Figure. Numbers of Spitz nevi vs melanomas diagnosed by respondents in pediatric patients during a 5-year period.
Tlougan BE, Orlow SJ, Schaffer JV. Spitz nevi: beliefs, behaviors, and experiences of pediatric dermatologists. JAMA Dermatol.. 2013;149(3). doi:10.1001/jamadermatol.2013.1124..
Tlougan BE, Orlow SJ, Schaffer JV. Spitz NeviBeliefs, Behaviors, and Experiences of Pediatric Dermatologists. JAMA Dermatol. 2013;149(3):283-291. doi:10.1001/jamadermatol.2013.1124
Author Affiliations: The Ronald O. Perelman Department of Dermatology and Department of Pediatrics, New York University School of Medicine, New York, New York.
Importance Controversy exists regarding strategies for diagnosis and management of Spitz nevi, a type of melanocytic neoplasm that most often develops in children.
Objective To determine the beliefs, behaviors, and experiences of pediatric dermatologists with regard to Spitz nevi.
Design Anonymous web-based survey.
Setting Private and academic dermatology practices.
Participants Respondents included 175 pediatric dermatologists from the United States and around the world, representing a 51.1% response rate (175 of 342). Analyses were limited to the 144 respondents whose practices included at least 50% children (younger than 18 years).
Main Outcome Measures Assessment of the following with regard to Spitz nevi: frequency of diagnosis, general beliefs, techniques used for evaluation (eg, dermoscopy and biopsy), management strategies, and observed outcomes.
Results Collectively, respondents had seen approximately 20 000 Spitz nevi; 67.6% (96 of 142) had diagnosed at least 6 Spitz nevi yearly, whereas 90.1% (128 of 142) had diagnosed no more than 2 prepubertal melanomas in the past 5 years. Ninety-six percent of respondents (95.8%; 136 of 142) categorized typical Spitz nevi as benign. Eighty percent of respondents (79.6%; 113 of 142) used dermatoscopy, and 96.5% (137 of 142) avoided partial biopsies of Spitz nevi. In children with a suspected Spitz nevus, clinical follow-up was chosen by 49.3% (69 of 140) of respondents for a small, stable nonpigmented lesion and by 29.7% (41 of 138) for a pigmented lesion with a typical starburst pattern seen via dermatoscopy. Predictors of clinical follow-up of the latter lesion included believing that Spitz nevi are not melanoma precursors (P = .04). Forty-seven percent (62 of 132) of respondents had observed involution of Spitz nevi. No deaths had resulted from the approximately 10 000 Spitz nevi or atypical spitzoid neoplasms seen by the 91 respondents with academic or hospital-based practices.
Conclusions and Relevance The results of our survey support conservative management of Spitz nevi in children, with clinical follow-up representing an option for typical lesions. This represents an important difference from strategies used for management of these lesions in adults.
A Spitz (spindle and epithelioid cell) nevus is a distinct type of melanocytic neoplasm that most often develops in children. Several large series have documented that 50% to 75% of patients with lesions that are diagnosed histopathologically as Spitz nevi are younger than 20 years.1- 5Quiz Ref IDA Spitz nevus usually presents as a solitary pink, red, or brown papule on the face or a limb. Initial growth tends to be rapid, which can be alarming to parents and physicians.6 Although they are benign neoplasms, Spitz nevi sometimes have histopathologic features that overlap with those of melanoma. This has led to controversies with regard to strategies for diagnosis (eg, when and how to perform biopsies) and management of Spitz nevi.
A survey conducted in 2001 on the management of Spitz nevi by dermatologists in the United States revealed a lack of consensus on the approach to these lesions.7 Only 20.2% of the respondents (77 of 381) were pediatric dermatologists. However, this subgroup was more likely than other dermatologists to encounter Spitz nevi in clinical practice, with 50% of pediatric dermatologists but only 20% of general dermatologists reporting that they saw at least 6 Spitz nevi per year.7
Children are much more likely to develop Spitz nevi than melanomas, whereas the converse is true for adults.3,5,8,9 As a result, the age of the patient represents an important consideration in the management of an enlarging skin lesion suspected to be melanocytic.8 Although approximately half of pediatric dermatologists and one-third of general dermatologists who participated in the 2001 survey indicated that they treat Spitz nevi in children differently from those in adults, specific questions about management of these lesions did not distinguish between pediatric and adult patients.7 In addition, the use of dermoscopy has increased dramatically among dermatologists during the past decade, and dermoscopic features that can aid in the recognition and longitudinal evaluation of Spitz nevi (especially pigmented variants) have been described.10- 12 This has led to new strategies for the clinical diagnosis and management of Spitz nevi in children.13- 15
We conducted a physician survey-based study to determine the current perspectives of pediatric dermatologists on Spitz nevi in children and adolescents; respondents included dermatologists practicing in a variety of settings in the United States and elsewhere in the world. Our survey assessed the beliefs, diagnostic approaches, and management behaviors of pediatric dermatologists regarding Spitz nevi in pediatric patients, along with their collective experiences with outcomes in these patients.
We conducted an anonymous, voluntary survey of 342 practicing dermatologists from the United States and around the world who were listed on the Society for Pediatric Dermatology website's general-access pediatric dermatologist locator (www.pedsderm.net/locator.php). These physicians were contacted via e-mail and invited to complete an online survey via the website www.surveymonkey.com. The survey invitation and link were sent in July 2010, and a reminder was delivered 3 weeks later; responses were collected for 6 weeks after the initial e-mail. The study was approved by the institutional review board of the New York University School of Medicine.
The survey consisted of 22 multiple-choice questions (eFigure), some of which offered the option of a free-written response or explanation. The questions assessed beliefs, practice behaviors (including general policies and 3 specific clinical scenarios), and experiences with regard to Spitz nevi in pediatric patients. Demographic information (eg, level of training, practice setting, and number of years in practice) was also obtained.
Data from each survey respondent (identified by a number) were recorded on an electronic source document. Descriptive frequencies were calculated to characterize physician demographics and survey responses. Pearson χ2 tests were performed (with commercially available software) to determine whether beliefs and behaviors differed among various demographic groups.
A total of 175 physicians completed the survey, a 51.1% response rate. Of respondents with at least 75% pediatric patients in their practices, 35.5% (39 of 110) diagnosed more than 10 Spitz nevi yearly compared with 21.3% (13 of 61) with less than 75% pediatric patients (P = .05). Table 1 lists characteristics of the 144 respondents whose practices included at least 50% pediatric patients (younger than 18 years) and their general experiences with Spitz nevi. Unless otherwise noted, analyses were limited to this pediatric dermatologist group; some questions were not answered by all of these respondents, resulting in denominators less than 144 for a subset of the calculations.
On the basis of the length of time each individual had been in practice (about 2000 years collectively), we estimated that the total number of Spitz nevi seen by the respondents during their careers was more than 20 000, with an extrapolated mean of approximately 10 Spitz nevi per respondent per year. In contrast, 64.1% (91 of 142) of respondents had diagnosed no prepubertal melanomas, and 90.1% (128 of 142) had diagnosed no more than 2 prepubertal melanomas (Figure) during the prior 5 years, with an estimated mean of less than 0.2 prepubertal melanomas per year of practice.
Of the pediatric dermatologists, 95.8% (136 of 142) categorized typical Spitz nevi as benign, 3.5% (5 of 142) were unsure how to classify these lesions, and 0.7% (1 of 142) considered typical Spitz nevi malignant. Among all respondents, dermatologists with practices composed of at least 75% pediatric patients were more likely to regard Spitz nevi as benign (97.2%; 105 of 108) than those with smaller proportions of their practices composed of pediatric patients (85.2%; 52 of 61; P = .009). The number of years in practice was not associated with categorization of Spitz nevi. Finally, 85.9% of respondents (122 of 142) did not think that typical Spitz nevi were melanoma precursors, 12.7% (18 of 142) were unsure, and 1.4% (2 of 142) thought they were melanoma precursors.
Of the pediatric dermatologists, 79.6% (113 of 142) used dermoscopy to evaluate lesions suspected to be Spitz nevi (Table 2), with a higher rate of use among those in practice for 5 years or less (100%; 31 of 31) than among those in practice for more than 5 years (71.2%; 79 of 111) (P < .001). More than 90% of respondents indicated that they had not always suspected the diagnosis when performing biopsies in lesions that turned out to be a Spitz nevi, and this was the most common reason for performing partial biopsies in such lesions (Table 2).
Management strategies chosen by respondents for 3 clinical scenarios involving suspected Spitz nevi are presented in Table 3. Predictors of clinical follow-up in scenario 2 (pigmented Spitz nevus) included having an academic practice (P = .03), believing that Spitz nevi are not melanoma precursors (P = .04), and never seeing an unambiguous melanoma in a child (P = .01).
Management plans chosen by respondents for 3 scenarios after biopsy of a histologically typical Spitz nevus in a pediatric patient are presented in Table 4. Among respondents to these questions, 79.9% (107 of 134) had not recommended a sentinel lymph node biopsy (SLNB) for any patient with a Spitz nevus or atypical spitzoid neoplasm (ASN) during the past 5 years, 17.2% (23 of 134) had referred 1 or 2 such patients for SLNB, and only 3.0% (4 of 134) had referred 3 or more patients. In addition, 68.7% (92 of 134) of respondents believed that nothing is gained from performing SLNB in patients with typical Spitz nevi or ASNs, and 30.6% (41 of 134) thought that SLNB could provide reassurance if results were negative. Much smaller proportions of the respondents believed that SLNB results in patients with Spitz nevi would indicate a poorer prognosis if positive (8.2%; 11 of 134) or guide the decisions to perform a “therapeutic” lymph node dissection (6.0%; 8 of 134) and/or administer adjuvant therapy (3.0%; 4 of 134).
Among respondents to these questions, 47.0% (62 of 132) had observed clinical and/or histologic evidence of regression of Spitz nevi, whereas only 2.3% (3 of 130) and 6.9% (9 of 130) had seen extracutaneous sequelae (other than positive SLNB results) of a typical Spitz nevus or ASN, respectively (Table 5). Data were provided for 17 patients with Spitz nevi or ASNs and positive SLNB results (Table 6), all of whom were alive with no evidence of disease after a mean follow-up of 2 years (range, 0-6 years). Among 8 additional patients who presented with clinically evident regional lymph node involvement and were followed up for a mean of 5.5 years (range, 1-10 years), 6 had no evidence of disease, 1 had recurrent lymph node involvement, and 1 died after development of distant metastases (Table 6).
Two respondents had each cared for a child or adolescent whose death was attributed to a lesion initially diagnosed as a Spitz nevus or ASN. Death associated with Spitz nevus or ASN was thus extremely rare, occurring in association with less than 1 in 10 000 lesions seen by the respondents and none of the approximately 10 000 lesions seen by the 91 respondents overall with academic or hospital-based practices.
The results of our study confirm that Spitz nevi are commonly seen by pediatric dermatologists. Two-thirds of pediatric dermatologists who responded to our survey (96 of 142) and half of those who responded to the 2001 Spitz nevus survey estimated seeing at least 6 Spitz nevi per year.7 Of our respondents, 95.8% (136 of 142), including 100% (76 of 76) of those with an academic practice composed primarily (≥75%) of children, believed that typical Spitz nevi are benign compared with only 74.2% (279 of 376) of all respondents to the 2001 survey7 (P < .001). This difference probably reflects the large predominance of Spitz nevi over melanomas (including spitzoid melanomas) in children, whereas the opposite is true in adults.3,8 Despite the existence of atypical spitzoid neoplasms that cannot be definitively classified based on their histologic features, precluding reliable prediction of their behavior, dermatopathologists generally agree that the typical Spitz nevus is benign and can be histologically diagnosed with certainty.16- 20
The proportion of our respondents who used dermoscopy to evaluate suspected Spitz nevi (79.6%; 113 of 142) represented a marked increase from the proportion of American Academy of Dermatology fellows in a 2001 survey about dysplastic nevi who used dermoscopy for pigmented lesions (23.1%; 103 of 446) (P < .001).21 Dermoscopy thus provided our respondents with a diagnostic tool that had not been widely used by dermatologists at the time of the 2001 Spitz nevus survey.7 The rate of dermoscopy usage in our survey was also higher than the 48.0% (1555 of 3238) of American Academy of Dermatology fellows in a 2009 survey of dermoscopy use in general (P < .001).10
Many of our respondents believed that dermoscopy was helpful in distinguishing nonpigmented as well as pigmented Spitz nevi from other entities in the differential diagnosis. However, despite dermoscopy usage, more than one-third of our respondents had failed to consider the diagnosis clinically in more than 25% of their biopsies of Spitz nevi during the past 5 years. Quiz Ref IDSeveral studies22- 25 using histopathologic databases have shown that the correct diagnosis is clinically suspected in no more than 20% of Spitz nevi that are biopsied. In one recent series, the clinical diagnosis was dermatofibroma in 10.9% (38 of 349) and pyogenic granuloma in 6.6% (23 of 349) of the Spitz nevi submitted for histologic examination.22 Our respondents most often mistook Spitz nevi for pyogenic granulomas, juvenile xanthogranulomas, and warts, reflecting their pediatric patient populations.
Most of our respondents realized that optimal histologic evaluation of a Spitz nevus requires a complete specimen, which enables assessment of reassuring features such as symmetry, circumscribed lateral margins, and maturation with depth.16,17,24,26 However, many had performed a partial biopsy of a Spitz nevus, usually when they did not recognize the lesion to be melanocytic and therefore performed a superficial biopsy or curettage. An increased index of suspicion for nonpigmented Spitz nevi and the use of dermoscopy for preoperative assessment of lesions initially presumed to be nonmelanocytic would help ensure that adequate histologic specimens are obtained when Spitz nevi are biopsied. Quiz Ref IDAs noted by many of our respondents, the recognition that a stable lesion in a child is a Spitz nevus (rather than, for example, a wart or pyogenic granuloma) would in many instances allow surgical procedures to be avoided. In contrast, biopsy should be performed of lesions with unusually rapid growth or other concerning features (see following). Although a complete biopsy is preferable, partial sampling of a larger lesion (eg, in a cosmetically sensitive site or that would require general anesthesia for excision) can be considered, especially when a nonmelanocytic neoplasm is thought to be more likely than a Spitz nevus.
In the 2001 Spitz nevus survey, 93% of respondents said that they would perform a biopsy of any suspected Spitz nevus, and the techniques that they selected for this general indication varied considerably.7 We and other authors13- 15,26 believe that the decision to perform a biopsy of a lesion thought to be a Spitz nevus and the type of biopsy performed should depend on characteristics of both the lesion (eg, pattern of growth, size, morphological characteristics, and location) and the patient (eg, age). Our survey therefore provided 3 common clinical scenarios with differences in key aspects of these features. This yielded information on the respondents' approaches to diagnosis and management in distinct, real-life clinical situations with pediatric patients.
Clinical scenario 1 involved a preschool-aged child with a small, stable, nonpigmented Spitz nevus in a cosmetically sensitive facial site. We believe that following the lesion clinically is the preferable option in this situation, and almost half of respondents agreed. In clinical scenario 2, the classic dermoscopic findings of a pigmented Spitz nevus enabled a more definitive clinical diagnosis to be established. Multiple longitudinal studies11- 13,27- 29 using dermoscopy have shown that pigmented Spitz nevi have a tendency to involute spontaneously (see later discussion). This has led to recommendations that pigmented Spitz nevi with classic dermoscopic features be followed clinically in children younger than 12 years.13- 15 However, less than one-third of our respondents (31.2% of those who used dermoscopy and 24.1% of those who did not) would follow such a lesion clinically. In the future, increased awareness among dermatologists of the dermoscopic features of pigmented Spitz nevi will probably lead to fewer biopsies of these lesions in children. Clinical scenario 3 involved a larger, nonpigmented Spitz nevus in an adolescent. Recently proposed guidelines agree that biopsy is generally recommended for suspected Spitz nevi (particularly amelanotic papulonodular lesions) in postpubertal patients and for those with atypical features, such as large size (eg, >1 cm), rapid evolution (especially if previously stable), asymmetry, or ulceration in patients of any age.13- 15
The biopsy techniques chosen by our respondents also differed depending on the clinical setting. A punch biopsy with intent to remove the nevus completely was the preferred method for the smaller lesions in scenarios 1 and 2, whereas elliptical excision was selected most often for the larger lesion in scenario 3. In the 2001 Spitz nevus survey, total excision was favored (43% of respondents), followed by shave/saucerization and punch/partial excision (33% and 22% of respondents, respectively).7 However, no clinical information (eg, lesion size) was provided, and some of these categories overlapped or were ambiguous (eg, a punch biopsy can completely remove a smaller lesion). In a retrospective study24 on surgical management of Spitz nevi, shave biopsy specimens were more likely to have positive margins (67%; usually the deep margin) than specimens from elliptical excisions (28%) or punch biopsies (21%; usually the lateral margin).
Two-thirds of our respondents would observe a child clinically after complete excision of a typical Spitz nevus. Such follow-up, which would not generally be needed after removal of other types of typical nevi, reflects a higher level of anxiety regarding even benign spitzoid lesions. Quiz Ref IDOnly 30.4% of our respondents would reexcise a typical Spitz nevus after biopsy if the histologic margins were positive but there was no clinical evidence of residual lesion. This is similar to the reexcision rate for typical Spitz nevi with positive biopsy margins (38%) observed in the study24 on surgical management of Spitz nevi, which documented a higher reexcision rate for ASNs (78%). In the 2001 survey,7 69% of respondents indicated that they would reexcise an incompletely removed Spitz nevus, but the survey question did not specify the type of Spitz nevus (typical or atypical) or the amount of residual lesion. As in our study, most respondents stated that they would use 1- to 2-mm margins for the reexcision. In a series of 30 incompletely removed Spitz nevi,30 no histologic evidence of residual nevus was found in 5 of 6 lesions that were reexcised, and none of the other lesions recurred during a mean follow-up of 5 years. We agree with the authors of that study and other experts16,18,30 that reexcision is not necessary after incomplete excision when the biopsy sample allows the diagnosis of a typical Spitz nevus to be established with certainty.
Most of our respondents did not believe that SLNB was useful for patients with Spitz nevi or ASNs, and all 17 patients with positive SLNB results were disease free after a mean follow-up of 2 years. Although SLNB results can help determine the prognosis of conventional melanomas in adults, positive results have not been shown to have prognostic significance for pediatric melanomas or ASNs in patients of any age. In a total of 8 series31- 38 including results of SLNB in at least 10 patients (adult and/or pediatric) with ASNs, 39.2% (67 of 171) of the biopsies had positive results, but all of the patients were alive and disease free after a mean follow-up of approximately 3 years. Of note, small aggregates of melanocytes within regional lymph nodes do not necessarily represent metastatic melanoma, because such deposits (intracapsular or intratrabecular more often than parenchymal) have been observed in association with nevi (nodal nevi, including Spitz, banal, and blue subtypes).19,39,40
There is no evidence that further lymph node dissection is of therapeutic benefit for patients with positive SLNB results in the setting of an ASN.19 The utility of this intervention is questionable even for adults with conventional melanomas and positive SLNB results. Lymph node dissection can result in long-term complications such as lymphedema. Likewise, adjuvant systemic therapies have substantial associated morbid effects but no proven benefit in patients with spitzoid lesions.19,34,41
A 30% to 40% likelihood of SLNB positivity has also been documented in several series of pediatric melanomas, similar to that seen with ASNs.19,41- 45 Although prepubertal children with melanoma diagnoses tend to have thicker tumors (which are often spitzoid) and a higher rate of positive SLNB results than adolescents or adults, survival in many series is paradoxically longer in prepubertal patients (overall mean 5-year survival, approximately 90% vs approximately 50% in adolescents).19,42- 47 Because it can be difficult to determine whether some ASNs represent Spitz nevi (with benign behavior) or spitzoid melanomas (with malignant behavior) based on their histologic features, series of “pediatric melanomas” undoubtedly include both types of lesions. Spitz nevi, ASNs, and spitzoid melanomas are thought to exist on a spectrum that is biologically distinct from that of banal nevi, dysplastic nevi, and conventional (eg, superficial spreading) melanomas.19,48,49 Spitzoid melanomas comprise a very small portion of lesions on the Spitz spectrum in children. Identifying molecular markers of ASNs with aggressive behavior is an important goal, with promising newer methods, including array-based comparative genomic hybridization.38,50,51
Considering the controversy surrounding spitzoid lesions, it is reassuring that our survey respondents had cared for only 2 patients who were thought to have died because of a lesion initially diagnosed as a Spitz nevus or ASN. This represents an extremely small fraction of the approximately 20 000 Spitz nevi diagnosed by these dermatologists. Interestingly, a recent study49 on the long-term outcomes of Spitz-spectrum lesions found that adults with Spitz nevi had an increased likelihood of developing a separate primary melanoma, suggesting a general propensity for melanocytic neoplasia (analogous to that in patients with dysplastic nevi) and the need for periodic total-body skin examination. Whether this association exists for Spitz nevi in children remains to be determined.
In contrast, many of our respondents had observed clinical, dermoscopic, and/or histologic evidence of involution of Spitz nevi (Table 5). Quiz Ref IDAs already noted, longitudinal dermoscopic evaluation of pigmented Spitz nevi has documented their tendency to evolve from a starburst pattern to a reticular or homogeneous pattern and/or regress over a period of a few months to several years.13,28,29 Recently, a large pediatric study27 of nonpigmented as well as pigmented lesions with clinical and dermatoscopic features characteristic of Spitz nevi showed that 79.7% (51 of 64 nevi) underwent involution during a mean follow-up of 25 months.
The opinions, diagnostic strategies, approaches to treatment, and collective experiences of the pediatric dermatologists who completed our survey support conservative management of Spitz nevi in pediatric patients. The results of this survey and recent longitudinal studies suggest that clinical follow-up is an appropriate option for lesions with the appearance and behavior of a typical Spitz nevus (including the pigmented subtype) in children. Dermoscopy has emerged as an important tool in the evaluation of patients with lesions suspected to be Spitz nevi. It is hoped that continued investigation of the molecular characteristics of neoplasms on the Spitz spectrum will help provide a better marker for rare lesions with malignant behavior.
Correspondence: Julie V. Schaffer, MD, The Ronald O. Perelman Department of Dermatology, New York University School of Medicine, 240 E 38th St, 11th Floor, Dermatology H-100, New York, NY 10016 (email@example.com).
Accepted for Publication: September 6, 2012.
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Tlougan, Orlow, and Schaffer. Acquisition of data: Tlougan, Orlow, and Schaffer. Analysis and interpretation of data: Tlougan, Orlow, and Schaffer. Drafting of the manuscript: Tlougan and Schaffer. Critical revision of the manuscript for important intellectual content: Orlow and Schaffer. Statistical analysis: Schaffer. Administrative, technical, and material support: Orlow and Schaffer. Study supervision: Orlow and Schaffer.
Conflict of Interest Disclosures: None reported.