C Slater1, T Preston, L T Weaver. 1. Division of Developmental Medicine, University of Glasgow, Yorkhill Hospitals, Glasgow, UK. C.Slater@clinmed.gla.ac.uk
Abstract
BACKGROUND: 13C-breath tests are noninvasive tools to measure gastrointestinal function and nutritional interventions. Calculation of percentage dose recovered of 13C in exhaled breath requires knowledge of CO2 production rate (VC02). A resting value is usually assumed, but this can underestimate VC02 because subjects are unlikely to remain at rest during tests that last for many hours. There is a need for a method to estimate nonresting VC02 during 13C-breath tests. OBJECTIVE: To calibrate a heart rate monitor to continually estimate VC02 during 13C-breath tests. DESIGN: Proof of concept study. SUBJECTS: Eight healthy adults, 10 healthy children and six children with cystic fibrosis. METHODS: Heart rate and VC02 were measured simultaneously at resting and nonresting levels. A new calibration method (smoothing heart rate and fitting a sigmoid function) was compared with published methods. A [ 3C]acetate breath test was used to demonstrate the range of physical activity during breath tests. RESULTS: The new calibration method was more accurate than existing methods (mean bias -0.0002%, 95% confidence interval (CI) -0.0007, 0.0003% of the mean measured VC02). Smoothing heart rate gave a more precise estimate of VC02 and a more accurate estimate of resting energy expenditure (mean bias -0.09, 95% Cl -0.22, 0.05 mmol CO2 min-' m-2 body surface area) than using raw data (mean bias -0.21, 95% Cl -0.38, -0.04 mmol CO2 min' m-2 body surface area). Physical activity level ranged from 1.0 to 2.5 in children, and 1.0 to 1.5 in adults. CONCLUSION: Use of smoothed HR with a sigmoid function provides an accurate method of estimating nonresting VC02 during 13C-breath tests.
BACKGROUND:13C-breath tests are noninvasive tools to measure gastrointestinal function and nutritional interventions. Calculation of percentage dose recovered of 13C in exhaled breath requires knowledge of CO2 production rate (VC02). A resting value is usually assumed, but this can underestimate VC02 because subjects are unlikely to remain at rest during tests that last for many hours. There is a need for a method to estimate nonresting VC02 during 13C-breath tests. OBJECTIVE: To calibrate a heart rate monitor to continually estimate VC02 during 13C-breath tests. DESIGN: Proof of concept study. SUBJECTS: Eight healthy adults, 10 healthy children and six children with cystic fibrosis. METHODS: Heart rate and VC02 were measured simultaneously at resting and nonresting levels. A new calibration method (smoothing heart rate and fitting a sigmoid function) was compared with published methods. A [ 3C]acetate breath test was used to demonstrate the range of physical activity during breath tests. RESULTS: The new calibration method was more accurate than existing methods (mean bias -0.0002%, 95% confidence interval (CI) -0.0007, 0.0003% of the mean measured VC02). Smoothing heart rate gave a more precise estimate of VC02 and a more accurate estimate of resting energy expenditure (mean bias -0.09, 95% Cl -0.22, 0.05 mmol CO2 min-' m-2 body surface area) than using raw data (mean bias -0.21, 95% Cl -0.38, -0.04 mmol CO2 min' m-2 body surface area). Physical activity level ranged from 1.0 to 2.5 in children, and 1.0 to 1.5 in adults. CONCLUSION: Use of smoothed HR with a sigmoid function provides an accurate method of estimating nonresting VC02 during 13C-breath tests.