INT J TUBERC LUNG DIS 18(4):492–498

Q 2014 The Union

http://dx.doi.org/10.5588/ijtld.13.0694

Smoking increases risk of recurrence after successful anti-

tuberculosis treatment: a population-based study

Y-F. Yen,*† M-Y. Yen,‡ Y-S. Lin,§ Y-P. Lin,¶ H-C. Shih,‡ L-H. Li,‡ P. Chou,† C-Y. Deng#
*
Section of Infectious Diseases, Taipei City Hospital, Taipei City Government, Taipei, †Community Medicine
Research Center and Institute of Public Health, National Yang-Ming University, Taipei, ‡Department of Disease
Control and Prevention, Taipei City Hospital, Taipei City Government, Taipei, §Department of Pharmacy, Taipei
Veterans General Hospital, Taipei, ¶Taipei Databank for Public Health Analysis, Taipei City Hospital, Taipei City
Government, Taipei, #Institute of Hospital and Health Care Administration, National Yang-Ming University, Taipei,
Taiwan

SUMMARY

O B J E C T I V E : To investigate whether tobacco smoking
increases the risk of tuberculosis (TB) recurrence and
identify factors associated with TB recurrence among
adults who had successfully completed anti-tuberculosis
treatment in Taipei, Taiwan.
M E T H O D S : Recurrence was defined as a new clinical or
microbiological diagnosis of TB requiring the start of a
new course of treatment in a patient who had
satisfactorily completed treatment for a previous TB
episode. Cox proportional hazard models were used to
calculate adjusted hazard ratios (aHRs) for recurrence.
R E S U LT S : We followed 5567 adults for recurrence after
successful anti-tuberculosis treatment. The mean age
was 58.5 years; 62.9% were male. Overall, 84 (1.5%)
had a recurrence of TB during follow-up. The incidence
of TB recurrence was 4.9 episodes/1000 person-years of
follow-up. Cox proportional hazards regression showed
that after controlling for other variables, the risk of TB
recurrence among subjects who smoked .10 cigarettes a
day was double that of never/former smokers. Other
independent risk factors significantly associated with TB
recurrence were homelessness (aHR 3.75, 95%CI 1.17–
12.07), presence of comorbidities (aHR 2.66, 95%CI
1.22–5.79) and a positive acid-fast bacilli smear (aHR
2.27, 95%CI 1.47–3.49).
C O N C L U S I O N : Smoking .10 cigarettes a day was
significantly associated with TB recurrence. To reduce
the risk of recurrence, we recommend including effective
measures of smoking cessation in TB control pro-
grammes, as recommended by the World Health
Organization Stop TB Strategy.
K E Y W O R D S : tuberculosis; recurrence; smoking;
Taiwan

TUBERCULOSIS (TB) can recur and lead to TB
transmission even after successful treatment. The
average reported incidence rates of recurrent TB (per
100 000 person-years [py]) worldwide were respec-
tively 3010 (95% confidence interval [CI] 2230–3970)
and 2290 (95%CI 1730–2940) at 6 and 12 months
after treatment completion.1 Recurrent TB cases
constitute a particular challenge to public health
systems, in part due to their higher rates of multidrug
resistance2,3 and mortality4 than new cases. Moreover,
retreatment regimens for recurrent TB tend to be more
expensive than first-line treatment for a new episode,
and can thus put further strain on national TB control
budgets.
Three recent literature reviews have reported that
while smoking is an important factor for TB develop-
ment, the association between tobacco smoking and
TB recurrence is not clear and has seldom been
investigated.5–7 To our knowledge, only four recent
studies have evaluated this association: three found a
significant association between smoking and TB
recurrence,8–10 while one did not.11 Moreover, these
findings are limited due to a possible misclassification
of TB cases, given the fact that TB patients were
identified using chest radiography (CXR) alone,9 the
small number of recurrent cases10,11 and inadequate
adjustment for potential confounders such as treat-
ment duration,8–11 clinical findings (e.g., cavity on
CXR)8–11 and underlying diseases (e.g., human
immunodeficiency virus [HIV] infection).8,9 These
studies also varied in their classification of smoking
status, as non-smoker and smoker,8,11 non-smoker, ex-
smoker and current smoker,9 and ‘never smoking or
quit smoking’ and ever smoking.10 A better under-

Correspondence to: Chung-Yeh Deng, Institute of Hospital and Health Care Administration, National Yang-Ming
University, 155, Section 2, Ni-Long Street, Taipei, Taiwan 11221. Tel: (þ886) 2 2826 7390. Fax: (þ886) 2 2822 1942. e-mail:
cydeng@ym.edu.tw
Article submitted 18 September 2013. Final version accepted 5 December 2013.

standing of the incidence of, and factors associated
with, TB recurrence might help in identifying the most
vulnerable populations and developing effective con-
trol measures that lower the rate of recurrence.
We estimated the incidence of TB recurrence and
analysed associated risk factors, with particular
emphasis on smoking, among adults who had
successfully completed anti-tuberculosis treatment
in Taipei, Taiwan, from 2005 to 2010.
METHODS
Study population and data source
This retrospective cohort study used TB surveillance
data from Taipei, Taiwan. In Taiwan, active cases of
pulmonary TB must be reported to the Taiwan Centers
for Disease Control (CDC) within 24 h, and other TB
cases within 7 days of diagnosis. After notification,
trained case managers use a structured questionnaire
to interview patients about their sociodemographic
characteristics, clinical findings, underlying diseases,
admission history and anti-tuberculosis treatment. TB
patients in Taiwan are required by law to be
monitored until treatment success, default, treatment
failure, transfer or death, as defined by the World
Health Organization (WHO).12
In this study, both initial and recurrent TB cases
were defined using clinical and/or laboratory find-
ings.13,14 Clinical cases included those with manifes-
tations consistent with TB (e.g., prolonged fever,
wasting) and exclusion of other differential diagnoses
by diagnostic evaluation.13 Laboratory definitions
included 1) Mycobacterium tuberculosis isolated
from a clinical specimen, or 2) acid-fast bacilli
(AFB) identified in a clinical specimen from patients
with clinical manifestations consistent with TB. This
study included Taiwanese adults aged 18 years
diagnosed with TB in Taipei during the period 2005–
2010. After successful anti-tuberculosis treatment, all
study subjects were followed until TB recurrence,
death or survival without recurrent TB. The follow-
up period lasted until 31 December 2011. The
Taiwan mortality registry was consulted to identify
follow-up cases who died during the study period.
This project was approved by the Institutional
Review Board of the Taipei City Hospitals, Taipei,
Taiwan.
Outcome variable
TB recurrence was defined as a new clinical or
microbiological diagnosis of TB requiring the start of
a new course of treatment in a patient who had
successfully completed treatment for a previous TB
episode.10
Explanatory variables
Explanatory variables included subject sociodemo-
graphic factors (age, sex, marital status, education
Smoking increases risk of TB recurrence 493
level, homelessness, current smoking status, alcohol
use and unemployment), clinical findings (AFB smear
status and TB culture, cavities on CXR, pleural
effusion, extra-pulmonary TB), comorbidities, in-
cluding malignancy and HIV infection, source of
notification, year of diagnosis and treatment course.
Education level was categorised as no education,
elementary school, high school and university or
higher. Unemployment was defined as unemployed
status at the first TB episode. Current smoking status
was categorised as never smoked or quit smoking,
smoking 1 – 10 cigarettes a day and smoking .10
cigarettes a day. Source of notification was defined as
the department that reported the TB case, i.e., general
wards, intensive care units or out-patient services.
Treatment course was categorised as 9 and .9
months.15
Statistical analysis
Cox proportional hazards regression models were
used to investigate risk factors associated with
recurrent TB. Unadjusted hazard ratios (HRs) for
sociodemographics, comorbidities, clinical findings
and treatment regimen were calculated using univar-
iate analysis. Kaplan-Meier curves were created and
comparisons were assessed based on the factors
associated with TB recurrence. All variables found
to be significant (P , 0.10) using univariate analysis
were considered for inclusion in multivariate analysis.
Backward-stepwise Cox regression analysis was
performed to produce the final model. HRs and
adjusted HRs (aHRs) with 95% confidence intervals
(CIs) are reported to show the strength and direction
of associations. We also tested the interaction
between smoking status and other covariates in the
multivariate analysis. All analyses were conducted
using the SPSS statistical software package version
21.0 (Statistical Product and Service Solutions,
Chicago, IL, USA).
RESULTS
In total, 7143 adult TB cases were reported to the
Taipei TB Control Department from 2005 to 2010
(Figure 1). Of these, 1504 individuals died during
anti-tuberculosis treatment, 13 were lost to follow-
up, 25 were still receiving treatment, 7 experienced
treatment failure and 27 were transferred out of
Taipei City; these 1566 cases were all excluded from
subsequent analyses. The remaining 5567 successful-
ly treated patients were included in the study. The
overall mean age was 58.5 years (range 18–98);
62.9% were male.
Of the 5567 study subjects, 84 (1.5%) developed
recurrence during 17 166 py of follow-up, which
corresponds to an incidence of TB recurrence of 4.9
episodes/1000 py (95%CI 3.8–5.9) of follow-up.
Table 1 shows the number of recurrent TB cases by
494 The International Journal of Tuberculosis and Lung Disease

Figure 1 Study flow diagram. TB ¼ tuberculosis.

year of follow-up: 75% of TB recurrences occurred
within 3 years of follow-up, and the number of
individuals who developed TB recurrence decreased
during the remainder of the follow-up period.
Compared with never/former smokers, the inci-
dence of TB recurrence was significantly higher
among study subjects who reported smoking .10
cigarettes a day (log-rank statistic, P ¼ 0.001; Figure
2).
Table 2 shows the factors associated with TB
recurrence in univariate analysis. Cox proportional
hazards analysis showed that factors associated with
TB recurrence were male sex, homelessness, current
smoking of .10 cigarettes a day (reference: never/
former smoking), presence of comorbidities and
being AFB smear- or TB culture-positive.
Backward-stepwise Cox regression analysis (Table
Table 1 Number and proportion of recurrences by year of
follow-up, Taipei, Taiwan
Events % of all cases
Year of follow-up n of recurrence
1 20
2 28
3 15
4 8
5 10
6 3
3) showed that after controlling for subjects’ socio-
demographics, clinical findings, underlying diseases,
admission history and TB course, the risk of TB
recurrence among study subjects who smoked .10
cigarettes a day was double that of never/former
smokers. Other independent risk factors significantly
associated with TB recurrence were homelessness
(aHR 3.75, 95%CI 1.17–12.07), presence of comor-
bidities (aHR 2.66, 95%CI 1.22–5.79) and a positive
AFB smear (aHR 2.27, 95%CI 1.47–3.49). The
interaction terms between smoking status and other
covariates were not statistically significant in the
multivariate analysis.
DISCUSSION
In this large cohort study of 4569 adults successfully
treated for TB, 84 (1.5%) developed a recurrence of
TB during the period 2005–2010. To our knowledge,
this is the first study to evaluate the incidence of TB
recurrence in Taiwan. The overall recurrence rate of
4.9/1000 py among adults with successful anti-
tuberculosis treatment is approximately 10 fold that
23.8 of new TB development among the general popula-
33.3
17.9 tion of Taipei.16 The TB recurrence rate in Taipei is
9.5 lower than that in Spain (5.3/1000 py).11 In addition,
11.9 we found that 75% of cases of TB recurrence
3.6
occurred in the first 3 years of follow-up. TB is much

Figure 2 Risk of TB recurrence by smoking status, Taipei, Taiwan. TB ¼ tuberculosis.
more likely among adults with a history of TB; TB
patients should thus be strictly monitored even after
completing treatment, particularly during the first 3
years.
Previous studies have shown that TB recurrence
rates might depend on background TB incidence.17,18
In a comparison with other countries with similar
low-to-moderate TB incidence, the TB recurrence
rate in Taipei was lower than that in Singapore (5.7
episodes/1000 py),19 but higher than that in Poland
(2.3 episodes/1000 py).20 International literature also
suggests that background TB incidence corresponds
to the rate of recurrence due to reinfection.21,22 As we
were unable in the present study to use techniques to
genotype TB by DNA strain, which would have
enabled us to distinguish reinfection from relapse of
the same TB episode, we estimate that the proportion
of reinfection in our 2005–2010 population (mean
incidence 48.5/100 000) among those with recurrent
TB was approximately 32.3%, per Wang et al.’s
formula.21
We found that current smokers of .10 cigarettes a
day had a higher risk of recurrence. Our study might
have underestimated this risk because we compared
current smokers with never/former smokers instead
of with never smokers only, and smoking might still
have a latent effect on TB recurrence among former
smokers who quit only 3–5 years before diagnosis of
TB recurrence.10 A previous study reported that
former smoking and TB recurrence might be posi-
tively associated.9 To reduce the risk of recurrence,
effective smoking cessation measures should be
incorporated into TB control programmes.
The study also evaluated the association between
smoking and TB recurrence after stratifying patients
by sex, TB culture and AFB smear. Backward-stepwise
Cox regression analysis showed that smoking .10
cigarettes a day was significantly associated with TB
recurrence among male and culture- and AFB-positive
Smoking increases risk of TB recurrence 495
patients. While patients who smoked .10 cigarettes a
day had a higher risk of TB recurrence among female,
culture-negative or AFB-negative patients, these asso-
ciations were not statistically significant, which may
have been due to insufficient sample size; further
research is therefore needed.
In accordance with the WHO DOTS programme,23
current TB control policies in Taiwan attempt to
achieve a treatment success rate of 85% in treating
people with TB.16 Population-based TB control
requires more than simply treating active disease;
prevention strategies are also urgently required.24
Given the concerns regarding the increased risk of TB
recurrence among smokers, interventions for smok-
ing cessation should be integrated into current TB
control programmes. For example, health care
providers and TB case managers at clinics and local
health centres should proactively inform TB patients
that smoking increases the risk of recurrence.25
Moreover, incorporating smoking cessation measures
into the DOTS programme could help promote
quitting if treatment observers are well trained in
advising patients to cease smoking.26
Previous studies have shown that smoking was
significantly associated with the development of active
TB,27 likely because nicotine inhibits the production
of tumour necrosis factor alpha by macrophages in the
lungs, thereby increasing susceptibility to the devel-
opment of progressive disease from latent M. tuber-
culosis infection.28,29 This mechanism might also
explain the development of recurrence.
This study found that homeless people had a higher
risk of TB recurrence despite having completed anti-
tuberculosis treatment. Previous research showed
that homeless people are among the most disadvan-
taged social groups and are susceptible to infectious
diseases due to malnutrition and unhealthy living
conditions (e.g., on the street, in crowded shelters).30
Moreover, delayed diagnosis and poor adherence to
496 The International Journal of Tuberculosis and Lung Disease

Table 2 Unadjusted HRs for factors associated with recurrent TB, Taipei, Taiwan, 2005–2010
Recurrent TB
Subjects
Factor n n (%) HR (95%CI) P
Age, years
18–34 948 19 (2.0) 1
35–49 907 16 (1.8) 0.93 (0.48–1.80) 0.819
50–64 1235 18 (1.5) 0.81 (0.42–1.54) 0.511
65 2477 31 (1.3) 0.79 (0.45–1.41) 0.427
Sex
Female 2063 21 (1.0) 1
Male 3504 63 (1.8) 1.90 (1.16–3.11) 0.011
Year of diagnosis
2005 1082 29 (2.7) 1
2006 985 21 (2.1) 0.90 (0.51–1.59) 0.709
2007 1016 15 (1.5) 0.78 (0.41–1.48) 0.439
2008 856 10 (1.2) 0.75 (0.35–1.59) 0.45
2009 803 6 (0.7) 0.69 (0.27–1.72) 0.423
2010 825 3 (0.4) 0.94 (0.27–3.28 0.924
Marital status
Unmarried 1306 27 (2.1) 1
Married 4242 57 (1.3) 0.70 (0.44–1.10) 0.698
Unknown 19 0 — —
Education level
No education 335 5 (1.5) 1
Elementary school 1103 19 (1.7) 0.99 (0.37–2.64 0.977
High school 1940 29 (1.5) 0.86 (0.33–2.21) 0.747
University or higher 1652 25 (1.5) 0.86 (0.33–2.23) 0.749
Unknown 537 6 (1.1) 0.58 (0.18–1.89) 0.365
Unemployed status
No 1971 30 (1.5) 1
Yes 3578 54 (1.5) 1.07 (0.68–1.67) 0.776
Unknown 18 0 — —
Homelessness
No 5521 81 (1.5) 1
Yes 46 3 (6.5) 5.67 (1.79–17.95) 0.003
Current smoking, cigarettes/day
None 4582 59 (1.3) 1
1–10 333 4 (1.2) 1.01 (0.37–2.77) 0.991
.10 652 21 (3.2) 2.54 (1.54–4.17) ,0.001
Any alcohol use
No 5088 75 (1.5) 1
Yes 464 9 (1.9) 1.41 (0.71–2.81) 0.331
Unknown 15 0 — —
Comorbidities*
No 5326 77 (1.4) 1
Yes 241 7 (2.9) 2.93 (1.35–6.37) 0.007
Acid-fast bacilli smear
Negative 3644 38 (1.0) 1
Positive 1923 46 (2.4) 2.43 (1.58–3.74) ,0.001
TB culture
Negative 2097 21 (1.0) 1
Positive 3470 63 (1.8) 2.14 (1.30–3.50) 0.003
Cavities on CXR
No 4550 64 (1.4) 1
Yes 1017 20 (2.0) 1.35 (0.82–2.24) 0.236
Pleural effusion on CXR
No 4989 77 (1.5) 1
Yes 578 7 (1.2) 0.83 (0.38–1.80) 0.633
Extra-pulmonary TB
No 4903 77 (1.6) 1
Yes 664 7 (1.1) 0.63 (0.29–1.36) 0.236
Source of notification
Out-patient services 3513 49 (1.4) 1
Ordinary ward 1961 33 (1.7) 1.29 (0.83–2.01) 0.257
Intensive care unit 93 2 (2.2) 1.77 (0.43–7.30) 0.427
Duration of treatment, months
9 3158 42 (1.3) 1
.9 2409 42 (1.7) 1.46 (0.95–2.23) 0.081
*Including malignancies and human immunodeficiency virus infection.
HR ¼ hazard ratio; TB ¼ tuberculosis; CI ¼ confidence interval; CXR ¼ chest radiograph.

Table 3 Adjusted HRs for factors associated with recurrent TB,
Taipei, Taiwan, 2005–2010
Factors Adjusted HR*
Current smoking, cigarettes/day
None 1
1–10 0.86
.10 2.04
Homelessness 3.75
Comorbidities† 2.66
AFB positivity 2.27
*Backward-stepwise Cox regression analysis after controlling for sex,
homelessness, current smoking, comorbidities, AFB smear, TB culture, and
duration of treatment.
†
Including malignancies and human immunodeficiency virus infection.
HR ¼ hazard ratio; TB ¼ tuberculosis; CI ¼ confidence interval; AFB ¼ acid-fast
bacilli.
medical treatment are highly likely in the case of
recurrent TB among the homeless.31 The homeless
also fear further discrimination in case of TB
recurrence.32 To ensure cooperation in follow-up
examinations and treatment, TB control programmes
should therefore include a comprehensive strategy to
improve adherence to the TB tracing system among
the homeless.
In line with the results reported by a previous study,3
we found that the presence of comorbidities (e.g.,
HIV) was significantly associated with TB recurrence.
However, we found no significant association between
TB recurrence and pretreatment cavitation, which
confirmed the results of one study33 but contradicted
those of two other reports.34,35
Two limitations should be considered when inter-
preting the present findings: 1) the absence of
information on strain typing made it impossible to
differentiate between endogenous reactivation and
exogenous reinfection, and 2) 26.2% of the initial TB
cases were identified by clinical diagnosis rather than
by AFB smear or culture. Although this might have
resulted in overdiagnosis of TB, such overdiagnosis is
unlikely to be widespread, as the Taiwan CDC holds
monthly expert meetings on TB to examine the
diagnosis of every ambiguous TB case.16
CONCLUSIONS
Continuing to smoke .10 cigarettes a day was
significantly associated with TB recurrence. This
finding has important implications for TB control
policy. While the DOTS strategy has become the
standard treatment protocol for TB control in most
countries,23 current TB control strategies fail to
account for the increased risk of recurrence due to
smoking. To reduce the risk of recurrence, it is thus
imperative to incorporate effective smoking cessation
measures into TB control programmes, as recom-
mended by the WHO Stop TB Strategy.
Smoking increases risk of TB recurrence 497
Acknowledgements
The authors wish to thank registered nurses J-C Hsiao, M-Y Chen,
95%CI and C-C Pan for interviewing the subjects and reviewing their
medical records; and the Research Office for Health Data,
Department of Education and Research, Taipei City Hospital,
Taiwan, for their valuable contributions to data management and
0.31–2.38
1.22–3.41 statistical analysis.
Conflict of interest: none declared.
1.17–12.07
1.22–5.79
1.47–3.49 References
1 Panjabi R, Comstock G W, Golub J E. Recurrent tuberculosis
and its risk factors: adequately treated patients are still at high
risk. Int J Tuberc Lung Dis 2007; 11: 828–837.
2 Yu C C, Chang C Y, Liu C E, Shih L F, Hsiao J H, Chen C H.
Drug resistance pattern of Mycobacterium tuberculosis
complex at a medical center in central Taiwan, 2003–2007. J
Microbiol Immunol Infect 2010; 43: 285–290.
3 El Sahly H M, Wright J A, Soini H, et al. Recurrent tuberculosis
in Houston, Texas: a population-based study. Int J Tuberc Lung
Dis 2004; 8: 333–340.
4 Murray J, Sonnenberg P, Shearer S C, Godfrey-Faussett P.
Human immunodeficiency virus and the outcome of treatment
for new and recurrent pulmonary tuberculosis in African
patients. Am J Respir Crit Care Med 1999; 159: 733–740.
5 Slama K, Chiang C Y, Enarson D A, et al. Tobacco and
tuberculosis: a qualitative systematic review and meta-analysis.
Int J Tuberc Lung Dis 2007; 11: 1049–1061.
6 Bates M N, Khalakdina A, Pai M, Chang L, Lessa F, Smith K R.
Risk of tuberculosis from exposure to tobacco smoke: a
systematic review and meta-analysis. Arch Intern Med 2007;
167: 335–342.
7 Lin H H, Ezzati M, Murray M. Tobacco smoke, indoor air
pollution and tuberculosis: a systematic review and meta-
analysis. PLOS Med 2007; 4: e20.
8 Thomas A, Gopi P G, Santha T, et al. Predictors of relapse
among pulmonary tuberculosis patients treated in a DOTS
programme in South India. Int J Tuberc Lung Dis 2005; 9: 556–
561.
9 Jee S H, Golub J E, Jo J, Park I S, Ohrr H, Samet J M. Smoking
and risk of tuberculosis incidence, mortality, and recurrence in
South Korean men and women. Am J Epidemiol 2009; 170:
1478–1485.
10 d’Arc Lyra Batista J, de Fatima Pessoa Militao de Albuquerque
M, de Alencar Ximenes R A, Rodrigues L C. Smoking increases
the risk of relapse after successful tuberculosis treatment. Int J
Epidemiol 2008; 37: 841–851.
11 Millet J P, Orcau A, de Olalla P G, Casals M, Rius C, Cayla J A.
Tuberculosis recurrence and its associated risk factors among
successfully treated patients. J Epidemiol Community Health
2009; 63: 799–804.
12 World Health Organization: Global tuberculosis control. WHO
Report 1999. WHO/CDS/CPC/TB/99.259. Geneva, Switzer-
land: WHO, 1999.
13 Wang F D, Chang C H, Su W J, et al. Screening of hospital
workers for pulmonary tuberculosis in a medical center in
Taiwan. Infect Control Hosp Epidemiol 2006; 27: 510–511.
14 Lai C C, Wang H C. Diagnostic performance of enzyme-linked
immunospot assay and whole-blood interferon-gamma assay
for the diagnosis of extrapulmonary tuberculosis. J Microbiol
Immunol Infect 2011; 44: 406–407.
15 American Thoracic Society, Centers for Disease Control and
Prevention, Infectious Diseases Society of America. Treatment
of tuberculosis. MMWR Recomm Rep 2003; 52 (RR-11): 1–77.
16 Taiwan Centers for Disease Control and Prevention.
Surveillance and statistics of reportable infectious diseases.
Taipei, Taiwan: Taiwan CDC, 2004. http://203.65.72.83/ch/dt/
upload/QC/QC_index.htm Accessed January 2014.
498 The International Journal of Tuberculosis and Lung Disease
17 van Rie A, Warren R, Richardson M, et al. Exogenous
reinfection as a cause of recurrent tuberculosis after curative
treatment. N Engl J Med 1999; 341: 1174–1179.
18 Cacho J, Perez Meixeira A, Cano I, et al. Recurrent tuberculosis
from 1992 to 2004 in a metropolitan area. Eur Respir J 2007;
30: 333–337.
19 Singapore Tuberculosis Service/British Medical Research
Council. Assessment of a daily combined preparation of
isoniazid, rifampin, and pyrazinamide in a controlled trial of
three 6-month regimens for smear-positive pulmonary
tuberculosis. Am Rev Respir Dis 1991; 143: 707–712.
20 Snider D E, Graczyk J, Bek E, Rogowski J. Supervised six-
months’ treatment of newly diagnosed pulmonary
tuberculosis using isoniazid, rifampin, and pyrazinamide
with and without streptomycin. Am Rev Respir Dis 1984;
130: 1091–1094.
21 Wang J Y, Lee L N, Lai H C, et al. Prediction of the tuberculosis
reinfection proportion from the local incidence. J Infect Dis
2007; 196: 281–288.
22 Shen G, Xue Z, Shen X, et al. The study recurrent tuberculosis
and exogenous reinfection, Shanghai, China. Emerg Infect Dis
2006; 12: 1776–1778.
23 World Health Organization. Global tuberculosis control, 2011.
Geneva, Switzerland: WHO, 2011. WHO/HTM/TB/2011.16.
http://whqlibdoc.who.int/publications/2011/9789241564380_
eng.pdf. Accessed January 2014.
24 Brewer T F, Heymann S J. To control and beyond: moving
towards eliminating the global tuberculosis threat. J Epidemiol
Community Health 2004; 58: 822–825.
25 Secades-Villa R, Alonso-Perez F, Garcia-Rodriguez O,
Fernandez-Hermida J R. Effectiveness of three intensities of
smoking cessation treatment in primary care. Psychol Rep
2009; 105: 747–758.
26 Sereno A B, Soares E C, Lapa E S J R, et al. Feasibility study of a
smoking cessation intervention in directly observed therapy
short-course tuberculosis treatment clinics in Rio de Janeiro,
Brazil. Rev Panam Salud Publica 2012; 32: 451–456.
27 Lin H H, Ezzati M, Chang H Y, Murray M. Association between
tobacco smoking and active tuberculosis in Taiwan: prospective
cohort study. Am J Respir Crit Care Med 2009; 180: 475–480.
28 Boelaert J R, Gomes M S, Gordeuk V R. Smoking, iron and
tuberculosis. Lancet 2003; 362: 1243–1244.
29 Davies P D, Yew W W, Ganguly D, et al. Smoking and
t uberculosis: the epidemiological association and
immunopathogenesis. Trans R Soc Trop Med Hyg 2006; 100:
291–298.
30 Beijer U, Wolf A, Fazel S. Prevalence of tuberculosis, hepatitis C
virus, and HIV in homeless people: a systematic review and
meta-analysis. Lancet Infect Dis 2012; 12: 859–870.
31 Burman W J, Cohn D L, Rietmeijer C A, Judson F N, Sbarbaro J
A, Reves R R. Noncompliance with directly observed therapy
for tuberculosis. Epidemiology and effect on the outcome of
treatment. Chest 1997; 111: 1168–1173.
32 Story A, van Hest R, Hayward A. Tuberculosis and social
exclusion. BMJ 2006; 333: 57–58.
33 British Thoracic Association. A controlled trial of six months’
chemotherapy in pulmonary tuberculosis. Second report: results
during the 24 months after the end of chemotherapy. Am Rev
Respir Dis 1982; 126: 460–462.
34 Yamagishi F, Toyota M. [Research and control of relapse
tuberculosis cases]. Kekkaku 2009; 84: 767–768. [Japanese]
35 Benator D, Bhattacharya M, Bozeman L, et al. Rifapentine and
isoniazid once a week versus rifampicin and isoniazid twice a
week for treatment of drug-susceptible pulmonary tuberculosis
in HIV-negative patients: a randomised clinical trial. Lancet
2002; 360: 528–534.

O B J E C T I F : Evaluer si le fait de fumer augmente le
risque de récurrence de tuberculose (TB) et identifier les
facteurs associés à cette récurrence parmi des patients
qui avaient terminé avec succès leur traitement anti-
tuberculeux à Taipei, Taiwan.
M E T H O D E S : Une récurrence a été définie comme un
nouveau diagnostic clinique ou microbiologique de TB
nécessitant la mise en route d’un nouveau protocole de
traitement chez un patient qui avait achevé avec succès le
traitement de l’épisode de TB précédent. Le modèle de
risque proportionnel de Cox a permis de mesurer les
ratios de risque ajusté (aHR) de récurrence.
R E S U L T A T S : Nous avons suivi 5567 adultes à la
recherche d’une récurrence après un traitement réussi.
Leur âge moyen était de 58,5 ans et 62,9% étaient des
hommes. Parmi eux, 84 (1,5%) ont eu une récurrence de
TB pendant le suivi. L’incidence de la récurrence était
4,9 épisodes pour 1000 personnes-années de suivi. La
OBJETIVO: Investigar si el tabaquismo aumenta el
riesgo de sufrir una recaı́da de la tuberculosis (TB) y
examinar los factores asociados con la misma en los
adultos que habı́an completado con éxito el tratamiento
antituberculoso en Taipéi, Taiwán.
M É T O D O S : Se definió la recaı́da como un nuevo
diagnóstico clı́nico o microbiológico de TB que exige
el comienzo de otro ciclo terapéutico en un paciente que
habı́a completado de manera satisfactoria el tratamiento
de un episodio previo de TB. Se aplicaron modelos
proporcionales de Cox con el fin de medir los cocientes
de riesgos ajustados (aHR) de recaı́da.
R E S U LT A D O S : Se vigiló la aparición de recaı́da en 5567
adultos después de un tratamiento antituberculoso
eficaz. El promedio de la edad fue 58,5 años y 62,9%
fueron hombres. En general, 84 pacientes presentaron
recaı́da durante el seguimiento (1,5%); la incidencia de
recaı́da de TB fue 4,9 episodios por 1000 años-persona
Smoking increases risk of TB recurrence i
RESUME
régression de risques proportionnels de Cox a montré,
après contrôle des autres variables, que le risque de
récurrence de TB parmi les patients qui fumaient .10
cigarettes par jour était le double de celui des patients qui
n’avaient jamais fumé ou avaient arrêté. Les autres
facteurs de risque ind épendants significativement
associés à la récurrence de la TB étaient l’absence de
domicile fixe (aHR 3,75 ; IC95% 1,17–12,07), la
présence d’autres pathologies (aHR 2,66 ; IC95%
1,22–5,79) et un frottis positif aux bacilles acido-
résistants (aHR 2,27 ; IC95% 1,47–3,49).
C O N C L U S I O N : Le fait de fumer .10 cigarettes par
jour était significativement associé à la récurrence de la
TB. Nous avons recommandé d’inclure des mesures
efficaces d’arrêt du tabac dans les programmes de lutte
contre la TB afin de réduire ce risque, comme le
préconise la stratégie Stop TB de l’Organisation
Mondiale de la Santé.
RESUMEN
de seguimiento. El modelo de riesgos proporcionales de
Cox puso en evidencia que tras corregir otras variables,
el riesgo de recaı́da en las personas que fumaban .10
cigarrillos por dı́a era dos veces mayor que el riesgo de
las personas que nunca habı́an fumado o que habı́an
abandonado el hábito. Otros factores independientes
que se asociaron de manera significativa con el riesgo de
recaı́da de la TB fueron la carencia de domicilio (aHR
3,75; IC95% 1,17 - 12,07), las enfermedades
concomitantes (aHR 2,66; IC95% 1,22 - 5,79) y una
baciloscopia positiva (aHR 2,27; IC%95 1,47 - 3,49).
C O N C L U S I Ó N : Fumar .10 cigarrillos por dı́a se asoció
de manera significativa con la recaı́da de TB. A fin de
disminuir el riesgo de recaı́da se recomienda integrar a
los programas de lucha contra la TB, medidas eficaces
que favorezcan el abandono del tabaquismo, como lo
promueve la Estrategia Alto a la Tuberculosis de la
Organización Mundial de la Salud.