Indirect calorimetry is a tool that allows for the non-invasive assessment of metabolism through the
measurement of the consumption of oxygen and the production of carbon dioxide. However, its use in intensive
care patients presents technical limitations and difficulties. This study presents a frequency analysis of the
oxygen consumption and carbon dioxide production signals as an alternative tool for assessing metabolism
and provides qualitative information relating the spectral densities of these signals with the type of substrate
and metabolic pathways in intensive care patients. This study included 10 patients from the Intensive Care
Unit (ICU) after a 12 h fasting period, of ages ranging between 42 and 80 years old, receiving mechanical
ventilatory assistance (in assisted-control mode). Oxygen consumption, carbon dioxide production and
electrocardiographic signal tendencies were registered after the continuous administration of a 500 ml infusion
of a 5% glucose solution in a 30 min window and in another 30 min window after removing the infusion. The
spectral densities of the heart rate variability, oxygen consumption and the production of carbon dioxide were
obtained using a periodogram. The characteristic bandwidths and their frequency and potency mass centers
were gotten from the spectral densities for oxygen consumption and the production of carbon dioxide; the
Sympathetic-Vagal Index (SVI) was gotten from the spectral density of the time series of RR intervals.
Characteristic bandwidths were obtained at both low and high frequencies for oxygen consumption and
the production of carbon dioxide. Using ANOVA variance analysis, they were compared with the four time
windows, with the SVI and the 15-30 and 45-60 min windows corresponding to the times with greater and lesser
amount of glucose in the patient. No significant statistical differences were found. However, this study
provides relevant qualitative information suggesting a relationship between spectral density bandwidths related
to oxygen consumption and carbon dioxide production to metabolism. The results of this study are relevant
and suggest that there are characteristic bandwidths in the spectral densities of oxygen consumption and
carbon dioxide production related to metabolic pathways. These signals can be useful as auxiliary tools for the
treatment of critical patients.