ORIGINAL

Eifler RV Ultrasonography ARTICLE

in visceral fat and steatosis

The role of ultrasonography in the measurement of subcutaneous

and visceral fat and its correlation with hepatic steatosis*
O papel da ultrassonografia na medida da gordura subcutânea e visceral e sua correlação com a esteatose
hepática

Roberto Velloso Eifler1

Abstract Objective: To evaluate the sonographic measurement of subcutaneous and visceral fat in correlation with the grade of
hepatic steatosis. Materials and Methods: In the period from October 2012 to January 2013, 365 patients were
evaluated. The subcutaneous and visceral fat thicknesses were measured with a convex, 3–4 MHz transducer transversely
placed 1 cm above the umbilical scar. The distance between the internal aspect of the abdominal rectus muscle and the
posterior aortic wall in the abdominal midline was considered for measurement of the visceral fat. Increased liver
echogenicity, blurring of vascular margins and increased acoustic attenuation were the parameters considered in the
quantification of hepatic steatosis. Results: Steatosis was found in 38% of the study sample. In the detection of moderate
to severe steatosis, the area under the ROC curve was 0.96 for women and 0.99 for men, indicating cut-off values for
visceral fat thickness of 9 cm and 10 cm, respectively. Conclusion: The present study evidenced the correlation between
steatosis and visceral fat thickness and suggested values for visceral fat thickness to allow the differentiation of normality
from risk for steatohepatitis.
Keywords: Abdominal fat; Steatosis; Metabolic syndrome; Ultrasonography.

Resumo Objetivo: Avaliar as medidas ultrassonográficas da gordura subcutânea e da gordura visceral em comparação com o
grau de esteatose hepática. Materiais e Métodos: Foram avaliados 365 pacientes entre outubro de 2012 e janeiro de
2013. A gordura subcutânea e a gordura visceral foram medidas com transdutor convexo de 3 a 4 MHz colocado trans-
versalmente 1 cm acima da cicatriz umbilical. Para a gordura visceral, considerou-se a distância entre a face interna do
músculo reto abdominal e a parede posterior da aorta na linha média do abdome. A quantificação da esteatose hepá-
tica levou em consideração o aumento da ecogenicidade do fígado, a perda de definição das margens vasculares e o
aumento da atenuação acústica. Resultados: Em 38% da amostra constatou-se esteatose. Na detecção de esteatose
moderada a severa, a área sob a curva ROC foi 0,96 para mulheres e 0,99 para homens, indicando pontos de corte
para a espessura da gordura visceral de 9 cm e de 10 cm, respectivamente. Conclusão: Comprovou-se correlação
entre esteatose e espessura da gordura visceral e foram sugeridos valores para a espessura da gordura visceral que
permitem distinguir a normalidade da esteatose simples e do risco de esteato-hepatite.
Unitermos: Gordura abdominal; Esteatose; Síndrome metabólica; Ultrassonografia.

Eifler RV. The role of ultrasonography in the measurement of subcutaneous and visceral fat and its correlation with hepatic steatosis.
Radiol Bras. 2013 Set/Out;46(5):273–278.

INTRODUCTION Reaven(2) observed that obesity was many times greater risk for cardiovascular mor-
times associated with diseases such as ar- bidity than individuals without MS(4).
In the last decades, particularly in the re- terial hypertension, hyperglycemia, dys- Non alcoholic fatty liver disease
cent years, the rate of obesity has increased lipidemia and, mainly, that such hemody- (NAFLD) corresponds to the clinical mani-
among the populations in developed and namic and metabolic alterations had a com- festation of metabolic syndrome, covering
emerging Western countries, both in chil- mon connection with insulin resistance. a wide spectrum of liver lesions from be-
dren and adults(1). Early in the eighties, Based on such observations the concept of nign steatosis to non alcoholic steatohepa-
metabolic syndrome (MS) was consoli- titis (NASH) with high risk for progression
* Study developed at the Unit of Ultrasonography, Funda-
dated. to cirrhosis and hepatocellular carcinoma(5).
ção Riograndense Universitária de Gastroenterologia (Fugast), The prevalence of MS is currently esti- NAFLD can be diagnosed by imaging stud-
Porto Alegre, RS, Brazil.
1. MD, Specialist in Radiology and Imaging Diagnosis, Fun-
mated to be between 20% and 25% of the ies such as ultrasonography (US), com-
dação Riograndense Universitária de Gastroenterologia (Fugast), general population, with increasing indi- puted tomography (CT) and magnetic reso-
Porto Alegre, RS, Brazil.
Mailing Address: Dr. Roberto Velloso Eifler. Rua Cabral, 983/
ces(3). The relevance of the diagnosis of MS nance imaging (MRI). Although US dem-
302, Rio Branco. Porto Alegre, RS, Brazil, 90420-121. E-mail: became clear once its relationship with onstrates lower accuracy than CT and MRI
[email protected].
cardiovascular diseases was confirmed. in the detection of NAFLD, the low cost
Received March 7, 2013. Accepted after revision May 27,
2013. Individuals with MS present two to three and easy make this method the most uti-

Radiol Bras. 2013 Set/Out;46(5):273–278 273

0100-3984 © Colégio Brasileiro de Radiologia e Diagnóstico por Imagem
Eifler RV Ultrasonography in visceral fat and steatosis

lized diagnostic tool in the initial evalua- least the association between visceral fat spite of showing a promising pathway,
tion of liver parenchymal alterations(6). and MS components is already well estab- most of them remained as isolated experi-
US, like CT and MRI, cannot differen- lished(16,17). Imaging studies are the meth- ments lacking follow-up. The author’s con-
tiate steatosis from NASH, which still re- ods of choice to evaluate and quantify vis- clusion is that this is a further example of
mains as a capacity of the histopathologi- ceral fat. In 1983, Tokunaga et al. established what happens nowadays in the medicine
cal study(7). However, steatosis can be criteria for evaluating body fat by CT. Cur- field: the knowledge compartmentaliza-
quantified by US as mild (grade 1), mod- rently, CT is considered as the gold stan- tion. The evaluation of visceral fat is highly
erate (grade 2) and severe (grade 3), and dard for evaluating intra-abdominal fat (16), interesting for different medical specialties
such quantification is fairly correlated with but in truth it is highly expensive, lacks prac- such as gastroenterology, hepatology, endo-
CT, MRI and histopathological analysis(8,9). ticality and exposes the patient to ionizing crinology, cardiology and other medical
According to some authors(10), US may radiation. Criteria for evaluating visceral fat specialties devoted to the study of obesity
present 89% sensitivity and 93% specific- with good accuracy by MRI were also de- and metabolism. However, it is not in-
ity in the identification of steatosis. On the veloped, but this method, more than CT, is cluded in the sonographic routine. Sono-
other hand, other authors report sensitivity subjected to artifacts and, also, its variation graphists follow correct protocols which,
up to 94% and specificity up to 100%(11). coefficient is also higher(18). Anthropomet- however, are failing in meeting the specific
The criteria for quantifying steatosis ric measurements are most frequently uti- necessities of an increasing number of spe-
present little variations among the several lized in the evaluation of body fat, but such cialties.
authors. The classification proposed by method has demonstrated to be incapable The author’s objective is adding infor-
Saadeh et al.(9) is a good example: of differentiating visceral from subcutane- mation to the work of those researchers,
– Grade 1 (mild): diffuse increase in liver ous fat, besides the relatively high intra and particularly the Brazilian ones, such as
echogenicity, with normal visualization interobserver variability(19). Leite et al.(23), Ribeiro Filho et al.(24) and
of intrahepatic vessels and of the dia- In contrast to the disadvantages of CT, Diniz et al.(25), who are concerned in defin-
phragm. MRI and anthropometric measurements, ing normal and risk values for subcutane-
– Grade 2 (moderate): Blurred visualiza- US has shown to be a simple, low-cost ous and visceral fat thickening. The corre-
tion of intrahepatic vessels and of the method without radiation risk, and with lation with the grade of steatosis estab-
diaphragm. already proved reproducibility and reliabil- lished by US, whose criterion is already
– Grade 3 (severe): intrahepatic vessels, ity in the quantification of visceral fat(20–22). well defined, will be utilized by the authors
diaphragm and the posterior region of The sonographic technique consists in of the present study for determining a cut-
the liver cannot be visualized. separately measuring the abdominal, sub- off value to identify those individuals with
Recent studies correlating US with his- cutaneous and visceral fat thickness with a at higher risk for steatohepatitis.
topathological analysis confirm that US is 3–4 MHz transducer placed 1 cm above the
a relevant noninvasive tool for evaluating umbilical scar. The studies developed by
MATERIALS AND METHODS
NAFLD(12,13) and intend to demonstrate Armellini et al.(20) have demonstrated that
that the normal or grade 1 liver echogeni- the visceral fat thickness measured by this In the present study, 365 patients of a
city rules out NAFLD and removes the ne- technique is fairly correlated with the area clinic specialized in gastroenterology in
cessity of liver biopsy(14,15). of such tissue quantified by CT. In 2002, Porto Alegre, RS, Brazil were consecu-
In the last years, another important role Leite et al. defined the values of 8 cm as tively and randomly assessed regardless of
has been assigned to US in the evaluation visceral fat thickness for women and 9 cm their complaints and symptoms, in the pe-
of MS, besides the quantification of steato- for men as associated with a higher cardio- riod from October 15, 2012 to January 25,
sis. As already seen, obesity is closely re- vascular risk(23). In 2003, Ribeiro Filho et 2013. Only chronic diseases such as alco-
lated to the diagnosis of MS. On the other al. proposed the value of 7 cm for the di- holism, cirrhosis, lymphoma, and renal fail-
hand, it was observed that obesity, as usu- agnosis of visceral obesity in women(24). In ure, besides history of bariatric or aesthetic
ally evaluated by anthropometric measure- 2009, Diniz et al., in a study involving 50 surgery determined exclusion from the
ments, presented idiosyncrasies, for ex- patients indistinct of gender, established present study.
ample, individuals with low body mass mean values with respective standard de- All the patients were evaluated by an ex-
índex might present high incidence of typi- viations corresponding to 2.64 cm ± 1.37 perienced sonographist whose daily prac-
cal alterations of MS. This has called the for subcutaneous fat thickness and 6.84 cm tice at least for ten years has involved the
attention to the fact that the fat distribution, ± 2.38 for visceral fat thickness(25). quantification of steatosis and during the
rather than the total excess of fat, would be Considering the author’s activities in- preceding ten months had measured subcu-
related to the insulin resistance and, con- volving US in the investigation of gastroen- taneous and visceral fat on a daily basis.
sequently, to the metabolic syndrome(3). terological conditions, the mentioned stud- The measurement of perirenal fat was not
Since then, the relationship between sub- ies originality, simplicity and efficacious- considered because of its still controversial
cutaneous/gluteofemoral/visceral fat and ness have called his attention. However, reproducibility.
the insulin action have been studied exten- such studies have called the author’s atten- The scans were performed with a single
sively and nowadays one can affirm that at tion principally because of the fact that, in Toshiba Nemio XG US apparatus with a

274 Radiol Bras. 2013 Set/Out;46(5):273–278

Eifler RV Ultrasonography in visceral fat and steatosis

convex, 3–4 MHz transducer. The utiliza-

tion of a linear transducer for measuring
subcutaneous fat was not considered nec-
essary since the differences between mea-
surements with convex and linear transduc-
ers were non-significant, so the technique
proposed by Radominski et al.(21) was uti-
lized.
The criteria for steatosis quantification
followed the classification developed by
Saadeh et al.(9), except for the classification
of mild steatosis (grade 1) where the pres-
ence of small focal hypoechoic areas on the
liver parenchyma, particularly in the peri-
portal and perivascular regions, prevailed
over the distinction between the hepatic
echotexture and the right kidney cortex A
echotexture which is fairly dependent on
subjectivity and on an ideally standardized
renal echotexture. Focal hypoechoic areas
are caused by venous systems (generally
cholecystic or parabiliary) independent or
relatively independent from the portal sys-
tem which locally reduce the portal perfu-
sion and, consequently, determine focal
metabolic changes which generate areas
free from adipose tissue accumulation(26).
Measurements of subcutaneous and vis-
ceral fat thickness were performed with the
patient positioned in dorsal decubitus and
right arm elevation, and the convex 3–4
MHz transducer cross-sectionally placed
on the midline, 1 cm above the umbilical
scar, during the expiratory phase, without
pressure on the abdomen in order not to
distort the measurement(25). Subcutaneous B
fat thickness corresponded to the distance Figure 1. A: Scheme of measurements of subcutaneous and visceral fat thickness. B: Sonographic im-
in centimeters between the skin and the age showing measurements of subcutaneous and visceral fat thickness.
anterior surface of the linea alba which is
the tendinous raphe that unites the two and ellipsoid image; for the purposes of tion was established between the patients’
halves of the rectus abdominus muscle. measurement, such fat is included in the sex and presence of steatosis (Table 1).
Visceral fat thickness corresponded to the visceral fat thickness. The aorta is usually About 38% of the study population pre-
distance in centimeters between the poste- located at left from the midline, and once sented steatosis, without statistically signifi-
rior surface of the linea alba and the plane its image is identified, a horizontal line is cant difference between men and women.
of the posterior aortic wall (Figure 1). It is drawn, passing through its posterior wall up Based on data showing that US may
convenient to observe that the linea alba is to the midline. At the intersection of such present high sensitivity and specificity in
many times thick, allowing the distinction a line with the line from the linea alba (first the identification of steatosis(11), and consid-
between the anterior and posterior surfaces, caliper) the second caliper is positioned to ering that normal liver echogenicity prac-
or many times thin, showing up on the im- measure the visceral fat thickness. tically rules out histological NAFLD(14,15),
ages as a trace. In the latter case, the trace subcutaneous and visceral fat thicknesses
is considered as the anterior and posterior were evaluated in steatosis-free individu-
RESULTS
surfaces for the purpose of measurement. als. For the women the subcutaneous fat
Generally, there is accumulation of extra- In the present study, 365 patients in the thickness was 2.51 cm ± 1.94, and for the
peritoneal fat on the midline, right under age range between 16 and 92 years (mean men, 2.14 cm ± 1.38 (p = 0.004; Student t
the linea alba, showing up as hypoechoic 52.9 years) were evaluated, and a correla- test). Visceral fat thickness was 5.38 cm ±

Radiol Bras. 2013 Set/Out;46(5):273–278 275

Eifler RV Ultrasonography in visceral fat and steatosis

Table 1 Distribution of steatosis by sex. The 55 men with steatosis were distrib-
Number Steatosis-free Steatosis 1 Seatosis 2 Steatosis 3
uted according to degree of fatty infiltration
(Table 5).
Ptcients n % n % n % n % n %
Similarly to the women, subcutaneous
Women 234 64.1 149 63.7 49 20.9 28 12.0 8 3.4 fat thickness among the different grades of
Men 131 35.9 76 58.0 32 24.4 17 13.0 6 4.6 steatosis in men did not present any statis-
Total 365 100 225 61.7 81 22.2 45 12.3 14 3.8 tically significant difference. On the other
hand, a progressive and statistically signifi-
cant increase in visceral fat thickness was
2.74 for the women and 6.78 cm ± 2.66 for thickness variation among age groups, observed with the increase in steatosis.
the men (p < 0.001; Student t test). such variation was not considered as clini-
It was possible to observe that the sub- cally relevant. Statistical analysis
cutaneous fat is on average 17% thicker in Mean visceral fat thickness for all the Subcutaneous fat was disregarded for
women than in steatosis-free men. On the steatosis-free men was 6.78 cm ± 2.66. lacking statistical significance, and the
other hand, visceral fat was on average 26%
thicker in men than in steatosis-free women. Table 2 Distribution according age range of subcutaneous and visceral fat among steatosis-free
On the basis of such data, the authors de- women.
cided to pay special attention to the differ- Subcutaneous fat Visceral fat
ence between men and women as subcuta- Steatosis-free women (cm) (cm)
neous and visceral fat were quantified per Up to 29 years (28 patients): mean = 23.8 years 2.36 ± 2.32 4.75 ± 2.16
age range. 30 to 69 years (97 patients): mean = 50.2 years 2.63 ± 1.90 5.38 ± 2.60
≥ 70 years (24 patients): mean = 77.1 years 2.21 ± 1.44 6.08 ± 3.20
Female individuals
p* 0.106 0.002
Based on studies demonstrating that
* Variance analysis (ANOVA), post-hoc comparison – Tukey’s test.
obesity presents its highest peak between
the ages of 45 and 64 years, 149 steatosis-
free women were evaluated up to achieve Table 3 Distribution of subcutaneous and visceral fat according grade of steatosis among women.
a definitive distribution among three dis- Subcutaneous fat Visceral fat
tinctive and more homogeneous age groups Female individuals (cm) (cm)
(Table 2).
Steatosis 1 (49 patients): mean = 59.2 years 3.08 ± 1.86 8.31 ± 3.16
The mean visceral fat thickness for all
Steatosis 2 (28 patients): mean = 57 years 3.46 ± 2.14 10.04 ± 1.14
the 149 steatosis-free women was 5.38 cm
Steatosis 3 (8 patients): mean = 63 years 2.38 ± 1.48 13.13 ± 4.34
± 2.74. No statistically significant variation
p* 0.019 < 0.001
was observed in subcutaneous fat thickness
among age groups in steatosis-free women. * Variance analysis (ANOVA), post-hoc comparison – Tukey’s test.

The 85 women with steatosis were dis-

tributed according the degree of fatty infil- Table 4 Distribution according age range of subcutaneous and visceral fat in steatosis-free men.
tration (Table 3).
Subcutaneous fat Visceral fat
A progressive and statistically signifi- Steatosis-free men (cm) (cm)
cant increase in visceral fat thickness was
Up to 29 years (12 patients): mean = 22.9 years 2.17 ± 2.06 6.00 ± 2.08
observed with the increase in the grade of
30 to 69 years (45 patients): mean = 49.6 years 2.18 ± 1.16 7.04 ± 2.72
steatosis in women. On the other hand, the
≥ 70 years (19 patients): mean = 78.1 years 2.05 ± 1.40 6.63 ± 2.52
subcutaneous fat thickness did not present
p* 0.800 < 0.045
any relevant alteration among the different
grades of steatosis. * Variance analysis (ANOVA), post-hoc comparison – Tukey’s test.

Male individuals
Table 5 Distribution of subcutaneous and visceral fat according grade of steatosis in men.
As among women, the sample of 131
Subcutaneous fat Visceral fat
male steatosis-free patients was divided Male individuals (cm) (cm)
into three age groups (Table 4).
Also, no significant variation was dem- Steatosis 1 (32 patients): mean = 56.7 years 2.19 ± 1.38 8.72 ± 2.70

onstrated in subcutaneous fat thickness in Steatosis 2 (17 patients): mean = 49.0 years 2.29 ± 2.10 11.24 ± 1.80

the different age groups among steatosis- Steatosis 3 (6 patients): mean = 54.1 years 2.50 ± 3.28 14.00 ± 3.10

free men. Despite the statistical signifi- p* 0.741 < 0.001

cance observed as regards visceral fat * Variance analysis (ANOVA), post-hoc comparison – Tukey’s test.

276 Radiol Bras. 2013 Set/Out;46(5):273–278

Eifler RV Ultrasonography in visceral fat and steatosis

analysis regarding visceral fat thickness atosis and visceral fat thickness(13,25), the 2. For men, visceral fat thickness ≥ 8 cm
was based on the area under the ROC curve author has tried to define values for ab- suggests the presence of mild steatosis; and
to define: a) cut-off value among steatosis- dominal fatty tissue thickness that could ≥ 10 cm, suggests moderate to severe ste-
free individuals and individuals with mild differentiate normality from mild steatosis atosis with risk for steatohepatitis and
steatosis (S1); b) an optimum cut-off value and suspicion of steatohepatitis(14). It has metabolic syndrome.
among individuals at no risk and individu- already been demonstrated that the accu- The risk for steatohepatitis with visceral
als at risk for steatohepatitis. For this pur- mulation of visceral fat is related to the fat thickness > 9 cm for women and > 10
pose, the authors have opted for utilizing development of steatohepatitis, and that cm for men raises the possibility of indica-
the group of steatosis-free and S1 individu- such accumulation continuously influences tion for liver biopsy.
als as a reference against which to appraise histological alteration in NAFLD from Such values practically superimpose
the group including individuals with mod- early in the process of fat deposition in themselves on those reported by Leite et
erate (S2) and severe (S3) steatosis, accord- hepatocytes to the onset of inflammatory al.(23), who have studied 422 patients, in-
ing to the study developed by Shannon et alterations(29). One of the advantages of vestigating the correlation between visceral
al.(15), which has demonstrated that 92% of measuring the visceral fat thickness is the fat thickness and cardiovascular risk. Such
patients with S2 and S3 at US presented simplicity and objectiveness of the method, study has indicated a cut-off value of 7 cm
moderate to severe steatosis at biopsy; and allowing the avoidance of subjective vari- to predict a moderate cardiovascular risk
that most steatosis-free patients or patients ables in the quantification of steatosis that for both female and male individuals, while
with S1 at US did not present any steatosis is very operator dependent. Both steatosis the present study has found a cut-off value
or presented only mild steatosis at biopsy. and abdominal visceral fat are correlated of 7 cm for women and 8 cm for men to
The results are shown on Table 6. independently of cardiometabolic risk, but indicate the presence of mild steatosis. The
the associations are stronger for visceral fat study developed by Leite et al.(23) has also
than for steatosis(30). defined cut-off values of 8 cm for women
DISCUSSION
In the 365 patients included in the present and 9 cm for men to predict high cardiovas-
Considering that the prevalence of both study, no statistically significant difference cular risk. The present study has established
MS(3) and NAFLD(27) is around 20% to was observed in subcutaneous fat thickness cut-off values of 9 cm for women and 10
25% in the general population, and that, in between male and female individuals, age cm for men to predict high risk for steato-
patients with NAFLD selected for liver bi- groups and in relation to the presence or not hepatitis.
opsy (criterion: steatosis at US and TGP/ of steatosis. On the other hand, visceral fat Ribeiro Filho et al.(24), who have also
TGO serum levels > 1.5 above the normal thickness did not present any statistically developed an exemplary study correlating
level in more than two occasions at a one- significant difference between age groups visceral fat thickness with metabolic syn-
month interval), steatohepatitis without fi- in steatosis-free men, but, among women, drome components, have assessed 100 fe-
brosis may affect 31.1% of the individuals, it did present a significant increase between male patients, defining 7 cm as cut-off
and steatohepatitis with fibrosis may reach young (< 29 years) and adult women, and value above which the patients present car-
27% of cases(28), the authors have tried to between adult and elderly women (> 70 diovascular risk.
establish a noninvasive, low-cost method years). However, among patients with ste- As regards the study developed by Diniz
to aid in the identification of patients at atosis, visceral fat thickening presented sig- et al.(25) intended principally to evaluate the
higher risk for presenting the progressive nificant increase, proportional to the grade interobserver variability of the sonographic
forms of such diseases. of steatosis, both in men and women. method, the fact of not having differenti-
With basis on the US accuracy to detect Based on the ROC curve analysis, the ated men and women as well as grades of
steatosis(10,11), one should consider the pos- following cut-off values were obtained: steatosis puts their values obtained for vis-
sibility of utilizing such imaging method as 1. For women, visceral fat thickness ceral fat thickness (6.84 cm ± 2.38) within
an initial tool in the screening for NAFLD ≥ 7 cm suggests the presence of mild ste- the spectrum of the present study.
even in the presence of normal levels of atosis; and ≥ 9 cm suggests moderate to se- Previous studies approaching the
hepatic enzymes(15). On the other hand, vere steatosis with risk for steatohepatitis evaluation of visceral fat by US are practi-
considering the relationship between ste- and metabolic syndrome. cally concentrated on validating US as

Table 6 Cut-off values for mild steatosis and moderate to severe steatosis by the ROC curve analysis.

Women AUC Cut-off Specificity Sensitivity Positive predictive value Negative predictive value
N×1 0.91 7 cm 77% 84% 77.2% 83.7%
N and 1 × 2 and 3 0.96 9 cm 88% 94% 98.9% 59.6%

Men AUC Cut-off Specificity Sensitivity Positive predictive value Negative predictive value
N×1 0.84 8 cm 66% 78% 65.8% 78.1%
N and 1 × 2 and 3 0.99 10 cm 90% 100% 100% 67.6%

N, steatosis-free; 1, grade 1 steatosis; 2, grade 2 steatosis; 3, grade 3 steatosis; AUC, area under ROC curve.

Radiol Bras. 2013 Set/Out;46(5):273–278 277

Eifler RV Ultrasonography in visceral fat and steatosis

compared with CT, MRI and anthropom- as a predictor of steatohepatitis and meta- Ultrasonographic quantitative estimation of he-
etry. Armellini et al.(20), for example, is the bolic syndrome. patic steatosis in children with NAFLD. J Pediatr
Gastroenterol Nutr. 2011;53:190–5.
first reference for all studies involving vis-
REFERENCES 16. Yoshizumi T, Nakamura T, Yamane M, et al. Ab-
ceral fat and US, and even utilized three dominal fat: standardized technique for measure-
1. Ogden CL, Carroll MD, Curtin LR, et al. Preva-
different parameters to measure visceral fat lence of overweight and obesity in the United
ment at CT. Radiology. 1999;211:283–6.
from the linea alba: the splenic vein, the an- States, 1999-2004. JAMA. 2006;295:1549–55. 17. Wajchenberg BL. Subcutaneous and visceral adi-
pose tissue: their relation to the metabolic syn-
terior aortic wall and the posterior aortic 2. Reaven GM. Banting lecture 1988. Role of insu-
drome. Endocr Rev. 2000;21:697–738.
wall. Currently, the posterior aortic wall is lin resistance in human disease. Diabetes. 1988;
37:1595–607. 18. van der Kooy K, Seidell JC. Techniques for the
utilized because it is believed that intra-ab- measurement of visceral fat: a practical guide. Int
3. Ribeiro Filho FF, Mariosa LS, Ferreira SRG, et al.
dominal thickness values based on such J Obes Relat Metab Disord. 1993;17:187–96.
Gordura visceral e síndrome metabólica: mais que
19. Vasques ACJ, Priore SE, Rosado LEFPdL, et al.
parameter are more representative of the uma simples associação. Arq Bras Endocrinol
Metabol. 2006;50:230–8. Utilização de medidas antropométricas para a
visceral fat amount. avaliação do acúmulo de gordura visceral. Revista
4. Isomaa B, Almgren P, Tuomi T, et al. Cardiovas-
The measurement of subcutaneous fat cular morbidity and mortality associated with the
de Nutrição. 2010;23:107–18.
thickness did not demonstrate to be statis- metabolic syndrome. Diabetes Care. 2001;24: 20. Armellini F, Zamboni M, Robbi R, et al. Total and
683–9. intra-abdominal fat measurements by ultrasound
tically relevant neither in men nor in and computerized tomography. Int J Obes Relat
women. It may be utilized as treatment fol- 5. Farrell GC, Larter CZ. Nonalcoholic fatty liver
Metab Disord. 1993;17:209–14.
disease: from steatosis to cirrhosis. Hepatology.
low-up in individual cases. In spite of the 2006;43(2 Suppl 1):S99–112. 21. Radominski RB, Vezozzo DP, Cerri GG, et al. O
apparent disparity between the increase in uso da ultra-sonografia na avaliação da distribui-
6. Ma X, Holalkere NS, Kambadakone RA, et al. Im-
ção de gordura abdominal. Arq Bras Endocrinol
visceral fat and increase in subcutaneous aging-based quantification of hepatic fat: meth-
Metabol. 2000;44:5–12.
ods and clinical applications. Radiographics.
fat with the progression of obesity, the 22. Hirooka M, Kumagi T, Kurose K, et al. A tech-
2009;29:1253–77.
proposition of an index to measure the vis- 7. Schwenzer NF, Springer F, Schraml C, et al. Non-
nique for the measurement of visceral fat by ul-
trasonography: comparison of measurements by
ceral fat/subcutaneous fat ratio has not invasive assessment and quantification of liver
ultrasonography and computed tomography. In-
found statistical relevance. steatosis by ultrasound, computed tomography
tern Med. 2005;44:794–9.
and magnetic resonance. J Hepatol. 2009;51:433–
The limitation of the present study is re- 45. 23. Leite CC, Wajchenberg BL, Radominski R, et al.
lated to the lack of correlation with serum Intra-abdominal thickness by ultrasonography to
8. Joy D, Thava VR, Scott BB. Diagnosis of fatty
predict risk factors for cardiovascular disease and
parameters and with histological analysis. liver disease: is biopsy necessary? Eur J Gastroen-
its correlation with anthropometric measure-
terol Hepatol. 2003;15:539–43.
ments. Metabolism. 2002;51:1034–40.
9. Saadeh S, Younossi ZM, Remer EM, et al. The
24. Ribeiro-Filho FF, Faria AN, Azjen S, et al. Meth-
CONCLUSION utility of radiological imaging in nonalcoholic
ods of estimation of visceral fat: advantages of ul-
fatty liver disease. Gastroenterology. 2002;123:
trasonography. Obes Res. 2003;11:1488–94.
The present study demonstrated the cor- 745–50.
25. Diniz ALD, Tomé RAF, Debs CL, et al. Avalia-
relation between steatosis and measure- 10. Joseph AE, Saverymuttu SH, al-Sam S, et al.
ção da reprodutibilidade ultrassonográfica como
Comparison of liver histology with ultrasonogra-
ment of visceral fat thickness in 365 pa- phy in assessing diffuse parenchymal liver dis-
método para medida da gordura abdominal e vis-
ceral. Radiol Bras. 2009;42:353–7.
tients, suggesting the preference for the ease. Clin Radiol. 1991;43:26–31.
26. Yoshimitsu K, Honda H, Kuroiwa T, et al. Un-
latter in the evaluation of NAFLD, not only 11. Andrade JO, Andrade MHF, Andrade GF, et al.
usual hemodynamics and pseudolesions of the
because of its objectiveness and reproduc- Quantification of liver echogenicity for ultra-
noncirrhotic liver at CT. Radiographics. 2001;21
sonographic classification of nonalcoholic fatty
ibility, but also because it allows the defi- Spec No:S81–96.
liver disease. GED Gastroenterol Endosc Dig.
nition of differentiated quantitative criteria 2006;25:159–64. 27. Ong JP, Younossi ZM. Epidemiology and natural
history of NAFLD and NASH. Clin Liver Dis.
for men and women, as well as for differ- 12. Dasarathy S, Dasarathy J, Khiyami A, et al. Va-
2007;11:1–16, vii.
lidity of real time ultrasound in the diagnosis of
ent age groups. Additionally, it allows a 28. Cotrim HP, Parise ER, Oliveira CP, et al. Nonal-
hepatic steatosis: a prospective study. J Hepatol.
more accurate treatment follow-up. 2009;51:1061–7. coholic fatty liver disease in Brazil. Clinical and
histological profile. Ann Hepatol. 2011;10:33–7.
As regards visceral fat, values were sug- 13. Hamaguchi M, Kojima T, Itoh Y, et al. The sever-
ity of ultrasonographic findings in nonalcoholic 29. Eguchi Y, Mizuta T, Sumida Y, et al. The patho-
gested for normality and risk for steato- logical role of visceral fat accumulation in steato-
fatty liver disease reflects the metabolic syndrome
hepatitis for both male and female indi- and visceral fat accumulation. Am J Gastroenterol. sis, inflammation, and progression of nonalco-
viduals. 2007;102:2708–15. holic fatty liver disease. J Gastroenterol. 2011;46
Suppl 1:70–8.
Furthermore, it is suggested that the 14. El-Koofy N, El-Karaksy H, El-Akel W, et al. Ul-
trasonography as a non-invasive tool for detection 30. Liu J, Fox CS, Hickson D, et al. Fatty liver, ab-
measurement of visceral fat thickness is in- dominal visceral fat, and cardiometabolic risk fac-
of nonalcoholic fatty liver disease in overweight/
cluded in the routine abdominal ultra- obese Egyptian children. Eur J Radiol. 2012;81: tors: the Jackson Heart Study. Arterioscler
sonography studies, considering the prac- 3120–3. Thromb Vasc Biol. 2011;31:2715–22.
ticality and efficaciousness of this method 15. Shannon A, Alkhouri N, Carter-Kent C, et al.

278 Radiol Bras. 2013 Set/Out;46(5):273–278