Reliability of quantitative cultures of protected specimen brush after freezing.

Abstract

Protected specimen brush (PSB) is considered to be one of the standard methods for diagnosing mechanical ventilator-acquired pneumonia at a threshold value > or = 10(3) cfu/ml. Nevertheless, this procedure requires immediate cultures which are not always possible 24 h per day. We therefore wanted to appreciate the diagnostic value of delayed quantitative cultures after specimen freezing. PSB was performed by fiberoptic bronchoscopy on 43 mechanically ventilated patients with suspicion of nosocomial bronchopneumonia. After PSB procedure, two aliquots of 0.5 ml were prepared. One aliquot was plated immediately on different culture media (Group 1). A second aliquot was frozen at -80 degrees C for 24 h, then plated on the same culture media as Group 1 (Group 2). All samples were incubated for 48 h. The diagnostic value threshold of PSB was 10(3) cfu/ml. A total of 47 samples were performed on 43 patients. In Group 1, cultures from PSB were positive in 26 samples and revealed 41 species yielding > or = 10(3) cfu/ml. In Group 2, PSB cultures were positive in 24 samples and revealed 36 species yielding > or = 10(3) cfu/ml. Despite a mean decrease in bacterial count of 1.00 +/- 1.44 log 10 (p < 0.001), most important for Streptococcus pneumoniae and Escherichia coli (respectively 3.22 +/- 2.21 log10 and 2.41 +/- 0.52 log 10), sensitivity and specificity of quantitative cultures after specimen freezing, compared with immediate cultures, were 88% and 100% respectively. We concluded that specimens from PSB could be frozen at -80 degrees C with good reliability except for S. pneumoniae and E. coli, enabling PSB procedure to be performed around the clock.