Prevalence and clinical associations of Staphylococcus aureus small-colony variant respiratory infection in children with cystic fibrosis (SCVSA): a multicentre, observational study

Background: Staphylococcus aureus is the bacterium cultured most often from respiratory secretions of people with cystic fibrosis. Both meticillin-susceptible S aureus and meticillin-resistant S aureus (MRSA) can adapt to form slow-growing, antibiotic-resistant isolates known as small-colony variants that are not routinely identified by clinical laboratories. We aimed to determine the prevalence and clinical significance of S aureus small-colony variants and their subtypes among children with cystic fibrosis. Methods: The Small Colony Variant Staphylococcus aureus (SCVSA) study was a 2-year longitudinal study of children aged 6–16 years at five US cystic fibrosis centres, using culture methods sensitive for small-colony variants. Children were eligible if they had a documented diagnosis of cystic fibrosis and a minimum of two cystic fibrosis clinic visits and two respiratory cultures in the previous 12 months at enrolment. Participants attended clinic visits quarterly, at which respiratory tract samples were taken and measures of lung function (percentage of predicted forced expiratory volume in 1 s [FEV₁] and frequency of respiratory exacerbations) were recorded. We determined the prevalence of small-colony variants and their subtypes, and assessed their independent associations with lung function and respiratory exacerbations using linear mixed-effects and generalised estimating equation logistic regression models. Analyses included both univariate models (unadjusted) and multivariate models that adjusted for potential confounders, including age, sex, race, baseline microbiology, treatment with CFTR modulator, and CTFR genotype. Findings: Between July 1, 2014, and May 26, 2015, we enrolled 230 children. Participants were followed-up for 2 years, with a mean of 6·4 visits (SD 1·14) per participant (range 2–9 visits) and a mean interval between visits of 3·94 months (SD 1·77). Across the 2-year period, S aureus small-colony variants were detected in 64 (28%) participants. Most (103 [56%] of 185) of the small-colony variants detected in these participants were thymidine dependent. Children with small-colony variants had significantly lower mean percentage of predicted FEV₁ at baseline than did children without small-colony variants (85·5 [SD 19] vs 92·4 [SD 18·6]; p=0·0145). Small-colony variants were associated with significantly lower percentage of predicted FEV₁ throughout the study in regression models, both in univariate analyses (regression coefficient −7·07, 95% CI −12·20 to −1·95; p=0·0068) and in multivariate analyses adjusting for potential confounders (−5·50, −10·51 to −0·48; p=0·0316). Small colony variants of the thymidine-dependent subtype had the strongest association with lung function in multivariate regression models (regression coefficient −10·49, −17·25 to −3·73; p=0·0024). Compared with children without small-colony variants, those with small-colony variants had significantly increased odds of respiratory exacerbations in univariate analyses (odds ratio 1·73, 95% CI 1·19 to 2·52; p=0·0045). Children with thymidine-dependent small-colony variants had significantly increased odds of respiratory exacerbations (2·81, 1·69–4·67; p=0·0001), even after adjusting for age, sex, race, genotype, CFTR modulator, P aeruginosa culture status, and baseline percentage of predicted FEV₁ (2·17, 1·33–3·57; p=0·0021), whereas those with non-thymidine-dependent small-colony variants did not. In multivariate models including small-colony variants and MRSA status, P aeruginosa was not independently associated with lung function (regression coefficient −4·77, 95% CI −10·36 to 0·83; p=0·10) and was associated with reduced odds of exacerbations (0·54, 0·36 to 0·81; p=0·0028). Only the small-colony variant form of MRSA was associated with reduced lung function (−8·44, −16·15 to −0·72; p=0·0318) and increased odds of exacerbations (2·15, 1·24 to 3·71; p=0·0061). Interpretation: Infection with small-colony variants, and particularly thymidine-dependent small-colony variants, was common in a multicentre paediatric population with cystic fibrosis and associated with reduced lung function and increased risk of respiratory exacerbations. The adoption of small-colony variant identification and subtyping methods by clinical laboratories, and the inclusion of small-colony variant prevalence data in cystic fibrosis registries, should be considered for ongoing surveillance and study. Funding: The Cystic Fibrosis Foundation and the National Institutes of Health.