Nonmelanoma Skin Cancer

We report the case of a 6-week-old girl who presented with a pedunculated embryonal rhabdomyosarcoma arising in a giant congenital melanocytic nevus (GCMN) on her lower back. There was no associated leptomeningeal involvement. The patient underwent surgical resection of the rhabdomyosarcoma at age 2 months, with subsequent chemotherapy consisting of actinomycin D and vincristine. No recurrences of metastases of tumor have been noted at 5 months of age.

Giant congenital melanocytic nevus (GCMN), defined as measuring or predicted to measure greater than 20 cm in diameter by adulthood, uncommonly occurs in newborns.1 Concerns arise regarding malignant degeneration, with melanoma developing in 4.5% to 8.5% of cases.2,3 Other malignancies rarely have been reported in a GCMN, including rhabdomyosarcoma, melanosarcoma, undifferentiated spindle cell tumor, neuroblastoma, melanoblastoma, myoblastic sarcoma, liposarcoma, and malignant peripheral nerve sheath tumor.1,4 We report a case of rhabdomyosarcoma arising in a GCMN. 

CASE REPORT

A 6-week-old infant girl presented with an 18x13-cm pigmented hairy nevus extending from the lower back onto the abdomen. A rapidly enlarging, soft pink pedunculated polyp measuring 3 cm in diameter had been present centrally at birth (Figure 1). An ultrasound of the spine revealed no abnormalities.

The polyp was surgically resected, along with a portion of the GCMN. Results of the histologic examination disclosed a rhabdomyosarcoma, embryonal type, located superficially within the dermis (Figure 2). The deep and lateral margins showed no evidence of tumor. Immunohistochemistry results of the neoplastic cells showed marked positivity for desmins, smooth muscle actin, and vimentin; results of tests for S100, neuron-specific enolase, chromogranins, synaptophysin, leukocyte common antigen, and CD57 were negative. Myoglobin stain was noncontributory.

No evidence of metastatic disease was noted on computed tomography of the chest, abdomen, and pelvis; bone scan; and bone marrow examination. The patient was started on a chemotherapy regimen consisting of actinomycin D and vincristine.

Comment

Although controversy exists regarding the precise risk of malignant degeneration in GCMN, it presents a major concern. Often, extensive surgical procedures are undertaken to eliminate or manage the risk of malignancy, primarily to minimize the risk of melanoma, the most commonly associated malignancy.5

Rhabdomyosarcoma is the most common soft tissue sarcoma and the third most common extracranial solid tumor of childhood. It is responsible for up to 25% of malignant cutaneous neoplasms in the pediatric population.6 Most tumors occur in the head and neck region, as well as the genitourinary tract; other areas include the extremities and trunk. Rhabdomyosarcoma occurring in the skin de novo may present as an asymptomatic, firm red nodule or as a lobulated mass with a shiny erythematous surface and telangiectatic vessels.6,7 The differential diagnosis includes hemangioma, lymphangioma, hygroma, lymphoma, leukemia cutis, amelanotic melanoma, and angiofibroma.7

The 4 most common histologic subtypes of rhabdomyosarcoma include the following: embryonal, embryonal-botryoid, alveolar, and pleomorphic. The embryonal and embryonal-butryoid subtypes account for 50% to 60% of childhood rhabdomyosarcomas. The alveolar pattern is seen more commonly in older children, and the pleomorphic pattern is seen more in adults.7

Immunohistochemical stains used to identify rhabdomyosarcoma include the intermediate filaments vimentin and desmins. Vimentin highlights undifferentiated cells of mesenchymal origin, whereas desmins become positive as a cell becomes committed to rhabdomyoblastic differentiation. Because myoglobin is also found in skeletal muscle, myoglobin stains less consistently in rhabdomyosarcomas. This is most likely a result of its expression later in the development of skeletal muscle.7,8 Other muscle-specific proteins used in identification are Myo-D, myogenin, muscle- specific actin, and Z-band protein.9

Staging of rhabdomyosarcoma by the Intergroup Rhabdomyosarcoma Study (IRS) initially was based on a clinical grouping that sorted patients according to the extent of disease after surgical resection: group I, complete resection; group II, microscopic residual disease after gross resection, with or without nodal involvement; group III, gross residual disease after surgery; and group IV, metastatic disease at diagnosis. Now, the IRS is incorporating a more prognostically significant TNM staging. The TNM Staging System, in addition to clinical grouping, includes favorable versus unfavorable sites, along with increasing the stage number for bulky disease measuring greater than 5 cm in diameter or for nodal involvement. Favorable sites include the orbit, eyelid, head and neck, nonparameningeal, and genitourinary areas not including bladder or prostate. Unfavorable areas include extremities, bladder, prostate, parameningeal, retroperitoneum, and, as in our patient, the trunk.8,9 According to the latest IRS staging, our patient falls into the clinical group I, TNM stage II category, which places her in the low-risk rhabdomyosarcoma group.

The treatment of rhabdomyosarcoma in a GCMN consists of surgical resection and combination chemotherapy with or without radiation therapy, depending on the clinical grouping of the patient. According to the IRS definition of low-risk patients, the chemotherapy regimen consists of actinomycin D and vincristine with or without cyclophosphamide, depending on stage and location. Radiation therapy is added for patients with groups II through IV disease and for some unfavorable locations with a high risk of recurrence.9,10

Rhabdomyosarcoma arising in a GCMN has been reported rarely, with only 4 cases to date in the literature (Table).4,5,11,12 Three of the 4 cases occurred in infants, with a fatal outcome reported in one of the infants. In 3 of 4 cases, the trunk was indicated as the site of the rhabdomyosarcoma, with only one case associated with leptomeningeal involvement. In all cases, the rhabdomyosarcoma was excised, with subsequent chemotherapy in 2 of the 4 previously reported cases.

The occurrence of rhabdomyosarcoma arising in a GCMN could be explained by the pluripotential nature of undifferentiated neural crest cells. These cells likely maintain the ability for rhabdomyoblastic differentiation. This mesenchymal derivative originates from the cranial or cephalic portion of the neural crest. The term ectomesenchyme applies to the potential of the ectoderm to give rise to mesenchymal tissue and indeed may be the pathway by which a rhabdomyosarcoma arises.5

Conclusion

We report a fifth case of rhabdomyosarcoma arising in a GCMN. Our patient underwent surgical resection of the rhabdyomyosarcoma at age 2 months, with subsequent chemotherapy consisting of actinomycin D and vincristine. She has tolerated outpatient chemotherapy well, with no gross evidence of local recurrence, and will complete 12 months of therapy for her low-risk rhabdomyosarcoma. 

We report the case of a 6-week-old girl who presented with a pedunculated embryonal rhabdomyosarcoma arising in a giant congenital melanocytic nevus (GCMN) on her lower back. There was no associated leptomeningeal involvement. The patient underwent surgical resection of the rhabdomyosarcoma at age 2 months, with subsequent chemotherapy consisting of actinomycin D and vincristine. No recurrences of metastases of tumor have been noted at 5 months of age.

Giant congenital melanocytic nevus (GCMN), defined as measuring or predicted to measure greater than 20 cm in diameter by adulthood, uncommonly occurs in newborns.1 Concerns arise regarding malignant degeneration, with melanoma developing in 4.5% to 8.5% of cases.2,3 Other malignancies rarely have been reported in a GCMN, including rhabdomyosarcoma, melanosarcoma, undifferentiated spindle cell tumor, neuroblastoma, melanoblastoma, myoblastic sarcoma, liposarcoma, and malignant peripheral nerve sheath tumor.1,4 We report a case of rhabdomyosarcoma arising in a GCMN. 

CASE REPORT

A 6-week-old infant girl presented with an 18x13-cm pigmented hairy nevus extending from the lower back onto the abdomen. A rapidly enlarging, soft pink pedunculated polyp measuring 3 cm in diameter had been present centrally at birth (Figure 1). An ultrasound of the spine revealed no abnormalities.

The polyp was surgically resected, along with a portion of the GCMN. Results of the histologic examination disclosed a rhabdomyosarcoma, embryonal type, located superficially within the dermis (Figure 2). The deep and lateral margins showed no evidence of tumor. Immunohistochemistry results of the neoplastic cells showed marked positivity for desmins, smooth muscle actin, and vimentin; results of tests for S100, neuron-specific enolase, chromogranins, synaptophysin, leukocyte common antigen, and CD57 were negative. Myoglobin stain was noncontributory.

No evidence of metastatic disease was noted on computed tomography of the chest, abdomen, and pelvis; bone scan; and bone marrow examination. The patient was started on a chemotherapy regimen consisting of actinomycin D and vincristine.

Comment

Although controversy exists regarding the precise risk of malignant degeneration in GCMN, it presents a major concern. Often, extensive surgical procedures are undertaken to eliminate or manage the risk of malignancy, primarily to minimize the risk of melanoma, the most commonly associated malignancy.5

Rhabdomyosarcoma is the most common soft tissue sarcoma and the third most common extracranial solid tumor of childhood. It is responsible for up to 25% of malignant cutaneous neoplasms in the pediatric population.6 Most tumors occur in the head and neck region, as well as the genitourinary tract; other areas include the extremities and trunk. Rhabdomyosarcoma occurring in the skin de novo may present as an asymptomatic, firm red nodule or as a lobulated mass with a shiny erythematous surface and telangiectatic vessels.6,7 The differential diagnosis includes hemangioma, lymphangioma, hygroma, lymphoma, leukemia cutis, amelanotic melanoma, and angiofibroma.7

The 4 most common histologic subtypes of rhabdomyosarcoma include the following: embryonal, embryonal-botryoid, alveolar, and pleomorphic. The embryonal and embryonal-butryoid subtypes account for 50% to 60% of childhood rhabdomyosarcomas. The alveolar pattern is seen more commonly in older children, and the pleomorphic pattern is seen more in adults.7

Immunohistochemical stains used to identify rhabdomyosarcoma include the intermediate filaments vimentin and desmins. Vimentin highlights undifferentiated cells of mesenchymal origin, whereas desmins become positive as a cell becomes committed to rhabdomyoblastic differentiation. Because myoglobin is also found in skeletal muscle, myoglobin stains less consistently in rhabdomyosarcomas. This is most likely a result of its expression later in the development of skeletal muscle.7,8 Other muscle-specific proteins used in identification are Myo-D, myogenin, muscle- specific actin, and Z-band protein.9

Staging of rhabdomyosarcoma by the Intergroup Rhabdomyosarcoma Study (IRS) initially was based on a clinical grouping that sorted patients according to the extent of disease after surgical resection: group I, complete resection; group II, microscopic residual disease after gross resection, with or without nodal involvement; group III, gross residual disease after surgery; and group IV, metastatic disease at diagnosis. Now, the IRS is incorporating a more prognostically significant TNM staging. The TNM Staging System, in addition to clinical grouping, includes favorable versus unfavorable sites, along with increasing the stage number for bulky disease measuring greater than 5 cm in diameter or for nodal involvement. Favorable sites include the orbit, eyelid, head and neck, nonparameningeal, and genitourinary areas not including bladder or prostate. Unfavorable areas include extremities, bladder, prostate, parameningeal, retroperitoneum, and, as in our patient, the trunk.8,9 According to the latest IRS staging, our patient falls into the clinical group I, TNM stage II category, which places her in the low-risk rhabdomyosarcoma group.

The treatment of rhabdomyosarcoma in a GCMN consists of surgical resection and combination chemotherapy with or without radiation therapy, depending on the clinical grouping of the patient. According to the IRS definition of low-risk patients, the chemotherapy regimen consists of actinomycin D and vincristine with or without cyclophosphamide, depending on stage and location. Radiation therapy is added for patients with groups II through IV disease and for some unfavorable locations with a high risk of recurrence.9,10

Rhabdomyosarcoma arising in a GCMN has been reported rarely, with only 4 cases to date in the literature (Table).4,5,11,12 Three of the 4 cases occurred in infants, with a fatal outcome reported in one of the infants. In 3 of 4 cases, the trunk was indicated as the site of the rhabdomyosarcoma, with only one case associated with leptomeningeal involvement. In all cases, the rhabdomyosarcoma was excised, with subsequent chemotherapy in 2 of the 4 previously reported cases.

The occurrence of rhabdomyosarcoma arising in a GCMN could be explained by the pluripotential nature of undifferentiated neural crest cells. These cells likely maintain the ability for rhabdomyoblastic differentiation. This mesenchymal derivative originates from the cranial or cephalic portion of the neural crest. The term ectomesenchyme applies to the potential of the ectoderm to give rise to mesenchymal tissue and indeed may be the pathway by which a rhabdomyosarcoma arises.5

Conclusion

We report a fifth case of rhabdomyosarcoma arising in a GCMN. Our patient underwent surgical resection of the rhabdyomyosarcoma at age 2 months, with subsequent chemotherapy consisting of actinomycin D and vincristine. She has tolerated outpatient chemotherapy well, with no gross evidence of local recurrence, and will complete 12 months of therapy for her low-risk rhabdomyosarcoma. 

We report the case of a 6-week-old girl who presented with a pedunculated embryonal rhabdomyosarcoma arising in a giant congenital melanocytic nevus (GCMN) on her lower back. There was no associated leptomeningeal involvement. The patient underwent surgical resection of the rhabdomyosarcoma at age 2 months, with subsequent chemotherapy consisting of actinomycin D and vincristine. No recurrences of metastases of tumor have been noted at 5 months of age.

Giant congenital melanocytic nevus (GCMN), defined as measuring or predicted to measure greater than 20 cm in diameter by adulthood, uncommonly occurs in newborns.1 Concerns arise regarding malignant degeneration, with melanoma developing in 4.5% to 8.5% of cases.2,3 Other malignancies rarely have been reported in a GCMN, including rhabdomyosarcoma, melanosarcoma, undifferentiated spindle cell tumor, neuroblastoma, melanoblastoma, myoblastic sarcoma, liposarcoma, and malignant peripheral nerve sheath tumor.1,4 We report a case of rhabdomyosarcoma arising in a GCMN. 

CASE REPORT

A 6-week-old infant girl presented with an 18x13-cm pigmented hairy nevus extending from the lower back onto the abdomen. A rapidly enlarging, soft pink pedunculated polyp measuring 3 cm in diameter had been present centrally at birth (Figure 1). An ultrasound of the spine revealed no abnormalities.

The polyp was surgically resected, along with a portion of the GCMN. Results of the histologic examination disclosed a rhabdomyosarcoma, embryonal type, located superficially within the dermis (Figure 2). The deep and lateral margins showed no evidence of tumor. Immunohistochemistry results of the neoplastic cells showed marked positivity for desmins, smooth muscle actin, and vimentin; results of tests for S100, neuron-specific enolase, chromogranins, synaptophysin, leukocyte common antigen, and CD57 were negative. Myoglobin stain was noncontributory.

No evidence of metastatic disease was noted on computed tomography of the chest, abdomen, and pelvis; bone scan; and bone marrow examination. The patient was started on a chemotherapy regimen consisting of actinomycin D and vincristine.

Comment

Although controversy exists regarding the precise risk of malignant degeneration in GCMN, it presents a major concern. Often, extensive surgical procedures are undertaken to eliminate or manage the risk of malignancy, primarily to minimize the risk of melanoma, the most commonly associated malignancy.5

Rhabdomyosarcoma is the most common soft tissue sarcoma and the third most common extracranial solid tumor of childhood. It is responsible for up to 25% of malignant cutaneous neoplasms in the pediatric population.6 Most tumors occur in the head and neck region, as well as the genitourinary tract; other areas include the extremities and trunk. Rhabdomyosarcoma occurring in the skin de novo may present as an asymptomatic, firm red nodule or as a lobulated mass with a shiny erythematous surface and telangiectatic vessels.6,7 The differential diagnosis includes hemangioma, lymphangioma, hygroma, lymphoma, leukemia cutis, amelanotic melanoma, and angiofibroma.7

The 4 most common histologic subtypes of rhabdomyosarcoma include the following: embryonal, embryonal-botryoid, alveolar, and pleomorphic. The embryonal and embryonal-butryoid subtypes account for 50% to 60% of childhood rhabdomyosarcomas. The alveolar pattern is seen more commonly in older children, and the pleomorphic pattern is seen more in adults.7

Immunohistochemical stains used to identify rhabdomyosarcoma include the intermediate filaments vimentin and desmins. Vimentin highlights undifferentiated cells of mesenchymal origin, whereas desmins become positive as a cell becomes committed to rhabdomyoblastic differentiation. Because myoglobin is also found in skeletal muscle, myoglobin stains less consistently in rhabdomyosarcomas. This is most likely a result of its expression later in the development of skeletal muscle.7,8 Other muscle-specific proteins used in identification are Myo-D, myogenin, muscle- specific actin, and Z-band protein.9

Staging of rhabdomyosarcoma by the Intergroup Rhabdomyosarcoma Study (IRS) initially was based on a clinical grouping that sorted patients according to the extent of disease after surgical resection: group I, complete resection; group II, microscopic residual disease after gross resection, with or without nodal involvement; group III, gross residual disease after surgery; and group IV, metastatic disease at diagnosis. Now, the IRS is incorporating a more prognostically significant TNM staging. The TNM Staging System, in addition to clinical grouping, includes favorable versus unfavorable sites, along with increasing the stage number for bulky disease measuring greater than 5 cm in diameter or for nodal involvement. Favorable sites include the orbit, eyelid, head and neck, nonparameningeal, and genitourinary areas not including bladder or prostate. Unfavorable areas include extremities, bladder, prostate, parameningeal, retroperitoneum, and, as in our patient, the trunk.8,9 According to the latest IRS staging, our patient falls into the clinical group I, TNM stage II category, which places her in the low-risk rhabdomyosarcoma group.

The treatment of rhabdomyosarcoma in a GCMN consists of surgical resection and combination chemotherapy with or without radiation therapy, depending on the clinical grouping of the patient. According to the IRS definition of low-risk patients, the chemotherapy regimen consists of actinomycin D and vincristine with or without cyclophosphamide, depending on stage and location. Radiation therapy is added for patients with groups II through IV disease and for some unfavorable locations with a high risk of recurrence.9,10

Rhabdomyosarcoma arising in a GCMN has been reported rarely, with only 4 cases to date in the literature (Table).4,5,11,12 Three of the 4 cases occurred in infants, with a fatal outcome reported in one of the infants. In 3 of 4 cases, the trunk was indicated as the site of the rhabdomyosarcoma, with only one case associated with leptomeningeal involvement. In all cases, the rhabdomyosarcoma was excised, with subsequent chemotherapy in 2 of the 4 previously reported cases.

The occurrence of rhabdomyosarcoma arising in a GCMN could be explained by the pluripotential nature of undifferentiated neural crest cells. These cells likely maintain the ability for rhabdomyoblastic differentiation. This mesenchymal derivative originates from the cranial or cephalic portion of the neural crest. The term ectomesenchyme applies to the potential of the ectoderm to give rise to mesenchymal tissue and indeed may be the pathway by which a rhabdomyosarcoma arises.5

Conclusion

We report a fifth case of rhabdomyosarcoma arising in a GCMN. Our patient underwent surgical resection of the rhabdyomyosarcoma at age 2 months, with subsequent chemotherapy consisting of actinomycin D and vincristine. She has tolerated outpatient chemotherapy well, with no gross evidence of local recurrence, and will complete 12 months of therapy for her low-risk rhabdomyosarcoma. 

Granuloma faciale (GF) is an uncommon inflammatory condition characterized by reddish brown papules and plaques that usually involve the facial area. Extrafacial lesions are rare. Histologically, the lesions are marked by leukocytoclastic vasculitis and extensive fibrin deposition. There are a variety of treatment options available for GF.

Granuloma faciale (GF) is an uncommon disease of unknown etiology presenting with asymptomatic cutaneous papules, nodules, and plaques of the face. However, there has been some confusion and evolution of the term granuloma faciale. In 1937, the term eosinophilic granuloma of the skin was used to describe ulcerated lesions of the mouth and anus that were associated with tuberculosis.1,2 Cobane et al3 used the term facial granuloma with eosinophilia. Confusion continued until Lever and Leeper4 subdivided eosinophilic granuloma of the skin into 3 groups. The first group consisted of cases with cutaneous lesions associated with and histologically identical to eosinophilic granuloma of the bone (Langerhans cell histiocytosis). The second group included cases with cutaneous lesions consisting of torpid, asymptomatic purplish patches on the face. The last group was cases with a preponderance of eosinophils in a granulomatous infiltrate that were associated with a variety of diseases.1,4 A case described by Wigley5 in 1945 as Sarcoid of Boeck may, in fact, be the first reported case of GF.6 It was Pinkus,7 however, who in 1952 finally suggested the term granuloma faciale.

GF is characterized by papules, nodules, and plaques that are usually solitary, though multiple or disseminated lesions can be seen.8 The lesions usually are soft, elevated, well-circumscribed nodules ranging in size from millimeters to centimeters.3 The lesions can be shades of red, brown, or purple and may darken with sun exposure. Lesions usually have a smooth surface with follicular accentuation but may have telangiectasias. Ulceration or crusting rarely occurs. These lesions usually develop very slowly and remain unchanged, though occasionally they will involute.6 Tenderness, burning, and pruritus also have been reported.9,10

Although extracutaneous involvement is rare, there are reports of oral mucosa/upper airway involvement. GF can occur in patients of any age, sex, or race but is usually found in middle-aged Caucasian men.1,6,11-13 

Materials and Methods

The clinical presentation and treatment of 38 patients with GF was reviewed. Patient information was obtained from a database covering the past 10 years of patient activity at a large dermatopathology laboratory in Michigan. GF was diagnosed by clinical and histologic criteria including an infiltrate composed of lymphocytes, neutrophils, histiocytes, and eosinophils in the superficial and deep dermis and fibrin deposition within dermal blood vessels confirming the presence of vascular damage. The patients were reviewed according to demographics, location, treatment, and prognosis. Follow-up was obtained through a questionnaire sent to each referring physician. To our knowledge, this review represents the largest series of patients with GF ever described. 

Results

The clinical features of the patients are summarized in the Table. Their ages ranged from 28 to 85 years, with a mean age of 51.9 years for men and 55.4 years for women. The study consisted of 24 men and 14 women, which is in concordance with an early study showing a predominance of GF in men.1 One man and one woman (patients 1 and 7) had multiple lesions. Interestingly, five men (patients 19, 22, 25, 29, and 38) had extrafacial lesions; there are very few reported cases of extrafacial GF.

Comment

GF is characterized by solitary nodules and plaques.1 Most lesions are asymptomatic red, brown, or purple nodules that are soft, elevated, well-circumscribed, and slow to develop (Figure 1).

Lesions usually are seen on the face but may occur elsewhere.14 Particularly, lesions may appear on light-exposed areas, and some lesions are photoexacerbated.10 Sites of predilection are the sides (30%) and tip (7%) of the nose, preauricular area (22%), cheeks (22%), forehead (15%), and helix of the ear (4%).10,15 In a clinical and histopathologic review, Pedace and Perry1 described 21 cases of GF seen at the Mayo Clinic from 1945 to 1965. They noted a predominance of the lesions on the face (nose, forehead, malar, preauricular and postauricular areas, and chin) and less frequently on the forearms and elsewhere.

A review of the literature reveals 14 reported cases of extrafacial involvement.14-18 The clinical aspects of facial and extrafacial lesions are similar.8 Extrafacial lesions usually are found on the trunk and proximal extremities.19 Extrafacial lesions have been reported as isolated findings and in conjunction with facial lesions.17,18 The infrequent reports of extrafacial lesions may reflect either the inconspicuous nature of extrafacial lesions or a failure to examine patients specifically for the lesions.8

Extrafacial GF may be difficult to differentiate from erythema elevatum diutinum (EED). In fact, some authors believe EED and GF represent different parts of the spectrum of the same disease.20 Both are rare chronic forms of cutaneous small vessel vasculitis and may share some pathogenic mechanisms, but there are several clinical and histologic differences. EED is characterized by multiple lesions localized on the extensor surfaces of extremities in an acral, bilateral, and symmetrical distribution. Bulla formation and hemorrhagic crusting may be seen. The trunk is usually spared, and facial lesions are rare. Histopathologic features of EED include a dense superficial and deep polymorphous dermal infiltrate where neutrophils are prominent and eosinophils are scanty or absent. A grenz zone of normal collagen beneath the epidermis rarely exists and the epidermis is not always spared. EED may be associated with systemic conditions, primarily gammopathies. EED shows an excellent response to dapsone.15,21,22

There is a relatively large clinical differential diagnosis for GF including lupus erythematosus, polymorphous light eruption, fixed drug eruption, benign and malignant lymphoid proliferations, sarcoidosis, granuloma annulare, foreign body reaction, tinea faciei, insect bite reaction, juvenile xanthogranuloma, mastocytoma, Spitz nevus, EED, mycosis fungoides, basal cell carcinoma, histiocytosis X, and rosacea.1,6,23-26

There is one case report of Trichophyton rubrum causing histologic changes similar to GF.26 There also has been a case of GF mimicking rhinophyma.27

Histologic findings of GF show a normal epidermis that may be thinned and flattened by underlying infiltrate (Figure 2). There is a narrow grenz zone between the epidermis and the dermal inflammatory infiltrate consisting of lymphocytes, eosinophils, and neutrophils with leukocytoclasis (Figure 3). The infiltrate usually is distributed diffusely in the upper two thirds of the dermis. Fibrin deposition around blood vessels is evidence of vasculitis (Figure 4). In the later fibrotic stage, perivascular fibrin deposition predominates, and the number of inflammatory cells is greatly reduced.6

Microscopically, the primary differential diagnosis is EED, insect bite reaction, cutaneous lymphoma, or leukocytoclastic vasculitis. The exact pathogenesis is unclear, but some consider it a variant of leukocytoclastic vasculitis. Immunoglobulins, fibrin, and complement can be found at the dermal-epidermal junction and around blood vessels on direct immunofluorescence.28-30

GF usually lacks systemic symptoms or laboratory findings other than rare peripheral eosinophilia.31 Immunohistochemical analysis revealed the majority of lymphocytes to be helper T-cell lymphocytes. The cells stained strongly with antibodies against IL-2 receptor and with antibodies to lymphocyte functional antigen-1 α. Overlying keratinocytes did not stain with intracellular adhesion molecule-1 or HLA-DR, which may account for the presence of the grenz zone in GF. These findings suggest that a γ–interferon-mediated process may play some role in the pathogenesis of this disorder.32

GF is known to be resistant to therapy. Numerous physical modalities and medical therapeutics have been tried. Laser therapy, including the CO2,33 argon,34 pulsed dye, and long-pulsed tunable dye lasers,35-38 all have been attempted with varying success. A study showed that lesions treated with a CO2 laser and dermabrasion had a more even texture compared with lesions treated with electrosurgery alone. Healing times were similar between lesions treated with electrosurgery and CO2 laser; however, lesions treated with dermabrasion healed more quickly.39 Studies of patients treated with an argon laser resulted in total resolution of plaques of GF but had a remaining white collagenous scar.34

A case report by Elston37 showed complete resolution of 3 lesions of GF when treated with a pulsed dye laser after the patient failed topical corticosteroids and oral dapsone. A case report by Ammirati et al35 of a patient treated with the 585-nm pulsed dye laser showed clinical eradication of the lesion at 6-year follow-up. Another report by Welsh et al36 showed good results when GF was treated with the pulsed dye laser. Recently, the long-pulsed tunable dye laser was used successfully with no scarring.38 Other modalities that have been used include surgical excision,7,10 dermabrasion, superficial ionizing radiation,6,10 topical psoralen plus UV light,40 cryosurgery,41 intralesional corticosteroids,42 combined cryosurgery and intralesional steroid injection,43 and electrodesiccation.10,39

Medical treatment has included intralesional gold, colchicine, isoniazid, corticosteroids, potassium arsenite, testosterone, antimalarials, dapsone, and clofazimine.9,10,25,27,34,39,44-46 Most medical therapies have shown varying success. No controlled trials are available because of the rarity of the condition.

In our review, most patients were treated with topical and intralesional steroids with varying results that ranged from mild improvement to complete resolution. Because there is a lack of scarring with steroid therapy, we recommend this as a good first-line therapy. Although none of the patients were treated with pulsed dye laser therapy, review of the literature demonstrates favorable results with this treatment modality. Pulsed dye laser should be considered as an alternative therapy.

Granuloma faciale (GF) is an uncommon inflammatory condition characterized by reddish brown papules and plaques that usually involve the facial area. Extrafacial lesions are rare. Histologically, the lesions are marked by leukocytoclastic vasculitis and extensive fibrin deposition. There are a variety of treatment options available for GF.

Granuloma faciale (GF) is an uncommon disease of unknown etiology presenting with asymptomatic cutaneous papules, nodules, and plaques of the face. However, there has been some confusion and evolution of the term granuloma faciale. In 1937, the term eosinophilic granuloma of the skin was used to describe ulcerated lesions of the mouth and anus that were associated with tuberculosis.1,2 Cobane et al3 used the term facial granuloma with eosinophilia. Confusion continued until Lever and Leeper4 subdivided eosinophilic granuloma of the skin into 3 groups. The first group consisted of cases with cutaneous lesions associated with and histologically identical to eosinophilic granuloma of the bone (Langerhans cell histiocytosis). The second group included cases with cutaneous lesions consisting of torpid, asymptomatic purplish patches on the face. The last group was cases with a preponderance of eosinophils in a granulomatous infiltrate that were associated with a variety of diseases.1,4 A case described by Wigley5 in 1945 as Sarcoid of Boeck may, in fact, be the first reported case of GF.6 It was Pinkus,7 however, who in 1952 finally suggested the term granuloma faciale.

GF is characterized by papules, nodules, and plaques that are usually solitary, though multiple or disseminated lesions can be seen.8 The lesions usually are soft, elevated, well-circumscribed nodules ranging in size from millimeters to centimeters.3 The lesions can be shades of red, brown, or purple and may darken with sun exposure. Lesions usually have a smooth surface with follicular accentuation but may have telangiectasias. Ulceration or crusting rarely occurs. These lesions usually develop very slowly and remain unchanged, though occasionally they will involute.6 Tenderness, burning, and pruritus also have been reported.9,10

Although extracutaneous involvement is rare, there are reports of oral mucosa/upper airway involvement. GF can occur in patients of any age, sex, or race but is usually found in middle-aged Caucasian men.1,6,11-13 

Materials and Methods

The clinical presentation and treatment of 38 patients with GF was reviewed. Patient information was obtained from a database covering the past 10 years of patient activity at a large dermatopathology laboratory in Michigan. GF was diagnosed by clinical and histologic criteria including an infiltrate composed of lymphocytes, neutrophils, histiocytes, and eosinophils in the superficial and deep dermis and fibrin deposition within dermal blood vessels confirming the presence of vascular damage. The patients were reviewed according to demographics, location, treatment, and prognosis. Follow-up was obtained through a questionnaire sent to each referring physician. To our knowledge, this review represents the largest series of patients with GF ever described. 

Results

The clinical features of the patients are summarized in the Table. Their ages ranged from 28 to 85 years, with a mean age of 51.9 years for men and 55.4 years for women. The study consisted of 24 men and 14 women, which is in concordance with an early study showing a predominance of GF in men.1 One man and one woman (patients 1 and 7) had multiple lesions. Interestingly, five men (patients 19, 22, 25, 29, and 38) had extrafacial lesions; there are very few reported cases of extrafacial GF.

Comment

GF is characterized by solitary nodules and plaques.1 Most lesions are asymptomatic red, brown, or purple nodules that are soft, elevated, well-circumscribed, and slow to develop (Figure 1).

Lesions usually are seen on the face but may occur elsewhere.14 Particularly, lesions may appear on light-exposed areas, and some lesions are photoexacerbated.10 Sites of predilection are the sides (30%) and tip (7%) of the nose, preauricular area (22%), cheeks (22%), forehead (15%), and helix of the ear (4%).10,15 In a clinical and histopathologic review, Pedace and Perry1 described 21 cases of GF seen at the Mayo Clinic from 1945 to 1965. They noted a predominance of the lesions on the face (nose, forehead, malar, preauricular and postauricular areas, and chin) and less frequently on the forearms and elsewhere.

A review of the literature reveals 14 reported cases of extrafacial involvement.14-18 The clinical aspects of facial and extrafacial lesions are similar.8 Extrafacial lesions usually are found on the trunk and proximal extremities.19 Extrafacial lesions have been reported as isolated findings and in conjunction with facial lesions.17,18 The infrequent reports of extrafacial lesions may reflect either the inconspicuous nature of extrafacial lesions or a failure to examine patients specifically for the lesions.8

Extrafacial GF may be difficult to differentiate from erythema elevatum diutinum (EED). In fact, some authors believe EED and GF represent different parts of the spectrum of the same disease.20 Both are rare chronic forms of cutaneous small vessel vasculitis and may share some pathogenic mechanisms, but there are several clinical and histologic differences. EED is characterized by multiple lesions localized on the extensor surfaces of extremities in an acral, bilateral, and symmetrical distribution. Bulla formation and hemorrhagic crusting may be seen. The trunk is usually spared, and facial lesions are rare. Histopathologic features of EED include a dense superficial and deep polymorphous dermal infiltrate where neutrophils are prominent and eosinophils are scanty or absent. A grenz zone of normal collagen beneath the epidermis rarely exists and the epidermis is not always spared. EED may be associated with systemic conditions, primarily gammopathies. EED shows an excellent response to dapsone.15,21,22

There is a relatively large clinical differential diagnosis for GF including lupus erythematosus, polymorphous light eruption, fixed drug eruption, benign and malignant lymphoid proliferations, sarcoidosis, granuloma annulare, foreign body reaction, tinea faciei, insect bite reaction, juvenile xanthogranuloma, mastocytoma, Spitz nevus, EED, mycosis fungoides, basal cell carcinoma, histiocytosis X, and rosacea.1,6,23-26

There is one case report of Trichophyton rubrum causing histologic changes similar to GF.26 There also has been a case of GF mimicking rhinophyma.27

Histologic findings of GF show a normal epidermis that may be thinned and flattened by underlying infiltrate (Figure 2). There is a narrow grenz zone between the epidermis and the dermal inflammatory infiltrate consisting of lymphocytes, eosinophils, and neutrophils with leukocytoclasis (Figure 3). The infiltrate usually is distributed diffusely in the upper two thirds of the dermis. Fibrin deposition around blood vessels is evidence of vasculitis (Figure 4). In the later fibrotic stage, perivascular fibrin deposition predominates, and the number of inflammatory cells is greatly reduced.6

Microscopically, the primary differential diagnosis is EED, insect bite reaction, cutaneous lymphoma, or leukocytoclastic vasculitis. The exact pathogenesis is unclear, but some consider it a variant of leukocytoclastic vasculitis. Immunoglobulins, fibrin, and complement can be found at the dermal-epidermal junction and around blood vessels on direct immunofluorescence.28-30

GF usually lacks systemic symptoms or laboratory findings other than rare peripheral eosinophilia.31 Immunohistochemical analysis revealed the majority of lymphocytes to be helper T-cell lymphocytes. The cells stained strongly with antibodies against IL-2 receptor and with antibodies to lymphocyte functional antigen-1 α. Overlying keratinocytes did not stain with intracellular adhesion molecule-1 or HLA-DR, which may account for the presence of the grenz zone in GF. These findings suggest that a γ–interferon-mediated process may play some role in the pathogenesis of this disorder.32

GF is known to be resistant to therapy. Numerous physical modalities and medical therapeutics have been tried. Laser therapy, including the CO2,33 argon,34 pulsed dye, and long-pulsed tunable dye lasers,35-38 all have been attempted with varying success. A study showed that lesions treated with a CO2 laser and dermabrasion had a more even texture compared with lesions treated with electrosurgery alone. Healing times were similar between lesions treated with electrosurgery and CO2 laser; however, lesions treated with dermabrasion healed more quickly.39 Studies of patients treated with an argon laser resulted in total resolution of plaques of GF but had a remaining white collagenous scar.34

A case report by Elston37 showed complete resolution of 3 lesions of GF when treated with a pulsed dye laser after the patient failed topical corticosteroids and oral dapsone. A case report by Ammirati et al35 of a patient treated with the 585-nm pulsed dye laser showed clinical eradication of the lesion at 6-year follow-up. Another report by Welsh et al36 showed good results when GF was treated with the pulsed dye laser. Recently, the long-pulsed tunable dye laser was used successfully with no scarring.38 Other modalities that have been used include surgical excision,7,10 dermabrasion, superficial ionizing radiation,6,10 topical psoralen plus UV light,40 cryosurgery,41 intralesional corticosteroids,42 combined cryosurgery and intralesional steroid injection,43 and electrodesiccation.10,39

Medical treatment has included intralesional gold, colchicine, isoniazid, corticosteroids, potassium arsenite, testosterone, antimalarials, dapsone, and clofazimine.9,10,25,27,34,39,44-46 Most medical therapies have shown varying success. No controlled trials are available because of the rarity of the condition.

In our review, most patients were treated with topical and intralesional steroids with varying results that ranged from mild improvement to complete resolution. Because there is a lack of scarring with steroid therapy, we recommend this as a good first-line therapy. Although none of the patients were treated with pulsed dye laser therapy, review of the literature demonstrates favorable results with this treatment modality. Pulsed dye laser should be considered as an alternative therapy.

Granuloma faciale (GF) is an uncommon inflammatory condition characterized by reddish brown papules and plaques that usually involve the facial area. Extrafacial lesions are rare. Histologically, the lesions are marked by leukocytoclastic vasculitis and extensive fibrin deposition. There are a variety of treatment options available for GF.

Granuloma faciale (GF) is an uncommon disease of unknown etiology presenting with asymptomatic cutaneous papules, nodules, and plaques of the face. However, there has been some confusion and evolution of the term granuloma faciale. In 1937, the term eosinophilic granuloma of the skin was used to describe ulcerated lesions of the mouth and anus that were associated with tuberculosis.1,2 Cobane et al3 used the term facial granuloma with eosinophilia. Confusion continued until Lever and Leeper4 subdivided eosinophilic granuloma of the skin into 3 groups. The first group consisted of cases with cutaneous lesions associated with and histologically identical to eosinophilic granuloma of the bone (Langerhans cell histiocytosis). The second group included cases with cutaneous lesions consisting of torpid, asymptomatic purplish patches on the face. The last group was cases with a preponderance of eosinophils in a granulomatous infiltrate that were associated with a variety of diseases.1,4 A case described by Wigley5 in 1945 as Sarcoid of Boeck may, in fact, be the first reported case of GF.6 It was Pinkus,7 however, who in 1952 finally suggested the term granuloma faciale.

GF is characterized by papules, nodules, and plaques that are usually solitary, though multiple or disseminated lesions can be seen.8 The lesions usually are soft, elevated, well-circumscribed nodules ranging in size from millimeters to centimeters.3 The lesions can be shades of red, brown, or purple and may darken with sun exposure. Lesions usually have a smooth surface with follicular accentuation but may have telangiectasias. Ulceration or crusting rarely occurs. These lesions usually develop very slowly and remain unchanged, though occasionally they will involute.6 Tenderness, burning, and pruritus also have been reported.9,10

Although extracutaneous involvement is rare, there are reports of oral mucosa/upper airway involvement. GF can occur in patients of any age, sex, or race but is usually found in middle-aged Caucasian men.1,6,11-13 

Materials and Methods

The clinical presentation and treatment of 38 patients with GF was reviewed. Patient information was obtained from a database covering the past 10 years of patient activity at a large dermatopathology laboratory in Michigan. GF was diagnosed by clinical and histologic criteria including an infiltrate composed of lymphocytes, neutrophils, histiocytes, and eosinophils in the superficial and deep dermis and fibrin deposition within dermal blood vessels confirming the presence of vascular damage. The patients were reviewed according to demographics, location, treatment, and prognosis. Follow-up was obtained through a questionnaire sent to each referring physician. To our knowledge, this review represents the largest series of patients with GF ever described. 

Results

The clinical features of the patients are summarized in the Table. Their ages ranged from 28 to 85 years, with a mean age of 51.9 years for men and 55.4 years for women. The study consisted of 24 men and 14 women, which is in concordance with an early study showing a predominance of GF in men.1 One man and one woman (patients 1 and 7) had multiple lesions. Interestingly, five men (patients 19, 22, 25, 29, and 38) had extrafacial lesions; there are very few reported cases of extrafacial GF.

Comment

GF is characterized by solitary nodules and plaques.1 Most lesions are asymptomatic red, brown, or purple nodules that are soft, elevated, well-circumscribed, and slow to develop (Figure 1).

Lesions usually are seen on the face but may occur elsewhere.14 Particularly, lesions may appear on light-exposed areas, and some lesions are photoexacerbated.10 Sites of predilection are the sides (30%) and tip (7%) of the nose, preauricular area (22%), cheeks (22%), forehead (15%), and helix of the ear (4%).10,15 In a clinical and histopathologic review, Pedace and Perry1 described 21 cases of GF seen at the Mayo Clinic from 1945 to 1965. They noted a predominance of the lesions on the face (nose, forehead, malar, preauricular and postauricular areas, and chin) and less frequently on the forearms and elsewhere.

A review of the literature reveals 14 reported cases of extrafacial involvement.14-18 The clinical aspects of facial and extrafacial lesions are similar.8 Extrafacial lesions usually are found on the trunk and proximal extremities.19 Extrafacial lesions have been reported as isolated findings and in conjunction with facial lesions.17,18 The infrequent reports of extrafacial lesions may reflect either the inconspicuous nature of extrafacial lesions or a failure to examine patients specifically for the lesions.8

Extrafacial GF may be difficult to differentiate from erythema elevatum diutinum (EED). In fact, some authors believe EED and GF represent different parts of the spectrum of the same disease.20 Both are rare chronic forms of cutaneous small vessel vasculitis and may share some pathogenic mechanisms, but there are several clinical and histologic differences. EED is characterized by multiple lesions localized on the extensor surfaces of extremities in an acral, bilateral, and symmetrical distribution. Bulla formation and hemorrhagic crusting may be seen. The trunk is usually spared, and facial lesions are rare. Histopathologic features of EED include a dense superficial and deep polymorphous dermal infiltrate where neutrophils are prominent and eosinophils are scanty or absent. A grenz zone of normal collagen beneath the epidermis rarely exists and the epidermis is not always spared. EED may be associated with systemic conditions, primarily gammopathies. EED shows an excellent response to dapsone.15,21,22

There is a relatively large clinical differential diagnosis for GF including lupus erythematosus, polymorphous light eruption, fixed drug eruption, benign and malignant lymphoid proliferations, sarcoidosis, granuloma annulare, foreign body reaction, tinea faciei, insect bite reaction, juvenile xanthogranuloma, mastocytoma, Spitz nevus, EED, mycosis fungoides, basal cell carcinoma, histiocytosis X, and rosacea.1,6,23-26

There is one case report of Trichophyton rubrum causing histologic changes similar to GF.26 There also has been a case of GF mimicking rhinophyma.27

Histologic findings of GF show a normal epidermis that may be thinned and flattened by underlying infiltrate (Figure 2). There is a narrow grenz zone between the epidermis and the dermal inflammatory infiltrate consisting of lymphocytes, eosinophils, and neutrophils with leukocytoclasis (Figure 3). The infiltrate usually is distributed diffusely in the upper two thirds of the dermis. Fibrin deposition around blood vessels is evidence of vasculitis (Figure 4). In the later fibrotic stage, perivascular fibrin deposition predominates, and the number of inflammatory cells is greatly reduced.6

Microscopically, the primary differential diagnosis is EED, insect bite reaction, cutaneous lymphoma, or leukocytoclastic vasculitis. The exact pathogenesis is unclear, but some consider it a variant of leukocytoclastic vasculitis. Immunoglobulins, fibrin, and complement can be found at the dermal-epidermal junction and around blood vessels on direct immunofluorescence.28-30

GF usually lacks systemic symptoms or laboratory findings other than rare peripheral eosinophilia.31 Immunohistochemical analysis revealed the majority of lymphocytes to be helper T-cell lymphocytes. The cells stained strongly with antibodies against IL-2 receptor and with antibodies to lymphocyte functional antigen-1 α. Overlying keratinocytes did not stain with intracellular adhesion molecule-1 or HLA-DR, which may account for the presence of the grenz zone in GF. These findings suggest that a γ–interferon-mediated process may play some role in the pathogenesis of this disorder.32

GF is known to be resistant to therapy. Numerous physical modalities and medical therapeutics have been tried. Laser therapy, including the CO2,33 argon,34 pulsed dye, and long-pulsed tunable dye lasers,35-38 all have been attempted with varying success. A study showed that lesions treated with a CO2 laser and dermabrasion had a more even texture compared with lesions treated with electrosurgery alone. Healing times were similar between lesions treated with electrosurgery and CO2 laser; however, lesions treated with dermabrasion healed more quickly.39 Studies of patients treated with an argon laser resulted in total resolution of plaques of GF but had a remaining white collagenous scar.34

A case report by Elston37 showed complete resolution of 3 lesions of GF when treated with a pulsed dye laser after the patient failed topical corticosteroids and oral dapsone. A case report by Ammirati et al35 of a patient treated with the 585-nm pulsed dye laser showed clinical eradication of the lesion at 6-year follow-up. Another report by Welsh et al36 showed good results when GF was treated with the pulsed dye laser. Recently, the long-pulsed tunable dye laser was used successfully with no scarring.38 Other modalities that have been used include surgical excision,7,10 dermabrasion, superficial ionizing radiation,6,10 topical psoralen plus UV light,40 cryosurgery,41 intralesional corticosteroids,42 combined cryosurgery and intralesional steroid injection,43 and electrodesiccation.10,39

Medical treatment has included intralesional gold, colchicine, isoniazid, corticosteroids, potassium arsenite, testosterone, antimalarials, dapsone, and clofazimine.9,10,25,27,34,39,44-46 Most medical therapies have shown varying success. No controlled trials are available because of the rarity of the condition.

In our review, most patients were treated with topical and intralesional steroids with varying results that ranged from mild improvement to complete resolution. Because there is a lack of scarring with steroid therapy, we recommend this as a good first-line therapy. Although none of the patients were treated with pulsed dye laser therapy, review of the literature demonstrates favorable results with this treatment modality. Pulsed dye laser should be considered as an alternative therapy.

We report the case of a 43-year-old woman with basal cell carcinoma (BCC) of the umbilicus. Although BCC is a common skin tumor, only 2 cases of BCC arising within the umbilicus have been reported previously. Our review of the literature shows that truncal BCCs frequently develop in younger patients, often grow larger, and are associated with an increased risk for developing multiple nonmelanoma skin cancers. Therefore, we advocate a low threshold for performing biopsies on umbilical lesions that are atypical in appearance, course, or response to therapy.

We report the case of a 43-year-old woman with basal cell carcinoma (BCC) of the umbilicus. Although BCC is a common skin tumor, only 2 cases of BCC arising within the umbilicus have been reported previously. Our review of the literature shows that truncal BCCs frequently develop in younger patients, often grow larger, and are associated with an increased risk for developing multiple nonmelanoma skin cancers. Therefore, we advocate a low threshold for performing biopsies on umbilical lesions that are atypical in appearance, course, or response to therapy.

We report the case of a 43-year-old woman with basal cell carcinoma (BCC) of the umbilicus. Although BCC is a common skin tumor, only 2 cases of BCC arising within the umbilicus have been reported previously. Our review of the literature shows that truncal BCCs frequently develop in younger patients, often grow larger, and are associated with an increased risk for developing multiple nonmelanoma skin cancers. Therefore, we advocate a low threshold for performing biopsies on umbilical lesions that are atypical in appearance, course, or response to therapy.