Page 28 - ADULT-ONSET ASTHMA PREDICTORS OF CLINICAL COURSE AND SEVERITY

ADULT-ONSET ASTHMA – PREDICTORS OF CLINICAL COURSE AND SEVERITY
FACTORS INFLUENCING ASTHMA SEVERITY IN ADULTS
ASPECTS OF ASTHMA SEVERITY
Classi ca on of asthma severity52 is based on the minimum level of asthma medica on needed to prevent the disease from becoming uncontrolled or which remains uncontrolled despite high dose therapy 53-55. Uncontrolled asthma can be either: poor symptom control (e.g. asthma control ques onnaire (ACQ)-score >1.5 56), frequent or severe exacerba ons or increased air ow limita on 57. Compared to childhood asthma, adult-onset asthma tends to be more severe 5, 58. Studies inves ga ng the prognosis of asthma severity in general and the di erent aspects of uncontrolled asthma will be addressed here, except for air ow limita on, which was addressed in the previous paragraph.
PREDICTORS OF INCREASE IN ASTHMA SEVERITY
Several predictors of increased asthma severity in adult-onset asthma have been described, for instance smoking, increasing age, high symptom scores and low lung func on. A er a follow-up of 70 months in a cohort of 250 pa ents, a low FEV1 and increasing age were related to increased asthma severity (based on GINA severity grading). Pa ents with severe asthma at the end of follow-up also showed a signi cant increase in BMI during the study 59. This study did not nd an associa on between asthma severity and smoking, whereas others did. Polosa et al. found a dose-response rela on between pack years and the development of moderate-severe asthma in rhini s pa ents, with an odds ra o (OR) 2.9 (95% CI 1.1-7.5) for 11-20 pack year and OR 5.6 (95% CI 1.4-21.7) for >20 pack years 60. In our own cohort of 200 adults with a recent diagnosis of asthma we found the number of pack years and lower FEV1/FVC as univariate predictors of an increase in asthma severity a er 2 years follow-up. The only independent predictor in the mul variate analysis was pack years, where every ten pack years smoked gave an OR 1.4 (95% CI, 1.02-1.91) for developing more severe asthma 61. Another study with a mixed popula on of adults with asthma (both childhood and adult-onset asthma) reported a high symptom score, low FEV1 and ICS use at baseline to be strongly related to severe asthma a er 9 years follow-up. Furthermore the strongest prognos c factors in a mul variate model were chronic cough or worsening of cough, high IgE level or hospitalisa on history. The authors found no rela on between age of onset and asthma severity, hence results might be extrapolated to adult-onset asthma pa ents separately 62.
Finally, the predic ve ability of phenotypes found by cluster analysis is limited. Cluster analysis has been used to iden fy di erent asthma phenotypes in several cross-sec onal studies 5, 23, 63-65. One prospec ve study inves gated the prognos c value of these clusters in a mixed asthma popula on and showed that severe asthma clusters were not able to predict disease course with regard to asthma control, exacerba ons rate or treatment requirements 66. Whether pa ents s ll belong to the same phenotypes a er a certain period di ers between asthma phenotypes 40, 67. Pa ents with adult-onset asthma were more likely to belong to
26