Abstract
Rationale: Rates of viral respiratory infection (VRI) are similar in people with cystic fibrosis (CF) and the general population; however, the associations between VRI and CF pulmonary exacerbations (PEx) require further elucidation.Objectives: To determine VRI prevalence during CF PEx and evaluate associations between VRI, clinical presentation, and treatment response.Methods: The STOP2 (Standardized Treatment of Pulmonary Exacerbations II) study was a multicenter randomized trial to evaluate different durations of intravenous antibiotic therapy for PEx. In this ancillary study, participant sputum samples from up to three study visits were tested for respiratory viruses using multiplex polymerase chain reactions. Baselines and treatment-associated changes in mean lung function (percent predicted forced expiratory volume in 1 s)(,) respiratory symptoms (Chronic Respiratory Infection Symptom Score), weight, and C-reactive protein were compared as a function of virus detection. Odds of PEx retreatment within 30 days and future PEx hazard were modeled by logistic and Cox proportional hazards regression, respectively.Results: A total of 1,254 sputum samples from 621 study participants were analyzed. One or more respiratory viruses were detected in sputum samples from 245 participants (39.5%). Virus-positive participants were more likely to be receiving CF transmembrane conductance regulator modulator therapy (45% vs. 34%) and/or chronic azithromycin therapy (54% vs. 44%) and more likely to have received treatment for nontuberculous Mycobacterium infection in the preceding 2 years (7% vs. 3%). At study visit 1, virus-positive participants were more symptomatic (mean Chronic Respiratory Infection Symptom Score, 53.8 vs. 51.1), had evidence of greater systemic inflammation (log(10) C-reactive protein concentration, 1.32 log(10) mg/L vs. 1.23 log(10) mg/L), and had a greater drop in percent predicted forced expiratory volume in 1 second from the prior 6-month baseline (5.8 vs. 3.6). Virus positivity was associated with reduced risk of future PEx (hazard ratio, 0.82; 95% confidence interval, 0.69-0.99; P = 0.034) and longer median time to next PEx (255 d vs. 172 d; P = 0.021) compared with virus negativity.Conclusions: More than one-third of STOP2 participants treated for a PEx had a positive test result for a respiratory virus with more symptomatic initial presentation compared with virus-negative participants, but favorable long-term outcomes. More refined phenotyping of PEx, taking VRIs into account, may aid in optimizing personalized management of PEx.Clinical trial registered with www.clinicaltrials.gov (NCT02781610).