Dietary glycemic factors, insulin resistance,

and adiponectin levels in acne vulgaris

Aslı Aksu Çerman, MD,a Ezgi Aktaş, MD,a I_lknur Kıvanç Altunay, MD,a Janset Erkul Arıcı, MD,a
Aysın Tulunay, PhD,c and Feyza Yener Ozturk, MDb
Istanbul, Turkey

See related letter on page 220

Background: There is increasing evidence to support the relationship between acne vulgaris and diet.

Objective: The aim of this study was to investigate possible associations among dietary glycemic index,
glycemic load, milk consumption, insulin resistance, and adiponectin levels in the pathogenesis of acne vulgaris.

Methods: The dietary glycemic index, glycemic load, milk consumption, fasting glucose, insulin, insulin-
like growth factor)-1, insulin-like growth factor binding protein-3, adiponectin, and homeostasis model
assessment of insulin resistance values of 50 patients with acne vulgaris and 36 healthy control subjects
were measured.

Results: Glycemic index and glycemic load levels were significantly higher (P = .022 and P = .001,
respectively) and serum adiponectin levels were significantly lower (P = .015) in patients with acne than in
the control subjects. There was an inverse correlation between serum adiponectin concentration and
glycemic index (P = .049, r =
0.212).

Limitations: This study used a cross-sectional design and the study population was limited to young,
nonobese adults.

Conclusion: A high-glycemic-index/-load diet was positively associated with acne vulgaris. Adiponectin
may be a pathogenetic cofactor contributing to the development of the disease. Further research on
adiponectin levels in patients with acne in terms of development of insulin resistance might be important in
this possible relationship. ( J Am Acad Dermatol 2016;75:155-62.)

Key words: acne vulgaris; adiponectin; forkhead box class O1; glycemic index; glycemic load; insulin
resistance; insulin-like growth factor-1; insulin-like growth factor binding protein-3; mammalian target of
rapamycin complex-1; milk consumption.

A cne vulgaris is a chronic inflammatory dis-
ease of the pilosebaceous unit.1 Recent
studies have shown a relationship between
acne and nutritional factors; accumulating evidence
from epidemiologic and controlled dietary studies
suggest that high-glycemic-load diets and milk
consumption in particular might promote the deve-
lopment or exacerbation of acne vulgaris.2-7 Some
authors are of the opinion that a high-glycemic-index
diet induces hyperinsulinemia, which subsequently
elicits endocrine responses and enhances androgen
synthesis, ultimately affecting the development of
acne through mediators such as androgens, insulin-
like growth factor (IGF)-1, and IGF binding protein
(IGFBP)-3.8,9 IGF-1 is known to stimulate the key
factors involved in acne pathogenesis, including
keratinocyte proliferation, sebocyte proliferation,
and sebum production.3,9 Smith et al10 showed that
a low-glycemic-load diet results in improvements in
acne severity and insulin sensitivity, as expressed by

From the Dermatologya and Endocrinologyb Departments, Şişli
Hamidiye Etfal Training and Research Hospital; and Immu-
nology Department, Marmara University Faculty of Medicine.c
Funding sources: None.
Conflicts of interest: None declared.
Accepted for publication February 23, 2016.
Reprint requests: Aslı Aksu Çerman, MD, Şişli Etfal Egitim ve Araştirma
Hastanesi, Etfal S, Halaskargazi Cad 34371 Şişli-Istanbul/Turkey.
E-mail: aksuasli@hotmail.com.
Published online April 6, 2016.
0190-9622/$36.00
Ó 2016 by the American Academy of Dermatology, Inc.
http://dx.doi.org/10.1016/j.jaad.2016.02.1220

155
156 Çerman et al J AM ACAD DERMATOL
JULY 2016

the homeostasis model assessment of insulin resis- Patients with a history of any topical or systemic
tance (HOMA-IR), suggesting that nutrition-related acne therapy were excluded from the study.
lifestyle factors might play a role in the pathogenesis Subjects with a history of eating disorders, acute or
of acne. Acne can be related to some endocrine chronic infection, diabetes mellitus (DM), hyperten-
diseases; the most common of these diseases in sion, ischemic heart disease, thyroid disorders,
females is polycystic ovarian syndrome, which is chronic renal or liver disease, polycystic ovarian
characterized by peripheral insulin resistance and syndrome, or amenorrhea were also excluded. All
hyperinsulinemia.11 participants signed a written
Adiponectin is an informed consent form
CAPSULE SUMMARY
adipocyte-derived hormone after being provided with a
that is produced mainly by d High-glycemic-index/-load diets have full explanation of the
subcutaneous fat; it exhibits been implicated in acne pathogenesis. purpose and nature of the
important anti-inflammatory, study, and all of its related
antioxidant, and antidiabetic
d Hypoadiponectinemia associated with
procedures. The study was
effects.12,13 Dietary glycemic a high-glycemic-index/-load diet may
approved by the Şişli
index and glycemic load augment the inflammatory response
Hamidiye Etfal Training and
have been shown to be in patients with acne.
Research Hospital local
inversely associated with adi- d A low-glycemic-index/-load diet may ethics committee.
ponectin concentrations.14,15 have therapeutic potential in acne Factors related to glucose
Adiponectin inhibits proin- management. metabolism and energy bal-
flammatory cytokines and ance that were previously
induces antiinflammatory implicated in the develop-
ones, and it down-regulates adhesion molecule ment of acne and glycemic index and glycemic load
expression, suppresses toll-like receptors and their were examined as possible etiologic factors. The
ligands, and increases insulin sensitivity.13,16 glycemic index values of foods containing carbo-
The objective of this study was to investigate the hydrates were calculated in relation to glucose or
relationships among acne, adiponectin, dietary gly- white bread as a reference food based on the
cemic factors, and insulin resistance. patient’s postprandial blood glucose response and
blood insulin levels. Glycemic load combines the
METHODS glycemic index value and the carbohydrate amount
Participants and protocol to quantify the overall estimated glycemic effect of
The study cohort was designed as 50 patients with standard portion sizes of foods. We also aimed to
acne vulgaris and 36 healthy control subjects, suit- investigate the relationships among dietary glyce-
able for testing for 1 adiponectin enzyme-linked mic index, glycemic load, and acne. For this pur-
immunosorbent assay (ELISA) kit to be worked on pose, the dietary patterns of all participants were
the same day, consisting of 86 plates, when test determined from self-reported weighed/measured
plates were excluded. The acne vulgaris group was food records compiled over the previous 7 days.
recruited from patients admitted to the dermatology Total calorie intake per day, along with mean
outpatient clinic at Şişli Hamidiye Etfal Training and glycemic load and glycemic index and total
Research Hospital, Istanbul, Turkey, and the healthy amounts of carbohydrates, protein, and lipids,
control group from hospital staff volunteers. Subjects were calculated using Australia-specific dietary
with normal body mass index (BMI) (18.5-25 kg/m2) analysis software (FoodWorks; Xyris Software,
were included in the study. Highgate Hill, Australia). Dietary glycemic index
The patients with acne vulgaris were subdivided and glycemic load were calculated using
into the following 3 categories according to the the following equations: dietary glycemic index =
International Consensus Conference on Acne classi- S(glycemic index for food item 3 proportion of
fication system: mild (few to several comedones, total carbohydrate contributed by item) and dietary
papules, and pustules; no nodules); moderate glycemic load = S([glycemic index for food
(several comedones, papules, and pustules; few to item 3 its carbohydrate content in grams]/100).
several nodules); and severe (numerous comedones, The glycemic index values used glucose as the
papules, and pustules; many nodules). At the initial reference, and Sydney University’s glycemic index
visit, each patient’s age, sex, weight, height, and and glycemic index database (produced by Sydney
duration of acne were recorded. BMI was calculated University) was accessed from its World Wide Web
as weight (kg)/height (m2). site (http://www.glycemicindex.com).
J AM ACAD DERMATOL
VOLUME 75, NUMBER 1
Abbreviations used:
BMI: body mass index
DM: diabetes mellitus
ELISA: enzyme-linked immunosorbent assay
HOMA-IR: homeostasis model assessment of
insulin resistance
IGF: insulin-like growth factor
IGFBP: insulin-like growth factor binding
protein
mTORC1: mammalian target of rapamycin
complex-1
Each participant was investigated for the presence
of insulin resistance. HOMA-IR levels were calcu-
lated according to the following formula: fasting
glucose (mmol/L) 3 fasting insulin (U/mL)/22.5.
Levels greater than 3 were accepted as indicators of
insulin resistance.
Laboratory analyses
Venous blood samples were drawn from the
participants between 09:00 and 11:00 hours after a
12-hour fasting period to assess glucose, insulin, IGF-
1, IGFBP-3, and adiponectin levels. Serum glucose
level measurements were performed with a Roche/
Hitachi analyzer (Cobas c 701; Roche Diagnostics,
Mannheim, Germany) using an enzymatic method.
Insulin concentrations were measured with an elec-
trochemiluminescence immunoassay (Cobas e 601;
Roche Diagnostics). IGF-1 and IGFBP-3 levels were
measured with an immunometric chemilumines-
cence assay (Immulite 2000; Siemens Healthcare
Diagnostics, Los Angeles, CA). Serum adiponectin
levels were assessed by ELISA with a commercially
available adiponectin ELISA kit (AssayMax human
adiponectin ELISA kit, catalog no. EA2500-1;
Assaypro LLC, St Charles, MO). Absorbance was
read at 450 nm in an ELISA plate reader. Serum
adiponectin levels were expressed as ng/mL, and the
minimum detectable dose of adiponectin was
accepted as 0.7 ng/mL.
Statistical analysis
Statistical calculations were performed using the
Number Cruncher Statistical System 2007 (NCSS,
Kaysville, UT) and Power Analysis and Sample Size
2008 statistical software (NCSS). The data were
described statistically in terms of mean 6 SD, median
and range, or frequencies (number of cases) and
percentages when appropriate. A Shapiro-Wilk test
and the visual assessment of the histograms Q-Q
plots, box plots, skewness, and kurtosis were used to
differentiate between normal and non-normal distri-
butions. Regarding these analysis, the data of glucose
levels, IGFBP-3, adiponectin, and glycemic load
Çerman et al 157
Table I. Demographic characteristics of subjects
Acne vulgaris Healthy control
n = 50 n = 36 P
Age, y, mean 6 SD 18.80 6 3.20 19.06 6 3.49 .747*
Sex, n (%) .636y
Male 22 (44) 14 (38.9)
Female 28 (56) 22 (61.1)
BMI, kg/m2, 21.16 6 1.86 21.69 6 1.58 .07*
mean 6 SD
Acne severity, n (%)
Mild 13 (26)
Moderate 14 (28)
Severe 23 (46)
BMI, Body mass index.
*P value for the Mann-Whitney U test comparing age and BMI
among patients with acne vulgaris and healthy control subjects.
y
P value for the Pearson x 2 test comparing gender ratio among
patients with acne vulgaris and healthy control subjects.
variables revealed normal distribution. The
between-group comparisons of these variables
were evaluated with independent samples t test,
whereas the data of the other variables were
nonnormally distributed and were evaluated with
Mann-Whitney U test. The Kruskal-Wallis test was
used to compare patients by acne severity. Pearson
x2 test was used to compare qualitative variables.
P values less than or equal to .05 were considered
statistically significant and P values less than or equal
to .001 were considered highly significant.
RESULTS
There were no substantial differences in gender
ratio, age, or BMI for each variable between the
patients with acne vulgaris and the healthy volun-
teers (P [ .05). There were 13 patients (26%) with
mild, 14 patients (28%) with moderate, and 23
patients (46%) with severe acne vulgaris (Table I).
The dietary glycemic index and glycemic load,
milk and dairy product consumption, serum fasting
glucose, insulin, HOMA-IR, IGF-1, IGFBP-3, and
adiponectin levels of the patients with acne vulgaris
and the control subjects are summarized in Table II.
Although milk and dairy product consumption,
serum glucose, insulin, IGF-1, IGFBP-3, and
HOMA-IR values of the acne vulgaris and control
groups did not differ significantly (P [.05), glycemic
index and glycemic load values were significantly
higher in the acne vulgaris group than in the healthy
control subjects (P = .022 and P = .001, respectively)
(Fig 1). In addition, mean 6 SD serum adiponectin
concentration was significantly lower in the patients
with acne vulgaris than in the healthy control sub-
jects (9.93 6 2.29 ng/mL
1 vs 11.28 6 2.74 ng/mL
1;
P = .015) (Table II, Fig 2). Serum adiponectin
158 Çerman et al J AM ACAD DERMATOL
JULY 2016

Table II. Dietary glycemic index, glycemic loads, milk consumption, and serum glucose, insulin, homeostasis
model assessment of insulin resistance, insulin-like growth factor-1, insulin-like growth factor binding protein-3,
and adiponectin levels of patients with acne vulgaris and healthy control subjects
Acne vulgaris Healthy control
n = 50 n = 36 P
Dietary glycemic index, mean 6 SD 47.42 6 6.60 44.52 6 6.58 .022*y
Dietary glycemic load, mean 6 SD (median) 79.04 6 23.31 (76.50) 63.36 6 18.50 (62.50) .001zx
Milk consumption [3 dd/wk, n (%) 23 (28.75) 17 (47.2) .911*
Glucose, mg/dL, mean 6 SD 84 6 7.72 83.67 6 8.33 .849z
Insulin, U/mL, mean 6 SD (median) 9.72 6 6.26 (8.40) 9.88 6 4.30 (8.91) .384*
HOMA-IR, mean 6 SD (median) 2.03 6 1.49 (1.70) 2.05 6 0.93 (1.82) .370*
IGF-1, ng/mL, mean 6 SD 312.08 6 109.79 317.33 6 113.25 .723*
IGFBP-3, ng/mL, mean 6 SD 4.97 6 0.83 4.94 6 0.84 .905z
Adiponectin, ng/mL, mean 6 SD 9.93 6 2.29 11.28 6 2.74 .015yz

dd, Average daily portion; HOMA-IR, homeostasis model assessment of insulin resistance; IGF, insulin-like growth factor; IGFBP, insulin-like
growth factor binding protein.
*Mann Whitney U test.
y
P \ .05.
z
Independent samples t test.
x
P \ .01.

Fig 1. Mean 6 SD glycemic index and glycemic load

values of the study groups. P \.05 compared with healthy
control subjects.
concentrations showed an inverse correlation with
glycemic index values (P = .049, r =
0.212) (Fig 3).
When the patients with acne vulgaris were eval-
uated according to disease severity, serum insulin,
IGF-1, and IGFBP-3 levels were higher in the patients
with severe acne vulgaris (mean 6 SD; 10.19 6 5.88,
327.04 6 102.6, and 5.14 6 0.7, respectively) than in
patients with mild acne vulgaris (mean 6 SD;
7.8 6 3.5, 279.5 6 126.2, and 4.6 6 0.9, respectively),
but the difference was not statistically significant
(P = .269, P = .239, and P = .176, respectively)
(Table III). Milk consumption did not differ signifi-
cantly among the acne vulgaris subgroups (P = .596)
Fig 2. Serum levels of adiponectin in patients with acne
vulgaris and healthy control subjects. P \ .05 compared
with healthy control subjects.
(Table III). The glycemic index values were signifi-
cantly higher in the patients with moderate and
severe acne vulgaris than in the patients with mild
acne vulgaris (P = .035) (Fig 4). There was a positive
correlation between acne severity and glycemic
index value (P = .014, r = 0.345) (Fig 5).
The multivariate linear regression analysis to
evaluate the relative contribution of BMI, age, and
gender ratio to glycemic index revealed that the
most important contributing factor was being in the
acne group (P \ .001, r = 0.215, 95% confidence
J AM ACAD DERMATOL
VOLUME 75, NUMBER 1
Fig 3. Serum adiponectin concentrations were signifi-
cantly and inversely correlated with glycemic index values
in the entire study group (r =
0.212, P = .049).
interval
5.769 to
0.038). The same analyses for
glycemic load and adiponectin levels also showed
that the most important contributing factor was
being in the acne group (P \ .001, r = 0.343, 95%
confidence interval
24,995 to
6.363; P \ .001,
r = 0.262, 95% confidence interval 0.272-2.433,
respectively).
DISCUSSION
Recent studies have suggested that as diets
Westernize, acne prevalence increases.8 Three major
compounds of the Western diet have been identified
in acne pathogenesis: (1) hyperglycemic carbohy-
drates, (2) milk and dairy products, and (3) saturated
fats.17-19 Recent reviews have provided a new
perspective on nutrient signaling in acne vulgaris
by both high glycemic load and increased insulin,
IGF-1, and leucine signaling because of dairy protein
consumption.5,17,20 The authors have proposed that
increased nutrient-sensitive kinase mammalian target
of rapamycin complex-1 (mTORC1) activity and
decreased nuclear levels of forkhead box class O1
transcription factor might aggravate or promote acne
development.5,17,20 Both situations enhance sebo-
cyte proliferation and sterol regulatory element-
binding protein-1c-mediated sebaceous lipogenesis,
and stimulate the expression of peroxisome
proliferator-activated receptor-g and the secretion
of androgen hormones and the kinase S6 kinase-1
(the important downstream substrates of
mTORC1).5,17-20 S6 kinase-1-mediated phosphoryla-
tion of insulin receptor substrate-1 down-regulates
insulin/IGF-1 signaling and then induces insulin
resistance.20 Nutrient-mediated stimulation of
mTORC1- S6 kinase-1 signaling explains the reported
associations among acne, Western diet, insulin
Çerman et al 159
resistance, and conditions associated with the meta-
bolic syndrome.17-20
Smith et al21 reported on a randomized, controlled
trial investigating the effect of a low-glycemic-index/-
load diet compared with a high-glycemic-index/-
load diet on acne. The low-glycemic-index/-load
groups had a significant decrease in acne count and
free androgen index, and a significant increase in
insulin sensitivity and IGFBP-1 compared with the
high-glycemic-index/-load groups. A cross-sectional
study identified higher dietary glycemic index among
participants with moderate to severe acne compared
with those with no or mild acne.22
In our study, even after adjustment for BMI, age,
and gender ratio, the dietary glycemic index and
glycemic load values were significantly higher in the
patients with acne vulgaris than in the healthy control
subjects. In addition, the glycemic index values were
significantly higher in patients with moderate and
severe acne vulgaris than in patients with mild acne
vulgaris, and it was positively correlated with disease
severity. This result has persuaded us to acknowl-
edge that a high-glycemic-index diet might play a role
as a pathogenetic cofactor that triggers the develop-
ment of acne vulgaris and increases its severity.
Although there was no statistically significant differ-
ence in the consumption of milk and dairy proteins
between the patients with acne vulgaris and the
control group, this result might not reflect long-term
consumption habits, as the diet records that provided
the data only covered the 7 days before the study.
Not all cross-sectional studies have demonstrated
an association between a low-glycemic-index/-load
diet and acne.23,24 Kaymak et al24 reported that no
significant differences were observed between pa-
tients with acne and control subjects in terms of
serum glucose, insulin, IGF-1, IGFBP-3, overall
glycemic index, dietary glycemic load, and leptin
levels. However, that study has been appraised for
using a diet assessment tool that was never validated,
and it focused only on carbohydrates. In addition,
only 5 patients (10.2%) in that study had severe
disease, making it statistically difficult to investigate
the association among the parameters of the study. In
our study, serum glucose, insulin, IGF-1, IGFBP-3,
and HOMA-IR values did not differ significantly
between the acne vulgaris and control groups.
When the patients with acne vulgaris were grouped
according to disease severity, serum insulin, IGF-1,
and IGFBP-3 concentrations were higher in the
patients with severe acne vulgaris than in those
with mild acne vulgaris, but the differences were
not statistically significant. This finding might be
explained by a true lack of difference or a wide
distribution variance.
160 Çerman et al J AM ACAD DERMATOL
JULY 2016

Table III. Dietary glycemic index, glycemic load, milk consumption, serum glucose, insulin, homeostasis model
assessment of insulin resistance, insulin-like growth factor-1, insulin-like growth factor binding protein-3, and
adiponectin levels according to acne severity
Acne severity
Mild, n = 13 Moderate, n = 14 Severe, n = 23
mean 6 SD (med) mean 6 SD (med) mean 6 SD (med) P
Glucose 83.46 6 8.04 (83.0) 84.21 6 8.59 (84.5) 84.17 6 7.32 (85.0) .874*
Insulin 7.80 6 3.46 (7.01) 10.74 6 8.55 (7.4) 10.19 6 5.88 (9.4) .269*
HOMA-IR 1.58 6 0.71 (1.5) 2.27 6 2.06 (1.6) 2.14 6 1.4 (1.9) .347*
IGF-1 279.54 6 126.16 (259.0) 317.71 6 106.99 (278.5) 327.04 6 102.56 (327.0) .239*
IGFBP-3 4.55 6 0.92 (4.4) 5.06 6 0.85 (5.4) 5.14 6 0.70 (5.0) .176*
Adiponectin 9.4 6 1.34 (9.9) 10.33 6 2.27 (10.0) 9.98 6 2.72 (10.2) .520*
Glycemic index 44.19 6 5.39 (44.4) 48.06 6 3.34 (48.6) 48.87 6 8.13 (50.0) .035*y
Glycemic load 75.85 6 25.43 (69.00) 81.43 6 27.84 (83.00) 79.39 6 19.73 (82.00) .806*
n (%) n (%) n (%)
Sex
Female 4 (30.8) 9 (64.3) 15 (65.2) .103z
Male 9 (69.2) 5 (35.7) 8 (34.8)
Milk
[3 dd/wk 7 (53.8) 6 (42.8) 13 (56.5) .596z
\3 dd/wk 6 (46.2) 8 (57.2) 10 (43.5)

dd, Average daily portion; HOMA-IR, homeostasis model assessment of insulin resistance; IGF, insulin-like growth factor; IGFBP, insulin-like
growth factor binding protein; med, median.
*Kruskal-Wallis test.
y
P \ .05.
z
Pearson x 2 test.

Fig 4. Glycemic index values in patients with acne

vulgaris according to disease severity. P \ .05 compared
mild acne vulgaris.

Adiponectin is the most abundant adipocyto-

kine in human beings, and it has been found to
improve impairments in glucose metabolism by
increasing insulin sensitivity.25 It inhibits mTORC1
activity by activating 59 adenosine monophosphate
(AMP)-activated protein kinase.26 It also has anti-
inflammatory effects that include the ability to sup-
press proinflammatory molecules and modulate
the expression of anti-inflammatory cytokines.13,16
Adiponectin levels are inversely correlated with
obesity and BMI in the general population.27
Recently, numerous studies have reported that a
high-glycemic-index/-load diet is associated with
Fig 5. Correlation of glycemic index values with acne
severity (r = 0.345, P = .014).
lower adiponectin concentrations.14,15,27 In our
study, the serum adiponectin levels of the patients
with acne vulgaris were significantly lower than
those of the healthy control subjects. Furthermore,
serum adiponectin levels showed a negative corre-
lation with glycemic index values. We hypothesize
that a hyperglycemic diet may lead to adiponectin
reduction, which may be related with the activation
of mTORC1 and the development or exacerbation of
J AM ACAD DERMATOL Çerman et al 161
VOLUME 75, NUMBER 1

Fig 6. Possible role of adiponectin in the pathogenesis of hyperglycemic diet-induced acne.

AMPK, 59 adenosine monophosphate (AMP)-activated protein kinase; FoxO1, forkhead box
class O1; IL, interleukin; IRS, insulin receptor substrate; LXR, liver X receptor; mTORC1,
mammalian target of rapamycin complex 1; NLRP3, Nod-like receptor family, pyrin domain
containing 3 inflammasome; PPAR, peroxisome proliferator-activated receptor; S6K1, S6 kinase
1; SG, sebaceous gland; SREBP, sterol regulatory element-binding protein; TG, triglyceride;
TLR2, toll-like receptor-2.

acne vulgaris. On the other hand, decreased adipo-
nectin levels might suppress anti-inflammatory cyto-
kines and activate proinflammatory cytokines (Fig
6). In 2002, Cordain et al8 reported that the Kitavan
Islanders of Papua, New Guinea, and the Ache
hunter-gatherers of Paraguay never develop acne
and they have markedly lower rates of obesity, DM,
hyperlipidemia, and cardiovascular diseases. The
authors suggested that the absence of acne in these
societies might be a direct consequence of their low-
glycemic-load diets. In 2012, Del Prete et al28 found a
close relationship between inflammatory acne and
insulin resistance in male subjects. In study by
Nagpal et al,29 the prevalence of insulin resistance
was significantly higher in male patients with post-
adolescent acne compared with the control subjects,
however the metabolic syndrome prevalence was
similar in both groups. Our cohort did not include
patients with postadolescent acne. Recent studies
have suggested that hypoadiponectinemia might
play an important role in insulin resistance, DM,
and atherosclerosis.13,16,25 Therefore, reduced adi-
ponectin in patients with acne might be a
contributing factor in the development of DM and
cardiovascular disease. Further research on adipo-
nectin levels in patients with acne in terms of the
development of DM and cardiovascular comorbidity
might highlight this possible relationship; this sug-
gestion should be verified through other studies,
particularly those involving adult patients with
postadolescent acne and obese patients with acne.
Several methodological limitations within this
study warrant further discussion. First, only young
adults were studied, and the sample size was decided
according to the size of the adiponectin ELISA kit
and not a power analysis. Second, obesity, as a
confounder, was an exclusion criterion and our
results applied only to nonobese patients with acne
vulgaris. Third, diet tracking was limited to a week
and might not represent a long-term diet trend.
162 Çerman et al
Fourth, this research study used a cross-sectional
design; because of the nature of this study design,
interpretations of the results are limited. Lastly, as
Smith et al21 indicated in their study, using the fasting
insulin index in small intervention trials might result
in an underestimation or overestimation of the
relationship between acne and the degree of hyper-
insulinemia; therefore, it should be noted as a
limitation of the study.
In conclusion, to our knowledge, this is the first
study to compare adiponectin levels and its associ-
ation with dietary factors in patients with acne
vulgaris and healthy control subjects. We do not
suggest that hypoadiponectinemia is a disease-
specific finding, but that it is most likely associated
with a high-glycemic-index/-load diet and might
possibly serve as a cofactor that augments the
inflammatory response in patients with acne vulga-
ris. Low-glycemic-index/-load diets may have thera-
peutic potential in the treatment of acne vulgaris,
because of the beneficial metabolic and immuno-
logic effects of these diets.
The scientific guidance of E. Cerman is deeply appre-
ciated. English-language editing of this manuscript was
provided by Scribendi.
REFERENCES
1. Thiboutot D, Gollnick H, Bettoli V, et al. New insights into the
management of acne: an update from the Global Alliance to
Improve Outcomes in Acne Group. J Am Acad Dermatol. 2009;
60:S1-S50.
2. Spencer EH, Ferdowsian HR, Barnard ND. Diet and acne: a
review of the evidence. Int J Dermatol. 2009;48:339-347.
3. Danby FW. Nutrition and acne. Clin Dermatol. 2010;28:598-604.
4. Bowe WP, Joshi SS, Shalita AR. Diet and acne. J Am Acad
Dermatol. 2010;63:124-141.
5. Melnik BC. Dietary intervention in acne: attenuation of
increased mTORC1 signaling promoted by Western diet.
Dermatoendocrinol. 2012;4:20-32.
6. Cordain L. Implications for the role of diet in acne. Semin Cutan
Med Surg. 2005;24:84-91.
7. Cordain L, Eades MR, Eades MD. Hyperinsulinemic diseases of
civilization: more than just syndrome X. Comp Biochem
Physiol A Mol Integr Physiol. 2003;136:95-112.
8. Cordain L, Lindeberg S, Hurtado M, Hill K, Eaton SB,
Brand-Miller J. Acne vulgaris: a disease of Western civilization.
Arch Dermatol. 2002;138:1584-1590.
9. Melnik BC, Schmitz G. Role of insulin, insulin-like growth
factor-1, hyperglycemic food and milk consumption in the
pathogenesis of acne vulgaris. Exp Dermatol. 2009;18:
833-841.
10. Smith RN, Mann NJ, Braue A, Makelainen H, Varigos GA. A
low-glycemic-load diet improves symptoms in acne vulgaris
J AM ACAD DERMATOL
JULY 2016
patients: a randomized controlled trial. Am J Clin Nutr. 2007;86:
107-115.
11. Franks S. Polycystic ovary syndrome. N Engl J Med. 1995;333:
853-861.
12. Lim S, Quon MJ, Koh KK. Modulation of adiponectin as a
potential therapeutic strategy. Atherosclerosis. 2014;233:
721-728.
13. Ouchi N, Walsh K. Adiponectin as an anti-inflammatory factor.
Clin Chim Acta. 2007;380:24-30.
14. Pischon T, Girman CJ, Rifai N, Hotamisligil GS, Rimm EB.
Association between dietary factors and plasma adiponectin
concentrations in men. Am J Clin Nutr. 2005;81:780-786.
15. Loh BI, Sathyasuryan DR, Mohamed HJ. Plasma adiponectin
concentrations are associated with dietary glycemic index in
Malaysian patients with type 2 diabetes. Asia Pac J Clin Nutr.
2013;22:241-248.
16. Fantuzzi G. Adiponectin in inflammatory and immune-mediated
diseases. Cytokine. 2013;64:1-10.
17. Melnik BC. Linking diet to acne metabolomics, inflammation,
and comedogenesis: an update. Clin Cosmet Investig Dermatol.
2015;8:371-388.
18. Danby FW. Acne: causes and practical management. Oxford:
Wiley; 2015:49-83.
19. Melnik BC, John SM, Plewig G. Acne: risk indicator for
increased body mass index and insulin resistance. Acta Derm
Venereol. 2013;93:644-649.
20. Melnik BC, Zouboulis CC. Potential role of FoxO1 and mTORC1
in the pathogenesis of Western diet-induced acne. Exp
Dermatol. 2013;22:311-315.
21. Smith RN, Mann NJ, Braue A, Makelainen H, Varigos GA. The
effect of a high-protein, low glycemic-load diet versus a
conventional, high glycemic-load diet on biochemical
parameters associated with acne vulgaris: a randomized,
investigator-masked, controlled trial. J Am Acad Dermatol.
2007;57:247-256.
22. Burris J, Rietkerk W, Woolf K. Relationships of self-reported
dietary factors and perceived acne severity in a cohort
of New York young adults. J Acad Nutr Diet. 2014;114:384-392.
23. Reynolds RC, Lee S, Choi JY, et al. Effect of the glycemic index of
carbohydrates on acne vulgaris. Nutrients. 2010;2:1060-1072.
24. Kaymak Y, Adisen E, Ilter N, Bideci A, Gurler D, Celik B. Dietary
glycemic index and glucose, insulin, insulin-like growth
factor-1, insulin like growth factor binding protein-3, and
leptin levels in patients with acne. J Am Acad Dermatol. 2007;
57:819-823.
25. Caselli C, D’Amico A, Cabiati M, Prescimone T, Del Ry S,
Giannessi D. Back to the heart: the protective role of
adiponectin. Pharmacol Res. 2014;82:9-20.
26. Lamming DW. Diminished mTOR signaling: a common
mode of action for endocrine longevity factors. Springerplus.
2014;3:735.
27. Silva FM, de Almeida JC, Feoli AM. Effect of diet on
adiponectin levels in blood. Nutr Rev. 2011;69:599-612.
28. Del Prete M, Mauriello MC, Faggiano A, et al. Insulin resistance
and acne: a new risk factor for men? Endocrine. 2012;42:
555-560.
29. Nagpal M, De D, Handa S, Pal A, Sachdeva N. Insulin resistance
and metabolic syndrome in young men with acne. JAMA
Dermatol. 2015;23:1-5.