(276) Effects of quantity and quality of daily airway clearance treatments on lung function in children and young people with cystic fibrosis: Results from Project Fizzyo

Background: Many children and young people with CF (CYPwCF) get advice about using different positive expiratory pressure (PEP) devices to clear sticky mucus from their lungs. However, little is known about how the quantity or quality of these daily airway clearance treatments (ACTs) at home impact lung function over time.

Methods: Project Fizzyo was a longitudinal remote monitoring observational study, which used electronic pressure sensors to record real-time breath-by-breath data during usual ACTs undertaken at home over 16 months [1]. All participants (6-16y) used one of four PEP ACT devices (Acapella Choice, Aerobika, AstraTech PEP, Pari PEP) at least once a day as part of their routine physiotherapy.

The number of ACTs per day and breaths per ACT were recorded for each CYPwCF. Where data from prescribed ACTs were expected but not received, the session was labelled ‘no ACT’. ACTs were labelled as ‘good’ quality if they met the criteria, with generous margins, for common advice on PEP device use: average mid-expiratory breath pressure between 5-25cmH₂O and expired breath length between 1.5-3seconds. If an ACT was recorded, but these criteria were not met, then the ACT was labelled ‘poor’ quality.

ACT data were reconciled with longitudinal clinical data for each CYPwCF and individualized polynomial regressions for FEV₁ %predicted were used to match the time intervals of the ACT data. A linear mixed-effects model was used to evaluate the effect of different components of ACT: quantity (number of ACTs, breaths per ACT) and quality (expired breath pressures and lengths) on FEV₁. The model accounted for disease severity, age, intravenous antibiotics and modulator therapy use.

Results: 145 CYPwCF were recruited from 3 UK CF centres and 50,892 full ACTs from 137 CYPwCF were recorded and analysed. Of 76,308 prescribed ACTs, a total of 16,681 were labelled as ‘good’ ACTs (22%), 34,211 were ‘poor’ ACTs (45%) and 25,416 were labelled ‘no ACTs’ (33%).

In isolation, there was no measurable impact of ACT quantity on FEV₁ over time, such that those performing more ACTs or breaths per ACT had comparable FEV₁ to those with fewer ACTs or breaths per ACT. However, there was a significant impact of ACT quality on FEV₁, with those performing ‘good’ ACTs having significant, clinically important effects on FEV₁ (extrapolated >5%predicted per year), compared with those doing ‘no ACTs’ or ‘poor’ ACTs. In addition, quantity mattered when analysed in combination with quality: those with accumulatively more ‘good’ quality ACTs did significantly better than those with fewer. Conversely, those with more ‘poor’ quality ACTs or ‘no ACTs’ were worse off than those with fewer.

Conclusions: Regular ‘good’ quality ACTs were more important than ACT quantity in providing significant FEV₁ benefits, even after adjusting for disease severity, age, antibiotics and modulator therapies. However, fewer than a quarter of CYPwCF regularly achieved ‘good’ quality ACTs, suggesting many CYPwCF may not realise their techniques are poor. These new findings have important implications for the way clinicians teach, measure and monitor effective ACTs in the future, in order to ensure benefits are optimised and deleterious effects of poor ACTs are avoided.

Acknowledgements: Supported by UCL Rosetrees Stoneygate prize (M712), CF Trust Award (CEA010), UCL Partners and HEFCE Higher Education Innovation Fund (KEI2017-01-04) and the NIHR GOSH BRC.

References: [1] Raywood E et al., BMJ Open 2020;10:e039587 doi:10.1136/bmjopen-2020-039587