#17. A Comparison Between Indirect Calorimetry and Predictive Equations

Hi, I am Dr. Robert Bilkovski. Welcome to this podcast series on Indirect Calorimetry. In this episode, we will go over a comparison between indirect calorimetry and predictive equations.

The goals of this podcast are to share:

• The principles of indirect calorimetry and a high-level understanding of the Weir Formula
• The basis of predictive equations and their limitations
• Limitations associated with indirect calorimetry

In this next segment, we will go into more detail on how the various means energy expenditure can be assessed at the bedside. Moreover, the strengths and/or limitations of each method will be discussed.

To start, Resting Energy Expenditure can be determined through the use of predictive equations that use a combination of body weight, age, or height.

One of the most widely recognized and used predictive equations is the Harris Benedict equation, which uses a combination of body weight, height, and age to predict energy expenditure. This predictive equation was first introduced in the early 20^th century.

Other equations include The World Health Organization, the WHO-2 and the Penn State, just to name a few.

Of note, some predictive equations have been modified for use with pediatric patients¹.

Predictive equations do not incorporate patient-specific information beyond those mentioned earlier. In contrast indirect calorimetry utilizes the respiratory gas measurements, namely oxygen and carbon dioxide to determine Resting Energy Expenditure.

Indirect calorimetry incorporates these gas measurements into a formula called the Weir Formula in order to inform on energy expenditure.

The specifics of the Weir formula are too complex to describe in this podcast, but simply include VCO2, VO2 and the 24-hour urinary nitrogen measurement. However, urinary nitrogen represents roughly 4% of the total energy expenditure and is often excluded.

A final method to measure expenditure is via use of the Fick method.

This measurement requires the placement of a pulmonary artery catheter, which is invasive and limits widespread use.

The Fick method relies on the measurement of cardiac output using a principle called thermodilution combined with measurement of both arterial and mixed venous oxygen content.

The Fick method is beyond further discussion in this podcast, most notably, given that it requires a pulmonary artery catheter and has a sizeable inherent error. In fact, that error can be up to 15% due to the variation in cardiac output which occurs during a respiratory cycle².

Therefore, let us focus on predictive equations and indirect calorimetry. The convenience of predictive equations is readily apparent for energy expenditure determinations could be obtained without additional technology and interpretive skills, however, use in critical care environments has shown them to be sub-optimal when compared to indirect calorimetry³.

Similarly, studies conducted by Karlsson and Jotterand demonstrated that the predictive equations were inaccurate compared to indirect calorimetry in the elderly and pediatric populations, respectively^4,5.

In a largest study conducted comparing predictive equations and indirect calorimetry, with over 1400 patients assessed, Zusman showed that all equations tested had relatively poor correlation and agreement with indirect calorimetry.

The correlation ranged between 0.36 to 0.54, while agreement was between 0.3 and 0.5.

In total, eight predictive equations were assessed, and the percentage of error was at or greater that 20% amongst all equations studied.

The author concluded that: “both underfeeding and overfeeding are harmful and that optimizing nutrition to patient-specific needs is an urgent task” and “the optimal way to define caloric goals is ideally preferred using indirect calorimetry.”

The author flagged one important limitation of predictive equations when compared to the use of indirect calorimetry - the predictive equations failed to capture the dynamic metabolic changes that occurred during a patient's critical illness, where repeated measures of indirect calorimetry can better capture these changes over time.

These authors helped to validate that indirect calorimetry provides both timely and better assessment of nutritional needs, compared to predictive equations. It is important to note that indirect calorimetry has some limitations that may impact accuracy⁶.

The limitation to indirect calorimetry is categorized into four main areas that are - leaks, high FiO2, hemodynamic shifts and humidity in the breathing circuit⁷.

Leaks can occur in an array of places within the breathing circuit, including the endotracheal tube, the breathing circuit, and the ventilator itself.

There are inherent inaccuracies as a result of the Haldane transformation, which is used to calculate inhaled gas volumes from exhaled gas measurements, when the inspiratory fraction of oxygen is high for the difference between inspired and exhaled oxygen concentrations become very small. Thus, the indirect calorimetry measurements become more error-prone at increasing FiO2, notably above 70%;

The presence of moisture and humidity within the breathing circuit can have a negative impact on assessing volume measurements.

Lastly, hemodynamic shifts such as fluid challenges or hemodialysis can impact cardiac output, and in turn, energy expenditure determinations.

In closing, indirect calorimetry is the preferred means to evaluate the caloric needs of critically ill patients for it has been shown to be more accurate compared to predictive equations and can be used throughout the course of illness as metabolic needs of a patient will change.

In future podcasts, the clinical use cases and ASPEN guidelines will be discussed in addition to the principles of steady state determination as applied to indirect calorimetry measurement.

Thank you for listening to this podcast on a comparison between Indirect Calorimetry and predictive equations. In the next podcast we will go over clinical use cases and review ASPEN guidelines.

References

1. Jotterand Chaparro, C., et al. (2017). "Performance of Predictive Equations Specifically Developed to Estimate Resting Energy Expenditure in Ventilated Critically Ill Children." J Pediatr 184: 220-226.e225
2. Oshima, T., et al. (2017). "Indirect calorimetry in nutritional therapy. A position paper by the ICALIC study group." Clin Nutr 36(3): 651-662.
3. [Source: Zusman, O., et al. (2016). "Resting energy expenditure, calorie and protein consumption in critically ill patients: a retrospective cohort study." Crit Care 20(1): 367.
4. Jotter and Chaparro, C., et al. (2018). "Estimation of Resting Energy Expenditure Using Predictive Equations in Critically Ill Children: Results of a Systematic Review." JPEN J Parenter Enteral Nutr 42(6): 976-986.
5. Karlsson, M., et al. (2017). "Ability to predict resting energy expenditure with six equations compared to indirect calorimetry in octogenarian men." Exp Gerontol 92: 52-55.
6. Zusman, O., et al. (2016). "Resting energy expenditure, calorie and protein consumption in critically ill patients: a retrospective cohort study." Crit Care 20(1): 367.
7. [Source: Oshima, T., et al. (2017). "Indirect calorimetry in nutritional therapy. A position paper by the ICALIC study group." Clin Nutr 36(3): 651-662.]