Separation between the chest wall and subpleural lung lesions: A two-step method to preoperatively exclude invasion or focal pleural adhesion by multidetector computed tomography.

PURPOSE: To develop and assess a non-invasive two-step method for evaluating the relationship between the parietal pleura and peripheral pulmonary lesions to preoperatively exclude invasion or focal pleural adhesion by multidetector computed tomography (CT).

METHODS: Twenty-six patients with pulmonary peripheral lesions who underwent surgical lung resection between May and December 2017 were enrolled in this study. Routine CT was performed in the inspiratory phase in the supine position. Additional CT examinations were performed both in inspiratory and expiratory phases in the affected-side-up lateral position. Axial, sagittal, and coronal images were reconstructed from the CT data. In the first-step analysis, we evaluated the separation between the chest wall and subpleural lung lesions (separation) by comparing inspiratory- and expiratory-phase images obtained in the affected-side-up lateral position. When the separation was absent, we performed a second-step analysis, where we compared images obtained in the supine position during routine CT with those obtained in the affected-side-up lateral position and subsequently assessed the presence and absence of the separation.

RESULTS: In the first-step analysis, the separation was observed in 21 lesions, which were categorised as showing "no invasion" or "no focal adhesion" on the basis of histological findings. After the second-step analysis, the separation was absent in three lesions and present in two; the latter two lesions were categorised as showing "no invasion" or "no focal adhesion" on the basis of operative and histological findings. Of the three lesions that did not exhibit the separation in either step of the analysis, two were diagnosed as exhibiting parietal pleural invasion on the basis of histological findings, while the third was categorised as showing "no invasion" or "no focal adhesion" on the basis of operative and histological findings. The sensitivity, specificity, positive and negative predictive values, and accuracy of this two-step method were 96% (95% confidence interval [CI]: 79-100%), 100% (95% CI: 16-100%), 100%, 67% (95% CI: 23-93%), and 96% (95% CI: 80-100%), respectively.

CONCLUSIONS: Our two-step method is especially useful for excluding the parietal pleural involvement of peripheral pulmonary lesions. Even when the separation between the chest wall and subpleural lung lesions was limited, the change in position was useful for observing the separation and excluding parietal pleural involvement. This novel two-step method also has the advantage of being simple, cost-effective, and universally available.