SuPAR and PAI-1 in critically ill, mechanically ventilated patients

Abstract

Purpose: SuPAR (soluble urokinase plasminogen activator receptor) and PAI-1 (plasminogen activator inhibitor 1) are active in the coagulation-fibrinolysis pathway. Both have been suggested as biomarkers for disease severity. We evaluated them in prediction of mortality, acute lung injury (ALI)/acute respiratory distress syndrome (ARDS), sepsis and renal replacement therapy (RRT) in operative and non-operative ventilated patients. Methods: We conducted a prospective, multicenter, observational study. Blood samples and data of intensive care were collected. Mechanically ventilated patients with baseline suPAR and PAI-1 measurements were included in the analysis, and healthy volunteers were analysed for comparison. Receiver operating characteristics (ROC), logistic regression, likelihood ratios and Kaplan-Meier analysis were performed. Results: Baseline suPAR was 11.6 ng/ml (quartiles Q1-Q3, 9.6-14.0), compared to healthy volunteers with suPAR of 0.6 ng/ml (0.5-11.0). PAI-1 concentrations were 2.67 ng/ml (1.53-4.69) and 0.3 ng/ml (0.3-0.4), respectively. ROC analysis for suPAR 90-day mortality areas under receiver operating characteristic curves (AUC) 0.61 (95 % confidence interval (CI): 0.55-0.67), sepsis 0.68 (0.61-0.76), ALI/ARDS 0.64 (0.56-0.73) and RRT 0.65 (0.56-0.73). Patients with the highest quartile of suPAR concentrations had an odds ratio of 2.52 (1.37-4.64, p = 0.003) for 90-day mortality and 3.16 (1.19-8.41, p = 0.02) for ALI/ARDS. In non-operative patients, the AUC's for suPAR were 90-day mortality 0.61 (0.54-0.68), RRT 0.73 (0.64-0.83), sepsis 0.70 (0.60-0.80), ALI/ARDS 0.61 (0.51-0.71). Predictive value of PAI-1 was negligible. Conclusions: In non-operative patients, low concentrations of suPAR were predictive for survival and high concentrations for RRT and mortality. SuPAR may be used for screening for patients with potentially good survival. The association with RRT may supply an early warning sign for acute renal failure. © 2012 Springer-Verlag Berlin Heidelberg and ESICM.