Dietary glycemic index and glucose, insulin, insulin-like growth factor-I, insulin-like growth factor binding

protein 3, and leptin levels in patients with acne

Yesim Kaymak, MD,a Esra Adisen, MD,b Nilsel Ilter, MD,b Aysun Bideci, MD,c

Demet Gurler,d and Bulent Celik, PhDe

Ankara, Turkey

Background: Several isolated observations have suggested that acne can develop in groups when a highglycemic index diet is adopted.

Objective: This study was designed to examine associations among daily diet glycemic index, glycemicloads, serum insulin levels, and acne.

Methods: A total of 49 patients with acne and 42 healthy control subjects were included in the study. At theinitial visit, fasting glucose, insulin, insulin-like growth factor-I, insulin-like growth factor binding protein 3,and leptin levels were measured. A voluntary self-completed questionnaire was administered andparticipants were asked how frequently they consumed the specified amount of food. Overall glycemicindex and dietary glycemic load were calculated.

Results: No significant differences were observed between patients with acne and control subjects inserum glucose, insulin, leptin levels, overall glycemic index, or dietary glycemic load.

Limitations: The information and data obtained from this questionnaire were limited to patients’ ownrecollections.

Conclusion: Results of this study indicate that dietary glycemic index, glycemic load, and insulin levels do nothave a role in pathogenesis of acne in younger patients. ( J Am Acad Dermatol 2007;57:819-23.)

An association between diet and acne haslong been postulated and research in thisfield has been accelerating during the past

few years.1-3 Earlier reports investigating the impactof diet on acne had controversial results, with someindicating diets with carbohydrate and fat worsenacne,4 and others indicating no such relationship.5,6

From the Medical Health Centera; Faculty of Medicine, Department

of Dermatology,b Department of Pediatric Endocrinology,c and

Nutrition and Dieteticsd; and Vocational Education Facultye;

University of Gazi.

Funding sources: None.

Conflicts of interest: None declared.

Accepted for publication June 17, 2007.

Reprint requests: Yesim Kaymak, MD, Hosxdere cad. Sxair Baki Sok.

2/5 Y. Ayrancı-Ankara 06540 Turkey. E-mail: yesimkaymak@

yahoo.com.

Published online July 28, 2007.

0190-9622/$32.00

ª 2007 by the American Academy of Dermatology, Inc.

doi:10.1016/j.jaad.2007.06.028

Some researchers have suggested a relationshipbetween milk and acne.4,7 Recently, it has beensuggested that ingesting milk may stimulate endog-enous insulin-like growth factor (IGF)-I and IGF-Imay mediate some of the effects of comedogenicfactors, and high levels of IGF-I increase productionof sebum.7-10

Several isolated observations suggested that acnecan develop in groups when a high glycemic indexdiet is adopted.11 High glycemic foods or high car-bohydrate diet ([55% of energy from carbohydrate)in healthy men and boys elevates plasma insulinconcentrations, and may cause long-term hyperin-sulinemia and insulin resistance.12 Hyperinsulinemicdiets were thought to represent a previously

Abbreviations used:

ELISA: enzyme-linked immunosorbent assayIGF: insulin-like growth factorIGFBP: insulin-like growth factor binding protein

819

J AM ACAD DERMATOL

NOVEMBER 2007

820 Kaymak et al

unrecognized environmental factor in the develop-ment of acne by their influence on follicular epithe-lial growth, keratinization, and androgen-mediatedsebum secretion.13 Some authors suggest that highglycemic index diet-induced hyperinsulinemia elicitsendocrine responses, and promotes unregulatedtissue growth and enhanced androgen synthesis,which finally affect the development of acne throughmediators such as androgens, IGF-I, IGF bindingprotein (IGFBP)-3 and retinoid signaling path-ways.13-22

Serum leptin levels in human beings depend onsex, age, and body fat mass. Oral food intake and dietcomposition such as carbohydrate-rich food influ-ence leptin concentrations.23 Some observationshave demonstrated that leptin directly controls thesecretory activity of human ovarian cells with inter-relationship between leptin and the IGF/IGFBPsystem.24

The aim of this study was to investigate the effectof high glycemic index diets on pathogenesis ofacne.

METHODSParticipants and protocol

This prospective cohort study was designed toexamine associations among glycemic index ofdaily diet, serum insulin levels, and acne. A total of49 university students with severe, moderate, ormild acne seen in our outpatient department and42 healthy students who did not have acne (controlgroup) were included in the study. The local ethicscommittee approved the study protocol.

At the initial visit the patients’ age, sex, weight,height, and duration of acne were recorded. Bodymass index was calculated by weight (in kilograms)divided by square of height (in meters) as a measureof total adiposity.

Acne grading was performed using the Interna-tional Consensus Conference on Acne Classificationsystem25 with the following categories: mild, few toseveral comedones, papules, and pustules, and nonodules; moderate, several comedones, papules,pustules, and few to several nodules; and severe,numerous comedones, papules, and pustules, andmany nodules. The face, chest, neck, and back of allpatients with acne were examined closely underbright lighting.

The glycemic index, originally developed in 1981,is a relative comparison of the potential of variousfoods or combinations of foods to increase bloodglucose, based on equal amounts of carbohydrate inthe food.26 Glycemic index values are generallyreproducible representing regional differences.26

In 1997, the concept of glycemic load (glycemic

index 3 carbohydrate content/serving size) wasintroduced to assess the potential of a food toincrease blood glucose. Dietary glycemic load rep-resents the quality and quantity of carbohydrates andtheir interaction with a higher glycemic index havinga greater impact at higher carbohydrate intakes.27

We also aimed at investigating the associationsamong dietary glycemic index, glycemic load, andacne. For this purpose a voluntary self-completedquestionnaire was administered to all patients withacne and control subjects. Questionnaires weredesigned for this study by a dietitian. We examinedlow, moderate, and high glycemic index foods thathad been hypothesized to be associated with acne.Participants were asked how frequently they con-sumed the specified amount of food. A commonlyused portion size was specified for each food (eg,slice, piece, dessert spoon, spoonful, water glass,portion). There are no glycemic index values givenfor meat, poultry, fish, vegetables, cheese, or eggsbecause these foods contain little or no carbohy-drate. Even in large amounts, these foods wheneaten alone are not likely to induce a significantincrease in blood glucose.

Data from the food frequency questionnaire wereused to estimate overall glycemic index and glycemicload. Glycemic index values of foods were obtainedfrom published reports based on studies in NorthAmerica.28 Overall glycemic index was calculated bymultiplying the carbohydrate content (in grams) of agiven food item by the number of servings per day of

Table I. Sociodemographic characteristics of acneand control groups

Acne Control Test

Sex, No. (%)Female 30 (61.2) 26 (61.9) Chi-square

= 0.004;P = .947

Male 19 (38.8) 16 (38.1)Total 49 (53.8) 42 (46.2)

Age, y, mean (SD)Female 21.56 (1.38) 22.12 (1.33) t = �1.505;

P = .138Male 21.95 (2.09) 22.19 (1.60) t = �0.375;

P = .710Total 21.71 (1.68) 22.14 (1.42) t = �1.299;

P = .197BMI, mean (SD)

Female 21.12 (2.53) 21.68 (2.50) t = �0.826;P = .412

Male 21.83 (2.17) 25.18 (2.73) t = �4.043;P = .000*

Total 21.39 (2.40) 23.01 (3.08) t = �2.809;P = .006*

BMI, Body mass index.

*P \ .05.

J AM ACAD DERMATOL

VOLUME 57, NUMBER 5

Kaymak et al 821

Table II. Overall glycemic index, dietary glyemic loads and serum glucose, insulin, insulin-like growth factor-I,insulin-like growth factor binding protein 3, and leptin levels of acne and control groups

Acne (n = 49) Control (n = 42)

Mean (SD) Mean (SD) t P

Overall glycemic index 62.18 (11.73) 66.42 (11.07) �1.763 .081Dietary glycemic load 252.46 (139.33) 291.28 (176.11) �1.173 .244Glucose, mg/dL 84.80 (8.01) 81.40 (8.47) 1.960 .053Insulin, �U/mL 12.79 (7.20) 12.63 (7.03) 0,106 .916IGF-I, ng/mL 323.03 (139.97) 397.27 (121.03) �2.626 .010*IGFBP-3, ng/mL 59.51 (9.59) 49.97 (7.27) 4.244 .000*Leptin, ng/mL 20.44 (19.01) 20.01 (15.31) 0.118 .906Homa IR 2.68 (1.53) 2.58 (1.57) 0.288 .774

IGF, Insulin-like growth factor; IGFBP, IGF binding protein.

*P \ .05.

that food item and its glycemic index value, summingall food items reported, and dividing by the totalcarbohydrate in the diet. Total glycemic load wascalculated by multiplying the carbohydrate contentof a given food item by the number of servingsconsumed per day and its glycemic index value andsumming the values for all food items reported.29

When the reported glycemic index values for foodswere observed to vary across studies,28 we used themean of the reported values of glycemic index forthat food.

Each participant in the acne and control groupswere investigated for the presence of insulin resis-tance. Insulin resistance index (Homa-IR) levelswere calculated acording to the formula: (serumglucose/18 3 insulin)/22.5. Levels above 3 wereaccepted as indicators of insulin resistance.30

Analytic methodsAt the initial visit, fasting venous blood samples

were obtained from a forearm vein with needletechnique for glucose, insulin, IGF-I, IGFBP-3, andleptin levels. Tests were carried out in the laborato-ries of our department of biochemistry and depart-ment of pediatric endocrinology by commercial kits.

Measurement of serum glucose levels was per-formed using an autoanalyzer (Abbott-Aeroset,Chicago, Ill) with the Randox kits (DiagnosticSystem Laboratories Inc, Webster, Tex). The insulinenzyme-linked immunosorbent assay (ELISA) is anenzymatically amplified one-step sandwich-typeimmunoassay (Diagnostic System Laboratories Inc).IGFBP-3 levels were measured by DSL-10-6600Active IGFBP-3 ELISA (Diagnostic SystemLaboratories Inc). Serum IGF-I levels were measuredby DSL-10-2800 nonextraction IGF-I ELISA(Diagnostic System Labotatories Inc). Serum leptinlevels were measured by DSL-10-23100i human lep-tin ELISA (Diagnostic System Labotatories Inc).

Statistical analysisThe results were expressed as mean 6 SD. Statis-

tical comparisons were made using the Student t test,nonparametric chi square, and Kruskal-Wallis tests. Pless than .05 was accepted as the significance level.

RESULTSAmong the 49 university students with acne, 19

(38.8%) were male and 30 (61.2%) were female. Theages were between 19 and 34 years (mean: 20.37 6

1.944 years) and the duration of the condition lastedfrom 1 to 10 years (mean 3.7). Seventeen patients(34.6%) had mild, 27 (55.1%) had moderate, and 5(10.2%) had severe acne. Control group constitutedof 42 people, 16 (39.1%) male and 36 (61.9%) female.Patient and control groups were matched accordingto their age and sex. Table I shows the sociodemo-graphic characteristics and body mass index of thesample. There was no significant difference betweenthe two groups with respect to age, sex, or body massindex (P [ .05) (Table I).

Overall glycemic index, dietary glycemic loadsand serum fasting glucose, insulin, IGF-I, IGFBP-3,and leptin levels of patients with acne and controlsubjects are summarized in Table II. No significantdifferences in levels of overall glycemic index wereobserved between patients with acne and controlsubjects (P[.05). Serum glucose, insulin, and leptinlevels of acne and control groups were not signifi-cantly different (P [ .05). There were no significantdifferences between acne and control groups ac-cording to Homa-IR values (P[.05) and none of thepatients had insulin resistance. The levels of IGF-Iwere significantly lower and IGFBP-3 concentrationswere significantly higher in control subjects com-pared with the patients with acne (P\.05) (Table II).

There were no significant differences amongmild, moderate, and severe acne groups accordingto overall glycemic index, glycemic loads and fasting

J AM ACAD DERMATOL

NOVEMBER 2007

822 Kaymak et al

glucose, insulin, IGF-I, IGFBP-3, and leptin levels(P [ .05).

Overall glycemic index levels of patients withacne whose disease duration was more than 2 yearswere significantly higher than the ones less than 2years (P \ .05).

DISCUSSIONNon-Westernized diets are known to be free of

processed foods, cereal grains, dairy products, re-fined sugars, and refined oils and almost entirelycomprise unprocessed fresh fruits, vegetables, andlean meats, fish, and seafood.31 Dietary attitudes ofthe Turkish population resembles that of the Westernpopulation especially in the last decade. This resem-blence is even moreso in university students regard-ing consumption of fast foods. Therefore, higherincidence of acne could be expected related torecent dietary attitudes in the Turkish population.However, in contrast to the hypothesis regardinghyperinsulinemic, high glycemic index diets playinga role in the pathogenesis of acne, in this studydietary glycemic index, glycemic load, and insulinlevels in patients with acne were not higher than incontrol subjects. In addition, there were no signifi-cant differences between acne and control groupsregarding the serum glucose, IGF-I, and leptin levels.

Because the glycemic index, insulin, and leptinlevels of the acne group were not different from thecontrol group, it can be suggested that they do notplay a major role in pathogenesis of acne in thissample. There is a distinct demographic differencebetween populations with acne vulgaris, which isprincipally a teenage disorder, and those showinginsulin resistance, which is seen mainly in over-weight adults. Women with postadolescent acnemaintain elevated serum concentrations of IGF-I32

and are mildly insulin resistant.33 Therefore, lack ofhyperinsulinemia and insulin resistance in thisyoung sample may address factors other than hy-perinsulinemia in the pathogenesis of acne. Hormo-nal changes, genetic susceptibility, stress, andenvironmental factors should be considered. Theresults of this study do not reveal a relationshipbetween high glycemic index diets and acne, butindicate a relationship between duration of acne andhigh glycemic index. However, the nature of thepossible relationship between duration of acne andhigh glycemic index diets should be investigated inlong-term prospective studies.

This study suggests that the presence of acne in ayounger population may be highly associated withfactors other than dietary glycemic index or seruminsulin levels. However, this suggestion should beverified by other studies involving, especially, adult

patients with postadolescent acne and obese pa-tients with acne.

REFERENCES

1. Tan J, Vasey K, Fung K. Beliefs and perceptions of patients with

acne. J Am Acad Dermatol 2001;44:439-5.

2. Belisario JC. Acne vulgaris: its etiology and treatment, 1951.

Australas J Dermatol 2000;41(Suppl):23-49.

3. Thiboutot DM, Strauss JS. Diet and acne revisited. Arch

Dermatol 2002;138:1591-2.

4. Robinson HM. The acne problem. South Med J 1949;42:

1050-60.

5. Fulton JE, Plewig G, Kligman AM. Effect of chocolate on acne

vulgaris. JAMA 1969;210:2071-4.

6. Anderson PC. Foods as the cause of acne. Am Fam Physician

1971;3:102-3.

7. Adebamowo CA, Spiegelman D, Danby FW, Frazier L, Willett

WC, Holmes MD. High school dietary dairy intake and teenage

acne. J Am Acad Dermatol 2005;52:207-14.

8. Wolf R, Matz H, Orion E. Acne and diet. Clin Dermatol

2004;22:387-93.

9. Danby FW. Acne and milk, the diet myth, and beyond. J Am

Acad Dermatol 2005;52:360-2.

10. Deplewsky D, Rosenfield RL. Role of hormones in piloseba-

ceous unit development. Endocr Rev 2000;21:363-92.

11. Bendiner E. Disastrous trade-off: Eskimo health for white

‘‘civilization.’’ Hosp Pract 1974;9:156-89.

12. Holt SH, Miller JC, Petocz P. An insulin index of foods: the

insulin demand generated by 1000-kJ portions of common

foods. Am J Clin Nutr 1997;66:1264-76.

13. Cordain L, Lindeberg S, Hurtado M, Hill K, Eaton SB, Brand-

Miller J. Acne vulgaris: a disease of Western civilization. Arch

Dermatol 2002;138:1584-90.

14. Woo-Sam PC. Cohesion of horny cells during comedo forma-

tion: an electron microscopic study. Br J Dermatol 1977;97:

609-15.

15. Ostman EM, Liljeberg Elmstahl HG, Bjorck IM. Inconsistency

between glycemic and insulinemic responses to regular and

fermented milk products. Am J Clin Nutr 2001;74:96-100.

16. Barbieri RL, Smith S, Ryan KJ. The role of hyperinsulinemia in

the pathogenesis of ovarian hyperandrogenism. Fertil Steril

1998;50:197-212.

17. Cara JF. Insulin-like growth factors, insulin-like growth factor

binding proteins and ovarian androgen production. Horm Res

1994;42:49-54.

18. Goulden V, McGeown CH, Cunliffe WJ. The familial risk of

adult acne: a comparison between first-degree relatives of

affected and unaffected individuals. Br J Dermatol 1999;141:

297-300.

19. Zoubolis CC, Xia L, Akamatsu H, Seltmann H, Fritsch M,

Hornemann S, et al. The human sebocyte culture model

provides new insights into development and management of

seborrhoea and acne. Dermatology 1998;196:21-31.

20. Brismar K, Fernqvist-Forbes E, Wahren J, Hall K. Effect of

insulin on the hepatic production of insulin-like growth factor-

binding protein-1 (IGFBP-1), IGFBP-3, and IGF-1 in insulin-

dependent diabetes. J Clin Endocrinol Metab 1994;79:872-8.

21. Lee KW, Cohen P. Nuclear effects: unexpected intracellular

actions of insulin-like growth factor binding protein-3.

J Endocrinol 2002;175:33-40.

22. Wong WW, Copeland KC, Hergenroeder AC, Hill RB, Stuff JE,

Ellis KJ. Serum concentrations of insulin, insulin-like growth

factor-1 and insulin-like growth factor binding proteins are

different between white and African American girls. J Pediatr

1999;135:296-300.

J AM ACAD DERMATOL

VOLUME 57, NUMBER 5

Kaymak et al 823

23. Kılıc M, Taskın E, Ustundag B, Aygun AD. The evaluation of

serum leptin level and other hormonal parameters in children

with severe malnutrition. Clin Biochem 2004;37:382-7.

24. Saygılı F, Oge A, Yılmaz C. Hyperinsulinemia and insulin

insensitivity in women with nonclassical congenital adrenal

hyperplasia due to 21-hydroxylase deficiency: the relationship

between serum leptin levels and chronic hyperinsulinemia.

Horm Res 2005;63:270-4.

25. Pochi PE, Shalita AR, Strauss JS, Webster SB, Cunliffe WJ, Katz

HI, et al. Report of the consensus conference on acne

classification: Washington, DC, March 24 and 25, 1990. J Am

Acad Dermatol 1991;24:495-500.

26. Attia N, Tamborlane WV, Heptulla R, Maggs D, Grozman A,

Sherwin RS, et al. The metabolic syndrome and insulin-like

growth factor I regulation in adolescent obesity. J Clin

Endocrinol Metab 1998;83:1467-71.

27. Liu S, Willett WC. Dietary glycemic load and atherotrombotic

risk. Curr Atheroscler Rep 2002;4:454-61.

28. Foster-Powell K, Holt SH, Brand-Miller JC. International table of

glycemic index and glycemic load values: 2002. Am J Clin Nutr

2002;77:5-56.

29. Mullins JF, Naylor D. Glucose and the acne diathesis: an

hypothesis and review of pertinent literature. Tex Rep Biol

Med 1962;20:161-75.

30. Matthews DR, Hosker JP, Rudensky AS, Naylor BA, Treacher DF,

Turner RC. Homeostasis model assessment, insulin resistance

and beta-cell function from fasting plasma glucose and insulin

concentration in man. Diabetologia 1985;28:412-9.

31. Cordain L. Implications for the role of diet in acne. Semin

Cutan Med Surg 2005;24:84-91.

32. Kiens B, Richter EA. Types of carbohydrate in an ordinary diet

affect insulin action and muscle substrates in humans. Am J

Clin Nutr 1996;63:47-53.

33. Ludwig DS. The glycemic index: physiological mechanism

relating to obesity, diabetes, and cardiovascular disease. JAMA

2002;8:2414-23.