SciELO - Scientific Electronic Library Online

vol.77 issue5Molecular Genetics of Epidermolysis BullosaClinical aspects of epidermodysplasia verruciformis author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Anais Brasileiros de Dermatologia

Print version ISSN 0365-0596On-line version ISSN 1806-4841

An. Bras. Dermatol. vol.77 no.5 Rio de Janeiro Sept./Oct. 2002 



Acanthosis nigricans in obese women in a mixed-race population: a marker of metabolic disturbances*



Leila Maria Batista AraújoI; Adriano Moura Costa de ViveirosII; Renata Cruz LopesII; Aldenice de Carvalho VianaIII; Rosa T FukuiIV; Mileni J M UrsichV

IProfessor of Endocrinology at the Universidade Federal da Bahia Faculty of Medicine
IIAcademics of the Universidade Federal da Bahia Faculty of Medicine
IIITechnical Assistant of the Endocrinology Laboratory at the Hospital Universitário Professor Edgard Santos, Salvador, Bahia State
IVBiochemist at the Medicina da Universidade de São Paulo Faculty of Medicine
VProfessor of Endocrinology at the Universidade de São Paulo Faculty of Medicine





BACKGROUND: Acanthosis nigricans (AN) has been associated with various metabolic and endocrine disturbances.
OBJECTIVE AND METHODS: The objective of this study is to evaluate the frequency of metabolic syndrome clusters in a group of obese mixed-race women with AN as compared to a group without AN.
SUBJECTS AND METHODS: 481 women consecutively admitted to an outpatient obesity clinic were studied: 388 with AN, and 93 without AN. Except for 20 diabetic patients, all patients were submitted to an oral glucose tolerance test (75 g).

RESULTS: The skin color distribution was 34.5% white, 38.9% mulatto and 26.6% black. The global frequency of AN was 80.7%. AN frequency was significantly higher in black versus white and black versus mulatto (90.6 % and 66.9%, p = 0.000006). It was also higher in mulatto versus white (86% and 66.9%, p<0,02). The AN group was younger (35 + 10 years versus 38 + 10 years, p < 0,01) and heavier (41 + 6 kg/m2 versus 39 + 6 kg/m2, p<0.01). It consisted of larger waist circumference and higher frequency android obesity, with type 2 diabetes (11% versus 4.3%, p=0.05), higher fasting insulin levels and insulin resistance (HOMA IR) than the group without AN. The frequencies of diastolic hypertension and disturbances of total cholesterol and triglyceride levels in the group with AN were similar to the group without AN.
CONCLUSION: Among obese women from a multi-race population, a higher frequency of AN was observed in black and mulatto women. A higher number of metabolic syndrome clusters was observed in the group with AN than in the group without AN. Thus, obese patients with AN should be targeted for screening metabolic disturbances, even at a young age.

Key words: Acanthosis nigricans; diabetes mellitus non-insulin-dependent; metabolism; obesity; racial stocks; insulin resistance.




Acanthosis nigricans (AN) is a dermatological condition characterized by thickening of the skin, hyperpigmentation and accentuated skin lines, generating a rough and velvety aspect at the affected site. Histologically, hyperkeratosis, accentuated projecting of dermal papillae and discrete thickening of the epidermal layers are commonly observed.1Although it may occur at any site on the body surface, the most affected area is the posterior neck region, followed by the axillae, the side of the neck, flexible surfaces of the limbs, periumbilical and inframammary regions, mucous membrane of the oral cavity or even, in rare cases, the soles of the feet and palms of the hands.1

AN may be divided into two broad forms, malignant and benign. The malignant form is a marker of abdominal neoplasias, particularly gastric adenocarcinoma. Benign forms can be idiopathic, hereditary or drug induced, as well as caused by endocrine diseases.1

Endocrinopathies are the main causes of AN, with obesity as the most common disturbance, frequently associated with hyperinsulinism, diabetes mellitus and insulin resistance.2-15 Other endocrine disturbances associated with AN are described as Cushing's disease, policystic ovaries, thyroidopathies, hirsutism, Addison's disease, acromegaly, among others, some of which occur along with insulin resistance. 5,6,8,11,13

The prevalence of AN in non select populations varies from 7 to 74%, according to age, race, frequency of type and degree of obesity and concomitance with endocrinopathy.9-11 In 34 obese individuals of both sexes, in a predominantly black population (59%), Hud et al.10 described AN prevalence in 74% of patients.

Given the high prevalence of AN in obese patients, the objective of this article was to evaluate the frequency of metabolic complications and its relation with the skin color of obese women with and without AN in a mixed-race population.



A total of 481 obese women were evaluated, ranging in age from 15 to 68 years and in body mass index (BMI = weight/height2) from 33 to 47 kg/m2. The patients spontaneously sought care at the obesity outpatient clinic of the Hospital Universitário Professor Edgard Santos, in Salvador, Bahia State. The population frequenting this clinic consists predominantly of women, which is the reason why men were excluded from the study. Also excluded were asthmatic and psychotic patients, and carriers of chronic diseases, except for those with a metabolic syndrome. All were submitted to a protocol that included clinical and biochemical evaluation prior to initiating treatment for obesity.

The criterion for defining AN was clinical inspection, continually monitored by one of the authors (Figure A and B).

For the classification of race the following phenotypic characteristics were considered (type of hair, shape of ears, nose and lips and skin color), given that there is a broad mixture of races, predominantly mulattos and blacks of African origin in Salvador. Moreover, the difference between white and light-mulatto, and dark-mulatto and black is often difficult to define.17

The waist circumference was determined along the mean line between the lower rib margin and iliac crest, using an appropriate plastic measuring tape. Android obesity was determined by the waist circumference in relation to the hips > 0.85.

The present of diastolic arterial hypertension was ascertained when the diastolic arterial pressure was found to be equal or greater than 85 mmHg in two determinations, either by using the Tycos Large Cuff Blood Pressure Meter or history of hypotensive drug.

With the exception of 20 patients, who were already diagnosed with diabetes mellitus, all the others were submitted to the oral glucose tolerance test (OGTT, 75 g) with the collection of blood for insulin and glucose doses at times: 0.60 and 120 minutes, while maintaining a diet of more than 150 g of carbohydrates three days prior to the test. The glucose was measured in the plasma by the glucose-oxidase method, and World Health Organization (WHO) criteria were applied for the diagnosis of diabetes mellitus or glucose intolerance. 16

The insulin was measured in the serum by radioimmunoassay, using the pig anti-insulin antibody, purchased from Sigma (n=130), or specific human anti-insulin, purchased from Linco (n=164).

The insulin resistance was evaluated by the method of glucose homeostasis (HOMA IR), which considers the glycemic and insulinemic levels of fasting (HOMA IR={insulin (uU/ml) x glucose (mmol/l]/22.5}. 18

The lipid profile was evaluated in all of the patients. The total-cholesterol, HDL-cholesterol, LDL-cholesterol and triglyceride determinations were carried out by the enzymatic method. The criterion for dyslipidemia was adopted from the 2nd Brazilian Congress of Dyslipidemia.18

Fasting lipid profile was evaluated and II Brazilian Dyslipidemia Consensus criterias were applied.19

Regarding the statistical analysis, Epi info, version 6.0 was applied which compared metabolic disturbances between the groups of women with and without AN by means of variance analysis and proportions test. Differences were considered significant when p<0.05. Analysis of multiple logistic regression with the aid of the Statistical Package for Social Studies program (SPSS, version 9.0) was applied for computing the odds ratio in relation to the presence of AN, taking age, BMI, waist circumference, diabetes and diastolic hypertension and race as variables.



The clinical and laboratory data of patients are exhibited in Table 1.

The race distribution in the 481 obese women studied revealed 34.5% whites, 38.9% mulattos and 26.6% blacks. The presence of AN in these patients was 80.7%. The frequency of AN in white, mulatto and black races was 66.9%, 86.1% and 90.6%, respectively (Chart). The proportions test showed that the frequency of AN was significantly higher in black women in comparison to whites and black versus mulattos (p=0.000000 and 0.000006). It was also higher in mulatto women versus whites (p<0.02). The analysis of multiple regression showed that the frequency of AN was three times greater in mulattos (odds ratio, OR=3.1; Confidence Intervals, CI 95% (1.7 - 5.4) and five times higher in the black race in relation to the white (OR=5.0; CI 95% (2.4 -10), regardless of age and BMI.



In the group of women with AN, average age was significantly lower, and the BMI significantly higher than in the group without AN (p < 0.01 for both). The waist circumference and frequency of android obesity were significantly greater in the group with AN (p < 0.001 and p=0.05, respectively).

The prevalence of type-2 diabetes with AN was 11.1% in contrast to 4.3% in patients without this condition, revealing a statistically significant difference (p < 0.05). This increase was three times higher, controlling age, BMI and race, with OR=3.1; CI 95% (1.1 - 9.8). A tendency toward higher frequency of impaired glucose tolerance was observed in the group with AN in relation to the group without AN (14.9% and 9.7%, respectively p=0.18).

Levels of fasting insulin and insulin resistance (HOMA IR) were greater for the group with AN than for the group without AN (p=0.05 for both), Table 1.

Frequencies of diastolic hypertension, total-cholesterol equal or above 200 mg%, LDL-cholesterol equal of greater than 130 mg%, HDL-cholesterol < 35 mg% and triglycerides equal or greater than 200 mg% were similar in both groups (Table 1).



Various papers have considered AN as a marker for many endocrine disturbances, among which are diabetes mellitus and glucose intolerance, frequently associated with obesity.5,6,11,14,20,21

In the present study, obese women were observed to have a higher frequency of diabetes mellitus and android obesity, higher levels of fasting insulin and insulin resistance as well as a tendency to greater presence of glucose intolerance in the women with AN in relation to those without.

Stuart et al11,14 suggested that AN is a marker for easily detecting diabetes mellitus. In a study of 89 African-Americans with AN type 2 diabetes was observed at a frequency of 21.3%.14 In a study of 406 Mexican-Americans of both sexes conducted by Burke et al.,15 type 2 diabetes was observed in 27% of patients with AN and 19.7% in patients without AN, the difference of which was not significant. In the study, a lower frequency was observed (11.1% in patients with AN and 4.3% in those without AN). The frequency differences pointed to by the authors could be related to genetic or environmental factors, because only Mexican-Americans were included.

In the present study, the levels of fasting insulin as well as the degree of insulin resistance as evaluated by HOMA IR were greater in the AN group than in the group without AN. Burke et al. also observed higher levels of fasting insulin in Mexican-Americans with AN compared with patients without AN. Hud et al.10observed levels of fasting insulin to be twice as high in patients with AN than in those without AN. Stuart et al.11 showed individuals from indigenous peoples as having a higher level of fasting insulin than in individuals without AN.

Few studies until now have referred to lipid changes in patients with AN. In this study, there were no significant differences detected in the lipid profile frequency (total-cholesterol, LDL and HDL-cholesterol, and triglycerides). Burke et al.15 in the aforementioned study showed statistical difference only in relation to the HDL-cholesterol levels, which were lower in individuals with AN than in those without AN. As the population studied included obese and non-obese patients, with a significantly higher frequency of obesity in the groups of individuals with AN, it became difficult to judge whether variations in the HDL-cholesterol indices were due to the presence of AN or to variable obesity. Nonetheless, the current study shows no statistical differences found in the proportions of total-cholesterol, HDL and LDL-cholesterol and triglycerides beyond the recommended limits.

The presence of diastolic arterial hypertension in this study was not significantly higher in patients with AN in relation to those without AN. Arterial hypertension is also a marker of the metabolic syndrome, related to hyperinsulism.22-24 In Burke et al.15the study observed a higher frequency of diastolic arterial hypertension, which was nonetheless not observed in relation to systolic pressure.

As for race, the present study showed a clear predominance of AN in women of pardo and black races in relation to white. Hud et al.10 studied the prevalence of AN in 34 obese adults in a population comprised of 58% of blacks. They observed the presence of this condition in 74% of patients, the black race being more affected than that of the white (85% versus 57%). In a study of 1,412 obese children, with 31% white, 25% Hispanic and 43% black, Stuart et al.9 demonstrated a 7.1% AN frequency in the population studied. This frequency was higher in obese children (28%). AN was also observed in 0.5% of white children, 5.7% of Hispanic and 13.3% of African Americans. In American indigenous children from Texas and Nebraska, AN frequency was described as 19% in the Omaha and Winnebago peoples to 38% in the Alabama-Coushatta tribe, with insulin levels twice as high in the group with AN. 9,11

The authors thus concluded that in obese patients with AN there is frequent association with diabetes mellitus, android obesity, hyperinsulinism, insulin resistance and the tendency to a higher frequency of glucose intolerance. The results did not indicate a significant relation between AN and dyslipidemia. A higher frequency of AN in black and mulatto women than in whites was also observed. In this way the importance of metabolic evaluation in obese patients with AN is confirmed, due to being an easily detectable lesion upon physical examination and to the possibility of being a predictor of metabolic syndrome clusters.



The authors would like to express their thanks to the CNPq for awarding them the "auxilio integrado" grant (no 523837/95-0); to Aldenice Viana, laboratory technician, for collecting the glucose tolerance tests at the Hospital Universitário Professor Edgard Santos; to Ivanise Maria Santana Silva, secretary, for her devoted assistance; to Rosa T. Fukui for insulin determinations in the Lim-18, of the Universidade de São Paulo Faculty of Medicine; and, to Eduardo Martins Netto for the statistical analysis.



1. Schwartz RA. Acanthosis nigricans. J Am Acad Dermatol 1994; 31:1-19.         [ Links ]

2. Kahn CR, Flier JS, Bar RS, et al. The syndromes of insulin resistance and Acanthosis nigricans: insulin-receptor disorders in man. N Engl J Med 1976; 294:739-45.        [ Links ]

3. Flier JS. Metabolic importance of Acanthosis nigricans. Arch Dermatol 1985; 121:193-4.        [ Links ]

4. Peters EJ, Stuart CA, Prince MJ. Prevalence of Acanthosis nigricans and obesity: acquired and intrinsic defects in insulin action. Metabolism Clin Exp 1986; 35:807-13.        [ Links ]

5. Dunaif A, Graf M, Mandeli J, Laumas V, Dobrjanski A. Characterization of groups of hyperandrogenic women with Acanthosis nigricans, impaired glucose tolerance, and/or hyperinsulinemia. J Clin Endocrinol Metab 1987; 65:499.        [ Links ]

6. Matsuoka LY, Wortsman J, Gavin JR, Goldman J. Spectrum of endocrine abnormalities associated with Acanthosis nigricans. Am J Med 1987; 83:719-25.        [ Links ]

7. Flier JS, Eastman RC, Minaker KL, Matteson D, Rowe JW. Acanthosis nigricans in obese women with hyperinsulinemia. J Clin Endocrinol Metab 1987; 65:499.        [ Links ]

8. Barth JH, Ng LL, Wojnarowska F. Acanthosis nigricans, insulin resistance and cutaneous virilism. Br J Dermatol 1988; 118:613.         [ Links ]

9. Stuart CA, Pate CJ, Peters EJ. Prevalence of Acanthosis nigricans in an unselected population. Am J Med 1989, 87:269-72.         [ Links ]

10. Hud JA Jr, Cohen JB, Wagner JM, Cruz PD. Prevalence and significance of Acanthosis nigricans in an adult obese population. Arch Dermatol 1992; 128:941- 4.        [ Links ]

11. Stuart CA, Smith MM, Gilkison CR, Shaheb S, Stahn RM. Acanthosis nigricans among native Americans: an indicator of high diabetes risk. Am J Publ Health 1994; 84:1839-42.        [ Links ]

12. Panidis D, Skiadopoulos S, Rousso D, Joannides D, Panidou E. Association of Acanthosis nigricans with insulin resistance in patients with polycystic ovary syndrome. Brit J Dermatol 1995; 132:936-41.        [ Links ]

13. Esperanza LE, Fenske NA. Hyperandrogenism, insulin resistance, and Acanthosis nigricans (HAIR-AN) syndrome: Spontaneous remission in a 15-year-old-girl. J Am Acad Dermatol 1996; 34:892-97.        [ Links ]

14. Stuart CA, Gilkison CR, Keenan BS, Nagamani M. Hyperinsulinemia and Acanthosis nigricans in African Americans. J Natl Med Assoc 1997; 89:523-7.        [ Links ]

15. Burke PB, Hazuda HP, Hale DE, Stern MP. A quantitative scale of Acanthosis nigricans. Diabetes Care 1999; 22:1655-59.        [ Links ]

16. World Health Organization. diabetes mellitus: Report of a WHO Study Group. Geneva, World Health Organization, 1995 (Teach Rep Ser, nº 727).        [ Links ]

17. Krieger H, Morton NE, Mi MP, Azevêdo E, Maia AF, Yasuda N. Racial admixture in north-eastern Brazil. Am Hum Genet Loud 1965; 25:113-6.        [ Links ]

18. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RI. Homeostase model assessment: insulin resistance and insulin concentrations in man. Diabetologia 1985; 28:412-9.        [ Links ]

19. Sociedade Brasileira de Cardiologia. II Consenso Brasileiro sobre Dislipidemia. Arq Bras Card 1996; 67,113-8.        [ Links ]

20. Gilkson C & Stuart CA. Assessment of patients with Acanthosis nigricans skin lesion for hyperinsulinemia, insulin resistance and diabetes risk. Nurse Pract 1992; 17:26-44.        [ Links ]

21. Moller DE, Flier JS. Insulin resistance mechanisms, syndromes, and implications. N Engl J Med 1991; 325:938-48.         [ Links ]

22. Reaven GM. Role of insulin resistance in human disease. Diabetes 1988; 37:1595.        [ Links ]

23. Kaplan NN. The deadly quarter. Upper body obesity, glucose intolerance, hipertriglyceridaemia and hypertension . Arch Intern Med 1989; 49: 1514.         [ Links ]

24. Schmidt MI, Watson RL, Duncan BB, Metcalf P, Brancati FL, Sharrett R, Davis CE, Heirs G. Clustering of dyslipidemia, hiperuricemia, diabetes and hypertension and its association with fasting insulin and central and overall obesity in a general population. Metabolism Clin Exp 1996;45: 699-706.        [ Links ]



Leila M B Araújo
Rua Augusto Viana s/n, 6º andar - Endocrinologia
Hospital Universitário Professor Edgard Santos
Salvador Bahia 40110-160
Fax: (71) 247-8492

Received in June, 19th of 2001.
Approved by the Consultive Council and accepted for publication in February, 15th of 2002.



* Work done at Endocrinology Service of the Professor Edgard Santos University Hospital, Universidade Federal da Bahia Faculty of Medicine and the Laboratório de Investigação (Lim 18) of the Universidade de São Paulo Faculty of Medicine.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License