P240 Association between adherence to azithromycin and dornase alpha and lung function decline in adult cystic fibrosis (CF) patients: a two-year analysis

Abstract

Objectives: Poor adherence to prescribed treatments is well documented in CF and has been shown to be a predictor of intravenous antibiotic (IVAB) usage and lung function, but not other health outcomes. The aim of this study was to examine the relationship between medication adherence and lung function decline as well as pulmonary exacerbations in adults with CF over a 24-month period. Methods: Patients prescribed azithromycin or dornase alpha over the 24-month period had medical, pharmacy and spirometry records retrospectively audited. Pharmacy refill records were used to calculate the medication possession ratio (MPR) of each medication. A composite MPR was calculated for all patients adjusted for patients prescribed both medications. Lung function decline was measured using all percent predicted (pp)FEV1 data available from the same period. Longitudinal analysis with a linear mixed model was used to determine the relationship in change in ppFEV1 and MPR adjusted for predictive variables. Negative binomial regression, was used to predict pulmonary exacerbations treated with IVAB or both IVAB and oral antibiotics after adjusting for baseline ppFEV1 over the same period. Results: In total 84 patients (mean [SD] age 29.8 +/- 8.8, median [IQR] ppFEV1 63 [51-83]) were prescribed azithromycin (84%), dornase alpha (65%) or both (49%) for the 24-month period. The mean (SD) MPR rate for azithromycin and dornase alpha was 0.61 +/- 0.29 and 0.57 +/- 0.29 respectively and the composite MPR was 0.58 +/- 0.27. Composite MPR was not a significant predictor of the change in slope of ppFEV1. A significant relationship between baseline ppFEV1 and composite MPR (p = 0.049) was observed after adjusting for age, BMI and gender at baseline. The composite MPR was not found to be a predictor of antibiotic courses. Conclusion: Adherence to dornase alpha and azithromycin was poor, and was negatively associated with baseline ppFEV1 but did not predict a change in ppFEV1 over 24 months.