Abstract
The aim of this study was to show the efficacy described in the scientific literature of lung ultrasound (LU) during video-assisted thoracic surgery (VATS) to determine the location and characterization of pulmonary nodules. The results showed that intraoperative LU is especially useful in localizing lung lesions by VATS with a sensitivity close to 100%. It was also shown to be useful in planning very precise lung resections, reducing the resection of healthy lung tissue. Although general criteria could be established for the degree of benignity/malignancy of lung lesions based on the ultrasound patterns obtained, the great variability observed in these patterns does not offer sufficient guarantees to make a reliable diagnosis. In this sense, the application of Doppler ultrasound or the utilization of Intraoperative Contrast-Enhanced Ultrasound (Io-CEUS) in a completely collapsed lung can be advantageous for the accurate localization and characterization of pulmonary lesions. Even so, at present, definitive confirmation of the nature of lung lesions usually requires biopsy and histopathological study. The development of artificial intelligence algorithms that integrate the results of histological analyses with various types of ultrasonographic patterns (based on pulmonary aeration obtained, as well as Doppler and Io-CEUS records) will likely represent the future of ultrasonographic differentiation of these lesions.