Simulation of lung lesions for validating the sonography of the flooded lung.

Abstract

The quality of sonography of a unilaterally flooded lung needs to be validated on lesions of different echogenicity, size and subpleural position. Lesions were simulated in 12 young pigs with three different methods. After transbronchial (method 1) or transpleural puncture (method 2), diverse substances were injected into the lung. After 4 weeks, the thorax was opened and the lung flooded for the sonographic location of the lesions. In method 3, pulmonary lesions were simulated in an acute experiment after thoracotomy by transpleural injection or by filling of a Fogarty catheter balloon and were located sonographically. Transbronchial injection of alcohol invariably led to subsegment atelectasis. Only 25% of thoracoscopically controlled transpleural injections produced focal lesions in experiments in which the animals survived. Representative lesions were found only after alcohol injections. Transpleural injection of blood or a blood/Echovist suspension (method 3) simulated isoechogenic or echo-rich lesions with indistinct boundaries. By filling a Fogarty catheter balloon with saline solution or Echovist suspension, we succeeded in simulating echo-free or echo-rich lesions with smooth contours, located in different subpleural depths. After unilateral lung flooding, sonography successfully detected the locations of all these lesions and revealed their correlation with functional structures. Sonography of the flooded lung might be helpful in the intraoperative location of lesions, especially in the context of video-assisted thoracoscopic surgery.