Abstract
Background:
In susceptible individuals, exercise can be a potent trigger of bronchoconstriction resulting in symptoms which are commonly diagnosed as exercise-induced asthma (EIA). Empiric trials of inhaled corticosteroids (ICS) are often employed as treatment in preventing EIA symptoms, yet there is marked variability in treatment response. This heterogeneity may be explained by the differing pathological mechanisms which predispose to exercise-induced bronchoconstriction (EIB). These include airway inflammation, which itself is heterogeneous. Patients with eosinophilic airway inflammation, compared to noneosinophilic, demonstrate greater protection against EIB with regular ICS therapy. Additionally, the degree of sputum eosinophilia correlates with the severity of EIB. Exhaled nitric oxide (FENO) is a non-invasive surrogate biomarker for eosinophilic airway inflammation. Increased levels of FENO are associated with the presence and severity of EIB, and in patients with non-specific respiratory symptoms can predict steroid responsiveness. Investigating the potential ability of FENO measurements to identify patients with EIA symptoms likely to have a favourable response to ICS would reduce empiric prescribing, and is therefore clinically important.
Hypothesis:
Patients with EIA symptoms and high FENO are more likely to respond to ICS treatment, compared to those with low FENO
Aims:
1. Calculate the predictive utility of FENO measurements in patients with EIA symptoms and airway hyper-responsiveness (AHR) for response to ICS
2. Compare the effectiveness of ICS in the management of patients with EIA symptoms with low versus high FENO
3. Confirm that pre-treatment measurement of FENO is an important way to approach the management of patients with EIA symptoms.
Methods:
Patients with EIA symptoms and AHR to mannitol and/or exercise challenge were enrolled. A randomised, crossover, placebo-controlled trial of budesonide 800μg b.d was undertaken. Each treatment period was one month in duration, with an intervening two week washout. Patients were allocated to a low or high FENO group based on their pre-treatment off steroid measurement, using 45ppb as the cut-point. The following endpoints were measured at baseline and after each treatment arm: FENO, spirometry, AHR to mannitol and exercise challenges, Asthma Control Questionnaire score, and Borg Dyspnoea Score.
Results:
Forty five symptomatic patients were screened and seventeen fulfilled the eligibility criteria. FENO had a high predictive utility for steroid responsiveness (ROC AUC=0.833). The optimum cut-point for FENO to predict steroid responsiveness in this population was 41.0ppb with corresponding sensitivity, specificity, positive and negative predictive values of 78.6%, 66.7%, 91.7% and 40.0% respectively.
Analyses by FENO stratification revealed there were no significant improvements in any of the measured endpoints following budesonide in the low FENO group, except for FENO itself (33.4ppb vs. 17.6ppb; p=0.006). In contrast, patients with a high baseline FENO, demonstrated a reduced FENO (76.5ppb vs. 36.1ppb; p=0.007) and reduced AHR to mannitol (PD15 to mannitol increased; 193mg vs. 443mg; p=0.010). Improvements in asthma control score and AHR to exercise challenge approached, but did not reach, significance (p=0.071 and 0.063 respectively).
Conclusions:
Patients with a high pre-treatment FENO demonstrated clinical improvements following treatment with budesonide, whereas those with a low FENO showed no improvement. Pre-treatment FENO measurements may be used to predict whether patients with EIA respiratory symptoms will respond to ICS treatment. These data support the use of this simple test to aid clinical decisions in the management EIA.