Prevalence of common and atypical melanocytic nevi in Turkish children

ARTICLE

Auteur(s) : Melih Akyol¹, Abuzer Gaffar Atli¹, Sedat Özçelik¹, Ziynet Çinar², Filiz Altioğlu Çig¹, Hüdaverdi Bircan³

¹Dermatology Department, School of Medicine of Cumhuriyet University, 58040-Sivas, Turkey
²Biostatistics Department, School of Medicine of Cumhuriyet University, 58040-Sivas, Turkey
³Business Department, School of Economics, Cumhuriyet University, 58140-Sivas, Turkey

accepté le 15 Mars 2008

The incidence and mortality of cutaneous melanoma is increasing rapidly all around the world. Recently, Pinarbasi et al. [1] reported an increase of 11.7% within the last 5-year period, and 14.4% within the previous 5-year period (from 1994 to 2003) in melanoma cases in the population of Antalya in Turkey. Therefore, it is important to define the high-risk groups for the diagnosis and treatment of cutaneous melanoma [2].

Common melanocytic nevi (CMN) are benign tumors that are found in about 80-100% of the general population. The prevalence of atypical melanocytic nevi (AMN), diagnosed with a biopsy, is about 2-20% in the white race, and 1.5% in children [3]. CMN and AMN (especially) can be possible precursors of melanoma and they are accepted as independent risk factors [2, 4]. It is possible to designate the relationship between the number and distribution of CMN on the whole body and the cutaneous melanoma [2, 5, 6].

Defining the factors affecting nevus development, and having knowledge of the prevalence of CMN and AMN for each geographic region is imperative. The aim of the present study, the first to be performed in our country, was to estimate the prevalence of CMN and AMN and to identify some of the phenotypic factors, gender, and age associated with the presence of CMN and AMN in primary school children in a Turkish population.

Methods

Sociodemographic characteristics of the children were noted. The study was based on skin examination. The children gave oral assent at the time of the physical examination. Dermatologic examinations were done by two dermatologists in the school infirmaries or in rooms that were prepared for consultation. Children who were absent at the first examination, were examined at the second visit to the school.

Skin types were determined as I-VI according to Fitzpatrick. The hair colors were classified as black, brown, blonde, or red and the eye colors were classified as brown-black, hazel, green or blue. In this study, we did not examine the effects of sun exposure characteristics. With the exception of scalp and anogenital regions, the other anatomical regions were classified and evaluated as head-neck, ventral trunk, dorsal trunk, upper extremities, lower extremities, hands and feet. Brown-black discrete lesions of two millimeters or larger were counted as CMN. To minimize the error in differentiation of CMN from other pigmented lesions, two experienced clinicians counted the nevi separately. As there are no absolute criteria for diagnosis of CMN, any darkly pigmented lesion larger than two millimeters was counted as CMN if it was not considered to be a solar lentigo lesion (light brown pigmented lesions with irregular outlines located on sun-exposed areas) or freckles (poorly defined lesions light to medium brown, less than three millimeters) defined by relatively specific clinical criteria. Lentigo simplex was not excluded because there are no absolute clinical criteria to distinguish it from CMN. If a lesion had three or more of the following symptoms it was recorded as AMN: (a) ≥ 5 mm in diameter, (b) poorly defined borders that tend to fade into adjacent normal skin, (c) irregular borders, (d) highly variable color, (e) presentation of macular and papular areas in a single lesion.

In the statistical evaluation of the results, SPSS (version 10.0) and Epi Info (version 3.4.3) were used. For statistical analysis the number of CMN and AMN were used as dependent variables. The mean values of age according to gender were compared by independent-samples t test. The counts for CMN and AMN for different factor classes were reported by median nevus counts, including interquartile range. The number of children with at least one CMN, and AMN by anatomic sites, and by phenotypical characteristics, were noted as numeric and percentile values. The number of children having at least one CMN by anatomical sites, in both genders, was compared by chi-square test (with odds ratios).

For all predictor variables, including age and gender, logistic regression analysis was performed, and CMN were treated as a dichotomous variable according to whether the presence of one or more clinically CMN. Logistic regression-derived odds ratios, together with their 95% CI, were obtained. Some variables had small numbers. Therefore the numbers of blue and green eyes (light eye colour), blonde and red hair (light hair colour), and skin type I and II (light skin colour) were put together, and an analysis with this composed data was made.

First, we checked for multicollinearity. Because tolerance scores are not available through the logistic regression command in SPSS, we computed these values through the linear regression command, using CMN presence as the dependent variable, and gender, age groups, hair and eye colour, and skin type as independent variables. We checked correlations between the variables to see if there might be multicollinearity problems. The correlation matrix obtained indicated strong correlations between eye color and hair color, and between skin type and hair colour (p = 0.01). Additionally, tolerance values were calculated. That the tolerance value is low (1-R²) was accepted as a problem with multicollinearity. The tolerance value of hair colour was 0.89 < 1-R² (0.95). Therefore hair colour was eliminated. Then, a logistic regression analysis was carried out for the remaining variables.

For each variable, a reference category was determined. In this case, four variables (gender, age, skin types and eye colour) were all simultaneously included in the model, and logistic regression analysis was completed.

Results

The median count of total CMN and AMN counts in all cases were 2 (ranging from zero to 52), and 0 (ranging from zero to 8), respectively. The median counts of CMN in children with brown-black, hazel, blue, and green eyes were 2 (25-74 percentiles, 0-4), 2 (25-75 percentiles, 1-5), 2 (25-75 percentiles, 0.75-3.25), and 2 (25-75 percentiles, 1-4), respectively. The median counts of CMN in children with brown, black, blonde, and red hair were 2 (25-75 percentiles, 0-4), 2 (25-75 percentiles, 1-4), 2 (25-75 percentiles, 0-4), and 4.5 (2.5-5.75), respectively. The median counts of CMN in children with skin type I, II, III, and IV were 3 (25-75 percentiles, 2-5), 2 (25-75 percentiles, 1-5), 2 (25-75 percentiles, 0-4), and 1 (25-75 percentiles, 0-4), respectively. The median counts of CMN were 2 (25-75 percentiles, 0-4) in both genders.

Table 2 shows the effects of all variables, including phenotypical characteristics, considered in the bivariate analysis. In the logistic regression model, while gender and age produced weak odds ratios, light skin type emerged as moderate risk factor. Eye colour was not a significant predictor. Boys had more CMN than girls. We found an increase of CMN with age.

Table 3 shows the number of children with at least one CMN, and AMN by anatomical sites. According to these results, while the number of CMN seems to be higher on the ventral and dorsal trunk regions in boys, the number of CMN is higher on the hands and feet in girls. While the number of CMN seems to be higher on the head-neck region, the number of AMN seems to be higher on the ventral and dorsal trunk.
Table 1 The number of children with at least one CMN, and one AMN on any anatomical site by phenotypical characteristics

^**Denotes the reference category for each variable.

Table 3 The number of children with at least one CMN, and AMN by anatomic sites

Discussion

In previous studies it has been shown that, in patients with dysplastic nevus syndrome, there was a higher concentration of CMN in sun-exposed vs sun-protected areas [8, 11]. In addition, the number of nevi may increase or decrease with the amount of sun exposure [12-14]. In our study, the head-neck region had the highest number of CMN (40.5%) and it was the anatomical region that, for the subjects of our study, was more exposed to sunlight. As we did not define the risk factors for development of CMN, such as history of sunburns, and freckles, our research, just like the previous studies, could not clearly figure out the effect of sunlight on CMN development [8, 11-14]. This may be considered as a limitation of our study.

Gender variations in nevus density with higher nevus counts in boys compared to girls have been documented before [15-18]. Gallagher et al. [15] found that the densities of nevus in the ventral trunk and head-neck regions are higher in adult males compared to females. However, the nevus density in the upper and lower extremities is higher for females. It is not clear why girls have higher CMN count on their hands and feet. That the truncal region is exposed to sunlight in boys more than girls may explain why boys have a higher CMN count on their ventral and dorsal trunk. However the results of this study do not include the effects of sun exposure on the development of CMN.

Again, research performed on students in Holland, Switzerland and Australia has shown that CMN is higher in males and increases with age [16-18]. The results of our study show that CMN is more frequent in boys and its number increases with age. Also, as opposed to other research [8, 19, 20], the number of AMN does not increase with age.

Pigmentary factors indicate the children who are at a high risk for CMN development. The children with skin type II and fair hair colour are the most prone to develop many CMN. Skin type I has lower nevus counts, which might be due to better protection of these sun-sensitive children by their parents [9]. That skin types I and II were most prone to develop many CMN is shown to be important in our study, because people in our region do have not comprehensive information about sun-protection, the children with skin types I and II may not be sufficiently protected from the sun. However we did not use a questionnaire form, including questions about sun protection methods. This was another limitation of our study.

Dulon et al. [21] found that the median number of total nevus counts was 11, and no nevus was identified on 222 children (1.9%). In a study from Western Australia, English et al. [22] reported that the mean nevus count at age 10 was 7. Additionally, Autier et al. [23] found a median of 6 nevi ≥ 2 mm per child. Although the prevalence of CMN in our study is similar to those of other populations, we found that the median number of CMN counts in children in Anatolia was lower than those of the previous studies from other geographical regions. Sivas has a cold climate, and the people must be covered by clothes for more than six months of the year. This indispensable sun-protection may explain these lower CMN counts.

The estimated prevalence of clinically atypical nevi ranges from 7 to 18% in different populations. These variations, at least in part, may arise from the use of different diagnostic criteria in different studies. Additionally, differences in sun exposure among populations of different studies may also contribute to these variations. The predilection of atypical nevi for intermittently sun-exposed areas (especially the trunk), their positive association with a history of painful sunburn (in which the pain lasts more than two days) or blistering sunburn, and the finding that people with clinically atypical nevi often have sun-sensitive skin types, suggest that the development of dysplastic nevi could be related to acute, intense, sun exposure [24]. Because the highest number of CMN was on the head-neck region, and the highest number of AMN was on the ventral and dorsal trunk, no relationship between CMN and AMN according to anatomical region can be defined.

Different results were found in studies that analyzed the effects of hair color, skin type, and eye colour on nevus prevalence [17, 25-27]. Gallagher et al. [28] did not find a relationship between light skin colour or eye colour and nevus prevalence. In other studies, it was proposed that individuals having light skin, eye and hair colour may have a high nevus prevalence [17, 25-27]. Also, in our study, there was a significant relationship between light skin colour and the presence of CMN, in our population, but we did not find a relationship between light eye colour and CMN prevalence.

The results of our study and further studies which take sun-exposure characteristics and pigmentary factors into account may facilitate the determination of additional factors affecting nevus development and may help to design preventive strategies for cutaneous melanoma in our geographical area.

Acknowledgements

References

2 Augustsson A, Stierner U, Suurkula M, Rosdahl I. Prevalence of common and dysplastic naevi in a Swedish population. Br J Dermatol 1991; 124: 152-6.

3 Rozycki JL, Barret TL. The dysplastic nevus. Curr Probl Dermatol 2003: 170-5.

4 Naeyaert JM, Brochez L. Dysplastic nevi (clinical practice). N Engl J Med 2003; 349: 2233-40.

5 Bauer J, Garbe C. Acquired melanocytic nevi as risk factor for melanoma development. A comprehensive review of epidemiological data. Pigment Cell Res 2003; 16: 297-306.

6 Rodenas JM, Delgado-Rodriguez M, Farinas-Alvarez C, Herranz MT, Serrano S. Melanocytic nevi and risk of cutaneous malignant melanoma in Southern Spain. Am J Epidemiol 1997; 145: 1020-9.

7 Sivas Meteorology Station Directorship, Ministery of Environment and Forest, The Turkish Republic.

8 Valiukeviciene S, Miseviciene I, Gollnick H. The prevalence of common acquired melanocytic nevi and the relationship with skin type characteristics and sun exposure among children in Lithuania. Arch Dermatol 2005; 141: 579-86.

9 Bauer J, Büttner P, Wiecker TS, et al. Risk factors of incident melanocytic nevi: a longitudinal study in a cohort of 1232 young German children. Int J Cancer 2005; 115: 121-6.

10 Enta T, Kwan TY. Melanocytic nevi in sun-protected Canadian Hutterite children. Arch Dermatol 1998; 134: 379-81.

11 Kopf AW, Lindsay AC, Rogers GS, et al. Relationship of nevocytic nevi to sun exposure in dysplastic nevus syndrome. J Am Acad Dermatol 1985; 12: 656-62.

12 Armstrong BK, De Klerk NH, Holman CDJ. The aetiology of common acquired melanocytic nevi: constitutional variables, sun exposure, and diet. J Natl Cancer Inst 1986; 77: 329-35.

13 Augustsson A, Stierner U, Rosdahl I, et al. Melanocytic nevi in relation to sun exposure, and site-spesific counts predicting total number of nevi. Acta Derm Venereol (Stockh) 1992; 72: 123-7.

14 Harrison SL, Buettner PG, MacLennan L. The North Queensland “sun-safe clothing” study: Design and baseline results of a randomized trial to determine the effectiveness of sun-protective clothing in preventing melanocytic nevi. Am J Epidemiol 2005; 161: 536-45.

15 Gallagher RP, McLean DI, Yang CP, Coldman AJ, Silver HK, Spinelli JJ, Beagrie M. Anatomic distribution of acquired melanocytic naevi in white children. A comparison with melanoma: The Vancouver mole study. Arch Dermatol 1990; 125: 466-71.

16 Rampen FH, Van den Meeren HL, Boezeman JB. Frequency of moles as a key to melanoma incidence. J Am Acad Dermatol 1986; 15: 1200-3.

17 Sigg C, Pelloni F. Frequency of acquired melanocytic nevi and their relationship to skin complexion in 939 school children. Dermatologica 1989; 179: 123-8.

18 Kelly JW, Rivers JK, MacLennan R, Harrison S, Lewis AE, Tate BJ. Sunlight: A major factor associated with the development of melanocytic nevi in Australian school children. J Am Acad Dermatol 1994; 30: 40-8.

19 Rivers JK, MacLennan R, Kelly JW, et al. The Eastern Australian childhood nevus study: prevalence of atypical nevi, congenital nevus-like nevi, and other pigmented lesions. J Am Acad Dermatol 1995; 32: 957-63.

20 Augustsson A, Stierner U, Rosdahl I, et al. Common and dysplastic naevi as risk factors for cutaneous malignant melanoma in a Swedish population. Acta Derm Venereol 1991; 71: 518-24.

21 Grob JJ, Gouvernet J, Aymar D, et al. Count of benign melanocytic nevi as a major indicator of risk for nonfamilial nodular and superficial spreading melanoma. Cancer 1990; 66: 387-95.

22 Dulon M, Weichenthal M, Blettner M, et al. Sun exposure and number of nevi in 5- to 6-year-old European children. J Clin Epidemiol 2002; 55: 1075-81.

23 English DR, Milne E, Simpson JA. Sun protection and the development of melanocytic nevi in children. Cancer Epidemiol Biomarkers Biol 2005; 14: 2873-6.

24 Kang S, Barnhill RL, Mihm MC, et al. Melanoma risk in individuals with clinically atypical nevi. Arch Dermatol 1994; 130: 999-1001.

25 Pope DJ, Sorahan T, Marsden JR, Ball PM, Grimley RP, Peck IM. Benign pigmented nevi in children. Prevalence and associated factors: The West Midlands, United Kingdom mole study. Arch Dermatol 1992; 128: 1201-6.

26 English DR, Armstrong BK. Melanocytic nevi in children. I. Anatomic sites and demographic and host factors. Am J Epidemiol 1994; 139: 390-401.

27 English DR, Armstrong BK. Melanocytic nevi in children II: Observer variation in counting nevi. Am J Epidemiol 1994; 139: 402-7.

28 Gallagher RP, McLean DI. The epidemiology of acquired melanocytic nevi. A brief review. Dermatol Clin 1995; 13: 595-603.