Pulmonary artery catheter use is associated with reduced mortality in severely injured patients: A National Trauma Data Bank analysis of 53,312 patients

Objective: To evaluate the association between pulmonary artery catheter (PAC) use and mortality in a large cohort of injured patients. We hypothesized that PAC use is associated with improved survival in critically injured trauma patients. Design. Retrospective database analysis. Setting. A total of 268 level 1 trauma centers from across the United States. Patients. A total of 53,312 patients admitted to the intensive care units of the trauma centers participating in the National Trauma Data Bank maintained by the American College of Surgeons. Measurements and Main Results: The National Trauma Data Bank was queried to identify patients aged 16-90 yrs with complete data on base deficit, and Injury Severity Score (n = 53,312). Patients were initially divided into two groups: those managed with a PAC (n = 1,933) and those managed without a PAC (n = 51,379). Chi-square and Student's t-test analysis were utilized to explore group differences in mortality. In a second analysis, groups were stratified by base deficit, Injury Severity Score, and age to further explore the influence of injury severity on PAC use and mortality. In addition, a logistic regression model was developed to assess the relationship between PAC use and mortality after adjusting for differences in age, mechanism, injury severity, injury pattern, and co-morbidities. Overall, patients managed with a PAC were older (45.8 +/- 21.3 yrs), had higher Injury Severity Score (28.4 +/- 13.5), worse base deficit (-5.2 +/- 6.5), and increased mortality (PAC, 29.7%; no PAC, 9.8%; p <.001). However, after stratification for injury severity, PAC use was associated with a survival benefit in four subgroups of patients. Each of these groups had advanced age or increased injury severity. Specifically, patients aged 61-90 yrs, with arrival base deficit worse than -11 and Injury Severity Score of 25-75, had a decrease in the risk of death with PAC use (odds ratio, 0.33; 95% confidence interval, 0.17-0.62). Three additional groups had a similar decrease in the risk of death with PAC use: odds ratio, 0.60 (95% confidence interval, 0.43-0.83), 0.82 (95% confidence interval, 0.44-1.52), and 0.63 (95% confidence interval, 0.40-0.98). Logistic regression analysis demonstrated a decreased mortality when a PAC was used in the management of patients with the following severe injury characteristics: Injury Severity Score of 25-75, base deficit of less than -11, or age of 61-90 yrs (odds ratio, 0.593; 95% confidence interval, 0.437-0.805). Conclusions. Trauma patients managed with a PAC are more severely injured and have a higher mortality. However, severely injured patients (injury Severity Score, 25-75) who arrive in severe shock, and older patients, have an associated survival benefit when managed with a PAC. This is the first study to demonstrate a benefit of PAC use in trauma patients.