A dietary survey was carried out on an Antarctic base over a period of 6 to 12 months on twelve individuals. Three methods of determining individual food and energy intake were used-weighing and recording of food as eaten in combination with food tables, dietary recall, and bomb calorimetry of duplicate meals. Use of weighed food intakes and food composition tables underestimated energy intake by 7% when compared to analysis of duplicate meals by bomb calorimetry. One week was found to be the most practical period over which intake should be measured to determine "habitual" food intake.

Two different makes of bioimpedance spectrometer (UniQuest-SEAC SFB-3 and Xitron 4000B) were used for a series of measurements on volunteers and patients in intensive care. Although each machine was accurate over the frequency range 5 to 500 kHz when bench tested on model resistor-capacitor circuits, significant differences in their recorded impedance parameters appeared when used in vivo, especially on intensive care patients. A series of laboratory tests was performed on each machine simulating the situation in vivo to identify possible reasons for these differences. Whilst stray capacitance in the environment was identified as the major contributor to variability in high-frequency performance, interaction between electrode impedance and lead positioning was also a factor. The observed phase shift with frequency or time delay (Td) used in the Xitron modeling software appears to be the result of a time constant caused by stray capacitance and so is unlikely to have any biological meaning. Significant differences in the in vivo numerical values produced by bioimpedance spectrometers may be attributed to instrument design, data processing and, in particular, the clinical environment.

An energy balance survey was conducted on an Antarctic base on 12 individuals for periods varying between 6 and 12 months. Energy expenditure was estimated using two activity/dairy card techniques and mean 24-hr heart rate used in conjunction with a variety of derived heart rate/energy expenditure regression equations. The accuracy of the various techniques was determined by comparison with estimates obtained using energy intakes and changes in body fat. No method of measuring energy expenditure was accurate enough to predict fat gain or loss. Errors in individuals were large and random but the mean error in the estimation of energy expenditure of the community was reasonable. When using the dairy card there was no difference between using measured values for energy costs of specific activities and using values taken from the literature. Using heart rate was no less accurate than the diary card if energy expenditure was predicted from a regression line derived from heart rate and log energy expenditure, and the technique was more acceptable to the subjects.

Indirect calorimetry estimates energy expenditure from measurement of respiratory gas exchange volumes. This paper considers the design and evaluation of an indirect calorimeter, the Europa GEM, suitable for use in nutritional research. The calorimeter is of the ventilated hood, flow-through type and is intended for use with spontaneously breathing patients. Our aim was to develop an accurate, flexible instrument with a high level of automation. Performance was assessed in a laboratory simulation using reference gas injections (n = 24) producing a mean error of 0.3 +/- 2% in oxygen consumption (VO2), 1.8 +/- 1% in carbon dioxide production (VCO2) and 1.4 +/- 1.5% in respiratory quotient (RQ). In order to investigate the effect of FeCO2 on error multiplication a further subdivision (n = 8) of tests at FeCO2 = 0.5%, 0.75% and 1% was made by modulating the air flow through the hood. However, the predicted increase in system accuracy with increasing FeCO2 was not apparent in practice.

No abstract