Ventilator-associated pneumonia (VAP) is a significant clinical entity that affects up to 20% of patients requiring mechanical ventilation, leading to an increase in significant morbidity and mortality in the intensive care unit (ICU) (Howroyd et al., 2024). Some risk factors of VAP have been identified, such as older age, tracheostomy, and longer duration of mechanical ventilation (Rosenthal et al., 2025). There is no single best definition for the diagnosis of VAP, and current diagnostic methods rely on combinations of chest radiology, clinical signs and symptoms, and/or quantitative microbiological testing (Fally et al., 2024). Rapid identification of mechanically ventilated patients with an increased risk of VAP may accelerate the diagnosis and develop control measures (Frondelius et al., 2024). Better markers of VAP should accelerate the diagnosis, aid in antibiotic treatment, and evaluate the prognosis.

Biomarker-based diagnosis of VAP has been commonly studied using two sample types: serum and bronchoalveolar lavage fluid (BALF) (Pinilla-Gonzalez et al., 2023; Yang et al., 2025). Procalcitonin (PCT) is the precursor structure of the calcitonin hormone with 116 amino acids, and it has been described as a host-response biomarker with clinical value for diagnosing pneumonia, bacterial peritonitis, and sepsis (Paudel et al., 2020). Previous evidence showed PCT as a pertinent biomarker for VAP diagnosis and a helpful tool for antibiotic discontinuation (Cortes et al., 2021). However, a recent systematic review fails to provide sufficient evidence to support the role of PCT in the routine assessment of patients with VAP (Alessandri et al., 2021). Even in BALF, PCT concentration seemed to be less valuable in discriminating VAP and inflammation (Linssen et al., 2008; Grover et al., 2014). Presepsin, also known as soluble CD14 subtype (sCD14-ST), is a 13-kDa truncated form of soluble CD14 (sCD14) with 64 amino acid residues, which exhibits favorable diagnostic accuracy in sepsis across neonates, children, and adults (Formenti et al., 2024; Peng et al., 2025). Increasing evidence showed a higher diagnostic accuracy offered by presepsin than other conventional biomarkers, including PCT, high-sensitivity C-reactive protein (hs-CRP), and IL-6 during bacterial sepsis (Kweon et al., 2014). The plasma level of presepsin was found to be increased after bacterial infections and decreased following antibiotic treatment in critically ill patients (Galliera et al., 2019). Measurements of presepsin in the samples of tracheal aspirate obtained from intubated newborns suggested that presepsin could serve as a complementary marker in diagnosing early-onset neonatal pneumonia (Savic et al., 2018). Lipopolysaccharide-binding protein (LBP) is an acute-phase protein that plays an essential role in the pulmonary immune response to gram-negative infection (Laryushina et al., 2024). LBP level on the 1st day of suspected sepsis has shown a better diagnostic value of sepsis than PCT and soluble CD14 in critically ill neonates and children (Pavcnik-Arnol et al., 2007). LBP has been studied in combination with PCT and presepsin to diagnose and monitor VAP and sepsis, showing significant clinical value (Mussap et al., 2011; Rumende and Mahdi, 2013). However, there is limited evidence showing combinations of presepsin, PCT, and LBP for VAP diagnosis in mechanically ventilated patients. In this study, we examined serum and BALF levels of presepsin, PCT, and LBP in mechanically ventilated patients with VAP and also compared their diagnostic values alone or in combination.

The study demonstrates that combining presepsin and LBP, sampled in BALF, exhibits better diagnostic value for VAP in mechanically ventilated patients than that sampled in serum. Selecting a correct biological fluid for the measurement of specific predictors results in further improvement in a biomarker panel to diagnose the occurrence of VAP in critically ill patients with mechanical ventilation.

The role of PCT as a biomarker of VAP remains controversial in an earlier work. Song et al (Song et al., 2020). demonstrated that serum PCT is an early, sensitive, and highly specific high-risk monitoring index and has an early prediction value for VAP after cardiac valve replacement. However, in the study of Borkowska et al., no association was observed between PCT level and the occurrence of VAP (Zielinska-Borkowska et al., 2012). In our study, we determined a higher baseline level of PCT sampled in serum but not in BALF in patients developing VAP, aligning with a previous study in which serum but not alveolar PCT was helpful for diagnosing VAP and mortality (Duflo et al., 2002). The synthesis of PCT may be associated with an increased production of tumor necrosis factor-α (TNF-α), as both exhibited a similar expression pattern during infection (Mazaheri et al., 2022). In healthy and septic animals, PCT injection did not initiate or enhance the production of TNF-α, while TNF-α injection induced a 25-fold massive and sustained PCT increase (Whang et al., 2000). This suggests that PCT secretion is an “intermediary” rather than a “proximal” event in the sepsis cascade that requires a “primed” inflammatory background to exert its effect (Dahaba and Metzler, 2009). TNF-α concentration was significantly higher systemically than in the BALF of patients with acute respiratory distress syndrome (Agouridakis et al., 2002). Therefore, in our study, systemic cytokine production can explain the increased serum procalcitonin concentrations, whereas the low alveolar procalcitonin release could be explained by the lower production of the local mediators.

The levels of presepsin and LBP were likely to outperform PCT in evaluating infection severity and the prognosis of severe sepsis/septic shock (Mierzchala et al., 2011; Tsujimoto et al., 2018). Presepsin, a cell-surface glycoprotein, is constitutively expressed on the surface of multiple cells, including monocytes, neutrophils, and macrophages (Wu et al., 2019). It is also a high-affinity receptor specific for lipopolysaccharides (LPS) and LBP complexes, regulating LPS-triggered apoptosis (Tsukamoto et al., 2018). When presepsin binds the LBP complex, Toll-like receptor 4 (TLR4)-specific proinflammatory signaling is activated, followed by initiation of host response against infectious agents (Na et al., 2023). Presepsin is produced by proteolytic shedding of the membrane CD14 (mCD14) during cellular activation; thus, presepsin can be detected proportionately in the BALF followed by the release of the LPS–LBP–CD14 complex (Marcos et al., 2010). Although not studied in VAP patients, Mabrey et al (Mabrey et al., 2021). found that higher plasma soluble CD14 subtype was associated with worse clinical outcomes in critically ill patients. LBP, as a component of the bacterial outer membrane, recognizes lipopolysaccharides and transfers to presepsin, which plays a multifunctional role in potentiating the recognition, clearance, and killing of gram-negative bacteria (Taddonio et al., 2015). Its upregulation is considered part of the acute-phase response. Although LBP is known to respond to gram-negative components, a more expanded role for LBP as a general recognition molecule has been shown (Zweigner et al., 2006). Several bacterial surface components, such as lipoteichoic acid, from gram-positive pathogens are also recognized by this molecule. LBP regulated the complexation of lipoteichoic acid with CD14 and its biological activity toward immune cells. Lipoteichoic acid-like glycolipids isolated from spirochetes are recognized by LBP and initiate signaling in the presence of LBP (Schroder and Schumann, 2005). The upregulation of LBP in serum was found in patients with sepsis regardless of gram-negative, gram-positive, or fungal infections, suggesting that LBP may be less specific to gram-negative infection (Blairon et al., 2003). In our study, higher levels of serum LBP were determined in the gram-negative group than in the gram-positive group, suggesting the primary role of LBP in response to gram-negative infections in VAP. However, the diagnostic value of LBP in gram-negative or gram-positive infection was limited in our further subgroup analysis. Rumende et al (Rumende and Mahdi, 2013). demonstrated that combining PCT and LBP could serve as good prognostic markers to predict mortality in patients with VAP. IL-6 alone or in combination with IL-1β led to elevations in LBP and presepsin concentrations in culture supernatants of human bronchial epithelial cells, suggesting that airway epithelial cells produce LBP and presepsin after stimulus treatment (Regueiro et al., 2009). In our study, although presepsin and LBP were increased in both serum and BALF samples, two of them in BALF alone or in combination exhibited better diagnostic value for patients with confirmed VAP compared with their serum counterparts. Furthermore, combined analysis of the BALF presepsin and LBP also conferred a significant diagnostic value for gram-negative VAP.

The study has certain limitations. First, there was a lack of serial sampling to monitor the dynamic status of the studied biomarkers, which may limit their interpretation and translation into clinical applications. Second, the gold standard for diagnosis of pneumonia is still tissue histology; thus, the lack of a VAP gold standard may lead to misclassification of some patients with non-documented bacterial infection. Third, all patients in this study were recruited due to suspicion of VAP, and the absence of a control group of patients without suspected VAP may limit the clinical utility of the biomarkers studied. Fourth, VAP caused solely by viral infections was excluded in this study, which may limit the generalizability of our findings to virus-induced VAP. Fifth, although several indications for mechanical ventilation were reported in the study, the single-center design with a relatively small sample size may still have missed critically ill patients with other causes, which may also limit the generalizability of our results in other settings.

In conclusion, the study demonstrates that presepsin combined with LBP in BALF may be an adjunct to diagnose definitive VAP among mechanically ventilated patients with suspected VAP. Our findings support the incorporation of presepsin and LBP in BALF into the biomarker panel for definitive VAP in critically ill patients.