Assessment of mycobacteremia detection as a complementary method for the diagnosis of tuberculosis in HIV-infected patients.

Abstract

The purpose of this investigation was to assess the usefulness of mycobacteremia detection in human immunodeficiency virus (HIV) patients with suspected tuberculosis. The study included 47 patients with suspected tuberculosis and confirmed HIV infection. A first blood sample was incubated in a BACTEC 9050 MB system, while white blood cells isolation was performed on a second blood specimen before incubation in a BACTEC MGIT 960 system. The third specimen was taken from the affected organs of each patient according to their clinical profile. Twelve (25.5%) patients were positive for mycobacterial infection identified by any of the methods used. Ten (21.2%) were positive for Mycobacterium tuberculosis and 2 (4.3%) for M. avium. Six patients were diagnosed by the culture of specimen from affected organs only, whilst three other patients were positive exclusively for blood cultures. Three additional patients were diagnosed by both methods. Four patients with negative cultures were ultimately diagnosed with tuberculosis by measuring the adenosine deaminase levels. Mycobacteremia detection can be used to increase the sensitivity of the diagnosis of tuberculosis and other mycobacteria in patients with HIV. However, it cannot be used as the sole diagnostic method. Clinical specimen cultures do not provide 100% diagnostic accuracy and it is, therefore, critical to further improve the mycobacteria detection sensitivity.

Full text

Page 1

ARTICLE
Assessment of mycobacteremia detection
as a complementary method for the diagnosis
of tuberculosis in HIV-infected patients
J. Hernández & A. Jaramillo & G. I. Mejía & P. Barón &
V. Gomez & M. A. Restrepo & J. Robledo
Received: 19 January 2010 /Accepted: 28 July 2010 /Published online: 24 August 2010
# Springer-Verlag 2010
Abstract The purpose of this investigation was to assess
the usefulness of mycobacteremia detection in human
immunodeficiency virus (HIV) patients with suspected
tuberculosis. The study included 47 patients with suspected
tuberculosis and confirmed HIV infection. A first blood
sample was incubated in a BACTEC 9050 MB system,
while white blood cells isolation was performed on a
second blood specimen before incubation in a BACTEC
MGIT 960 system. The third specimen was taken from the
affected organs of each patient according to their clinical
profile. Twelve (25.5%) patients were positive for myco-
bacterial infection identified by any of the methods used.
Ten (21.2%) were positive for Mycobacterium tuberculosis
and 2 (4.3%) for M. avium. Six patients were diagnosed by
the culture of specimen from affected organs only, whilst
three other patients were positive exclusively for blood
cultures. Three additional patients were diagnosed by
both methods. Four patients with negative cultures were
ultimately diagnosed with tuberculosis by measuring the
adenosine deaminase levels. Mycobacteremia detection can
be used to increase the sensitivity of the diagnosis of
tuberculosis and other mycobacteria in patients with HIV.
However, it cannot be used as the sole diagnostic method.
Clinical specimen cultures do not provide 100% diagnostic
accuracy and it is, therefore, critical to further improve the
mycobacteria detection sensitivity.
Introduction
The increase in the global burden of tuberculosis (TB) in
recent decades led the World Health Organization (WHO)
to declare TB a global emergency in 1993. It is estimated
that one third of the world population is infected, making
TB an important cause of morbidity and mortality of
infectious origin [1]. The WHO estimates that 9.27 million
new cases of TB occurred in 2007 (139 per 100,000
population) [2]. The global case detection rate only reached
60% and the cure rate was 84%. Despite the implementation
of health policies to control the disease, factors such as
demographic growth, poverty, overcrowding, and homeless-
ness contribute to this problem [2].
Other factors that worsen the picture are the increase
in resistance to TB treatment regimen and the frequent
co-occurrence with human immunodeficiency virus (HIV)
[2]. The HIV epidemic and the global burden of TB are
closely intertwined. Among the 9.27 million incident cases
of TB in 2007, an estimated 1.37 million (14.8%) were
HIV-positive [2]. The majority of these co-infected people
were in resource-constrained countries. TB is the most
common cause of death in HIV-infected patients in the
developing world and accounts for up to a third of acquired
immunodeficiency syndrome (AIDS) deaths worldwide. It
accelerates progression of the clinical course of HIV
infection, leading to additional opportunistic infections
and death [3–5]. HIV greatly increases the likelihood of
both reactivating latent TB and progression to active TB
after primary infection, as people infected with HIV have a
J. Hernández:G. I. Mejía:P. Barón:V. Gomez:
M. A. Restrepo:J. Robledo (*)
Unidad de Bacteriología y Micobacterias,
Corporación para Investigaciones Biológicas (CIB),
Cra 72A No. 78B-141,
Medellín, Colombia
e-mail: jrobledo@cib.org.co
J. Hernández:A. Jaramillo:G. I. Mejía:J. Robledo
Escuela de Ciencias de la Salud,
Universidad Pontificia Bolivariana,
Medellín, Colombia
Eur J Clin Microbiol Infect Dis (2010) 29:1435–1441
DOI 10.1007/s10096-010-1023-y

Page 2

10- to 50-fold risk of TB either as reactivation or a new
infection compared to patients without HIV [6, 7].
In Colombia, there are close to 200,000 people living
with HIV and, although the incidence of the disease is
considered to be low (0.7%), the epidemic continues to rise
since 1983 when the first case was reported [8].
Although the same basic principles apply for patients with
and without HIV, the clinical manifestations of TB vary in
patients with HIV as the degree of immunosuppression
progresses.AstheCD4countdecreases,pulmonarycavitations
occur less frequently and miliary and extrapulmonary mani-
festations predominate. These specific characteristics make
diagnosis difficult, delay the initiation of proper treatment, and
increase the mortality to 30% [9, 10]. The “gold standard” for
the diagnosis of TB is the microbiological demonstration of
acid-fast bacilli (AFB) in body samples. The tests aimed at
identifying Mycobacterium tuberculosis are the same in
patients with and without HIV, but patients with HIV
frequently require more cumbersome methods, such as
biopsies, bone marrow aspirate, and blood cultures [11, 12].
The identification of M. tuberculosis and other mycobac-
teria from blood cultures has been very useful in patients
with HIV/AIDS. Various methods have been used for blood
cultures, such as MYCO/F LYTIC, BacT/ALERT MB, and
ISOLATOR 10 [13–16]. Several trials have reported the
usefulness of the isolation of mycobacteria from the blood of
HIV-positive patients. The rate of positivity depends on both
the incidence of HIV and TB, and can vary between 5 and
30%. Frequently, positive blood cultures are the only way to
make the diagnosis [17–19].
The goal of this study was to determine the presence of
mycobacteria in the peripheral blood of patients with HIV
and clinical suspicion of TB by microbiological demon-
stration of M. tuberculosis in two blood samples processed
by different methods compared with the conventional
cultures taken from the affected body organs.
Methods
Patients and samples collection
Our study was conducted at the Corporación para Inves-
tigaciones Biológicas (CIB) and the Hospital La Maria in
Medellin, Colombia, and was approved by the research
ethics committee of both institutions. Adult patients who
presented to the Hospital La Maria between June 1st 2006
and January 30th 2007 were eligible for inclusion in the
study. Each patient was interviewed, the study was
explained, and informed consent obtained. Inclusion criteria
were positive confirmatory test for HIV (by Western blot)
and clinical suspicion for pulmonary or extrapulmonary TB
and not being on treatment for TB. The clinical variables
considered were age, gender, time since HIV diagnosis, and
whether death occurred during the hospitalization. All of
the information was stored in a database and processed
using Epi Info 6 and DATA SETS (Centers for Disease
Control and Prevention [CDC], Atlanta, GA, USA).
Whole-blood culture
The samples processing is described in Fig. 1. Two blood
samples were taken from each patient. First, a 5-mL sample
of venous blood was inoculated into a BACTEC MYCO/F
LYTIC culture bottle and incubated in a BACTEC 9050 MB
(Becton Dickinson, Franklin Lakes, NJ, USA) instrument for
a maximum of 56 days and discarded if the result was
negative. If the culture was positive, the sample underwent a
process of isolation and identification of mycobacteria using
standard methods [20]. Briefly, Auramine–Rhodamine and
Kinyoun (when Auramine–Rhodamine was positive) stain-
ings for AFB identification were performed, followed by
inoculation in Middlebrook 7H11 Thin Layer Agar (Mdk
7H11-TLA) incubated in 6% CO2at 37°C for 4 weeks and
in Lowenstein–Jensen medium (LJ) at 37°C for 8 weeks.
Mycobacterial identification was carried out using previously
described standard procedures [21].
Processing and culture of WBC
The second sample of venous blood was obtained using a
Vacutainer CPT (Becton Dickinson, Franklin Lakes, NJ,
USA). These specimens were centrifugated for 20 min at
2,400 rpm at room temperature within the first 2 h following
venipuncture. After centrifugation, the thin layer of WBC was
collected and half was treated with a 30 mg/mL saponin
solution for 2 min at room temperature for the lysis of the
WBC. Saponin-treated and nontreated WBC were further
inoculated simultaneously into two liquid Middlebrook 7H9
(Mdk 7H9) media. The first is the fully automated MGIT 960
(Becton Dickinson, Franklin Lakes, NJ, USA) supplemented
with PANTA (polymyxin B, amphotericin B, nalidixic acid,
trimethoprim, and Azlocillin)and OADC (oleic acid, albumin,
dextrose, and catalase), both provided in the BD BACTEC™
MGIT™ 960 Supplement Kit. The second was an inhouse-
produced Mdk 7H9 supplemented with OADC only. Tubes
indicating growth in any of those two media were subcultured
onto Mdk 7H11-TLA plates [20]. Colonies growing in Mdk
7H11-TLAwere stainedwith Kinyounbeforea final biochem-
ical identification was performed as previously described.
Samples from affected organs
The third sample was taken from the affected organs. A
sputum specimen was systematically obtained from each
patient. If positive for mycobacteria, no further sample was
1436Eur J Clin Microbiol Infect Dis (2010) 29:1435–1441

Page 3

taken. If negative, additional specimens were taken for further
processing (bronco-alveolar lavage, cerebrospinal fluid
[CSF], pleural fluid, lymph node biopsy, testicle biopsy,
peritoneal liquid, and feces) according to the clinical
presentation of the disease. Respiratory nonsterile specimen
(sputum and bronco-alveolar lavage) were decontaminated
with a solution of N-acetyl-cysteine, sodium citrate 2.94%,
and sodium hydroxide 2%. Feces specimens were decontami-
nated with a 4% sodium hydroxide solution [21]. Neutraliza-
tion was obtained with phosphate-buffered saline. Smears of
each sample were stained for AFB and consequently
inoculated in LJ and MGIT 960 media. Colonies growing
in MGIT 960 were subcultured onto Mdk 7H11-TLA plates
[20]. The final identification of mycobacterial species was
performed as described for the blood samples.
Isolates characterization
AFB identification from the smears was performed according
to recommended methods [21]. The characteristics of M.
tuberculosis colonies in Mdk 7H11-TLA were identified by
initial morphology considering consistency and the tendency
to form cords in liquid and solid media [22–24]. The Mdk
7H11-TLA and LJ were observed twice a week for 4 and
8 weeks, respectively. The final identification of the colonies
in either medium was confirmed by growth and biochemical
tests as previously described [24, 25].
Patients’ final diagnosis
The final diagnosis of every patient was established by the
medical staff of the Hospital La Maria based on clinical
findings, microbiological confirmation, and additional tests,
such as adenosine deaminase (ADA) in pleural fluid [26].
Results
Clinical details of the patients
During the study period, 55 patients with confirmed HIV
infection and clinical suspicion for pulmonary or extrapulmo-
nary TB were included. Eight patients were excluded because
they did not meet the inclusion criteria; therefore, a total of 47
5 mL Blood
n = 47
Patients
HIV (+)
TB suspicious
BACTEC 9050 MB
WBC Isolation
n = 47
+
-
Discard
Mycobacteria Species
Identification
Mdk 7H11
L J
Auramine
Rhodamine /
Kinyoun
+
+
Saponin
No Saponin
Mdk 7H11
Mdk 7H9
MGIT 960 BD
+
Kinyoun
Mycobacteria Species
Identification
Sputum*
Broncho-alveolar liquid*
CSF
Pleural fluid
Peritoneal liquid
Lymph node biopsy
Testicle biopsy
Feces*
Decontamination*
Kinyoun /
Auramine
Rhodamine
LJ
BACTEC MGIT 960 BD
Mdk 7H11
Mycobacteria Species
Identification
+ +
+
+
5-8 mL Blood Other Samples
n = 47
n = 25
n = 11
n = 6
n = 2
n = 1
n = 1
n = 1
Fig. 1 Flow chart of the samples processing applied to the 47 patients
positive for HIV. A 5-mL whole-blood sample taken from each patient
was cultivated in a BACTEC 9050 MB system. If positive, LJ and
Mdk 7H11-TLA cultures were continued, followed by a Kinyoun and
Auramine–Rhodamine staining (left). WBC were isolated from the
second blood sample before being lysed with saponin. Saponin-treated
and control WBC were cultured in MGIT 960, Mdk 7H9, and Mdk
7H11-TLA media. Positive isolates were stained with Kinyoun and
Auramine–Rhodamine (center). Additionally, samples were taken from
the affected organs after clinical evaluation. Nonsterile sputum,
broncho-alveolar liquid, and feces were decontaminated with sodium
hydroxide and N-acetylcysteine (*). All samples were stained with
Kinyoun and Auramine–Rhodamine and further cultured simulta-
neously on LJ, MGIT 960, and Mdk 7H11-TLA media (right). For all
three types of samples, mycobacteria species identification was
performed according to described procedures
Eur J Clin Microbiol Infect Dis (2010) 29:1435–14411437

Page 4

patients were analyzed. Men accounted for 41 patients (88%)
and 6 (12%) were women. The average age was 37 years and
the mean time since the diagnosis of HIV was 35.5 months
(varying from 1 to 120 months). Thirteen (28%) patients died
during the study.
Mycobacterial detection from pulmonary
and extrapulmonary clinical specimens
The clinical specimens collected (bronco-alveolar lavage,
CSF, pleural fluid, lymph node biopsy, testicle biopsy,
peritoneal liquid, and feces) were processed for mycobac-
terial identification (Table 1). Three sputa were positive for
M. tuberculosis and three additional patients had a positive
M. tuberculosis culture in pleural fluid. The testicular
biopsy of a seventh patient was positive and M. tuberculosis
was detected in the broncho-alveolar liquid culture of an
eighth subject. M. avium was detected for only one patient
whom peritoneal fluid and feces cultures allowed the
isolation of the mycobacterium.
Detection of mycobacteremia in blood samples
of HIV-positive patients
The culture of whole-blood samples allowed the detection
of M. tuberculosis in 4 (9%) cases and M. avium in 2 (4%)
cases (Table 2). WBC isolation allowed the detection of M.
tuberculosis in only three patients and M. avium was
characterized in isolates from two patients. An assessment
of the effect of the WBC lysis by saponin was also performed
but it did not provide any significant improvement.
Patients’ final diagnosis
A final diagnosis of each patient was established and reported
according to the determined etiological factor (if determined)
and the type of disease (Table 3). TB was diagnosed in 14 of
the patients, while two had an M. avium mycobacteremia.
Pneumonia was diagnosed in eight patients.
Discussion
The total number of patients in whom mycobacteria were
isolated with any method was 12 (25.5%), ten of which were
infected by M. tuberculosis and two by M. avium. Myco-
bacteremia was detected in six patients (12.6%). Such a rate
of mycobacteremia is within the range of 3 to 50% described
in the literature [5, 17, 18, 27, 28]. Of the 12 patients who
presented a mycobacterial infection, blood culture was the
only method that allowed a proper diagnosis for three of
them. Inversely, the presence of mycobacteria could not be
detected in blood but only in a specimen from affected
organs for six patients. Finally, mycobacteria could be
detected by both blood cultures and traditional clinical
specimens simultaneously in only three patients. These data
point out a crucial disparity in the sensitivity and accuracy of
mycobacteria detection methods.
Traditionally, sputum is the first clinical sample taken
from a patient presenting pulmonary TB symptoms. Those
samples were positive for M. tuberculosis in only three
patients, which represents 30% of the patients in whom this
mycobacterium was ultimately detected. We collected
samples from other organs depending on the clinical
symptoms. Overall, clinical specimens taken from affected
organs allowed the detection of eight cases of TB. This
represents a rate of 80% if we consider the number of
patients in which M. tuberculosis was eventually detected
by any method. The last two patients in whom M.
tuberculosis was ultimately detected had no positive
sputum culture. However, both their whole-blood and
WBC cultures turned out to be positive. In this particular
case, mycobacteremia detection allowed to diagnose
patients who would have been considered to be negative
otherwise, improving the rate of M. tuberculosis detection
in two cases out of ten (20%).
Cultures of clinical samples taken from the affected organs
cannot be considered 100% sensitive, and, therefore, cannot
Table 1 Mycobacteria detection rates in affected organs. The number
of positive specimens for AFB smear and culture by any of the
performed methods are given
Clinical specimenPositives
M. tuberculosis M. avium
Sputum, n=47
Pleural fluid, n=6
Peritoneal fluid, n=2
Testicular edema, n=1
Lymph node, n=1
Broncho-alveolar liquid, n=25
CSF, n=11
Feces, n=1
3 (6.3%)
3 (50%)
0
1 (100%)
0
1 (4%)
0
0
0
0
1 (50%)
0
0
0
0
1 (100%)
Table 2 Mycobacteria detection rates in whole-blood and WBC
cultures. WBC treated with saponin and nontreated were cultured in
MGIT 960, Mdk 7H9, and Mdk 7H11-TLA media
Clinical specimen Positives
M. tuberculosisM. avium
Whole blood
WBC
4 (9%)
2 (16.6%)
3 (25%)
2 (4%)
2 (16.6%)
2 (16.6%)
Saponin
No saponin
1438 Eur J Clin Microbiol Infect Dis (2010) 29:1435–1441

Page 5

be recommended as the sole diagnostic method, and neither
can blood cultures. Whole-blood cultures allowed the detec-
tion of M. tuberculosis in only four patients out of the ten
who were eventually found to be positive. This represents a
rate of 40%, which is half of the positivity rate given by
organ samples cultures. WBC isolation and subsequent
culture did not allow the detection of any additional case.
It even proved to be less sensitive, as one case detected by
whole-blood culture could not be detected by WBC analysis.
WBC lysis with saponin, surprisingly, indicated that the
treatment decreased the sensitivity of the method, as one case
detected without WBC lysis could not be detected after lysis.
M. avium is also an important cause of morbidity and
mortality in patients with HIV. If M. avium was detected in
only two patients, neither of them had a positive sputum.
The first was diagnosed by feces and peritoneal fluid
cultures, as well as whole-blood and WBC cultures, whilst
the second patient had only his whole-blood and WBC
cultures resulting positive. Even though the number of M.
avium-positive patients is low, our results indicate that
blood culture doubled the mycobacteria detection rate.
Again, this sheds light on the fact that, if neither of the two
methods should be recommended as the sole diagnostic
tool, they appear to be complementary.
If we compare the time to the final result, whole-blood
culture was the method that took the longest to yield a
positive result (40 days ± 12). Cultures of clinical specimens
from the affected organs gave results in an average of 33 days
(±2). WBC isolation and cultures needed 24 days (±18) to
reveal a mycobacteria growth. If those methods were equally
accurate and sensitive, WBC analysis would clearly stand out
as being the fastest, particularly when the characterization is
achieved using the TLA method. Nevertheless, its use as a
complementary method can provide a preliminary and rapid
estimateofthefinaloutcome,anditcanbeextremelyusefulto
quickly start the appropriate treatment in positive patients.
The final diagnosis ofthe 47patientswas performed bythe
Hospital La Maria physicians. Ultimately, pulmonary and
extrapulmonary TB were diagnosed in 14 patients, which
means that four of them had been missed by every culture
method described in the present study. Those four patients
were eventually diagnosed by the measurement of ADA
levels as recommended by other studies [26, 29–32]. The
etiological factor could be identified in the majority of the
cases (Table 3). In countries with an important TB burden,
M. tuberculosis is a frequent cause of community-acquired
pneumonia (CAP) [33–35]. In a recent study in sub-Saharan
Africa, M. tuberculosis was the leading cause of CAP and
reflected the worsening TB epidemic in the region [36].
Also, the spectrum of HIV-associated opportunistic pneumo-
nias is broad and includes bacterial, mycobacterial, fungal,
viral, and parasitic pneumonias. Although pneumonias due
to Cryptococcus neoformans, Histoplasma capsulatum,
Coccidioides immitis, cytomegalovirus, and Toxoplasma
gondii are less frequent, their presence in the lung is often
indicative of disseminated disease and is associated with
significant mortality [37]. Weight loss was associated with
other clinical outcomes in seven patients. This finding
correlates with several reports of various forms of TB in
which a significant correlation between severe weight loss
and TB was reported [38–40].
A study evaluating the utility of blood cultures for
mycobacteria was performed in South Africa and included
71patientswithsuspectedTB,showingthat16(22%)patients
Etiological factorFinal clinical diagnosisClinical category
Candida albicansEsophageal candidiasis (1)
Oral candidiasis (1)
Cryptococcosis (2)
Histoplasmosis (5)
Herpes zoster (1)
Toxoplasmosis (2)
Latent syphilis (1)
Mycobacteriosis (2)
Tuberculosis (14)
Pneumonia (5)
Community-acquired pneumonia (2)
Necrotizing pneumonia (1)
Hodgkin’s lymphoma (1)
Kaposi’s sarcoma (1)
Weight loss (7)
Malnutrition (1)
Hematophagocytic syndrome (1)
Uveitis (1)
Fungal infection
Cryptococcus neoformans
Histoplasma capsulatum
Varicella zoster
Toxoplasma gondii
Treponema pallidum
Mycobacterium avium
Mycobacterium tuberculosis
Pneumocystis jiroveci
Undetermined
Viral disease
Parasitic disease
Bacterial infection
Pneumonia
Cancer
Physiological syndrome
Others
Table 3 Final diagnosis in 47
patients with confirmed HIVand
presumptive TB diagnosis. A
classification is established
according to the etiological
factor (when determined), the
final clinical diagnosis, and the
clinical category of the disease
Eur J Clin Microbiol Infect Dis (2010) 29:1435–14411439

Page 6

were bacteremic with M. tuberculosis and 7 (10%) with M.
avium [17]. All of the patients with mycobacteremia were
infected with HIV. Meanwhile, in Brazil, a study that
included 80 patients with HIV identified 7 (8.8%) with
mycobacteremia due to M. avium and 5 (6.2%) due to M.
tuberculosis [41]. In our study, we found more cases of
mycobacteremia due to M. tuberculosis than due to M. avium
(4.2%). These results clearly underscore the importance of
searching for other mycobacteria different from M. tubercu-
losis, especially in patients with HIV.
Fandinho et al. published a study in which three serial
blood samples were drawn for mycobacteremia in 50
patients with HIV and clinical suspicion of TB or
disseminated mycobacterial infection, and they obtained
results similar to studies which took only one sample [42].
Our results strongly support those findings and allow us to
formulate the same recommendations, as our study shows
that WBC cultures did not improve mycobacteria detection
compared to whole-blood culture. However, the detection
of mycobacteria in the blood of patients with HIV proved to
be very useful in improving the timely diagnosis of TB/
HIV coinfection and infection by other mycobacteria.
Acknowledgments
Guzman, and Elsa Zapata in the startup process of the present work, as
well as Nidia Correa, Carlos A. Agudelo, and Angela M. Tobon who
assessed this project. Also, we wish to acknowledge the support of the
medical staff from Hospital La Maria Dinamica Clinical Laboratory,
Hospital Pablo Tobon Uribe, Clinica Universitaria Bolivariana, and
Universidad Pontificia Bolivariana. This work was financed by CIDI
(797-11/05-44) from the Universidad Pontificia Bolivariana and the
Corporación para Investigaciones Biológicas (CIB).
The authors wish to thank Natalia Builes, Angela
References
1. World Health Organization (WHO) (2004) Global tuberculosis
control—surveillance, planning, financing. Fact sheet no. 104.
WHO, Geneva, Switzerland
2. World Health Organization (WHO) (2009) Global tuberculosis
control—epidemiology, strategy, financing. WHO Report 2009
WHO/HTM/TB/2009.411. Available online at: http://www.who.
int/tb/publications/global_report/2009/en
3. Lodha R, Kabra SK (2004) Newer diagnostic modalities for
tuberculosis. Indian J Pediatr 71(3):221–227
4. Centers For Disease Control and Prevention (CDC) (2000)
Update: nucleic acid amplification tests for tuberculosis. MMWR
Morb Mortal Wkly Rep 49(26):593–594
5. David ST, Mukundan U, Brahmadathan KN, John TJ (2004)
Detecting mycobacteraemia for diagnosing tuberculosis. Indian J
Med Res 119(6):259–266
6. World Health Organization (WHO) (2004) TB/HIV: a clinical
manual, 2nd edn. Available online at: http://whqlibdoc.who.int/
publications/2004/9241546344.pdf
7. Kirk O, Gatell JM, Mocroft A, Pedersen C, Proenca R, Brettle RP,
BartonSE,SudreP,PhillipsAN,LundgrenJD(2000)Infectionswith
Mycobacterium tuberculosis and Mycobacterium avium among
HIV-infected patients after the introduction of highly active
antiretroviral therapy. Am J Respir Crit Care Med 162:865–872
8. Joint United Nations Programme on HIV/AIDS (UNAIDS) (2009)
UNAIDS second independent evaluation 2002–2008. Final report.
Available online at: http://data.unaids.org/pub/BaseDocument/
2009/20091002_sie_final_report_en.pdf.
9. Havlir DV, Barnes PF (1999) Tuberculosis in patients with human
immunodeficiency virus infection. N Engl J Med 340:367–373
10. Lawn SD, Myer L, Orrell C, Bekker LG, Wood R (2005) Early
mortality among adults accessing a community-based antiretro-
viral service in South Africa: implications for programme design.
AIDS 19:2141–2148
11. Chacko S, John TJ, Babu PG, Jacob M, Kaur A, Mathai D (1995)
Clinical profile of AIDS in India: a review of 61 cases. J Assoc
Physicians India 43:(8):535–538
12. Cohn DL (2005) Subclinical tuberculosis in HIV-infected patients:
another challenge for the diagnosis of tuberculosis in high-burden
countries? Clin Infect Dis 40(10):1508–1510
13. Hanna BA, Walters SB, Bonk SJ, Tick LJ (1995) Recovery of
mycobacteria from blood in mycobacteria growth indicator tube
and Lowenstein–Jensen slant after lysis-centrifugation. J Clin
Microbiol 33(12):3315–3316
14. Hanscheid T, Monteiro C, Cristino JM, Lito LM, Salgado MJ
(2005) Growth of Mycobacterium tuberculosis in conventional
BacT/ALERT FA blood culture bottles allows reliable diagnosis
of mycobacteremia. J Clin Microbiol 43(2):890–891
15. Esteban J, de Gorgolas M, Santos-O’Connor F, Gadea I,
Fernández-Roblas R, Soriano F (2001) Mycobacterium tubercu-
losis bacteremia in a university hospital. Int J Tuberc Lung Dis 5
(8):763–768
16. McDonald LC, Archibald LK, Rheanpumikankit S, Tansuphas-
wadikul S, Eampokalap B, Nwanyanawu O, Kazembe P, Dobbie
H, Reller LB, Jarvis WR (1999) Unrecognised Mycobacterium
tuberculosis bacteraemia among hospital inpatients in less
developed countries. Lancet 354:1159–1163
17. von Gottberg A, Sacks L, Machala S, Blumberg L (2001) Utility
of blood cultures and incidence of mycobacteremia in patients
with suspected tuberculosis in a South African infectious disease
referral hospital. Int J Tuberc Lung Dis 5:80–86
18. Grinsztejn B, Fandinho FCO, Veloso VG, João EC, Lourenço
MC, Nogueira SA, Fonseca LS, Werneck-Barroso E (1997)
Mycobacteremia in patients with the acquired immunodeficiency
syndrome. Arch Intern Med 157:2359–2363
19. Murcia MI, León CI, de la Hoz F, Saravia J (2007) Asociación
micobacterias-VIH/SIDA en pacientes atendidos en un Hospital
Universitario en Bogotá, Colombia. Rev Salud Pública 9(1):97–105
20. Mejia GI, Castrillon L, Trujillo H, Robledo JA (1999) Microcolony
detection in 7H11 thin layer culture is an alternative for rapid
diagnosis of Mycobacterium tuberculosis infection. Int J Tuberc
Lung Dis 3(2):138–142
21. Kent BD, Kubica GP (1985) Public health mycobacteriology. A
guide for the level III laboratory. Centers for Diseases Control
(CDC), Atlanta, GA, pp 36–39, 47–69, 185–187
22. Welch DF, Guruswamy AP, Sides SJ, Shaw CH, Gilchrist MJ
(1993) Timely culture for mycobacteria which utilizes a micro-
colony method. J Clin Microbiol 31:2178–2184
23. Runyon EH (1970) Identification of mycobacterial pathogens
utilizing colony characteristics. Am J Clin Pathol 54:578–586
24. Leão SC, Martin A, Mejia GI, Palomino JC, Robledo J, Telles MAS,
Portaels F (2004) Practical handbook for the phenotypic and
genotypic identification of mycobacteria. European Commision
International Cooperation for Developing Countries, Concerted
Action-proyect ICA-4-CT-2001-10087, Brussels
25. Correa NE, Cataño JC, Mejía GI, Realpe T, Orozco B, Estrada S,
Vélez A, Vélez L, Barón P, Guzmán A, Robledo J (2010)
Outbreak of mesotherapy-associated cutaneous infections caused
by Mycobacterium chelonae in Colombia. Jpn J Infect Dis 63
(2):143–145
1440 Eur J Clin Microbiol Infect Dis (2010) 29:1435–1441

Page 7

26. Burgess LJ, Maritz FJ, Le Roux I, Taljaard JJ (1995) Use of
adenosine deaminase as a diagnostic tool for tuberculous pleurisy.
Thorax 50(6):672–674
27. Bouza E, Díaz-López MD, Moreno S, Bernaldo de Quirós JC,
Vicente T, Berenguer J (1993) Mycobacterium tuberculosis
bacteremia in patients with and without human immunodeficiency
virus infection. Ann Intern Med 153:496–500
28. Bacha HA, Cimerman S, de Souza SA, Hadad DJ, Mendes CM
(2004) Prevalence of mycobacteremia in patients with AIDS and
persistent fever. Braz J Infect Dis 8(4):290–295
29. Mathur PC, Tiwari KK, Trikha S, Tiwari D (2006) Diagnostic
value of adenosine deaminase (ADA) in tubercular serositis.
Indian J Tuberc 53:92–95
30. Strankinga WF, Nauta JJ, Straub JP, Stam J (1987) Adenosine
deaminase activity in tuberculous pleural effusions: a diagnostic
test. Tubercle 68(2):137–140
31. Shibagaki T, Hasegawa Y, Saito H, Yamori S, Shimokata K
(1996) Adenosine deaminase isozymes in tuberculous pleural
effusion. J Lab Clin Med 127(4):348–352
32. Koh KK, Kim EJ, Cho CH, Choi MJ, Cho SK, Kim SS, Kim MH,
Lee CJ, Jin SH, Kim JM, Nam HS, Lee YH (1994) Adenosine
deaminase and carcinoembryonic antigen in pericardial effusion
diagnosis,especially insuspectedtuberculouspericarditis.Circulation
89(6):2728–2735
33. Ishida T (2000) Etiology of community-acquired pneumonia
among adult patients in Japan. Jpn J Antibiot 53(Suppl B):3–12
34. Scott JA, Hall AJ, Muyodi C, Lowe B, Ross M, Chohan B,
Mandaliya K, Getambu E, Gleeson F, Drobniewski F, Marsh K
(2000) Aetiology, outcome, and risk factors for mortality among
adults with acute pneumonia in Kenya. Lancet 355:1225–1230
35. Liam CK, Pang YK, Poosparajah S (2006) Pulmonary tuberculo-
sis presenting as community-acquired pneumonia. Respirology
11:786–792
36. Nyamande K, Lalloo UG, John M (2007) TB presenting as
community-acquired pneumonia in a setting of high TB incidence
and high HIV prevalence. Int J Tuberc Lung Dis 11(12):1308–1313
37. Huang L, Crothers K (2009) HIV-associated opportunistic
pneumonias. Respirology 14(4):474–485
38. Mahuad C, Bozza V, Pezzotto SM, Bay ML, Besedovsky H, del
Rey A, Bottasso O (2007) Impaired immune responses in
tuberculosis patients are related to weight loss that coexists with
an immunoendocrine imbalance. Neuroimmunomodulation 14(3–
4):193–199
39. Sari R, Bayindir Y, Sevinc A, Bahceci F, Ozen S (2002) The triad
of weight loss, fever and night sweating: isolated bone marrow
tuberculosis, a case report. J Chemother 14(4):420–422
40. Hira SK, Dupont HL, Lanjewar DN, Dholakia YN (1998) Severe
weight loss: the predominant clinical presentation of tuberculosis
in patients with HIV infection in India. Natl Med J India 11
(6):256–258
41. Nakatani SM, Messias-Reason IJ, Burger M, Cunha CA (2005)
Prevalence of Mycobacterium avium and Mycobacterium tuber-
culosis in blood cultures of Brazilian AIDS patients after
introduction of highly active retroviral therapy. Braz J Infect Dis
9(6):459–463
42. Fandinho FC, Grinsztejn B, Veloso VG, Lourenço MC, Werneck-
Barroso E, João E, Nogueira SA, de Fonseca LS (1997) Diagnosis
of disseminated mycobacterial infection: testing a simple and
inexpensive method for use in developing countries. Bull World
Health Organ 75(4):361–366
Eur J Clin Microbiol Infect Dis (2010) 29:1435–14411441