Introduction Biomarkers have been investigated in order to speed up diagnosis of VAP, a common condition in ICU patients. TREM-1 is a protein involved in amplification of immune responses to bacterial and fungal infection and exists as soluble and surface forms.^{1 2} The diagnostic value of soluble TREM-1 in broncho-alveolar lavage fluid (BALF) in VAP is controversial.3 Therefore the utility of surface TREM-1 for diagnosing VAP in a two-compartment model (BALF and blood) was investigated.

Methodology Paired blood and BALF were obtained in consenting patients in the following groups: (1) Ventilated patients with VAP diagnosed on semi-quantitative microbiology and Clinical Pulmonary Infection Score (CPIS); (2) Ventilated patients without sepsis; (3) Day-case bronchoscopy patients without evidence of infection. Flow cytometry was performed on cell pellets derived from simultaneous BALF and blood samples. Surface TREM-1, CD11b (immune cell activation marker) and L-selectin (immune cell migration marker) levels were measured on monocytes and neutrophils. At the same time an inflammatory cytokine panel (comprising IL-1β, IL-6, IL-8 and soluble TREM-1) was measured by ELISA in the paired blood and BALF samples.

Results Expression of TREM-1 and CD11b on monocytes were significantly elevated in BALF samples obtained from the VAP patient group. There was no change in blood surface TREM-1 and CD11b levels between the different patient groups. The BALF/blood ratio of monocytic TREM-1 increased the discrimination between the VAP and non-VAP groups (Abstract S77 Figure 1). Soluble TREM-1 levels were not significantly different between the groups. There was no difference in peripheral blood white cell count, CRP, expression of CD11b, L-selectin and inflammatory cytokines between VAP and disease controls.

• Download figure
• Open in new tab
• Download powerpoint

Abstract S77 Figure 1

The ratio monocytic surface TREM-1 between BALF and blood (flow cytometry). 16 patients with VAP are compared with 8 ventilated non-septic control (VC) and 17 non-ventilated non-infected control patients (NVC). The median levels and IQRs are: VAP (0.92, 0.66–1.77), VC (0.21, 0.12–0.28) and NVC (0.36, 0.22–0.41). * and # p=0.0001.

Conclusion The BALF/blood ratio of monocytic surface TREM-1 discriminates between the VAP and non-VAP groups. Measurement of surface TREM-1 using a two-compartment index may have utility in diagnosing VAP.