Abstract
Background: Lung function decline over time is an important variable for patients with severe/difficult asthma and it is unclear how granulocytic cell counts in sputum are associated with this variable as few longitudinal studies include both biomarkers.
Aims and objectives: To analyse lung decline, post bronchodilator FEV1 was recorded over time in order to determine the best fitting mixed effect model to determine the role of granulocytes.
Methods: Data was collected form the Glenfield Hospital Severe asthma clinic, the mean duration of follow up and number of visits were 6 years and 2.7/years. A mixed effect model was applied to the data. Using individual patient mean and standard deviation over time, a statistical mixture/cluster analysis was implemented.
Results: For the best fitting mixed effects model FEV1 decline was -25.7ml/year, p=0.0001. The significant independent fixed effects included exacerbations, age of onset, log eosinophils (p<0.001). Three clusters were found in the log eosinophil cluster analysis. Cluster 1 described a low grade eosinophilic group. Cluster 2, a hyper eosinophilic group and cluster 3, a neutrophilic/non-eosinophilic group. The clusters differed in their frequency of exacerbation/decline.
Conclusions: Eosinophils were found to be a significant predictor for FEV1 decline. Clustering eosinophil variables found that patients are either consistently eosinophilic over time, consistently non eosinophilic over time or have a large amount of eosinophilic variation.
- © 2011 ERS