Nonbronchoscopic bronchoalveolar lavage: A diagnostic tool for ventilator-associated pneumonia

ISSUE: Significant mortality and morbidity is attributed to ventilator-associated pneumonia (VAP). Diagnosis of bacterial and fungal VAP is complicated by contaminating upper airway microbial flora and by nonspecific diagnostic criteria. False-positive VAP diagnosis is estimated at 23%–65%. Smear and culture results are critical to identifying the responsible pathogen and selection of appropriate antimicrobial therapy. Care must be taken in assessing intubated patients to distinguish among tracheal colonization, upper respiratory tract infection, or normal colonization and early onset pneumonia. Inaccurate culture results can lead to inappropriate antibiotic therapy and the emergence or selection of resistant microorganisms. Quantitative cultures of lower respiratory tract secretions may differentiate contaminates from pathogens. PROJECT: This medical center is a level one-trauma center with a 44-bed surgical intensive care unit (SICU). Ventilator days average 630 per month. In 2004, respiratory therapists were trained to perform nonbronchoscopic bronchoalveolar lavage (NBBAL), BAL Cath™ (Ballard Medical Products, Draper, Utah). The NBBAL procedure consisted of the blind placement of a double lumen catheter into the lung to retrieve a bronchoalveolar lavage (BAL) fluid specimen. RESULTS: Between April 2004 and December 2004, 158 BAL specimens were collected in the SICU, a 66% increase in quantitative respiratory specimens compared with those collected between April 2003 and December 2003 (prior to the use of NBBAL). Microbial results from 69 NBBAL quantitative cultures showed 36 samples with pathogens only, 18 with no pathogens, and 15 with likely contamination or colonizers. Respiratory sputum cultures were not routinely collected at the time of the NBBAL. When available, sputum cultures from same patients were identified and evaluated. LESSONS LEARNED: NBBAL use increased validity of suspected respiratory infection by 66%. The presence of pathogens was substantial in 52% of patients and excluded in 26% of patients and required interpretation of likely contaminates or colonizers in 22%. Overall, 36% fewer microbial species were included in the NBBAL samples than in paired sputum samples submitted on the same patient. NBBAL is less expensive, more readily available because a skilled bronchoscopist is not required, and less invasive while providing improved assessment of VAP in the ICU setting.