Journal of Tuberculosis Research, 2014, 2, 173-180

Published Online December 2014 in SciRes. http://www.scirp.org/journal/jtr

http://dx.doi.org/10.4236/jtr.2014.24022

Risk Factors for Hemoptysis in Pulmonary

Tuberculosis Patients from Southern China:
A Retrospective Study
Shouyong Tan1*, Danxiong Sun1,2, Tiantuo Zhang3, Yanqiong Li1, Yuanyuan Cao4,
Moses M. Njire4, Changwei Wang4, Tianyu Zhang4*
1
State Key Laboratory of Respiratory Diseases, Department of Internal Medicine, The Guangzhou Chest
Hospital, Guangzhou, China
2
Respiratory Department, The First People Hospital of Yunnan Province, Kunming, China
3
Department of Internal Medicine, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou,
China
4
State Key Laboratory of Respiratory Diseases, Guangzhou Institutes of Biomedicine and Health, Chinese
Academy of Sciences, Guangzhou, China
Email: *[email protected], *[email protected]

Received 8 October 2014; revised 9 November 2014; accepted 18 November 2014

Copyright 2014 by authors and Scientific Research Publishing Inc.

This work is licensed under the Creative Commons Attribution International License (CC BY).
http://creativecommons.org/licenses/by/4.0/

Abstract
Clinical characteristics of tuberculosis (TB) patients from southern China with pulmonary tuber-
culosis hemoptysis (PTH) were analyzed retrospectively in order to improve the diagnosis of TB,
reduce mortality and prevent the transmission of TB. A total of 1227 cases of pulmonary TB pa-
tients hospitalized in the Third Affiliated Hospital of Sun Yat-sen University and Guangzhou Chest
Hospital from January to December of 2011 were analyzed retrospectively. 1) The male/female
ratio of the 1227 tuberculosis cases was 2.15:1. There were 403 cases (32.8%) of PTH with a male/
female ratio of 3.03:1. 2) The ratio of patients with PTH to those with TB was designated as Rh.
The Rh in the male group (36.2%, 303 cases) was higher than that in the female group (25.6%, 100
cases, risk ratio (RR) = 1.41, P 0.001). 3) The Rh in the elderly group (60 years old, 20.3%, 56
cases) was lower than that in the younger patients group (20 - 39 years old, 45.4%, 189 cases, RR =
2.51, P 0.001). 4) The Rh in initial treatment group (29.6%, 296 cases) was lower than that in the
retreatment group (46.9%, 107 cases, RR = 1.58, P 0.001). 5) The Rh in sputum-positive TB pa-
tients (44.5%, 297 cases) was significantly higher than that in the smear-negative TB patients
(18.9%, 106 cases, RR = 2.35, P 0.001). 6) The Rh of patients with lung lesions range < 3 lung
fields (31.7%, 105 cases) was not significantly different with that of patients with lung lesions
range 3 lung fields (33.3%, 298 cases, RR = 1.05, P = 0.96 > 0.05). 7) The Rh of patients with cavi-
ties (51.8%, 309 cases) was higher than that of patients without cavities (14.9%, 94 cases, RR =
3.48, P 0.001). Male, young, retreated, sputum-positive TB patients and those with cavitary TB
*
Corresponding authors.

How to cite this paper: Tan, S.Y., Sun, D.X., Zhang, T.T., Li, Y.Q., Cao, Y.Y., Njire, M.M., Wang, C.W. and Zhang, T.Y. (2014)
Risk Factors for Hemoptysis in Pulmonary Tuberculosis Patients from Southern China: A Retrospective Study. Journal of
Tuberculosis Research, 2, 173-180. http://dx.doi.org/10.4236/jtr.2014.24022
S. Y. Tan et al.

were more predisposed to PTH in southern China. TB patients with such characteristics should be
sensitized and accorded good care.

Keywords
Tuberculosis, Hemoptysis, Risk Factors

1. Introduction
Hemoptysis is the expectoration or spitting of blood originating from any part of the respiratory tract, usually
from hemorrhage in the lung parenchyma (pulmonary alveoli) and the bronchial arteries [1]. It may be a symp-
tom of several diseases, more or less severe, but its appearance induces concern to the patient and requires a full
diagnostic investigation. Bronchiectasis, pulmonary tuberculosis (TB) and lung cancer are the main causes of
hemoptysis. The frequency of each disease as a cause of hemoptysis varies in different series according to the
geographical area. Pulmonary TB is an important cause of hemoptysis in developing countries [2], whereas, in
developed countries, bronchiectasis, lung cancer and bronchitis are the main causes. TB may cause hemoptysis
either in active state (cavitary lesions, rupture of pulmonary artery aneurysms) or as a late sequelae (rupture of
aneurysms or secondary to bronchiectasis). Rupture of Rasmussens aneurysm can occur with active disease or
as a late finding. It occurs when there is rupture of ectatic portions of the pulmonary arteries traversing thick-
walled cavities [3].
Kralingen et al. [4] from Nepal have reported 63 cases of hemoptysis patients, among which the most com-
mon (65%) patients were infected with TB and its sequelae, followed by pneumonia (17%), bronchitis (13%)
and lung cancer (5%). However, TB is not the main cause of hemoptysis any more in some medium-level de-
veloped countries. Fidan et al. [4] from Turkey reported 108 cases of hemoptysis with lung cancer as the most
common cause (34.3%), followed by bronchiectasis (25.0%), TB (17.6%), pneumonia (10.2%) and pulmonary
embolism (4.6%). In developed countries, bronchiectasis, lung cancer and bronchitis have been the main cause
of hemoptysis mainly due to the good control of TB. In other studies, Porzeziska et al. [5] from Poland retros-
pectively analyzed 431 cases of hemoptysis patients with tumor as the most common cause (40.4%). Tsouma-
kidou et al. [6] analyzed 184 cases of hemoptysis from Greece and found bronchiectasis (26%) and chronic
bronchitis (23%) as the main causes followed by acute bronchitis (15%) and lung cancer (13%). TB is still the
main cause of hemoptysis in some developed countries. For example, Pires et al. [7] reported 221 adult hemop-
tysis patients cases from Portugal and found that the most common cause was TB sequelae (22.2%), followed by
bronchiectasis (15.8%), lung cancer (13.9%), TB (10%) and pneumonia (4.5%). China is a developing country
with high incidence of TB with pulmonary TB being the main cause of hemoptysis.
Typical symptoms of TB may include cough, sputum, hemoptysis companied with anorexia, weakness, night
sweat, chills and fever. The CT image for TB is usually not obvious even with enhanced scan, and most patients
show slight hemoptysis while others could have moderate and even severe hemoptysis [8]. A recent study in
China reported hemoptysis as the second cause of death in pulmonary TB patients [9]. Hemoptysis mainly re-
sults in asphyxiation and sometimes causes hemorrhagic shock or both. Pulmonary tuberculosis hemoptysis (PTH)
can also cause complications, such as TB dissemination and pulmonary atelectasis. In addition, patients with
PTH may transmit TB more easily, as PTH patients have a higher M. tuberculosis-positive rate [10], which has
also been reported in this study. The M. tuberculosis-positive pulmonary TB patients are infectious TB patients.
Therefore, we analyzed the clinical characteristics of PTH patients in southern China in order to improve the
diagnosis and treatment of PTH and to prevent the transmission of TB.

2. Materials and Methods

2.1. Ethical Approval
Individual participants gave written informed consent before enrollment in the study. All the patient information
was routinely collected and recorded by attending physicians. All patient procedures were reviewed and ap-
proved by the Ethics Committees of Guangzhou Chest Hospital and the Third Affiliated Hospital of Sun Yat-sen
University.

174
 S. Y. Tan et al.

2.2. Setting and Study Population

A total of 1227 cases of TB patients hospitalized at the Guangzhou Chest Hospital and the Third Affiliated Hos-
pital of Sun Yat-sen University from January to December of 2011 were analyzed retrospectively. Guangzhou
Chest Hospital is the largest TB treatment center in southern China, which admits about 70,000 patients/year
while the latter is a general hospital, with about 5000 patients admissions with respiratory diseases/year. All pa-
tients were initially treated as inpatients, but some persons were followed on an outpatient basis after hospitali-
zation. Medical records from in-patient and out-patient treatments were reviewed for patients demographics,
TB treatment history, comorbidities, chest radiographs, smear test and M. tuberculosis cultivation. Patients were
enrolled as they needed to be hospitalized and with clear demographic records. Patients with indeterminate test
results, missing data, and data outliers were excluded. All the participants in this study were Han race.

2.3. Criteria and Definitions

1. Inclusion criteria for TB: Compliance with active TB (hereinafter referred to as TB) patients. Exclusion cri-
teria for TB: TB patients with one of the following diseases: lung cancer, nontuberculosis mycobacteria bron-
chiectasis, pulmonary aspergillosis and non-tuberculosis mycobacteria lung disease.
2. TB diagnostic criteria: This was based on the TB diagnosis and treatment guidelines [11].
(1) M. tuberculosis-positive pulmonary TB indicates those that meet one of the following 4 criteria: sputum
smear microscopy for acid-fast bacilli positive twice; sputum smear microscopy for acid-fast bacilli positive
once and active TB lesion showed in chest radiographs; sputum smear microscopy for acid-fast bacilli positive
once and M. tuberculosis culture positive once; sputum smear microscopy for acid-fast bacilli negative but M.
tuberculosis culture positive once and active TB lesion showed in chest radiographs.
(2) Smear-positive pulmonary TB indicates those that meet one of the following 3 criteria: sputum smear mi-
croscopy for acid-fast bacilli positive twice; sputum smear microscopy for acid-fast bacilli positive once and ac-
tive TB lesion showed in chest radiographs. In short, M. tuberculosis-positive pulmonary TB indicates smear-
positive or M. tuberculosis culture-positive; smear-positive pulmonary TB means smear-positive only.
(3) M. tuberculosis-negative pulmonary TB is sputum smear microscopy for acid-fast bacilli negative three
times, M. tuberculosis culture negative once and meet 3 items of the following criteria 1) to 6) or either of 7) or
8). 1) Typical clinical symptoms of TB and chest X-ray findings; 2) Anti-tuberculosis treatment is effective; 3)
Other non-tuberculosis lung diseases were excluded clinically; 4) PPD (5IU) test is strongly positive and serum
anti-tuberculosis antibodies are positive; 5) Sputum M. tuberculosis PCR test and probe detection are positive; 6)
TB confirmed by extra-pulmonary pathology; 7) Acid-fast mycobacteria is detected in bronchial alveolus lavage
(BAL) fluid; 8) Bronchus or lung histopathology was confirmed with TB lesions.
(4) Initial treated pulmonary TB was confirmed with one of the following criteria: 1) Never used anti-tuber-
culosis drug before; 2) Failed to complete the first course of anti-tuberculosis treatment; 3) Irregular chemothe-
rapy less than a month.
(5) Retreatment TB was confirmed with one of the following criteria: 1) Initial treatment failed; 2) Recur-
rence after completion of regular chemotherapy or directly observed treatment short-course (DOTS); 3) New
case but with irregular chemotherapy more than a month; 4) New lesions emerged or recurrence or deterioration
of the rudimental lesion after pulmonary resection.
3. Please refer to the clinical TB diagnosis and treatment guidelines [12] for defining the index of hemoptysis.
1) Small hemoptysis: <100 ml/24hr; 2) Middle hemoptysis: 100 ml and <500 ml/24hr, or 100 - 300 ml/hr; 3)
Heavy hemoptysis: >500 ml/24hr or >300 ml/hr. Here the volume of hemoptysis was calculated according to
biggest volume during a patients course of disease.
4. Exclusion criteria for PTH: Smear and culture were negative and the symptom got better after anti-infla-
mation treatment for 2 weeks.

2.4. Statistical Analysis

Statistical analysis was performed using SPSS 14.0 statistical software (SPSS, Chicago). For qualitative data, 2
test was used. U test was used to compare the sample rate and the overall rate. P < 0.05 was considered statisti-
cally significant.

175
S. Y. Tan et al.

3. Results
3.1. Demographic Characteristics of PTH Patients
Among the 1227 cases of pulmonary TB patients recruited in this study, 837 were males and 390 were females
with a ratio of 2.14. Their gender and age distribution is showed in Table 1. A total of 403 cases were PTH pos-
itive. Male patients had higher Rh (the ratio of patients with PTH/those with TB within the same group, 36.2%)
than that of female patients (25.6%) with a risk ratio (RR) = 1.41, P 0.001. Most PTH patients fell into the 20
age 39 group (46.9%, Table 1) and the number of PTH patients decreased continuously along with the in-
creasing age (>20, Figure 1(a)). TB patients also had a relatively higher Rh at the 20 age 39 group (45.4%,
189 cases) than the older patients at the 60 age group (20.3%, 56 cases, RR = 2.51, P 0.001). The Rh de-
clined continuously along with patients increased ages starting from age 20 (Figure 1(b)).
There were 244 cases (60.55%) of small hemoptysis, 126 cases (31.27%) of middle hemoptysis and 33 cases
(8.19%) of heavy hemoptysis. The constituent ratio of patients with small hemoptysis was significantly higher
than that of patients with middle hemoptysis (2 = 29.313, P < 0.001) and that of patients with heavy hemoptysis
(2 = 14.292, P < 0.001). The constituent ratios of patients with middle hemoptysis and patients with heavy he-
moptysis showed no statistical significance. Among the PTH patients, there were 94 cases (23.33%) combined
with TB bronchiectasis and 75 cases (18.61%) combined with endobronchial TB.

Table 1. Demographic data of 1227 TB patients.

Pulmonary TB
Risk ratio
95% CI P value
(RR)
Hemoptysis (%) N = 403 No hemoptysis (%) N = 824

Male 303 (75.2) 534 (64.8) 1.41

Sex 1.17 - 1.70 <0.001
Female 100 (24.8) 290 (35.2) Reference

Age 19 44 (10.9) 81 (9.8)

(years) 20 - 39 189 (46.9) 227 (27.6) 2.51 1.98 - 3.17 <0.001

40 - 59 114 (28.3) 263 (31.9) 1.67 1.28 - 2.19 <0.001

60 56 (13.9) 253 (30.7) Reference

(a) (b)

Figure 1. a) PTH patients age distribution. The ratio = number of PTH patients in the sub-
group/403 (the total number of PTH patients). b) Rh in different PTH patients age groups.

176
 S. Y. Tan et al.

3.2. Treatment History and Sputum Bacteriological Characteristics of PTH Patients

The treatment history and sputum bacteriological characteristics of the 1227 pulmonary TB patients are shown
in Table 2. Among the 999 (81.4%) pulmonary TB patients in the initial treatment group, there were 296 cases
of PTH (Rh = 29.6%). There were 107 cases of PTH (Rh = 46.9%) among the 228 (18.6%) pulmonary TB pa-
tients in the retreatment group. The Rh in the latter group was significantly higher than that in the former one (2
= 25.189, P < 0.001).
There were 667 sputum-positive TB patients (54.4%) including 297 PTH (Rh = 44.5%) and 560 smear-nega-
tive TB patients (45.6%) including 106 PTH (Rh = 18.9%). The Rh of sputum-positive PTH patients was sig-
nificantly higher than that of the sputum-negative PTH patients (2 = 40.691, P < 0.001).

3.3. Chest Radiographic Characteristics of PTH Patients

The chest radiographic results of all the 1227 pulmonary TB patients are summarized in Table 3. There were
331 (27.0%) TB patients with lung lesion range < 3 lung fields, including 105 PTH patients (Rh = 31.7%). TB
patients with lung lesion range 3 lung fields were 896 (73.0%), including 298 PTH patients (Rh = 33.3%). The
Rh in these 2 groups showed no significant difference, i.e., there was no significant correlation between scope of
pulmonary lesions and hemoptysis (2 = 2.588, P > 0.10). However, Rh in pulmonary TB patients with cavities
is much higher than that in those without cavities (2 = 189.71, P < 0.001).

4. Discussion
There were relatively more hospitalized male TB patients than females (2.15:1). Various reasons could possibly
explain this difference: firstly, men are more exposed to environments contaminated with M. tuberculosis be-
cause of their relatively higher mobility to the work stations and other social places. On the other hand, women
are less mobile as they are often confined to family household chores, thus less exposed to M. tuberculosis. Se-
condly, women occupy the lower socio-economic cadres in the society, and thus not many could be easily found
in the hospitals. Thirdly, the pathogenecity and virulence of M. tuberculosis in humans may be related to gender,
in which men may be more vulnerable to TB than women due to less competent immune potency [13]. Fourthly,
the fact that more smokers are males could partially account for more male TB patients than females. In fact,
recent studies have described an association between tobacco smoke and the risk of developing TB [14] [15].
This study also showed that there were more male PTH patients (3.03:1) than there were females, similar to oth-
er studies reported in the literature [16].

Table 2. 1227 cases of early retreatment TB patients and bacteriological conditions.

Pulmonary TB
Risk ratio
Hemoptysis (%) No hemoptysis (%) 95% CI P value
(RR)
N = 403 N = 824

History of Initial treatment 296 (73.4) 703 (70.4) Reference

pulmonary TB Retreatment 107 (26.6) 121 (29.6) 1.58 1.32 - 1.90 <0.001

Sputum Smear-positive TB 297 (73.7) 370 (44.9) 2.35 1.97 - 2.81 <0.001
bacteriology Smear-negative TB 106 (26.3) 454 (55.1) Reference

Table 3. Imageological change of 1227 cases with TB.

Pulmonary TB
Hemoptysis (%) No hemoptysis (%) Risk ratio 95% CI P value
N = 403 N = 824
<3 lung fields 105 (26.1) 226 (27.4) Reference
Lesion range
3 lung fields 298 (73.9) 598 (72.6) 1.05 0.87 - 1.26 0.96

Cavity 309 (79.6) 287 (34.8) 3.48

Cavity 2.91 - 4.16 <0.001
No cavity 94 (20.4) 537 (65.2) Reference

177
S. Y. Tan et al.

This study also showed that age was possibly an important independent risk factor for PTH, which was con-
sistent with findings from the U.S. by Achkar et al. [17]. Patients with 20 age 39 were more prone to he-
moptysis which could be attributed to the following reasons: firstly, such patients did not pay enough attention
to respiratory symptoms, such as cough and sputum [8]. Most patients could not have been aware of the severity
of hemoptysis until it happened, after which they sought medical attention. On the other hand, the elderly people
undergo regular health examinations for age-related diseases, and thus the probable earlier detection and treat-
ment of lung diseases reducing the incidence of hemoptysis. Secondly, immunity difference between the young
and the elderly TB patients could also explain the findings [18]. People with hypoimmunity are more susceptible
to TB; however, strong immune response is a double-edged sword and could also be detrimental. The human
immune system attenuates with age with the lymphocytes becoming weaker in their function [19].
Heavy and middle hemoptysis are mainly observed in bronchiectasis, TB and relatively rare in other diseases.
Studies from different geographical regions have reported different causes for hemoptysis. In China, Yuan et al.
[20] reported heavy and middle hemoptysis in both bronchiectasis and TB patients, with middle hemoptysis also
occurring in pulmonary TB patients and in low numbers in lung cancer and pneumonia patients. However, Ree-
chaipichitkul et al. [3] from Thailand reported 101 cases of hemoptysis of which bronchiectasis was the main
cause (33.7%), followed by active TB (20.8%) and then lung cancer (10.9%). Pulmonary TB mainly results to
heavy and middle hemoptysis because the TB lesions usually block the blood vessels carrying blood under high
pressure resulting in breakage which can also result in ruptured Rasmussens aneurysm [21] [22]. Wu et al. [23]
reported 62 cases of heavy hemoptysis which were all associated with systemic circulation-pulmonary circula-
tion shunt. In cases of chronic lung infections, tumors, pulmonary embolism and congenital heart and lung dis-
eases, the blood in the pulmonary artery decreases. Therefore, compensatory hyperplasia of bronchial arteries
and anastomosis bronchiectasis increases blood in pulmonary circulation resulting into a systemic circulation-
pulmonary circulation shunt. This is a common phenomenon in pulmonary TB patients which make them have
heavy and middle hemoptysis. Here we showed that male gender and pulmonary TB were the main risk factors
for heavy and middle hemoptysis, which is consistent with an earlier report from Turkey which indicated TB as
an independent predictor of heavy hemoptysis [24].
In this study, hemoptysis accounted for 46.9% of retreated TB patients, and the corresponding Rh was signif-
icantly higher than that of the initial treatment TB group. Firstly, the reason could be that pulmonary TB in the
retreated group had complications, such as bronchiectasis and pulmonary aspergillosis. Bronchiectasis results
from chronic fibro-cavitary TB while pulmonary aspergillosis results from TB cavities being prone to secondary
Aspergillus infection. Secondly, TB patients had attenuated immune function. Therefore, the decontamination
function of bronchial epithelium was weakened, which was helpful for the fungi at the pars oralis pharyngis in-
vading downward into lung tissues easily. Thirdly, rifamycins, aminoglycosides, quinolones and other broad-
spectrum anti-TB drugs could kill both M. tuberculosis and other bacteria and easily lead to dysbacteriosis and
then eventual secondary pulmonary fungal infection [25].
TB cavities usually results from liquefied caseous necrotic lesions. Most TB cavities have collagen and cheese
as the outer and inner layers respectively, and granulation tissues containing a lot of capillaries in between them.
There is incomplete thrombosis in the blood vessels in the middle layer, which can be disrupted by TB infection
leading to hemoptysis and eventually heavy hemoptysis. Most TB cavities have Rasmussens aneurysms and
their rupture often lead to middle and even heavy hemoptysis. For TB patients without cavities, less blood ves-
sels are disrupted and thus less hemoptysis, especially middle and heavy hemoptysis cases. This is concordant
with the findings in this study. Therefore, cavitary TB is an important and very reasonable risk factor for PTH.
Different TB lesions contain different amount of M. tuberculosis bacilli. There are about 102 - 105 bacilli in a
solid TB lesion, while the bacilli in a TB cavity is between 107 and 109 [26]. As caseous TB lesions liquefy, M.
tuberculosis first multiply in macrophages and extracellular bacilli in the solid caseous TB lesions are sup-
pressed. M. tuberculosis can also be neutralized easily by macrophages in the living tissue [27]. Once the ca-
seous TB lesions liquefy, the necrotic lesions can then be exhausted to form cavities. The extracellular M. tu-
berculosis in the solid caseous TB lesions and those in the cavities then proliferate rapidly because of oxy-
gen-rich environments. Large amounts of M. tuberculosis bacilli in the cavities connected with bronchi expelled
outside by cough. In this study, we found that 79.6% of PTH cases with cavities were M. tuberculosis-positive
pulmonary TB patients, while the same scenario was only 54.3% in PTH patients without cavities. The differ-
ence was statistically significant (P < 0.001) which indicated that PTH patients with cavities were more likely M.
tuberculosis-positive. To the best of our knowledge, this is the first study which has performed a systemic anal-

178
 S. Y. Tan et al.

ysis of all the five factors of PTH. The obvious limitations of this study were that the patients were mainly from
southern China, especially Guangdong province and its periphery, and they were only enrolled as inpatients and
with clear demographic records. Multicentric studies including more pulmonary TB patients from different sites,
both inpatients and outpatients, and not only Han race patients should be considered in future studies.
In summary, PTH were more common in young (age 20 to 40 years old), male, sputum-positive, retreated and
cavitary TB patients in southern China. TB patients who are prone to hemoptysis should be sensitized on the
implications of TB infection and the easiness of contracting PTH. Cavitary TB patients are not only prone to
hemoptysis but also M. tuberculosis-positive and thus easier to transmit M. tuberculosis. Therefore, the man-
agement of such patients needs to be strengthened to prevent the rapid and extensive transmission of TB.

5. Conclusion
Hemoptysis is one of the most common symptoms in patients with respiratory disease and may be a life-threat-
ening condition. We intended to assess risk factors associated with hemoptysis in patients with tuberculosis (TB)
or post-TB sequelae in a country where TB burden is relatively high, because active TB has been shown to be a
major etiology of hemoptysis. Besides being a feature of active TB, hemoptysis can also be a manifestation of
complications such as cavitation, fibrosis, bronchiectasis, and mycetoma. Pulmonary tuberculosis hemoptysis is
related to sex, age, pathological change type, blood vessel damage spot and scope. Hemoptysis can suddenly
progress into massive hemoptysis depending on the patients status; therefore energetic cautionary measures
should be implemented right from the onset of bloody expectoration. To achieve this, then a deeper understand-
ing of the inter-related factors associated with pulmonary tuberculosis hemoptysis is needed so as to strengthen
cautionary measures and standardize its diagnosis and treatment.

Acknowledgements
Tianyu Zhang was supported by the Chinese Academy of Sciences One Hundred Talents Program (Category A),
the Key Program of the Chinese Academy of Sciences (KJZD-EW-L02) and the National Great Research Pro-
gram of China (2013ZX10003006).

Authors Contributions
Conceived and designed the experiments: ST, DS, TTZ, TYZ; Performed the experiments: ST, DS, TTZ, YL;
Analyzed the data: ST, DS, TTZ, YL, YC, MMN, CW, TYZ; Wrote the paper: ST, DS, YL, YC, MMN, CW,
TYZ.

References
[1] Pianosi, P. and Al-sadoon, H. (1996) Hemoptysis in Children. Pediatrics in Review, 17, 344-348.
[2] Jean-Baptiste, E. (2000) Clinical Assessment and Management of Massive Hemoptysis. Critical Care Medicine, 28,
1642-1647.
[3] Reechaipichitkul, W. and Latong, S. (2005) Etiology and Treatment Outcomes of Massive Hemoptysis. The Southeast
Asian Journal of Tropical Medicine and Public Health, 36, 474-480.
[4] Van-Kralingen, K., Van Kralingen-Heijboer, A., Zimmerman, M. and Postmus, P. (1995) Management of Hemoptysis
in a Third World City Hospital: A Retrospective Study. Tubercle and Lung Disease, 76, 344-348.
[5] Porzeziska, M., Gorzewska, A., Drozdowski, J., Sulzycka, M. and Somiski, J. (2005) Assessment of Hemoptysis
Etiology among Patients Hospitalized in Pneumonology Departament of Medical University of Gdansk in the Years
1998-2002. Polskie Archiwum Medycyny Wewnetrznej, 114, 658-663.
[6] Tsoumakidou, M., Chrysofakis, G., Tsiligianni, I., Maltezakis, G., Siafakas, N.M. and Tzanakis, N. (2006) A Prospec-
tive Analysis of 184 Hemoptysis CasesDiagnostic Impact of Chest X-Ray, Computed Tomography, Bronchoscopy.
Respiration, 73, 808-814.
[7] Soares, P.F., Teixeira, N., Coelho, F. and Damas, C. (2011) HemoptysisEtiology, Evaluation and Treatment in a
University Hospital. Revista Portuguesa de Pneumologia (English Edition), 17, 7-14.
[8] Uzun, O., Atasoy, Y., Findik, S., Atici, A.G. and Erkan, L. (201) A Prospective Evaluation of Hemoptysis Cases in a
Tertiary Referral Hospital. The Clinical Respiratory Journal, 4, 131-138.
[9] National Technical Steering Group of the Epidemiological Sampling Survey for Tuberculosis Office of the Nationwide

179
S. Y. Tan et al.

Epidemiological Sampling Survey for Tuberculosis (2002) Report on Nationwide Random Survey for the Epidemiol-
ogy of Tuberculosis in 2000. The Journal of the Chinese Antituberculosis Association, 24, 65-66.
http://dx.doi.org/10.3969/j.issn.1000-6621.2002.02.001
[10] Hirshberg, B., Biran, I., Glazer, M. and Kramer, M.R. (1997) Hemoptysis: Etiology, Evaluation, and Outcome in a
Tertiary Referral Hospital. Chest Journal, 112, 440-444.
[11] Tuberculosis Csf. (2001) Tuberculosis Diagnosis and Treatment Guidelines. Chinese Journal of Tuberculosis and Res-
piratory Diseases, 24, 70-74.
[12] Association, C.M. (2005) Clinical Practice Guidelines: Tuberculosis Volume. Peoples Medical Publishing House,
Beijing, 110 p.
[13] Uplekar, M., Rangan, S., Weiss, M., Ogden, J., Borgdorff, M. and Hudelson, P. (2001) Attention to Gender Issues in
Tuberculosis Control. International Journal of Tuberculosis and Lung Disease, 5, 220-224.
[14] Kolappan, C. and Gopi, P. (2002) Tobacco Smoking and Pulmonary Tuberculosis. Thorax, 57, 964-966.
http://dx.doi.org/10.1136/thorax.57.11.964
[15] Bates, M.N., Khalakdina, A., Pai, M., Chang, L., Lessa, F. and Smith, K.R. (2007) Risk of Tuberculosis from Exposure
to Tobacco Smoke: A Systematic Review and Meta-Analysis. Archives of Internal Medicine, 167, 335-342.
http://dx.doi.org/10.1001/archinte.167.4.335
[16] Li, Y. and Gu, Z. (2010) Clinical Analysis of 90 Cases of Pulmonary Tuberculosis Hemoptysis. China Medical Herald,
5, 105.
[17] Achkar, J. and Joseph, G. (2012) Independent Association of Younger Age with Hemoptysis in Adults with Pulmonary
Tuberculosis. International Journal of Tuberculosis and Lung Disease, 16, 897-902.
http://dx.doi.org/10.5588/ijtld.11.0758
[18] Prez-Guzmn, C., Vargas, M.H., Torres-Cruz, A. and Villarreal-Velarde, H. (1999) Does Aging Modify Pulmonary
Tuberculosis? A Meta-Analytical Review. Chest Journal, 116, 961-967. http://dx.doi.org/10.1378/chest.116.4.961
[19] Dorshkind, K. and Swain, S. (2009) Age-Associated Declines in Immune System Development and Function: Causes,
Consequences, and Reversal. Current Opinion in Immunology, 21, 404-407.
http://dx.doi.org/10.1016/j.coi.2009.07.001
[20] Yuan, Y.G.W. (2012) Diagnosis and Treatment of 167 Cases of Patients with Hemoptysis Analysis. Clinical Pulmo-
nary Medicine, 11, 2055-2055.
[21] Irodi, A. and Keshava, S. (2009) Rasmussens AneurysmUndue Importance to an Uncommon Entity? British Jour-
nal of Radiology, 82, 698. http://dx.doi.org/10.1259/bjr/87990037
[22] Zellwegera, C.C., Maillarda, J.O., Christend, G. and Auberta, J.D. (2004) Successful Embolization of Rasmussens
Aneurysm for Severe Haemoptysis. Swiss Medical Weekly, 134, 47-48.
[23] Wu, X.M., Lai, Q., Chen, Y.F., Ren, Y.M., Liang, R.G. and Soong, Y.Q. (2008) Relationship between Broncho-Pul-
monary Shunt and Massive Hemoptysis (a Report of 62 Cases). China Journal of Modern Medicine, 18, 939-943.
[24] Ozgl, M., Turna, A., Yildiz, P., Ertan, E., Kahraman, S. and Yilmaz, V. (2006) Risk Factors and Recurrence Patterns
in 203 Patients with Hemoptysis. Trk Tberkloz ve Toraks Dernei, 54, 243-248.
[25] Goble, M., Iseman, M.D., Madsen, L.A., Waite, D., Ackerson, L. and Horsburgh Jr., C.R. (1993) Treatment of 171 Pa-
tients with Pulmonary Tuberculosis Resistant to Isoniazid and Rifampin. New England Journal of Medicine, 328, 527-
532. http://dx.doi.org/10.1056/NEJM199302253280802
[26] Ma, Y. and Panyu, X. (2006) Tuberculosis. Peoples Health Publishing House, Beijing, 71.
[27] Sopko, D.R. and Smith, T.P. (2011) Bronchial Artery Embolization for Hemoptysis. Seminars in Interventional Radi-
ology, 28, 48-62. http://dx.doi.org/10.1055/s-0031-1273940

180