Marco Calandri

For interventional radiologists working in the thoracic field, 2025 has been marked by mixed emotions: the familiar excitement of innovation—an intrinsic hallmark of interventional radiology—alongside emerging questions that challenge established practices and compel critical reflection. Among the many publications released over the past year, two studies stand out. Although they address different topics, together they prompt reflection on the future direction of thoracic interventional radiology and, specifically, on its most widespread and foundational procedure: lung biopsy.

Background: Splenic artery embolization (SAE) has emerged as a key adjunct to non-operative management (NOM) in hemodynamically stable patients with blunt splenic trauma, yet variability in its application persists across centers. Objectives: The aim was to evaluate real-life clinical practices, techniques, and outcomes of SAE in blunt splenic trauma across multiple Italian trauma centers. Materials and Methods: This retrospective multicenter study analyzed data from 281 patients undergoing emergency SAE for blunt splenic trauma between January 2019 and December 2021. Demographics, imaging findings, embolization techniques, complications, and outcomes were collected and analyzed. Multivariate logistic regression was used to assess predictors of splenectomy. Results: The technical success rate was 100%, with a 9.6% rate of post-embolization splenectomy and a complication rate of 24.9% (including 5.7% splenic infarction and 3.2% rebleeding). Embolization was performed proximally (46.6%), distally (28.8%), or with a combined approach (24.6%). No significant correlation was found between embolization technique and splenectomy rate. Patients with AAST grade III injuries benefited from SAE with high technical success and low failure rates. Notably, 14.2% of patients underwent angiography despite negative CT, with a splenectomy rate of 10% in this subgroup. Multivariate analysis identified no independent predictors of splenectomy. Conclusions: SAE is a reliable and effective tool in the management of blunt splenic trauma, achieving high splenic salvage rates even in selected grade III injuries and CT-negative patients. In an era of precision medicine, interventional radiology should be regarded as a distinct and specific treatment modality, comparable to surgery, rather than being merely included within non-operative management (NOM).

Intraprocedural use of ablation confirmation software is critical and allows for quantitative measurement of the minimal ablative margin after percutaneous thermal ablation of liver tumors and immediate re-ablation of insufficient margins.

Objectives: Evaluating the pathological response and the survival outcomes of combined thermal ablation (TA) and transarterial chemoembolization (TACE) as a bridge or downstaging for liver transplantation (LT) in patients with hepatocellular carcinoma (HCC) > 3 cm. Materials and methods: A retrospective review encompassed 36 consecutive patients who underwent combined TA-TACE as bridging or downstaging before LT. Primary objectives included necrosis of the target lesion at explant pathology, post-LT overall survival (OS) and post-LT recurrence-free survival (RFS). For OS and RFS, a comparison with 170 patients subjected to TA alone for nodules 3 cm, particularly for bridging and downstaging to LT, achieving OS and RFS rates significantly exceeding 80% at 1, 3 and 5 years.

The purpose of this study was to evaluate the efficacy of unenhanced cone-beam computed tomography (CBCT) performed at the end of drug-eluting bead transarterial chemoembolization (DEB-TACE) in predicting HCC nodules’ early radiologic response to treatment, assessed using mRECIST criteria with a 30–60 day four-phase contrast-enhanced CT follow-up. Fifty-nine patients (81 lesions) subjected to DEB-TACE as exclusive treatment for HCC lesions (naive/relapse) between February 2020 and October 2021 were prospectively enrolled. In a post-interventional unenhanced CBCT procedure, two experienced radiologists evaluated for each lesion the overall intensity of the contrast media deposit, the homogeneity of the enhancement, and the presence of smooth and complete margins. The univariate analysis found that lesions with complete response (CR+) had a significantly higher incidence of clear and complete margins than CR− lesions (76.9% vs. 17.2%, p = 0.003) and a higher intensity score (67.3% vs. 27.6%, p = 0.0009). A Dmax p = 0.01). These features were confirmed as significant predictors for CR+ by multivariate binary logistic regression. The homogeneity of the enhancement did not affect the DEB-TACE outcome. Post-interventional unenhanced CBCT is effective in predicting early radiological response to DEB-TACE, since the presence of an intense contrast media deposit with clear and complete margins in treated HCC lesions is associated with CR.

The aim of our study is the development of an automatic tool for the prioritization of COVID-19 diagnostic workflow in the emergency department by analyzing chest X-rays (CXRs). The Convolutional Neural Network (CNN)-based method we propose has been tested retrospectively on a single-center set of 542 CXRs evaluated by experienced radiologists. The SARS-CoV-2 positive dataset (n = 234) consists of CXRs collected between March and April 2020, with the COVID-19 infection being confirmed by an RT-PCR test within 24 h. The SARS-CoV-2 negative dataset (n = 308) includes CXRs from 2019, therefore prior to the pandemic. For each image, the CNN computes COVID-19 risk indicators, identifying COVID-19 cases and prioritizing the urgent ones. After installing the software into the hospital RIS, a preliminary comparison between local daily COVID-19 cases and predicted risk indicators for 2918 CXRs in the same period was performed. Significant improvements were obtained for both prioritization and identification using the proposed method. Mean Average Precision (MAP) increased (p < 1.21 × 10⁻²¹ from 43.79% with random sorting to 71.75% with our method. CNN sensitivity was 78.23%, higher than radiologists’ 61.1%; specificity was 64.20%. In the real-life setting, this method had a correlation of 0.873. The proposed CNN-based system effectively prioritizes CXRs according to COVID-19 risk in an experimental setting; preliminary real-life results revealed high concordance with local pandemic incidence.