DOI: 10.1111/jdv.

14851 JEADV

ORIGINAL ARTICLE

Cutaneous tuberculosis in China – A multicentre

retrospective study of cases diagnosed between 1957 and
2013
J. Zhang,1,† Y.K. Fan,1,† P. Wang,1,† Q.Q. Chen,1 G. Wang,2 A.E. Xu,3 L.Q. Chen,4 R. Hu,1 W. Chen,5,6
Z.Q. Song,1 F. Hao1,*
1
Department of Dermatology and Venereology, Southwest Hospital, Third Military Medical University, Chongqing, China
2
Department of Dermatology and Venereology, Xijing Hospital, Forth Military Medical University, Xi’an, China
3
Department of Dermatology and Venereology, Third People’s Hospital of Hangzhou, Hangzhou, China
4
Department of Dermatology and Venereology, First People’s Hospital of Wuhan, Wuhan, China
5
€rich, Zurich, Switzerland
IZZ Immunologie-Zentrum Zu
6
€t Mu
Department of Dermatology and Allergy, Technische Universita €nchen, Munich, Germany
*Correspondence: F. Hao. E-mail: haofei62@medmail.com.cn

Abstract
Background China has one of the largest populations with tuberculosis worldwide. Cutaneous tuberculosis (CTB) is a
rare manifestation of mycobacterial infection. Although CTB is well described, it is important to periodically revisit the
prevailing clinical and epidemiological features in most populated countries such as China, India, and Indonesia, where
tuberculosis is still a major health problem.
Objective This retrospective study aimed to re-evaluate the CTB cases in China in the past 50 years to obtain a
comprehensive insight into this multiplex entity.
Methods Cases of diagnosed CTB with confirmed histology from four large medical centres in central China between
1957 and 2013 were collected and analysed, including demographic data, clinical manifestations and pathological
findings.
Results Of the 1194 cases enrolled, there were 666 (55.78%) and 528 cases (44.22%) of true CTB and tuberculids,
respectively. Erythema induratum of Bazin (EIB) was the most common CTB (35.8%), followed by lupus vulgaris (LV,
32.7%), tuberculosis verrucosa cutis (18.9%), papulonecrotic tuberculid (8.0%), scrofuloderma (2.8%), tuberculosis cutis
ulcerosa (1.3%), penile tuberculids (0.4%), and lichen scrofulosorum (0.1%). EIB was the predominant tuberculid
(80.87%), while LV the predominant true CTB (58.7%). The number of diagnosed CTB showed a decreasing trend in the
1960s and 1970s, then increased again, and peaked in the 1990s.
Conclusions Cutaneous tuberculosis is still a common problem in China. Chronologic changes in CTB cases reported
in China over the past 50 years may reflect the prevalence transition of overall tuberculosis. CTB has diverse clinical
presentations, and each subtype is characterized by specific gender predilection, duration, age, clinic and pathological
findings.
Received: 1 September 2017; Accepted: 10 January 2018

Conflict of interests:
None declared.

Funding source:
SWH2012LC04

Introduction Mycobacterium bovis.1,2 Different classifications of CTB are gen-

Cutaneous tuberculosis (CTB) is a rare form of extrapulmonary erally based on different infection routes and bacterial load in
mycobacterial infection in 1–2% of all tuberculosis cases, mainly skin lesions. It is most widely accepted to classify CTB into two
caused by Mycobacterium tuberculosis, and occasionally primary categories: true CTB and tuberculids. The former per-
tains to the multibacillary pole of the CTB spectrum, including
†These authors contributed equally to this work. tuberculous chancre, tuberculosis verrucosa cutis (TVC),

JEADV 2018, 32, 632–638 © 2018 European Academy of Dermatology and Venereology
Cutaneous tuberculosis in China 633

scrofuloderma (SFD), lupus vulgaris (LV), acute miliary tuber- Data analysis was performed using Microsoftâ Excel (Office
culosis, tuberculous gumma, tuberculosis ulcerosa cutis (TCU) 2010, Inc., Seattle, WA, USA) and SPSSâ 19.0 software (SPSS
and tuberculous cellulitis, where mycobacteria can be commonly Inc., Chicago, IL, USA). Quantitative data were presented by
detected in the lesional tissues via microscopy, culture or poly-   s, and proportional data were denoted as percentages. Dif-
X
merase chain reaction (PCR). Tuberculids represent the pau- ferences in the ratios and averages between groups were assessed
cibacillary end of the spectrum, in which the skin lesions are by Pearson’s chi-square test and t-test with a two-sided a value
considered to be induced by hypersensitivity reaction to the of 0.05.
mycobacterial antigens lodged in cutaneous blood vessels, and
mainly consist of papulonecrotic tuberculid (PNT), including Results
penile tuberculid (PT), erythema induratum of Bazin (EIB) and
lichen scrofulosorum (LS).1,3 A recently reported entity is super- Distribution of different subtypes of CTB in the cohort
ficial thrombophlebitic tuberculids displaying pathological man- True CTB was diagnosed in 666 (55.8%), while tuberculids were
ifestation of granulomatous infiltrate localized to the veins in the found in 528 (44.2%) of the cohort. Overall, EIB was the most
subcutaneous fat.4 Mycobacterial DNA has been demonstrated common CTB (35.8%), followed by LV (32.7%), TVC (18.9%),
by PCR in biopsied tissues of traditionally defined tuberculids.1 PNT (8.0%), SFD (2.8%), TCU (1.3%), PT (0.4%) and LS
Due to perplexing and diversifying clinical presentations, diag- (0.1%). EIB was the predominant tuberculid (80.9%), while LV
nosis of CTB can pose great challenge for dermatologists in daily the major true CTB (58.7%). In general, the male-to-female
practice, with the misdiagnosis rate reported to be in the range gender ratio was 1:1.4 (502 versus 692), while TVC was
of 33–50%.5 male-predominant and EIB female-predominant (Table 1).
China bears one of the largest disease burdens of tuberculosis
in the world. The Global Tuberculosis Report 2016 by WHO Disease duration and age of the patients
estimated the incidence accounting for 9.1% of the totally The duration of CTB from disease onset to diagnosis varied with
affected population worldwide, and ranking next to India and different subtypes. The longest duration was LV, with a mean
Indonesia.6 Between 1979 and 2010, five national epidemiologi- course of 9.6 years, followed by TCU (7.6 years), TVC
cal surveys of tuberculosis conducted every 5–10 years in China (7.6 years), SFD (5.1 years), EI (3.3 years), PNT (3.1 years) and
revealed a decreasing trend in the overall incidence rate of active PT (2.2 years) (Table 1). Up to 77.6% of the enrolled patients
TB, from 717/100 000 in 1979 to 459/100 000 in 2010. The were between 10 and 49 years of age, with the highest incidence
results were heterogeneous, in terms of geographic areas and rate at age 10–39 years, representing 61.9% of all cases (Fig. 1).
social groups, whereby the incidence rate even increased in some Paediatric cases under age 15 accounted for 6%.
socio-economically underdeveloped provinces in western China.
It is noteworthy that the declining rate in TB incidence has slo- Chronologic changes in CTB
wed down in recent years, with the severity concurrent HIV Data analysis of cases from Chongqing Metropole revealed that
infection/AIDS becoming increasingly prominent.7–11 This study 334 CTB were diagnosed during the first 15 years between 1955
was aimed to focus on CTB in China on the national scale. and 1969, accounting for nearly half of the total 703 cases diag-
nosed at the hospital involved. From 1970 to 1979, due to special
Patients and methods turbulent social settings in China, diagnosis of CTB decreased
Medical records of 1194 CTB cases, diagnosed between 1957 and abruptly, with fragmented and unreliable medical records during
2013 and confirmed by histopathology, were retrieved from four this period. Reported cases of CTB had increased since 1980, but
large medical centres in China, covering a total population of remained at a low level until 2000 (93 cases, 13.2% totally).
approximately 180 millions, including Southwest Hospital of the From 2000 to 2012, 271 (38.6%) cases of CTB were diagnosed
Third Military Medical University, Chongqing, Xijing Hospital (Fig. 2a). A similar ascending trend was found in data from Xi’
of the Fourth Military Medical University, Xi’an, Shaanxi, The Shanxi Province after 1980: Between 1983 and 1997, a total of 67
First Hospital of Wuhan, Hubei, and The Third People’s Hospi- cases of CTB were reported (58 true tuberculosis and nine tuber-
tal of Hangzhou, Zhejiang. The study was approved by the Ethics culids); while 217 cases of CTB were diagnosed in the next
Committees of each hospital involved. Diagnosis of CTB was 15 years between 1998 and 2012 (Fig. 2b).
made in cases with (i) complete clinical records; (ii) typical clini-
cal manifestations confirmed by histology; (iii) success in tracing Clinical features and pathological findings of CTB
the evidence of infectious pathogens by Mantoux test, T-SPOT, subtypes
or culture, and biochemical assay authentication; and (iv) Different subtypes of CTB in our cases demonstrated specific
responsiveness to anti-tuberculosis therapies. Data concerning clinical features that could serve as key diagnosis clues. For
antibiotic susceptibility and treatment follow-up were frag- example, LV occurred mainly in the facial area (53.4%), fol-
mented and could not be analysed. lowed by neck and upper limbs (Figs 3 and 4). TVC was

JEADV 2018, 32, 632–638 © 2018 European Academy of Dermatology and Venereology
634 Zhang et al.

Table 1 Pooled general demographic data of CTB patients from four centres in China

True cutaneous TB (n = 666) Tuberculids (n = 528)

LV TVC SFD TCU EI PNT PT LS
Case 391 (32.7) 226 (18.9) 33 (2.8) 16 (1.3) 427 (35.8) 96 (8.0) 4 (0.3) 1 (0.1)
number
(%)
Mean 39.6  17.9 35.3  15.5 33.7  15.9 38.8  17.2 36.7  15.0 28.2  10.7 50.8  15.8 31
age of (0–89) (5–57) (5–64) (4–66) (6–77) (4–54) (34–71)
patients
(range)
Gender 1:1.7 4:1 1.5:1 1:1 1:4.9 2:1 – Female
ratio
(male:
female)
Mean 9.6  10.3 year 7.6  6.7 year 5.1  8.5 year 7.6  13.3 year 3.3  5.3 year 3.1  4.2 year 2.2  1.2 year 3 year
duration (7 day–50 year) (10 day–30 (1 month–30 (1 month–50 (3 day–36 (10 day–20 (6 month–4
(range) year) year) year) year) year) year)

(a) 100 True CTB

Tuberculids
≥60 80

Case number
LV
50-59 60
TVC
Age groups

40-49 SFD
TCU 40
30-39 EI
PNT 20
20-29 PT
LS 0

2010-13
1965-69
1960-64

1970-74

1980-84
10-19
1955-59

1985-89

2005-09
1990-94

2000-04
1975-79

1995-99
0-9

0 50 100 150 200 250 Time

Case number
True CTB
(b) 80 Tuberculids
Figure 1 Age distribution of patients with CTB and the distribu-
tion of different types of CTB in each age group. 60
Case number

40

featured by plaques (48.3%) and verrucous hyperplasia (42.0%),

20
accompanied by erosions and ulcers, and displaying predilection
for extremities such as feet (35.4%), upper limbs, and gluteal 0
1983-87
1978-82

1988-92

1998-02
1973-77

2003-07
1993-97

2008-12

areas (33.2%). SFD was characterized by ulcers, sinuses, plaques,

subcutaneous nodules, and sometimes masses, which usually
emerged on the upper trunk (28.1%) and extremities (upper Time
limbs 28.1%, lower limbs 25.0%). TCU commonly involved
perianal area (68.8%; Fig. 5). PT is a variant type of PNT, fea- Figure 2 (a) Number of CTB diagnoses made in different time
tured by erythema, papules, nodules and erosions affecting the intervals in centre 1 (Chongqing). (b) Number of diagnosed cases
prepuce and glans penis (Fig. 6, Tables 2 and 3). of CTB in centre 2 (Xi’an) from 1973 to 2012.

Discussion
The present study indicated that there were more cases of true CTB, while in less developed regions, the opposite is usually
CTB than tuberculids diagnosed in China, and their relation- observed.12–25 In review of the worldwide reports on CTB, SFD
ships evolved with time. The evolution can be explained by the (38.9%) was the most common type of true CTB, followed by
ever-changing proportion of populations with latent mycobacte- LV (36.8%) and TVC (21.1%), whereas EIB was the most preva-
rial infection. In population with a higher human development lent tuberculid (63.7%), followed by PNT (11.5%). Table 4
index, such as Japan and Hong Kong, tuberculids dominate the summarizes and compares the distribution of subtypes of true

JEADV 2018, 32, 632–638 © 2018 European Academy of Dermatology and Venereology
Cutaneous tuberculosis in China 635

Figure 3 A female patient whose whole facial area was implicated Figure 6 Penis tuberculid caused by tuberculosis of the epi-
by lupus vulgaris leading to severe disfiguration. didymis.

The age distribution pattern of CTB correlated well with that of

the overall tuberculosis disease, which ranged from 15 to
49 years.26 In our cohort, the mean ages of the PT and LV
patients were greater than those of the TVC, SFD, PNT and EI
patients, respectively (P < 0.05), while PNT patients were signif-
icantly younger than those with most other subtypes of CTB
(P < 0.005), except for SFD. In other studies, the average age of
LV patients was the highest among patients with true CTB in
India (male patients alone), United Kingdom, Turkey and South
Korea, ranging between 31.7 and 51.5 years.13,21,23,25 The mean
Figure 4 Lupus vulgaris. The central area has undergone excision
age of SFD patients was substantially lower than that of patients
and skin flap grafting because it was misdiagnosed as an
angioma. with other subtypes of true CTB, ranging between 23.4 and
45.8 years. In India, early onset of SFD was claimed to be attrib-
uted to intake of unpasteurized milk in children.15 PNT mainly
involved young adults at the age of 20–30 years, with the lowest
mean age in the CTB spectrum (Table 5).
Our study showed an overall female predisposition to CTB,
mainly due to a higher proportion of female patients with EIB.
Available results also indicated a female majority in EIB.27 In
true CTB, female patients were slightly outnumbered by male
patients (female-to-male ratio 1.1), while in tuberculids, female
patients were dominating (female-to-male ratio 2.6). Studies of
tuberculids in Japan and Hong Kong showed the same pattern,
with female-to-male ratio of 2.6 and 11.9, respectively.12,13,17
Figure 5 Tuberculosis cutis ulcerosa. Large ulceration on the bor- The duration of CTB is relatively long, which is usually mea-
der of the skin and mucosa. sured in years. In our results, the mean duration of true CTB
(8.7  9.4 years) was significantly longer than that of tuber-
culids (3.2  5.1 years) (P < 0.001). Among different subtypes
CTB and tuberculids in reported cases from different countries. of true CTB, the duration of LV (9.6  10.3 years) was signifi-
Our results differed from this overall pattern, especially regard- cantly longer than that of TVC (7.6  6.7 years), (P < 0.005),
ing true CTB, which may be due to the difference in the preva- and TVC markedly longer than that of SFD (5.1  8.5 years)
lence of tuberculosis and the varied immunity status among (P < 0.05). Existing data also suggested that LV exhibited the
different populations. longest duration in most regions investigated. A drastic example
Each subtype of CTB could be observed in different age was an 87-year-old woman with a ‘birthmark’ on the gluteal
groups, but the rate was slightly higher in young adults. The area, which was later proved to be a LV.28 The relatively long
mean age of patients with various types of CTB was found to duration of LV may be explained by its slow progression, indo-
range from 20 to 40 years, as also reflected in our study.12,13,17,24 lent nature, frequently delayed diagnosis due to low economic

JEADV 2018, 32, 632–638 © 2018 European Academy of Dermatology and Venereology
636 Zhang et al.

Table 2 Comparison of skin lesion types among different CTB types gathered from four centres in China (case/percentage)

Lesion types True CTB (n = 574)* Tuberculids (n = 496)*

LV (n = 354) TVC (n = 174) SFD (n = 30) TCU (n = 16) EI (n = 403) PNT (n = 88) PT (n = 4) LS (n = 1)
Erythema 120 (33.9) 22 (12.6) – 6 (37.5) 99 (24.6) 12 (13.6) 1 (25.0) –
Nodules 67 (18.9) 16 (9.2) 9 (30.0) 2 (12.5) 334 (82.9) – 1 (25.0) –
Papules 78 (22.0) 129 (74.1) – – – 73 (83.0) 2 (50.0) 1 (100.0)
Plaques 136 (38.4) 84 (48.3) 9 (30.0) 3 (18.8) 47 (11.7) – – –
Ulcerations 45 (12.7) 19 (10.9) 12 (40.0) 12 (75.0) 36 (8.9) – 1 (25.0) –
Fistula – – 3 (10.0) 2 (12.5) – – – –
Abscess – – 2 (6.7) – – – – –
Hyperplasia 5 (1.4) 73 (42.0) – – – – –
Scars 35 (9.9) 6 (3.5) 2 (6.7) 1 (6.3) 7 (1.7) 12 (13.6) – –
Atrophy 15 (4.2) 2 (1.2) – – 7 (1.7) – – –
Desquamation – – – – – – – –
Lichenification – – – – – – – –
Mass 3 (0.9) 4 (2.3) 8 (26.7) – – – – –

*Case number of patients who had skin lesions records.

Table 3 Comparison of implicated sites of different CTB types gathered from four centres in China (case/percentage)

Implicated Sites True CTB (n = 583)* Tuberculids (n = 496)*

LV (n = 354) TVC (n = 181) SFD (n = 32) TCU (n = 16) EI (n = 403) PNT (n = 88) PT (n = 4) LS (n = 1)
Lower limbs 20 (5.7) 11 (6.1) 8 (25.0) – 387 (96.0) 66 (75.0) – –
Upper limbs 28 (7.9) 52 (28.7) 9 (28.1) – 32 (7.9) 68 (77.3) – –
Trunk 4 (1.1) 5 (2.8) 9 (28.1) – – 51 (58.0) – 1 (100.0)
Neck 28 (7.9) – 6 (18.8) – – – – –
Buttock 23 (6.5) 60 (33.2) – – – 21 (23.9) – –
Feet – 64 (35.4) – – 11 (2.7) – – –
Hands 14 (4.0) 22 (12.2) – – – – – –
Perianal area – 12 (6.6) – 11 (68.8) – – – –
Perineum – – – 3 (18.8) – – – –
Labia – – – 2 (12.5) – – – –
Penis – – – – – – 4 (100.0) –
Nose 26 (7.3) – – 1 (6.3) – – – –
Face 189 (53.4) – 6 (18.8) – – – – –
Auricles 31 (8.8) – – – – 1 (1.1) – –

*Case number of patients who had skin lesions records.

status of most affected patients.29,30 The duration of tuberculids China was relatively low from 1980 to 1999, because the
was much shorter than that of true CTB, but the difference healthcare system was in constant improvement after the
between diverse subtypes of tuberculids was insignificant social turmoil in the 1960s and 1970s. However, in both
(P > 0.8). countries, the incidence of CTB increased again after 2000.
The incidence of TB is determined by interplay between Many factors may influence the reported incidence, such as
social factors, evolution of pathogen and epidemiological land- improved diagnostic level and disease awareness of physicians,
scape of other diseases.1,26 Mostly, CTB is part of systemic TB strengthened willingness of patients to seek professional help,
infection; therefore, the incidence of CTB is corresponding to and enhanced availability and accessibility of medical care. All
that of the entire TB infection, and CTB incidence is usually of these may provide reasonable explanation to the increased
higher in countries or regions with inadequate TB control and number of diagnosed cases with peak at around 2005 in China
management theoretically. In Japan, the incidence of CTB was (Fig. 2a and b). Further studies are required to confirm
elevated during and after World War II, but declined by whether this epidemiologic change would reflect the true state
entering 1955.12 Our data showed the incidence of CTB in of TB infection in China.

JEADV 2018, 32, 632–638 © 2018 European Academy of Dermatology and Venereology
Cutaneous tuberculosis in China 637

Table 4 Summary of distribution of the CTB spectrum reported by clinical retrospective studies around the world

Types True cutaneous tuberculosis (number/percentage)

LV TVC SFD TB Gumma Primary TB Complex TB Chancre Miliary TB TCU
Japan12 14.2% (188) 10.8% (143) 14.0% (185) – 0.2% (2) – – –
India13–15 53.7% (239) 6.29% (28) 25.62% (114) 3.4% (15) – – – –
Ethiopia16 10.9% (22) 3.0% (6) 70.8% (143) 8.9% (18) – – – –
Hong Kong17 4.1% (6) 4.1% (6) – – – – – 1.4% (2)
Morocco18 12.0% (26) 7.4% (16) 71.8% (155) – 1.4% (3) – 1.4% (3)
Spain19 61.1% (22) 2.8% (1) 11.1% (4) – – – 11.1% (4) 13.9% (5)
Pakistan20 21.6% (16) 8.1% (6) 64.9% (48) – – – – –
United Kingdom21 12.8% (6) – 55.3% (26) – – – 4.3% (2) 2.1% (1)
South Africa22 19.6% (18) 1.1% (1) 17.4% (16) – – 1.1% (1) – 1.1% (1)
Turkey23 38.5% (5) – 61.5% (8) – – – – –
Bangladesh24 40.0% (160) 50.0% (200) 10.0% (40) – – – – –
S. Korea25 13.3% (2) – 66.7% (10) 13.3% (2) – – – –
Total 710 (36.8%) 407 (21.1%) 749 (38.9%) 35 (1.8%) 2 (0.1%) 4 (0.2%) 6 (0.3%) 12 (0.6%)
Types Tuberculids (number/percentage) Sum
PNT PT EIB LS EN PTN True TB Tuber-culids Total
Japan 12
12.8% (169) 0.3% (4) 40.5% (536) 0.1% (1) – 7.3% (96) 518 (39.1%) 806 (60.9%) 1324
India13–15 2.0% (9) – 0.5% (2) 6.1% (27) 2.5% (11) – 396 (89.0%) 49 (11.0%) 445
Ethiopia16 – – 1.0% (2) 5.5% (11) – – 189 (93.6%) 13 (6.4%) 202
Hong Kong17 2.8% (4) – 87.6% (127) – – – 14 (9.7%) 131 (90.3%) 145
Morocco18 6.0% (13) 203 (94.0%) 13 (6.0%) 216
Spain19 36 – 36
Pakistan20 5.4% (4) 70 (94.6%) 4 (5.4%) 74
United Kingdom21 2.1% (1) – 10.6% (5) – 6.4% (3) – 37 (78.7%) 10 (21.3%) 47
South Africa22 41.3% (38) – 21.7% (20) 3.3% (3) – – 37 (40.2%) 61 (66.3%) 92
Turkey23 – – – – – – 13 – 13
Bangladesh24 – – – – – – 400 – 400
S. Korea25 – – – – – – 14 – 15
Total 221 (11.5%) 4 (0.4%) 692 (63.7%) 42 (3.9%) 14 (1.3%) 96 (8.8%) 1927 1087 3009

Table 5 Summary of mean duration of various types of CTB reported by clinical retrospective studies around the world

Types LV SFD TVC OTB Miliary TCU/ EI PNT Total

TB Abscess
Hong Kong17 5.6 year – 11.3 year 5.2 year – – 11 month 1 year 1.6 year
(0.5–10 year) (8 month–40 (0.5–10 year) (2 day–20 (1–24 month)
year) year)
Spain19 12.6  18.3 year 3.8  5.6 2 month 1 year 9.2  4.0 5.4  5.9 – – –
month day month
Turkey23 4.7  7.0 year 1.0  1.4 – – – – – – –
S. Korea25 9.0  4.2 month 6.8  7.5 – – – – – – 6.6  6.7 month
month
Morocco18 – – – – – – – – 1 month–12 year

Our study has some limitations. Microbiological evidence Data regarding multidrug resistance and concurrence with HIV/
through bacterial culture or PCR examination was absent in AIDS were lacking or incomplete for further analysis.
most of the cases because these essays have not been performed In conclusion, chronologic changes in CTB in China over the
regularly clinically in any centre of our study till now. This hos- past 50 years may reflect the transition of prevalence of overall
pital-based time point retrospective study carried sampling and tuberculosis in this period of time. Due to its diverse clinical
diagnostic bias, and could not represent the true prevalence in presentations, CTB is usually misdiagnosed or neglected, even
general population, or follow the natural history of the disease. by dermatologists. Different subtypes of CTB were characterized

JEADV 2018, 32, 632–638 © 2018 European Academy of Dermatology and Venereology
638 Zhang et al.

by conspicuous gender predilection, duration, age, and typical 14 Kumar B, Muralidhar S. Cutaneous tuberculosis: a twenty-year prospec-
clinical manifestations and pathological findings. Due to the tive study. Int J Tuberc Lung Dis 1999; 3: 494–500.
15 Kumar B, Rai R, Kaur I et al. Childhood cutaneous tuberculosis: a study
accelerating and intensifying globalization, international cooper- over 25 years from northern India. Int J Dermatol 2001; 40: 26–32.
ation of dermatologists with regard to epidemiology, diagnosis, 16 Terranova M, Padovese V, Fornari U et al. Clinical and epidemiological
treatment and prevention of CTB is warranted. study of cutaneous tuberculosis in Northern Ethiopia. Dermatology 2008;
217: 89–93.
17 Ho CK, Ho MH, Chong LY. Cutaneous tuberculosis in Hong Kong: an
References update. Hong Kong Med J 2006; 12: 272–277.
1 Caminero J. Guia de la Tuberculosis Para Medicos Especialistas. Interna- 18 Zouhair K, Akhdari N, Nejjam F et al. Cutaneous tuberculosis in Mor-
tional Union against Tuberculosis and Respiratory Diseases, Paris, 2003: occo. Int J Infect Dis 2007; 11: 209–212.
35–37. 19 Marcoval J, Alcaide F. Evolution of cutaneous tuberculosis over the past
2 Dostrovsky A, Sagher F. Dermatological complication of BCG vaccina- 30 years in a tertiary hospital on the European Mediterranean coast. Clin
tion. Br J Dermatol 1963; 75: 180–190. Exp Dermatol 2013; 38: 131–136.
3 Santos JB, Figueiredo AR, Ferraz CE et al. Cutaneous tuberculosis: epi- 20 Yasmeen N, Kanjee A. Cutaneous tuberculosis: a three year prospective
demiologic, etiopathogenic and clinical aspects – part I. An Bras Dermatol study. J Pak Med Assoc 2005; 55: 10–12.
2014; 89: 219–228. 21 Yates VM, Ormerod LP. Cutaneous tuberculosis in Blackburn district
4 Hara K, Tsuzuki T, Takagi N et al. Nodular granulomatous phlebitis of (U.K.): a 15-year prospective series, 1981–95. Br J Dermatol 1997; 136:
the skin: a fourth type of tuberculid. Histopathology 1997; 30: 129–134. 483–489.
5 Zhang H, Liu Y, Liu C et al. Retrospective analysis of 28 cutaneous tuber- 22 Visser AJ, Heyl T. Skin tuberculosis as seen at Ga-Rankuwa Hospital. Clin
culosis cases. China J Lepr Skin Dis 2005; 21: 108–109. Exp Dermatol 1993; 18: 507–515.
6 WHO. Global Tuberculosis Report 2016. Geneva: World Health Organi- 23 Kivanc-Altunay I, Baysal Z, Ekmekci TR et al. Incidence of cutaneous
zation, 2016. tuberculosis in patients with organ tuberculosis. Int J Dermatol 2003; 42:
7 Duanmu H. Report on fourth national epidemiological sampling survey 197–200.
of tuberculosis. Chin J Tuberc Respir Dis 2002; 25: 3–7. 24 Choudhury AM, Ara S. Cutaneous tuberculosis–a study of 400 cases. Ban-
8 Duanmu H, Liu Y, Shi H et al. Analysis of the third national epidemio- gladesh Med Res Counc Bull 2006; 32: 60–65.
logical survey of tuberculosis. Dis Surveillance 1995; 10: 215–217. 25 Min KW, Ko JY, Park CK. Histopathological spectrum of cutaneous
9 Li L, Duanmu H. The epidemic of childhood tuberculosis in China. Natl tuberculosis and non-tuberculous mycobacterial infections. J Cutan
Med J Chin 2004; 84: 1678–1680. Pathol 2012; 39: 582–595.
10 Wang Y. The fifth national tuberculosis epidemiological survey in 2010. 26 Frieden TR, Sterling TR, Munsiff SS et al. Tuberculosis. Lancet 2003; 362:
Chin J Antituber 2012; 34: 485–508. 887–899.
11 Onozaki I, Law I, Sismanidis C et al. National tuberculosis prevalence 27 Rademaker M, Lowe DG, Munro DD. Erythema induratum (Bazin’s dis-
surveys in Asia, 1990–2012: an overview of results and lessons learned. ease). J Am Acad Dermatol 1989; 21: 740–745.
Trop Med Int Health 2015; 20: 1128–1145. 28 Woo PN, Batta K, Tan CY, Colloby P. Lupus vulgaris diagnosed after
12 Hamada M, Urabe K, Moroi Y et al. Epidemiology of cutaneous tubercu- 87 years, presenting as an ulcerated ‘birthmark’. Br J Dermatol 2002; 146
losis in Japan: a retrospective study from 1906 to 2002. Int J Dermatol (3): 525–526.
2004; 43: 727–731. 29 Hill MK, Sanders CV. Cutaneous tuberculosis. Microbiol Spectr 2017; 5.
13 Sharma S, Sehgal VN, Bhattacharya SN et al. Clinicopathologic spectrum http://doi.org/0.1128/microbiolspec.TNMI7-0010-2016.
of cutaneous tuberculosis: a retrospective analysis of 165 Indians. Am J 30 Sellami K, Boudaya S, Chaabane H et al. Twenty-nine cases of lupus vul-
Dermatopathol 2015; 37: 444–450. garis. Med Mal Infect 2016; 46(2): 93–95.

JEADV 2018, 32, 632–638 © 2018 European Academy of Dermatology and Venereology