Abstract
We evaluated 819 isolates referred to us as "Burkholderia cepacia" from cystic fibrosis (CF) clinics and research laboratories from five countries; 28 (3.4%) were not B. cepacia. A further 12 (1.5%) organisms appeared to be other Burkholderia species, but identification could not be confirmed by conventional means. The most prevalently misidentified organisms were Stenotrophomonas maltophilia, Alcaligenes xylosoxidans, and Comamonas acidovorans. Many of these organisms grew on oxidation-fermentation polymyxin-bacitracin-lactose (OFPBL) and Pseudomonas cepacia agars, selective media currently used for B. cepacia isolation. We developed a new medium, B. cepacia selective agar (BCSA), which is more enriched for the growth of B. cepacia yet which is more selective against other organisms than currently available selective agars. A total of 190 of 191 (99.5%) isolates of B. cepacia from patients with CF grew on BCSA without vancomycin, whereas 100% grew on OFPBL agar and 179 (94.2%) grew on P. cepacia agar. Of 189 other gram-negative and gram-positive organisms tested, 10 (5.3%) grew on BCSA without vancomycin. The addition of vancomycin to BCSA lowered the false positivity rate to 3.7% without further inhibition of B. cepacia. The false positivity rates for OFPBL and P. cepacia agars were 19.6 and 13.8%, respectively. Isolates of B. cepacia from CF patients grew most quickly on BCSA, with 201 of 205 (98.0%) being readily visible within 24 h, whereas 182 (88.8%) grew on OFPBL agar and 162 (79.0%) grew on P. cepacia agar within 24 h. We propose that the use of BCSA will allow investigators to overcome many of the difficulties associated with the identification of B. cepacia and should be considered for use as a primary isolation agar for specimens from patients with CF.