This is the first study to use artificial intelligence quantitative lung imaging software to assess the effects of DE exposure on pulmonary vessels. In this study, we analyzed the pulmonary vessels in chest CT of subjects with occupational DE exposure and histopathological changes in chronic DE-exposed rats, which indicating that long-term exposure to DE might lead to the remodeling of small pulmonary vessels through promoting the expression of MIF.
As the most popular traffic pollutant, DE has brought profound health effects on airway inflammation and remodeling, however, there is limited research on the pulmonary vascular effects. In this study, we employed the A-View® artificial intelligence platform to analyze the CT images of the occupational DE exposed populations. As results, more small vessels were captured by CT in the DETs group due to proliferation and hypertrophy of smooth muscle cells in small vessels, as well as higher volumes of small pulmonary vessels compare to the control members [9, 38] .Consistent to this results, several cohort studies associated with traffic pollution, such as ozone [39], black carbon [40], also shown increased pulmonary vascular resistance and adverse effect on the number of pulmonary vessels. Furthermore, there are positive correlation between amount of DE exposure and some imaging parameters of small pulmonary vessels, in a cross-sectional area of 0–2 mm2, not in 2-5mm2(Supplementary Table 2). To be noticed, since most pulmonary artery remodeling started from small pulmonary vessels, this remodeling is asymptomatic until the pulmonary hypertension occur with enormous lost (roughly 50–70%) of the pulmonary vascular bed due to emphysema or other reasons [41].
In previous study, we had shown that DE exposure can lead to small airway wall thickening without lumen stenosis, occurring before airflow limitation[20]. Here we demonstrated that small airway remodeling was not directly linked to small pulmonary vessel remodeling due to DE exposure. Consistent with this, pulmonary vascular dysfunction and remodeling have been observed before airflow limitation in cigarette smoke animal or smokers [42] [43]. Therefore, the lung health effects of DE exposure are all from distal portion of airway or pulmonary vessels.
To reveal the mechanisms of the vessel remodeling of DE exposure, we conduct a chronic DE exposure animal experiment. Our study showed that DE-exposed rats had increased thickness of the small pulmonary vessels, more collagen deposition around the small pulmonary vessels, and higher RVSP. Of course, this DE exposure concentration (3mg/m3) is notably higher than the 430.8ug/m3 DETs concentration in the workshop, may explain the rapid appearance of airway and pulmonary vessel changes in the animal model. Consistent with our results, Davel et al reported that environmental particulate exposure has led to continuous contracted small pulmonary vessels with increase of endothelin [44], and Liu et al demonstrated the thickening of pulmonary artery wall, as well as apoptosis of vascular endothelial cells and proliferation of vascular smooth muscle cells [9]. However, the mechanisms of DE exposure on vascular remodeling have not been clarified yet.
As a pro-inflammatory and pro-proliferative factor, MIF was secreted by various cell types, including epithelial cells, T cells, macrophages, as well as impaired the airway epithelial and smooth muscle cells [45]. In our experiments, MIF in both the lung tissue and BALF was increased in the DE rats. In turn, MIF can recruit and activate a substantial number of alveolar macrophages and pulmonary interstitial macrophages and then resulting the release of various inflammatory mediators (such as TNF-a, IL-6) and angiogenic factors (such as PDGF, VEGF), which play a crucial role in the development of pulmonary vascular remodeling [46].
The A-view system utilized in this study offers fully automatic analysis and segmentation of pulmonary vessels, and can analysis through low-dose CT scans[14]. In comparison to the initial chest imaging platform created by the Applied Chest Imaging Laboratory of Brigham and Women's Hospital [12, 39], this system has the advanced ability to identify smaller pulmonary vessels [14], potentially allowing for earlier detection of individuals exhibiting indicative of pulmonary vascular dysfunction. Furthermore, the Apollo automated lung quantification system by VIDA requires contrast media injection for detecting smaller pulmonary vessels [47, 48], posing challenges for patients with renal insufficiency or critically ill patients. Besides utilizing small pulmonary vessel volume as an assessment metric, this system has introduced a novel indicator, the number of blood vessels, as a diagnostic criterion, offering a new approach to evaluating small pulmonary vessels.
To be honestly, there are still certain limitations for pulmonary vascular analysis by A-view system. Firstly, it is unable to differentiate between small pulmonary arteries and veins without contrast media. Secondly, due to the similarity in tissue density between blood vessel walls and blood on CT, it cannot measure the thickness of small blood vessel walls or the extent of luminal stenosis. Further research should be carried out to improve the resolution and sensitivity of CT scan or new detective methods.