Childhood cutaneous tuberculosis: A 20-year retrospective study in Tunis
Dermatology department of Habib Thameur Hospital, Tunis, Tunisia. email@example.com
Despite prevention programs, tuberculosis is still endemic in developing countries. We assessed the epidemiologic and clinical profiles of childhood cutaneous tuberculosis in our dermatology department from 1981 to 2000 and compared it to previous Tunisian reports and to the relevant literature. This is a retrospective study over a 20-year period (1981-2000) in a large teaching hospital of the capital. Patients included were below age 15 years. Diagnosis was based upon clinical examination, tuberculin reaction, histopathology and response to antitubercular therapy. There were 26 patients with cutaneous tuberculosis, 0.1 percent of the total number of dermatology outpatients for that time period. Of these 26, seven (27 %) were immunocompetent. There were four boys and three girls and the mean age was 9.5 years. Three patients had lupus vulgaris, three had scrofuloderma, and one child had orificial tuberculosis. Six out of seven children were BCG vaccinated. There was no family history of tuberculosis. The Mantoux reaction was positive in six children. There was no systemic organ involvement in all cases. All patients were treated successfully with triple or quadruple anti-tubercular drugs for 4-11 months. Compared to a previous Tunisian report conducted over an 8-year period in the seventies, the incidence of childhood cutaneous tuberculosis has decreased. In that report, scrofuloderma was the most frequent form. Currently the incidence of lupus vulgaris has reached that of scrofuloderma, demonstrating the increase of the clinical pattern associated with strong immunity. All children had localized disease and responded to antimycobcterial chemotherapy.
Extra-pulmonary tuberculosis constitutes about 10 percent of all cases of tuberculosis, and cutaneous tuberculosis makes up only a small proportion of these cases. Despite prevention programs, tuberculosis is still progressing endemically in developing countries.
The aim of our study was to assess the epidemiologic and clinical profiles of childhood cutaneous tuberculosis in a dermatology department of the capital and to compare it to previous Tunisian reports and to the relevant literature.
Patients and methods
We carried out a retrospective study on on cutaneous tuberculosis for the 20-year period from 1981 to 2000. Our hospital is a large urban teaching hospital. Patients included were below age 15 years. A detailed history was taken with particular reference to BCG vaccination and tuberculosis in the family. Clinical features including the cutaneous and systemic examination were recorded. Investigation included hemogram, hepatic function, and renal function. To rule out tuberculosis in other organs, chest X-ray and sputum-smear examination for acid-fast bacilli (AFB) on 3-consecutive days were done for all patients. Skin biopsy was performed in all cases and examined after staining with hematoxylin and eosin. The Mantoux test was performed by intradermal injection of 0.1 ml of purified tuberculin on the volar surface of the forearm; induration was measured after 72 hours.
We identified 26 patients with cutaneous tuberculosis over the 20-year period. These patients formed 0.1 percent of the total dermatology outpatients. Seven (27 %) of these 26 cases were children. Table 1 summarizes the main clinical features of these patients. There were 4 boys and 3 girls (boy to girl ratio of 1.3). The mean age was 9.5 years. The duration of the lesions ranged from 1 month to 2 years. Of the various patterns of cutaneous tuberculosis seen, three children had lupus vulgaris, three children had scrofuloderma of the neck, and one boy had perianal tubercular ulcer (Fig. 1).
In our patients with lupus vulgaris, the face was affected in two cases and the arm in one case. One lesion developed at the site of BCG vaccination (Fig. 2). Six patients had been given BCG vaccination; only one was unvaccinated. There was no family history of tuberculosis and no systemic organ involvement in all cases. All patients were HIV- negative. The Mantoux reaction was positive in six children and varied from 7 to 16 mm in diameter. One patient with scrofuloderma had a negative reaction.
Histopathology revealed tuberculoid granuloma in six children, caseation necrosis was found in two cases. In one case, histpathological report showed nonspecific features in the form of chronic inflammatory infiltrate.
All patients were treated with quadruple antitubercular drugs for 8 months with combined isoniazide and rifampin plus streptomycin and ethambutol or pyrazynamide the first 2 months. The lesions improved on treatment. After a mean follow-up period of 12 months, disfiguring scars were seen in three cases (including atrophic scar, keloid scar, and retractile scar).
Cutaneous tuberculosis represents 1.5 percent of all cases of extrapulmonary tuberculosis . The incidence of childhood cutaneous tuberculosis in an Indian series conducted over 25 years was estimated at 18.7 percent . A recent study from the south part of Tunisia involving 63 patients with cutaneous tuberculosis identified 16 children (25.3 %) . In the present series, 27 percent of the total number of patients with skin tuberculosis were children. These results compared to a previous Tunisian study conducted on 130 patients with cutaneous tuberculosis involving an 8-year period (1970-1977) in which there were 94 children (72 %), (Kamoun MR. Analytic study of 130 cases of cutaneous tuberculosis. Medicine thesis, Tunis, 1980). Taken together, these studies demonstrate that the incidence of childhood cutaneous tuberculosis in Tunisia has decreased. In contrast to the Ramesh series in India , boys outnumbered girls in our study.
Before 1977, scrofuloderma was the most frequent form among the childhood cutaneous tuberculosis in Tunisia. It represents 57 percent of the various patterns; lupus vulgaris represents only 30 percent (Kamoun MR. Analytic study of 130 cases of cutaneous tuberculosis. Medicine thesis, Tunis, 1980). Now the incidence of lupus vulgaris has reached that of scrofuloderma, demonstrating the increase of the clinical pattern of cutaneous tuberculosis associated with strong immunity, a finding probably related to the improvement in health conditions and general availability of vaccination programs countrywide since 1959.
Occasionally lupus vulgaris occurs at the site of BCG vaccination [5, 6, 7], suggesting exogenous inoculation of the infection. In this study, one child developed lupus vulgaris of the arm following inoculation with BCG vaccine (case 1). Misdiagnosis, neglect, or late diagnosis of cutaneous tuberculosis in children may result in extensive disease. In our children, we have no delay in diagnosis nor extensive disease. Because it is a paucibacillary form of tuberculous infection, culture is often negative and the diagnosis is mainly based on the Mantoux test, the histopathologic picture, and the response to chemotherapy. Where facilities permit, the polymerase chain reaction can improve diagnostic accuracy. Treatment is based on triple or quadruple anti-tubercular therapy for minimum 6 months . In contrast to the Ramesh series where 62 percent of patients failed to complete therapy , all affected children in our study underwent systemic antitubercular therapy without interruption. After treatment, disfiguring scars are frequent in children. These scars were seen in half of our patients. All children had localized disease and responded to antimycobacterial chemotherapy.
Cutaneous manifestations of tuberculosis are polymorphous in both children and adults. Scrofuloderma and lupus vulgaris were the commonest forms found in our study. Thanks to the national antituberculosis program (in place since 1959) and the generalization of BCG vaccination, the incidence of childhood cutaneous tuberculosis has decreased from that of the last decade.
References1. Kumar B, Muralidhar S. Cutaneous tuberculosis: A-Twenty-year prospective study. Int J Tuberc Lung Dis 1999;3:494-500.
2. Kumar B, Rai R, Kaur I, Sahoo B, Muralidhar S, Radotra BD. Childhood cutaneous tuberculosis: a study over 25 years from northern India. Int J Dermatol. 2001;40:26-32. PubMed
3. Zahaf A, Boudaya S, Khemakhem M, Bouassida S, Turki H. La tuberculose cutanée de l'enfant. Ann Derm Venereol 2001;128 :1s49.
4. Ramesh V, Misra RS, Beena KR, Mukherjee A. A study of cutaneous tuberculosis in children. Pediatr dermatol 1999; 16 (4) :264-9. PubMed
5. Dangoisse C, Song M. Management of cutaneous complications of BCG vaccine. Ann Dermatol Venereol. 1990;117:45-51. PubMed
6. Vittori F, Groslafeige C. Tuberculosis lupus after BCG vaccination. A rare complication of the vaccination. Arch Pediatr 1996;3: 457-9. PubMed .
7. Bhardwaj P, Mahajan V. Lupus vulgaris. Indian Pediatr 2003; 40:902-3. PubMed
8. Morand JJ, Cuguillière A et Sayag J. Tuberculose cutanée. Encycl Méd Chir (Elsevier, Paris), Dermatologie, 98-360-A-10,1999, 12p.
© 2006 Dermatology Online Journal