In this study, the performance of air-containing space and its specific patterns in differentiating neoplastic GGNs from non-neoplastic ones was compared based on a larger sample. It was found that the air-containing space, air bronchogram, and BLL signs were significantly more common in neoplastic GGNs than in non-neoplastic ones. Therefore, each of these signs can be considered as an indicator of neoplastic GGNs. Among these signs, the BLL sign demonstrated the highest specificity (93.6%) among the common morphological features. Interestingly, in neoplastic GGNs, the occurrence of BLL was not closely correlated with tumor size, attenuation on CT images, or the degree of invasiveness. Overall, any pattern of air-containing space is valuable in differentiating GGNs, and the BLL sign can be considered as a specific and independent indicator for neoplastic GGNs. These findings provide precise information for the diagnosis and management of GGNs.

The air bronchogram sign was frequently observed in both neoplastic and non-neoplastic GGNs, probably reflecting the preservation of the underlying pulmonary architecture. The difference in the prevalence is probably because most non-neoplastic GGNs are infiltrative and therefore opacify the pulmonary airspaces more easily; however, in neoplastic GGNs, cellular infiltrates were less common [18]. Furthermore, there was no significant difference in the incidence of natural or dilated and distorted bronchi between neoplastic and non-neoplastic GGNs [23]. The present study confirmed these previous findings, suggesting that the presence of the air bronchogram sign and changes in the internal bronchus cannot be considered specific predictors of neoplastic GGNs.

In contrast, the presence of the BLL sign was predominantly observed in neoplastic GGNs, which can be attributed to the lepidic tumor growth pattern that leads to incomplete filling of the normal parenchyma, dilated bronchium, and enlarged alveolar spaces [22]. When compared to the air bronchogram sign, although the BLL sign was less common in neoplastic GGNs, it may be a more specific indicator, supporting previous speculation based on smaller sample size [18]. However, it is worth noting that the BLL sign was not exclusively detected in tumors. Some non-neoplastic GGNs also exhibited the BLL sign, but more than a half cases were of the marginal type, which is rarely observed in neoplastic GGNs. Additionally, almost all non-neoplastic GGNs with the BLL sign did not show concomitant air bronchogram. Therefore, the presence of a marginal BLL sign may indicate a non-neoplastic GGN, while the coexistence of the BLL and air bronchogram signs strongly suggests a tumor. These findings were not reported previously and should be verified in the future studies.

In this study, the differences in the prevalence of air bronchogram and BLL sign in neoplastic and non-neoplastic GGNs are consistent with previous research [18, 19, 23, 26]. However, there are some studies that have reported different findings. For example, one study that included 94 non-neoplastic GGNs and 1,840 neoplastic GGNs found that BLL and air bronchogram signs were more common in neoplastic GGNs. However, after adjusting for baseline characteristics, only the BLL sign was found to be more common in matched neoplastic PSNs [19]. Another study involving 33 non-neoplastic GGNs and 47 neoplastic GGNs found that only the air bronchogram sign was more common in neoplastic pGGNs. The incidence rates of the BLL sign in non-neoplastic and neoplastic GGNs were similar, regardless of whether they were pure or mixed [11]. Furthermore, both the BLL and air bronchogram signs were rarely detected in transient pure GGNs [33]. These differences in findings may be attributed to variations in study methods or relatively small sample sizes.

Among neoplastic GGNs, the BLL sign was more frequently observed in large, part-solid, and highly aggressive tumors. However, its occurrence did not correlate with tumor size, CT attenuation, or invasiveness. Similar findings have been reported in previous studies, where the occurrence rate of the BLL sign in invasive adenocarcinoma and pre-invasive lesions showed no significant difference, regardless of whether they were pGGNs or PSNs [34, 35]. Therefore, the diagnostic value of the BLL sign in distinguishing between pre-invasive and invasive GGNs is limited [15, 36]. It can be concluded that the occurrence of the BLL sign in tumors is highly random, and this sign serves as an independent indicator with significant value only in differentiating GGNs.

Among neoplastic GGNs, the BLL sign was more commonly detected in the large, part-solid, and highly aggressive ones. However, its occurrence did not correlate with the tumor size, attenuation on CT images, or degree of invasiveness. Similar findings have been reported in previous studies, where the occurrence rate of the BLL sign in invasive adenocarcinoma and pre-invasive lesions showed no significant difference, regardless of whether they were pGGNs or PSNs [34, 35]. Therefore, the diagnostic value of the BLL sign in distinguishing between pre-invasive and invasive GGNs is limited [15, 36]. It can be concluded that the occurrence of the BLL sign in tumors is highly random, and this sign serves as an independent indicator with significant value only in differentiating GGNs.

Though the air bronchogram and BLL signs are valuable indicators for distinguishing GGNs, they are not commonly observed, especially the BLL sign. Therefore, it is important to consider other morphological characteristics in the diagnosis of GGNs. A study has suggested that factors such as nodule size, CT attenuation, lesion border, and margin type in solitary GGNs can be useful in differentiating lung cancer from benign lesions. It has been confirmed that the presence of a well-defined border and a higher average CT value are associated with malignancy in pGGNs, and the presence of a larger size, well-defined border, and spiculated margin can aid in the differential diagnosis of malignant PSNs [19]. However, in this study, any type of air-containing space, particularly the BLL sign, showed significantly higher specificity compared to other morphological features in diagnosing neoplastic GGNs. Regarding the smaller GGNs without the aforementioned indicators, the air containing space, BLL in particular, could be considered an important clue for discriminating them.

Note that the “BLL sign” rather than the “vacuole sign” was used for describing localized low-density area in GGNs in this study because the latter is typically defined as round or irregular air attenuation with a diameter of 1–2 mm in a nodule [22]. However, in the cases included in this study, the diameter of the low-density areas reached up to 15 mm. Despite the size difference, we believe that the nature of the low-density areas in GGNs is similar, regardless of their size. Furthermore, it is important to exercise caution when interpreting the presence of the BLL sign in patients with emphysema. In these cases, the presence of the BLL sign may be attributed to pre-existing internal bullae surrounded by the lesions, leading to a potentially “false” BLL sign. Therefore, when distinguishing GGNs, particularly in patients with emphysema, the presence of the BLL sign should be interpreted with caution.

GGNs are a common type of pulmonary disease that can exhibit various characteristics and frequently require differential diagnosis. However, compared to solid nodules, GGNs have fewer CT features such as lobulation, spiculation, and pleural indentation, which makes their differential diagnosis more challenging. Therefore, there is a need for more specific CT features to improve differentiation. Previous and the present studies have shown that air containing space, including air bronchogram and BLL, is more frequently observed in neoplastic lesions. However, there has been no research on their specificity in diagnosis, leading to the neglect of their value in differential diagnosis. In the present study, based on a large sample, it was found that although the sensitivity of BLL was not high, its specificity reached 93.6%. This finding confirms its importance in differential diagnosis, especially for the inexperienced radiologists, as it improves their diagnostic accuracy of GGNs to some certain extent. The future trend in imaging diagnosis is precise diagnosis, which requires the exploration of more specific imaging features for accurate differentiation. Additionally, incorporating more specific features discovered in clinical studies into artificial intelligence software for lung nodule assessment may potentially enhance its ability to predict the neoplastic or non-neoplastic nature of GGNs in the future.

This study has several limitations that should be acknowledged. Firstly, it is important to note that this was a retrospective study conducted on a larger sample size. Therefore, it is crucial to validate these findings through prospective studies or in a real clinical setting. Secondly, there may be patient selection bias in this study as it only included surgically treated GGNs with pathological results. This means that the conditions of air-containing space in non-surgically treated GGNs, or those that resolve on their own, are unknown. Thirdly, while this study evaluated the performance of the air bronchogram and BLL sign in differentiating GGNs, it did not provide a comprehensive evaluation of other morphological features that have been investigated in prior studies. Finally, given the low sensitivity of the BLL sign, it is recommended to combine it with other clinical and radiological features such as lobulation, spiculation, and pleural indentation signs to accurately assess the possibility of GGNs in clinical practice.