| As one of the important type of traffic trauma and fall injuries from height, pulmonary contusion caused by blunt impact on chest is usually accompanied by multiple injuries and related closely to acute respiratory dysfunction syndrome (ARDS) that would result in very poor prognosis. Therefore, it is vital to make a correct and prompt diagnosis of pulmonary contusion. CT is recognized as the most accurate and sensitive means for diagnosis of pulmonary contusion. Severe pulmonary contusion is characterized by apparent hemorrhage, tissue edema and vessel destructions that are helpful for a definite diagnosis. Until now, we are still not clear if the location and the severity of pulmonary contusion are analogous with manifestation of CT. As for patients with relatively moderate trauma, symptoms remain atypical and CT manifestation shows no obvious sign of pulmonary contusion, when the pulmonary contusion may be neglected easily. Under such circumstance, if the patients are under an incorrect treatment, pulmonary contusion would deteriorate rapidly, cause severe pulmonary edema, respiratory dysfunction even result in death. Therefore, early and accurate diagnosis of pulmonary contusion is significant importance. There are still not definite and direct evidence to prove that the status of pulmonary edema and hemorrhage is coincident with CT signs. Thus, this research aims to explore the relationship between CT appearance and real features of lung injuries after impacting through making and comparing 3D reconstruction of the pathological digitized sectional images of the lungs and CT graphs on the basis of a suitable animal model of chest blunt impact injury. Materials and Methods: 1. Constructing swine model of simple pulmonary contusion. Seven miniature swines weighing (18.2±2.9) kg was under fasting for 12 hours before teat, intubated through femoral and pulmonary arteries and impacted at the dexter chest with BIM-ⅡBioimpactor at the speed of 19 m/s and compression rate of 15% for establishment of animal model. Hemodynamic data collected before and 0.25, 3 and 6.5 hours after injury as well as the artery blood gas were analyzed. CT was done 6.5 hours after impacting followed the swines being sacrificed to perform dissection and pathological observation. 2. The reconstruction and contrast of pathological digitized sectional image and CT photographs. One miniature swine weighing 21 kg (65 cm long) was injured as the procedure described above. Posterior to 6.5 hours, CT scanning was carried out. Then, the blood in pulmonary vascular cavity was flushed and lung tissues fixed by formaldehydum perfusion through pulmonary artery. After being kept at -30 ℃in 0.9% saline for 1 week, the thorax was milled by 1mm per layer using TK26350 milling machine at -25 ℃and the pathologic digitized sectional images for thorax of miniature swine were collected with high clearance digital photoghrapy layer by layer. All the images were processed by photoshop 7.0 to make sure that the localized points and the size of a standard image are in coincidence with the others. Then, the serial images of CT and pathological digitized sectional images were processed by 3D-doctor software. The boundary of dexter lung and injured region of the images were segmented by multiple borderline-drawing methods. Meanwhile, the corresponding layer between CT and pathological digitized sectional images was contrasted and the size of injured region measured. Thereafter, the boundary of the lung and the injured region were reconstructed to calculate the volume of dexter lung and the interior injured region respectively as well as the ratio of the latter to the former. The reconstructed images were compared between the dexter lung and the interior injured region. Results and conclusions: 1. Miniature swine model of pulmonary contusion was constructed successfully by BIM-ⅡBioimpactor at the speed of 19 m/s and the compression rate of 15%. The injured region of this model is localized, with mild or moderate injury severity. 2. The results of CT almost consist with the appearance of the general sample near the impact point. But in the dorsal part of the lower and middle lobes of the dexter lung, we have found 3 circumstances: (1) Hemorrhage was showed up both in CT and generalsample; (2)Hemorrhage emerged in CT rather than in general sample; (3) Hemorrhage was suspicious in CT but obvious in general sample. 3. The bleeding area in 2D images of CT is less than that in PDSI dramatically, and ratio of the volume of the injured region to the lung that of in 3-D images of CT is far more lower than that of PDSI. 4. PDSI and the 3-D reconstruction images of it keep the original configuration basically and have a better comparability with CT images. The experimental technique we used hasn't destroyed the pulmonary pathologic configuration and is a feasible and credible way to observe the hemorrhage of lung impacted by blunt force. 5. There was only slight density increase in the dorsal region of the lower and middle lobes of the dexter lung by CT scanning after blunt impacting. Nonetheless, according to the lung PDSI, there found much larger areas with obvious bleeding in above mentioned areas, as shows that there is great error when CT is used to make a diagnose in this region. So, we must pay much more attention to the patients suspected of pulmonary contusion manifested by CT so as to avoid rapid exacerbation caused by improper dealing. |
PDF Full Text Request