© ERS 2019. Introduction: Diagnosing ventilator-associated pneumonia (VAP) remains challenging. Soluble urokinase plasminogen activator receptor (suPAR) has prognostic value in critically ill patients with systemic infection. We hypothesised that plasma suPAR levels accurately predict development of VAP. Methods: This observational, multicentre, prospective cohort study compared patients at risk for VAP with a control group. Plasma and tracheal aspirate samples were collected. Plasma suPAR levels were measured on the day of diagnosis and 3 days before diagnosis. Results: The study included 24 VAP patients and 19 control patients. The suPAR concentration measured 3 days before diagnosis was significantly increased in VAP patients versus matched samples of control patients (area under the receiver operating characteristic curve (AUC) 0.68, 95% CI 0.52-1.00; p=0.04). Similar results were found on the day of diagnosis (AUC 0.77, 95% CI 0.6-0.93; p=0.01). Plasma suPAR was significantly higher in deceased patients (AUC 0.79, 95% CI 0.57-1.00; p<0.001). Combining suPAR with the Clinical Pulmonary Infection Score, C-reactive protein and/or procalcitonin led to a significantly increased discriminative accuracy for predicting VAP and an increased specificity. Conclusions: suPAR can be used to diagnose VAP with a fair diagnostic accuracy and has a moderate prognostic accuracy to be used in critically ill intensive care unit patients. Its performance improves when added to other clinically available biomarkers (C-reactive protein and procalcitonin) or scoring systems (Clinical Pulmonary Infection Score and Sepsis-related Organ Failure Assessment).