Article Text

Download PDFPDF

P254 Validation of telemedicine spirometry
  1. R Peat,
  2. P Szymczyk,
  3. D Russell,
  4. D Nazareth,
  5. M Shaw,
  6. MJ Walshaw
  1. Liverpool Heart and Chest Hospital, Liverpool, UK


Introduction Spirometry is considered one of the most important tests for the monitoring of CF and is commonly performed at each clinic visit. With the emerging appetite for telemedicine as a viable alternative to traditional hospital outpatient visits, the suitability of an inexpensive handheld spirometer was assessed for home use.

Methods We compared the accuracy of a handheld spirometer (COPD6, Vitalograph, UK) with a standard laboratory based spirometer (Spirostik, Geratherm, Germany) in 41 consecutive adult CF patients (mean age 39 years [SD 11.4], mean predicted FEV1 60% [23], 16 male) attending our regional CF centre. All were experienced at performing spirometry. They were randomly assigned to either device and carried out 3 manoeuvres before repeating the test session the alternative spirometer. Both testing sessions were supervised and performed in the same clinic room. Patient coaching and feedback was withheld from the handheld spirometry session. Intra-session analysis of flow volume tracings was carried out for lab based testing sessions.

Results Acceptable paired results were obtained in 36 patients (88%). The mean Spirostik FEV1 was 2.14L (variability 0.06 [2.3%]) and COPD6 2.02L (variability 0.09 [4.4%]); mean difference between devices 0.12L (5.6%). Eight of 36 (22%) results were lower on the handheld device.

Analysis of the flow volume tracing was required for 6 patients (16%) to either improve technique (3), or QC/eliminate artefact (3): where artefact was not eliminated, FEV1 was overestimated by 0.08 to 0.09L (∼4.5%).

Conclusions The handheld device was easy to operate and accuracy and repeatability was considered to be acceptable for a telemedicine/home use application. However, differences between traditional clinic/laboratory spirometry results and telemedicine spirometry results were observed. When comparing spirometry results clinicians should have an appreciation of expected differences. Additional variables may exist where patients are unsupervised in a home setting.

Abstract P254 Figure 1

Bland-Altman plot

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.