Introduction and Objectives At the start of the COVID-19 pandemic, paediatric asthma patients attending our clinic (Royal Brompton Hospital) were given NuvoAir spirometers to monitor lung function at home. Previous studies have shown good agreement between home and clinic spirometers however these studies were carried out in the hospital setting, with spirometry supported by clinical staff. There have been no studies examining the real-world reliability of this clinical tool. We hypothesised that NuvoAir was a valid alternative to a conventional clinic spirometer (Vitalograph Alphatouch). This study aimed to investigate the technical reliability and validity of NuvoAir spirometer in paediatric asthma patients, in comparison to Vitalograph.

Methods We carried out a retrospective longitudinal study involving 43 patients. Three pre-pandemic Vitalograph sessions and three NuvoAir sessions were recorded for each patient. Spirometry grades were manually assessed according to the American Thoracic Society (ATS)/European Respiratory Society (ERS) guidelines. FEV₁ and FVC measurements were converted to z-scores using the Global Lung Function Initiative equations. The data were normally distributed and consequently analysed using a paired Student’s T-test, Pearson’s Correlation Coefficient and Bland-Altman plots.

Results Spirometry from 43 patients (mean age 13.9 years) was analysed. NuvoAir spirometry grade distributions were comparable to those from Vitalograph; 70.5% and 79.1% of grades were deemed acceptable according to ATS/ERS guidelines. Mean FEV₁ z-scores (mean ±standard deviation) were -1.05 ±1.39 and -1.11 ±1.34 (p=0.650), whilst FVC z-scores were 0.01 ±1.20 and -0.33 ±1.20 (p=0.019) for NuvoAir and Vitalograph respectively. Pearson’s correlation coefficients were r=0.78 (p<0.0001) and r=0.59 (p<0.0001) for FEV₁ and FVC respectively. Bland-Altman 95% limits of agreement were -1.71 to 1.59 for FEV₁ z-scores and -2.24 to 1.52 for FVC z-scores with estimated bias levels at -0.06 and -0.36 respectively.

• Download figure
• Open in new tab
• Download powerpoint

Abstract P88 Figure 1

(a) Bland-Altman plot depicting mean and difference in FEV1 z-scores from Vitalograph and NuvoAir spirometers (n=43). The maroon dashed line indicates the estimated bias level, whilst the maroon dotted lines indicate the 95% limits of agreement. (b) Bland-Altman plot depicting mean and difference in FVC z-scores from Vitalograph and NuvoAir spirometers (n=43). The maroon dashed line indicates the estimated bias level, whilst the maroon dotted lines indicate the 95% limits of agreement

Conclusion Our real-world study demonstrates that self-performed NuvoAir home spirometry is a reliable clinical tool and produces valid lung function values when compared to clinic spirometry. Home spirometry using the NuvoAir device supports a hybrid model of face-to-face and virtual consultations and is a useful addition to remote monitoring tools as telemedicine continues to evolve.