ARDS after pulmonary contusion: Accurate measurement of contusion volume identifies high-risk patients

Background. The pathophysiology of pulmonary contusion (PC) is poorly under stood, and only minimal advances have been made in management of this entit y over the past 20 years. Improvement in understanding of PC has been hinde red by the fact that there has been no accurate way to quantitate the amoun t of pulmonary injury. With this project, we examine a method of accurately measuring degree of PC by quantifying contusion volume relative to pulmona ry function and outcome. Methods. Patients with PC from isolated chest trauma who had admission ches t computed tomographic scan were identified from the registry of a Level I trauma center over a 1.5-year period. Subsequently, prospective data on all patients admitted to the intensive care unit with PC during a 5-month peri od were collected and added to the retrospective database. Using computer-g enerated three-dimensional reconstruction from admission chest computed tom ographic scan, contusion volume was measured and expressed as a percentage of total lung volume. Admission pulmonary function variables (Pao(2)/FiO(2) , static compliance), injury descriptors (chest Abbreviated Injury Score, I njury Severity Score, injury distribution), and indicators of degree of sho ck (admission systolic blood pressure, admission base deficit) were documen ted. Outcomes included maximum positive end-expiratory pressure, ventilator days, pneumonia, and acute respiratory distress syndrome (ARDS). Results. Forty-nine patients with PC (35 bilateral) were identified. The av erage severity of contusion was 18% (range, 5-55%). Patients were classifie d using contusion volume as severe PC (greater than or equal to 20%, n = 17 ) and moderate PC (< 20%, n = 32). Injury Severity Score was similar in the severe and moderate groups (23.3 vs. 26.5, p = 0.33), as were admission Gl asgow Coma Scale score (12 vs. 13, p = 0.30), admission blood pressure (131 vs. 129 mm Hg, p = 0.90), and admission Pao(2)/Fio(2) (197 vs. 255, p = 0. 14). However, there was a much higher rate of ARDS in the severe group as c ompared with the moderate group (82% vs. 22%, p < 0.001). There was a trend toward higher pneumonia rate in the severe group, with 50% of patients in the severe group developing pneumonia as compared with 28% in the moderate group (p = 0.20). Conclusion Extent of contusion volumes measured using three-dimensional rec onstruction allows identification of patients at high risk of pulmonary dys function as characterized by development of ARDS. This method of measuremen t may provide a useful tool for the further study of PC as well as for the identification of patients at high risk of complications at whom future adv ances in therapy may be directed.