Published literature indicates a rising prevalence of childhood obesity in both developed and developing countries.12 In a national household screening program study, involving 12,701 children aged 1 to 18 years in different areas of Saudi Arabia from 1994 to 1998, the overall prevalence of overweight was 10.68 and 12.7 per cent and that of obesity was 5.98 and 6.74 per cent in boys and girls, respectively. The highest frequency was in the Eastern Province (27.6% and 13.8% for overweight and obesity in girls, respectively).3 Overweight and obesity reported among adolescent females aged 12-19 years in Al- Khobar, Eastern Province was 28%.4 Skin diseases are reported as frequent findings in the obese patients. To our knowledge, there is no report in the literature on the prevalence of dermatological lesions in obese schoolchildren.
A cross-sectional study of 126 female and 30 male obese patients aged 16 to 89 years was conducted in Mexico City.5 The cutaneous findings with statistical significance found were: plantar hyperkeratosis, acanthosis nigricans, striae, skin tags and keratosis pilaris.
Another study, conducted in France,6 showed the following skin findings in obesity: acanthosis nigricans, skin tags, signs of hyperandrogeny, striae distensae, stasis acroangiodermatitis, leg ulcers, lymphoedema and intertrigo. In a USA7 study, the skin disorders in the obese included stretch marks, acanthosis nigricans, hirsutism, intertrigo and multiple papillomas. All these studies were conducted on adult obese patients in hospital. To the best of the investigator's knowledge, no study has been conducted to investigate these conditions in schoolchildren in Saudi Arabia. Therefore, the aim of this study was to determine the most common skin disorders among female schoolchildren in primary and intermediate schools in Al-Khobar, Eastern Saudi Arabia.
MATERIAL AND METHODS
This was a cross-sectional study conducted in the Al-Khobar from January to March 2003. The target population was composed of all female schoolchildren in regular public and private primary and preparatory schools (total number of 28766 schoolchildren). The estimated sample size was based on the proportion of schoolchildren in the pilot study suffering from skin diseases. Accordingly, the final study sample size with a total number of 2239 schoolchildren was determined. A multi-stage stratified random sampling technique with proportional allocation was used as follows:
First; schools were sub-classified by type (governmental/private) and educational levels (primary/preparatory). Using the random digits table, ten governmental and five private schools from every educational level were selected yielding the desired sample size from a total of 30 schools. A simple random sample was then used to select the classes.
In the second stage, a systematic random sampling technique was used to select the desired number of schoolchildren from randomly selected classes. Skin diseases were classified according to the modified 10th revision of the international classification of disease (ICD-10).8 Informed consent for inclusion in the study was obtained from the parents by invitation accompanied by a letter explaining the purpose and details of the study. The questionnaires consisted of information on the schoolchildren and two questions on the frequency of change of underwear and the number of baths taken per week. All the selected schoolchildren were examined by the investigator in a private room during the day in the same school. The entire body except the thighs and perineum, was exposed. To compensate for the effect on the result of some skin diseases such as intertrigo in the obese children, the girls were asked during examination if they had any complaints or lesion in those areas. The initially undiagnosed cases were reexamined in the same school by a female consultant dermatologist. The following anthropometric measurements were included: (1) Weight: This was measured by the interviewer, using auto-calibrated measuring scale. The schoolchildren were bare-footed and lightly clothed (school uniform). The weight was recorded to the nearest 100 grams. (2) Height: This was measured using the measuring cartoon scale attached to the wall. Each subject was requested to stand straight without shoes, with the heels together. Buttocks, shoulders and occiput were made to touch the vertical wall. The height was measured to the nearest centimeter. (3) BMI: This was calculated using the formula: BMI= Weight (in Kg)/Height (in M2). The BMI interpretation was based on the table of a standard definition for overweight and obesity in children9 which does not include underweight in children (classified as normal, overweight and obese). To include underweight growth chart of body mass index-for-age percentiles: girls, 2 to 20 years (CDC growth charts)10 was used.
For the growth chart of body mass index-for age percentiles, the following established percentile cut-off points were used: Underweight = BMI-for-age < 5th percentile; At risk of overweight = BMI-for-age 85th percentile to < 95th percentile; Overweight = BMI-for-age ≥ 95th percentile.
All variables were checked for accuracy and completeness, and were coded. Data was then entered into a personal computer and the Statistical Products of Service Solutions (SPSS) version 10 was used for data entry and analysis. The appropriate statistical analytical techniques were performed. Frequency distribution tables were constructed for the distribution of various skin conditions. All the necessary approvals were obtained from the relevant authorities before the conduct of the study. Positive skin disease cases were treated and referred to the dermatology clinic for follow up and treatment.
The mean height of the sample schoolchildren was 136.85 ± 15.12 cm, while the mean weight was 37.43 ± 16.06 Kg. The mean BMI of the sample schoolchildren was 19.21 ± 5.17. Consequently, 11.3% of the sample school children were found to be obese, 20% were overweight and 11% underweight.
There were significant associations between obesity and the frequency of underwear change and the number of baths taken per week, (p < 0.001). Obese schoolchildren changed their underwear (mean 6.35 ± 2.56 sd times/wk) and took baths (mean 4.68 ± 2.19 sd baths/wk) more frequently than non-obese schoolchildren (mean 5.98 ± 2.58 sd and 4.36 ± 2.06 sd, respectively).
Using the standard definition,9Table 1 shows the skin conditions found to be statistically associated with obesity. The skin diseases found to be significantly more prevalent among obese schoolchildren than children of normal weight were acanthosis nigricans, folliculitis, and stria.
Using the same definition,9Figure 1 shows the common skin diseases in obese schoolchildren. Using the CDC growth chart definition,10Table 2 shows the skin diseases found to be statistically associated with underweight schoolchildren. Dandruff, acne, K. pilaris and insect bites were less prevalent among underweight schoolchildren, while herpes simplex, cheilitis, and cherry angioma were more prevalent.
The overall response rate in this study was 100%. In the present study, the common skin findings associated with obesity were dandruff, acne, xerosis, acanthosis nigricans, folliculitis, alopecia, striae and callosity. Of these skin diseases, acne, dandruff and acanthosis nigricans and striae represented the highest prevalence (30.9%, 23.8%, 14.1% and 6.4% respectively).
Acne and dandruff associated with insulin resistance and hyperandrogenism were the most common skin conditions among the overweight and obese schoolchildren. In this study, the prevalence of acne and dandruff in the obese was 30.9% and 23.8%, respectively. No studies have recorded the prevalence of dandruff in obese people.
The prevalence of acne in this study was higher than that reported in obese adults in Mexico (25.1%)5 and Egypt (6.8%).11 This is probably because acne tends to be more prevalent in the younger age group.
Acanthosis nigricans (AN) was a finding that is consistent in all studies of obese persons. It is characterized by a hyperpigmented velvety cutaneous thickening affecting localized areas of the skin in obese persons.12 AN is frequently associated with obesity, endocrinopathies, malignancy, genetic syndromes and the use of some drugs.7 AN is a reliable cutaneous marker of hyperinsulinemia in obese individuals.13 High concentrations of insulin stimulates DNA synthesis and cell proliferation through insulin-like growth factor (IGF-1) receptors which present in skin keratinocytes, causing the development of AN.14
AN in the present study was present was found in 14.1% of the obese schoolchildren, which was lower than what was reported from Brazil15 (76%), Mexico City5 (64.2%), and USA13 (74%), but higher than reported in Egypt11 (0.6%). The difference could be related to age group and the severity of obesity.
In this study, the prevalence of striae, which is due to skin over-extension, was low compared to the what was reported in obese adults in Mexico (89%)5 and among children in Taiwan16 (40%), but higher than in adult Egyptians (0.8%).11
Folliculitis caused by excessive fat folds in an obese child and favoring humidity and maceration with bacterial and fungal overgrowth was the 5th most common skin condition (2.4%) in overweight and obese children in the study. This was not recorded in the Mexican or Egyptian studies.
Planter hyperkeratosis (callosity) is considered a stigma of morbid obesity resulting from the effect of the pressure of excess weight.5 The prevalence here was 2%, which was lower than that reported among obese adults in Mexico (75.2%).5
The result of the study revealed an association between obesity and alopecia and xerosis. Since most cases of alopecia are due to trauma, it was expected to be more prevalent in obese schoolchildren as their excess weight predisposed them to more accidents. Obese schoolchildren in the study changed their underwear and took baths more frequently than non-obese schoolchildren. This depleted the skin's natural oils and lead to xerosis.
In the present study, using the CDC growth chart with 85th and 95th percentiles as cut-off points the prevalence of underweight was 11%. Dandruff, acne and keratosis pilaris were significantly less common among underweight schoolchildren, most probably because 83% of those underweight were in primary school, a time when sebaceous glands activity is not yet well established.
Herpes simplex may be present in the underweight children as a result of their susceptibility to infection because of low immunity, while cheilitis is seen in severely malnourished children as a result of iron deficiency.
In conclusion, the incidence of overweight (20%) and obesity (11%) was high. In this study, the skin conditions that were statistically associated with obesity included: dandruff, acne, xerosis, acanthosis nigricans, folliculitis, alopecia, stria, and callosity. Acanthosis nigricans is a reliable marker of hyperinsulinemia in obese children who may be at a greater risk for developing atherosclerotic and cardiovascular complications. Therefore, children with acanthosis nigricans should be screened for these conditions. One of the limitations of the study was that the thighs and the perineum were not examined. This may, therefore, have distorted the results of some skin diseases such as intertrigo in obese children. Health education programs on skin diseases and obesity should be directed at all schoolchildren, their families and teachers. Malnutrition, be it under or over nutrition in schoolchildren should be dealt with. Effective school-based programs for prevention, early detection and treatment of malnutrition should be initiated. Further similar studies involving male and female schoolchildren at different educational levels in various parts of the kingdom should be encouraged.
The authors are grateful to all female schoolchildren who participated in the study as well as their families.
1. Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia Am J Clin Nutr. 2002;75(6):971–7
2. Mary C, Pinki S, Julian H, Jenny W. Increasing prevalence of obesity in primary school children: cohort study BMJ. 2001;322(7294):1094–95
3. El-Hazmi M, Warsy A. The prevalence of obesity and overweight in 1-18 year-old Saudi children Annals Saud Med. 2002;122(5-6):303–7
4. Abahussain NA, Musaiger AO, Nicholls PJ, Stevens R. Nutritional status of adolescent girls in the eastern province of Saudi Arabia
Nutr Health. 1999;13(3):171–7
5. Garcia HL. Dermatological complications of obesity Am J Clin Dermatol. 2002;3(7):497–506
6. Martalo O, Pierard C, Scheen A, Pierard GE. Skin diseases and obesity Rev Med Liege. 2003;58(2):73–6
7. Bray GA. Health hazards of obesity Endocrinol Metab Clin North Am. 1996;25(4):907–19
8. Karen P. Essentials of ICD-10-AM An information package for clinicians and users of coded data, Skin. 98 NovAccessed January,2003 Available at: http://www2.fhs.usyd.edu.au/ncch/clined/
9. Cole T, Bllizzi M, Flegal K, Dietz W. Establishing a standard definition for child overweight and obesity worldwide: international survey BMJ. 2000;320:1240–1243
10. . National Center for Health Disease Control and prevention CDC growth charts: United States.Accessed Jun,2003 Available at: http://www.pregnancyweekly.com/topics/set2/chart16.pdf
11. Mousa AM, Soliman MM, Hamza MR. Obesity and skin diseases J Egypt Public Health Assoc. 1977;52(2):65–74
12. Kay S, Jen B, Richard A, Howard P, Alexander S Color atlas & synopsis of pediatric dermatology. 2002International ed The McGraw-Hill Companies:Inc
13. Hud JA, Cohen JB, Wagner JM, Cruz PD. Prevalence and significance of acanthosis nigricans in an adult obese population Arch Dermatol. 1992;128(7):941–4
14. Rendon MI, Gruz PD, Sontheimer RD, Bergstrsser PR. Acanthosis nigricans: a cutaneous marker of tissue resistance to insulin J Am Acad Dermatol. 1989;21:461–9
15. Araujo MB, Porto MV, Netto EM, Ursich M J. Association of acanthosis nigricans with race and metabolic disturbances in obese women Braz J Med Biol Res. 2002;35(1):59–64
16. Chen W. Childhood obesity in Taiwan Dermatol. 1994;188(2):108–12