Background Non-adherence to inhaled combination therapy (ICT) is a major contributor to poor control in difficult asthma1; however identifying non-adherence in this population is challenging. We have shown that suppression of fractional exhaled nitric oxide (FeNO), following directly observed inhaled corticosteroid can identify non-adherence2; it has previously been suggested that a single high FeNO measurement can identify non-adherence in difficult asthma.

Methods We performed a retrospective analysis of patients attending a difficult asthma clinic between January 2007 and December 2010. Prescription refill data, patient demographics, FeNO and lung function were collated. ICT prescription refill ratio % was calculated as the number of doses refilled/number of doses prescribed over 6 months × 100.1 Non-parametric correlation analysis was performed. Tests for non-adherence defined as ICT prescription refill cut-offs of <80% and <50% were assessed.

Results One hundred and forty-six patients underwent systematic evaluation during the time period. Patient characteristics are displayed in Abstract S10 table 1. FeNO level and ICT prescription refill ratio did not correlate significantly (r=−0.11, p=0.2). For FeNO >45 ppb, there was a trend with ICT prescription refill ratio of <80% (p=0.06); significant in patients not prescribed oral steroids (n=99, p=0.016). No relationship existed between FeNO >100 ppb at both prescription refill ratio cut-offs. Using FeNO >45 ppb to define non-adherence, negative (NPV) and positive predictive values (PPV) were: 66% and 61% for <80% ICT prescription refill ratio; 71% and 33% for <50% prescription refill ratio. Using FeNO >100 ppb, NPV and PPV were: 55% and 52% for <80%; 69% and 32% for <50% prescription refill ratio.

Conclusion While there is a weak relationship between a single point-in-time FeNO measurement and non-adherence to ICT in difficult asthma, this cannot be used as an alternative to our previously described FeNO suppression test.