Airway tracers and biomarkers for aspiration detection and VAP prediction in ICU patients with tracheal tubes: a systematic review and diagnostic accuracy meta-analysis.

Background: Ventilator-associated pneumonia (VAP) remains an important complication in mechanically ventilated patients in intensive care units (ICUs). Early subglottic microleakage may precede overt infection but remains difficult to detect at the bedside. Airway tracers and biochemical biomarkers have been proposed as bedside tools for detecting upstream microleakage and predicting downstream VAP, but their diagnostic value remains uncertain because studies differ in sampling matrices, thresholds, reference standards, and endpoints. Methods: Six databases were systematically searched from inception to 1 March 2026. Studies evaluating airway tracers or biomarkers in critically ill adults receiving invasive mechanical ventilation or with artificial airways were included under a dual-endpoint framework, comprising upstream microleakage or aspiration-related event detection and downstream VAP prediction. Study data were extracted, and risk of bias was assessed using the QUADAS-2 tool. For markers with sufficient clinical and methodological comparability, quantitative analyses included bivariate random-effects pooling, variance decomposition, stepwise sensitivity analyses, and Bayes' theorem-based post-test probability projection. Results: Sixteen studies were included. Blue-dye-based visual methods were summarized descriptively because the available studies represented distinct aspiration paradigms and did not meet the criteria for quantitative pooling. FEES-validated tracheostomy studies reported relatively high sensitivity for swallowing-related transglottic aspiration, whereas enteral-tracer or reflux-aspiration monitoring studies were limited by surrogate reference standards and/or low visual detectability for continuous, low-volume microleakage. One mechanically ventilated enteral-feeding cohort was retained as narrative evidence only because it did not provide a direct 2 × 2 diagnostic accuracy comparison of visual blue-dye detection against the microsphere reference standard. For VAP prediction, airway α-amylase showed a pooled sensitivity of 78.1%, specificity of 70.3%, and an area under the curve of 0.811. Conclusions: Current airway tracers and biomarkers do not appear suitable as standalone diagnostic triggers for VAP. However, airway α-amylase may have potential as a negative-predictive screening tool. Future research should standardize volumetric airway sampling and prospectively validate clinically actionable rule-out algorithms to support early antimicrobial de-escalation.