Frequency and Dermoscopic Features of Congenital Melanocytic Naevi in Antalya, Turkey and Review of the Literature

Background: Congenital melanocytic naevi (CMN) are present at birth or they develop during the first year of life. In the literature, there is no study prospectively determining the prevalence and dermoscopic patterns of CMN among 0-12 months old infants. There is no data about its frequency in Turkish population. Dermoscopic diagnostic criteria for CMN are not clear. Aim: The aim of this study is to determine the frequency of CMN in our region and to define their dermoscopic properties. Methods: 4289 consecutive infants aged between 0 and 12 months, admitted to pediatric outpatient clinics of Akdeniz University Hospital were included in this study. Infants were first examined by a pediatrician and those with naevi were consulted to a dermatologist. Lesions of CMN were evaluated with camerated dermatoscopy device and images were recorded to a computer. Results: CMN were detected in 0,5% (n:20) of cases. Lower extremity was the most common location (60%), followed by scalp (15%), trunk (10%), upper extremity (10%) and more than one location (5%). Naevi diameter was 1.5-20 cm in 11 cases, smaller than 1,5cm in seven cases and greater than 20cm in two cases. In dermoscopic examination dot structure (n:18) and globular pattern (n:15) were mostly observed. Other findings included multifocal hypopigmentation pattern (n:6), reticular pattern (n:4), homogeneous pattern (n:4), cobblestone pattern (n:2), and parallel furrow pattern (n:1). Conclusion: This is the first prospective study to determine the prevalence and dermoscopic patterns of CMN among 0-12 months old infants admitted to a pediatric outpatient clinic. The frequency of CMN in our region is similar with some of the previous studies. Our Dermoscopic results support that CMN appearing in younger ages is more often characterized by a globular pattern and dot structure.


Introduction
Congenital melanocytic nevi (CMN) are present at birth or they develop during the first year of life. In current literature frequency is reported as 1-6% [1]. Although CMN can be seen all over the world, clinical and epidemiological features can show regional differences. To date, little is known about the frequency and dermoscopic patterns of congenital nevi in a Turkish population. In Elazığ, the researchers evaluated dermatological findings of 150 babies during the first three days of life. However, the sample size was too small to determine the frequency of CMN [2].
CMN have a risk of developing malignant melanoma [3]. Melanoma in children generally occur extremely rare and the increased risk of melanoma is associated with large and giant CMN [4]. Although most melanomas develop before the age of 10 years, a recent study showed that it can develop later. Therefore, Yun et al. recommend lifelong follow-up of giant CMN due to melanoma risk [5]. Dermoscopy is widely used in the diagnosis and follow-up of CMN [6]. Few studies have investigated the dermoscopic features of CMN [7][8][9][10]. There is only one study on nevi onset in the first years of life [10]. In this study, a total of 103, two year-old children with the presence of at least one naevus attending to dermatology clinic were recruited for the study. However, the study had some limitations, as lesions with a diameter of <3 cm were included and those located on mucosal or subungual sites were excluded.
In the literature, there is no study prospectively determining the prevalence and dermoscopic patterns of CMN among 0-12 months old infants. The aim of this study was to determine prevalence, clinical and dermoscopic features of CMN in our region among 0-12 months old infants admitted to a pediatric outpatient clinic.

Methods
For an estimated prevalence of 1% with a 0.80 statistical power, 95% confidence interval, and statistical significance set at p<0.05, required sample size was 2000. Between November 2007 and September 2009, 4289 consecutive patients between 0 and 12 months of age, admitted to Akdeniz University Hospital's Well Child Clinic were evaluated in the study. Infants were first examined by a pediatrician and patients with nevi were consulted to dermatology clinic. The families of the patients gave informed consent. The study protocol was approved by the Ethics Committee of Akdeniz University, School of Medicine (Project No: 2007.04.0103.004). Genders and ages of cases, location, size and color of nevi and history of getting phototherapy were recorded. Skin types of the patients were determined according to Fitzpatrick skin phototype classification system [11].
Lesions of CMN were evaluated with a camerated dermatoscopy device (Heine Delta 20, Optotechnik GmbH & co. KG, Herrsching, Germany with Canon Powershot A540) by EK and AAK together. The dermatoscope was attached to the camera by Heine-Fotoadapter. The contact plate of the dermoscope was moistened with immersion fluid before photography. Photographic data on the memory card were transferred to patient's folders on the computer. Dermoscopic structures were evaluated according to Braun et al. [12] article reviewing the principles of dermoscopy.

Statistical Analysis
Descriptive analysis of the sample was performed, including percentages for categorical variables and the median age and standard deviation.

Results
CMN was detected in 20 infants admitted to Akdeniz University's Well Child Clinic. Nine of the patients were female and 11 were male, showing a slight male predominance (male to female ratio 1.22). Median age ± standard deviation was 196 ± 215 days. Lesions were light brown in 15 and dark brown in five nevi.
The shortest diameter was 0.3 cm and the longest diameter was above 20 cm. One case had a history of receiving phototherapy. Six cases (30%) had type 1 and 14 (70%) had type 2 skin-type characteristics. Patients' age, gender, skin types, localization and clinical features of the lesions are shown in Table 1.
In dermoscopic examination dot structure (n:18) and globular pattern (n:15) were mostly observed. Other findings included multifocal hypopigmentation pattern (n:6), reticular pattern (n:4), homogeneous pattern (n:4), cobblestone pattern (n:2), and parallel furrow pattern (n:1). In the patient 3 who had parallel furrow pattern, the lesions were located on the palmar/plantar surfaces of the hands and feet. On the other hand, patients with multifocal hypopigmentation pattern had lesions on the areas with dense hair follicules. In some of the cases there were more than one dermoscopic properties. Dermoscopic features are given in Table 1 and commonly observed dermoscopic patterns are presented in one multipart figure.
The patients did not have any melanoma during a follow-up period of 5 years.

Discussion
CMN was detected in 0.5% of cases. There is only a few studies about the prevalence of CMN in child population. We were unable to find any study that provides prevalence and dermoscopic patterns of CMN among 0-1 year old children. Therefore, we compared our findings with studies investigating prevalence in newborns. In Turkey, there is only one report about frequency of CMN [2]. In Elazığ, a city located in Eastern Anatolia, Ozturk et al. evaluated dermatological findings of 150 babies in the first three days and found one CMN (0.6%) among these newborns. However, sample size was very limited for determining the frequency of CMN and dermoscopic findings, and size or location of the lesion were not specified. Studies including the frequency of CMN in newborns are summarized in Table 2. In some European countries [13][14][15][16][17][18][19][20][21][22][23], studies reported prevalence rates ranging from 0.2% to 6%. These large differences can be due to the clinical heterogeneity of the studies, in terms of methods and geographical and genetic variations. The lowest incidence of the disease has been reported from the Sweden as 0.2% [15]. In this study, the data was obtained from a revised International Classification of Diseases (ICD) database. The data entry had been made for the patient when the lesion was biopsied. Thus, the data may not be good enough for the true frequency in the population. Interestingly, in the study of Lorenz et al. [18] from Germany, the incidence was notably high (6%) when compared to the other European countries. Increased frequency of CMN in Germany shows that the genetic and geographic factors may play an important role for the development of the disease.
Our report is similar with some of the previous studies. As seen in Table 2, there are regional differences in the Mediterranean coast. Our city, which has a sunny and humid climate, is a major immigration area especially in summer for seasonal laborers from the Eastern part of Turkey. The regional factors can depend on the genetic background of the populations. A recent study by Kinsler et al. [24] also supports the genetical background in the phenotype of the CMN. They noted that there were significantly fewer blonde/light brown-haired individuals in the CMN cohort than in control groups and they identified associations between families with CMN-affected children and the red hair pigmentary phenotype, and associations between melanocortin-1-receptor gene variants and the presence and extent of CMN. In our study, all children had a fair pigmentary type. It is hard to make definitive comparisons with the skin types of the general population of Turkish people because of the lack of previous publication on skin type [25][26][27]. Moreover, the Fitzpatrick skin phototype is determined by constitutional colour and the result of exposure to ultraviolet radiation and it is hard to describe the phototype in children aged 0-12 months in the aspect of lack of tanning and exposition to sunburns. However, Tiftikcioglu et al. [25] remarked most of the population of their clinic had type 2 and 3 skins ranging from Fitzpatrick type 1 to 5. Akyol et al. [26] showed that common melanoctic nevus in children was more frequent in children having a lighter skin type, and their numbers increase with age.
In terms of lesion size, studies in newborns showed that small CMN was the most common type ( Table 2). In our study, medium CMN was more common. This discrepancy with the others may be due to the enrollment of a different age group (0-12 months olds) in our study. These studies also reported that the trunk was the most common anatomical location in newborns. Stinco et al. 's [10] studies show that CMN can be located on the lower extremities in the first years of life. These findings suggest that the most common localization areas of CMN may change over time with age and with the continuation of development of the nevi and the individual.  Several studies investigated the dermoscopic features of CMN in child population [6][7][8][9][28][29][30][31]. Studies including the dermoscopic features are summarized in Table 3.

Studies
Country  In these studies, patients attending dermatology or pediatric dermatology clinics were recruited and age groups differed from birth to 15 years of age. Oliveria et al. [7] and Zalaudek et al. [8] found a marked predominance of globular pattern (82%) among children between 0-15 years of age and at 12 years of age, respectively. Stinco et al. [10] showed that the predominant dermoscopic patterns were globular (CMN present at birth 51%; naevi appearing after birth during the first 2 years of life 58%) and reticular (CMN present at birth 28%; naevi appearing after birth during the first 2 years of life 14%). Lesions with a diameter of >3 cm or that were located on mucosal or subungual sites were excluded from their study.
Anguilera et al. [9] investigated the number and dermoscopic patterns of melanocytic nevi among individuals of 1-15 years of age. They found that 81.1% had nevi with a globular pattern and 52.8% with a reticular pattern. They reported the rate of each pattern was increasing with age. Regarding the pattern predominance, they observed that homogeneous pattern which is more common in younger ages was decreasing with age, while reticular pattern was more predominant in older ages and the frequency of a predominant globular pattern was constant among all ages. According to this, age might be a factor for having a dominant reticular pattern, but not for having a dominant globular pattern.
Mangononi et al. [32], in their study, investigated the dermoscopic, histological and immunohistochemical cancerous features in acquired melanocytic nevi that have been repeatedly exposed to UVA or UVB. They reported that nevi exposed to NB-UVB or UVA1 showed statistically significant increase in size and changes in their dermoscopic features, including overall darkening, increased reticular pattern and increased number and size of brown globules and dots. Unfortunately, the size of the dataset of our study, with only 20 CMN, is inadequate to perform a statistical analysis on dermoscopic patterns or distribution to specific body sites.
CMN can develop to other patterns in the course of naevus evolution [10,33,34]. Specific dermoscopic patterns can be observed in some anatomic localizations depending on skin histology  [28][29][30][35][36][37]. In our study, we observed parallel furrow pattern in palmar surfaces of hands and feet, and multifocal hypopigmentation pattern in areas with high concentration of hair follicles due to perifollicular hypopigmentation.
Our study is the first study to show clinical features and dermoscopic patterns of CMN in children from our population. In addition, the design of the study prospectively determine the prevalence and dermoscopic patterns of CMN among 0-12 months old infants admitted to a pediatric outpatient clinic and makes this study different from previous reports in the literature.