Cumulative respiratory pathogen detection burden by multiplex PCR is associated with mortality in a TB-enriched ICU cohort.

Rapid multiplex polymerase chain reaction (mPCR) is increasingly used in intensive care units (ICUs), but the prognostic value of cumulative pathogen detection burden in critically ill patients with suspected pulmonary tuberculosis (TB) or complex respiratory infections remains unclear. We retrospectively studied 130 ICU patients with severe respiratory symptoms and suspected TB or complex respiratory infections. Bronchoalveolar lavage fluid (BALF) was analyzed by mPCR within 48 h of admission. Multivariable logistic and Cox proportional hazards regression models evaluated the association between total pathogen detection count and mortality after adjustment for age, Acute Physiology and Chronic Health Evaluation II score, prior anti-infective therapy, and pre-ICU hospitalization duration. The mPCR positivity rate was 93.8%, with mixed-category co-detections in 46.9%. The most frequently detected targets were IMPORTANCE: Rapid molecular assays frequently detect multiple respiratory pathogens in critically ill patients, but the clinical interpretation of these complex results remains challenging. In this single-center intensive care unit (ICU) cohort enriched for suspected tuberculosis and complex respiratory infections, bronchoalveolar lavage fluid multiplex polymerase chain reaction showed frequent mixed-category pathogen detections. A higher cumulative pathogen detection count was associated with increased short-term mortality after adjustment for illness severity, prior anti-infective therapy, and pre-ICU hospitalization duration. These findings suggest that pathogen detection complexity may provide a risk-stratification signal beyond simple pathogen identification. However, molecular detection does not necessarily indicate active infection, and the observed association should not be interpreted as evidence that reducing detection burden alone would improve outcomes. Prospective multicenter studies incorporating adjudicated infection status, antimicrobial exposure, host immune status, and semi-quantitative pathogen-load metrics are needed to validate whether this metric can be incorporated into ICU prognostic models.