Ultrasound assessment of pleural invasion via visceral pleura interruption and arch distance-to-tumour diameter ratio in a rabbit model of subpleural lung cancer.

Background: Pleural invasion of lung cancer is an adverse factor for treatment and prognosis but difficult to assess during clinical staging. We investigated the feasibility of assessing the arch distance-to-perpendicular tumour diameter ratio (A/Dper; "per" stands for perpendicular diameter) and visceral pleura interruption on ultrasound (US) images for evaluating pleural invasion in lung cancer models. Methods: The lung cancer model was constructed in thirty rabbits. The length of the interface between the primary tumour and neighbouring structures and the tumour diameters were measured on US images. The integrity of the visceral and parietal pleura was observed. The pathological tissue was evaluated with elastic tissue staining and hematoxylin and eosin staining. Receiver operating characteristic (ROC) curves were used to assess the ratios. The diagnostic efficacy of the ratios and pleural interruption were calculated. Results: The cut-off value of A/Dper for distinguishing PL3 from PL2 tumours was 1.55, with the sensitivity, specificity, and accuracy and the area under the ROC curve of 100.0%, 87.5%, 92.9%, and 0.979, respectively. Evaluation of the continuity of the visceral pleura on US images revealed 12 of the 13 (92.3%) PL0+1 tumours and 12 of the 14 (85.7%) PL2 or PL3 tumours. The combination of them for evaluating pleural invasion achieved an excellent consistency between the pathological and US results. Conclusions: Visceral pleura interruption and A/Dper on US images are potential indications for evaluating pleural invasion, and should be further validated in more clinical trials.