Special Article

Normal-weight obesity syndrome: diagnosis, prevalence, and

clinical implications
Lana P. Franco, Carla C. Morais, and Cristiane Cominetti

The growing concern about the impact of overweight on health has led to studies
that shed light on types of obesity other than the classic model based on body
mass index. Normal-weight obesity syndrome is characterized by excess body fat in

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

individuals with adequate body mass index (18.5–24.9 kg/m2). This condition
increases the risk of cardiovascular morbidity and mortality and other conditions
associated with chronic diseases, such as insulin resistance, hypertension, and dysli-
pidemia. The aims of this review are to define the diagnostic criteria for normal-
weight obesity syndrome and to examine the risks associated with this condition in
order to promote preventive measures and early treatment for affected individuals.

INTRODUCTION have either an appropriate body fat percentage or an ex-

cess of fat that might be masked by the normal BMI.3
Excessive body fat is responsible for at least 2.8 million Excessive body fat despite normal body weight has
deaths per year worldwide. Overweight and obesity in- been described as normal-weight obesity syndrome
crease the risk of other noncommunicable diseases, (NWO).4 Normal-weight obesity is a condition in which
such as cardiovascular disease, diabetes mellitus, and a person has an adequate BMI but an increased body fat
certain types of cancer.1 The metabolic effects of body percentage and a greater risk of developing noncommu-
fat accumulation on blood pressure, lipid profile, insu- nicable chronic diseases. Although investigations are re-
lin sensitivity, inflammatory and oxidative states, and cent, estimates show that about 30 million Americans
on gene expression patterns have been a frequent target are affected by NWO.5,6 This prevalence is not well es-
of scientific research. tablished, however, and there is wide variation between
The World Health Organization (WHO) defines the studies carried out thus far. This variation is attrib-
obesity as excessive body fat accumulation, which is as- uted to aspects such as ethnic differences, diverse meth-
sociated with clear risks to health. To overcome the dif- odologies used to assess body composition, and different
ficulties associated with measuring and classifying the cutoff points established for the diagnosis of NWO.
percentage of body fat, the WHO established the body Individuals with NWO tend to develop certain char-
mass index (BMI) as the parameter for identifying over- acteristic health conditions, such as low-grade proinflam-
weight and obesity.2 matory status, increased oxidative stress, insulin
Body mass index, however, does not allow the as- resistance, and dyslipidemia, which lead to a higher risk
sessment of body composition, as it does not differenti- of cardiovascular disease, metabolic syndrome, and
ate fat-free mass from adipose tissue. Thus, an cardiovascular-related death.3,4,6–10 Furthermore, nutri-
individual with normal BMI (18.5–24.9 kg/m2) may genetic studies suggest that NWO may interfere with the

Affiliation: L.P. Franco and C. Cominetti are with the Postgraduate Program in Nutrition and Health, Faculty of Nutrition, Federal University
of Goias, Goi^ania, Goias, Brazil. C.C. Morais is with the Postgraduate Program in Health Sciences, Faculty of Medicine, Federal University of
Goias, Goi^ania, Goias, Brazil.
Correspondence: C. Cominetti, Faculty of Nutrition, Federal University of Goias, Rua 227, Quadra 68 s/n, Setor Leste Universitario, Goi^ania,
Goias, Brazil CEP 74.605-080. Email: ccominetti@ufg.br. Phone: þ55-(62)-3209-6270, ext 210.
Key words: body composition, inflammation, insulin resistance, obesity, oxidative stress.
C The Author(s) 2016. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For
V
Permissions, please e-mail: journals.permissions@oup.com.

doi: 10.1093/nutrit/nuw019
558 Nutrition ReviewsV Vol. 74(9):558–570
R
relationship between genetic polymorphisms and an in- metabolic disorders of lesser magnitude, which are usu-
dividual’s health.7,11–13 Considering that body composi- ally undiagnosed. Moreover, an increase in visceral fat
tion is not routinely assessed in outpatient care, it is is not always present.3 As a result, individuals with
important to characterize NWO syndrome and to iden- NWO are unaware of the risks to which they are ex-
tify the health risks associated with this condition. Clarity posed, in contrast to metabolically obese but normal-
about the diagnosis, prevalence, and clinical implications weight subjects, who present major metabolic changes
of NWO syndrome will help healthcare providers incor- that result in signs and symptoms as well as subsequent
porate specific actions into healthcare plans – including diagnosis. Indeed, individuals with NWO might eventu-
the periodic assessment of adiposity, even in people with ally present as being metabolically obese but of normal
adequate BMI – in order to maintain health and reduce weight if they are not correctly diagnosed.
the risk of disease in NWO populations.
Therefore, the aims of this review are to character- PREVALENCE AND CRITERIA FOR DIAGNOSIS

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

ize NWO syndrome, to examine the diagnostic criteria,
and to highlight factors associated with the risk of dis- As noted previously, NWO prevalence data differ be-
ease development in affected individuals. tween studies because of ethnic differences in popula-
tions and the lack of consensus on diagnostic criteria.
DEFINITIONS AND RELATED CONCEPTS Normal-weight obesity has been studied more com-
monly among women, in whom it is highly prevalent,
Overweight and obesity are defined as the excessive ac- suggesting that sex hormones may play a role in the de-
cumulation of body fat, which represents a risk to velopment of this condition.10
health, according to the WHO.2 The most commonly The study that first described NWO established
used parameter for classifying overweight and obesity normal BMI and a high percentage of body fat as the di-
in a population is BMI. The WHO also recommends agnostic criteria.4 The authors used dual-energy X-ray
the use of this index in association with measurement absorptiometry to measure body fat percentage, and the
of waist circumference and calculation of waist-to-hip cutoff point for diagnosis was 30%. The study only in-
ratio.1 cluded females in whom no metabolic changes or non-
Although BMI has high specificity to detect exces- communicable diseases had been previously detected,
sive body adiposity, its sensitivity is low. An important and NWO prevalence was not evaluated.
limitation is that BMI has no ability to distinguish fat- Studies have shown that individuals with NWO
free mass from fat mass; consequently, it cannot be used may or may not present with changes in other anthro-
to assess the distribution of body fat. Sex, age, genetic pometric parameters, such as waist circumference,
and environmental factors all influence body fat distribu- waist-to-hip ratio, waist-to-height ratio, and percentage
tion, and visceral (intra-abdominal or android) fat is a of android or gynoid fat. Therefore, these parameters
risk factor for the development of noncommunicable are typically used to provide complementary informa-
diseases, regardless of the total body fat content. In addi- tion when evaluating individuals clinically, but they are
tion, BMI cutoff points do not consider differences in not suitable for diagnosing NWO.3,4,6,9,10,15
the body proportions of different populations.5,14 Another consideration is that variation in the body
The need to properly diagnose individuals with ex- composition pattern may depend on ethnicity.5,14 For
cessive body fat prompted De Lorenzo et al.4 to define example, adjustment of the BMI reference value for the
NWO, a condition characterized by weight within nor- Asian population may help prevent overestimation of
mal limits according to BMI, but with a high percentage the prevalence of overweight and obesity. On the other
of body fat. hand, the use of the universal BMI cutoff point set by
It is important to differentiate individuals with the WHO has allowed the comparison of data from dif-
NWO from those who are metabolically obese but of ferent epidemiological studies in order to establish the
normal weight, a condition that has been studied for relation between BMI and the risk of noncommunicable
some decades. Individuals who are metabolically obese diseases.1,14
but of normal weight have an appropriate weight ac- Kim et al.15 found an NWO prevalence of 36%
cording to BMI but have changes related to metabolic (men) and 29% (women) in Koreans with normal
syndrome. These individuals usually have excessive vis- weight (BMI, 18.5–22.9 kg/m2 for Asians16). This study
ceral fat, hyperinsulinemia, insulin resistance, dyslipide- in Koreans included 5313 men and 6904 women older
mia, high blood pressure, and an increased risk for type than 20 years, and dual-energy X-ray absorptiometry
2 diabetes, cardiovascular events, and cardiovascular was used to measure the percentage of body fat. The au-
death. Normal-weight obesity differs from metabolic thors established the lowest percentage of body fat pre-
obesity with normal weight in that it is associated with sented by overweight individuals (BMI, 22.9 kg/m2) as

Nutrition ReviewsV Vol. 74(9):558–570

R
559
the cutoff point for NWO. Thus, men with a body fat cardiovascular conditions, as it is associated with the
percentage 20.6% and women with a body fat percent- development of dyslipidemia, insulin resistance, and
age 33.4% were included in the NWO group.15 changes in blood pressure as well as pro-oxidative ef-
Thus far, only one study about NWO in Latin fects and the exacerbation of low-grade chronic proin-
America has been published.9 It included 1222 young flammatory status. An increase in body fat percentage is
adults aged 23 to 25 years with normal BMI, 55.3% of usually linked to sedentary behavior and poor eating
whom were women. The prevalence of NWO among habits, which further aggravate the problem.1
the total sample was 9.1% (9.2% for men and 9.0% for Cardiovascular disease is the leading cause of death
women). Body fat was estimated by measurement of worldwide, accounting for more than 17.5 million
skinfold thickness, and the cutoff point for NWO was deaths in 2012 and estimated to cause approximately
the 90th percentile of the sum of the subscapular and 22.2 million deaths in 2030.26 Recent studies have de-
triceps skinfolds, applying Slaughter’s equation.17 scribed increased cardiovascular risk in subjects with

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

Some studies have used the quartiles method to di- NWO, emphasizing the importance of characterizing
agnose NWO. In these investigations, volunteers were NWO syndrome and the metabolic changes associated
separated into tertiles of body fat percentage according with it.5,8,21,27
to sex, and diagnosis of NWO was based on the upper
tertile cutoff point.6,8,10 Using the tertile criterion for di- Blood pressure, lipid profile, and homocysteine
agnosis, Romero-Corral et al.6 evaluated 6171 concentrations
Americans (3129 women) older than 20 years and
found an overall NWO prevalence of 33.4%. Body fat In an attempt to establish the relationship between
percentage was evaluated by bioelectrical impedance, NWO and cardiovascular risk, some authors have in-
and the cutoff point (upper tertile) used for classifica- vestigated lipid profile, blood pressure, apolipoproteins,
tion of NWO was 23.1% for men and 33.3% for women. and homocysteine concentrations in subjects with
From the results, the authors estimate that 30 million NWO. Some found an increased risk of cardiovascular
Americans have NWO. disease and mortality in this population. This relation-
Using the highest tertile of body fat percentage ship was observed only in women, suggesting a specific
(26% for men and 38% for women) as the cutoff association with sex and a possible involvement of hor-
value, Marques-Vidal et al.10 found an NWO prevalence monal metabolism.6,10
of 2.8% in men and 5.4% in women. The study popula- Shea et al.25 and Kim et al.21 evaluated the presence
tion consisted of 6188 Caucasians aged 35 to 75 years of cardiovascular risk factors in subjects with NWO. In
and classified into various BMI ranges. Kang et al.8 also the first study, Canadian individuals were evaluated for
used the upper tertile of body fat percentage for NWO the presence of hypertension, dyslipidemia, increased
classification (23.5% for men and 29.2% for homeostasis model assessment of insulin resistance
women), but they did not evaluate the prevalence of (HOMA-IR), and ultrasensitive C-reactive protein.
NWO syndrome. Those who presented two or more changes were con-
Marques-Vidal et al.18 used bioelectrical impedance sidered to be at high risk. Classifying individuals into
and various diagnostic cutoff points to calculate body fat tertiles of body fat percentage resulted in a greater prev-
percentage and to assess the prevalence of NWO, which alence of risk among those in the middle and upper ter-
ranged from 0.1% to 41.1% in women and from 0.1% to tiles (12.0% and 19.5%, respectively) when compared
3.6% in men. This study and the others mentioned above with those in the lowest tertile (7.4%).25 In the second
confirm the lack of consensus regarding NWO diagnos- study, Korean individuals with NWO had a higher risk
tic criteria and, hence, the discrepancies in the reported of developing one or more cardiovascular risk factors
data. Table 13,4,6,8–11,15,18–25 shows the differences be- (hypertension, dyslipidemia, and/or hyperglycemia)
tween studies. In light of this lack of consensus, there is a than individuals with an appropriate body fat percent-
need to establish cutoff points for body fat percentages to age (odds ratio [OR] ¼ 1.63, 95%CI 1.21–2.19 in men;
define obesity. Moreover, it is urgent that appropriate di- and OR ¼ 1.56, 95%CI 1.36–1.8 in women).21
agnostic criteria for NWO are established, given that ex- Among individuals with NWO syndrome, both
cessive body fat is associated with the development of mean blood pressure and prevalence of hypertension
disease, especially cardiovascular disease. seem to be higher.6,8,10,20 Blood pressure above 115/
75 mmHg has been linearly associated with increased
CONDITIONS ASSOCIATED WITH EXCESSIVE BODY FAT mortality from cardiovascular disease in the general
population. Indeed, hypertension is a major risk factor
Excessive body fat is a major risk factor for the de- for the development of heart failure and coronary artery
velopment of noncommunicable diseases such as disease.28 Small increases in blood pressure are related

560 Nutrition ReviewsV Vol. 74(9):558–570

R
 Table 1 Diagnostic criteria for normal-weight obesity (NWO), prevalence of NWO, and observations in subjects with NWO in 16 studies
Reference Sample Evaluation Diagnostic criterion Prevalence of NWO Observations in NWO subjects
method
BMI (kg/m2) Percent body fat
4
De Lorenzo et al. (2006) 74 Italian women DXA 18.5–24.9 >30% in women Not evaluated "LDL-C to HDL-C ratio

R
16–74 y "HDL-C
De Lorenzo et al. (2007)3 60 Italian women DXA 18.5–24.9 >30% in women Not evaluated "TNF-a, "IL-1a
20–68 y "IL-1b, "IL-6
"IL-8
Di Renzo et al. (2006)11 75 Italian women DXA 18.5–24.9 >30% in women Not evaluated #BMR
20–45 y

Nutrition ReviewsV Vol. 74(9):558–570

Di Renzo et al. (2010)19 60 Italian women DXA 18.5–24.9 >30% in women Not evaluated #NO2/NO3
20–35 y #Glutathione
"Lipid peroxides
#Nonprotein antioxidant capacity
Kang et al. (2014)8 82 NWO Koreans Bioelectrical 18.5–24.9 Upper tertile: Not evaluated "Visceral fat
33 men impedance 23.5% in men; "Systolic BP
49 women 29.2% in women "Fasting glucose
82 controls "Triacylglycerols
#HDL-C
"Subclinical vascular inflammation
Kim et al. (2014)15 12 217 adult Koreans: DXA 18.5–22.9 20.6% in men; Among normal-BMI "WC, "TC
5313 men 33.4% in women subjects: #HDL-C, "triacylglycerols
6904 women 32% total "Systolic BP (men)
Mean age, 44.6 y 36% in men; 29% in "Fasting glucose
women "HOMA-IR
"Diabetes (men)
"CVRF, "SAH
"Dyslipidemia
"Metabolic syndrome
Kim et al. (2015)20 2078 adult Koreans Bioelectrical 18.5–24.9 Upper tertile: 8.3% "PWV
1141 men impedance 25.4% in men; "Risk for soft coronary plaques
1026 women 31.4% in women
Mean age 6 SD, 53 6 9 y
Kim et al. (2013)21 12 386 Koreans Bioelectrical 18.5–24.9 25% in men; 30% in 12.9% total "Systolic BP, "WC
6534 men impedance women 4.5% in men; 21.9% in "TC, "triacylglycerols
5852 women women "LDL-C, "CVRF
30–49 y "Fasting glucose
"Fasting insulin
"HOMA-IR
"SAH (men)
"Hyperglycemia
"Dyslipidemia
Kosmala et al. (2012)22 73 NWO Polish adults DXA 18.5–24.9 20–39 y: Not evaluated "Fasting insulin
95 controls 19% in men; 32% in "HOMA-IR, "LDL-C, #HDL-C,
Mean age 6 SD, 38 6 7 y women "triacylglycerols

561
(continued)

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

 Table 1 Continued

562
Reference Sample Evaluation Diagnostic criterion Prevalence of NWO Observations in NWO subjects
method 2
BMI (kg/m ) Percent body fat
40–59 y: "CRP > WC
21% for men; 33% in "Abdominal fat
women "Android/gynoid fat rate
60–79 y: Myocardial dysfunction
24% for men; 35% in
women
Madeira et al. (2013)9 1222 adult Brazilians Skinfold 18.5–24.9 TSF þ SSF  P90, corre- Among normal-BMI "Prevalence of metabolic syn-
546 men thickness sponding to 23.1% in subjects: drome, insulin resistance, and
676 women men; 33.3% in 9.1% total hyperinsulinemia
23–25 y women 9.2% in men; 9.0% in "WC, #HDL-C
women "Hypertriglyceridemia
Marques-Vidal 1523 Caucasians Bioelectrical 18.5–24.9 1) 30% overall 1) 3.2% in men; 10.1% in Not evaluated
et al. (2008)23,a 648 men impedance 2) 29.1% in men; 37.2% women
875 women in women 2) 3.4% in men; 2.5% in
Mean age 6 SD, 38 6 17 y 3) 26%–31% in men; women
39%–43% in women 3) 3.9% in men; 0.6% in
(according to age) women
4) 26.8%–32.6% in men, 4) 2.6% in men; 0.5% in
35.4%–44.4% in women
women (according to Among normal-BMI
age) subjects:
1) 6.9% in men; 15.0% in
women
2) 7.2% in men; 3.7% in
women
3) 5.6% in men; 0.9% in
women
4) 3.9% in men; 0.7% in
women
Marques-Vidal 6188 Caucasians Bioelectrical 18.5–24.9 1) 30% overall Among normal-BMI Not evaluated
et al. (2008)18,b 3213 women impedance 2) 29.1% for men; subjects:
2912 men 37.2% in women 1) 0.7% in men; 27.8% in
35–75 y 3) 26%–31% in men; women
39%–43% in women 2) 1.0% in men; 6.9% in
(according to age) women
4) 28.1%–32.6% in men; 3) 0.8% in men; 1.4% in
35.9%–44.4% in women

R
women (according to 4) 0.6% in men; 2.5% in
age) women
5) Fat mass index: 5) 0.1% in men; 0.1% in
women
(continued)

Nutrition ReviewsV Vol. 74(9):558–570

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022
 Table 1 Continued
Reference Sample Evaluation Diagnostic criterion Prevalence of NWO Observations in NWO subjects
method
BMI (kg/m2) Percent body fat
8.3 kg/m2 in men;

R
11.8 kg/m2 in
women
1) De Lorenzo et al.
(2006)
2) Zhu et al. (2003)
3) Gallager et al. (2000)

Nutrition ReviewsV Vol. 74(9):558–570

4) Kyle et al. (2001)
5) Kyle et al. (2003)
Marques-Vidal et al. (2010)10 6188 Caucasians Bioelectrical 18.5–24.9 Upper tertile: 2.8% in men; 5.4% in "Cardiovascular risk (women)
3213 women impedance 26% in men; 38% in women "Systolic BP
2912 men women Dyslipidemia
35–75 y "Hyperhomocysteinemia
"Leptin (women)
Miazgowski and 145 Polish women DXA 18.5–24.9 2 analyses: Not evaluated "Diastolic BP
Safranow (2013)24 20–40 y 30% (De Lorenzo "Triacylglycerols
et al., 2006) "LDL-C
Upper tertile: 35% #HDL-C
"Android and gynoid fat
Romero-Corral et al. (2010)6 6171 North Americans Bioelectrical 18.5–24.9 Upper tertile: Among normal-BMI sub- Not evaluated
3129 women impedance 23.1% in men; jects: 33.4%
3042 men 33.3% in women
Mean age 6 SD, 41.3 6 0.31 y
Shea et al. (2012)25 977 Canadians DXA 18.5–24.9 Upper tertile: Among normal-BMI sub- "Systolic BP
785 women 20.8% in men; jects: 33.7% "Triacylglycerols
192 men 35.0% in women "HOMA-IR
20–79 y "CRP
"Prevalence of cardiometabolic
changes

Abbreviations: BMI, body mass index; BMR, basal metabolic rate; BP, blood pressure; CRP, C-reactive protein; CVRF, cardiovascular risk factors; DXA, dual-energy X-ray absorptiometry; HDL-C,
high-density lipoprotein; HOMA-IR, homeostasis model assessment of insulin resistance; IL-1a, interleukin 1a; IL-1b, interleukin 1b; IL-6, interleukin 6; IL-8, interleukin 8; LDL-C, low-density li-
poprotein cholesterol; NO2/NO3, nitrite to nitrate ratio; NWO, normal-weight obesity; PWV, pulse wave velocity; SAH, systemic arterial hypertension; TC, total cholesterol; TNF-a, tumor ne-
crosis
a
factor a; TSF, triceps skinfold; SSF, subscapularis skinfold; WC, waist circumference.
Indicates study evaluated NWO prevalence according to several cutoff points established by different authors. Numbered percentages listed in the Percent body fat column correspond to
numbered prevalences in the Prevalence of NWO column.
b
Indicates study evaluated NWO prevalence according to several cutoff points established by different authors. These cutoff points, as well as the prevalence found by each author, are num-
bered according to the studies listed in the Percent body fat column. These studies are cited in the reviews by Marques-Vidal (2008).18

563
Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022
to the pathophysiological mechanisms of endothelial Excessive body fat is usually related to lower HDL-
dysfunction, which can progress to the development of C concentrations, higher triacylglycerol levels, and
atherosclerosis.29 changes in LDL-C particles. In obesity, the lipolysis pro-
Normal-weight obesity has been associated with an cess is altered: there is greater lysis of triacylglycerols
increased risk of subclinical atherosclerosis in a study and a higher release of free fatty acids in the blood-
that evaluated 2078 Koreans with normal BMI and no stream, which inhibits the action of lipoprotein lipase
history of coronary artery disease. The cutoff points of and increases the hepatic synthesis of very low-density
body fat percentage used for NWO diagnosis were lipoprotein, with a consequent increase in plasma tria-
25.4% for men and 31.4% for women (upper tertile). cylglycerols. In this situation, there is increased produc-
After adjustments for age, sex, and smoking habits, in- tion of cholesteryl ester transfer protein, which
dividuals with NWO showed higher values of pulse promotes the exchange between triacylglycerol from
wave velocity and a higher prevalence of soft coronary very low-density. and cholesterol from high-density li-

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

plaques compared with those with an adequate body fat poproteins. Triacylglycerol-rich high-density lipopro-
percentage (1474.0 6 275.4 cm/s vs 1380.7 6 234.3 cm/s tein is degraded and, in turn, there is a decrease in
and 21.6% vs 14.5%, respectively). Furthermore, the blood levels of high-density lipoprotein. This results in
presence of NWO was considered an independent risk hypertriglyceridemia, in which low-density lipoprotein
factor for the development of soft coronary plaques becomes richer in triacylglycerol and contains a lower
(OR ¼ 1.46; 95%CI, 1.03–2.07).20 amount of cholesterol. However, the triacylglycerol of
The atherogenesis process contributes to higher car- these particles is hydrolyzed by hepatic lipase, which re-
diovascular risk. Dyslipidemia, evidenced by increased sults in smaller and denser low-density lipoprotein par-
levels of total cholesterol, low-density lipoprotein choles- ticles with a greater potential to cause
terol (LDL-C), and triacylglycerols or by reduced atherogenesis.31,32
concentrations of high-density lipoprotein cholesterol Hyperhomocysteinemia has also been linked to in-
(HDL-C) as well as by the accumulation of oxidized creased cardiovascular risk.33 Homocysteine is a sulfhy-
low-density lipoprotein particles in the endothelium, is dryl substance produced from methionine conversion
among the factors that favor this process.30 in the crosstalk with folate metabolism. When concen-
Some studies have linked NWO with greater risk of trations of homocysteine in the blood are high, homo-
dyslipidemia.6,8,10 In a study of 164 Koreans, Kang cysteine is oxidized and there is an increase in pro-
et al.8 reported that individuals with NWO had higher oxidant substances that favor modification of low-
triacylglycerol levels and lower HDL-C concentrations density lipoprotein particles; activation of nuclear factor
compared with individuals with normal BMI and body jB; reduction in vascular endothelium antioxidant ac-
fat percentage. In Brazil, Madeira et al.9 observed an as- tivity; bioavailability of nitric oxide, endothelin-1, and
sociation between now, low HDL-C levels (OR ¼ 1.65; prostaglandin I2; and inhibition of C-reactive pro-
95%CI, 1.11–2.47) and hypertriglyceridemia tein.34,35 Furthermore, hyperhomocysteinemia has been
(OR ¼ 1.93; 95%CI, 1.02–3.64). In a study of 6171 associated with accumulation of body fat. In vitro and
American adults of both sexes, Romero-Corral et al.6 in vivo experimental studies suggest that lipolysis is in-
reported that individuals with NWO had lower serum hibited at high levels of homocysteine, which stimulates
HDL-C concentrations, higher LDL-C and triacylgly- the protein kinase activated by adenosine monophos-
cerol levels, and altered ratios of apolipoprotein B to phate, resulting in further accumulation of fat in
apolipoprotein A1 when compared with subjects with adipocytes.36
normal BMI and adequate body fat percentage. They To date, only one study has evaluated homocysteine
also observed a proportional relationship between in- levels in subjects with NWO. Marques-Vidal et al.10 eval-
creased body fat percentage and the highest risk of dys- uated 6125 Swiss adults of both sexes and reported
lipidemia and cardiovascular disease mortality in higher mean plasma homocysteine levels in subjects with
women with NWO (HR ¼ 1.06 for each point increase NWO than in those with normal BMI and body fat per-
in body fat percentage; 95%CI, 1.01–1.12).6 centage (9.52 6 0.24 mmol/L vs 9.33 6 0.08 mmol/L).
On the other hand, De Lorenzo et al.4 did not find In addition to differences in blood pressure, lipid
differences between lipid profiles of adult Italians with profile, and cardiovascular risk, the relationship be-
NWO and those of the control group, although they tween NWO and increased risk of inflammation has
found a higher ratio of LDL-C to HDL-C, a marker of been studied. Higher plasma concentrations of inflam-
cardiovascular risk, in individuals with NWO matory biomarkers (interleukins, C-reactive protein,
(2.14 6 1.19 vs 1.29 6 0.37). In this study, however, the and tumor necrosis factor a [TNF-a]) are related to the
NWO group presented higher HDL-C levels development of atherosclerosis and, consequently, to
(1.92 6 0.36 mmol/L vs 1.57 6 0.33 mmol/L).4 increased cardiovascular risk.3,6,8

564 Nutrition ReviewsV Vol. 74(9):558–570

R
Proinflammatory state gamma. Similarly, Di Renzo et al.19 also found higher
concentrations of cytokines (TNF-a, IL-1a, IL-2b, IL-6,
White adipose tissue has been functionally classified as and IL-15) in Italian women with NWO than in those
an endocrine organ on the basis of its production and with normal weight and body fat percentage. No differ-
secretion of proatherogenic adipocytokines (eg, resis- ence in mean IL-10 levels between groups was observed.
tin), antiatherogenic adipocytokines (eg, adiponectin), Marques-Vidal et al.10 and Kang et al.8 evaluated C-re-
prothrombotic factors (eg, inhibitor of plasminogen ac- active protein levels in subjects with NWO and also
tivator type 1), and proinflammatory factors (eg, leptin, found no significant differences between groups. In
TNF-a, interleukin [IL]-1, IL-6).3,37 contrast, in a study of American individuals, NWO was
Some researchers have reported associations be- associated with higher C-reactive protein levels.6
tween NWO and concentrations of adipocytokines. In a Interleukins and TNF-a regulate several processes
study of Swiss men and women between 35 and 75 years in the body, including lipid and glucose metabolism as

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

of age, Marques-Vidal et al.10 verified higher leptin con- well as immune functions related to the body’s inflam-
centrations in women with NWO than in those with an matory response. Therefore, these cytokines may be in-
adequate body fat percentage. No significant differences volved in the development of insulin resistance and
in adiponectin concentrations or in the leptin to adipo- dyslipidemia when their metabolism is altered.
nectin ratio between groups were observed. Increased Interleukin 1, IL-18, IL-8, and TNF-a act as proinflam-
leptin levels in American individuals with NWO were matory cytokines, while IL-10 is an anti-inflammatory
also reported by Romero-Corral et al.6 cytokine that acts by controlling the immune response.
Leptin is a hormone that acts as a metabolic marker Interleukin 6 appears to have different functions, having
of energy storage. Consequently, it decreases appetite either a proinflammatory or an anti-inflammatory role,
while increasing thermogenesis, hepatic lipid oxidation, depending on the tissue in which it is expressed.31
and lipolysis in adipocytes and skeletal muscle. Obese In addition to cytokines, C-reactive protein is also
individuals typically have high levels of leptin; however, involved in atherosclerosis, apparently by facilitating
they seem to be resistant to this hormone’s action.37,38 monocyte adhesion to vascular endothelium and mono-
Adiponectin, in turn, is an antiatherosclerotic and cyte migration into the vessel wall. Moreover, C-reac-
insulin-sensitizing protein. It inhibits hepatic glucose tive protein is associated with reduced activity of
production, increases the uptake of glucose by muscle endothelial nitric oxide synthase and increased migra-
and the oxidation of fatty acids by muscle and liver tion of macrophages to atherosclerotic lesions. Thus, C-
cells, and increases energy expenditure. Adiponectin reactive protein is classified as a biomarker of inflam-
also seems to have a protective effect against insulin re- mation and may indicate increased cardiovascular
sistance, atherogenesis, and proinflammatory status.37,38 risk.39
However, to date, no study has found a relationship be- On the basis of the studies presented here, it is
tween adiponectin levels and NWO. speculated that subjects with NWO show altered levels
On the other hand, some authors have reported an of proinflammatory cytokines, which may contribute to
association between NWO and a proinflammatory an increased risk of developing noncommunicable dis-
state.3,6–8 Indeed, since adipocytes produce several eases. Adding support to this, Kang et al.8 reported the
proinflammatory cytokines, subjects with a high body risk of vascular inflammation to be about 3 times higher
fat percentage, such as those with NWO, would be ex- in subjects with NWO than in control subjects
pected to show increased levels of these cytokines when (OR ¼ 3.07; 95%CI, 1.29–7.30; P ¼ 0.01).
compared with individuals with an adequate body fat Another study revealed that individuals with NWO
percentage. Therefore, the risk of developing cardiovas- present intermediate levels of proinflammatory cyto-
cular disease and metabolic syndrome would be greater kines in comparison with levels observed in normal and
in those individuals.31 obese individuals. The authors suggested this finding
In a study of 60 Italian women aged 20 to 35 years, might be related to the increased percentage of body fat
De Lorenzo et al.3 identified a proinflammatory state in in individuals with NWO. They also suggest that indi-
those diagnosed with NWO, but not in the control viduals with NWO present a proinflammatory state that
group. This state was characterized by increased plasma is “under development”, as the cytokines found (TNF-a,
levels of TNF-a, IL-1a, IL-1b, IL-6, and IL-8. The au- IL-1a, IL-1b, and IL1-IL8) are characteristic of an acute
thors also observed a correlation between TNF-a and inflammatory response. This condition, however, can
IL-6 levels and body fat percentage (r ¼ 0.55, P < 0.005). progress to chronic inflammation, a feature of obesity
Other cytokines were evaluated, and there were no sig- and a strong indicator of cardiovascular risk.3
nificant differences between groups for levels of C-reac- Researchers are investigating the effectiveness of
tive protein, IL-10, IL-2, IL-12p70, or interferon measures to reduce cardiovascular risk in individuals

Nutrition ReviewsV Vol. 74(9):558–570

R
565
with NWO. Di Renzo et al.40 evaluated the effects of the compared with those with adequate body fat percent-
consumption of dark chocolate (70% cocoa), which has age. However, there were no differences in the other pa-
known antioxidant and anti-inflammatory properties, rameters measured (C-reactive protein, adiponectin,
in 15 adult women with NWO (aged 20–40 years). and markers of liver function).10
After a washout period, patients were given chocolate, Another study conducted in the province of
100 g/d for 1 week. At the end of the study, the authors Navarra, Spain, investigated the relationship between
verified a significant increase in HDL-C levels (D%, body fat percentage, insulin resistance, and prevalence of
10.41 6 13.53) as well as a reduction in the total choles- diabetes in 4828 Spanish adults (3186 women) aged 18–
terol to HDL-C ratio (D%, 11.45 6 7.03), in the LDL- 80 years.42 The sample comprised 587 individuals with
C to HDL-C ratio (D%, 11.70 6 8.91), and in levels of normal BMI, 1320 who were overweight, and 2921 who
IL-1 receptor antagonist (D%, 32.99 6 3.84), suggest- were obese. The group with normal BMI and prediabetes
ing that the consumption of concentrated cocoa may or type 2 diabetes had a higher percentage of body fat

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

have cardioprotective effects in this population.40 compared with normoglycemic individuals, a finding
Despite the findings of the above studies, Marques- consistent for both sexes (35.5 6 7.0% vs 30.3 6 7.7%
Vidal et al.10 questioned the relationship between NWO and 25.2 6 9.0% vs 19.9 6 8.0% for women and men, re-
and a proinflammatory state because of differences in spectively). It was noted that a high percentage of body
the studies’ data. However, Kang et al.8 emphasize that fat was also associated with inflammation and markers of
body fat percentage may be even more important than oxidative stress. The high levels of insulin secretion
BMI in identifying the risk of developing a chronic low- found in individuals with NWO is probably a compensa-
grade proinflammatory state and, consequently, cardio- tory response to reduced insulin sensitivity.42
vascular disease. Furthermore, it seems that hypoxia in adipose tis-
sue, found in individuals with excessive body fat, results
Insulin resistance in increased production of reactive oxygen species and
activation of kinases. This induces the expression of
The relationship between excessive body fat, insulin re- proinflammatory cytokines and insulin resistance. The
sistance, and type 2 diabetes has already been widely de- overproduction of reactive oxygen species can also re-
scribed. Both a very low body fat percentage and an sult in mitochondrial dysfunction in liver and skeletal
excessive body fat percentage are associated with insulin muscle, which can cause lipid accumulation in these tis-
resistance. This paradox highlights the complexity of sues and contribute to the maintenance of insulin
the mechanisms involved in insulin resistance. Adipose resistance.19
tissue has long been recognized as an endocrine gland,
having several functions in the body, including glucose Oxidative stress
metabolism and regulation of inflammatory status. As
already described, adipocytes secrete a large variety of Oxidative stress is characterized as an imbalance between
molecules, including hormones such as leptin, cytokines the production of reactive oxygen species and an organ-
such as TNF-a, and fatty acids. Thus, adipose tissue ism’s antioxidant status and DNA damage repair capacity.
plays an essential role in energy balance and glucose In addition to reactive oxygen species, reactive nitrogen
homeostasis.41 species and free radicals can also cause cell damage.43
Madeira et al.9 reported that individuals with Oxidative stress interferes with both pancreatic secretion
NWO are at increased risk of developing early met- of insulin and glucose uptake by muscle and adipose tis-
abolic syndrome, especially insulin resistance, indicat- sues. It also promotes damage to cell membranes, pro-
ing a need for careful clinical evaluation of these teins, and DNA. Thus, it is directly involved in increasing
individuals. In a study conducted in Brazil, logistic the risk of developing noncommunicable diseases such as
regression models were adjusted for age, sex, and cardiovascular disease and diabetes.44
skin color. Normal-weight obesity was associated with Excessive body fat is usually associated with in-
metabolic syndrome (OR ¼ 6.83; 95%CI, 2.84–16.47), creased secretion of cytokines by adipose tissue, which
insulin resistance (OR ¼ 3.89; 95%CI, 2.39–6.33), may result in the production of reactive oxygen species.
hyperinsulinemia (OR ¼ 2.17; 95%CI, 1.24–3.80), and Additionally, the excess of free fatty acids promotes a
increased waist circumference (OR ¼ 8.46; 95%CI, greater accumulation of energy (as glucose and lipids),
5.09–14.04), and it was related to low HDL-C levels and with consequent increases in the oxidation of these sub-
hypertriglyceridemia. strates and in the formation of free radicals. This entire
In a study of 6125 Swiss individuals (3213 women), process, if not controlled, generates inflammatory and ox-
women with NWO showed a 1.63-times higher risk of idative stress in the body.44 As a result, individuals with
developing hyperglycemia (95%CI, 1.10–2.42) NWO are likely to have higher levels of oxidative stress.

566 Nutrition ReviewsV Vol. 74(9):558–570

R
 Di Renzo et al.19 evaluated oxidative stress status in influence gene expression without promoting changes
60 Italian women classified into 3 groups: normal BMI, in the DNA sequence. Nutrigenetics is the study of how
NWO, and overweight/obesity. They reported that indi- genetic variations may interfere with the interaction be-
viduals with NWO, in addition to presenting a proin- tween diet and the development of diseases. Therefore,
flammatory state, are exposed to higher oxidative stress studies of nutritional genomics aim to support the es-
than those with an adequate body fat percentage. In this tablishment of individualized nutritional recommenda-
study, body fat percentage was correlated with the for- tions and genetically personalized nutrition.45,46
mation of lipid peroxides and lower glutathione levels. The human genome presents highly complex varia-
The authors highlight lipid peroxidation as an indicator tions. When these variations exist in at least 1% of the
of initial oxidative damage to cell membranes, lipopro- population, they are classified as polymorphisms. The
teins, and other structures formed by lipids. most common genetic variations are the single nucleo-
Additionally, individuals with NWO showed lower ni- tide polymorphisms (SNPs), characterized by changes

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

tric oxide concentrations and a reduced nonprotein an- in a single DNA nucleotide. There are common poly-
tioxidant capacity. These results were attributed to the morphisms that occur in up to 50% of the population.
beginning of systemic oxidative stress.19 The authors These genetic variations can either be silent and not
also suggest that inflammation and oxidative stress sta- change gene expression, or they may result in the pro-
tus develop long before the symptoms of metabolic syn- duction of altered proteins that can influence several
drome appear. Indeed, individuals with NWO are metabolic processes.47
usually not diagnosed with metabolic syndrome; how- Several studies have investigated the relationship
ever, as reported in the literature, the metabolic changes between genetic polymorphisms and the development
observed in this population are important and should of NWO in an attempt to better understand NWO and
be investigated. Finally, the authors highlight the need its associated health risks (Table 2).7,11–13
to explore the genetic profile of individuals with NWO Di Renzo et al.7 investigated the frequency of the
in order to elucidate the mechanisms involved in the re- variable-number tandem repeat polymorphism in the
lationship between body composition, inflammation, IL-1 receptor antagonist gene, an anti-inflammatory cy-
oxidative stress, and other NWO-associated changes.19 tokine that competes with IL-1 for its cell receptors.
Polymorphisms in this gene have been linked to auto-
NUTRIGENETICS AND NORMAL-WEIGHT immune diseases, immune system disorders, atheroscle-
OBESITY SYNDROME rosis, and gastric cancer, and individuals who carry the
A2 allele are at increased risk of developing these disor-
Nutritional genomics is the study of interactions be- ders.7,48 The study in 110 Italian women revealed a sim-
tween food components and the human genome, on ilar frequency of the A1 allele in all groups.7 The A2
molecular, cellular, and systemic levels. It comprises allele was identified more frequently in the NWO
nutrigenomics, nutrigenetics, and nutritional epige- (12.5%) and overweight/obesity groups (17.5%) than in
nomics. Nutrigenomics investigates the effects of food the normal-BMI group (6.7%). The A3 and A4 alleles
intake on the gene expression pattern. Nutritional epi- were not found in any group and the A5 allele was ob-
genomics is the study of epigenetic mechanisms that served only in normal-weight and NWO subjects, but

Table 2 Genetic polymorphisms associated with normal-weight obesity (NWO) syndrome

Reference SNP gene Genotype/ Prevalence in Prevalence in Associated metabolic
risk allele NWO individuals control group alterations
Di Renzo et al. (2007)7 IL-1 receptor A2 12.5% 6.7% "IL-1b in NWO and over-
antagonist weight/obesity groups
Di Renzo et al. (2006)11 MTHFR TT 4.2% 28.8% Not evaluated
IL-15Ra GG 14.3% 36.8%
Di Renzo et al. (2008)12 IL-6 GG 59.0% 61.0% Increased IL-6 levels only in
NWO or obese individuals
NWO GG carriers had higher
HOMA-IR compared with
C-allele carriers
Di Renzo et al. (2013)13 TNF-a GG 73.3% 76.5% Lean body mass was lower in
NWO group than in control
group
Abbreviations: HOMA-IR, homeostasis model assessment of insulin resistance; IL-1, interleukin 1; IL-1b, interleukin 1b; IL-6, interleukin
6; IL-15Ra, interleukin 15 receptor a; MTHFR, methylenetetrahydrofolate reductase; SNP, single nucleotide polymorphism; TNF-a, tumor
necrosis factor a.

Nutrition ReviewsV Vol. 74(9):558–570

R
567
not in overweight and obese individuals. In the NWO TNF-a levels were similar in both the NWO group and
and overweight/obesity groups, A2 allele carriers had the normal-BMI/body fat percentage group. The GG
higher levels of IL-1b compared with noncarriers. In genotype was associated with the development of sarco-
the normal-BMI group, there was no significant differ- penia only in the obese group. Among the GG carriers,
ence in IL-1b concentrations between A2 allele carriers those with NWO had a lower amount of lean body
and noncarriers. No significant differences in IL-1a lev- mass when compared with the control group.13
els between the groups were found.7 In another study, Di Renzo et al.12 evaluated the re-
Another study in Italian women investigated the lationship between the IL-6 174G/C SNP and the pro-
presence of polymorphisms in the methylenetetrahy- gression of insulin resistance in subjects with NWO.
drofolate reductase gene in subjects with NWO.11 Interleukin 6, a cytokine secreted by adipose tissue, has
Methylenetetrahydrofolate reductase is a regulatory en- proinflammatory and, in some cases, anti-inflammatory
zyme of the carbon cycle, and its activity is altered by roles. Polymorphisms in its gene appear to increase the

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

the C677T SNP (exchange of an alanine for a valine at risk of development of insulin resistance and type 2 dia-
codon 222), which gives rise to a more labile protein betes. In this study in 150 Italian women, the genotype
with reduced enzymatic function. Because of this distribution (GG/GC/CC) was similar between groups
change, carriers of the variant allele (T) present lower (NWO subjects and controls). Individuals with NWO
folate concentrations, higher homocysteine levels and, showed higher IL-6 levels than those with normal BMI
hence, higher risk of developing noncommunicable dis- and adequate body fat percentage. When the groups
eases. Di Renzo et al.,11 however, observed a lower fre- were separated according to genotype, the GG genotype
quency of the variant homozygous genotype (TT) in the was associated with increased IL-6 levels only in NWO
NWO group (4.2%) than in the control group (28.8%). or obese individuals. In the group with normal BMI
The same study evaluated the presence of the SNP and adequate body fat percentage, IL-6 levels were not
rs3136618 (G/A) in the IL-15 receptor a (IL-15Ra) gene different between genotypes. The authors reported a
in individuals with NWO. Interleukin 15 is a cytokine positive correlation between IL-6 concentrations and
involved in various inflammatory processes and appears body fat percentage in GG carriers, while this correla-
to interfere in the metabolic regulation of adipose and tion was negative in CC carriers. The authors also
muscle tissue by controlling the deposition of body fat showed that the presence of one or two C alleles pro-
and acting anabolically in skeletal muscle. This relation- vides protection against insulin resistance, which may
ship might be due to the presence of IL-15Ra in adipose be related to the lower IL-6 secretion in these individ-
tissue. Like the methylenetetrahydrofolate reductase uals. Regarding HOMA-IR, there were no differences
SNP, the variant homozygous genotype (GG) was less attributed to the genotype in the control group; how-
frequent in the NWO group (14.3%) than in the control ever, among subjects with NWO, GG carriers had
group (36.8%). Thus, the authors reported an associa- higher HOMA-IR than the C-allele carriers.12
tion between the NWO phenotype and the presence of The studies described here indicate that individuals
wild alleles of methylenetetrahydrofolate reductase and with NWO may present unique characteristics with re-
IL-1 receptor antagonist genes, indicating the need for gard to the presence of genetic polymorphisms and
further studies to evaluate the frequency of these poly- their relation to health. Further studies can help identify
morphisms in subjects with NWO as well as the possi- the risk markers that determine the vulnerability of
ble metabolic consequences.11 these individuals to the development of disease.
Di Renzo et al.13 evaluated the relationship between
the TNF-a G308A SNP and sarcopenia in subjects with CONCLUSION
NWO. Sarcopenia is defined as the loss of muscle mass
associated with a decline in strength and physical per- Normal-weight obesity is a metabolic condition recently
formance. Tumor necrosis factor-a, a cytokine involved described and widely studied in order to evaluate its as-
in chronic inflammation, is able to induce insulin resis- sociation with the risk of disease development.
tance, anorexia, and weight loss. Furthermore, TNF-a However, affected individuals are seldom identified
appears to be involved in the metabolic regulation of through routine healthcare because of the limited obe-
muscle tissue and in the communication between mus- sity classification, which is based solely on anthropo-
cle tissue and adipose tissue. Individuals with a high metric parameters (BMI). Accurate diagnosis of NWO
body fat percentage tend to have higher TNF-a levels. is paramount, as there is sufficient scientific evidence
In this study, the genotype frequencies were 73.68% GG linking excessive body fat to a higher risk of cardiovas-
and 26.32% GA. However, there was no significant dif- cular mortality and other associated health conditions.
ference in the frequency of these genotypes between the Although most studies evaluate only women, and
NWO group and the control group. HOMA-IR and some present controversial results, many of them have

568 Nutrition ReviewsV Vol. 74(9):558–570

R
pointed to clear health risks for individuals with NWO. 13. Di Renzo L, Sarlo F, Petramala L, et al. Association between 308 G/A TNF-a poly-
morphism and appendicular skeletal muscle mass index as a marker of sarcopenia
Further epidemiological studies on NWO are needed in in normal weight obese syndrome. Dis Markers. 2013;35:615–623.
order to refine the diagnostic criteria for different popu- 14. Romero-Corral A, Somers VK, Sierra-Johnson J, et al. Accuracy of body mass index
in diagnosing obesity in the adult general population. Int J Obes.
lations. Such research would help healthcare providers 2008;32:959–966.
to evaluate individuals with NWO and to counsel them 15. Kim MK, Han K, Kwon HS, et al. Normal weight obesity in Korean adults. Clin
Endocrinol. 2014;80:214–220.
about preventive care and early treatment. 16. Joseph L, Wasir JS, Misra A. Appropriate values of adiposity and lean body mass
indices to detect cardiovascular risk factors in Asian Indians. Diabetes Technol
Ther. 2011;13:899–906.
17. Slaughter MH, Lohman TG, Boileau RA, et al. Skinfold equations for estimation of
Acknowledgments body fatness in children and youth. Human Biol. 1988;60:709–723.
18. Marques-Vidal P, Pécoud A, Hayoz D, et al. Prevalence of normal weight obesity in
The authors thank André Hedlund for reviewing the Switzerland: effect of various definitions. Eur J Nutr. 2008;47:251–257.
19. Di Renzo L, Galvano F, Orlandi C, et al. Oxidative stress in normal-weight obese
English of the manuscript. syndrome. Obesity (Silver Spring). 2010;18:2125–2130.

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

20. Kim S, Kyung C, Park JS, et al. Normal-weight obesity is associated with increased
risk of subclinical atherosclerosis. Cardiovasc Diabetol. 2015;14:58. doi:10.1186/
Author contributions. L.P.F. and C.C.M drafted the s12933-015-0220-5.
manuscript. C.C. provided supervision and critical revi- 21. Kim JY, Han SH, Yang BM. Implication of high-body-fat percentage on cardiometa-
bolic risk in middle-aged, healthy, normal-weight adults. Obesity.
sion of the manuscript. All authors contributed to and 2013;21:1571–1577.
approved the final version of the manuscript. 22. Kosmala W, Jedrzejuk D, Derzhko R, et al. Left ventricular function impairment in
patients with normal-weight obesity: contribution of abdominal fat deposition,
profibrotic state, reduced insulin sensitivity, and proinflammatory activation. Circ
Cardiovasc Imaging. 2012;5:349–356.
Funding/support. Master’s (L.P.F.) and doctoral (C.C.M.) 23. Marques-Vidal P, Chiolero A, Paccaud F. Large differences in the prevalence of
students receive scholarships from the Brazilian funding normal weight obesity using various cut-offs for excess body fat. E Spen Eur E J
Clin Nutr Metab. 2008;3:e159–e162.
agency “Coordenaç~ao de Aperfeiçoamento de Pessoal de 24. Miazgowski T, Safranow K, Krzy_zanowska-Swiniarska B, et al. Adiponectin, visfatin
Nıvel Superior.” No specific funding was received for and regional fat depots in normal weight obese premenopausal women. Eur J
Clin Invest. 2013;43:783–790.
writing this review. 25. Shea JL, King MT, Yi Y, et al. Body fat percentage is associated with cardiometa-
bolic dysregulation in BMI-defined normal weight subjects. Nutr Metab
Cardiovasc Dis. 2012;22:741–747.
26. World Health Organization. Draft Action Plan for the Prevention and Control of
Declaration of interest. The authors have no relevant Noncommunicable Diseases 2013–2020. Geneva: World Health Organization;
interests to declare. 2013:1–50.
27. Jean N, Somers VK, Sochor O, et al. Normal-weight obesity: implications for
cardiovascular health. Curr Atheroscler Rep. 2014;16:464. doi:10.1007/s11883-014-
0464-7.
28. Lewington S, Clarke R, Qizilbash N, et al. Age-specific relevance of usual blood
REFERENCES pressure to vascular mortality: a meta-analysis of individual data for one million
adults in 61 prospective studies. Lancet. 2002;360:1903–1913.
1. World Health Organization. Global Status Report on Noncommunicable Diseases 29. Leong XF, Ng CY, Jaarin K. Animal models in cardiovascular research:
2010. Geneva: World Health Organization; 2014:1–280. hypertension and atherosclerosis. Biomed Res Int. 2015;2015. doi:10.1155/2015/
2. World Health Organization. Physical Status: the Use and Interpretation of 528757.
Anthropometry. Geneva: World Health Organization; 1995:1–452. WHO Technical 30. Seo H, Oh H, Park H, et al. Contribution of dietary intakes of antioxidants to
Report Series no. 854. homocysteine-induced low density lipoprotein (LDL) oxidation in atherosclerotic
3. De Lorenzo A, Del Gobbo V, Premrov MG, et al. Normal-weight obese syndrome: patients. Yonsei Med J. 2010;51:526–533.
early inflammation? Am J Clin Nutr. 2007;85:40–45. 31. Jung UJ, Choi MS. Obesity and its metabolic complications: the role of
4. De Lorenzo A, Martinoli R, Vaia F, et al. Normal weight obese (NWO) women: an adipokines and the relationship between obesity, inflammation, insulin resistance,
evaluation of a candidate new syndrome. Nutr Metab Cardiovasc Dis. dyslipidemia and nonalcoholic fatty liver disease. Int J Mol Sci. 2014;15:
2006;16:513–523. 6184–6223.
5. Oliveros E, Somers VK, Sochor O, et al. The concept of normal weight obesity. 32. Klop B, Elte JWF, Cabezas MC. Dyslipidemia in obesity: mechanisms and potential
Prog Cardiovasc Dis. 2014;56:426–433. targets. Nutrients. 2013;5:1218–1240.
6. Romero-Corral A, Somers VK, Sierra-Johnson J, et al. Normal weight obesity: a risk 33. Sengwayo D, Moraba M, Motaung S. Association of homocysteinaemia with
factor for cardiometabolic dysregulation and cardiovascular mortality. Eur Heart J. hyperglycaemia, dyslipidaemia, hypertension and obesity. Cardiovasc J Afr.
2010;31:737–746. 2013;24:265–269.
7. Di Renzo L, Bigioni M, Del Gobbo V, et al. Interleukin-1 (IL-1) receptor antagonist 34. Di Minno MN, Pezullo S, Palmieri V, et al. Genotype-independent in vivo oxidative
gene polymorphism in normal weight obese syndrome: relationship to body com- stress following a methionine loading test: maximal platelet activation in subjects
position and IL-1 a and b plasma levels. Pharmacol Res. 2007;55:131–138. with early-onset thrombosis. Thromb Res. 2011;128:43–48.
8. Kang S, Kyung C, Park JS, et al. Subclinical vascular inflammation in subjects with 35. Dragani A, Falco A, Santilli F, et al. Oxidative stress and platelet activation in sub-
normal weight obesity and its association with body fat: an 18 F-FDG-PET/CT study. jects with moderate hyperhomocysteinaemia due to MTHFR 677 C!T polymor-
Cardiovasc Diabetol. 2014;4:13–70. phism. Thromb Haemostasis. 2012;108:533–542.
9. Madeira FB, Silva AA, Veloso HF, et al. Normal weight obesity is associated with
36. Wang Z, Pini M, Yao T, et al. Homocysteine suppresses lipolysis in adipocytes by
metabolic syndrome and insulin resistance in young adults from a middle-income
activating the AMPK pathway. Am J Physiol Endocrinol Metabol.
country. PLoS One. 2013;8:1–9.
2011;301:703–712.
10. Marques-Vidal P, Pécoud A, Hayoz D, et al. Normal weight obesity: relationship
with lipids, glycaemic status, liver enzymes and inflammation. Nutr Metab 37. Hajer GR, van Haeften TW, Visseren FL. Adipose tissue dysfunction in obesity, dia-
Cardiovasc Dis. 2010;20:669–675. betes, and vascular diseases. Eur Heart J. 2008;29:2959–2971.
11. Di Renzo L, Bigioni M, Bottini FG, et al. Normal Weight Obese syndrome: role of 38. Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol
single nucleotide polymorphism of IL-1 5Ra and MTHFR 677C!T genes in the re- Metab. 2004;89:2548–2556.
lationship between body composition and resting metabolic rate. Eur Rev Med 39. Yousuf O, Mohanty BD, Martin SS, et al. High-sensitivity C-reactive protein and car-
Pharmacol Sci. 2006;10:235–245. diovascular disease. J Am Coll Cardiol. 2013;62:397–408.
12. Di Renzo L, Bertoli A, Bigioni M, et al. Body composition and -174G/C interleukin-6 40. Di Renzo L, Rizzo M, Sarlo F, et al. Effects of dark chocolate in a population of nor-
promoter gene polymorphism: association with progression of insulin resistance mal weight obese women: a pilot study. Eur Rev Med Pharmacol Sci.
in normal weight obese syndrome. Curr Pharm Des. 2008;14:2699–2706. 2013;17:2257–2266.

Nutrition ReviewsV Vol. 74(9):558–570

R
569
41. Ye L, Huang Y, Zhao L, et al. IL-1b and TNF-a induce neurotoxicity through glutamate 45. Camp KM, Trujillo E. Position of the Academy of Nutrition and Dietetics: nutritional
production: a potential role for neuronal glutaminase. J Neurochem. 2013;125:897–908. genomics. J Acad Nutr Diet. 2014;114:299–312.
42. Gomez-Ambrosi J, Silva C, Galofré JC, et al. Body adiposity and type 2 diabetes: in- 46. Ordovas JM, Corella D. Nutritional genomics. Annu Rev Genomics Hum Genet.
creased risk with a high body fat percentage even having a normal BMI. Obesity 2004;5:71–118.
(Silver Spring). 2011;19:1439–1444. 47. Fenech M, El-Sohemy A, Cahill L, et al. Nutrigenetics and nutrigenomics: view-
43. Lay SL, Simard G, Martinez MC, et al. Oxidative stress and metabolic pathologies: points on the current status and applications in nutrition research and practice. J
from an adipocentric pont of view. Oxid Med Cell Longev. 2014;2014:908539. Nutrigenet Nutrigenomics. 2011;4:69–89.
doi:10.1155/2014/908539. 48. Arman A, Soylu O, Yildirim A, et al. O polimorfismo VNTR no gene codificador do
44. Marseglia L, Manti S, D’Angelo G, et al. Oxidative stress in obesity: a critical com- antagonista do receptor da interleucina 1 esta associado com a doença arterial
ponent in human diseases. Int J Mol Sci. 2015;16:378–400. coronariana. Arq Bras Cardiol. 2008;91:293–298.

Downloaded from https://academic.oup.com/nutritionreviews/article/74/9/558/1752252 by guest on 16 February 2022

570 Nutrition ReviewsV Vol. 74(9):558–570

R