Abstract

With neurodevelopmental outcome of very low birthweight (VLBW) infants being adversely affected by inadequate nutrition during the first few weeks of life, there is an urgent need for more specific nutritional data on the sick VLBW ventilator dependent infant. The development of a new mass spectrometry gas analysis indirect calorimetry system which is non-invasive and can operate over several hours or days is described. Technical evaluation of each of the components of the system indicates a total random error of less than 5%. Systematic error was determined using gas infusions which simulated carbon dioxide production and oxygen consumption. The relative error in the measurement of carbon dioxide production was less than or equal to 1.5% (coefficient of variation (CV) 6.0%)) with carbon dioxide infusion rates ranging from 3.86 to 13.98 ml/min. The relative error in oxygen 'consumption' was less than or equal to 4.3% (CV 2.8%) for infusions of oxygen at rates of 7.5 to 14.80 ml/min. With nitrogen infusions simulating oxygen consumptions of 2.0 and 5.5 ml/min the relative error in the calculated nitrogen infusion was 1.5% (CV 4.1%) and 1.4% (CV 5.7%) respectively. Clinical studies on 10 infants demonstrated a mean energy expenditure of 161.7 kJ/kg/day and a respiratory quotient in excess of 1.0. The energy expenditure of ventilated VLBW infants may be less than previously indicated and the energy mix and nitrogen content of parenteral nutrition regimens recommended for these infants may be inappropriate.