Variable clinical course has been reported with the acquisition of Burkholderia cepacia in patients who have cystic fibrosis (CF). We hypothesized that the perceived worsening with B. cepacia may reflect the underlying severity of pulmonary disease at the time of acquisition. To test this hypothesis, we matched CF patients colonized with B. cepacia with CF patients not colonized with the organism. Two-year pre- and postacquisition data and long-term data were compared. Patients were matched for gender, age (+/- 1 yr), height (+/- 5 cm), weight (+/- 8 kg), percent predicted forced expiratory volume in one second (% pred FEV(1)) (+/- 10%), and pancreatic sufficiency status. Differences in rates of change pre- and postacquisition for FEV(1), FVC, weight, and frequency of intravenous courses were compared within pairs with the Wilcoxon signed rank test. Two-year and long-term survival was compared within pairs with the McNemar test. No significant differences were observed in mean annual rates of change in weight (0.33 and -0.28 kg/yr), % pred FEV(1) (-0.36 and -1.74%/yr), and percent predicted forced vital capacity (% pred FVC) (-3.80 and -2.32%/yr) between B. cepacia and control pairs in 2-yr and long-term postacquisition interval, respectively. Similar rates of change were noted for pre- to postacquisition intervals within pairs for weight (0.17 kg/yr), % pred FEV(1) (-0.16%/yr), % pred FVC (5.02 %/yr). There was a significantly higher rate of intravenous antibiotic courses in B. cepacia cases in the 2-yr and long-term postacquisition interval. Higher mortality was observed in the B. cepacia cases in the long term (p < 0.05). We conclude that colonization with B. cepacia does not necessarily adversely affect pulmonary status, but is associated with reduced long term survival. Whereas previous associations may be attributed to a propensity to colonize those who had more advanced disease, specific strain types of B. cepacia may have enhanced pathogenicity.