Reliability of Measurements Using Innocor Beath by Breath Analyser During a Maximal Exercise Test in Cystic Fibrosis Patients.

Bayfield KJ, McGovern M, Simpson AJ, Embley M, Cunningham S, Davies JC, Alton EW, Innes JA

Thorax, Vol 69, S2, Abstract P207

British Thoracic Society Winter Meeting, London, 2014


Cardiopulmonary exercising testing (CPET) is con- sidered the gold standard to study exercise capacity as an end- point in clinical trials. Originally the UKCFGTC used the shuttle walk test for exercise capacity measurement but this proved inappropriate for mild, fit cystic fibrosis (CF) patients in our trial cohort (FEV1 50–90% predicted). The Innocor device uses photoacoustic gas detection technology and offers metabolic meas- urement but has not previously been validated for CPET in CF. Aim To compare the Innocor with known reliable CPET machines to see if it is suitable to take forward into a multi dose clinical trial of gene therapy.


12 CF patients (7 Male, 14–47 years) participated in the study recruited from London and Edinburgh sites. They performed two incremental cycle ergometer exercise tests to exhaustion (adapted Godfrey protocol) with breath by breath analysis assessed using a reference system (Jaeger Masterscreen PFT, London; Pulmolink Medisoft, Edinburgh) or the Innocor device. All tests were randomly ordered, completed at least 24 h apart, with no more than two week’s separation.


VO2max and VE max were comparable between the Innocor and reference systems (p = 0.1790 and p = 0.7642 respectively; paired t tests). For VO2max, Bland Altman analysis showed the mean difference [Reference equipment-Innocor] was -0.026 l/min and the 95% confidence interval was -0.27 to 0.22 l/min (see Figure). In our experience the Innocor heart rate (HR) recording (derived from saturation monitor) was unreliable; reading low during exercise compared to ECG-derived HR.


This small study confirms the Innocor device can produce measures of VO2max comparable (95% confidence interval) with standard calibrated exercise systems in CF patients with mild to moderate lung disease. We found the method for Innocor to derive HR (pulse oximetry) was not reliable compared to reference ECG especially during heavy exercise. We were subsequently able to overcome this problem by interfacing the Innocor device with a separate electrocardiographic heart rate monitor.