Dermatoses in overweight and obese children and their relationship with insulin and skin color

The aim of the present study was to investigate the prevalence of obesity‐related dermatoses in obese children, and the association between these dermatoses and insulin resistance as well as skin color.


| INTRODUC TI ON
The World Health Organization defines overweight and obesity according to body mass index [BMI = weight (kg)/height (m 2 )].It is necessary to take the age and gender of the child into account in order to evaluate overweight and obesity in children.For this purpose, percentile curves are used.The percentile represents the order of a child among 100 children of the same age. 1 Accordingly, children are classified as underweight if the percentile value is <5%, normal if it is between 5% and 85%, overweight if it is between 85% and 95%, and obese if it is more than 95%. 2 When overweight and obesity are evaluated dermatologically, it affects all skin layers.It involves the physiology of the skin, the barrier mechanism of the skin, sweat gland function, sebum production, collagen structure, wound healing, subcutaneous adipose tissue structure, lymphatics, microcirculation, and macrocirculation.As a result of these, being overweight and obese may cause or exacerbate various dermatosis such as acanthosis nigricans (AN), keratosis pilaris (KP), hirsutism, acrochordon, plantar hyperkeratosis, striae distensae, adipose dolorosa, cellulite, intertrigo, candidiasis, dermatophytoses, folliculitis, necrotizing fasciitis, hidradenitis suppurativa, gout, and psoriasis. 3Clarifying the physiopathology of obesity-related dermatoses is essential in preventing these diseases and developing more effective treatment alternatives.
There are numerous studies on the evaluation of obesity-related dermatoses in adults in the literature; however, we found very few studies conducted on children.Recently, we have identified many overweight and obese child patients with previously undiagnosed insulin resistance presenting with skin manifestations in our daily practice.We observed that some dermatoses are more common in those with darker skin tones.Therefore, the aim of this study was to research the prevalence of obesity-related dermatoses are in overweight and obese children and the association between these dermatoses and insulin resistance and skin color.

| ME THODS
This prospective, observational, and cross-sectional study investigated obesity-related dermatoses in overweight and obese children.The weight and height of the children were measured.BMI was calculated by the formula BMI = body weight (kg)/height (m 2 ). 1 The percentile of BMI was obtained according to the percentile curves determined for Turkish children according to gender and age. 4 Skin type was defined by the scale developed by Fitzpatrick (FP) according to skin color. 5e plasma glucose and insulin levels of the children were measured after 10 h of fasting.Insulin resistance was calculated through the Homeostasis Model Assessment (HOMA) equation [HOMA: fasting insulin (μU/mL) × fasting plasma glucose (mg/dL)/405].Values at and above 2.5 were recorded as insulin resistance. 6

| Statistical analysis
The chi-squared test and Fisher's exact probability test were used for categorical comparisons of the groups for data evaluation.One-    2).

| RE SULTS
The dermatoses detected more common in the patient group than in the control group were KP, stria distensae, hyperhidrosis, AN, plantar hyperkeratosis, bacterial infections, and acrochordon, respectively.Furthermore, a correlation was found between dystrophic cellulite and hirsutism and higher BMI in girls (Table 2).
There was not any positive correlation between dystrophic cellulite and insulin resistance; a positive correlation was detected between dystrophic cellulite and obesity-overweight.There was no positive correlation between seborrheic dermatitis and obesityoverweight; however, a positive correlation was detected between seborrheic dermatitis and insulin resistance.
It was observed in this study that the incidence of KP, AN, and acrochordon increased as the skin color darkened according to the FP skin scale (Table 4).There was not any correlation between other dermatoses and skin color.However, according to our logistic regression analysis results, an increase in skin color does not increase the probability of acrochordon formation.(OR, 0.147; 95% CI, 0.018-1.200,p = 0.073).
According to the results of logistic regression analysis, higher  5).
According to the results of logistic regression analysis, having a higher BMI and FP 3 skin type (according to FP 1 skin type) significantly affect the development of KP (p < 0.05).Increase of each unit in BMI increases the risk of KP by 1.13 times (p = 0.001).Having FP 3 skin type increases the risk of developing KP 3.26 times (1/0.306)compared to FP 1 skin type (p = 0.015) (Table 6).

| DISCUSS ION
It was accepted that adipose tissue acts as a thermal insulation and a buffer against trauma.Studies conducted in recent years have shown that many adipokines secreted from adipose tissue have autocrine, paracrine, and endocrine effects.The adipose tissue plays an important role in regulating insulin resistance and inflammation through the pro-inflammatory cytokines secreted by adipose tissue.
Obesity causes dermatosis in many cases due to mass increase along with the adipokinins secreted by the adipose tissue. 7,80][11][12] In addition to being overweight and obese, we also examined the association between these dermatoses and insulin resistance and skin color.
The dermatoses seen more common in the patient group than in the control group of this study were KP, stria distensae, hyperhidrosis, AN, plantar hyperkeratosis, bacterial infections, and acrochordon, respectively.
KP usually presents as small, perifollicular, and punctate papules located on the extensor surfaces of the extremities.It is frequently seen in atopic individuals; the incidence increases in those with higher BMIs. 13In the study by Nino et al., 9 KP was observed at a rate of 17.5% in obese children, but it was not found to be correlated with weight gain.It has been stated that KP is common in overweight and obese adults and that insulin resistance may play a role in developing KP. 13 In our study, we found a correlation between KP, high BMI, and insulin resistance.Excessive androgen production is associated with hyperinsulinemia, a consequence of IR. 14 Pilar hyperkeratosis associated with increased androgen levels has been noted in individuals with IR. 15 Androgens, insulin, growth hormone, and IGF are generally higher in obese patients.It has been known for many years that they activate the sebaceous glands and increase the development and severity of acne vulgaris. 16,17It was found in the study conducted by Mirmirani et al. 10 and Al-Saeed et al. 12 on female children that the incidence of acne vulgaris in children increased with the increase in BMI.However, no association was detected between higher BMI, insulin resistance, and acne vulgaris in children in this study.We thought this might be related to the increased incidence of acne vulgaris in both the patient and control groups since most of the children in our study were in adolescence.
Striae can be considered as "scars" that occur due to damage to the dermal connective tissue containing collagen formed in response to the force of local stress. 180][11] We found a significant correlation between striae distensae formation and higher BMI.We also found a positive correlation between striae distensae formation and insulin resistance.
We think this is related to the frequent coexistence of obesity and insulin resistance.
Obese patients have deeper skin folds and thicker subcutaneous fat tissues.Therefore, more intense sweating occurs when the temperature increases due to friction and maceration.Furthermore, thicker adipose tissue can increase metabolic activity and heat production, prevent heat loss, and simultaneously increase the secretion of sympathomimetic adipokines, causing hyperhidrosis. 19,200][11] A positive association between hyperhidrosis and using insulin and diabetes mellitus is known. 21,22creased sympathetic activity has been associated with essential hyperhidrosis. 23,24In patients with diabetes mellitus, the balance between parasympathetic and sympathetic activity is disturbed and sympathetic activity becomes dominant. 25It is also known that there is a positive feedback mechanism between insulin resistance and sympathetic activity, but it is not clear which one induces which. 26In line with the literature, we found a positive correlation between hyperhidrosis, high BMI, and insulin resistance.
We found a positive correlation between AN, high BMI, and insulin resistance in children.In other studies, a positive correlation was found between obesity in children and the formation of AN. [9][10][11] Keratinocytes and fibroblasts express the insulin-like growth factor receptor, which can lead to hyperproliferative effects when stimulated.In this way, hyperinsulinemia causes cutaneous changes such as AN and acrochordons. 27 has been accepted that plantar hyperkeratosis is a physiological response due to mechanical trauma. 28[11] According to our results, plantar hyperkeratosis in children was associated with high BMI and insulin resistance.Increased epidermal growth factor receptor expression has been detected in patients with insulin resistance. 29 is known that obesity and insulin resistance play a role in the pathogenesis of dystrophic cellulite in a variety of ways. 30We found that dystrophic cellulite (lipodystrophy) was more common in  The effect of FP 3 relative to reference (FP 1).

c
The effect of FP 4 relative to reference (FP 1).d The effect of "has insulin resistance" relative to the reference (no insulin resistance).The effect of FP 2 relative to reference (FP 1). b The effect of FP 3 relative to reference (FP 1).

c
The effect of FP 4 relative to reference (FP 1).d The effect of "has insulin resistance" relative to the reference (no insulin resistance).

TA B L E 6
Evaluation of skin type, BMI, and insulin resistance effects in predicting keratosis pilaris development by logistic regression analysis.
children with insulin resistance and higher BMI.[11] Hirsutism is the most common symptom of hyperandrogenism.
Hyperandrogenism in obese people appears as a result of increased testosterone synthesis from increased adipose tissue and increased androgen production from the ovaries due to hyperinsulinemia.
Furthermore, insulin reduces the synthesis of sex hormone-binding globulin. 1,19,31As expected, we found increased hirsutism in children with insulin resistance and higher BMI.Similarly, Mirmirani and Carpenter 10 also found an increase in the incidence of hirsutism by obesity in children.
Obesity is associated with various bacterial infections, from less complicated infections such as folliculitis and furunculosis to serious infections such as erysipelas and necrotizing fasciitis that require hospitalization. 32,33In our study, only folliculitis was found among these bacterial infections.There was an increase in the incidence of folliculitis by higher BMI and insulin resistance.Similar studies found a positive correlation between obesity in children and the incidence of folliculitis. 10,12 did not find a correlation between seborrheic dermatitis and obesity; however, we found a positive correlation between seborrheic dermatitis and insulin resistance.Many studies in the literature emphasize the relationship between obesity and psoriasis, an inflammatory dermatosis such as seborrheic dermatitis. 3,34There is no study investigating the association between seborrheic dermatitis and obesity.The association between insulin resistance and inflammation has been explained in various ways, regardless of the relationship between obesity and inflammation. 35Insulin resistance may trigger and exacerbate seborrheic dermatitis as a trigger of inflammation.
Irritation of the skin due to friction is thought to cause acrochordon formation in obese individuals.Keratinocytes and fibroblasts express the insulin-like growth factor receptor, which causes hyperproliferative effects when stimulated.Thus, hyperinsulinemia leads to the formation of AN and acrochordons. 27In a study conducted on 156 obese adults, the severity of obesity with AN was higher, and AN formation was associated with increased insulin resistance. 36In our study, a positive correlation was found between obesity, insulin resistance, and the incidence of acrochordon in children.
We found the six varicose veins and three androgenetic alopecia of the dermatoses associated with obesity; however, we have not found any statistically significant difference in favor of higher BMI.We thought that this result was related to the fact that we performed the study on obese children.We believe that androgenetic alopecia due to hyperandrogenism and varicose veins due to the development of chronic venous insufficiency will be seen at later ages, if these children are followed up.
Hidradenitis suppurativa, lymphedema, stasis dermatitis, and adipose dolorosa which are associated with obesity were not found in the individuals participating in our study.These obesity-related dermatoses were not found in other studies on obese children.
Since hidradenitis suppurativa appears during the second and third decades, it is not seen in obese children. 37Adiposis dolorosa is absent in obese children whereas it is frequent in obese and postmenopausal women. 38 a study evaluating 729 ingrown nails, obesity was associated with ingrown nails. 39We did not find a positive correlation between ingrown toenails and obesity, but two cases of ingrown toenails were found among obese children.
According to the FP skin scale, we found a positive correlation between darkening skin color and KP, AN, and acrochordon.In children and adolescents, White race has not been found to be a risk factor for AN and acrochordons. 10,40According to our logistic regression analysis results, being dark-skinned increases the probability of AN and KP formation compared to being light-skinned.
However, according to our logistic regression analysis results, being dark-skinned does not increase the probability of acrochordon formation compared to being light-skinned.According to our results, FP3 was the skin type with the highest risk of developing AN and KP.AN occurs due to acanthosis, papillomatosis, and hyperkeratosis of the epidermis.The hyperpigmentation seen in AN occurs mainly due to hyperkeratosis. 27However, the quantitative skin color measurement study of Pattamadilok et al. shows that AN lesions contain higher melanin than normal skin. 41It has also been reported that hyperpigmentation caused by melanin in the epidermal basal layer can be seen in AN. 27 We could not find any information in the literature regarding the relationship between KP and melanin pigment.Explaining the physiopathological correlation between these diseases and skin color may lead to developing new treatment alternatives.

| CON CLUS ION
The study was conducted on overweight and obese children and normal-weight healthy children who were followed up and treated in Kahramanmaraş Sütçü İmam University Medical Faculty Hospital between October 2016 and May 2017.Approval was obtained to indicate that there was no ethical and scientific objection to the study from our local Clinical Research Ethics Committee (decision number 14 dated March 21, 2016).The present study was carried out in accordance with the Declaration of Helsinki.Children of families who gave consent for this study were included.Children under the age of 2 and over 18, syndromic children, children with endocrinopathy other than overweight and obesity, and children with a BMI below the fifth percentile were excluded from the study.
way ANOVA test was used to compare the data obtained by measurement.Ratios and frequencies represent data.The effects of different predictors determined by pairwise comparisons in predicting the occurrence of obesity-related dermatoses were evaluated by multivariate logistic regression analysis.Hosmer-Lemeshow test was used to evaluate model fit.Statistical significance was accepted as p < 0.05.The data were analyzed in SPSS package program version 22 (SPSS Inc.).
(n = 61), striae distensae (n = 57), hyperhidrosis (n = 43), AN (n = 42), plantar hyperkeratosis (n = 28), dystrophic cellulite (n = 28), hirsutism (n = 25), bacterial infection (n = 21), seborrheic dermatitis (n = 10), viral infection (n = 7), livedo reticularis (n = 7), acrochordon (n = 6), fungal infection (n = 6), varicose vein (n = 6), androgenetic alopecia (n = 3), ingrown nails (n = 2), and xerosis (n = 1), respectively (Table BMI, having FP 3 and FP 4 skin types (according to FP 1 skin type) significantly affect the development of AN (p < 0.05).An increase of each unit in BMI increases the risk of AN development by 1.24 times (p < 0.001).Having FP 3 skin type increases the risk of developing AN by 5.34 times (1/0.187)when compared to FP 1 skin type (p = 0.003).Having FP 4 skin type increases the risk of developing AN by 3.35 times (1/0.298)when compared to FP 1 skin type (p = 0.021) (Table Abbreviation: BMI, body mass index.a The effect of FP 2 relative to reference (FP 1). b Dermal symptoms of obesity and insulin resistance may precede the diagnosis of diabetes mellitus and may be the first manifestations of diabetes mellitus.Early recognition and management of these cutaneous conditions improve the patient's quality of life and may prevent the occurrence of long-term sequelae associated with insulin resistance and hyperglycemia.On the contrary, a more effective and rapid response to the treatment of these dermatoses is obtained by the treatment of insulin resistance and obesity.Clarifying the physiopathology of these dermatoses enables the development of more effective treatment alternatives.AUTH O R CO NTR I B UTI O N S Conceptualization, methodology, project administration, software, writing-original draft preparation: Mine Mujde Kus.Conceptualization, methodology, writing-original draft preparation: Mehmet Kamil Mulayim.Data curation, formal analysis, writingoriginal draft preparation: Celal Kus.Data curation, formal analysis, software: Adem Doganer.Conceptualization, methodology, supervision: Perihan Ozturk.Resources, data curation software: Fatih Temiz.Data curation, writing-reviewing and editing: Hulya Nazik.

Group Patient group (obese and overweight) (n = 111) Control group (n = 101) p
a Chi-squared test.TA B L E 1 Demographic characteristics of the participants.TA B L E 2 Descriptive statistics between patient group and control group of the dermatoses.a Chi-squared test; Fisher absolute probability test.
Evaluation of the effects of skin type, BMI, and insulin resistance in predicting the development of acanthosis nigricans by logistic regression analysis.
TA B L E 5Note: Dependent variable: keratosis pilaris; binary logistic regression: a: 0.05; Nagelkerke R 368.Data in bold indicate is a statistically significant intergroup difference (p < 0.05).Abbreviation: BMI, body mass index.a