Subcutaneous sarcoidosis: Is it a specific subsetof cutaneous sarcoidosis frequently associated

with systemic disease?

Iftikhar Ahmed, MD, and Sujata R. Harshad, MD

Rochester, Minnesota

Background: Skin is involved in 25% of cases of sarcoidosis. The lesions are specific and nonspecificdepending on the presence or absence of granulomas, respectively. Specific lesions are not thought to haveprognostic significance and are not associated with systemic disease.

Objective: We sought to evaluate for the presence or absence of systemic disease in patients withsubcutaneous sarcoidosis.

Methods: With diagnostic criteria of subcutaneous sarcoidosis, 33 cases were identified in the literatureand 21 cases in our institutional database. A retrospective clinical and pathologic review of these cases wasconducted.

Results: Subcutaneous sarcoidosis is characterized by a peak incidence during the fourth decade; femalepredisposition; asymptomatic to slightly tender lesions typically involving the upper extremities; cutaneouslesional clustering and multiplicity; autoimmune disease associations at time of diagnosis in a subset ofcases; systemic disease associations at diagnosis in most patients, typically consisting of bilateral hilaradenopathy; and a favorable response to oral corticosteroid therapy.

Limitations: Retrospective analysis with inadequate documentation of therapeutic regimens and theirresponses in some cases is a limitation of this study.

Conclusions: The confirmatory diagnosis of subcutaneous sarcoidosis depends on identifying pannicularnoninfectious sarcoidal or epithelioid granulomas with minimal lymphocytic inflammation. Subcutaneoussarcoidosis is the only specific subset of cutaneous sarcoidosis frequently associated with systemic disease.( J Am Acad Dermatol 2006;54:55-60.)

Sarcoidosis is a multisystem granulomatousdisorder of unclear origin.1 Cutaneous involve-ment is seen in 25% of cases typically at disease

outset.1 Skin lesions are heterogeneous and classi-fied as specific and nonspecific on the basis of thepresence or absence of sarcoidal granulomas,

From the Department of Dermatology, Mayo Clinic.

Dr Harshad was a Visiting Clinician and is currently on faculty at

St John’s Medical College, Bangalore, India.

Funding sources: None.

Conflicts of interest: None identified.

Accepted for publication October 2, 2005.

Reprints not available from the authors.

Correspondence to: Iftikhar Ahmed, MD, Department of Der-

matology, Mayo Clinic, 200 First St SW, Rochester, MN 55905.

Published online December 5, 2005.

0190-9622/$32.00

ª 2005 by the American Academy of Dermatology, Inc.

doi:10.1016/j.jaad.2005.10.001

respectively.2 It is thought that specific skin lesionsdo not have prognostic significance and do notcorrelate with the presence of systemic disease.3,4

Herein, we present a retrospective review of pub-lished cases of sarcoidosis in which we identifiedsubcutaneous sarcoidosis on the basis of the diag-nostic criteria of Vainsencher and Winkelmann.5 Wealso present a retrospective review of the singlelargest series of cases of subcutaneous sarcoidosisdiagnosed and managed at our institution. Ourobservations conclusively demonstrate the strongassociation between subcutaneous lesions and sys-temic disease in sarcoidosis.

METHODSAfter receiving approval from our institutional

review board, our dermatopathology files werereviewed for the years 1966 to 2001 for all cases

55

J AM ACAD DERMATOL

JANUARY 2006

56 Ahmed and Harshad

filed under the diagnoses of sarcoidosis, nonspecificgranulomas, and granulomatous panniculitis.Routine hematoxylin and eosinestained sections ofall cases and any available specially stained tissuesections (Gram, Fite, acid-fast, and methenaminesilver stains) were examined. The selection criteriafor cases of subcutaneous sarcoidosis were basedon the criteria of Vainsencher and Winkelmann,5

including the identification of epithelioid or sarcoi-dal granulomaswithminimal lymphocytic inflamma-tion involving predominantly the panniculus, withnegligible involvement of the hypodermis (Fig 1).None of the cases selected showed evidence ofmicro-organisms (bacterial, fungal, ormycobacterial)either on tissue culture or with histologic specialstains. In addition, all the cases were negative forextraneous material on polarization microscopy ofroutine sections. A total of 21 cases were identified.The clinical and laboratory data of these patientswere recorded. None of the selected cases have beenreported previously from this institution.

Fig 1. Epithelioid, or sarcoidal, granulomas with minimallymphocytic inflammation involving panniculus. Minimaldermal involvement is acceptable for diagnosis of sub-cutaneous sarcoidosis as long as panniculus is involved.(Hematoxylin-eosin stain; original magnification 3100.)

Fig 2. Multiple skin-colored, indistinct, subcutaneousnodules on back and ventral aspects of right forearm.Lesions had firm palpable quality. At patient evaluation,differential diagnosis included multiple lipomas.

RESULTSThe 21 patients whose cases were identified in

our database included 15 women and 6 men, with afemale:male ratio of 2.5:1. All the patients exceptone (from Latin America) were white. The mean agewas 46.3 years (range: 29-61 years). The duration ofdisease was known for 20 patients and was from1 week to 10 years at the time of diagnosis. A total of18 patients (90%) had disease duration longer than1 year. The family history was noncontributory for allpatients, and no obvious preceding cause was iden-tifiable in any patient.

The skin lesions were typically asymptomatic (12patients [57%]) to mildly tender (9 patients [43%])(Figs 2 and 3). The lesions were the color of normalskin in 6 of 20 patients (30%), erythematous in 10(50%), violaceous in 2 (10%), and hyperpigmented in2 (10%). Lesional multiplicity was usually observed([6 lesions in 18 of 21 patients [86%]). In 14 of these18 patients (78%), the dimensions of individuallesions were between 1 and 4 cm. Lesions werecharacteristically located on the extremities andusually bilateral and asymmetric (upper extremityin 21 patients, lower extremity in 16, trunk in 6,buttock in 2, and forehead in 1). In 20 of the 21patients (95%), more than one anatomic site wasinvolved, and all these patients had lesions on anupper extremity. Other types of lesions of sarcoido-sis coexisted with the subcutaneous lesions in 15patients (71%) and included plaques (6 patients),papules (4 patients), erythema nodosum (4 patients),and scar sarcoid (1 patient). In 16 of the 20 patients(80%) who had a systematic evaluation, a systemicdisease component was recognized at the timesubcutaneous sarcoidosis was diagnosed (Table I).All 16 of these patients had pulmonary involvement,which was documented with chest radiography.Bilateral hilar adenopathy was identified in 15 ofthe 16 patients (94%). In 6 of these cases,

Fig 3. Solitary large skin-colored subcutaneous noduleoverlying elbow joint. Lesion could be moved over jointand had firm palpable quality. At patient evaluation,differential diagnosis included inflamed bursa.

J AM ACAD DERMATOL

VOLUME 54, NUMBER 1

Ahmed and Harshad 57

Table II. Association between abnormallaboratory test results and abnormal chestradiographs in patients with subcutaneoussarcoidosis

Laboratory test

No. of patients with

abnormal (elevated)

result/total no. of

patients tested (%)

Abnormal chest

radiograph, no.

of patients

Angiotensin-convertingenzyme

3/11 (27) 3

Calcium 3/18 (17) 2Erythrocyte

sedimentation rate7/19 (37) 6

Gammaglobulin 3/11 (27) 1

Cases from Mayo Clinic database.

Table I. Systemic disease identified at outset in16 patients with subcutaneous sarcoidosis

Systemic disease No. of patients

Pulmonary 16Arthritis 3Peripheral neuropathy 2Renal 2Uveitis 2Parotitis 1Dactylitis 1Mucositis (nasal, oral, conjunctival) 3

Cases from Mayo Clinic database.

Table III. Response to treatment of patients with subcutaneous sarcoidosis

Type of response

No. of

patients

Primary treatment

(No. of patients) Secondary treatment (No. of patients)

Complete remission(skin and systemic)

3 NSAIDs (3) None

Complete remission(skin) and partialremission (systemic)

5 Oral corticosteroids (5) Clofazimine 1 methotrexate 1 hydroxychloroquine(Plaquenil) (1); intralesional corticosteroids (1)

Remission (skin) withsubsequent recurrences

4 Oral corticosteroids (4) Methotrexate 1 NSAIDs (1)

Partial remission 2 NSAIDs Dapsone (1)

Cases from Mayo Clinic database.

NSAID, Nonsteroidal anti-inflammatory drug.

paratracheal adenopathy and pulmonary infiltrateswere also present. Pulmonary infiltrates were iden-tified in the one patient with no hilar adenopathy.Involvement of other organ systems was docu-mented subjectively and objectively (Table I).Various autoimmune disease associations werenoted at the outset in 6 of 21 patients (29%) andincluded Hashimoto thyroiditis in two patients,rheumatoid arthritis in two patients, and ulcerativecolitis, systemic lupus erythematosus, and sicca syn-drome in one patient each. Only one patient hadtwo coexisting autoimmune diseases (Hashimotothyroiditis and ulcerative colitis) at diagnosis.Laboratory abnormalities were not usually notedeither at diagnosis or on subsequent follow-up(Table II) and, when noted, they typically occurredin conjunction with abnormal radiographic findings.Adequate follow-up (mean 3.5 years; range 1 month-27 years and 10 months) was available for 14 patients(67%). Of these 14 patients, 12 (86%) had completeremission of the skin lesions (Table III), and for 9 ofthese patients, oral corticosteroids were the primarymodality of therapy. The dosage of corticosteroids

used was from 20 to 40 mg daily, and responses weretypically seen 4 to 8 weeks after the initiation oftherapy.

DISCUSSIONSarcoidosis is a multisystem granulomatous dis-

order of unclear cause.1 It is a global disease thataffects both sexes in all races and age groups. Thepeak incidence typically occurs between the secondand third decades, with a second peak occurring inwomen between the fourth and sixth decades.6 Inthe United States, the annual racial incidence is 10to 14/100,000 for Caucasians and 35.5 to 64/100,000for African Americans, with the highest incidence(107/100,000) among African American women age30 to 39 years.7,8

Cutaneous involvement is seen in 25% of cases ofsarcoidosis, typically at the time of diagnosis.1 Skinlesions are categorized into specific and nonspecificlesions on the basis of the presence or absence ofsarcoidal granulomas, respectively.2 It had beenassumed that specific skin lesions of sarcoidosishave no prognostic significance and are not

J AM ACAD DERMATOL

JANUARY 2006

58 Ahmed and Harshad

Table IV. Features of 33 reported cases of subcutaneous sarcoidosis

Reference Sex/age, y Site of skin lesions Systemic disease

Darier and Roussy12 F/36 Trunk Enlarged LNs (axillary and inguinal)F/43 Trunk Enlarged LNs (axillary and inguinal)

Schaumann13 M/23 Trunk, extremities Enlarged LNs (hilar)Laplane14 F/51 Lower extremity Enlarged LNs (hilar)Gorl15 F/50 Trunk Parotitis, lung infiltrateGougerot et al16 M/70 Trunk, face, extremities Not reportedMaloney and Combes17 M/39 Trunk, buttock, lower extremity Not reported

F/11 Upper extremity Increased pulmonary markingsMarten and Warner18 M/21 Extremities Positive Kveim test, conjunctivitis, enlarged

LNs (hilar)Schirmer19 M/18 Neck, trunk, extremities Enlarged LNs (hilar)Clayton and Wood20 M/41 Face, trunk, extremities Enlarged LNs (hilar)Gross et al*21 M/20 Heak, neck, trunk, extremities Uveitis, hepatosplenomegaly, enlarged

LNs (hilar)Scadding22 F/30 Upper extremity Enlarged LNs (hilar)Vainsencher and Winkelmann5 F/45 Upper extremity Not known

F/54 Extremities Uveitis, parotitis, neuritisM/44 Face, trunk, extremities Enlarged LNs (hilar), parotitis, dactylitis, lacrimal

gland involvementF/50 Upper extremity Enlarged LNs (hilar)

Boyd and Andrews*23 M/24 Extremities Enlarged LNs (hilar and paratracheal)Marzano et al24 F/33 Upper extremity Enlarged LNs (hilar), elevated ACE levelGirao et al25 M/37 Lower extremity Enlarged LNs (hilar and mediastinal), elevated

ACE levelKuramoto et al26 M/24 Lower extremity, buttocks Enlarged LNs (hilar)Weltfriend et al27 F/74 Extremities Enlarged LNs (hilar), elevated ACE levelKroll et al28 F/52 Upper extremity Enlarged LNs (hilar), interstitial pulmonary

infiltrates, parotitis, positive Kveim testGallimore et al29 F/65 Trunk Not reportedHiggins et al30 F/61 Trunk, buttocks Elevated ACE level, positive Kveim test, parotitisCurco et al31 F/60 Extremities Elevated ACE level, dactylitisOhi et al32 M/33 Lower extremity Elevated ACE level, lung granulomasLosada et al33 M/52 Upper extremity, trunk Elevated ACE level, enlarged LNs (hilar)Voelter-Mahlknecht et al34 M/30 Trunk, groin Elevated ACE level, enlarged LNs (mediastinal)Barnadas et al35 F/38 Extremities Abnormal pulmonary function testsMingins et al36 F/70 Trunk Elevated ACE levelRuiz de Erenchun et al37 M/61 Extremities Elevated ACE level, parotitis, enlarged LNs (hilar

and mediastinal)Shidrawi et al38 M/35 Head, neck, extremities Elevated ACE level, enlarged LNs (hilar),

pulmonary interstitial disease

Cases were identified with use of the criteria of Vainsencher and Winkelmann.5

ACE, Angiotensin-converting enzyme; F, female; LN, lymph node; M, male.

*African American.

correlated with the presence of systemic disease.3,4

In contrast, nonspecific lesions such as erythemanodosum have been shown to portend a favorableprognosis, because of the greater likelihood ofspontaneous resolution of the disease when theseskin lesions occur.9,10 Herein, we present data dem-onstrating a strong association between subcutane-ous sarcoidosis (a specific subset of cutaneoussarcoidosis) and systemic disease.

The skin lesions in sarcoidosis are protean andcommonly include macules, papules, nodules,

plaques, infiltrative scars, and lupus pernio.11

Atypical presentations include ulcerative, psoriasi-form, verrucous, ichthyosiform, hypopigmented,and erythrodermic type lesions.11 Subcutaneous in-volvement is diagnosed by identifying noninfectioussarcoidal or epithelioid granulomas with minimallymphocytic inflammation involving predominantlythe panniculus.5 With the use of these criteria, weretrospectively identified 33 cases of subcutaneoussarcoidosis reported in the literature (Table IV).5,12-38

The clinical details of these cases, in particular the

J AM ACAD DERMATOL

VOLUME 54, NUMBER 1

Ahmed and Harshad 59

relation to systemic disease, were collated and arepresented herein. In addition, we used the samecriteria to identify 21 cases of subcutaneous sarcoid-osis diagnosed and managed at our institution over adefined period.

Patients identified in the literatureOf the 33 patients with subcutaneous sarcoidosis

we identified in the literature, 17 were female and16 were male (Table IV). The female:male ratio was1:1. The mean age was 42.3 years (range: 11-74years). All the patients except two19,22 were white.The extremities were involved in 26 patients (78.8%)and included the upper and lower extremities in 14,the upper extremity in 7, and the lower extremity in5. Other sites of involvement included the trunk in15 patients; the face, head, and neck in 6; and thebuttocks and groin in 4. More than one anatomic sitewas involved in 19 patients (58%). A systemic diseasecomponent was present at the outset and identifiedin 29 patients (88%) at the time the diagnosis ofsubcutaneous sarcoidosis was established. It con-sisted of lymphadenopathy in 21 patients (hilar in 18,hilar plus other in 3, and other in 3), parotitis in 6,lung infiltrates in 4, uveitis in 2, dactylitis in 2, andconjunctivitis, abnormal pulmonary function tests,lacrimal gland involvement, and hepatosplenomeg-aly in 1 patient each. An increased level of angioten-sin-converting enzyme was observed in 11 of 13patients (85%), usually in the presence of a systemicdisease component.

Patients at our institutionFor the 21 patients with subcutaneous sarcoidosis

who were identified in our institutional database,the mean age (46.3 years) was close to that of thepatients reported in the literature, but in contrast tothe reported patients, our cohort was predominantlyfemale. Similar to the reported patients, most of ourpatients were also white (except for one Hispanicpatient). No preceding causative factor or pertinentfamily history was recorded for any patient. Otherfeatures shared by the reported patients and those inour cohort included the common location of lesionson the extremities, typically the upper extremities,and the association with a systemic disease compo-nent at disease outset, notably bilateral hilar ade-nopathy (Table I). We also observed the coexistenceof other types of lesions (specific and nonspecific) ofcutaneous sarcoidosis in two thirds of patients withsubcutaneous involvement. However, the subcuta-neous lesions represented most of the lesions inthese patients and were the principal reason forreferral to dermatology. The confirmatory diagnosisof subcutaneous sarcoidosis depends on, as it did for

all our patients, the histologic demonstration ofnoninfectious sarcoidal or epithelioid granulomaswith minimal lymphocytic inflammation within thepanniculus. In our cohort, an association of subcu-taneous sarcoidosis with various autoimmune dis-eases (not previously recognized) was noted at theoutset in 6 of 21 patients (29%). Laboratory abnor-malities, including angiotensin-converting enzymelevels, were not frequently identified (Table II), butwhen noted, they were typically found in conjunc-tion with abnormal radiographic findings. Adequatefollow-up (mean: 3.5 years) was available for 14 ofour 21 patients (67%), and therapeutic responses ofat least the skin lesions to oral corticosteroids wereobserved in all the patients for whom corticosteroidswere the primary modality of therapy (Table III). Thedosages of oral corticosteroids varied from 20 to 40mg daily and responses were typically observedwithin 4 to 8 weeks after the initiation of therapy. Inall, 5 patients received treatment with various non-steroidal antiinflammatory drugs and 3 of them hadcomplete remission of skin and systemic lesions,usually from 6 to 8 weeks after treatment was started.Various secondary treatment modalities were usedwhen partial responses of skin and systemic lesionsoccurred with oral corticosteroids and nonsteroidalantiinflammatory drugs or when skin lesions re-curred after having completely regressed with initialtherapy (Table III). However, because the responsesto these secondary therapeutic agents were notclearly documented in our patients, the details ofthese treatments are not described.

SummarySubcutaneous sarcoidosis is a specific subset of

cutaneous sarcoidosis of unclear origin character-ized by: (1) a peak incidence during the fourthdecade; (2) female predisposition; (3) asymptomaticto slightly tender subcutaneous lesions typicallyinvolving the upper extremities; (4) cutaneous le-sional multiplicity and clustering; (5) autoimmunedisease associations at the outset in a subset ofpatients; (6) strong association with a systemic dis-ease component at the outset of disease, notablybilateral hilar adenopathy; (7) noninfectious pannic-ular sarcoidal or epithelioid granulomas with mini-mal lymphocytic inflammation; and (8) a favorableresponse to oral corticosteroid therapy.

REFERENCES

1. Newman LS, Rose CS, Maier LA. Sarcoidosis [published cor-

rection appears in N Engl J Med 1997;337:139]. N Engl J Med

1997;336:1224-34.

2. Burov EA, Kantor GR, Isaac M. Morpheaform sarcoidosis: report

of three cases. J Am Acad Dermatol 1998;39:345-8.

J AM ACAD DERMATOL

JANUARY 2006

60 Ahmed and Harshad

3. Hanno R, Needelman A, Eiferman RA, Callen JP. Cutaneous

sarcoidal granulomas and the development of systemic

sarcoidosis. Arch Dermatol 1981;117:203-7.

4. Mana J, Marcoval J, Graells J, Salazar A, Peyri J, Pujol R.

Cutaneous involvement in sarcoidosis: relationship to sys-

temic disease. Arch Dermatol 1997;133:882-8.

5. Vainsencher D, Winkelmann RK. Subcutaneous sarcoidosis.

Arch Dermatol 1984;120:1028-31.

6. Hillerdal G, Nou E, Osterman K, Schmekel B. Sarcoidosis:

epidemiology and prognosis; a 15-year European study. Am

Rev Respir Dis 1984;130:29-32.

7. Reich JM, Johnson RE. Incidence of clinically identified sar-

coidosis in a northwest United States population. Sarcoidosis

Vasc Diffuse Lung Dis 1996;13:173-7.

8. Rybicki BA, Major M, Popovich J Jr, Maliarik MJ, Iannuzzi MC.

Racial differences in sarcoidosis incidence: a 5-year study in a

health maintenance organization. Am J Epidemiol 1997;

145:234-41.

9. Mana J, Salazar A, Manresa F. Clinical factors predicting

persistence of activity in sarcoidosis: a multivariate analysis

of 193 cases. Respiration 1994;61:219-25.

10. Neville E, Walker AN, James DG. Prognostic factors predicting

the outcome of sarcoidosis: an analysis of 818 patients. Q J

Med 1983;52:525-33.

11. English JC III, Patel PJ, Greer KE. Sarcoidosis. J Am Acad

Dermatol 2001;44:725-43.

12. Darier J, Roussy G. Un cas de tumeurs benigned multiples:

sarcoıdes sous-cutanees ou tuberculides nodulaires hypoder-

miques. Ann Dermatol Syph 1904;5:144-9.

13. Schaumann J. Etude sur le lupus pernio et ses rapports avec

les sarcoıdes et la tuberculose (observation 50). Ann Dermatol

Syph 1916-1917;6:368-73.

14. Laplane ML. Un cas de sarcoıde hypodermique de la jambe.

Bull Soc Fr Dermatol Syph 1921;28:75-9.

15. Gorl P. Ein Beitrag zur Kasuiistik des subcutanen Sarkoids

Darier-Roussy. Arch Dermatol Syph 1924;148:130-41.

16. Gougerot H, Burnier R, Eliascheff O. Tuberculose pernio du nez

et tuberculoses ‘‘sarcoıdes’’ hypodermiques disseminees. Bull

Soc Fr Dermatol Syph 1932;39:218-20.

17. Maloney ER, Combes FC. Darier-Roussy’s sarcoid, with special

reference to its tuberculous etiology. Arch Dermatol Syph

1936;33:709-24.

18. Marten RH, Warner J. Sub-cutaneous nodular sarcoid. Trans

St Johns Hosp Soc 1967;53:160-1.

19. Schirmer D. Subcutaneous nodular sarcoidosis (sarcoid Darier-

Roussy) [in German]. Z Hautkr 1975;50:837-8.

20. Clayton R, Wood P. Subcutaneous nodular sarcoid. Dermato-

logica 1974;149:51-4.

21. Gross MD, Andriacchi F, Gordon R, Maddox D. Nodular

subcutaneous sarcoidosis. Arch Dermatol 1977;113:1442-3.

22. Scadding JG. Sarcoidosis. London: Eyre and Spottiswoode; 1967.

23. Boyd RE, Andrews BS. Sarcoidosis presenting as cutaneous

ulceration, subcutaneous nodules and chronic arthritis.

J Rheumatol 1981;8:311-6.

24. Marzano AV, Gasparini G, Caputo R, Alessi E. Subcutaneous

sarcoidosis following hypophysectomy for pituitary micro-

adenoma inducing Cushing’s disease. Int J Dermatol 1998;

37:798.

25. Girao L, Bajanca R, Feio AB, Apetato M. Systemic sarcoidosis

revealed by the coexistence of scar and subcutaneous sar-

coidosis. J Eur Acad Dermatol Venereol 2000;14:428-30.

26. Kuramoto Y, Shindo Y, Tagami H. Subcutaneous sarcoidosis

with extensive caseation necrosis. J Cutan Pathol 1988;15:

188-90.

27. Weltfriend S, Harth Y, Katz I. Subcutaneous sarcoidosis in a

patient with malignant carcinoid tumor of the colon. J Am

Acad Dermatol 1989;20:507-8.

28. Kroll JJ, Shapiro L, Koplon BS, Feldman F. Subcutaneous

sarcoidosis with calcification. Arch Dermatol 1972;106:894-5.

29. Gallimore AP, George CD, Lampert IA. Subcutaneous sarcoid-

osis mimicking carcinoma of the breast. Postgrad Med J

1990;66:677-8.

30. Higgins EM, Salisbury JR, Du Vivier AW. Subcutaneous sar-

coidosis. Clin Exp Dermatol 1993;18:65-6.

31. Curco N, Pagerols X, Vives P. Subcutaneous sarcoidosis with

dactylitis. Clin Exp Dermatol 1995;20:434-5.

32. Ohi T, Saijo S, Tagami H. Coexistence of subcutaneous

sarcoidosis of the sole and scar sarcoidosis. Acta Derm

Venereol 1996;76:500.

33. Losada A, Garcia-Doval I, de la Torre C, Cruces MJ. Sub-

cutaneous sarcoidosis worsened by cyclosporin treatment

for pyoderma gangrenosum. Br J Dermatol 1998;138:1103-4.

34. Voelter-Mahlknecht S, Benez A, Metzger S, Fierlbeck G. Treat-

ment of subcutaneous sarcoidosis with allopurinol. Arch

Dermatol 1999;135:1560-1.

35. Barnadas MA, Rodriguez-Arias JM, Alomar A. Subcutaneous

sarcoidosis associated with vitiligo, pernicious anemia and

autoimmune thyroiditis. Clin Exp Dermatol 2000;25:55-6.

36. Mingins C, Williams MR, Cox NH. Subcutaneous sarcoidosis

mimicking breast carcinoma. Br J Dermatol 2002;146:924-5.

37. Ruiz de Erenchun F, Vazquez-Doval FJ, Idoate M, Leache A,

Quintanilla E. Subcutaneous nodules as the first clinical

manifestation of sarcoidosis. Clin Exp Dermatol 1992;17:192-4.

38. Shidrawi RG, Paradinas F, Murray-Lyon IM. Sarcoidosis pre-

senting as multiple subcutaneous nodules. Clin Exp Dermatol

1994;19:356-8.