Abstract
An 82-year-old woman presented with a four-month history of an ulcerative plaque overlying her left neck. This lesion had developed as a subcutaneous nodule, gradually increased in size, and evolved into ulcers. Before visiting our Dermatology clinic, the patient had been diagnosed as having a bacterial abscess, but treatments with antibiotics were unsuccessful. The presence of a purulent discharge and prominent ulceration caused further confusion as bacterial abscess, and radiologic evaluation on computed tomography also led to the possibilities of secondary lesions from an abscess or malignancy. However, the characteristic appearance of her lesion allowed us to discern cutaneous tuberculosis, especially scrofuloderma. Based on clinical examinations, staining for acid-fast bacilli, and positive findings of polymerase chain reaction, a quick diagnosis of scrofuloderma was made. After that, she was treated successfully with anti-tuberculosis therapy and the ulcer healed. Our case highlights the problem of delayed diagnosis of scrofuloderma presenting as a bacterial abscess. In conclusion, having a high index of suspicion is needed to diagnose cutaneous tuberculosis correctly.
Cutaneous tuberculosis is a rare form of extrapulmonary tuberculosis, and it is often misdiagnosed because of a low index of suspicion1. Scrofuloderma results from the involvement and break-down of the skin overlying a tuberculous focus2. It begins as a painless subcutaneous nodule, called a cold abscess. With the continuous propagation of an underlying infection, it is characterized by fistula formation and pus discharge from a lymph node. The cervical lymphatic chains are most frequently involved2-4. Herein we report a case of scrofuloderma misdiagnosed as a bacterial abscess and note the importance of maintaining a high level of awareness.
An 82-year-old woman presented with a four-month history of an ulcerative plaque overlying her left neck. This lesion had developed as a subcutaneous nodule, which gradually increased in size and evolved into ulcers. The physical examination revealed a 3×5 cm, erythematous suppurative plaque on the surface of her neck. It was well-demarcated with elevated borders and deep undermined edges. The lesion also had central ulceration covered by a scant, cheesy, and purulent discharge (Fig. 1). There was no history of trauma. Systemic examination did not reveal any abnormalities, and there was no recent weight loss or night sweats.
Before visiting our Dermatology clinic, the patient had been diagnosed as having a bacterial abscess and was treated with antibiotics at two local clinics and by an ENT doctor at our hospital. Earlier therapeutic attempts with antibiotics (Banan; Meiact; cefditoren pivoxil) were unsuccessful and the lesion was sometimes reduced but never had disappeared entirely. Further investigations such as a computed tomography (CT) study of the neck, and fine needle aspiration cytology (FNAC) were done by the ENT doctor of our hospital to rule out malignancy. Also, she was referred to our clinic for the evaluation of chronic dermatologic disease. At that time, the characteristic appearance of her lesion allowed us to discern cutaneous tuberculosis, especially scrofuloderma and further workups including skin biopsy, polymerase chain reaction (PCR) for Mycobacterium tuberculosis, cultures (bacterial, fungal, and mycobacterial), and chest X-ray were carried out.
Histological examinations revealed epithelioid cell granulomas with inflammatory cell infiltrates in the dermis. Ziehl-Neelsen stain demonstrated acid-fast bacilli (Fig. 2). Cultures from the skin biopsy showed growth of Mycobacterium tuberculosis after 8 weeks. Fungal and standard bacterial cultures from the skin biopsy were negative. A PCR study for the detection of Mycobacterium tuberculosis was conducted on a biopsy specimen and turned out positive (Fig. 3). However, FNAC from the skin lesions on the neck showed only neutrophils and fibrin exudates.
A radiograph of the chest revealed 1 cm, small, calcified nodules on the right upper pulmonary lobe, and pulmonary tuberculosis was suspected (Fig. 4A). CT of the neck, which demonstrated peripheral enhancing low density lesions with extension to the cutaneous area, was found and ruled out an abscess or malignancy (Fig. 4B).
Based on clinical examinations, staining for acid-fast bacilli, and positive PCR findings, a rapid diagnosis of scrofuloderma was made. The patient was quickly started on an anti-tubercular treatment regimen that included isoniazid, rifampicin, ethambutol and pyrazinamide. The cutaneous lesion regressed in a slow but complete fashion. Her treatment with anti-tuberculosis therapy was successful and the ulcer healed after 18 months.
Cutaneous tuberculosis remains a rare infection, with an incidence of 3.5% reported among patients with organ tuberculosis5. Among them, scrofuloderma is a common form of cutaneous tuberculosis; it is seen in many developing countries and in some European countries5,6. It is caused by a lesion in the lymph node, bone, muscle or tendon that continuously spreads to the skin2-4.
The skin lesion first presents as painless, subcutaneous nodules, which is called a cold abscess. Then, the reddish-blue induration suppurates and breaks down, forming an ulcer with granulating tissue at the base. Sometimes sinuses may develop and communicate directly with areas of deep infection. Progression and scarring produce irregular adherent masses, densely fibrous in places, and fluctuant or discharging in others2-4. The disease is usually slow-growing, and has an insidious course and an unusual presentation; a chronically discharging sinus or a non-healing ulcer are possible4.
Scrofuloderma remains a diagnostic and therapeutic challenge because the clinical appearance of the lesions is not always characteristic, and a positive culture result is not always obtained1,7. This case highlights delayed diagnosis of scrofuloderma presenting as a bacterial abscess. Although our case presents typical features of scrofuloderma, the presence of purulent discharge, prominent ulceration, or features suggestive of a carbuncle can cause further confusion and misdiagnosis as a bacterial abscess7. Moreover, radiologic evaluation on CT also led to the possibility of secondary lesions from an abscess or malignancy. The delay in the diagnosis emphasizes the importance of maintaining a high level of suspicion.
In our case, scrofuloderma was confirmed by the presence of acid fast bacilli on AFB staining and a positive culture for Mycobacterium tuberculosis. Supportive evidence included the presence of epithelioid granulomas on histopathologic examination and a positive result on PCR for mycobacteria.
Scrofuloderma may be a manifestation of systemic tuberculous disease and can result from direct extension or hematogeneous spread of the infection6. The incidence of systemic involvement in adults with scrofuloderma has been reported to be as high as 35%6. Of note, using PCR to identify the infectious agents has increased the sensitivity and improved the overall detection rate among those with a diagnosis of cutaneous tuberculosis7,8. A further advantage of PCR is the possibility of earlier initiation of treatment7,8.
Histology may show marked caseation necrosis, with numerous bacteria usually in the deeper structures2,9. The center of the lesion is dominated by necrosis and abscess material with inflammatory infiltrates. In the epidermis, scarring and atrophic changes often predominate. Tubercle bacilli are usually easy to identify in the purulent discharge or biopsy tissue3.
Chemotherapy still remains the treatment of choice2,10. It aims to cure the disease as rapidly as possible in order to prevent relapses and the emergence of resistant strains. The standard 6-month regimen for adults includes isoniazid, rifampin, pyrazinamide, and ethambutol for the initial 2 months followed by isoniazid and rifampin for a further 4 months in the continuation phase2,10. When a therapeutic trial is undertaken in a case of cutaneous tuberculosis, 6 weeks of therapy with four drugs appears adequate to confirm the diagnosis11. The response to anti-tuberculous treatment alone as a primary therapy is usually good, but some authors have recommended wide surgical removal of all diseased tissue, combined with pharmacologic anti-tuberculous treatment2,3.
Our case serves as a reminder that chronic neck ulcers may be due to cutaneous tuberculosis and that pulmonary tuberculosis may be a focus for tuberculous infection. A chronic, non-healing ulcer is a common presentation of scrofuloderma. Nevertheless, scrofuloderma with concomitant bacterial infection frequently occur, preventing early diagnosis of tuberculosis if not suspected1,7. Therefore, a high index of suspicion is needed for the prompt diagnosis and management of this uncommon skin disorder.
When a therapeutic trial is undertaken in a case of cutaneous tuberculosis, 6 weeks of therapy with four drugs appears adequate to prove (or disprove) the diagnosis.
Figures and Tables
![]() | Fig. 2Epithelioid cell granulomas with inflammatory cell infiltrations in the dermis (H&E, ×100), Inset: Acid-fast bacilli on Ziehl-Neelsen staining (H&E, ×1,000). |
References
1. Lantos G, Fisher BK, Contreras M. Tuberculous ulcer of the skin. J Am Acad Dermatol. 1988. 19:1067–1072.


2. Tappeiner G. Wolff K, Goldsmith LA, Katz SI, Gilchrest BA, Paller AS, Leffell DJ, editors. Tuberculosis and infections with atypical mycobacteria. Fitzpatrick's dermatology in general medicine. 2008. 7th ed. New York: McGraw-Hill Medical;1768–1778.
3. Barbagallo J, Tager P, Ingleton R, Hirsch RJ, Weinberg JM. Cutaneous tuberculosis: diagnosis and treatment. Am J Clin Dermatol. 2002. 3:319–328.
4. Sehgal VN, Wagh SA. Cutaneous tuberculosis. Current concepts. Int J Dermatol. 1990. 29:237–252.
5. Kivanç-Altunay I, Baysal Z, Ekmekçi TR, Köslü A. Incidence of cutaneous tuberculosis in patients with organ tuberculosis. Int J Dermatol. 2003. 42:197–200.


6. Yates VM, Ormerod LP. Cutaneous tuberculosis in Blackburn district (U.K.): a 15-year prospective series, 1981-95. Br J Dermatol. 1997. 136:483–489.


7. Tan WP, Tang MB, Tan HH. Scrofuloderma from the acromioclavicular joint presenting as a chronic ulcer in an immunocompetent host. Singapore Med J. 2007. 48:e243–e245.
8. Hsiao PF, Tzen CY, Chen HC, Su HY. Polymerase chain reaction based detection of Mycobacterium tuberculosis in tissues showing granulomatous inflammation without demonstrable acid-fast bacilli. Int J Dermatol. 2003. 42:281–286.


9. Vashisht P, Sahoo B, Khurana N, Reddy BS. Cutaneous tuberculosis in children and adolescents: a clinicohistological study. J Eur Acad Dermatol Venereol. 2007. 21:40–47.

