A narrative review of preoperative CT for predicting spread through air spaces of lung cancer.

Background And Objective: Spread through air space (STAS) is a recognized mechanism of lung cancer invasion and is associated with patient prognosis. However, current diagnostic accuracy of bronchial cytology and intraoperative frozen section for STAS remains insufficient to meet clinical needs. Therefore, accurate and non-invasive preoperative prediction of STAS is critical for clinical decision-making. In this paper, we review and summarize recent studies on the role of computed tomography (CT) in predicting STAS in lung cancer, and discuss the limitations and future directions of related research in this field.

Methods: Relevant studies were identified through searches on the Web of Science, PubMed, Cochrane Library, and EMBASE. We included English-language articles published between July 2017 and June 2024, focusing on research related to STAS and CT.

Key Content And Findings: This review aimed to assess the viability of preoperative CT imaging for predicting STAS. Current studies suggest that traditional imaging signs enable the assessment of STAS, and with the development of artificial intelligence, the efficacy of STAS prediction models has been greatly enhanced by radiomics and deep learning methods. However, risk stratification studies remain limited and require further refinement through more comprehensive pathological definitions of STAS.

Conclusions: Preoperative CT imaging images demonstrated good predictive efficacy of STAS. However, further progress requires a more comprehensive definition of STAS and validation through large-sample, prospective, and multi-center studies to enhance its clinical applicability.