- Email: [email protected]
Patterns of distribution of giant congenital melanocytic nevi (GCMN): The 6B rule
Patterns of distribution of giant congenital melanocytic nevi (GCMN): The 6B rule Vanessa P. Martins da Silva, MD,a Ashfaq Marghoob, MD,b Ramon Pigem, MD,a Cristina Carrera, PhD,a,c Paula Aguilera, PhD,a Joan A. Puig-Butille, PhD,c,d Susana Puig, PhD,a,c and Josep Malvehy, PhDa,c Barcelona, Spain, and New York, New York Background: Garment-related terms have been used to describe the pattern of distribution of giant congenital melanocytic nevi (GCMN). Objective: We sought to describe patterns of distribution of GCMN and propose a classification scheme. Methods: Photographic records of patients with GCMN from the Hospital Clinic of Barcelona were analyzed and a classification based on observed GCMN distribution patterns was created. The classification was independently applied by 8 observers to cases found in the literature. The interobserver agreement was assessed. Results: Among 22 patients we observed 6 repeatable patterns of distribution of GCMN, which we termed the ‘‘6B’’: bolero (involving the upper aspect of the back, including the neck), back (on the back, without involvement of the buttocks or shoulders), bathing trunk (involving the genital region and buttocks), breast/belly (isolated to the chest or abdomen without involvement of bolero or bathing trunk distributions), body extremity (isolated to extremity), and body (both bolero and bathing trunk involvement). Our literature search found 113 cases of GCMN, which we were able to classify into 1 of the 6B patterns with an overall kappa of 0.89. Limitations: Some patterns occur infrequently with a dearth of images available for analysis. Conclusions: The anatomic distribution of GCMN occurs in 6 recognizable and repeatable patterns. ( J Am Acad Dermatol 2017;76:689-94.) Key words: benign skin tumors; classification; congenital melanocytic nevus; distribution; giant nevus; skin neoplasms. From the Melanoma Unit, Department of Dermatology, Hospital Clinic of Barcelona, University of Barcelonaa; Department of Dermatology, Memorial Sloan Kettering Cancer Center, New Yorkb; ‘‘Centro de Investigaci on Biomedica en Red en Enfermedades Raras, Instituto de Salud Carlos III,’’ Barcelonac; and Department of Biochemical and Molecular Genetics, Hospital Clınic, IDIBAPS, University of Barcelona.d The research at the Melanoma Unit of the Hospital Clinic of Barcelona is partially funded by grants 03/0019, 05/0302, 06/0265, 09/01393, and 12/00840 from Fondo de Investigaciones Sanitarias, Spain; by the Centro de Investigaci on Biom edica en Red de Enfermedades Raras, Instituto de Salud Carlos III, Spain; by the AGAUR 2009 SGR 1337 and AGAUR 2014 SGR 603 of the Catalan Government, Spain; by the European Commission under the Sixth Framework Program, contract number LSHC-CT-2006-018702 (GenoMEL), and under the Seventh Framework Program (Diagnoptics); by the National Cancer Institute of the US National Institutes of Health (CA83115); by de TV3 Foundation; by the Telemaraton Foundathe MARATO tion (research grant ‘‘Todos somos raros’’); and by the Leo Messi Foundation. The project ‘‘Phenotype and molecular characterization of large to giant congenital melanocytic nevi’’ has
received funding from the Spanish Federation of Neuromuscular Disease, the Spanish Federation of Rare Diseases, and Isabel Gemio Research Foundation for muscular dystrophy and other rare diseases through the call for research projects on rare diseasese2014 through the initiative ‘‘We are rare, all are unique.’’ The funders had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; in the decision of the submission of the manuscript; or in the preparation, review, or approval of the manuscript. Conflicts of interest: None declared. Accepted for publication May 30, 2016. Reprint requests: Josep Malvehy, PhD, Melanoma Unit, Department of Dermatology, Hospital Clinic of Barcelona, C/Villarroel, 170. Esc 3-4, 08036 Barcelona, Spain. E-mail: [email protected]. 0190-9622 Ó 2016 by the American Academy of Dermatology, Inc. Published by Elsevier, Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-ncnd/4.0/). http://dx.doi.org/10.1016/j.jaad.2016.05.042
689
J AM ACAD DERMATOL
690 Martins da Silva et al
APRIL 2017
Congenital melanocytic nevi (CMN) have an distribution classification with initial definitions and estimated prevalence of between 0.5% and 31.7%.1schematics termed as the ‘‘6B’’ (Fig 2). Two other 5 CMN larger than 20 cm are uncommon with an observers independently reclassified the same cases estimated incidence of between 1 in 20,000 to 1 in using the proposed classification scheme. Cases that 500,000 births.2,6 The current classification of CMN presented disagreement in the classification were (Fig 1) is based mainly on the maximum diameter jointly re-evaluated for consensus. This exercise that the nevus is predicted to attain during adultlead to fine-tuning of the classification and the final hood.7 A CMN is classified as schematic was created (Fig large-1 if it measures be2). Two different observers CAPSULE SUMMARY tween 20 to 30 cm, large-2 if independently classified the it measures between 30 and cases according to the Garment-related terms have been used 40 cm, giant-1 if it measures proposed final classification to describe the anatomic localization of between 40 and 60 cm, and scheme and interobserver giant congenital melanocytic nevi. giant-2 if greater than 60 cm agreement was assessed. A classification scheme, with good in maximal diameter.7 This interobserver agreement, is proposed to classification scheme also inPhase II standardize the distribution patterns of cludes additional morphoTo test the reproducibility giant melanocytic nevi. logic characteristics of the and applicability of the 6B CMN: anatomic localization, classification scheme, an atlas A simple, reliable, and repeatable degree of color heteroof published images of paclassification can help to improve geneity, surface rugosity, tients with GCMN was assemcommunication and patient risk hypertrichosis, dermal/subbled. All articles identified via stratification. cutaneous nodularity, and literature search using number of smaller satellite PubMed containing the terms nevi. ‘‘giant congenital nevi’’ and published between To highlight the distribution of giant CMN January 1, 1998, and April 30, 2014, were evaluated (GCMN), Bircher,8 in 1897, used terms such as for the presence of images of GCMN. Images were ‘‘bathing trunk,’’ ‘‘vest-like,’’ and ‘‘vest with collar.’’ selected if they met the following criteria: (1) images In 1965, Reed et al9 described 55 patients using terms showing the GCMN with a projected adult size larger such as ‘‘bathing trunk,’’ ‘‘shoulder stole’’ or than 40 cm in which the entire GCMN was visible with ‘‘cape-like,’’ ‘‘coat sleeve,’’ and ‘‘stocking-like.’’ In back, front, and/or side body sectors shown; and 2005, Torrelo et al10 separated 1188 nevi based on (2) images showing the main portion of the the following morphologic categories: round, patchy GCMN with a written description that adequately indented or triangular shape, agminated, diffuse helped the reader appreciate the full extent of its patchy, Blaschkolinear, block/flag-like, and distribution. Images excluded from the study garment-like distribution. atlas included those showing only a portion of Based on the garment-related terms used to the GCMN where the full extension of the GCMN describe the distribution patterns of large or GCMN, could not be deciphered, case reports that were we hypothesized that GCMN have repetitive recogdeemed not to represent CMN or GCMN, duplicate nizable patterns of distribution. The aim of this study images of the same patient in multiple publications, was to evaluate the distribution patterns of GCMN and and GCMN where portions of the nevus had been to determine whether recognizable patterns emerge. excised. The workflow for phase II of this study is shown in Fig 3. All cases included in the study atlas were METHODS classified according to the 6B classification scheme Phase I by 8 physicians. Two weeks later, 4 of the 8 A query of the medical records at the Hospital physicians classified all cases for a second time. Clinic of Barcelona was performed to identify They did not have access to the answers previously patients diagnosed with GCMN between 1975 and provided. The new answers were used to assess the 2013. The charts were reviewed to identify patients intraobserver agreement. who had full-body clinical photographs of CMN with a projected adult size larger than 40 cm and in whom Statistical analysis the full extent (location, distribution, and size) of the All statistical analyses were performed using GCMN was visible on the photographs. The included software (SPSS Statistics for Windows, Version 22.0, cases were analyzed by 2 observers to determine the IBM Corp, Armonk, NY). Descriptive data were patterns of distribution and to create an anatomic d
d
d
J AM ACAD DERMATOL
Martins da Silva et al 691
VOLUME 76, NUMBER 4
Abbreviations used: CMN: GCMN: NCM:
congenital melanocytic nevi giant congenital melanocytic nevi neurocutaneous melanosis
reported as mean 6 SD and/or as percentages. Interobserver and intraobserver agreement between the 2 evaluators in phase I was assessed by unweighted kappa. In phase II, the interobserver agreement between evaluators was assessed by Fleiss kappa. The results were interpreted based on the proposed 5 levels of Landis and Koch11 with values of 0 to .20 representing slight agreement, .21 to .40 fair agreement, .41 to .60 moderate agreement, .61 to .80 substantial agreement, and .81 to 1 almost perfect agreement. Ethical issues The study complied with the Declaration of Helsinki Principles and was approved by the Clinical Research Ethics Committee of the Hospital Clinic of Barcelona.
RESULTS Phase I In all, 24 patients with GCMN were identified after review of the medical records. Two patients were excluded because images revealed that significant portions of the GCMN had been excised. The remaining 22 patients ranged in age between 4 days and 51 years at the time the images were acquired. The majority of patients (68.2%) were younger than 10 years. Thirteen patients were female (59.1%). Four patients had neurocutaneous melanosis (NCM), 2 of whom had a primary central nervous system melanoma and died from their disease. The interobserver agreement between 2 observers using the initial 6B classification scheme was 0.758. After refinements, resulting in the final version of the classification, the overall agreement between 2 observers improved to 1. The kappa values for each of the 6B patterns in each phase of the study are presented in Table I. Fig 2 defines the 6B patterns of distribution via schematics and images depicting exemplar cases. The most frequent distribution pattern observed was bathing trunk (45.5%), followed by bolero (27.4%), back (13.6%), breast/ belly (4.5%), body (4.5%), and body extremity (4.5%). Phase II From the literature search we identified 275 potential articles containing images of GCMN. We ultimately found 113 images of cases of GCMN that
met our inclusion criteria. The overall interobserver agreement among the 8 observers resulted in a kappa of 0.891 (0.722-0.962). The 4 observers who evaluated the cases at 2 separate time points had intraobserver agreement kappa scores of 0.941, 0.942, 0.927, and 0.913 (mean 0.931). Most frequently observed was the bathing trunk pattern (39.8% of cases) followed by bolero (23%), back (17.7%), breast/belly (8%), body extremity (9.7%), and body (1.8%).
DISCUSSION Our study highlights that GCMN follow reproducible distribution patterns represented by the 6Bs (Fig 2). This scheme can be used to classify the distribution pattern of most/all GCMN reported in the literature with high levels of intraobserver and interobserver agreement. It has been shown that phenotypic characteristics of individuals with GCMN can help predict risk for developing NCM, melanoma, or other complications. The risk for developing melanoma correlates with the size of the CMN,12 and the number of satellite nevi has been shown to correlate with risk for developing NCM.13 The phenotype may also provide insight regarding the timing of events during embryogenesis. For example, ‘‘kissing nevi’’ on the eyelids (a CMN with one portion located on the upper eyelid and the other on the lower eyelid) indicate that the aberrant melanocyte migration took place between the 9th week of embryogenesis, when the upper and lower lids fuse, and the 20th week, when the lids separate.14,15 The location of the GCMN may also be a factor or a surrogate marker of risk. It has been suggested that the bathing trunk pattern may be associated with an increased risk for melanoma and NCM.9 The bathing trunk distribution encountered in 45.5% of our patients (n = 10) and in 39.8% (n = 45) of cases reported in the literature, compares with previously reported figures of 44% (n = 15)8 and 47.3% (n = 26).9 This pattern was observed in 33 of 39 published patients with melanoma and in 12 of 20 patients with NCM.9 It is noteworthy that our 2 patients who died from melanoma presented with a bathing trunk GCMN. Although the descriptor ‘‘bathing trunk’’ is commonly used, no clear or agreed upon definitions exist for this or other garment-related terms. Future studies incorporating molecular information together with genetic, environmental, and other factors related to embryogenesis may someday help explain why GCMN are distributed in these patterns. It is likely that a combination of factors including the type of mutation, the location of the progenitor
692 Martins da Silva et al
J AM ACAD DERMATOL
Fig 1. The criteria for congenital melanocytic nevi characterization.7 CMN, Congenital melanocytic nevi.
Fig 2. The 6B classification for giant congenital melanocytic nevi: definitions and clinical examples of the 6 patterns of distribution. *Gray area represents the extension variation that the nevi may present. **Satellite lesionseeven the large oneseshould be disregarded when classifying the body pattern.
APRIL 2017
J AM ACAD DERMATOL
Martins da Silva et al 693
VOLUME 76, NUMBER 4
Fig 3. Flow chart of the case selection process used in phase II. CMN, Congenital melanocytic nevi; GCMN, giant congenital melanocytic nevi.
Table I. Kappa value for each category in all the phases Pattern
Bolero Back Bathing trunk Breast/belly Body extremity Body
Phase I initial
Phase I final
Phase II
0.593 1 0.876 1 1 0.345
1 1 1 1 1 1
0.888 0.825 0.915 0.912 0.962 0.722
mutated cell, the timing of the mutation being expressed during embryogenesis, and the presence of polymorphisms all play a role in sculpting the
morphology and distribution patterns encountered in GCMN. Our study has limitations. GCMN are rare and some of the patterns described in the 6B scheme are seen infrequently. This may explain why some patterns had lower interobserver agreements. Multicenter studies with a larger number of cases are required to validate our findings. The 6B classification scheme for GCMN excluded nevi located on the head because very few nevi in this location were found to be larger than 40 cm. However, a classification scheme incorporating nevi located on the head should be considered. In conclusion, our study found that most GCMN can be categorized into 1 of 6 distinct anatomic
J AM ACAD DERMATOL
694 Martins da Silva et al
distribution patterns with good interobserver and intraobserver agreement. Based on our study we propose the standardization of terminology used to describe GCMN garment distribution patterns. We favor addition of the 6B classification scheme to the recently reported recommendations for the categorization of cutaneous features of CMN to facilitate collaborative research and allow for more reliable comparisons of study findings. We are grateful to our patients and relatives, who are always committed to collaborate with our studies; the dermatologists Alejandra Villarreal, Marisa Zimmermann, Luz Casta~ neda, Luz Calderon, Caroline Takigami, Sebastian Mercau, and Adriana Pe~ na y Gimena Bolomo for reviewing the cases; and the nurses Daniel Gabriel, Pablo Iglesias, and M. Eugenia Moliner, and technicians Abel Ca~ no, Mireia Domınguez, and Beatriz Alejo for their help.
APRIL 2017
5.
6.
7.
8. 9.
10.
11. REFERENCES 1. Viana ACL, Gontijo B, Bittencourt FV. Giant congenital melanocytic nevus. An Bras Dermatol. 2013;88(6):863-878. 2. Alikhan A, Ibrahimi OA, Eisen DB. Congenital melanocytic nevi: where are we now? Part I. Clinical presentation, epidemiology, pathogenesis, histology, malignant transformation, and neurocutaneous melanosis. J Am Acad Dermatol. 2012;67(4): 495.e1-495.e17. quiz 512-514. 3. Monteagudo B, Labandeira J, Le on-Mui~ nos E, et al. Prevalence of birthmarks and transient skin lesions in 1,000 Spanish newborns [in Spanish]. Actas Dermo-Sifiliogr aficas. 2011;102(4): 264-269. 4. Palau-Lazaro MC, Buendıa-Eisman A, Serrano-Ortega S. Prevalence of congenital nevus in 1000 live births in
12.
13.
14.
15.
Granada, Spain [in Spanish]. Actas Dermo-Sifiliogr aficas. 2008; 99(1):81. Aguilera P, Puig S, Guilabert A, et al. Prevalence study of nevi in children from Barcelona. Dermoscopy, constitutional and environmental factors. Dermatol Basel Switz. 2009;218(3):203-214. Lovett A, Maari C, Decarie J-C, et al. Large congenital melanocytic nevi and neurocutaneous melanocytosis: one pediatric center’s experience. J Am Acad Dermatol. 2009; 61(5):766-774. Krengel S, Scope A, Dusza SW, Vonthein R, Marghoob AA. New recommendations for the categorization of cutaneous features of congenital melanocytic nevi. J Am Acad Dermatol. 2013;68(3):441-451. Bircher MO. Zur Aetiologie des Nevus pilosus pigmentosus congenitus extensus. Arch F€ ur Dermatol Syph. 1897;41:195-225. Reed WB, Becker SW Sr, Becker SW Jr, Nickel WR. Giant pigmented nevi, melanoma and leptomeningeal melanocytosis: a clinical and histopathological study. Arch Dermatol. 1965;91:100-119. Torrelo A, Baselga E, Nagore E, Zambrano A, Happle R. Delineation of the various shapes and patterns of nevi. Eur J Dermatol EJD. 2005;15(6):439-450. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-174. Krengel S, Hauschild A, Sch€afer T. Melanoma risk in congenital melanocytic nevi: a systematic review. Br J Dermatol. 2006; 155(1):1-8. Marghoob AA, Dusza S, Oliveria S, Halpern AC. Number of satellite nevi as a correlate for neurocutaneous melanocytosis in patients with large congenital melanocytic nevi. Arch Dermatol. 2004;140(2):171-175. Hamming N. Anatomy and embryology of the eyelids: a review with special reference to the development of divided nevi. Pediatr Dermatol. 1983;1(1):51-58. Desai SC, Walen S, Holds JB, Branham G. Divided nevus of the eyelid: review of embryology, pathology and treatment. Am J Otolaryngol. 2013;34(3):223-229.
received funding from the Spanish Federation of Neuromuscular Disease, the Spanish Federation of Rare Diseases, and Isabel Gemio Research Foundation for muscular dystrophy and other rare diseases through the call for research projects on rare diseasese2014 through the initiative ‘‘We are rare, all are unique.’’ The funders had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; in the decision of the submission of the manuscript; or in the preparation, review, or approval of the manuscript. Conflicts of interest: None declared. Accepted for publication May 30, 2016. Reprint requests: Josep Malvehy, PhD, Melanoma Unit, Department of Dermatology, Hospital Clinic of Barcelona, C/Villarroel, 170. Esc 3-4, 08036 Barcelona, Spain. E-mail: [email protected]. 0190-9622 Ó 2016 by the American Academy of Dermatology, Inc. Published by Elsevier, Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-ncnd/4.0/). http://dx.doi.org/10.1016/j.jaad.2016.05.042
689
J AM ACAD DERMATOL
690 Martins da Silva et al
APRIL 2017
Congenital melanocytic nevi (CMN) have an distribution classification with initial definitions and estimated prevalence of between 0.5% and 31.7%.1schematics termed as the ‘‘6B’’ (Fig 2). Two other 5 CMN larger than 20 cm are uncommon with an observers independently reclassified the same cases estimated incidence of between 1 in 20,000 to 1 in using the proposed classification scheme. Cases that 500,000 births.2,6 The current classification of CMN presented disagreement in the classification were (Fig 1) is based mainly on the maximum diameter jointly re-evaluated for consensus. This exercise that the nevus is predicted to attain during adultlead to fine-tuning of the classification and the final hood.7 A CMN is classified as schematic was created (Fig large-1 if it measures be2). Two different observers CAPSULE SUMMARY tween 20 to 30 cm, large-2 if independently classified the it measures between 30 and cases according to the Garment-related terms have been used 40 cm, giant-1 if it measures proposed final classification to describe the anatomic localization of between 40 and 60 cm, and scheme and interobserver giant congenital melanocytic nevi. giant-2 if greater than 60 cm agreement was assessed. A classification scheme, with good in maximal diameter.7 This interobserver agreement, is proposed to classification scheme also inPhase II standardize the distribution patterns of cludes additional morphoTo test the reproducibility giant melanocytic nevi. logic characteristics of the and applicability of the 6B CMN: anatomic localization, classification scheme, an atlas A simple, reliable, and repeatable degree of color heteroof published images of paclassification can help to improve geneity, surface rugosity, tients with GCMN was assemcommunication and patient risk hypertrichosis, dermal/subbled. All articles identified via stratification. cutaneous nodularity, and literature search using number of smaller satellite PubMed containing the terms nevi. ‘‘giant congenital nevi’’ and published between To highlight the distribution of giant CMN January 1, 1998, and April 30, 2014, were evaluated (GCMN), Bircher,8 in 1897, used terms such as for the presence of images of GCMN. Images were ‘‘bathing trunk,’’ ‘‘vest-like,’’ and ‘‘vest with collar.’’ selected if they met the following criteria: (1) images In 1965, Reed et al9 described 55 patients using terms showing the GCMN with a projected adult size larger such as ‘‘bathing trunk,’’ ‘‘shoulder stole’’ or than 40 cm in which the entire GCMN was visible with ‘‘cape-like,’’ ‘‘coat sleeve,’’ and ‘‘stocking-like.’’ In back, front, and/or side body sectors shown; and 2005, Torrelo et al10 separated 1188 nevi based on (2) images showing the main portion of the the following morphologic categories: round, patchy GCMN with a written description that adequately indented or triangular shape, agminated, diffuse helped the reader appreciate the full extent of its patchy, Blaschkolinear, block/flag-like, and distribution. Images excluded from the study garment-like distribution. atlas included those showing only a portion of Based on the garment-related terms used to the GCMN where the full extension of the GCMN describe the distribution patterns of large or GCMN, could not be deciphered, case reports that were we hypothesized that GCMN have repetitive recogdeemed not to represent CMN or GCMN, duplicate nizable patterns of distribution. The aim of this study images of the same patient in multiple publications, was to evaluate the distribution patterns of GCMN and and GCMN where portions of the nevus had been to determine whether recognizable patterns emerge. excised. The workflow for phase II of this study is shown in Fig 3. All cases included in the study atlas were METHODS classified according to the 6B classification scheme Phase I by 8 physicians. Two weeks later, 4 of the 8 A query of the medical records at the Hospital physicians classified all cases for a second time. Clinic of Barcelona was performed to identify They did not have access to the answers previously patients diagnosed with GCMN between 1975 and provided. The new answers were used to assess the 2013. The charts were reviewed to identify patients intraobserver agreement. who had full-body clinical photographs of CMN with a projected adult size larger than 40 cm and in whom Statistical analysis the full extent (location, distribution, and size) of the All statistical analyses were performed using GCMN was visible on the photographs. The included software (SPSS Statistics for Windows, Version 22.0, cases were analyzed by 2 observers to determine the IBM Corp, Armonk, NY). Descriptive data were patterns of distribution and to create an anatomic d
d
d
J AM ACAD DERMATOL
Martins da Silva et al 691
VOLUME 76, NUMBER 4
Abbreviations used: CMN: GCMN: NCM:
congenital melanocytic nevi giant congenital melanocytic nevi neurocutaneous melanosis
reported as mean 6 SD and/or as percentages. Interobserver and intraobserver agreement between the 2 evaluators in phase I was assessed by unweighted kappa. In phase II, the interobserver agreement between evaluators was assessed by Fleiss kappa. The results were interpreted based on the proposed 5 levels of Landis and Koch11 with values of 0 to .20 representing slight agreement, .21 to .40 fair agreement, .41 to .60 moderate agreement, .61 to .80 substantial agreement, and .81 to 1 almost perfect agreement. Ethical issues The study complied with the Declaration of Helsinki Principles and was approved by the Clinical Research Ethics Committee of the Hospital Clinic of Barcelona.
RESULTS Phase I In all, 24 patients with GCMN were identified after review of the medical records. Two patients were excluded because images revealed that significant portions of the GCMN had been excised. The remaining 22 patients ranged in age between 4 days and 51 years at the time the images were acquired. The majority of patients (68.2%) were younger than 10 years. Thirteen patients were female (59.1%). Four patients had neurocutaneous melanosis (NCM), 2 of whom had a primary central nervous system melanoma and died from their disease. The interobserver agreement between 2 observers using the initial 6B classification scheme was 0.758. After refinements, resulting in the final version of the classification, the overall agreement between 2 observers improved to 1. The kappa values for each of the 6B patterns in each phase of the study are presented in Table I. Fig 2 defines the 6B patterns of distribution via schematics and images depicting exemplar cases. The most frequent distribution pattern observed was bathing trunk (45.5%), followed by bolero (27.4%), back (13.6%), breast/ belly (4.5%), body (4.5%), and body extremity (4.5%). Phase II From the literature search we identified 275 potential articles containing images of GCMN. We ultimately found 113 images of cases of GCMN that
met our inclusion criteria. The overall interobserver agreement among the 8 observers resulted in a kappa of 0.891 (0.722-0.962). The 4 observers who evaluated the cases at 2 separate time points had intraobserver agreement kappa scores of 0.941, 0.942, 0.927, and 0.913 (mean 0.931). Most frequently observed was the bathing trunk pattern (39.8% of cases) followed by bolero (23%), back (17.7%), breast/belly (8%), body extremity (9.7%), and body (1.8%).
DISCUSSION Our study highlights that GCMN follow reproducible distribution patterns represented by the 6Bs (Fig 2). This scheme can be used to classify the distribution pattern of most/all GCMN reported in the literature with high levels of intraobserver and interobserver agreement. It has been shown that phenotypic characteristics of individuals with GCMN can help predict risk for developing NCM, melanoma, or other complications. The risk for developing melanoma correlates with the size of the CMN,12 and the number of satellite nevi has been shown to correlate with risk for developing NCM.13 The phenotype may also provide insight regarding the timing of events during embryogenesis. For example, ‘‘kissing nevi’’ on the eyelids (a CMN with one portion located on the upper eyelid and the other on the lower eyelid) indicate that the aberrant melanocyte migration took place between the 9th week of embryogenesis, when the upper and lower lids fuse, and the 20th week, when the lids separate.14,15 The location of the GCMN may also be a factor or a surrogate marker of risk. It has been suggested that the bathing trunk pattern may be associated with an increased risk for melanoma and NCM.9 The bathing trunk distribution encountered in 45.5% of our patients (n = 10) and in 39.8% (n = 45) of cases reported in the literature, compares with previously reported figures of 44% (n = 15)8 and 47.3% (n = 26).9 This pattern was observed in 33 of 39 published patients with melanoma and in 12 of 20 patients with NCM.9 It is noteworthy that our 2 patients who died from melanoma presented with a bathing trunk GCMN. Although the descriptor ‘‘bathing trunk’’ is commonly used, no clear or agreed upon definitions exist for this or other garment-related terms. Future studies incorporating molecular information together with genetic, environmental, and other factors related to embryogenesis may someday help explain why GCMN are distributed in these patterns. It is likely that a combination of factors including the type of mutation, the location of the progenitor
692 Martins da Silva et al
J AM ACAD DERMATOL
Fig 1. The criteria for congenital melanocytic nevi characterization.7 CMN, Congenital melanocytic nevi.
Fig 2. The 6B classification for giant congenital melanocytic nevi: definitions and clinical examples of the 6 patterns of distribution. *Gray area represents the extension variation that the nevi may present. **Satellite lesionseeven the large oneseshould be disregarded when classifying the body pattern.
APRIL 2017
J AM ACAD DERMATOL
Martins da Silva et al 693
VOLUME 76, NUMBER 4
Fig 3. Flow chart of the case selection process used in phase II. CMN, Congenital melanocytic nevi; GCMN, giant congenital melanocytic nevi.
Table I. Kappa value for each category in all the phases Pattern
Bolero Back Bathing trunk Breast/belly Body extremity Body
Phase I initial
Phase I final
Phase II
0.593 1 0.876 1 1 0.345
1 1 1 1 1 1
0.888 0.825 0.915 0.912 0.962 0.722
mutated cell, the timing of the mutation being expressed during embryogenesis, and the presence of polymorphisms all play a role in sculpting the
morphology and distribution patterns encountered in GCMN. Our study has limitations. GCMN are rare and some of the patterns described in the 6B scheme are seen infrequently. This may explain why some patterns had lower interobserver agreements. Multicenter studies with a larger number of cases are required to validate our findings. The 6B classification scheme for GCMN excluded nevi located on the head because very few nevi in this location were found to be larger than 40 cm. However, a classification scheme incorporating nevi located on the head should be considered. In conclusion, our study found that most GCMN can be categorized into 1 of 6 distinct anatomic
J AM ACAD DERMATOL
694 Martins da Silva et al
distribution patterns with good interobserver and intraobserver agreement. Based on our study we propose the standardization of terminology used to describe GCMN garment distribution patterns. We favor addition of the 6B classification scheme to the recently reported recommendations for the categorization of cutaneous features of CMN to facilitate collaborative research and allow for more reliable comparisons of study findings. We are grateful to our patients and relatives, who are always committed to collaborate with our studies; the dermatologists Alejandra Villarreal, Marisa Zimmermann, Luz Casta~ neda, Luz Calderon, Caroline Takigami, Sebastian Mercau, and Adriana Pe~ na y Gimena Bolomo for reviewing the cases; and the nurses Daniel Gabriel, Pablo Iglesias, and M. Eugenia Moliner, and technicians Abel Ca~ no, Mireia Domınguez, and Beatriz Alejo for their help.
APRIL 2017
5.
6.
7.
8. 9.
10.
11. REFERENCES 1. Viana ACL, Gontijo B, Bittencourt FV. Giant congenital melanocytic nevus. An Bras Dermatol. 2013;88(6):863-878. 2. Alikhan A, Ibrahimi OA, Eisen DB. Congenital melanocytic nevi: where are we now? Part I. Clinical presentation, epidemiology, pathogenesis, histology, malignant transformation, and neurocutaneous melanosis. J Am Acad Dermatol. 2012;67(4): 495.e1-495.e17. quiz 512-514. 3. Monteagudo B, Labandeira J, Le on-Mui~ nos E, et al. Prevalence of birthmarks and transient skin lesions in 1,000 Spanish newborns [in Spanish]. Actas Dermo-Sifiliogr aficas. 2011;102(4): 264-269. 4. Palau-Lazaro MC, Buendıa-Eisman A, Serrano-Ortega S. Prevalence of congenital nevus in 1000 live births in
12.
13.
14.
15.
Granada, Spain [in Spanish]. Actas Dermo-Sifiliogr aficas. 2008; 99(1):81. Aguilera P, Puig S, Guilabert A, et al. Prevalence study of nevi in children from Barcelona. Dermoscopy, constitutional and environmental factors. Dermatol Basel Switz. 2009;218(3):203-214. Lovett A, Maari C, Decarie J-C, et al. Large congenital melanocytic nevi and neurocutaneous melanocytosis: one pediatric center’s experience. J Am Acad Dermatol. 2009; 61(5):766-774. Krengel S, Scope A, Dusza SW, Vonthein R, Marghoob AA. New recommendations for the categorization of cutaneous features of congenital melanocytic nevi. J Am Acad Dermatol. 2013;68(3):441-451. Bircher MO. Zur Aetiologie des Nevus pilosus pigmentosus congenitus extensus. Arch F€ ur Dermatol Syph. 1897;41:195-225. Reed WB, Becker SW Sr, Becker SW Jr, Nickel WR. Giant pigmented nevi, melanoma and leptomeningeal melanocytosis: a clinical and histopathological study. Arch Dermatol. 1965;91:100-119. Torrelo A, Baselga E, Nagore E, Zambrano A, Happle R. Delineation of the various shapes and patterns of nevi. Eur J Dermatol EJD. 2005;15(6):439-450. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-174. Krengel S, Hauschild A, Sch€afer T. Melanoma risk in congenital melanocytic nevi: a systematic review. Br J Dermatol. 2006; 155(1):1-8. Marghoob AA, Dusza S, Oliveria S, Halpern AC. Number of satellite nevi as a correlate for neurocutaneous melanocytosis in patients with large congenital melanocytic nevi. Arch Dermatol. 2004;140(2):171-175. Hamming N. Anatomy and embryology of the eyelids: a review with special reference to the development of divided nevi. Pediatr Dermatol. 1983;1(1):51-58. Desai SC, Walen S, Holds JB, Branham G. Divided nevus of the eyelid: review of embryology, pathology and treatment. Am J Otolaryngol. 2013;34(3):223-229.