Page 127 - ADULT-ONSET ASTHMA PREDICTORS OF CLINICAL COURSE AND SEVERITY

PREDICTORS FOR THE DEVELOPMENT OF PROGRESSIVE SEVERITY IN NEW-ONSET ADULT ASTHMA
RESULTS
BASELINE CHARACTERISTICS
Amongst the 200 pa ents ini ally included in the study, 72 subjects (36%) did no longer par cipate a er 2 years of follow-up. Most pa ents were lost-to-follow-up due to a lack of me (n=33), co-morbidi es (n=9), distance (n=5), language barrier (n=1) or unknown reasons (n=24). Baseline characteris cs of included pa ents and lost-to-follow-up pa ents are shown in Table 1. Included pa ents were more o en male (p=0.03), used higher doses of inhaled cor costeroids (p=0.05) and were more o en (ex-)smokers (p=0.03), compared to lost-to- follow-up pa ents.
ASTHMA SEVERITY
Seventeen pa ents (13.3%) showed an increase in asthma severity (increase in GINA score), whereas 53 (41.4%) of pa ents showed improvement (decrease in GINA score) a er two years as compared to baseline. Baseline characteris cs of these di erent groups are shown in Table 2. The pa ents whose asthma deteriorated used lower doses of inhaled cor costeroids, had milder asthma at baseline (p<0.01) and had smoked more pack years (p=0.02). These pa ents also tended to be more o en current smokers (p=0.078). There were no other signi cant di erences in baseline characteris cs including lung func on and in ammatory parameters. Table 3 presents the within-subject changes in the same variables a er two year follow-up, showing a signi cant drop in ACQ (p=0.049) despite increased ICS dose (p<0.001) in the deteriora on group.
PREDICTORS OF INCREASE IN ASTHMA SEVERITY
At rst, univariate logis c regression with mul ple variables was performed. Table 4 shows that FEV1/FVC and smoked pack years were signi cant predictors of an increase in asthma severity (increase in GINA score). An odds ra o (OR) of 1.5 (95% con dence interval (CI) 1.10- 2.02) was found for every 10 pack years a pa ent had smoked. When using a cut-o point of more than 20 pack years, an OR of 3.6 (95% CI 1.22-10.76) was found. Having a lower FEV1/FVC at baseline was predic ve for deteriora on as well. Every 10% lower FEV1/FVC was associated with an OR of 1.6 (95% CI 1.03-2.56). A cut-o point of FEV1/FVC <95% of the predicted value (mean of popula on) was associated with an OR of 3.9 (95% CI 1.30-11.98).
In order to calculate adjusted OR, the signi cant variables of the univariate analysis were ed into a logis c regression model. This showed that the only independent predictor for deteriora on of asthma severity was the number of smoked pack years with an OR of 1.4 (95% CI 1.02-1.91) for every 10 pack year smoked. A low FEV1/FVC at baseline was no longer signi cantly associated with asthma deteriora on (see Table 5).
125