Cutaneous Tuberculosis: a Twenty-Year Prospective Study
Total Page:16
File Type:pdf, Size:1020Kb
INT J TUBERC LUNG DIS 3(6):494–500 © 1999 IUATLD Cutaneous tuberculosis: a twenty-year prospective study B. Kumar, S. Muralidhar Department of Dermatology, Venereology and Leprology, Postgraduate Institute of Medical Education and Research, Chandigarh, India SUMMARY SETTING: A tertiary care hospital in northern India. the spectrum of cutaneous tuberculosis (22.1%), but OBJECTIVE: To study the patterns of clinical presenta- was seen more often in the presence of gumma and tion of cutaneous tuberculosis, to correlate them with scrofuloderma. There were more unvaccinated individ- Mantoux reactivity and BCG vaccination status, and to uals in the group with disseminated disease (80.3%) suggest a clinical classification based on these factors. than in those with localised disease (65.5%). DESIGN: Analysis of the records of patients with cuta- CONCLUSIONS: Lupus vulgaris was the most common neous tuberculosis who attended the hospital between clinical presentation, followed by scrofuloderma, tuber- 1975 and 1995. culids, tuberculosis verrucosa cutis and tuberculous RESULTS: A total of 0.1% of dermatology patients had gumma. Some patients presented more than one clinical cutaneous tuberculosis. Lupus vulgaris was the com- form of the disease. Classification of cutaneous tuber- monest form, seen in 154 (55%) of these patients, fol- culosis needs to be modified to include smear-positive lowed by scrofuloderma in 75 (26.8%), tuberculosis ver- and smear-negative scrofuloderma apart from the inclu- rucosa cutis in 17 (6%), tuberculous gumma(s) in 15 sion of disseminated disease. The presence of regional (5.4%) and tuberculids in 19 (6.8%). No correlation lymphadenopathy serves as a clinical indicator of dis- was found between Mantoux reactivity and the extent of seminated disease. Patients with disseminated disease disease (localised disease 63.6%, disseminated disease were less likely to have been BCG-vaccinated than those 67.9%). The presence of regional lymphadenopathy with localised disease. was an indication of dissemination of the disease (loca- KEY WORDS: cutaneous tuberculosis; disseminated tuber- lised disease 34.7%, disseminated disease 71.7%). Dis- culosis; Mantoux reactivity; BCG vaccination; treatment semination of the disease was observed in the whole of protocol TUBERCULOSIS, which is known to have existed We present a clinical analysis of patients with since 3000 BC, continues to pose a significant public cutaneous tuberculosis seen during the last twenty health problem even today, and kills around 3 million years. An attempt has also been made to extrapolate people annually.1 The emergence of the human immuno- its clinical spectrum to reach a more practical clinical deficiency virus (HIV) has led to a 20% increase in the classification. incidence of extra-pulmonary tuberculosis in the US.2 As cutaneous tuberculosis constitutes 1.5% of all MATERIALS AND METHODS cases of extra-pulmonary tuberculosis,3 it is natural that its incidence will increase proportionately. Clas- Patients with cutaneous tuberculosis seen in the Der- sification of a disease such as tuberculosis should matology Department of the Postgraduate Institute of reflect its immunopathogenesis, and should be able to Medical Education and Research (PGIMER), Chan- convey information on prognosis and aid the clinician digarh, India, between 1975 and 1995 were included in selecting an appropriate drug schedule. in the study. All the patients were also examined by Since 1896, when Darier4 classified cutaneous the senior author. The relevant details were recorded tuberculosis into true tuberculosis and tuberculids, on a predesigned proforma. Apart from a complete several attempts have been made by different workers physical examination, the haematological, hepatic to classify the disease based on the mode of spread of and renal functional profiles were analysed, in addi- the infection3 or on the rate of healing.5 Although tion to screening for evidence of concomitant tuber- each has its own merit, they do not fulfil all the needs culosis elsewhere in the body. Other investigations of the clinician. included chest X-ray, sputum examination, deposit of Correspondence to: Professor B Kumar, Head, Department of Dermatology, Venereology and Leprology, Postgraduate Institute of Medical Education and Research, Chandigarh 160-012, India. Tel: (ϩ91) 172 747 610 24. Fax: (ϩ91) 172 744 401/745 078. e-mail: [email protected] Article submitted 17 November 1997. Final version accepted 6 January 1999. Cutaneous tuberculosis: a 20-year prospective study 495 24 hour urine, and fine needle aspirate (FNAC) of Of the 19 patients who had tuberculids, respec- lymph nodes for acid-fast bacilli (AFB) by smear tively six, 10 and three had papulo-necrotic tubercu- and culture. Wherever indicated, barium meal follow- lids, erythema nodosum and lichen scrofulosorum. A through or enema, endometrial curettage or cerebrospi- small proportion of patients had more than one form nal fluid (CSF) examination were also undertaken. of the disease: lupus vulgaris and scrofuloderma were Biopsy was taken from the most prominent lesion seen in eight patients, lupus vulgaris and tuberculosis and stained with haematoxylin and eosin and for verrucosa cutis in two, and scrofuloderma and tuber- demonstration of AFB. Mantoux testing was done in culous gumma in one patient. all patients using 1 TU (0.02 mg) of PPD and read at There were 157 (56.1%) male and 123 (43.9%) 72 hours. Induration у10 mm was taken as sugges- female patients, with a male:female ratio of 1.27:1. tive of infection with Mycobacterium tuberculosis. Cor- Tuberculous gumma(s) were seen more often in females, relation between various parameters such as lympha- but all other types were more common in male denopathy and the dissemination of disease was done patients. TVC and tuberculids were equally distrib- by the 2 test. uted over the ages 20–60. Miliary cutaneous tuber- Patients were classified according to the morphol- culosis was seen in two women aged 50 and 17 years, ogy of lesions described in the standard texts.3 Wher- respectively. The mean age of male patients with ever any systemic organ (lungs, liver, spleen, central lupus vulgaris was 31.7 Ϯ 15.8 years, scrofuloderma nervous system [CNS], bone, etc.) was involved, the 26.5 Ϯ 15.02 years, tuberculous verrucosa cutis 27.8 Ϯ disease was classified as disseminated. 19.2 years, tuberculous gumma 18.8 Ϯ 2.9 and tuber- culids 33.4 Ϯ 18.6 years. The corresponding mean ages for female patients were 31.2 Ϯ 20.63, 23.4 Ϯ RESULTS 13.7, 31.5 Ϯ 23.3, 37.2 Ϯ 18.29 and 36.1 Ϯ 13.2, Of the 267 420 patients who attended the Dermatol- respectively. Most of the patients with scrofuloderma ogy Out-Patient Department (OPD) between 1975 and (98.6%), tuberculous gumma (92.8%) and tubercu- 1995, 280 had cutaneous tuberculosis, giving a preva- lids (72.2%) had sought medical attention within 5 lence rate of 0.1% among dermatology patients. years of onset. However, nearly one third of the With the exception of tuberculous chancre, all patients with lupus vulgaris (33.6%) and TVC forms of cutaneous tuberculosis were seen. Lupus (28.6%) had had the disease for more than 5 years vulgaris was the commonest form, seen in 154 patients before presenting for treatment. (55%), followed by scrofuloderma (SFD) in 75 In the investigation the erythrocyte sedimentation (26.8%), tuberculosis verrucosa cutis (TVC) in 17 rate (ESR) was elevated (Ͼ30 mm in the first hour by (6.%), tuberculous gumma(s) in 15 (5.4%) and tuber- the Westergren method) in 61 (21.8%) patients, and culids in 19 (6.8%) (Table 1). 14 (5%) patients were anaemic (HB Ͻ 10 gms%). Four patients developed lupus vulgaris following The Mantoux test was positive (Ͼ10 mm) in 170 bacille Calmette Guérin (BCG) vaccination, two pa- patients (66.7%) and negative in 85 (33.3%) for tients with scrofuloderma had dacrocystitis and one whom information was available. There was no sig- had dactylitis. Two laboratory workers handling tu- nificant difference in the incidence of Mantoux reac- bercle bacilli developed TVC on the palms after acci- tivity between those with localised or disseminated dental inoculation. Of the 15 patients with tubercu- disease (63.6% and 67.9%, respectively, Table 2). lous gumma, two had miliary tuberculosis and one Among the 272 patients for whom there was clin- orificial tuberculosis. Apart from the 15 patients with ical information on BCG vaccination status, 86 tuberculous gumma, 47 other patients had systemic patients had been BCG vaccinated and the remaining involvement; these 62 patients were classified as hav- 186 had not (Table 3). There were relatively more ing disseminated tuberculosis. unvaccinated patients with disseminated tuberculosis Table 1 Age distribution of patients with cutaneous tuberculosis Tuberculosis Age group Lupus verrucosa Tuberculous (years) vulgaris Scrofuloderma cutis gumma(s) Tuberculids Ͻ10 13 24 0 0 0 11–20 36 20 6 7 3 21–30 44 16 2 4 6 31–40 23 5 4 1 3 41–50 19 7 4 1 4 51–60 10 2 1 0 0 Ͼ60 9 1 — 2 3 Total 154 75 17 15 19 (%) (55) (26.8) (6) (5.4) (6.8) 496 The International Journal of Tuberculosis and Lung Disease Table 2 Mantoux reactivity Tuberculosis Lupus verrocosa Tuberculous Disease spectrum vulgaris Scrofuloderma cutis gumma Tuberculids Total Localised disease Positive 78 30 11 — 15 134 Negative 48 20 0 — 0 68 Disseminated disease Positive 14 15 1 6 0 36 Negative 6 7 0 4 0 17 Positive ϭ Ͼ10 mm; Negative ϭ р10 mm. (80.3%) compared to those with localised disease has hovered at 0.1–0.5% over the last few decades.6,7 (65.6%, P Ͻ 0.05, Table 3). Although the prevalence of cutaneous tuberculosis has Histopathology reports were available from the remained mostly constant, the recent upsurge in prev- records of 224 patients: 119 with lupus vulgaris, 67 alence of HIV infection would lead one to expect an with scrofuloderma, 17 with TVC, 12 with tubercu- increase in its incidence akin to that seen in the US, lous gumma and nine with tuberculids.