摘要

Background: Pulmonary exacerbations (PEx) are a major driver of morbidity and mortality in cystic fibrosis and reducing their frequency by extending the time between them is an important therapeutic goal. Although treatment decisions for exacerbations are often made based on dynamic changes in lung function, it is not clear if these changes truly impact future exacerbation risk. We analyzed adults with chronic Pseudomonas aeruginosa infection to determine whether changes in FEV1 or duration of intravenous antibiotic therapy were associated with time to the next pulmonary exacerbation. Methods: Medical records and Cystic Fibrosis Foundation Patient Registry data were examined retrospectively to assess whether various patient-specific demographic factors and exacerbation-specific characteristics were associated with time until next exacerbation using the Andersen-Gill model in order to control for previous exacerbation frequency history. Results: We examined 59 patients with 221 CF pulmonary exacerbations over a 3-year study period. Mean age was 28.2 years and mean baseline FEV1 was 62% predicted. In our univariable model, fall in FEV1 at onset of exacerbation (median absolute -3% predicted change), recovery of FEV1 with treatment (median absolute +3% predicted change) and duration of IV antibiotics (median 16 days) were not associated with time to next exacerbation (median 93.5 days). Paradoxically each one-year increase in age was associated with a reduction in hazard of PEx by 3% (HR 0.97, P = 0.03, 95% CI 0.95-1.00). Conclusions: FEV1 drop and recovery associated with onset and treatment of a CF pulmonary exacerbation or duration of intravenous antibiotics were not predictive of time until next exacerbation. Our finding that older age may be associated with decreased hazard of exacerbation is likely due to a healthy survivor effect and should be controlled for in clinical trials of pulmonary exacerbations.

全文