Energy Expenditure

EE: Narrative Reviews: Theoretical and Accepted Practice Applications (2005)

Brandi LS, Bertolini R, et al (1997).. Vo NM, Waycaster M, et al evaluated the effect of caloric overfeeding with high carbohydrates (based on RQ values measured by IC) on morbidity and mortality in 24 postoperative critically ill pt who received parenteral nutrition for a total of 1088 d.  This high calorie group (with RQ values >0.95) received 1.5 of measured RMR, while the normal caloric group (with RQ values <0.95) received 1.0.  Septic episodes and mortality were greater in the overfed group vs. the group receiving calories are equal to measured energy expenditure. [Source:  Vo NM, Waycaster M, Acuff RV, et al.  Effect of postoperative carbohydrate overfeeding.  Am Surg. 1987;53:632. Poster presentation: sample size of n=24] 

Elia M, Livesey G(1992).  O2 is a better predictor of energy expenditure than CO2 production.  When O2 consumption and CO2 exhalation are used o predict energy expenditure from a mixed diet providing fat, carbohydrate and alcohol (where alcohol provides 10% of energy), the estimated energy expenditure error is 0.464%.  If a constant value [i.e., Energy equivalent O2mix of 20.22 kJ/l (4.8 kcal/L)] is used to reflect a mixed diet composition of carbohydrate, fat, and protein when calculating energy expenditure in situations where the RQ is between 0.71 and 1.0, the maximum individual error is <4%.  Since the energy equivalent O2 for different fuels is much less variable than the energy equivalent of CO2, when different proportions of fuels are oxidized (e.g., fat-carbohydrate-protein-alcohol mixture), the variation in energy equivalent O2mix is 3- to 4-fold less than that for energy equivalent of CO2mix.  The percent variation in the Energy equivalent O2 between the highest and lowest value is 2.5%, while the variation in energy equivalent CO2diet and are 8.8%.  These errors are relatively small, even in the presence of nutrient imbalance, some workers simply report O2consumption as an index of energy expenditure

Standard values of RQ          




RQ Standard values

Energyequiv O2 kJ/l (kcal/L)

Energyequiv CO2 kJ/l(kcal/L)




21.10 (5.04)

21.12 (5.05)



19.61 (4.69)

27.62 (6.60)



19.48 (4.66)

23.33 (5.58)



20.33 (4.86)

30.49 (7.29)

Atypical fats





Medium chain triglycerides


26.64 (6.37)

19.69 (4.71)


Overall, a RQ below 1 indicates oxidation of fat and carbohydrate.  A RQ above 1.00 indicates carbohydrate oxidation and some conversion of carbohydrate to fat.

Feurer ID, Mullen JL (1986)   Measurements of the whole body RQ reflect net substrate oxidation at the time of measurement.  For example, 100% of oxidation from carbohydrate results in a net RQ of 1.0.  Ingestion of foods with mixed substrates (carbohydrate, protein, and fat) with subsequent oxidation results in an RQ of 0.85.  Finally, 100% of oxidation from fat results in an RQ of 0.70.  Finally, overfeeding resulting in lipogenesis may result in an RQ between 1.00-1.20.  In situations where measurements are taken when an individual is hyperventilating (i.e., in a non-steady state), RQ=>1.00.

Fung E (2000).  Respiratory quotients for anabolism/catabolism:

                Substrate              RQ

                Lipogenesis          1.0-1.2

                Carbohydrate        1.0

                Mixed substrate   0.85

                Protein   0.8

                Fat          0.7

                Ketogenesis          < 0.7

                Ethanol                  0.67

There are two assumptions.  The first assumption is that protein, fat, and carbohydrate are completely oxidized and little lipogenesis or ketogenesis are occurring.  Second, a direct relationship must exist between O2 and CO2 at alveolar and cellular levels.  The clinical instances where this assumption does not hold true is during periods of compensation for metabolic acidosis or alkalosis, hypoventilation, or hemodialysis.  [Analyst note:  Review article not abstracted due to not citing source for RQ values of anabolism and catabolism in Table 6 and limited discussion on RQ].

Jequier E, Acheson K, Schutz Y (1987).  Assumptions made when completing indirect calorimetry measures are that the urea pool and the bicarbonate pool remain stable or in equilibrium.  One hyperventilation occasion challenges this assumption.  For example, during hyperventilation, carbon dioxide is eliminated in excess of that produced by oxidative metabolism and there is a decrease in the bicarbonate pool.  CO2 elimination increases rapidly and exceeds concomitant O2 consumption resulting in a RQ>1.00.  The period of hyperventilation is followed by a compensatory period of hypoventilation during which metabolically produced CO2 is stored to re-equilibrate the bicarbonate pool.  Over a 45-minute measure, where the respiratory exchange measurements encompasses only one transient hyperventilation change, the mean values for CHO and fat oxidation will be correct.  However, considerable errors is incurred if a 5-minute measurement is taken when the hyperventilation began, during or at the end.  In contrast to CO2, VO2 changed very little over the 45 minute measure and thus had a smaller effect on energy expenditure calculations.

Specific metabolic situations where RQ is outside of range:

RQ <0.7

1.    Starvation

2.    Individuals receiving high-fat, low-carbohydrate hypocaloric diets,

3.    Alcoholics, and diabetes in the postabsorptive state.

4.   An increase in substrate flux through metabolic pathways that have low RQ, e.g.,                 gluconeogenesis and ketogenesis.

RQ > 1.0

1.   Net synthesis of fat from CHO (primarily measured in animals); but the importance of net lipogenesis in human nutrition is less evident.

(Sources:  Passmore R, Strong J et al, 1963.  The effect of overfeeding on two fat young women. Br J Nutr, 1963; Passmore R, Swindells YE.  Observations on the respiratory quotients and weight gain of man after eating large quantities of carbohydrate. Br J Nutr, 1963)

Matarese L (1997).  The RQ should be in physiologic range and consistent with the patient’s history and feeding.  [Analyst note:  Expert panel member recommendations and more current research reporting low sensitivity and reduced specificity limiting efficacy as an indicator of overfeeding and underfeeding (McClave SA, McClain CJ, et al, 2001) adapted Matarese L recommendations to use RQ for adjustment of feeding regimens].

Physiological considerations (i.e., oxidation) for RQ <0.71:

                Oxidation of ethanol and ketones

                Lipolysis (oxidizing fat stores)

                Diabetes mellitus


                High rates of urinary glucose excretion

Physiological with additional considerations (i.e., intervening factors such as illness or equipment changes) of RQ<0.71:


                Technical difficulties associated with actual measurement

Physiological causes (i.e., oxidation) for RQ >1.0:

                Excess CO2 production

                Hydrogen-ion buffering by bicarbonate-generating CO2

                Lipogenesis (building fat stores)

Physiological with additional considerations (i.e., intervening factors such as illness or equipment changes) of RQ >1.0:


                Metabolic alkalosis

                Postoperative period 6-8 hr after general anesthesia

                Adaptation to ventilator settings

[Sources:  Two animal models, 1 book; and primary studies:
Schultz Y, Ravussin E. Respiratory quotients lower than 0.70 in ketogenic diets.  Am J Clin Nutr. 1980;33:131C.;Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol. 1983;55:628-634; Weissman C, Kemper M, Askanazi J, Hymen AL, Kinney JM. Resting metabolic rate of the critically ill patient: measured versus predicted. J Anesthesiol. 1986;64:673-679.; Anderson CF, Loosbrock LM, Moxness KE.  Nutrient intake in critically ill patients: too many or too few calories?, Mayo Clin Proc, 1986)]

McClave SA, McClain CJ, Snider HL.  (2001).  The accuracy of RQ to be a measurement of substrate use cannot be substantiated and is of limited value to the clinician.  The physiologic range for RQ if from 0.67-1.3  [Cited source: Branson RS.  The measurement of energy expenditure: instrumentation, practical considerations, and clinical application.  Resp Care 1990;35: 640-659].  Values obtained outside of this range can only be generated through error.

Monitoring measured RQ throughout an IC measure provides a valid parameter of test validity.  [Cited sources:  Branson RS (as above) and McClave SA, Snider HL (abstracted below)].  An RQ elevated more than 1.0 in a patient receiving excess calories may indicate decreased respiratory tolerance, as increases in RQ were shown to correlate significantly with increased respiratory rate and decreased tidal volume (suggesting shallow rapid respiration in response to respiratory compromise).  The best utility of the RQ for the clinician is as a marker for test validity (confirming that the measured RQ is in the physiologic range) and as a marker of respiratory effect (detecting whether there are any changes in tidal volume and respiratory rate in response to over feeding).

McClave SA, Snider HL (1992).  When RQ is used for nutritional assessment, two assumptions are made:  the patient is in true steady state condition and all VCO2 measured reflects substrate utilization. 

 Individual Substrate Oxidation (RQ) values are:

                Glucose  1.0

                Fat          0.7

                Protein   0.8

By excluding protein, the nonprotein RQ (npRQ) provides a range of substrate utilization from 0.70 (indicating 100% fat utilization and 0% CHO utilization) to 1.0 (indicating 100% CHO oxidation and 0% fat oxidation).  A npRQ of 0.85 (i.e., midpoint) indicates 50% fat and 50% CHO oxidation.

(Cited Source:  Anderson CF, Loosbrock LM, Moxness KE.  Nutrient intake in critically ill patients: too many or too few calories? Mayo Clin Proc. 1986;61:853-858).

Alcohol or XX (unable to read word) metabolism may reduce npRQ below range to 0.67 and overfeeding with lipogenesis may increase the npRQ to 1.3.

(Cited Sources:  Feurer ID, Mullen JL. Bedside measurement of resting energy expenditure and respiratory quotient via indirect calorimetry.  Nutr Clin Prac. 1986; Weissman C, Kemper M et al,. Resting metabolic rate of the critically ill patients: measured versus predicted, 1986).

All potential sources of error should be scrutinized.  A 10% error in VO2 and a 10% error in VCO2 cause 7% and 3% errors, respectively, in the calculation of energy expenditure. (Source: Burzstein S, Elwyn DH, ASkanazi J.  Energy metabolism and indirect calorimetry in critically ill and injured patients.  Acute Care. 1988-1989:14-15:91-110).

Causes for discrepancies:

1.   Hyperventilation-through increase in CO2 from the work of breathing and              artifactual increased in VO2 from release of tissue stores. (Source:  Kinney, JM,            1987 Narrative review)

2.   Overfeeding leading to lipogenesis; exceptions occur in some patients were the    npRQ may plateau at 0.90 to 0.93 despite increasing levels of CHO infusion    above energy requirements.  This is thought to be related to hormonal counter-    regulatory mechanisms and defects in fat and CHO metabolism (Source;  Wolfe BM, Ruderman RL, Pollard A. Basic principles for surgical nutrition: Metabolic response to starvation, trauma, and sepsis.  In Dietal VM, ed. Nutrition in clinical surgery. Baltimore. Williams & Wilkins, 1985: 14-23.)

3.    Hypoventilation or underfeeding

4.    Diabetes mellitus

5.    Any time gluconeogenesis is present

6.    Sepsis after injury or surgery where glucose production andgluconeogenesis       continue despite hyperglycemia nd glucose infusion.

 7.   Hypothermia

 8.  Errors in calibration or leaks in the system

Peronnet I, et al. (1991.  The table of nonprotein RQ that is currently universally used is the one proposed in 1901 and modified in 1912.  Biochemical data on which the computations were made have been improved over the past decades; some inconsistencies can be found in the derivations of the values; therefore nonprotein RQ table should be slightly revised.  There is no apparent reason to continue using a table that contains systematic inconsistencies when it is as easy to use a table that is more consistent and in better accordance with biochemical and physical data currently accepted.  RQ associated with oxidation of glucose is 0.996 and complete oxidation of fatty acids is 0.704.  For a given RQ between the extremes (0.7036 to 0.9960), the table cannot be easily interpolated because the relationships between RQ and percent energy provided from glucose and fat oxidation and energy equivalent of oxygen are not linear.  [Analyst note:  See Evidence Analysis Worksheet for the updated RQ table].

Weissman C, Kemper (1995).  The ratio of CO2 elimination to O2 uptake reflects the net substrate utilization of the organism and is called respiratory quotient (RQ) when taken during rest.  Metabolism of each substrate results in different amounts of O2 consumed and CO2 produced:

RQ of individual substrates

                Substrate         RQ

                Alcohol            0.67

                Lipid                 0.71

                Protein              0.80

                Carbohydrate  1.00

RQs of 0.67 to 1.25 have been observed in humans, lower values reported in ketosis with ketonuria nd higher values in patients fed excess CHO and presumably converting it to fat.  (No reference given)  The RQ with its narrow physiological range – 0.7 to 1.25- with only a rare ketonuric patient dipping below 0.7, makes it a sensitive measure.