Abstract
Previous studies have found several biomarkers for acute respiratory distress syndrome (ARDS), but the accuracy of most biomarkers is still in doubt due to the occurrence of other comorbidities. In this systematic review and meta-analysis, we aimed to explore ideal ARDS biomarkers which can reflect pathophysiology features precisely and better identify at-risk patients and predict mortality. Web of Science, PubMed, Embase, OVID, and the Cochrane Library were systematically searched for studies assessing the reliability of pulmonary-originated epithelial proteins in ARDS. A total of 32 studies appeared eligible for meta-analysis, including 2654 ARDS/ALI patients in this study. In the at-risk patients' identification group, the highest pooled effect size was observed in Krebs von den Lungren-6 (KL-6) (SMD: 1.17 [95% CI: 0.55, 1.79]), followed by club cell proteins 16 (CC16) (SMD: 0.74 [95% CI: 0.01, 1.46]), and surfactant proteins-D (SP-D) (SMD: 0.71 [95% CI: 0.57, 0.84]). For the mortality prediction group, CC16 exhibited the largest effect size with SMD of 0.92 (95% CI: 0.42, 1.43). Meanwhile, the summary receiver operating characteristic (SROC) of CC16 for ARDS diagnosis reached an AUC of 0.80 (95% CI: 0.76, 0.83). In conclusion, this study provides a ranking system for pulmonary-originated epithelial biomarkers according to their association with distinguishing at-risk patients and predicting mortality. In addition, the study provides evidence for the advantage of biomarkers over traditional diagnostic criteria. The performance of biomarkers may help to clinically improve the ARDS diagnosis and mortality prediction.