Discussion

The results of the present study showed that in CIF, the lung field density near the interlobar fissure of the bilateral upper lobes was high-density in a greater number of cases. They also showed that less cases shifted from iso-density on inspiration to high-density on expiration in IIF than in CIF. This suggests that air moves from the lower lobe to the upper lobe through the fusion of the interlobar fissure.

Lobar collateral ventilation of the lungs was confirmed in experiments on dogs by Van Allen et al [1] in the 1930s and has since been demonstrated to be clinically correct. Frontal chest radiographs[7] were also discussed for a long time and used as the theoretical basis for lobar atelectasis[8]. For some time after CT was introduced, interlobar fissures could not be visualized because the image was reconstructed using a low spatial resolution reconstruction algorithm with a thickness of 10 mm. Using thin-section imaging via Thin-Section High-Resolution CTs and a high spatial resolution reconstruction algorithm, it became possible to visualize interlobar fissures[9]. With the advancement of helical imaging and multi-row detectors, 3D images could be easily created, and today, along with pulmonary arteries and veins, interlobar fissures can be automatically reconstructed using AI.

We will explain why we chose the upper lobe side between the upper and lower lobes to examine changes in lung density. As presented at RSNA in 1994 and published in the CD-ROM version of RadioGraphics[10], we conducted an experiment on the diffusion of formalin in the IIF fusion region using right lungs removed after death, and confirmed that formalin diffuses through the fusion of the interlobar fissure to the adjacent lobe.

Interlobar fissures have three locations on the right side and two locations on the left side. Accessory fissures may also exist, in addition to a combination of CIF and IIF. Although the degree of IIF varied, we decided not to consider it this time. We examined lung field density during inspiration and expiration only between the upper and lower lobes on both sides. In the case of CIF, a preliminary study confirmed that the largest change in lung field density during inspiration and expiration was in the upper lobe on the head side of the major fissure. For IIF, the upper lobe side on the extension line of the interlobar fissure to the mediastinum was defined as the region of interest, while for CIF, the upper lobe side near the interlobar fissure on the mediastinal side was defined as the region of interest to align with the IIF. The examination range was in a 1.5 cm diameter area on inspiration and expiration CT. For the density changes from iso density during inspiration and high density during expiration in the same case, when Pearson's chi-squared test was performed, a significant difference was obtained with a p-value of 0.01 or less on both sides. This means that in the CIF, the density on the upper lobe side increased during expiration. In other words, in the IIF, the lobar collateral ventilation at the fusion area suppressed the degree of density change on the upper lobe side.

Three microstructures in the lungs involved in collateral ventilation are known. The interalveolar pores of Kohn[11] were discovered in the late 19th century, while the bronchiole-alveolar channel of Lambert[12] and the interbronchiolar channel of Martin[13] were discovered in the mid-20th century. We believe that the fused area of the IIF of the pulmonary interlobar fissure plays a major role in lesion progression. The fused area of the interlobar fissure of the IIF is believed to exhibit the same behavior as that in the lobes of the lungs in response to physiological and pathological changes in the lungs[14].

Lung field density tended to be higher in the upper lobe than in the lower lobe during expiration. We believe that this is a common perception among those who view more inspiratory and expiratory CT images. In many cases of CIF, the lung field density on the upper lobe side increased during expiration; however, for IIF, it was confirmed in this study that even if the apical density increased during expiration, the density increase was suppressed near the fused area of the IIF of interlobar fissures. This suggests lobar collateral ventilation from the lower lobe to the upper lobe via the fused area of the IIF during expiration. If this is applied to BVRT (Bronchoscopic Volume Reduction Therapy)[15,16] using a one-way valve, even if the one-way valve is placed in an interlobar fissure case to reduce the volume of the overinflated pulmonary lobe, a collapsed lung will not be obtained if there is lobar collateral ventilation from the IIF. However, since the subjects were interstitial pneumonia patients, their lungs may be stiffer than normal, and if they were healthy, lobar collateral ventilation may have been observed more clearly. In addition, it may have become clearer upon upright CT[17], which is being put to practical use.

Although nearly a century has passed since the discovery of lobar collateral ventilation, the reason why it has not been confirmed in clinical cases until now is that there are few opportunities to take inspiration and expiration images on the same day. In addition, when investigating lobar collateral ventilation, we narrowed the research target to between the upper and lower lobes, which changed the most, analyzed the vicinity of the interlobar fissure, and compared it within the same case. We initially investigated the changes in inspiration and expiration at multiple locations, but the conditions were too complicated and we were unable to obtain clear-cut results.　This was partly because of the author's conviction that lobar collateral ventilation undoubtedly exists.

Ventilation, perfusion,  and lymphatic flow are thought to play important roles in the physiological action of the lungs, since they are factors involved in lobar collateral ventilation. In particular, we hope to elucidate the lymphatic flow[18,19,20] in the future. In addition, we took images in the supine position as usual; however, we are not sure whether this is the best way to understand the pathophysiology of the various pulmonary disease. An analysis using upright CT images will be therefore still required.