|| Checking for direct PDF access through Ovid
Methods containing only clinical information fail to meet the needs of prediction of acute lung injury (ALI) because of the relatively low positive predictive value. This study aimed to investigate the feasibility of using biomarkers as predictors of ALI in populations with severe sepsis/septic shock and to explore difference among biomarkers after adjustment for potential confounders.Serum specimens were collected from patients with severe sepsis/septic shock (n = 172) presented to the emergency department. Patients should be ruled out from the study if they were already suffering from ALI or if they deteriorated into ALI within 6 hours after specimen collection. The development of ALI of the remaining patients was tracked.Of all patients with severe sepsis/septic shock who encountered ALI more than 6 hours succeeding to specimen collection, 19 deteriorated into ALI. Elevation in serum interleukin 8 (IL-8) and Parkinson disease 7 (PARK7) levels had significant connection with higher risk of developing ALI (P = .006; P = .0001). Sepsis treatment and vasopressor application led to a robust connection between PARK7 and succeeding ALI development. Patients who deteriorated into ALI were distinguished accurately from patients who avoided ALI using PARK7 or Lung Injury Prediction Score (LIPS; area under the receiver operating characteristic curve [AUROC], 0.73 and 0.72 for each). Combination of PARK7 and LIPS ameliorated AUROC to 0.86 (vs 0.73, P = .05). On the contrary, serum soluble receptor for advanced glycation end products and von Willebrand factor made no contribution to the prediction of ALI development.Patients with PARK7 or IL-8 levels above normal are more vulnerable to ALI. Patients vulnerable to ALI can be distinguished with the combination of serum biomarkers and clinical prediction scores. In addition, the early rise in PARK7 emphasizes the importance of endothelial injury in the early pathogenesis of ALI.