Abstract

OBJECTIVE:

Ventilator-associated pneumonia (VAP) contributes significantly to morbidity and mortality in critically ill patients, but it can be difficult to diagnose. Clinical criteria, Clinical Pulmonary Infection Score, and quantitative culture of bronchoalveolar lavage have been used to distinguish between patients who are likely positive (sensitivity) and patients who are likely negative (specificity). Despite these test methods, patients continue to be misclassified. False-positive results may lead to inappropriate antibiotic use in patients. For those misclassified as test negative, appropriate treatment may be delayed. Biomarkers have been suggested as another method to enhance the ability to predict VAP. This article analyzes the evidence for the usefulness of 3 biomarkers that have been proposed as possible biomarkers of VAP: soluble triggering receptor expressed on myeloid type 1 cells, procalcitonin, and C-reactive protein.

METHODS:

A Medline search was conducted for the years between 1990 and 2009 to locate articles on the subject of biomarkers for predicting VAP in critically ill adult patients.

RESULTS:

Analysis of the literature does not currently support a clinical role for these biomarkers in predicting VAP. Variations in the diagnostic methods, antimicrobial use, cutoff values, and patient populations limit comparisons among the studies.

CONCLUSION:

Recommendations are offered to strengthen and standardize methods in future studies to clarify the utility of biomarkers for predicting VAP in specific patient populations.

Copyright © 2011 Elsevier Inc. All rights reserved.