SciELO - Scientific Electronic Library Online

vol.26 issue2Lipid and lipoprotein responses of dyslipidemic patients to exclusive nutritional counseling by gender and ageAgricultural production and potential and foods intended for schools in Goiás and Distrito Federal, located in the Midwest Region of Brazil author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand



  • English (pdf)
  • Article in xml format
  • How to cite this article
  • SciELO Analytics
  • Curriculum ScienTI
  • Automatic translation


Related links


Revista de Nutrição

Print version ISSN 1415-5273

Rev. Nutr. vol.26 no.2 Campinas Mar./Apr. 2013 



Which equation should be used to measure energy expenditure in HIV-infected patients?1


Qual equação utilizar para avaliar gasto energético de portadores do HIV?



Helena Siqueira VassimonI,*; Jacqueline Pontes MonteiroII; Alcyone Artioli MachadoIII; Francisco José Albuquerque de PaulaIII; Andressa KutschenkoIV; Alceu Afonso JordãoIII

IUniversidade de Franca, Programa de Pós-Graduação de Promoção de Saúde. Av. Dr. Armando Salles Oliveira, 201, Parque Universitário, 14404-600, Franca, SP, Brasil
IIUniversidade de São Paulo, Faculdade de Medicina de Ribeirão Preto, Departamento de Pediatria e Puericultura. Ribeirão Preto, SP, Brasil
IIIUniversidade de São Paulo, Faculdade de Medicina de Ribeirão Preto, Departamento de Clínica Médica. Ribeirão Preto, SP, Brasil
IVUniversidade de São Paulo, Faculdade de Medicina de Ribeirão Preto, Departamento de Medicina. Ribeirão Preto, SP, Brasil




OBJECTIVE: In view of the practical need to use equations for the evaluation of energy expenditure in HIV-infected patients, the objective of the present study was to determine the concordance between the energy expenditure values obtained by indirect calorimetry as the gold standard and those obtained by predictive equations elaborated from data for the healthy population: Harris-Benedict, Schofield and Cunningham, and by equations elaborated from data for HIV-infected patients: Melchior (1991-1993).
METHODS: The study was conducted at the Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto on 32 HIV-infected men under treatment with highly active antiretroviral therapy. Resting energy expenditure was measured by indirect calorimetry and estimated on the basis of measurement of O2 consumption and CO2 production.
RESULTS: Statistical analysis revealed weak concordance for the Harris-Benedict (0.38) and Cunningham (0.34) equations and satisfactory concordance for the Schofield equation (0.47). Only the two Melchior equations (1991 and 1993) showed strong concordance with the values obtained by indirect calorimetry (0.63 and 0.66, respectively) and could be used in practice.
CONCLUSION: The best equations seem to be population-specific, such as the Melchior equations elaborated for HIV-infected patients.

Indexing terms: Calorimetry, indirect. Energy metabolism. HIV.


OBJETIVO: Frente à necessidade prática da utilização de equações para avaliar gasto energético em portadores do HIV, o presente estudo avaliou a concordância dos valores de gasto energético obtidos pelo padrão ouro de calorimetria indireta com equações preditivas desenvolvidas a partir de população saudável: Harris-Benedict, Schofield e Cunningham e equações desenvolvidas a partir de portadores do HIV: Melchior (1991 a 1993).
MÉTODOS: O estudo foi realizado no Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto, com 32 homens portadores do HIV em terapia antirretroviral de alta potência. O gasto energético de repouso foi medido por calometria indireta e estimado a partir de medida de consumo de O2 e produção de CO2.
RESULTADOS: De acordo com análise estatística, a concordância foi fraca para equações de Harris-Benedict (0,38) e Cunningham (0,34) e satisfatória para Schofield (0,47). Apenas as duas equações de Melchior apresentaram forte concordância com valores obtidos por calometria indireta (respectivamente 0,63 e 0,66) e poderiam ser utilizadas na prática.
CONCLUSÃO:: As melhores equações parecem ser população específica, como as de Melchior, desenvolvidas para portadores do HIV.

Termos de indexação: Calorimetria indireta. Metabolismo energético. HIV.




In clinical practice, due to the increased life expectancy of Human Immunodeficiency Virus (HIV) infected patients, it is increasingly common to provide care for asymptomatic HIV-infected individuals who are generally of normal weight with preservation of muscle mass and with possible changes in adipose mass occurring in the presence of lipodystrophy1. Changes in body composition occur in the lipodystrophy syndrome, including lipoatrophy (loss of subcutaneous fat in certain parts of the body) and lipohypertrophy (accumulation of visceral fat in certain parts of the body) and metabolic changes such as dyslipidemia and insulin resistance, involving an increased risk for cardiovascular diseases2. On this basis, the need for nutritional monitoring becomes essential for this group, and a definition of the energy supply necessary for the maintenance of ideal weight is part of this process. Thus, it is necessary to develop practical tools in order to estimate the energy expenditure of this group of individuals.

Indirect Calorimetry (IC) is one of the most precise methods for the estimate of Resting Energy Expenditure (REE). However, the cost of the equipment needed for these determinations is high, its use requires trained personnel and the procedure is cumbersome, so that IC is more appropriate for scientific research3. In clinical practice it is possible to use predictive equations developed by means of validation studies which evaluate the variables that influence REE and its relationship with the quantitative measurement of this expenditure when evaluated by a more precise method3. Predictive equations are useful for a practical and low-cost estimate of REE. These equations are elaborated for specific populations taking into consideration factors such as body composition, age and gender4.

Many literature reports have compared predictive equations for specific populations to IC in order to determine the best equation for a given group5-7. Lopes et al.5 compared the Food and Agriculture Organization/World Health Organization/United Nations University, Harris and Benedict, Henry and Rees and Schofield equations in 30 subjects of both sexes with grade one obesity residing in the city of Porto Alegre (RS). According to the results compared to IC values, none of these equations guaranteed application to this population. With the same objective, Fet et al.6evaluated women ranging from normal weight to obesity and supported the notion that equations for the prediction of REE can induce to errors and seem to be population-specific. In the cited study, better concordance was observed with the equation elaborated from anthropometric measurements than with existing equations in the literature.

Regarding the HIV-infected population, only one article comparing equations and IC was detected in the literature. In a study of 70 HIV-infected men and 16 healthy men, Batterham et al.8 did not detect concordance between predictive equations and values obtained by IC and stated that the changes in body composition may be responsible, at least in part, for this observation.

In view of the practical necessity of using equations, the objective of the preset study was to evaluate the concordance between the resting energy expenditure values obtained by IC, the method considered to be the gold standard, and the estimates obtained using predictive equations.



This was an analytical study conducted at the University Hospital, Faculty of Medicine of Ribeirão Preto (HC/FMRP). The study was approved by the Research Ethics Committee of the Institution (HCRP nº 1991/2007) and all subjects gave written informed consent to participate. The participants were 32 HIV-infected men on Highly Active Antiretroviral Therapy (HAART) selected at the Acquired Immunodeficiency Syndrome Outpatient Clinic of the Unit Specializing in Infectious-Contagious Diseases (UETDI) of HC/FMRP. Inclusion criteria were age between 18 and 60 years, male gender, stable weight (tolerated variation of 10% in one year), a previous diagnosis of HIV infection, use of HAART for at least 6 months, and T CD4 cell count of more than 200cells/mm3. Exclusion criteria were severe thyroid or renal, cardiac pulmonary and hepatic changes, signs of opportunistic infections at the time of the study, use of glucocorticoids during the last year, age of less than 18 years or more than 60 years, female gender, patients not on HAART, and T CD4 cell count of less than 200cells/ mm3.

The mean age of the participants was 44±6 years, the time since the diagnosis of HIV infection was 8.5±4 years, and the time of HAART use was 7±4 years. All subjects had undetectable viral load and a T CD4 cell count of 482±224 cells/mm3. Mean body mass index was 24±3kg/m2.

IC and REE predictive equations

Resting energy expenditure was estimated using an IC apparatus (Sensor Medics Corporation, Yorba, Linda, CA, USA) by the analysis of consumed Oxygen Volume (VO2) and produced Carbonic Gas Volume (VCO2) measured in expired air. The participants were asked to breathe normally while lying down, with a transparent canopy covering their head and chest for 30 minutes. The temperature of the examination room was controlled and constant (23-25ºC) and, if a participant entered equilibrium, i.e, if he presented a <10% fluctuation in minutes of ventilation and oxygen consumption and a <5% fluctuation of the respiratory quotient, the procedure could be stopped after 15 minutes. VO2 and VCO2 values permitted the determination of the production of energy which is equal to energy expenditure in a situation of equilibrium, i.e., with no loss or gain in the form of heat or of chemistry. The value was expressed as calories, providing the REE in the fasting situation. The formula of Weir9was used to express the results as calories/day: REE (kcal/day)=5.5VO2+1.76VCO2.

The resting energy expenditure values obtained by IC were compared to those of the predictive values listed in Table 110-14. Depending on the equations available in the literature, the results can be reported in different energy units, i.e., calories, kilo joules or mega joules. After the results were obtained, various units were converted to kcal, considering that 1kcal is equivalent to 4.18 joules and 1 mega joule is equivalent to a 1,000 kilo joules.

Dual-energy X-ray Absorptiometry (DXA) was applied to all patients by a trained technician for the determination of the muscle mass values to be included in the equations, using a Hologic apparatus model QDR 4500W®. One of the authors (H. S. Vassimon) determined weight and height. The individuals were fasting, wearing light clothing and no shoes and were asked to remove jackets and coats, if present. Weight was measured with a Filizola® electronic scale of the platform type with a maximum capacity of 300.0kg and precision of 0.1kg. Height was measured with a stadiometer with 0.1cm precision, with the subjects standing with their heels touching, with their arms along the body, their back touching the vertical rod, and positioned in the center of the equipment. Body Mass Index (BMI) (weight in kg divided by height squared) was calculated in order to classify the nutritional status of the participants according to the World Health Organization.

Data were analyzed statistically with the aid of the SAS software, version 9.1. Continuous variables are reported as mean, standard desviation. To achieve the main objective of the study the St. Laurent15 coefficient was applied to test concordance between values (continuous variables), considering IC as the reference method for comparison. The results of concordance were analyzed according to the classification proposed by Landis & Koch16, whereby a coefficient of 0.81 to 1.00 indicates almost perfect concordance, a coefficient of 0.61-0.89 indicates strong concordance, a coefficient of 0.41 to 0.60 satisfactory concordance, a coefficient of 0.21 to 0.40 weak concordance, and a coefficient below 0.20 poor concordance. Bland & Altman17 graphs were obtained as another parameter for the evaluation of concordance of the data.



Table 2 shows the REE results estimated by IC and those obtained mathematically by means of equations. Analysis of the percent value obtained from the predictive equations compared to the total IC value revealed that only the Melchior equations of both 1991 and 1993 approached the real value. Greater discrepancies were observed between IC and the Cunningham10and Harris & Benedict11 equations.



Table 3 presents the St. Laurent15 coefficient obtained by comparing the values of the predictive equations to the values obtained by IC. It can be seen that only the REE values calculated with the Melchior equations showed strong concordance with those obtained by IC. Figures 1 and 2 contain graphs comparing the values obtained by the Melchior et al.12,13 equation to those obtained by IC.








The Harris & Benedict11 equation is still one of the main instruments used to evaluate REE in the general population, but the original validated study involved 136 men, 103 women and 94 children of North American origin considered to be healthy. Equations were developed for each gender, including only body weight, height and age as independent variables11. Schofield14considered jointly the data of 114 studies considered to be scientifically adequate, corresponding to a final sample of 7,173 healthy individuals (4,809 men and 2,364 women) mostly of European or North American origin. Equations were developed for each gender, including age and body weight as independent variables, without considering height. Cunningham10proposed an equation based on a review of studies published since 1980 evaluating REE and total energy expenditure in healthy populations of both genders and pointed out the influence of lean mass as a determinant variable of these values, including only lean mass in the equation.

The central point of the present study was the observation that most of the predictive equations developed for the general population were not sensitive for the estimate of REE in HIV-infected individuals. In this case, the equations underestimate REE compared to the values obtained by IC. Some factors that might explain the lack of concordance of the equations are: a) equations developed based on a healthy population not infected with HIV, b) clinical condition of hypermetabolism presented by the group evaluated, and c) the lack of inclusion of muscle mass as a variable of the equation. An equation is more precise for the specific population for which it was developed and in the case of the present study only two equations used HIV-infected subjects. In contrast, regarding hypermetabolism, this situation was observed both before and after the use of HAART18,19. Increased REE in HIV-infected individuals has been described since the onset of the disease but its causes have not been fully clarified. Some of the factors reported in the literature as being involved in this situation of hypermetabolism are: the action of HIV itself, the presence of opportunistic infections, the action of antiretroviral agents, and/ or the presence of lipodystrophy syndrome18,20-22. A previous study suggested that muscle tissue may be the site responsible for the increased energy expenditure in HIV-infected individuals with lipodystrophy syndrome23. With the objective of discriminating the participation of the various sites in REE in HIV-infected patients with lipodystrophy syndrome, Kosmiski et al.23 obtained a metabolic map of adipose tissue, skeletal muscle, brain, bone and residual mass (including intestine, lungs, liver, heart and kidneys) based on DXA values combined with specific metabolic rates for each site described in the literature. Multivariate regression analysis revealed that the estimated metabolic rate was elevated only in the skeletal muscle of HIV-infected patients with lipodystrophy syndrome and not in HIV-infected patients without the syndrome or in healthy controls. Thus, lean mass, which is considered to be the main determinant of REE for the general population, becomes a stronger variable for HIV-infected persons with lipodystrophy syndrome and therefore an important variable to be included in equations for REE estimates.

Within this context, the equations of Melchior et al.12,13 were the only ones showing strong concordance with IC values, having been developed for HIV-infected individuals and considering muscle mass as the variable. The equation published in 199113 was based on 50 HIV-infected individuals with a diagnosis of AIDS, with the presence of opportunistic infections or Kaposi sarcoma, with the presence of cachexia and the absence of antiretroviral use. In contrast, the equation published in 199312 was elaborated based on 129 malnourished HIV-infected patients with no secondary infections. Only the quantity of lean mass was included in these equations as the independent variable, supporting the importance of this variable for predictive REE equations in HIV-infected individuals. The limiting factor of the Melchior equation would be the question of malnutrition, a situation no longer currently present in most HIV-infected patients on HAART.



Of the five equations evaluated, the two equations of Melchior, developed from HIV-infected individuals, were those closest to IC values and could be applied in clinical practice for the estimate of the energy requirements of this group. The Schofield equation would be an alternative in cases in which the Melchior equation cannot be calculated. Most of the predictive equations of REE in the literature were inadequate for HIV-infected individuals. The best equations seem to be population-specific, such as those of Melchior, developed for HIV-infected individuals.


Research supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (Grant 2009/12449-7).


HS VASSIMON conceived the study and participated in writing the manuscript; JP MONTEIRO participated in the definition and execution of the project. FJA PAULA and AA MACHADO cooperated with data collection, providing equipment and patient selection. AA JORDÃO participated in orientation and in revision of the text. A KUTSCHENKO was responsible for the statistical analysis.



1. Vassimon HS, Jordao AA, Paula FJA, Machado AA, Monteiro JP. Comparison of bioelectrical impedance with skinfold thickness and X-ray absorptiometry to measure body composition in HIV-infected with lipodistrophy. Nutr Hosp. 2011; 26(3):458-64. doi: 10.3305/nh.2011.26.3.4794.         [ Links ]

2. Lichtenstein K, Balasubramanyam A, Sekhar R, Freedland E. HIV-associated adipose redistribution syndrome (HARS): Etiology and pathophysiological mechanisms. AIDS Res Ther. 2007; 4:14. doi: 10.1186/1742-6405-4-14.         [ Links ]

3. Haugen HA, Chan LN, Li F. Indirect calorimetry: A practical guide for clinicians. Nutr Clin Pract. 2007; 22(4):377-88.         [ Links ]

4. Wahrlich V, Anjos LA. Historical and methodological aspects of the measurement and prediction of basal metabolic rate: A review. Cad Saúde Pública. 2001; 17(4):801-17.         [ Links ]

5. Lopes AL, Silva, DP, Zacca R, Castro FAS, Oliveira AR. Validação de equações de predição da taxa metabólica basal por meio de calorimetria indireta em indivíduos obesos. Rev Bras Ativ Física Saúde. 2010; 15(4):243-38.         [ Links ]

6. Fett CA, Fett WC, Marchini JS. Resting energy expenditure measured vs. estimated and this relationship with body composition in women. Arq Bras Endocrinol Metabol. 2006; 50(6):1050-8.         [ Links ]

7. Rao Z, Wu X, Liang B, Wang M, Hu W. Comparison of five equations for estimating resting energy expenditure in Chinese young, normal weight healthy adults. Eur J Med Res. 2012; 17-26. doi: 10.1186/2047-783X-17-26.         [ Links ]

8. Batterham MJ, Morgan-Jones J, Greenop P, Garsia R, Gold J, Caterson I. Calculating energy requirements for men with HIV/AIDS in the era of highly active antiretroviral therapy. Eur J Clin Nutr. 2003; 57(2):209-17.         [ Links ]

9. Weir JB. New methods for calculating metabolic rate with special reference to protein metabolism. 1949. Nutrition. 1990; 6(3):213-21.         [ Links ]

10. Cunningham JJ. Body composition as a determinant of energy expenditure: A synthetic review and a proposed general prediction equation. Am J Clin Nutr. 1991; 54(6):963-9.         [ Links ]

11. Harris JA, Benedict FG. A biometric study of human basal metabolism. Proc Natl Acad Sci. 1918; 4(12):370-3.         [ Links ]

12. Melchior JC, Raguin G, Boulier A, Bouvet E, Rigaud D, Matheron S, et al. Resting energy expenditure in human immunodeficiency virus-infected patients: Comparison between patients with and without secondary infections. Am J Clin Nutr. 1993; 57(5):614-9.         [ Links ]

13. Melchior JC, Salmon D, Rigaud D, Leport C, Bouvet E, Detruchis P, et al. Resting energy expenditure is increased in stable, malnourished HIV-infected patients. Am J Clin Nutr. 1991; 53(2):437-41.         [ Links ]

14. Schofield WN. Predicting basal metabolic rate, new standards and review of previous work. Hum Nutr Clin Nutr. 1985; 39(Suppl 1):5-41.         [ Links ]

15. St Laurent RT. Evaluating agreement with a gold standard in method comparison studies. Biometrics. 1998; 54(2):537-45.         [ Links ]

16. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977; 33(1):159-74.         [ Links ]

17.Bland JM, Altman DG. Statisctical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986; 327(8476): 307-10.         [ Links ]

18. Batterham MJ. Investigating heterogeneity in studies of resting energy expenditure in persons with HIV/AIDS: A meta-analysis. Am J Clin Nutr. 2005; 81(3):702-13.         [ Links ]

19. Vassimon HS, Paula FJ, Machado AA, Monteiro JP, Jordao AA Jr. Hypermetabolism and altered substrate oxidation in HIV-infected patients with lipodystrophy. Nutrition. 2012; 28(9):912-6. doi: 10.1016/j.nut.2011.12.010.         [ Links ]

20. Hommes MJ, Romijn JA, Endert E, Sauerwein HP. Resting energy expenditure and substrate oxidation in human immunodeficiency virus (HIV)-infected asymptomatic men: HIV affects host metabolism in the early asymptomatic stage. Am J Clin Nutr. 1991; 54(2):311-5.         [ Links ]

21. Shevitz AH, Knox TA, Spiegelman D, Roubenoff R, Gorbach SL, Skolnik PR. Elevated resting energy expenditure among HIV-seropositive persons receiving highly active antiretroviral therapy. Aids. 1999; 13(11):1351-7.         [ Links ]

22. Kosmiski LA, Bessesen DH, Stotz SA, Koeppe JR, Horton TJ. Short-term energy restriction reduces resting energy expenditure in patients with HIV lipodystrophy and hypermetabolism. Metabolism. 2007; 56(2):289-95.         [ Links ]

23. Kosmiski LA, Ringham BM, Grunwald GK, Bessesen DH. Dual-energy X-ray absorptiometry modeling to explain the increased resting energy expenditure associated with the HIV lipoatrophy syndrome. Am J Clin Nutr. 2009; 90(6):1525-31.         [ Links ]



Received on: 24/2/2012
Final version on: 17/10/2012
Approved on: 12/11/2012



* Correspondence to/Correspondência para: HS VASSIMON. E-mail: <>.
1 Article elaborated from the thesis of HS VASSIMON, entitled "Oxidative stress and substrate oxidation in HIV-infected men with lipodystrophy syndrome". Universidade de São Paulo; 2011.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License