Development of a finite element-based injury metric for pulmonary contusion part I: model development and validation

Pulmonary contusion is the most commonly identified thoracic soft tissue injury in an automobile crash and after blunt chest trauma and affects 10-17% of all trauma admissions. The mortality associated with pulmonary contusions is significant and is estimated to be 10-25%. Thus, there is a need to develop a finite element model based injury metric for pulmonary contusion for the purpose of predicting outcome. This will enable current and future finite element models of the lung to incorporate an understanding of how stress and strain may be related to contusion injuries. This study utilizes 14 impacts onto male Sprague-Dawley rats. In 5 of these tests, a calibrated weight (46 g) is dropped from a height of 44 cm directly onto the lungs of intubated, anesthetized rats in situ. Contused volume is estimated from MicroPET scans of the lung and normalized on the basis of liver uptake of 18F-FDG. The lungs are scanned at 24 hours, 7 days, and 28 days (15 scans), and the contused volume is measured. In addition, 9 controlled mechanical tests on in situ rat lung are used for model development and validation. Identical impacts are performed on a finite element model of the rat lung. The finite element model is developed from CT scans of normal rat and scaled to represent average rat lung volume. First principal strain is chosen as a candidate injury metric for pulmonary contusion. The volume of contused tissue at the three time points measured using PET is compared to the strain level achieved by a corresponding volume in the finite element model. For PET scans (n=5 scans per time point), the average contusion volume was 4.2 cm3 at 24 hours, 2.8 cm3 at 7 days, and 0.39 cm3 at 28 days. These volumes were used to identify threshold peak first principal strain levels measured by the finite element model. Maximum first principal strain from the finite element model for the three volume levels (4.2, 2.8, and 0.39 cm3) was 3.5%, 8.8%, and 35% strain, respectively. Furthermore, the lung model exhibited exponential decay in principal strain threshold as more of the lung volume was considered, correlating to the precise and well defined volume of the contusion as it healed. The results of this study may be used to establish an injury metric to predict pulmonary contusion due to an impact to the lungs. The results may be used to improve finite element models of the human body, which may then be used to tune stiffnesses of interior components of automobiles and tune safety systems for maximum mitigation of this serious injury.