Reports of research work funded by grants prior to 2016
University of Otago Wellington
Predictions of Resting Energy Expenditure in Maori & Pacific Populations: The PREEMPt Study
B Corley, J Krebs, R Hall, R Carroll, M Weatheral
Centre for Endocrine Diabetes and Obesity Research
The PREEMPt study aims to quantify the limitations of predictions of resting energy expenditure in two New Zealand populations disproportionately affected by overweight and obesity. PREEMPt involves measuring whole body composition and resting energy expenditure. Body composition analysis is by means of densitometry, DEXA. The funding for these scans were generously awarded by Research for Life.
Resting energy expenditure measurements are by means of indirect calorimetry. The study of resting energy expenditure in populations requires high precision, validity and standardised testing conditions. In PREEMPt we are using a sophisticated calorimeter, the Promethion, a device that estimates energy expenditure based on measurement of the gaseous products of respiration; carbon dioxide, oxygen and water vapour. This device is located at the University of Otago Wellington, Centre for Translational Physiology. As PREEMPt is the first clinical trial to use the calorimeter, we undertook rigorous independent testing of the calorimeter prior to PREEMPt. This testing has recently been completed and the PREEMPt study is now underway.
In order to test the Promethion involves validation using inert gases. By simulating real oxygen consumption using nitrogen on multiple occasions and under a variety of conditions, we were able to determine how accurately it measures what it is intended to measure - oxygen depletion. The results of these validation studies not only indicate how well the Promethion performed, but also what degree of error we can expect when using it to estimate resting energy expenditure. By modifying our technique, we minimized this error. Currently the validity of the Promethion when quantifying oxygen depletion is 99% in the canopy hood configuration. Prior to validation testing and refinement the validity was less than 90%.
In addition to validity testing of the Promethion device we also performed a first test retest reliability study. Retest reliability is a measure of how likely is it that the Promethion readings will differ when measurements are obtained from a single participant across multiple visits.
The retest reliability of the Promethion calorimeter will be crucial for any prospective studies with a repeated measures design. A test-retest reliability study performed on ten healthy volunteers in preparation for PREEMPt has been completed. In order to conceptualise precision, the parameter “minimal real difference” was used. This parameter describes the minimum difference in O2 or CO2 measurements from one visit to the next that is necessary in order to be 95% certain a true difference in these gases actually exists. On initial testing the minimal real difference was 14% for CO2 and 18% for O2. This was considered poor retest reliability. Several modifications to the testing procedure have been implemented. Ideally we would aim to have a precision of 98% or greater in the absence of biological variation and a precision of 90% or greater in healthy stable human participants. A repeat evaluation of re-test reliability is currently underway.
We are now continuing with the PREEMPt study. The improved accuracy of our readings will enable us to maximise the yield from each participant studied. To date we have recruited 10 of 20 participants, three of whom have completed their study visits. It is envisaged that recruitment will be completed by August with data collection completed by September. The study is due to be concluded prior to December of this year.
Once generated, the results of the PREEMPt study will allow us to compare prediction of resting energy expenditure with more accurate measurements of energy expenditure. These results will be important for establishing the variance of resting energy expenditure in a sample of the NZ Maori and Pacific population. This will be the first step in an evaluation of the accuracy of resting energy expenditure equations and will inform future studies incorporating indirect calorimetry in these populations.