I
I
Pigmented Spitz nevi: Improvement of the diagnostic accuracy by epiluminescence microscopy Andreas Steiner, MD, Hubert Pehamberger, MD, Michael Binder, MD, and Klaus Wolff, MD Vienna, Austria
Background: Pigmented Spitz nevi have distinct clinical features but often may be difficult to differentiate from malignant melanoma by clinical criteria. Objective." The purpose of this study was to use a new noninvasive diagnostic approach to improve the clinical diagnosis of Spitz nevi. Methods: Epiluminescence microscopy (ELM) is a new, noninvasive technique for which criteria for the diagnosis of melanocytic tumors, benign and malignant, rec~.nt/yhave been established. These criteria were tested in an investigation of 54 pigmented Spitz nevi. Results: With ELM the accuracyof clinical diagnosis of pigmented Spitz nevi improved from 56% (clinical) to 93% (ELM). Conclusion." Our findings suggest that ELM criteria are useful to improve the accuracy of clinical diagnosis of Spitz nevi. (J AM ACADDERMATOL1992;27:697-701.) Pigmented Spitz nevi have distinct clinical features. T h e y usually are darkly and uniformly pigmented papules with a regular outline and a uniform appearance, l However, occasionally Spitz nevi may be larger than common nevi; may have an irregular polypoid or pedunculated outline; a scaly, crusted or eroded surface; and a variegated color. In addition, itching, tenderness, and bleeding may occur.2 Because these features are also characteristic of malignant melanoma, pigmented Spitz nevi may often be difficult to differentiate from malignant melanoma by clinical criteria alone. Epiluminescence microscopy* (ELM) is a noninvasive technique employed in the clinical diagnosis of early pigmented skin lesions. 3-9 With the ELM technique pigmented Spitz nevi exhibit a characteristic appearance. 34 We examined 54 pigmented Spitz nevi by ELM. These investigations made it possible to establish additional E L M criteria for the
*The term epiluminescence microscopy 3 derives from the epiluminescence projector, in which light shines on an object and the reiteeted image is projected. From the Department of Dermatology I, University of Vienna, Medical School. Accepted for publication April 20, 1992. Reprint requests: Hubert Pehamberger, MD, Department of Dermatology I, University of Vienna, Alser Strasse 4, A-1090 Vienna,Austria. 16/1/38749
diagnosis of Spitz nevi, which have improved diagnostic accuracy, particularly, in their differentiation from malignant melanoma. MATERIAL AND METHODS ELM was performed with a Wild M650 (Wild Heerbrugg AG, Heerbrugg, Switzerland) binocular surface microscope equipped with objectives of 91 mm working distance. Magnifications obtainable are • • • • and • All pigmented skin lesions were first examined for surface structure. They were then covered with immersion oil and a glass slide that was applied with slight pressure; this renders the epidermis translucent and allows study of the dermoepidermal junction zone. Photographs of the pigmented skin lesions were taken with an Olympus CM10 automatic camera, mounted on a side arm of the microscope. Lesions were first diagnosed clinically; consensus of at least two'of three investigators (A. S., H. P., and K. W.) was required for the clinical diagnosis to be considered established. Lesions were then examined with ELM and a diagnosis was made by means of the criteria of pattern analysis.a, 4 Again agreement of at least two of the three investigators was required. Subsequently lesions were excised and were subjected to histopathologic examination to verify the ELM diagnosis and to correlate features found by ELM. RESULTS
Pattern analysis of Spitz nevi For pattern analysis the terminology described previously3, 4 will be employed. When viewed with
697
698
Journal of the American Academy of Dermatology
Steiner et al.
9
L
,
,i?
Fig. 1. Oval flat homogenous pigmented Spitz nevus. Fig. 2. Oil immersion of Spitz nevus in Fig. 1 shows coffee bean-like appearance. Depigmentation in center is discrete. At periphery there is a typical rim of large brown globules (arrows). Outline is regular without radial streaming and pseudopods. 9Fig. 3. Pigmented Spitz nevus with macular and papular components. This lesion is much lighter than that shown in Fig. 1. Fig. 4. Oil immersion of Spitz nevus in Fig. 3 reveals bizarre retiform depigmentation in center, which dissolves into negative pigment network towards periphery (large arrow). Outline is regular without radial streaming or pseudopods; although there are black dots, they do not involve margin of lesion (small arrow). ELM, Spitz nevi appear uniform, regular, and round or oval (Figs. 1-6). They are relatively uniformly brown to black, except in areas of hypopigmentation. The margins of the lesions are well defined and regular. When the oil immersion technique is used, the uniformity is replaced by a subtle picture of variably pigmented areas. M a n y Spitz nevi (Table I) exhibit a characteristic coffee bean-like appearance (Fig. 2). Other lesions have a radiating form, looking like a starburst, that is, the pigment spreads regularly from the center to the margin (Figs. 2 and 6). The pigment network, when present, is prominent, regular and stops abruptly at the well-defined border of
the lesion. This pigment network m a y appear inverted in a few lesions in that it corresponds to a negative of the pigmented network seen in other pigmented skin lesions (negative pigment network) (Fig. 4). A characteristic finding of Spitz nevi is a central, retiform, bizarre depigrnentation (Fig. 4). Brown globules of different sizes as well as black dots in regular distribution may be found throughout the lesions. A rim of large brown globules at the periphery is a characteristic finding (Figs. 2 and 6). The margin of the lesion is well defined and regular without pseudopods or radial streaming; this is in contrast to the E L M appearance of small melanomas 3, 4 (Figs. 7 and 8).
Volume 27
Number 5, Part 1 November 1992
Epiluminescence microscopy: Spitz nevi 699
Fig. 5. Small pigmented skin lesion. Clinically it could be melanoma or Spitz nevus. Fig. 6. Oil immersion of lesion in Fig. 5 reveals faint bizarre depigmentation in center. Pigmentation is monomorphous. It spreads regularly from center tomargin and gives the lesion a radiated appearance (starburst lesion). There is a rim of large brown globules at the periphery (arrows), which by superimposition {small arrow) mimic, but by changing the focus clearly are not, pseudopods.3 Pseudopods are a sign of melanoma. Outline is regular without pseudopods, as stated, and without radial streaming. The lesion therefore is a Spitz nevus.
Table I. Epiluminescence microscopy appearance of pigmented Spitz nevus and melanoma Spitz nevi
General appearance Pigment network
Brown globules
Black dots Depigmentation Border
Melanoma
Monomorphous, starburst lesion, coffee bean-like appearance Prominent, regular, stops abruptly at periphery or thins out, "negative" pigment network Different size, regular throughout the lesion, rim of large brown globules at periphery around lesion Center or throughout lesion, regular distribution Bizarre, retiform in center No pseudopods, no radial streaming
Diagnostic features of malignant melanoma Melanomas (Table I) are polymorphous and irregularly pigmented. The pigment network is prominent and irregular, thins out at the periphery, or stops abruptly at the border of the lesion. Black dots may often be found only at the periphery and occur in irregular distribution. Depigmentation, when present, exhibits an irregular outline and is often located at the periphery of the lesions. Pseudo-
Polymorphous, multiple pattern Prominent, irregular, stops abruptly at periphery or thins out, no "negative" pigment network Different size, haphazardly spaced, no rim Often only at periphery, irregular distribution Irregular, often in periphery Often pseudopods, often radial streaming
pods and radial streaming (Fig. 8 ) are the most characteristic features of the margins of a melanoma.
ELM criteria and diagnostic accuracy The present study involved 54 histologically proven pigmented Spitz nevi. Thirty bad been clinically diagnosed as pigmented Spitz nevi, 12 as dysplastic nevi, 7 as compound nevi, and 5 as nodular melanomas (Table II). The diagnostic accuracy by
700
Journal of the American Academy of Dermatology
Steiner et al.
-
. ~
7 .
.
8 .
•
.
J
Fig. 7. Small pigmented skin lesion. Clinically it could be a melanoma or a pigmented Spitz nevus.
Fig. 8. Oil immersion of lesion in Fig. 7 reveals nonhomogeneous pigmentation with grayblue discoloration. Outline is irregular and radial streaming (arrow) is present. The lesion therefore is a melanoma, histologically classified as a superficial spreading melanoma (Clark level HI, Breslow 1.1 ram).
Table lI. Comparison of the clinical and E L M diagnoses of 54 histopathologically verified pigmented Spitz nevi Clinical diagnosis No. of lesions
Correct
1
ELM diagnosis
Incorrect
Correct
54*
30 56%
]
Incorrect
l
m
24]"
50 93%
45;
*All lesions were verified histopathologlcally and were pigmented Spitz nevi. tThe 24 incorrectly diagnosed lesions were histopathologicallymelanoma (5), dysplastic nevi (12), and compound nevi (7). ~The four incorrectly diagnosed lesions were histopatholog[eallymelanoma (2) and dysplastic nevi (2).
pure criteria was thus only 56%. In contrast, when E L M is used to examine the same lesions, 50 were diagnosed as pigmented Spitz nevi, 2 as dysplastic nevi, and 2 as melanomas (Table II). Thus the diagnostic accuracy was 93%. DISCUSSION W e 3, 4 and others s-9 have previously reported that E L M definitely improves the diagnostic accuracy of pigmented skin lesions. In the present study this is shown in more detail for pigmented Spitz nevi. Criteria have now been established but still have to be tested more extensively. This is important for small pigmented lesions that cannot be diagnosed by clinical criteria alone and in particular for the differentiation of pigmented Spitz nevi from malignant melanoma, dysplastic nevi, or common nevi. By applying E L M pattern analysis to Spitz nevi, we were able to increase the diagnostic accuracy from 56% to 93% (Table II). This is a distinct improvement, particularly because five of these Spitz nevi had clini-
cally been considered to be malignant melanoma. E L M correctly diagnosed three of these five lesions as Spitz nevi and in only two was the correct diagnosis missed. Both lesions revealed the E L M pigment pattern of malignant melanoma but histopathologically proved to be Spitz nevi. E L M is a new research technique that cannot and is not meant to claim absolute diagnostic reliability. Therefore it should not and cannot replace histopathologic examination for verification of the clinical diagnosis. However, it does help prevent misdiagnosis of questionable lesions of melanoma and may thus prevent unnecessarily radical surgery. REFERENCES 1. Rhodes AR. Neoplasmas: benign neoplasmas, hyperplasias and dysplasias of melanoeytes. In: Fitzpatrick TB, Eisen AZ, Wolff K, et al, eds. Dermatology in general medicine. 3rd ed. New York: McGraw-Hill, 1987:872-946. 2. Kopf AW, Andrade R. Benign juvenile melanoma. In: Kopf A, Andrade R, eds. Yearbook of dermatology, 19651966. Chicago: Year Book, 1966:7.
Volume 27 Number 5, Part 1 November 1992 3. Pehamberger H, Steiner A, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. I. Pattern analysis of pigmented skin lesions. J AM ACADDERMATOL 1987;17:571-83. 4. Steiner A, Pehamberger H, Wolff K. In vivo epiluminescence microscopyof pigmented skin lesions. II. Diagnosisof small pigmented skin lesions and early detection of malignant melanoma. J AM ACADDERMATOL1987;17:584-91. 5. Fritsch P, Pechlaner R. Differentiation of benign from malignant melanocytic lesions using incident light microscopy. In: Ackermann AB. Pathology of malignant melanoma. New York: Masson, 1981:301-12.
Epiluminescence microscopy: Spitz nevi 6. MacKie R. An aid to the preoperative assessment of pigmented lesionsof the skin. Br J Dermatol 1971;85:232-8. 7. Bahmer FA, Rohrer C. Rapid and simple macrophotography of the skin. Br J Dermatol 1986;114:135-6. 8. Soyer HP, Smolle J, Hoedl S, et al. Surface microscopy: a new approach to the diagnosis of cutaneous pigmented tumors. Am J Dermatopathol 1989;11:1-10. 9. Stolz W, Bilek P, Landthaler M, et al. Skin surface microscopy. Lancet 1989;2:864-5.
Atypical melanocytic nevi Clinical and histopathologic predictors o f residual tumor at reexcision Lisa M. Cohen, MD, Steven J. Hodge, MD, Lafayette G. Owen, MD, and Jeffrey P. Callen, M D Louisville, Kentucky
Background:The appropriate method for surgical management of melanocytic lesions with disordered architecture and melanocytic atypia (formerly dysplastic nevi) has been controversial. Physicians often reexcise these lesions after primary removal because of their potential relation to malignant melanoma. The outcomes of these reexcisions and the original biopsy specimens have not been previously examined. Objective:The purpose of this study was to examine re,excision specimens and their respective original specimens to determine whether there were any characteristics predictive of the presence of residual nevus cells (RNCs) on reexcision. Methods:One hundred eighty-nine reexcision specimens of atypical melanocytic lesions were evaluated for this study. The original specimens were examined for specific histopathologic features without knowledge of the findings on reexcision. Clinical characteristics were also examined. Results. Of the 189 reexcision specimens, 47 (24.9%) contained RNCs. The proportion of specimens with RNC on reexcision was significantly greater if the original lesion was removed by punch biopsy rather than by shave or elliptical excision (38.3% vs 22.0% vs 10.5%, respectively; p < 0.03). Lesions located on the chest had a higher likelihood of RNCs than those on the back or leg (52.2% vs 21.7%; p = 0.009; 52.2% vs 9.7%, p < 0.002, respectively). Mean age was greater in those with RNCs at reexcision than those without (43.6 vs 37.9 years, respectively; p < 0.0001). The proportion of specimens with RNCs at reexcision was greater in those that had both lateral margins involved than in those that had nevus cells in either one or neither of the lateral margins in the original biopsy specimens (39.7% vs 24.0% vs 7.8%, respectively; p = 0.0005). One of the 189 reexcision specimens (0.5%) contained melanoma, although the original histopathologic diagnosis was an atypical melanocytic nevus. Conclusion."We identified several clinical and histopathologic factors that are strongly associated with the presence of RNCs on reexcision. (J AM ACAD DERMATOL 1992;27:701-6.) In 1978, three simultaneous reports of dysplastic melanocytic nevi in melanoma-prone families emerged, referred to as the B-K mole syndrome or From the Divisionof Dermatology,Universityof Louisville. Acceptedfor publicationMay 11, 1992. Reprint requests:JeffreyP. Callen, MD, 310 E. Broadway,Suite200, Louisville,KY 40202. 16/1/39167
familial atypical multiple mole-melanoma ( F A M M M ) syndrome, t3 Subsequently, sporadic cases of dysplastic nevi associated with malignant melanoma ( M M ) were reported, and the t e r m dysplastic nevus syndrome ( D N S ) was applied. 4 This syndrome was classified into four types, namely, nonfamilial (sporadic) and familial with or without M M , in order of presumed increasing melanoma risk. s T h e lifetime risk of developing M M in the 701
I
Pigmented Spitz nevi: Improvement of the diagnostic accuracy by epiluminescence microscopy Andreas Steiner, MD, Hubert Pehamberger, MD, Michael Binder, MD, and Klaus Wolff, MD Vienna, Austria
Background: Pigmented Spitz nevi have distinct clinical features but often may be difficult to differentiate from malignant melanoma by clinical criteria. Objective." The purpose of this study was to use a new noninvasive diagnostic approach to improve the clinical diagnosis of Spitz nevi. Methods: Epiluminescence microscopy (ELM) is a new, noninvasive technique for which criteria for the diagnosis of melanocytic tumors, benign and malignant, rec~.nt/yhave been established. These criteria were tested in an investigation of 54 pigmented Spitz nevi. Results: With ELM the accuracyof clinical diagnosis of pigmented Spitz nevi improved from 56% (clinical) to 93% (ELM). Conclusion." Our findings suggest that ELM criteria are useful to improve the accuracy of clinical diagnosis of Spitz nevi. (J AM ACADDERMATOL1992;27:697-701.) Pigmented Spitz nevi have distinct clinical features. T h e y usually are darkly and uniformly pigmented papules with a regular outline and a uniform appearance, l However, occasionally Spitz nevi may be larger than common nevi; may have an irregular polypoid or pedunculated outline; a scaly, crusted or eroded surface; and a variegated color. In addition, itching, tenderness, and bleeding may occur.2 Because these features are also characteristic of malignant melanoma, pigmented Spitz nevi may often be difficult to differentiate from malignant melanoma by clinical criteria alone. Epiluminescence microscopy* (ELM) is a noninvasive technique employed in the clinical diagnosis of early pigmented skin lesions. 3-9 With the ELM technique pigmented Spitz nevi exhibit a characteristic appearance. 34 We examined 54 pigmented Spitz nevi by ELM. These investigations made it possible to establish additional E L M criteria for the
*The term epiluminescence microscopy 3 derives from the epiluminescence projector, in which light shines on an object and the reiteeted image is projected. From the Department of Dermatology I, University of Vienna, Medical School. Accepted for publication April 20, 1992. Reprint requests: Hubert Pehamberger, MD, Department of Dermatology I, University of Vienna, Alser Strasse 4, A-1090 Vienna,Austria. 16/1/38749
diagnosis of Spitz nevi, which have improved diagnostic accuracy, particularly, in their differentiation from malignant melanoma. MATERIAL AND METHODS ELM was performed with a Wild M650 (Wild Heerbrugg AG, Heerbrugg, Switzerland) binocular surface microscope equipped with objectives of 91 mm working distance. Magnifications obtainable are • • • • and • All pigmented skin lesions were first examined for surface structure. They were then covered with immersion oil and a glass slide that was applied with slight pressure; this renders the epidermis translucent and allows study of the dermoepidermal junction zone. Photographs of the pigmented skin lesions were taken with an Olympus CM10 automatic camera, mounted on a side arm of the microscope. Lesions were first diagnosed clinically; consensus of at least two'of three investigators (A. S., H. P., and K. W.) was required for the clinical diagnosis to be considered established. Lesions were then examined with ELM and a diagnosis was made by means of the criteria of pattern analysis.a, 4 Again agreement of at least two of the three investigators was required. Subsequently lesions were excised and were subjected to histopathologic examination to verify the ELM diagnosis and to correlate features found by ELM. RESULTS
Pattern analysis of Spitz nevi For pattern analysis the terminology described previously3, 4 will be employed. When viewed with
697
698
Journal of the American Academy of Dermatology
Steiner et al.
9
L
,
,i?
Fig. 1. Oval flat homogenous pigmented Spitz nevus. Fig. 2. Oil immersion of Spitz nevus in Fig. 1 shows coffee bean-like appearance. Depigmentation in center is discrete. At periphery there is a typical rim of large brown globules (arrows). Outline is regular without radial streaming and pseudopods. 9Fig. 3. Pigmented Spitz nevus with macular and papular components. This lesion is much lighter than that shown in Fig. 1. Fig. 4. Oil immersion of Spitz nevus in Fig. 3 reveals bizarre retiform depigmentation in center, which dissolves into negative pigment network towards periphery (large arrow). Outline is regular without radial streaming or pseudopods; although there are black dots, they do not involve margin of lesion (small arrow). ELM, Spitz nevi appear uniform, regular, and round or oval (Figs. 1-6). They are relatively uniformly brown to black, except in areas of hypopigmentation. The margins of the lesions are well defined and regular. When the oil immersion technique is used, the uniformity is replaced by a subtle picture of variably pigmented areas. M a n y Spitz nevi (Table I) exhibit a characteristic coffee bean-like appearance (Fig. 2). Other lesions have a radiating form, looking like a starburst, that is, the pigment spreads regularly from the center to the margin (Figs. 2 and 6). The pigment network, when present, is prominent, regular and stops abruptly at the well-defined border of
the lesion. This pigment network m a y appear inverted in a few lesions in that it corresponds to a negative of the pigmented network seen in other pigmented skin lesions (negative pigment network) (Fig. 4). A characteristic finding of Spitz nevi is a central, retiform, bizarre depigrnentation (Fig. 4). Brown globules of different sizes as well as black dots in regular distribution may be found throughout the lesions. A rim of large brown globules at the periphery is a characteristic finding (Figs. 2 and 6). The margin of the lesion is well defined and regular without pseudopods or radial streaming; this is in contrast to the E L M appearance of small melanomas 3, 4 (Figs. 7 and 8).
Volume 27
Number 5, Part 1 November 1992
Epiluminescence microscopy: Spitz nevi 699
Fig. 5. Small pigmented skin lesion. Clinically it could be melanoma or Spitz nevus. Fig. 6. Oil immersion of lesion in Fig. 5 reveals faint bizarre depigmentation in center. Pigmentation is monomorphous. It spreads regularly from center tomargin and gives the lesion a radiated appearance (starburst lesion). There is a rim of large brown globules at the periphery (arrows), which by superimposition {small arrow) mimic, but by changing the focus clearly are not, pseudopods.3 Pseudopods are a sign of melanoma. Outline is regular without pseudopods, as stated, and without radial streaming. The lesion therefore is a Spitz nevus.
Table I. Epiluminescence microscopy appearance of pigmented Spitz nevus and melanoma Spitz nevi
General appearance Pigment network
Brown globules
Black dots Depigmentation Border
Melanoma
Monomorphous, starburst lesion, coffee bean-like appearance Prominent, regular, stops abruptly at periphery or thins out, "negative" pigment network Different size, regular throughout the lesion, rim of large brown globules at periphery around lesion Center or throughout lesion, regular distribution Bizarre, retiform in center No pseudopods, no radial streaming
Diagnostic features of malignant melanoma Melanomas (Table I) are polymorphous and irregularly pigmented. The pigment network is prominent and irregular, thins out at the periphery, or stops abruptly at the border of the lesion. Black dots may often be found only at the periphery and occur in irregular distribution. Depigmentation, when present, exhibits an irregular outline and is often located at the periphery of the lesions. Pseudo-
Polymorphous, multiple pattern Prominent, irregular, stops abruptly at periphery or thins out, no "negative" pigment network Different size, haphazardly spaced, no rim Often only at periphery, irregular distribution Irregular, often in periphery Often pseudopods, often radial streaming
pods and radial streaming (Fig. 8 ) are the most characteristic features of the margins of a melanoma.
ELM criteria and diagnostic accuracy The present study involved 54 histologically proven pigmented Spitz nevi. Thirty bad been clinically diagnosed as pigmented Spitz nevi, 12 as dysplastic nevi, 7 as compound nevi, and 5 as nodular melanomas (Table II). The diagnostic accuracy by
700
Journal of the American Academy of Dermatology
Steiner et al.
-
. ~
7 .
.
8 .
•
.
J
Fig. 7. Small pigmented skin lesion. Clinically it could be a melanoma or a pigmented Spitz nevus.
Fig. 8. Oil immersion of lesion in Fig. 7 reveals nonhomogeneous pigmentation with grayblue discoloration. Outline is irregular and radial streaming (arrow) is present. The lesion therefore is a melanoma, histologically classified as a superficial spreading melanoma (Clark level HI, Breslow 1.1 ram).
Table lI. Comparison of the clinical and E L M diagnoses of 54 histopathologically verified pigmented Spitz nevi Clinical diagnosis No. of lesions
Correct
1
ELM diagnosis
Incorrect
Correct
54*
30 56%
]
Incorrect
l
m
24]"
50 93%
45;
*All lesions were verified histopathologlcally and were pigmented Spitz nevi. tThe 24 incorrectly diagnosed lesions were histopathologicallymelanoma (5), dysplastic nevi (12), and compound nevi (7). ~The four incorrectly diagnosed lesions were histopatholog[eallymelanoma (2) and dysplastic nevi (2).
pure criteria was thus only 56%. In contrast, when E L M is used to examine the same lesions, 50 were diagnosed as pigmented Spitz nevi, 2 as dysplastic nevi, and 2 as melanomas (Table II). Thus the diagnostic accuracy was 93%. DISCUSSION W e 3, 4 and others s-9 have previously reported that E L M definitely improves the diagnostic accuracy of pigmented skin lesions. In the present study this is shown in more detail for pigmented Spitz nevi. Criteria have now been established but still have to be tested more extensively. This is important for small pigmented lesions that cannot be diagnosed by clinical criteria alone and in particular for the differentiation of pigmented Spitz nevi from malignant melanoma, dysplastic nevi, or common nevi. By applying E L M pattern analysis to Spitz nevi, we were able to increase the diagnostic accuracy from 56% to 93% (Table II). This is a distinct improvement, particularly because five of these Spitz nevi had clini-
cally been considered to be malignant melanoma. E L M correctly diagnosed three of these five lesions as Spitz nevi and in only two was the correct diagnosis missed. Both lesions revealed the E L M pigment pattern of malignant melanoma but histopathologically proved to be Spitz nevi. E L M is a new research technique that cannot and is not meant to claim absolute diagnostic reliability. Therefore it should not and cannot replace histopathologic examination for verification of the clinical diagnosis. However, it does help prevent misdiagnosis of questionable lesions of melanoma and may thus prevent unnecessarily radical surgery. REFERENCES 1. Rhodes AR. Neoplasmas: benign neoplasmas, hyperplasias and dysplasias of melanoeytes. In: Fitzpatrick TB, Eisen AZ, Wolff K, et al, eds. Dermatology in general medicine. 3rd ed. New York: McGraw-Hill, 1987:872-946. 2. Kopf AW, Andrade R. Benign juvenile melanoma. In: Kopf A, Andrade R, eds. Yearbook of dermatology, 19651966. Chicago: Year Book, 1966:7.
Volume 27 Number 5, Part 1 November 1992 3. Pehamberger H, Steiner A, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. I. Pattern analysis of pigmented skin lesions. J AM ACADDERMATOL 1987;17:571-83. 4. Steiner A, Pehamberger H, Wolff K. In vivo epiluminescence microscopyof pigmented skin lesions. II. Diagnosisof small pigmented skin lesions and early detection of malignant melanoma. J AM ACADDERMATOL1987;17:584-91. 5. Fritsch P, Pechlaner R. Differentiation of benign from malignant melanocytic lesions using incident light microscopy. In: Ackermann AB. Pathology of malignant melanoma. New York: Masson, 1981:301-12.
Epiluminescence microscopy: Spitz nevi 6. MacKie R. An aid to the preoperative assessment of pigmented lesionsof the skin. Br J Dermatol 1971;85:232-8. 7. Bahmer FA, Rohrer C. Rapid and simple macrophotography of the skin. Br J Dermatol 1986;114:135-6. 8. Soyer HP, Smolle J, Hoedl S, et al. Surface microscopy: a new approach to the diagnosis of cutaneous pigmented tumors. Am J Dermatopathol 1989;11:1-10. 9. Stolz W, Bilek P, Landthaler M, et al. Skin surface microscopy. Lancet 1989;2:864-5.
Atypical melanocytic nevi Clinical and histopathologic predictors o f residual tumor at reexcision Lisa M. Cohen, MD, Steven J. Hodge, MD, Lafayette G. Owen, MD, and Jeffrey P. Callen, M D Louisville, Kentucky
Background:The appropriate method for surgical management of melanocytic lesions with disordered architecture and melanocytic atypia (formerly dysplastic nevi) has been controversial. Physicians often reexcise these lesions after primary removal because of their potential relation to malignant melanoma. The outcomes of these reexcisions and the original biopsy specimens have not been previously examined. Objective:The purpose of this study was to examine re,excision specimens and their respective original specimens to determine whether there were any characteristics predictive of the presence of residual nevus cells (RNCs) on reexcision. Methods:One hundred eighty-nine reexcision specimens of atypical melanocytic lesions were evaluated for this study. The original specimens were examined for specific histopathologic features without knowledge of the findings on reexcision. Clinical characteristics were also examined. Results. Of the 189 reexcision specimens, 47 (24.9%) contained RNCs. The proportion of specimens with RNC on reexcision was significantly greater if the original lesion was removed by punch biopsy rather than by shave or elliptical excision (38.3% vs 22.0% vs 10.5%, respectively; p < 0.03). Lesions located on the chest had a higher likelihood of RNCs than those on the back or leg (52.2% vs 21.7%; p = 0.009; 52.2% vs 9.7%, p < 0.002, respectively). Mean age was greater in those with RNCs at reexcision than those without (43.6 vs 37.9 years, respectively; p < 0.0001). The proportion of specimens with RNCs at reexcision was greater in those that had both lateral margins involved than in those that had nevus cells in either one or neither of the lateral margins in the original biopsy specimens (39.7% vs 24.0% vs 7.8%, respectively; p = 0.0005). One of the 189 reexcision specimens (0.5%) contained melanoma, although the original histopathologic diagnosis was an atypical melanocytic nevus. Conclusion."We identified several clinical and histopathologic factors that are strongly associated with the presence of RNCs on reexcision. (J AM ACAD DERMATOL 1992;27:701-6.) In 1978, three simultaneous reports of dysplastic melanocytic nevi in melanoma-prone families emerged, referred to as the B-K mole syndrome or From the Divisionof Dermatology,Universityof Louisville. Acceptedfor publicationMay 11, 1992. Reprint requests:JeffreyP. Callen, MD, 310 E. Broadway,Suite200, Louisville,KY 40202. 16/1/39167
familial atypical multiple mole-melanoma ( F A M M M ) syndrome, t3 Subsequently, sporadic cases of dysplastic nevi associated with malignant melanoma ( M M ) were reported, and the t e r m dysplastic nevus syndrome ( D N S ) was applied. 4 This syndrome was classified into four types, namely, nonfamilial (sporadic) and familial with or without M M , in order of presumed increasing melanoma risk. s T h e lifetime risk of developing M M in the 701
