Early Respiratory Bacterial Detection and Antistaphylococcal Antibiotic Prophylaxis in Young Children with Cystic Fibrosis

Abstract

Consensus is lacking regarding antistaphylococcal antibiotic prophylaxis use for young children with cystic fibrosis. Prophylaxis is recommended in the United Kingdom, but it is recommended against in the United States. To test the hypothesis that antistaphylococcal antibiotic prophylaxis is associated with a decreased risk of Staphylococcus aureus acquisition but no increased risk of Pseudomonas aeruginosa acquisition. We undertook a longitudinal observational study of children with cystic fibrosis who were recruited from birth (or from their first registry entry in the period) and followed until the age of 4 years (1,500 d) using 2000-2009 data from the UK Cystic Fibrosis Trust and Cystic Fibrosis Foundation registries. Children were excluded if they had a positive culture result for S. aureus or P. aeruginosa, or if they were receiving inhaled antibiotics, at the first encounter. Time to first S. aureus and P. aeruginosa detection in the UK/U.S. cohorts was compared using a Cox proportional hazards model. A UK-based analysis compared the same for those receiving flucloxacillin with those who received no prophylaxis. We included the following covariates: sex, age at registry entry, dornase alfa use, genotype, and center size. The primary analysis comprised 1,074 UK and 3,677 U.S. children. The risk of first detection was greater in U.S. children than in UK children for S. aureus (hazard ratio [HR], 5.79; 95% confidence interval [CI], 4.85, 6.90; P < 0.001) and P. aeruginosa (HR, 1.92; 95% CI, 1.65, 2.24; P < 0.001). In the UK analysis, we compared 278 children receiving flucloxacillin and 306 receiving no prophylaxis. Flucloxacillin was not associated with a reduced risk of S. aureus detection (HR, 1.22; 95% CI, 0.74, 2.0; P = 0.43), but it was associated with an increased risk of P. aeruginosa detection (HR, 2.53; 95% CI, 1.71, 3.74; P < 0.001). None of the covariates significantly affected the risk estimate in either analysis. The risk of first detection of S. aureus and P. aeruginosa was greater in the United States than in the United Kingdom. In the United Kingdom, the risk of first P. aeruginosa detection was increased among those receiving flucloxacillin compared with those who received no prophylaxis. These observational findings should be examined in randomized controlled trials.