SciELO - Scientific Electronic Library Online

 
vol.21 número2Una mayor presencia de dietistas-nutricionistas en el sector público podría mejorar la eficiencia del sistema sanitario público españolPrevalencia de anemia ferropénica en deportistas seleccionados del Instituto Peruano del Deporte durante el año 2013: estudio transversal índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google

Compartir


Revista Española de Nutrición Humana y Dietética

versión On-line ISSN 2174-5145versión impresa ISSN 2173-1292

Rev Esp Nutr Hum Diet vol.21 no.2 Pamplona abr./jun. 2017

https://dx.doi.org/10.14306/renhyd.21.2.252 

Research Articles

Can anthropometry measure the body fat of people living with HIV/AIDS?. A systematic review

¿Puede la antropometría medir la grasa corporal de las personas que viven con el VIH/SIDA?. Revisión sistemática

Nathalia Sernizon Guimarãesa  *  , Maria Arlene Faustob  , Adriana Maria Kakehasia  , Anderson Marliere Navarroc  , Unaí Tupinambása 

a Universidade Federal de Minas Gerais, Brazil.

b Universidade Federal de Ouro Preto, Minas Gerais, Brazil.

c Universidade de São Paulo, Brazil.

ABSTRACT

Introduction:

Assessment of the quantity and distribution of body fat in people living with HIV/AIDS is of great importance in clinical practice, due to the association of body fat changes with clinical conditions. The aim of this systematic review was to answer the central question: Can anthropometry accurately measure the body fat in people living with HIV/AIDS?

Material and Methods:

Systematic review carried out using four databases: Medline, LILACS, Scopus and BDTD.

Results:

Of the 581 studies found, 11 met the eligibility criteria. To assess the validate of anthropometry, only two studies employed regress analysis to development of predictive body fat equations in people living with HIV/AIDS and nine studies employed correlation analysis. This coefficient only measures the strength of the relation between two variables, and there is not concordance between them and therefore, these studies did not accurately evaluate whether or not the anthropometric information showed good concordance with the gold standard. The other two studies developed five equations to evaluate the total fat and limbs (arm, leg and trunk) in people living with HIV/AIDS using antiretrovirals and showed R2 between 0.50 and 0.83.

Conclusions:

Further research needs to be conducted to answer the central question of this review, as the small number of articles that applied the correct statistical test and the absence of research on people living with HIV/AIDS without the use of antiretrovirals.

Keywords: Anthropometry; Body Composition; Adipose Tissue; HIV; Acquired Immunodeficiency Syndrome

RESUMEN

Introducción:

La evaluación de la cantidad y distribución de la grasa corporal en personas que viven con el VIH/SIDA es de gran importancia en la práctica clínica, debido a la asociación de los cambios de grasa corporal con condiciones clínicas. El objetivo de esta revisión es responder a la pregunta central: ¿Puede la antropometría medir con precisión la grasa corporal en las personas que viven con el VIH/SIDA?

Material y Métodos:

Revisión sistemática llevada a cabo por medio de cuatro bases de datos: Medline, LILACS, Scopus y BDTD.

Resultados:

De los 581 estudios encontrados, 11 cumplieron con los criterios de elegibilidad. Para evaluar la validación de la antropometría, sólo dos estudios emplearon análisis de regresión para el desarrollo de las ecuaciones de predicción de grasa corporal en las personas que viven con el VIH/SIDA y nueve estudios emplearon análisis de correlación. Este coeficiente sólo mide la fuerza de la relación entre dos variables, y no hay concordancia entre ellos y, por lo tanto, estos estudios no evaluaron con precisión si la información antropométrica mostró buena concordancia con el estándar de oro. Los otros dos estudios desarrollaron cinco ecuaciones para evaluar la grasa total y en las extremidades (brazos, piernas y tronco) en personas que viven con el VIH/SIDA y usan antirretrovirales y mostraron R2 entre 0,50 y 0,83.

Conclusiones:

Más investigación debe llevarse a cabo para responder a la pregunta central de esta revisión, dado el pequeño número de artículos en que se aplicó la prueba estadística correcta y la ausencia de investigaciones sobre personas que viven con el VIH/SIDA que no usan antirretrovirales.

Palabras clave: Antropometría; Composición Corporal; Tejido Adiposo; VIH; Síndrome de Inmunodeficiencia Adquirida

INTRODUCTION

Over the course of 2015, there were approximately 2.1 million new cases of HIV infection, making a total of 36.7 million people living with HIV/AIDS (PLWHA) worldwide. Approximately, 20 million of these individuals do not make use of antiretroviral therapy1,2.

When compared to the general population, people living with HIV/AIDS undergo more frequent changes to their body composition, principally in relation to the quantity and distribution of body fat3,4. This redistribution of body fat is referred to as lipodystrophy or lipodystrophy associated with HIV and is subdivided into lipoatrophy, lipohypertrophy or mixed form. Lipoatrophy is characterized by the reduction of fat in the face, arms, legs and buttocks. Lipohypertrophy is characterized by the accumulation of fat in the abdomen, back, neck and breast area. Mixed form is characterized the two forms described above5. These morphological changes in body fat have multifactorial causes such as duration of HIV infection, type of medicine used in the antiretroviral therapy, duration of exposure to antiretroviral therapy, genetic predisposition or lifestyle (physical inactivity and inadequate diet)6,8.

Current studies shows that changes in fat distribution by region, especially intra-abdominal adipose tissue, have been associated with the incidence of dyslipidemia, insulin resistance, metabolic syndrome, type 2 diabetes mellitus and hepatic steatosis9,10. These metabolic changes can lead to an increase in morbimortality from cardiovascular diseases11. For these reasons, assessing body fat distribution and determining the quantity of fat in PLWHA is of vital importance in clinical practice.

Assessing the quantity of body fat can be performed using methods with different levels of sensitivity, specificity, clinical practicality and cost12,13. Dual energy X-Ray absorptiometry (DXA) and computed tomography (CT) are considered “gold standard” methods in estimating the body composition of individuals and quantifying body fat13,14. Nevertheless, as with MRI and ultrasound, DXA and CT are body composition assessment techniques less favored in clinical practice due to the high costs involved, including the acquisition of the appropriate equipment, use of specific software, trained professionals, and the regular expenses incurred in the maintenance and calibration of the machines15.

In comparison to the gold standard methods (DXA or CT), anthropometry is widely used in contexts of limited resources due to its low cost, shorter execution time and greater simplicity16. This technique is used by health professionals in clinical practice with the objective of recording body measurements such as weight, height, skinfolds and body circumferences16. When linked to indexes or predictive equations, these body measurements, also referred to as anthropometric indicators, can provide information on the quantity of the individual’s fat mass17,18.

Several studies have been carried out with the objective of investigating the accuracy of anthropometric measurements in the description of body fat quantity in different populations, justified by the need to gather measurements in a shorter timeframe, with lower cost and greater simplicity 19,20,21,22. Therefore, this review proposes a response to the following central question: Can anthropometry accurately measure the body fat of PLWHA?

MATERIAL AND METHODS

The search for information and the presentation and interpretation of data were carried out based on the PRISMA-P method23. The PROSPERO protocol24 of this systematic review was registered in the Centre for Reviews and Dissemination of the University of York, under number CRD42015025347 and may be consulted at: http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42015025347.

Observational and intervention studies were included that evaluated anthropometry through indicators, anthropometric indexes and predictive equations and were compared with at least one of the methods considered as gold standard in assessing body fat, namely, DXA or CT, in the PLWHA group aged 18 to 60 years.

The exclusion criteria were divided into: Group 1 - duplicate articles; Group 2 - studies that only evaluated fat free body mass, such as bone, water or muscle, not considering the assessment of the individuals’ body fat or when evaluated facial fat; Group 3 - absence of a comparison between anthropometry and the gold-standard methods (DXA or CT) or theme not connected to the objective of this review; Group 4 - study designs: narrative, systematic reviews or meta-analyses, experimental studies carried out on animals, report or case series; Group 5 - Individuals using corticosteroids or anabolic steroids, studies performed on pregnant women or nursing mothers, people living with HIV/AIDS with chronic infections. There was no restriction in relation to language and publication year of the studies.

Studies were identified by means of five electronic databases: (I) OVID-Medline (1982 to July 2015); (II) LILACS (2000 to July 2015); (III) Scopus (1982 to July 2015) and (IV) Brazilian Digital Library of Theses or Dissertations (2001 to July 2015). Selection of the search terms (keywords or descriptors) was done through a consultation of the Health Sciences Descriptors (DeCS), Medical Subject Headings (MeSH) and Emtree.

In every database, the descriptors shown in Table 1 were subdivided into three groups (assessment method for body composition/body fat changes/HIV-AIDS) and were then matched up using Boolean search operators: inverted commas, brackets, “AND” and “OR”.

Table 1 Search strategy used in the Medline-OVID database 

1. exp Anthropometry/
2. exp Absorptiometry, Photon/
3. Tomography, X-Ray Computed/
4. 1 and 2
5. 1 and 3
6. Fat Body.mp.
7. Adipose Tissue.mp.
8. Abdominal Fat.mp.
9. Intra-Abdominal Fat.mp.
10. Subcutaneous Fat.mp.
11. Body Composition.mp.
12. Body Fat Distribution.mp.
13. 6 or 7 or 8 or 9 or 10 or 11 or 12
14. Acquired Immunodeficiency Syndrome.mp.
15. HIV.mp. 16. 14 or 15
17. 4 or 5
18. 13 and 16 and 17

All exclusion stages of the studies were carried out independently by two authors of this review, with the objective of identifying studies that potentially met the inclusion criteria described previously. Any disagreement on the eligibility of the studies was resolved by a third reviewer.

RESULTS

Selection and overall characteristics of the studies: The information search returned 581 articles. 101 duplicate studies were excluded and, after applying the inclusion criteria through the reading of titles and abstracts, 557 were discarded. In the text analysis stage, 13 studies were excluded. The absence of a comparison between anthropometry and the gold-standard methods was the main reason for exclusion in all stages. In the end, 11 studies published between 1993 and 2015 were selected for the systematic review (Figure 1).

n: sample; LILACS: Latin American and Caribbean Health Sciences Literature; PubMed: International Literature on Health Sciences; SCIELO: Scientific Electronic Library Online. Stage 1: Removal of duplicates by reading titles; Stage 2: Application of further exclusion criteria by reading titles; Stage 3: Application of exclusion criteria by reading abstracts; Final stage: Application of exclusion criteria by textual analysis

Figure 1 Stages of the selection process for article inclusion in the systematic review about anthropometry and others methods assessing body fat in people living with HIV/AIDS. 

The types of study designs noted in the 11 selected articles were: cross-sectional (n=09)25,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33, cohort (n=01)34 and case-control (n=1)35 (Table 2).

Table 2 Summary of the comparative study results for anthropometric information and gold standard methods in assessing body fat in HIV infected individuals. 

Reference Population and Study Design Anthropometric Information Diagnostic Exam Estimated Fat Statistical Analysis Results
Beraldo y Cols., 2015 (Brazil) n = 100 Indicators: weight, height, WC, AC, HC, TC, CC, BSF, TSF, SSF, ICSF, LegL. DXA % Arm Fat, Multiple Linear Regression % Arm Fat= -1.499+(0.021x W)+(0.018xAC)+(0.023x TSF)+(0.002xA). R2=0.66
43.6 years % Leg Fat, % Trunk Fat= -18.043+(0.114x)+(0.169xWC)+(0.117xIC SF)+(0.038xA). R2=0.76
100% ♂ % Trunk Fat. % Leg Fat= -7.346+(0.022xW)+(0.134xTC)+0.015xLegL. R2=0.50
Cross-sectional
Florindo y Cols., 2008 (Brazil) n = 15 Indicators: BSF, TSF, SSF, ICSF, AxSF. ASF, CSF. DXA e CT % Total Fat, Multiple Linear Regression and Pearson Correlation % Total Fat ♂: 3.385+0.279*(AxSF+SSF). R2=0.83
36.9 years Equations: Durnin & Womersley, HIVE, Siri. Visceral Fat, % Total Fat : -24.343+0.736*(ICSF+ASF+CSF).
66.6% (n=10)♂ Subcutaneous Fat, R2=0.81
Cross-sectional Abdominal Fat. % Fat HIVE vs % Fat TC: r=0.69 p=0.012
Aghdassi y Cols., 2007 (Canada) n = 47 Indicators: WC, BSF, TSF, SSF. DXA % Total Fat. Pearson Correlation BMI vs % Total Fat: r=0.628 p<0.01
49.2 years Indexes: BMI, WHR, ΣSF: BSF+TSF+SSF WC vs % Total Fat: r=0.784 p<0.01
100% ♂ WHR vs % Total Fat: r=0.525 p<0.01
Cross-sectional BSF vs % Total Fat: r=0.538 p<0.01
TSF vs % Total Fat: r=0.669 p<0.01
SSF vs % Total Fat: r=0.665 p<0.01
ΣSFs vs % Total Fat: r=0.759 p<0.01
(WHR>0.9)ΣSFs vs % Total Fat: r=0.775 p<0.001
(WHR<0.9)ΣSFs vs % Total Fat: r=0.497 p<0.316
Batterham y Cols., 1999 (Australia) n = 36 Equations: Sloan, Wilmore, Forsyth, Katch, Durnin & Womersley, Thorland, Withers. DXA % Total Fat. Pearson Correlation Sloan vs % Total Fat: r=0.847 p<0.001
42.6 years Wilmore vs % Total Fat: r=0.769 p<0.001
100% ♂ Forsyth vs % Total Fat: r=0.786 p=0.001
Cross-sectional Katch vs % Total Fat: r=0.848 p<0.001
Durnin vs % Total Fat: r=0.828 p=0.002
Thorland vs % Total Fat: r=0.849 p<0.001
Withers vs % Total Fat: r=0.810 p<0.001
Antunes y Cols., 2011 (Brazil) n = 26 Indicators: HC, WC, CC, TC, BSF, TSF, SSF, ICSF. DXA Arm Fat (kg), Leg Fat (kg), Trunk Fat (kg). Pearson Correlation TSF vs Arm Fat(kg): r=0.605 p<0.01;
48.6 years Indexes: Arm fat area. % Arm Fat, TSF vs Arm Fat(%): r=0.833 p<0.01
76.9% (n=20)♂ % Leg Fat, WC vs Trunk Fat(kg): r=0.833 p<0.01;
Cross-sectional % Trunk Fat. WC vs Trunk Fat(%): r=0.583 p<0.01
CC vs Leg Fat(kg): r=0.328 p=0.10;
CC vs Leg Fat(%): r=0.133 p=0.51
TC vs Leg Fat(kg): r=0.482 p<0.01;
TC vs Leg Fat(%): r=0.367 p=0.06
Florindo y Cols., 2004 (Brazil) n = 15 Indicators: WC, HC, BSF, TSF, SSF, ICSF, AxSF, ASF, CSF. CT Visceral Fat, Subcutaneous Fat, Abdominal Fat. Pearson Correlation WC vs Visceral Fat: r=0.61 p<0.037
36.6 years** Indexes: WHR. WC vs Subcutaneous Fat: r=0.88 p<0.001
66.6% (n=10)♂ WC vs Abdominal Fat: r=0.89 p<0.001
Cross-sectional WHR vs Visceral Fat: r=0.74 p<0.006
WHR vs Subcutaneous Fat: r=0.61 p<0.035
WHR vs Abdominal Fat: r=0.75 p<0.005
Meininger y Cols., 2002 (EUA) n = 41 Index: WHR. DXA e CT % Total Fat and %Trunk/LimbFat. Pearson Correlation WHR vs. Abdominal Fat: r=0.72 p<0.0001
43 years Visceral Fat, Subcutaneous Fat, Abdominal Fat. WHR vs % Total Fat: r=0.38 p=0.012
100% ♂ WHR vs Trunk/Limb: r=0.68 p<0.0001
Control case
Mulligan y Cols., 2006 (EUA) n = 157 Indicators: weight. WC, HC, TC, AC. DXA Total Fat(kg), Arm Fat(kg), Lower Limb Fat(kg), Trunk Fat(kg), Leg Fat(kg) Spearman Correlation weight vs Fat(kg): r=0.724 p<0.001
S.I WC vs Fat (kg): r=0.616 p<0.001
87% (n=136)♂ HC vs Fat(kg): r=0.557 p<0.001
Cohort- 64 months TC vs Fat(kg): r=0.556 p<0.001
S.I AC vs Fat: r=0.639 p<0.001
weight vs Trunk Fat(kg): r=0.743 p<0.001
WC vs Trunk Fat(kg): r=0.638 p<0.001
HC vs Trunk Fat(kg): r=0.573 p<0.001
TC vs Trunk Fat(kg): r=0.500 p<0.001
AC vs Trunk Fat(kg): r=0.589 p<0.001
weight vs Lower Limb Fat(kg): r=0.631 p<0.001
WC vs Lower Limb Fat(kg): r=0.540 p<0.001
HC vs Lower limb Fat (kg): r=0.504 p<0.001
TC vs Lower Limb Fat(kg): r=0.555 p<0.001
AC vs Lower Limb Fat(kg): r=0.603 p<0.001
weight vs Arm Fat(kg): r=0.560 p<0.001
WC vs Arm Fat(kg): r=0.558 p<0.001
HC vs Arm Fat(kg): r=0.402 p<0.001
TC vs Arm Fat(kg): r=0.496 p<0.001
AC vs Arm Fat(kg): r=0.575 p<0.001
weight vs Leg Fat(kg): r=0.619 p<0.001
WC vs Leg Fat(kg): r=0.510 p<0.001
HC vs Leg Fat(kg): r=0.501 p<0.001
TC vs Leg Fat(kg): r=0.534 p<0.001
AC vs Leg Fat(kg): r=0.579 p<0.001
Segatto y Cols., 2012 (Brazil) n = 67 Indicators: WC, HC, TC. DXA Trunk Fat(g). Pearson Correlation ♂ BMI vs Trunk Fat: r=0.77 p<0.01
43.6 years Indexes: BMI, CI, WHR, WHeR, WTR. WHR vs Trunk Fat: r=0.60 p<0.01
58.2% (n=39)♂ CI vs Trunk Fat: r=0.52 p<0.01
Cross-sectional WHeRvs Trunk Fat: r=0.80 p<0.01
WTR vs Trunk Fat: r=0.58 p<0.01
♀BMI vs Trunk Fat: r=0.67 p<0.01
WHR vs Trunk Fat: r=0.52 p<0.01
CI vs Trunk Fat: r=0.58 p<0.01
WHeR vs Trunk Fat: r=0.87 p<0.01
WTR vs Trunk Fat: r=0.35 p>0.05
Siqueira y Cols., 2011 (Brazil) n = 32 Index: ΣSF: BSF+TSF+SSF+ICSF. DXA % Total Fat. Pearson Correlation LIPO+: ΣSF vs % Total Fat: r=0.46 p>0.05
44.5 years** LIPO-: ΣSF vs % Total Fat: r=0.79 p<0.001
100% ♂
Cross-sectional
Wang y Cols., 1993 (EUA) n = 18 Equations: Steinkamp, Durnin & Womersley. DXA % Total Fat. Pearson Correlation Steinkamp vs % Total Fat: r=0.82 p<0.05
41 years Durnin vs % Total Fat: r=0.69 p<0.05
100% ♂
Cross-sectional

WC: waist circumference; AC: arm circumference; HC: hip circumference; TC: thigh circumference; CC: calf circumference; BSF: biceps skinfold; TSF: triceps skinfold; SSF: subscapular skinfold; ICSF: iliac crest skinfold; LegL: leg length; AxSF: axillary skinfold; ASF: abdominal skinfold; CSF: calf skinfold; HIVE: equations for estimating fat mass in HIV/AIDS subjects; BMI: Body Mass Index; WHR: waist/hip ratio; WHeR: Waist/ height ratio; WTR: Waist/thigh ratio; ΣSF: sum of skinfolds; CI: conicity index; DXA: Dual energy X-Ray Absorptiometry; CT: computed tomography; W: weight (kg); A: age (years); R2: coefficient of determination; LIPO+: presence of lipodystrophy syndrome; LIPO-: absence of lipodystrophy syndrome.

Six studies were exclusively focused on men27,28,32,33,35 and five studies were on both sexes22,25,26,27,29,30,31,34.

Mean HIV infection diagnosis time was assessed in only five studies25,27,29,32,35 and was equal to 10 years, with minimum and maximum range of 8 and 12.5 years for infection. Mean duration of antiretroviral therapy was observed in eight studies25,26,27,29,30,32,34,35. Among the 11 selected studies, only one was carried out on PLWHA who had never undergone antiretroviral therapy34 (Table 2).

Anthropometric parameters and measures used in the studies: The anthropometric parameters and measures evaluated in the 11 selected studies were expressed in indicators, indexes or predictive equations for fat quantity.

The anthropometric indicators used in the studies were: weight, height, waist circumference (WC), arm circumference (AC), hip circumference (HC), thigh circumference (TC), calf circumference (CC), biceps skinfold (BSF), triceps skinfold (TSF), subscapular skinfold (SSF), iliac crest skinfold (ICSF), axillary skinfold (AxSF), abdominal skinfold (ASF), calf skinfold (CSF), thigh skinfold (ThSF), and leg length (LegL).

The anthropometric indexes presented by the studies were: Body mass index (BMI), waist/hip ratio (WHR), sum of skinfolds (ΣSF: BSF + TSF + SSF or ΣSF: BSF + TSF + ICSF + SSF); Arm fat area (AFA); conicity index (CI), waist/height ratio (WHeR), and waist/thigh ratio (WTR).

The equations of Durnin & Womersley; HIVE (equations for estimating fat mass in HIV/AIDS subjects26); Siri; Sloan, Wilmore, Forsyth, Katch, Thorland, Withers, Steinkamp were used with the objective of calculating the percentage of body fat based on anthropometric information validated in different populations (Table 3)

Table 3 Predictive equations of body fat in people living with HIV/AIDS used by three studies of this review. 

Reference Equation
Durnin & Womersley, 1974 1.1765-0.0744 Log(ΣSF: BSF+TSF+SSF+ICSF)
HIVE, 2008 ♂: 3.385-0.279*(AxSF+SSF)
HIVE, 2008 ♀: -24.323+0.736*(ICSF+ASF+CSF)
Siri, 1961 %G = [(4.95/D) - 4.50] x 100
Sloan, 1967 1.1043-0.001327(ThSF)-0.00131(SSF)
Wilmore, 1969 ♀ 18-48 years: D= 1.06234 - 0.00068 (SSF) - 0.00039 (TSF) - 0.00025 (AxSF)
Wilmore, 1969 ♂ 17-37 years: D= 1.08543 - 0.000886 (AC) - 0.00040 (AxSF) 1.10647-0.00162(SSF)-
Forsyth & Sinning, 1973 0.00144(AC)-0.00077(TSF)+0.000071(AxSF)
Katch & McArdle, 1973 1.09665-0.00103(TSF)-0.00056(SSF)-0.00054(AC)
Thorland, 1984 1.1136-0.00154(TSF+SSF+AxSF)+0.00000516(TSF+SSF+AxSF) 2
Withers, 1987 1.0988-0.0004(BSF+TSF+SSF+ICSF+AC+AxSF+CC)

D: body density; ΣSF: sum of skinfolds; BSF: biceps skinfold; TSF: triceps skinfold; ICSF: iliac crest skinfold; SSF: subscapular skinfold; ASF: abdominal skinfold; CSF: calf skinfold; ThSF: thigh skinfold; %F: fat percentage; AC: abdominal circumference; CC: calf circumference; AxSF: axillary skinfold.

Statistical methods used to compare collected data with the gold standard: Two studies25,26 developed predictive equations to estimate the body fat of PLWHA through linear regression analysis.

Nine studies27,28,29,30,31,32,33,34,35 used correlation coefficients (Pearson and Spearman) to assess the association between quantity of body fat using anthropometry in comparison to the gold standard (DXA or CT) for PLWHA.

Principal results: Beraldo y Cols.25, noted that the predictive equation composed of weight, age, AC and TSF corresponded to 66% of arm fat variability measured by DXA, while the predictive equation that used weight, age, WC and ICSF corresponded to 76% of trunk fat calculated by DXA.

In stratifying the sample by sex, the study performed by Florindo y Cols.26 noted that, in male people living with HIV/AIDS, the comparison of body fat percentage measured by DXA and the sum of ASF and SSF accounted for 83% of data variability. For female people living with HIV/AIDS, the comparison between the sum of ICSF, ASF and CSF and the body fat percentage measured by DXA explained 81% of data variability.

In comparison with anthropometric indexes or indicators, predictive equations for fat showed stronger correlations with total body fat percentage, especially the expressions from Thorland (r=0.849), Katch (r=0.848), Sloan (r=0.847), Steinkamp (r=0.82), Durnin (r=0.828; r=0.69)28,33.

When evaluating isolated anthropometric indicators, waist circumference was the anthropometric measurement that showed strongest association with fat percentage (r=0.853; r=0.784)27,34.

TSF showed a strong and positive association with arm fat percentage (r=0.833)29. Trunk fat was strongly associated to waist circumference (r=0.833 and r=0.854)29,34 and strong correlations were not noted between leg fat and calf or thigh circumference29,31,34.

When evaluating fat types by CT, Florindo y Cols.30 noted a strong correlation between WC and abdominal fat (r=0.89) as well as subcutaneous fat (r=0.88). However, on evaluating the relation between WC and visceral fat, moderate correlation was noted (r=0.61). Visceral fat showed greater correlation with waist/hip ratio (r=0.74)30.

DISCUSSION

There have been few studies validating the use of anthropometric techniques as predictors of body fat in HIV infected individuals. Among the investigated studies, only two to develop predictive equations for body fat in PLWHA. The main limitation of one of these studies is its sample size (n=15)30, which makes it necessary to validate these equations, stratified by sex, using a larger sample. The study by Beraldo y Cols.25 investigated men living with HIV/AIDS and presented three mathematical equations involving the fat percentage in the arm, trunk and legs. The equation to estimate the amount of fat in the arm involved the patient’s weight, arm circumference, triceps skinfold thickness and age, with estimated value of R2=0.66. To evaluate the trunk fat, the study cited propos equation involving weight, age, waist circumference and skinfold suprailiac (R2=0.76). Finally, the equation estimating the amount of fat in the legs, in percentage, involving weight and waist circumference (R2=0.50). Besides the investigation be unilateral in relation to sex, this study did not show a sub analysis with only HIV-infected individuals without the use of antiretroviral therapy, cannot be applied these equations found in these individuals25.The other studies (n=9) used correlation coefficients to compare anthropometric data with information gathered using the gold standard. This coefficient only measures the strength of the relation between two variables, and not concordance between them36. It is possible to obtain a high correlation coefficient and, at the same time, have the data return poor concordance. The statistical test of the correlation coefficient is irrelevant when concordance between continuous measurements is assessed36. The investigated studies did not accurately evaluate whether or not the anthropometric information showed good concordance with the gold standard.

In this studies that evaluated correlation coefficients, two studies evaluated BMI and sum of skinfolds in PLWHA. The use of this index as anthropometric information, poorly sensitive and specific to metabolic assessment, has encouraged researchers to develop other anthropometric methods to estimate the quantity of body fat22. In this review, among the development and use of other indexes to determine fat quantity through the anthropometry of PLWHA, we noted that the predictive equations for body fat and WC were strongly correlated with total body fat. On assessing fat types, we noted a strong correlation between WC and abdominal fat (r=0.89) and subcutaneous fat (r=0.88), suggesting that WC can be a good indicator for the quantification of fat, particularly abdominal fat, and could be used in the prior diagnosis of abdominal lipohypertrophy.

About the sum of the four skinfolds, this has been used to predict the quantity of fat, through predictive equations, based on the relation between subcutaneous fat, internal fat and body density37,38. In the light of the redistribution of body fat in PLWHA, it is necessary to discuss the feasibility of calculating the sum of skinfolds in order to diagnose changes, given that these predictive equations were developed and validated for healthy individuals and athletes22. It is worth emphasizing that we did not note any associations between the sum of skinfolds by body region in relation to fat types (total abdominal, visceral and central subcutaneous fat), demonstrating the need to reassess this method.

This review presented as a limitation the fact that they have not been researched congress summaries, as part of the gray literature. Publication bias, which occurs due to non-publication of studies with negative results, may have influenced the findings of this review.

CONCLUSIONS

This review found that nine of 11 investigated studies did not evaluate the correct statistical analysis use to assess if anthropometric information had good agreement with the gold standard, since the statistical tests used did not evaluate the correlation between continuous measures. Two studies that would answer central question of this review have proposed equations that used anthropometric information only for individuals using antiretroviral therapy, invalidating the answer to PLWHA without antiretroviral therapy. As a result of limitations in statistical treatment and sample size that studies selected in this review have is difficult to propose or not the assertion that anthropometry is a suitable method for the evaluation of the fat in PLWHA, especially without the use of antiretroviral therapy.

REFERENCES

1 UNAIDS. Global AIDS Response Progress Reporting 2016. Génova, Suiza: UNAIDS; 2016. [ Links ]

2 UNAIDS. People Living with HIV Stigma Index [Internet]. [citado 13 de diciembre de 2015]. Disponible en: Disponible en: http://www.stigmaindex.org / [ Links ]

3 Kibirige D, Ssekitoleko R. Endocrine and metabolic abnormalities among HIV-infected patients: a current review. Int J STD AIDS. 2013; 24(8): 603-11. [ Links ]

4 Stanley TL, Grinspoon SK. Body composition and metabolic changes in HIV-infected patients. J Infect Dis. 2012; 205(Suppl 3): S383-390. [ Links ]

5 Carr A, Emery S, Law M, Puls R, Lundgren JD, Powderly WG, et al. An objective case definition of lipodystrophy in HIV-infected adults: a case-control study. Lancet. 2003; 361(9359): 726-35. [ Links ]

6 Grunfeld C, Saag M, Cofrancesco J, Lewis CE, Kronmal R, Heymsfield S, et al. Regional adipose tissue measured by MRI over 5 years in HIV-infected and control participants indicates persistence of HIV-associated lipoatrophy. AIDS. 2010; 24(11): 1717-26. [ Links ]

7 Jacobson DL, Knox T, Spiegelman D, Skinner S, Gorbach S, Wanke C. Prevalence of, evolution of, and risk factors for fat atrophy and fat deposition in a cohort of HIV-infected men and women. Clin Infect Dis. 2005; 40(12): 1837-45. [ Links ]

8 Lichtenstein K, Balasubramanyam A, Sekhar R, Freedland E. HIV-associated adipose redistribution syndrome (HARS): etiology and pathophysiological mechanisms. AIDS Res Ther. 2007; 4: 14. [ Links ]

9 Valente AMM, Reis AF, Machado DM, Succi RCM, Chacra AR. Alterações metabólicas da síndrome lipodistrófica do HIV. Arq Bras Endocrinol Metabol. 2005; 49(6): 871-81. [ Links ]

10 Signorini DJHP, Monteiro MCM, de Andrade MFC, Signorini DH, Eyer-Silva WA. What should we know about metabolic syndrome and lipodystrophy in AIDS? Rev Assoc Med Bras (1992). 2012; 58(1): 70-5. [ Links ]

11 Hermsdorff HHM, Monteiro JBR. Gordura visceral, subcutânea ou intramuscular: onde está o problema? Arq Bras Endocrinol Metabol . 2004; 48(6): 803-11. [ Links ]

12 Lukaski HC. Methods for the assessment of human body composition: traditional and new. Am J Clin Nutr. 1987; 46(4): 537-56. [ Links ]

13 Wang ZM, Pierson RN, Heymsfield SB. The five-level model: a new approach to organizing body-composition research. Am J Clin Nutr . 1992; 56(1): 19-28. [ Links ]

14 Mazess RB, Barden HS, Bisek JP, Hanson J. Dual-energy x-ray absorptiometry for total-body and regional bone-mineral and soft-tissue composition. Am J Clin Nutr . 1990; 51(6): 1106-12. [ Links ]

15 da Costa RF. Composição corporal: teoria e prática da avaliação. São Paulo, Brasil: Manole; 2001. [ Links ]

16 WHO Expert Committee on Physical status. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. World Health Organ Tech Rep Ser. 1995; 854. [ Links ]

17 Oppliger RA, Nielsen DH, Vance CG. Wrestlers’ minimal weight: anthropometry, bioimpedance, and hydrostatic weighing compared. Med Sci Sports Exerc. 1991; 23(2): 247-53. [ Links ]

18 Rezende FAC, Rosado LEFPL, Priore SE, Franceschini SCC. Aplicabilidade de equações na avaliação da composição corporal da população brasileira. Rev Nutr. 2006; 19(3): 357-67. [ Links ]

19 de Moraes MM, da Veiga GV. Acurácia da gordura corporal e do perímetro da cintura para predizer alterações metabólicas de risco cardiovascular em adolescentes. Arq Bras Endocrinol Metabol . 2014; 58(4): 341-51. [ Links ]

20 Ribeiro EAG, Leal DB, de Assis MAA. Acurácia diagnóstica de índices antropométricos na predição do excesso de gordura corporal em crianças de sete a dez anos. Rev Bras Epidemiol. 2014; 17(1): 243-54. [ Links ]

21 Vasques ACJ, Priore SE, Rosado LEFPL, Franceschini S do CC. Utilização de medidas antropométricas para a avaliação do acúmulo de gordura visceral. Rev Nutr. 2010; 23(1): 107-18. [ Links ]

22 Rezende FAC, Rosado LEFPL, Franceschinni SCC, Rosado GP, Ribeiro RCL. Aplicabilidade do índice de massa corporal na avaliação da gordura corporal. Rev Bras Med Esporte. 2010; 16(2): 90-4. [ Links ]

23 Shamseer L, Moher D, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015; 349: g7647. [ Links ]

24 Moher D, Booth A, Stewart L. How to reduce unnecessary duplication: use PROSPERO. BJOG. 2014; 121(7): 784-6. [ Links ]

25 Beraldo RA, Vassimon HS, Navarro AM, Foss-Freitas MC. Development of predictive equations for total and segmental body fat in HIV-seropositive patients. Nutrition. 2015; 31(1): 127-31. [ Links ]

26 Florindo AA, Borelli A, Latorre MRDO, Segurado AAC, Rocha MS. Validation of equations of skinfold thickness for fat mass estimation in hiv/aids subjects: a comparison of dual energy x-ray absorptiometry and computed tomography of abdômen. Rev Bras Ativ Fís Saúde. 2008; 13(2): 75-83. [ Links ]

27 Aghdassi E, Arendt B, Salit IE, Allard JP. Estimation of body fat mass using dual-energy x-ray absorptiometry, bioelectric impedance analysis, and anthropometry in HIV-positive male subjects receiving highly active antiretroviral therapy. JPEN J Parenter Enteral Nutr. 2007; 31(2): 135-41. [ Links ]

28 Batterham MJ, Garsia R, Greenop P. Measurement of body composition in people with HIV/AIDS: a comparison of bioelectrical impedance and skinfold anthropometry with dual-energy x-ray absorptiometry. J Am Diet Assoc. 1999; 99(9): 1109-11. [ Links ]

29 Antunes R, Siqueira H, Jordão AA, de Albuquerque FJ, Marchioli A, Freitas F de, et al. Anthropometry and bioelectrical impedance analysis compared to dual-photon absorptiometry for the assessment of body composition of hiv-seropositive patients. Rev Chil Nutr. 2011; 38(4): 404-13. [ Links ]

30 Florindo AA, Latorre MRDO, dos Santos ECM, Borelli A, Rocha MS, Segurado AAC. Validação de métodos de estimativa da gordura corporal em portadores do HIV/AIDS. Rev Saude Publica. 2004; 38(5): 643-9. [ Links ]

31 Segatto AFM, Freitas IF, dos Santos VR, Alves KCLRP, Barbosa DA, Filho AMP, et al. Indices of body fat distribution for assessment of lipodysthrophy in people living with HIV/AIDS. BMC Res Notes. 2012; 5: 543. [ Links ]

32 Siqueira H, Jordao AA, Albuquerque FJ, Artioli A, Pontes J. 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. [ Links ]

33 Wang J, Kotler DP, Russell M, Burastero S, Mazariegos M, Thornton J, et al. Body-fat measurement in patients with acquired immunodeficiency syndrome: which method should be used? Am J Clin Nutr . 1992; 56(6): 963-7. [ Links ]

34 Mulligan K, Parker RA, Komarow L, Grinspoon SK, Tebas P, Robbins GK, et al. Mixed patterns of changes in central and peripheral fat following initiation of antiretroviral therapy in a randomized trial. J Acquir Immune Defic Syndr. 2006; 41(5): 590-7. [ Links ]

35 Meininger G, Hadigan C, Rietschel P, Grinspoon S. Body-composition measurements as predictors of glucose and insulin abnormalities in HIV-positive men. Am J Clin Nutr . 2002; 76(2): 460-5. [ Links ]

36 Petroski EL. Desenvolvimento e validação de equações generalizadas para a predição da densidade corporal. 1995. [Tesis de Doctorado]. [Santa María, Brasil]: Universidade Federal de Santa Maria; 1995. [ Links ]

37 Durnin JV, Womersley J. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J Nutr. 1974; 32(1): 77-97. [ Links ]

38 Blyth S. Karl Pearson and the Correlation Curve. Int Stat Rev. 1994; 62(3): 393-403. [ Links ]

Received: June 09, 2016; Accepted: September 13, 2016

* Dirección para correspondencia: nasernizon@hotmail.com

Authors state that there are no conflicts of interest in preparing the manuscript.

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License