentre, multi-equipment studies are necessary.

Purpose To evaluate safetey and efficacy of an ultra low dose (ULD) protocol implementation during lung radiofrequency ablations. Methods All patients who underwent a lung RFA were consecutively and retrospectively included between November 2017 and January 2021. 30 patiens were treated with a historical standard protocol (SP), 31 with an ULD protocol. Patient characteristics, dose indicators, technical efficacy (minimal margin MM, recurrence during the follow up) and complications (pneumothorax, alveolar hemorrhage, hemoptysis) were recorded. Results 61 patients were included (median age, 65 [54. 73], 33 women). There was no significant difference in the characteristics of the patients according to the type of protocol, except for the type of anesthesia. Even if the number of helical acquisitions had not significantly changed, all the dose indicators significantly decreased by 1.5-fold to 3-fold. The median dose-length-product and effective dose were respectively 465mGy.cm [315; 554] and 6.5mSv [4.4; 7.8] in the SP group versus 178 mGy.cm [154; 267] and 2.5mSv [2.2; 3.7] in the ULD group, (p<.001). The ULD group presented a lower intra-operator variability and a better inter-operator alignment. The MM were not significantly different in the 2 groups (4.6mm versus 5mm, p=.16). One local recurrence was observed in each group, at 8 months for the SP and at one year for the ULD group (p=1). No significant difference was found in the complications rates. Conclusion Implmenting an ultra-low dose protocol during lung RFA may allow with the same efficacy, a reduction of dose indicators, intra- and inter-operator alignments, without increasing complications rates.