Special precautions reduce oropharyngeal contamination in bronchoalveolar lavage for bacteriologic studies

Despite the use of quantitative culture, oropharyngeal contamination of bronchoalveolar lavage (BAL) specimens is still a factor that limits the usefulness of this technique in the diagnosis of lower respiratory tract infection. To investigate whether special precautions could reduce contamination, 20 noninfected patients undergoing diagnostic bronchoscopy were randomized into 2 groups of 10 patients: BAL was performed routinely in group R and with special precautions in group P. These precautions consisted of giving topical lidocaine by inhalation rather than by bolus injection, and passing the bronchoscope used for BAL through a previously inserted endotracheal tube. Quantitative culture of BAL specimens showed that 5 patients in group R (50%), but none of the patients in group P (0%), had at least 1 organism recovered in concentrations greater than or equal to 10(4) colony-forming units CFU/ml (p = 0.016). Fifteen of 39 isolates (38.5%) in group R and none of 18 isolates in group P (0%) were present in concentration greater than or equal to 10(4) CFU/ml (p = 0.001). We conclude that oropharyngeal contamination of BAL specimens can be minimized by adopting special precautions during the procedure and by using quantitative culture with 10(4) CFU/ml as the cut-off point. This may increase the specificity of the technique in the diagnosis of lower respiratory tract infection without reducing its sensitivity.