ABSTRACT
Portal vein thrombosis (PVT) represents a restriction or occlusion of the portal vein by a blood clot, which can appear in liver cirrhosis, inherited or acquired thrombophilia, malignancies, abdominal infection, abdominal inflammation, and injury to the portal vein; it can evolve to local venous extension, recanalization, or portal cavernoma (PC). This research represents an observational study of patients admitted with a diagnosis of PVT between January 2018 and December 2022. We assessed the rate of and risk factors for PC. In total, 189 patients with PVT were included; the rate of PC was 14.8%. In univariate and multivariate analysis, the main risk factors for the presence of PC were etiology (thrombophilia, myeloproliferative disorders, local inflammatory diseases, and idiopathic causes), prior PVT, and complete versus incomplete or single-branch portal obstruction. In patients with superior mesenteric vein (SMV) thrombosis, distal obstruction was more prone to PC than proximal obstruction. The main predictive factors were etiology, prior PVT, complete PVT obstruction, and no prior non-selective beta-blocker (NSBB) use; in patients with SMV thrombosis, the distal extension was more significantly associated with the risk of PC. We propose a composite score for the prediction of PC which includes etiology, prior diagnosis of PVT, prior NSBB use, complete versus incomplete PVT, and distal versus proximal SMV thrombosis, with good accuracy (AUC 0.822) and an estimated sensitivity of 76.92% and specificity of 82.39% at a cut-off value of 4.
ABSTRACT
Contrast-enhanced ultrasound (CEUS) is widely used in the characterization of liver tumors; however, the evaluation of perfusion patterns using CEUS has a subjective character. This study aims to evaluate the accuracy of an automated method based on CEUS for classifying liver lesions and to compare its performance with that of two experienced clinicians. The system used for automatic classification is based on artificial intelligence (AI) algorithms. For an interpretation close to the clinical setting, both clinicians knew which patients were at high risk for hepatocellular carcinoma (HCC), but only one was aware of all the clinical data. In total, 49 patients with 59 liver tumors were included. For the benign and malignant classification, the AI model outperformed both clinicians in terms of specificity (100% vs. 93.33%); still, the sensitivity was lower (74% vs. 93.18% vs. 90.91%). In the second stage of multiclass diagnosis, the automatic model achieved a diagnostic accuracy of 69.93% for HCC and 89.15% for liver metastases. Readers demonstrated greater diagnostic accuracy for HCC (83.05% and 79.66%) and liver metastases (94.92% and 96.61%) compared to the AI system; however, both were experienced sonographers. The AI model could potentially assist and guide less-experienced clinicians to discriminate malignant from benign liver tumors with high accuracy and specificity.
ABSTRACT
Recent advances in the field of ultrasonography offer promising tools for the evaluation of liver tumors. We aim to assess the value of multimodal ultrasound in differentiating hepatocellular carcinomas (HCCs) from other liver lesions. We prospectively included 66 patients with 72 liver tumors. The histological analysis was the reference standard for the diagnosis of malignant liver lesions, and partially for benign tumors. All liver lesions were assessed by multiparametric ultrasound: standard ultrasound, contrast-enhanced ultrasound (CEUS), the point shear wave elastography (pSWE) using shear wave measurement (SWM) method and real-time tissue elastography (RTE). To diagnose HCCs, CEUS achieved a sensitivity, specificity, accuracy and positive predictive value (PPV) of 69.05%, 92.86%, 78.57% and 93.55%, respectively. The mean shear-wave velocity (Vs) value in HCCs was 1.59 ± 0.29 m/s, which was lower than non-HCC malignancies (p < 0.05). Using a cut-off value of 1.58 m/s, SWM achieved a sensitivity of 54.76%, and 82.35% specificity, for differentiating HCCs from other malignant lesions. The combination of SWM and CEUS showed higher sensitivity (79.55%) compared with each technique alone, while maintaining a high specificity (89.29%). In RTE, most HCCs (61.53%) had a mosaic pattern with dominant blue areas corresponding to type "c" elasticity. Elasticity type "c" was 70.59% predictive for HCCs. In conclusion, combining B-mode ultrasound, CEUS, pSWE and RTE can provide complementary diagnostic information and potentially decrease the requirements for other imaging modalities.
