Journal of the Formosan Medical Association (2016) 115, 825e836

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.jfma-online.com

REVIEW ARTICLE

Early diagnosis of spinal tuberculosis

Chang-Hua Chen a,b,c,*, Yu-Min Chen d, Chih-Wei Lee e,
Yu-Jun Chang f, Chun-Yuan Cheng g, Jui-Kuo Hung h

a
Division of Infectious Diseases, Department of Internal Medicine, Changhua Christian Hospital,
Changhua City, Taiwan, ROC
b
Center for Infectious Diseases Research, Changhua Christian Hospital, Changhua City, Taiwan, ROC
c
Department of Nursing, College of Medicine and Nursing, Hung Kuang University, Taichung County,
Taiwan, ROC
d
Department of Pharmacy, Changhua Christian Hospital, Changhua City, Taiwan, ROC
e
Department of Medical Imaging, Changhua Christian Hospital, Changhua City, Taiwan, ROC
f
Epidemiology and Biostatics Center, Changhua Christian Hospital, Changhua City, Taiwan, ROC
g
Division of Neurosurgery, Department of Surgery, Changhua Christian Hospital, Changhua City,
Taiwan, ROC
h
Department of Orthopedic Surgery, Changhua Christian Hospital, Changhua City, Taiwan, ROC

Received 23 December 2015; received in revised form 28 June 2016; accepted 2 July 2016

KEYWORDS Spinal tuberculosis (STB) is a common manifestation of extrapulmonary tuberculosis (TB). STB
algorithm; accounts for around 2% of all cases of TB and around 15% of extrapulmonary TB cases. The
decision-making; World Health Organization has proposed a global strategy and targets for TB prevention, care,
early diagnosis; and control after 2015. Under this strategy, patients will receive standard care according to
review; the recommendations and guidelines after confirmation of STB diagnosis. However, current
risk assessment; recommendations and guidelines focus on disease and medication therapy management, and
spine; recommendations for early detection or decision-making algorithms regarding STB are lacking.
tuberculosis In this review, we identified five key components for early diagnosis: (1) risk factors for STB; (2)
common symptoms/signs of STB; (3) significant neuroradiological findings of STB; (4) significant
laboratory findings of STB, including positive interferon-g release assays and nonpyogenic evi-
dence in initial laboratory data; and (5) significant clinical findings of STB. Individualized
consideration for each patient with STB is essential, and we hope that the algorithm esta-
blished in this review will provide a valuable tool for physicians who encounter cases of STB.
Copyright ª 2016, Formosan Medical Association. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

Conflicts of interest: The authors have no conflicts of interest relevant to this article.
* Corresponding author. Division of Infectious Diseases, Department of Internal Medicine, Changhua Christian Hospital, 135 Nan Hsiao
Street, Changhua, Taiwan, ROC.
E-mail address: chenchanghuachad@gmail.com (C.-H. Chen).

http://dx.doi.org/10.1016/j.jfma.2016.07.001
0929-6646/Copyright ª 2016, Formosan Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under the
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
826
Introduction: tuberculosis, spinal
tuberculosis, and World Health Organization
strategies
Although tuberculosis (TB) is an ancient disease, it remains
a major global public health problem.1,2 After two decades
of progress toward TB elimination, annual decreases of 13
cases per 100,000 persons in Taiwan3 and at least 0.2 cases
per 100,000 persons in the USA4,5 have been reported.
Although pulmonary TB (PTB) is the most common form,
spinal TB (STB) is one of the oldest reported diseases in
humans and a common manifestation of extrapulmonary TB
(ETB).6 In 2011, there were an estimated 8.7 million cases
of TB globally, with 1.4 million TB-related deaths.7 Addi-
tionally, in Taiwan, there were 12,338 tuberculosis cases
(53 cases per 100,000 persons) and 626 TB-related deaths
(2.7 cases per 100,000 persons) in 2012.8 STB comprises
around 4% of all TB cases and around 15% of all ETB cases
within Taiwan.8
In response to global strategy and targets post 2015,
Taiwan Strategic and Technical Working Group for TB
performed costebenefit analyses and those analyses
were mainly divided into two parts. First, the preferred
utilization of a chest radiograph survey among high-risk
populations should be justified. Second, a model for the
endorsement of new diagnosis tools for TB patients and
latent TB infection patients should be established and
its practicability assessed.9 According to a World Health
Organization (WHO) report,8 the WHO post-2015
strategy proposal contains three components (Table
S1).10 Patients with a confirmed STB diagnosis receive
standard care according to those recommendations and
guidelines (Table S2). However, these recommendations
and guidelines focus on disease and medication therapy
management and lack significant evidence and
constructive recommendations for early detection as
well as decision-making algorithms regarding STB. Thus,
there is currently a lack of adequate evidence,
recommendations, and guidelines for the early diagnosis
of STB.
Effective vaccination is a mainstay of long-term policies
to combat and control the TB epidemic. The existing bacille
CalmetteeGuérin vaccine has been shown to protect
against disseminated TB in young children11,12; however, its
protective effects against PTB are variable.12e14 Modified
TB vaccines include the Mycobacterium W15 and
recombinant bacille CalmetteeGuérin vaccines16 and new
vaccines include the modified Vaccinia Ankara virus
expressing antigen 85A,17 adenovirus 35-vectored TB
vaccine candidate AERAS-402,18 H4:IC31,19 and M72/AS01
formulations.20,21 However, these new TB vaccines require
further evaluation and clinical trials to demonstrate their
efficiency and safety. Thus, early diagnosis and effective
treatment are the other essential long-term strategies for
controlling the TB epidemic. The first component of the
WHO post-2015 strategy is early diagnosis of TB,10 and a
rapid TB diagnosis is listed as the second main indication for
recommended diagnostic tests (Table S3).22 This review
focuses on early diagnosis of STB.
C.-H. Chen et al.
Clinical aspects of STB
Natural clinical course
Spinal involvement usually results from the hematogenous
spread of Mycobacterium tuberculosis into the cancellous
bone tissue of the vertebral bodies. The primary infection
site comes from either a pulmonary focus or other extrap-
ulmonary foci, such as the lymph nodes.23e26 Predisposing
factors for STB include many categories, such as previous
TB infection, malnutrition, alcoholism, diabetes mellitus,
and human immunodeficiency virus infection.6,25,27
Clinical presentation
Delays in diagnosis of STB are common. In a report by
Sternbach,28 the average time from presentation to diag-
nosis is 1 year and 7 months. Typically, the onset of
symptoms is insidious, and disease progression is slow. The
duration of symptoms prior to diagnosis may range from 2
weeks to several years.6,25 To date, 2040 patients with STB
have been reported worldwide (Table S4). Male patients
comprise 53% of the STB-positive population, with a mean
age of 43.4 years (median, 43 years; Table S4). The clinical
presentation and findings of physical examinations depend
on the site and stage of the disease, presence of compli-
cations, and constitutional symptoms.6,25,29 The most
commonly reported symptoms are back pain (1438/2040,
70.4%), fever (667/2040, 32.7%), body weight loss (620/
2040, 30.3%), neurological abnormalities (315/2040,
30.1%), and night sweats (390/2040, 19.1%; Table S4).
Additionally, the thoracic spine is the most frequently
involved segment (Table S4). The complications after STB
included syringomyelia, permanent neurological deficits,
and spinal osseous defects.1,25 Paraplegia is the most
devastating complication of STB.6,25 The incidence of
neurological deficit varies from 23% to 76%.30
Clinical laboratory diagnosis
STB diagnosis is confirmed by typical clinical presentation
along with systemic constitutional manifestation, evidence
of past exposure to TB or concomitant visceral TB, and
neuroimaging modalities.6,25,31e33 Montoux skin tests and
hematological investigations, such as complete blood
count, erythrocyte sedimentation rate, enzyme-linked
immunosorbent assay, and polymerase chain reaction
(PCR), are also helpful in diagnosing STB.6,25,32,34 Bone
tissue or abscess samples stained for acid-fast bacilli,
mycobacterial organisms isolated from culture, and
computed tomography (CT)-guided or ultrasonography-
guided needle biopsy or surgical biopsy are also widely
used.6,25,35,36
Imaging diagnosis
Plain radiography is usually performed initially in patients
suspected to have STB, and plain radiograph images show
Early diagnosis of spinal tuberculosis
bird-nest appearance characteristics of the aneurysmal
phenomenon.37 Magnetic resonance imaging (MRI) is the
modality of choice for imaging of STB. STB can be suggested
over pyogenic spondylodiscitis on the basis of several
characteristic MRI features. Typical MRI abnormalities
include spinal lesions that originate from the vertebral
endplate, involve the anterior vertebral body corner, show
evidence of subligamentous spread, exhibit multiple
vertebral bodies but preserved discs, and show extensive
paraspinal abscess formation, abscess calcification, and
vertebral destruction or vertebral body collapse.37 The
mean rates of abnormal findings in different imaging mo-
dalities are 15% in plain spinal radiography, 100% in MRI, and
100% in CT (Table S4). For radiolucent lesions to be visible
on plain radiographic images, there must be 30% bone
mineral loss.25,38,39 CT is more effective for defining the
shape and calcification of soft tissue abscesses than plain
radiography because CT provides much better visualization
of the bony details of irregular lytic lesions, sclerosis, disc
collapse, and disruption of bone circumference.6 However,
CT is less accurate for defining the epidural extension of the
disease and its effect on neural structures compared with
MRI. Hypointense T1-weighted and hyperintense T2-
weighted signal intensities, short tau inversion recovery
sequences, and administration of gadolinium with dieth-
ylene tri-amine penta-acetic acid are useful techniques for
differentiation of TB from pyogenic spondylitis .6,25,40,41
Treatment
Early diagnosis and appropriate empirical anti-TB agents
combined with surgery are associated with an excellent
prognosis.25,42e44 Surgical treatment is performed for
decompression, debridement, and fusion 25,43,45 and has
been used in w75.7% of diagnosed patients (Table S4), at
least 80% of whom have shown improvement (Table S4).
Physicians are not only familiar with the common clinical
features of STB but also have an in-depth understanding of
the disease; early diagnosis and effective treatment can
minimize sequelae and improve clinical outcomes. The
cornerstone of successful management of STB is early
detection and timely judicious medical interven-
tion.28,41,46,47 Since accurate and effective treatment and
early diagnosis are essential, but clinical evidence required
for early diagnosis of STB is still lacking. Hence, in this
review, we aimed to provide a summary of our under-
standing of STB, emphasizing early diagnostic evidence
from clinical clues, various modalities of medical imaging,
and recent advances in laboratory examinations, as well as
our clinical experiences with STB.
Summary of information regarding early
diagnosis of STB
Most patients with STB are not accurately diagnosed during
the initial stages of the disease due to delays from both
the patient and doctor48; such delays may result in sig-
nificant spinal damage.37 The golden standard method for
diagnosis of STB is positive M. tuberculosis culture from
tissue. Clinical judgment for STB has shown both poor
sensitivity and specificity. Moreover, the time to detection
827
of mycobacterial growth may be 4e6 weeks using tradi-
tional culture methods, including methods that utilize
LowensteineJensen medium; however, faster culture of
mycobacterial isolates has been reported, including
quantitative tissue culture on solid media, an automated
broth culture system from tissue specimens, quantitative
nucleic acid amplification of M. tuberculosis from tissue
samples, and measurement of TB antigen concentrations
in tissue samples. Nucleic acid amplification tests (NAATs)
provide a reliable method for increasing the specificity of
diagnosis (ruling in disease), but exhibit poor sensitivity
and cannot be used to rule out disease.49 Because we have
focused on early diagnosis, we have not provided a liter-
ature search of current culture methods in this review.
However, advanced clinical findings, laboratory methods,
and medical imaging modalities could be helpful for early
detection of STB. Notably, diagnosis of STB is a major
challenge for physicians because of the nonspecific, wide
spectrum of clinical presentations; this can result in
delayed diagnosis and risk of significant potential
morbidity and mortality due to the progression of the
disease and subsequent complications. Early diagnosis and
treatment is the key to avoiding this long-term disability.
Therefore, we searched the literature to summarize the
tools available for advanced detection and early diagnosis
(Table 1).
A rapid biomarker-based nontissue-based test
In an effort to develop more accurate tools for immunological
diagnosis of tuberculosis, two mycobacterial proteins, cul-
ture filtrate protein-10 and early secretory antigenic target-
6, have been evaluated.54,61,62 These immunodiagnostic
tests, the whole-blood interferon-g (IFN-g) enzyme-linked
immunosorbent assay (QuantiFERON-TB Gold; Cellestis Ltd,
Chadstone, Victoria, Australia) and the enzyme-linked
immunospot assay (T-SPOT.TB; Oxford Immunotec, Oxford,
UK), can quantitatively measure IFN-g production by lym-
phocytes specific to M. tuberculosis-specific immunodo-
minant antigens, which are encoded by the RD1 region of the
pathogen. Another commercially available IFN-g release
assay (IGRA), the QuantiFERON-TB Gold in-tube assay
(QFTGIT; Cellestis Ltd), is able to measure IFN-g production
specific to the immunodominant antigens TB7.7, early
secretory antigenic target-6, and culture filtrate protein-10.
A commercially available T-SPOT.TB assay, the ELISPOT assay
(i.e., T-SPOT. TB; Oxford Immunotec),63 has been devel-
oped. In addition, detection of lipoarabinomannan and A60
mycobacterial antigen have been reported.50,51 These
immunological diagnostic tools have been evaluated for
diagnosis of ETB, including STB, infections.
Based on a report by Kumar et al,46 in combination with
radiological criteria, bone scan, and enzyme-linked
immunosorbent assay, the QuantiFERON assay was pre-
dictive of tuberculosis in 90% of cases. A report by Yuan
et al53 showed that the ELISpot assay is a useful adjunct to
current tests for diagnosis of STB. Feng et al52 reported
that IGRA could function as a powerful immunodiagnostic
test to explore PTB and ETB, and T-SPOT.TB was shown to
be equally sensitive for determining PTB and ETB. Based
on the data of Cho et al,64 the T-SPOT.TB assay is more
828 C.-H. Chen et al.

Table 1 Summary of advanced information for early diagnosis of spinal tuberculosis.

Items Description Sensitivity/ PPV/NPV Comment and references
specificity (%) (%)
Gold standard
Mycobacterial Direct test of tissue for NM NM For confirmation of STB
culture result M. tuberculosis complex diagnosis24,25,32
/TB culture of tissue
Reference standard
Histopathological Granulation NM NM For suspected STB
result inflammation, caseous diagnosis24,25,32
necrosis
TB smear result TB smear of tissue NM NM For suspected STB diagnosis
24,25,32

Index tests
A rapid biomarker-based nontissue-based test
Enzyme-linked Detection of A60 74/NM NM Although sensitivity is 63%
immunosorbent mycobacterial antigen and specificity is 96%, it is
assay not a useful laboratory test
for diagnosis of active TB50
Selected panel of TB Detection of 15.6/99.3 96.1/NM Could serve as an important
antibodies lipoarabinomannan, tool for diagnosis of TB,
38-kDa antigen, and especially ETB51
16-kDa antigen
Immunodiagnostic Uses enzyme-linked 88.9/50.0 90.9 /NM The T-SPOT.TB test had a
tests, the whole- immunospot assays (T- higher sensitivity than the
blood interferon-g SPOT.TB; Oxford QFT-GIT assay for diagnosing
enzyme-linked Immunotec, Oxfordshire, ETB57
immunosorbent UK) to detect ETB
assay Uses blood T-SPOT.TB to 100/55 62/NM The T-SPOT.TB was more
(QuantiFERON-TB detect osteoarticular TB sensitive in patients with
Gold; Cellestis Ltd, chronic forms of ETB, such
Chadstone, Victoria, as lymph node or
Australia) and the osteoarticular TB (93%, 95%
enzyme-linked CI: 83e97%), than in
immunospot assay patients with acute forms of
(T-SPOT.TB; Oxford ETB, such as TB meningitis
Immunotec, Oxford, or miliary TB (79%, 95% CI:
UK) 66e87%, p Z 0.03)56
Uses QuantiFERON assay 84/95 90/NM ELISA, MRI of the spine, and
to detect STB bone scanning findings
provide a reasonably certain
diagnosis in 90% of cases,
providing the grounds for
initiation of a trial to
determine the safety of
antituberculous
chemotherapy46
Enzyme-Linked Uses CFP10/ESAT6 fusion 82.7/87.2 89.6/79.1 Yuan et al showed that this
ImmunoSpot protein as antigen to method has high sensitivity
(ELISpot) assay detect spinal TB and specificity and is an
specimens effective technique for
auxiliary diagnosis of STB53
Measures the IFN-g 82.8/81.3 80.0/83.9 The ELISpot assay is a useful
response to ESAT-6 and adjunct to current tests for
CFP-10 in T cells in diagnosis of atypical STB54
samples of peripheral
blood mononuclear cells
of STB specimens
Early diagnosis of spinal tuberculosis 829

Table 1 (continued )
Items Description Sensitivity/ PPV/NPV Comment and references
specificity (%) (%)
Uses IFN-g to detect STB 86.7/61.9 61.9/86.7 The ELISpot assay could
specimens provide useful support in
diagnosing skeletal TB, and
STB could be excluded
based on a negative ELISpot
assays67
Rapid tissue-based tests
Polymerase chain Uses PCR to detect STB 100/71.4 NM Although numbers in the
reaction specimens study by Van der Spoel van
Dijk et al. were too low to
effectively draw
statistically valid
conclusions, the importance
of the relevance of PCR for
rapid detection of low
numbers of acid-fast bacilli
and confirmation of
mycobacterial infection in
spinal biopsies has been
established.56
Analysis of 94.7/83.3 94.7/83.3 Twenty-five formaldehyde-
formaldehyde-fixed, fixed, paraffin-embedded
paraffin-embedded tissue blocks from vertebral
tissue by PCR biopsy specimen materials
with presumptive diagnosis
of tuberculous spondylitis
were analyzed, with an
accuracy of 92%.57
Xpert MTB/RIF Uses Xpert MTB/RIF to 81.3/99.8 NM Positive likelihood ratios:
assay, an automated detect ETB 490.5; negative: 0.2.
amplification Although the role of culture
system remains central in the
microbiological diagnosis of
ETB, the sensitivity of Xpert
in rapidly diagnosing the
disease makes it a much
better choice compared
with smear microscopy72
Uses GeneXpert MTB/RIF 77.3/98.2 NM Overall, the sensitivity and
(Xpert) to detect ETB specificity of the Xpert
assay are very high in
culture-positive specimens.
The high sensitivity is not
self-evident because only
small amounts of DNA are
expected in any
extrapulmonary clinical
sample71
Xpert MTB/RIF 81/99.6 NM The results of this study
suggest that the Xpert test
also shows good potential
for the diagnosis of ETB and
that its ease of use makes it
applicable for countries
where TB is endemic70
(continued on next page)
830
Table 1 (continued )
Items Description
Advanced medical imaging
Fluorine-18- Fluorine-18-
fluorodeoxyglucose fluorodeoxyglucose
positron emission positron emission
tomography/ tomography/computed
computed tomography imaging
tomography imaging
MRI Diagnosis based on
positive histopathology
results along with
supportive MRI
Note: NM: not mentioned.
CFP10 Z culture filtrate protein 10; CT Z computed tomography; DNA Z deoxyribonucleic acid; ELISA Z enzyme-linked immunospot
assay; ESAT6 Z early secretory antigen target 6; ETB Z extrapulmonary tuberculosis; IFN-g Z Interferon-g; IGRA Z Interferon-g
release assays; NPV Z negative predictive value; MRI Z magnetic resonance imaging; PCR Z polymerase chain reaction; PET/CT Z
positron emission tomography/computed tomography; PPV Z positive predictive value; PTB Z pulmonary tuberculosis; QFT-IT Z
QuantiFERON-TB Gold In-Tube; TB Z tuberculosis; STB Z spinal tuberculosis; SUV Z standardized uptake value; TST Z tuberculin skin
test.
sensitive in patients with chronic forms of ETB, such as
lymph node or osteoarticular TB (93%; 95% confidence
interval, 83e97%) than in patients with acute forms of
ETB, such as TB meningitis or miliary TB (79%; 95% confi-
dence interval, 66e87%; p Z 0.03). Additionally, studies
by Lai et al55 and Lai and Wang65 found that the T-SPOT.TB
assay was more sensitive than the QFTGIT for the diag-
nosis of ETB, particularly in culture-confirmed TB cases.
This suggests that the T-SPOT.TB assay may be a helpful
adjunct diagnostic tool in patients with suspected ETB.
However, Fan et al66 showed that IGRAs have limited value
as diagnostic tools to screen and rule out ETB, particularly
in low/middle-income countries. The immune status of
patients does not affect the diagnostic accuracy of IGRAs
for ETB. Rangaka et al67 showed that IGRAs do not have
high accuracy for the prediction of active TB, although use
of IGRAs in some populations may reduce the number of
patients considered for preventive treatment. Although
IGRA could be a powerful immunodiagnostic test for
evaluation of ETB, these new methods require further
investigations to determine their accuracy and safety in
different populations, particularly patients with STB.
Future studies are needed to address different epidemi-
ological and clinical settings, evaluate the performance of
common commercial assays in head-to-head comparisons,
and assess the role of adding more TB-specific antigens on
improving diagnostic sensitivity.
Rapid tissue-based tests
The promise of rapid diagnosis exists with the wide imple-
mentation of molecular platforms, such as PCR.56,57
C.-H. Chen et al.
Sensitivity/ PPV/NPV Comment and references
specificity (%) (%)
90/NM NM The maximum SUVs of early-
phase PET/CT may be
complementary to MRI for
differentiating pyogenic and
tuberculous spondylitis and
reflecting the activity of
infectious spondylitis60
NM NM MRI findings showed
primarily shadowed areas
with features such as disc
destruction and thecal or
cord compression. MRI
scanning could be used for
early detection of spinal TB,
which can reduce disability
and deaths in patients37
Turnaround times have been reported to be as short as 24
hours.68 Based on a report by Monni et al,69 the sensitivities
of culture and PCR are comparable (77% and 72%, respec-
tively). Some newer and more sensitive NAATs, including
the amplified M. tuberculosis direct test and Xpert MTB/RIF
(Xpert) assays, may enhance M. tuberculosis detection in
specimens. Xpert (Cepheid, Sunnyvale, CA, USA) is a fully
automated real-time heminested PCR system implementing
molecular beacon technology for the diagnosis of PTB
infection.
According to reports by Pai et al,70,71 commercial
NAATs may have a potential role in confirming the diag-
nosis of tuberculous meningitis and tuberculous pleuritis,
although their overall low sensitivity precludes their use
to rule out those types of ETB with certainty. Tortoli
et al72 evaluated the performance of the Xpert system in
many different extrapulmonary specimens in a country
with a low incidence of TB and found that, in comparison
with a reference standard consisting of combination of
culture and clinical diagnosis of TB, the Xpert system
exhibited an overall sensitivity and specificity of 81.3%
and 99.8%, respectively. The results of Hillemann et al’s59
study suggest that the Xpert test also shows good poten-
tial for the diagnosis of ETB and that its ease of use makes
it applicable for countries where TB is endemic. In addi-
tion, the sensitivity and specificity of the Xpert assay are
very high in culture-positive specimens. The high sensi-
tivity is not self-evident because only small amounts of
DNA are expected in any extrapulmonary clinical sam-
ple.58 Currently, there are also major clinical problems
with obtaining tissue specimens for diagnosis in most pa-
tients suspected of having STB.
Early diagnosis of spinal tuberculosis 831

Table 2 Early clinical presentation, laboratory results, and neuroradiological findings of patients with confirmed spinal
tuberculosis
Patient 1 2 3 4
Age (y)/sex 78/F 66/M 82/F 57/M
History of TB exposure þ þ þ þ
Initial laboratory data
White cell count (cells/mL) 4800 6200 5800 6800
ESR (mm/h) 45 45 38 68
CRP (mg/dL) 24.9 7.9 4.5 5.5
Albumin (g/dL) 4.2 3.2 3.5 3.6
Procalcitonin (ng/mL) 0.04 0.03 0.04 0.02
Positive IGRA þ þ þ þ
Initial neuroradiological findings
Originates from vertebral endplate þ þ þ þ
Involves the anterior vertebral body corner þ þ þ þ
Subligamentous spread þ þ þ þ
Multiple vertebral bodies involved & preserved disc þ þ þ þ
Extensive paraspinal abscess formation þ þ þ þ
Calcification of abscess þ þ þ þ
Vertebral destruction, vertebral body collapse (Gibbus deformity) þ þ þ þ
Definitive diagnosis
TB smear of tissue/MtbcDT of tissue/TB culture of tissue þ/þ/MTBC þ/þ/MTBC þ/þ/MTBC þ/þ/MTBC
Histopathological results GI GI, CN GI, CN GI, CN
No good response to traditional anti-bacterial regimens a þ þ þ þ
Patient delay (d)/doctor delay of ATA (d)/total delay (d) 60/5/65 30/5/35 32/5/37 60/3/63
ATA Z anti-TB agent; CN Z caseous necrosis; CRP Z C-reactive protein; ESR Z erythrocyte sedimentation rate; F Z female; GI Z
granulation inflammation; IGRA Z interferon- g release assays; M Z male; Mo Z month; MtbcDT Z Mycobacterium tuberculosis
complex direct test; TB Z tuberculosis .
a
No good response to traditional antibacterial regimens meant that clinical signs and symptoms and the results of laboratory ex-
aminations did not improve after 3 days of administration of combination antibacterial regimens (anti-Staphylococcal agent with an anti-
Pseudomonal agent, such as oxacillin plus aminoglycosides).

Advanced medical imaging mycobacterial infections with histopathological findings

Lee et al60 reported MRI to be superior to positron emission
tomography and CT in differentiating between STB and
pyogenic spondylitis. MRI could be used for early detection
of STB, which could reduce the occurrence of disability and
death in these patients.37 MRI is the earliest diagnostic
modality for STB; however, confirmatory diagnosis can only
be made on the basis of biopsy or culture results.37,73
Our experiences
A literature review revealed sparse research concerning
early diagnosis of STB; therefore, we summarized our
experience to analyze factors associated with early diag-
nosis of STB. Our experience included patients hospitalized
at Changhua Christian Hospital System, Changhua City,
Taiwan who were diagnosed with STB from January 1, 2010
to December 31, 2013. All adult patients diagnosed with
STB by all of the following methods were included: positive
vertebral osseous acid-fast bacilli; positive TB culture from
fine needle aspiration cytology specimens; and character-
istic pathologic findings. All patients diagnosed with STB
received standard anti-TB agents, according to the Taiwa-
nese guidelines for TB diagnosis and treatment.8
Five cases were diagnosed as having STB during the study
period. Four of these cases were confirmed to have
confirming TB, while the other case was a non-TB myco-
bacterial infection. Thus, only four patients with confirmed
STB infections were enrolled in this study. The early clinical
presentations, laboratory results, and neuroradiological
findings of these patients are listed in Table 2. The neuro-
radiological findings of a patient with STB are shown in
Figure 1. All patients received standard anti-TB agents. Two
underwent surgical interventions for neurological deficits,
and all survived.
We noted five key components for early diagnosis: (1)
risk factors for STB; (2) common symptoms/signs of STB; (3)
significant neuroradiological findings of STB; (4) significant
laboratory findings of STB, including positive IGRA and
nonpyogenic evidence in initial laboratory data; and (5)
significant clinical findings of STB. STB has a wide variety of
manifestations. However, we identified five key indicators
for STB (Table 2), which were consistent with the obser-
vations in previous reports.74e78 According to a study in the
Netherlands by Jutte et al,48 there was a mean delay in the
diagnosis of nonspinal bone and joint TB of 35 weeks. In the
current study, the total delay duration ranged from 25 days
to 65 days. Early diagnosis of STB is important to prevent
negative outcomes.77 Wares et al77 described the nonspe-
cific radiological features of STB. However, we identified
seven key radiological findings observed in the lesions of
patients with STB (Table 2). Imaging is important for diag-
nosis and treatment of STB, and the diagnostic roles of
832 C.-H. Chen et al.

Figure 1 Neuroradiological findings of a patient with spinal tuberculosis. (A). Postgadolinium T1-weighted magnetic resonance
imaging with fat saturation, sagittal view, showing enhancement of the T9 and T10 vertebrae with apposing endplates erosion, and
enhanced soft tissue spreading along the anterior subligamentous space, from T7 to T12 level. Mild enhancement of the T8 inferior
and T11 superior vertebral bodies are also noted, but the T8/9 and T10/11 discs are preserved. (B). Postgadolinium T1-weighted
magnetic resonance imaging with fat saturation, axial (at T9e10 level) view, showing extensive paraspinal abscess (arrow). (C).
Computed tomography scan of the thoracic spine axial view showing calcification of the paraspinal abscess (arrow).

various imaging modalities have been discussed previ-
ously.79 MRI and spiral CT are currently the most commonly
used imaging methods.
An algorithm for decision-making
Advanced clinical findings, laboratory methods, and
medical imaging could be helpful for early detection of
STB (Table 1). However, the availability and utility of
these methods are variable worldwide, including in
Taiwan. A successful experience to use an algorithmic
approach for patient management, and their results
enable early identifies candidates at high risk of TB
reactivation for anti-TB chemoprophylaxis.80 We have
developed a decision-making algorithm for the diagnosis
and management of STB based on our literature review
and our experience; we expect that this algorithm may be
useful at most institutes and facilities (Figure 2).1,74e79
Risk assessment for patients with suspicion of STB should
include the five key components. The risk assessment
elements are modified based on our experiences and
previous reports.74e76,78,81 MRI is one of the most useful
diagnostic methods for STB, often allowing an earlier
diagnosis than conventional methods. Rational orders for
MRI could reduce medical resource waste, according to
Taiwan’s National Health Insurance regulations. Patients
undergoing conservative treatment without surgery should
be monitored closely with serial clinical examinations (at
least weekly), ESR assessment (every 2e4 weeks), and MRI
or CT (depending on the clinical situation and level of risk
of the patient in order to assess abscess size and extent,
until resolution). Patients take anti-TB agents regularly
prior to meals under the directly observed treatment,
short course program.82 Optimized care involves risk
assessment of concurrent conditions, re-assessment of
follow-up conditions, and administration of appropriate
anti-TB agents with or without surgery.
Accurate diagnosis and adequate treatment of STB can
improve patient outcomes. Hence, the algorithm esta-
blished in this report (Figure 2) also includes decision-
making for the diagnosis and management of STB.1,74e79
Emergent surgical decompression should be initiated for
Early diagnosis of spinal tuberculosis 833

Figure 2 Algorithm for diagnosis and treatment of spinal tuberculosis. a Immediate investigations included complete medical
history, physical examinations, neurological examinations, laboratory examinations, and imaging examinations. b The grading
system score for spondylodiscitis is shown in Table S5. c Poor response was defined as a lack of reduction in clinical symptoms
and signs following administration of anti-TB agents. d Scoring for describing disability status based on functional disability
classification is shown in Table S6. CBC Z complete blood cell count; CRP Z C-reactive protein; CT Z computed tomography;
CXR Z chest X-ray; ESR Z erythrocyte sedimentation rate; F/U Z follow-up; FDC Z functional disability classification;
HIV Z human immunodeficiency virus; IGRA Z interferon-gamma release assay; LFT Z liver function test; MRI: magnetic
resonance imaging; PTB: pulmonary tuberculosis; RFT: renal function test; SD: spondylodiscitis; STB: spinal tuberculosis; WBC:
white blood cell count.

patients with infection-induced moderate to severe
neurological compromise for the duration of anti-TB
agent treatment. The neurological status at the time of
presentation is a critical factor for treatment decision-
making and patient outcomes. If the imaging results
suggest STB, the decision for operative management
should be based on risk assessment for the failure of
medical treatment according to the grading system for
spondylodiscitis (Table S5). Patients receiving medical
treatment alone may receive CT-guided drainage of the
target lesions. They should also be monitored and re-
assessed for signs of medical treatment failure. If these
834
signs are detected, surgical decompression should be
promptly arranged.
Future studies
This review included a large number of studies; however,
continued research on this topic is necessary to define early
diagnosis of STB more accurately. Large-scale studies and
clinical trials are also required in order to develop and
validate more accurate modules, matrixes, or algorithms
for early diagnosis of STB based on advances in medical
technologies. In order to realize the incremental value of
combinations of tests, particularly for samples of tissue
specimens, it will be necessary to carry out large, pro-
spective, well-designed studies recruiting representative
samples of patients with STB. Longitudinal cohort studies to
confirm the positive predictive value of IGRAs for subse-
quent development of active TB and ETB (including STB)
should also be performed. M. tuberculosis transmission is
limited because of genetic evolution and deviation from
different geographic regions and ethnic populations in
Taiwan, and it remains to be investigated.83
Conclusion
In conclusion, TB is preventable and curable, and its
elimination would have widespread health, economic, and
social benefits. STB is a great challenge to physicians
because of the nonspecific, wide spectrum of clinical
presentations, which result in delayed diagnosis and risk of
significant morbidity and mortality. In this review, we have
summarized knowledge on early diagnosis of STB. A high
degree of clinical acumen and evaluation based on the five
indicators outlined in this review are important for STB
diagnosis and for starting treatment with anti-TB agents,
which is the current mainstay of treatment. In addition, MRI
is one of the most useful diagnostic methods for STB.
Individualized consideration of each patient with STB is
essential, and we hope that the algorithm developed in this
report (Figure 2) will be a valuable tool for physicians.
Acknowledgments
We thank the TB case managers of Changhua Christian
Hospital for their assistance in data collection and the staff
of the Infection Control Committee, Clinical Microbiological
Laboratory, Department of Medical Records, and Depart-
ment of Computer of Changhua Christian Hospital for
providing information for statistical analysis. We also wish
to express their gratitude to Dr Yung-Jen Yang, who was
extremely helpful and provided invaluable assistance and
support. We thank Dr Fu-Chou Chen at Taichung Veterans
General Hospital for reviewing this article. Grant support
was received from Changhua Christian Hospital (104-CCH-
IRP-001, 105-CCH-IRP-001).
Appendix A. Supplementary data
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.jfma.2016.07.001.
C.-H. Chen et al.
References
1. Chen YH, Lin CB, Harnod T, Wu WT, Yu JC, Chen IH, et al.
Treatment modalities for tuberculosis of the spine: 22 years’
experience in east Taiwan. Formos J Surg 2013;46:189e94.
2. Mousa HAL. Bones and joints tuberculosis. Bahrain Med Bull
2007;29:1e9.
3. Lo HY, Chou P, Yang SL, Lee CY, Kuo HS. Trends in tuberculosis
in Taiwan, 2002e2008. J Formos Med Assoc 2011;110:501e10.
4. Centers for Disease Control. Reported tuberculosis in the United
States. Atlanta, GA: US Department of Health and Human Ser-
vices, CDC. Available from: http://www.cdc.gov/tb/statistics/
reports/2014/default.htm [Accessed October 6, 2014].
5. Salinas JL, Mindra G, Haddad MB, Pratt R, Price SF, Langer AJ.
Leveling of tuberculosis incidencedUnited States, 2013-2015.
MMWR Morb Mortal Wkly Rep 2016;65:273e8.
6. Ansari S, Amanullah MF, Ahmad K, Rauniyar RK. Pott’s spine:
diagnostic imaging modalities and technology advancements. N
Am J Med Sci 2013;5:404e11.
7. World Health Organization. Stop TB Partnership. The global
plan to stop TB 2011e2015. Available from: http://www.
stoptb.org/global/plan [Accessed June 15, 2016].
8. Centers for Disease Control DOH, R.O.C. (Taiwan). Taiwan
guidelines for TB diagnosis and treatment. 5th ed. Taipei:
Centers for Disease Control, Department of Health, R.O.C.
(Taiwan); 2011.
9. Chan PC, Chen CH, Chang FY. External review of the National
Tuberculosis Program and the development of strategy and
targets post 2015 in Taiwan. J Formos Med Assoc 2014;113:
775e7.
10. World Health Organization. WHO report on the proposed post-
2015 TB Strategy and targets. Available from: http://apps.
who.int/gb/ebwha/pdf_files/EB134/B134_12-en.pdf?uaZ1.
[Accessed November 29, 2013].
11. Trunz BB, Fine PEM, Dye C. Effect of BCG vaccination on
childhood tuberculous meningitis and miliary tuberculosis
worldwide: a meta-analysis and assessment of cost-effective-
ness. Lancet 2006;367:1173e80.
12. Rodrigues LC, Diwan VK, Wheeler JG. Protective effect of BCG
against tuberculous meningitis and miliary tuberculosis: a
meta-analysis. Int J Epidemiol 1993;22:1154e8.
13. Fine PE. Variation in protection by BCG: implications of and for
heterologous immunity. Lancet 1995;346:1339e45.
14. Colditz GA, Berkey CS, Mosteller F, Brewer TF, Wilson ME,
Burdick E, et al. The efficacy of bacillus CalmetteeGuerin
vaccination of newborns and infants in the prevention of
tuberculosis: meta-analyses of the published literature. Pedi-
atrics 1995;96:29e35.
15. Katoch K, Singh P, Adhikari T, Benara SK, Singh HB,
Chauhan DS, et al. Potential of Mw as a prophylactic vaccine
against pulmonary tuberculosis. Vaccine 2008;26:1228e34.
16. Hoft DF, Blazevic A, Abate G, Hanekom WA, Kaplan G, Soler JH,
et al. A new recombinant bacille CalmetteeGuérin vaccine
safely induces significantly enhanced tuberculosis-specific im-
munity in human volunteers. J Infect Dis 2008;198:1491e501.
17. Tameris MD, Hatherill M, Landry BS, Scriba TJ, Snowden MA,
Lockhart S, et al. Safety and efficacy of MVA85A, a new
tuberculosis vaccine, in infants previously vaccinated with
BCG: a randomised, placebo-controlled phase 2b trial. Lancet
2013;381:1021e8.
18. Kagina BM, Tameris MD, Geldenhuys H, Hatherill M, Abel B,
Hussey GD, et al. The novel tuberculosis vaccine, AERAS-402, is
safe in healthy infants previously vaccinated with BCG, and
induces dose-dependent CD4 and CD8T cell responses. Vaccine
2014;32:5908e17.
19. Geldenhuys H, Mearns H, Miles DJ, Tameris M, Hokey D, Shi Z,
et al. The tuberculosis vaccine H4:IC31 is safe and induces a
Early diagnosis of spinal tuberculosis
persistent polyfunctional CD4 T cell response in South African
adults: A randomized controlled trial. Vaccine 2015;33:3592e9.
20. Montoya J, Solon JA, Cunanan SRC, Acosta L, Bollaerts A,
Moris P, et al. A randomized, controlled dose-finding Phase II
study of the M72/AS01 candidate tuberculosis vaccine in
healthy PPD-positive adults. J Clin Immunol 2013;33:
1360e75.
21. Leroux-Roels I, Leroux-Roels G, Ofori-Anyinam O, Moris P, De
Kock E, Clement F, et al. Evaluation of the safety and immu-
nogenicity of two antigen concentrations of the
Mtb72F/AS02(A) candidate tuberculosis vaccine in purified
protein derivative-negative adults. Clin Vaccine Immunol 2010;
17:1763e71.
22. World Health Organization. High-priority target product pro-
files for new tuberculosis diagnostic. Report of a consensus
meeting. Available from: http://apps.who.int/iris/bitstream/
10665/135617/1/WHO_HTM_TB_2014.18_eng.pdf?
uaZ1&uaZ1. [Accessed April 28, 2014].
23. Boachie-Adjei O, Squillante RG. Tuberculosis of the spine.
Orthop Clin North Am 1996;27:95e103.
24. Schirmer P, Renault CA, Holodniy M. Is spinal tuberculosis
contagious? Int J Infect Dis 2010;14:e659e66.
25. Raviglione MC. Tuberculosis. In: Kasper D, Fauci A, Hauser S,
Longo D, Jameson J, Loscalzo J, editors. Harrison’s principles
of internal medicine. 19th ed. New York, NY: McGraw-Hill;
2015. http://accessmedicine.mhmedical.com/content.aspx?
bookidZ1130&SectionidZ79737003 [Accessed March 4, 2016].
26. Mok JH, Kim KU, Park HK, Lee MK. Extensively drug-resistant
tuberculosis presenting as primary lymphadenitis eroding into
the trachea in an immunocompetent patient. J Formos Med
Assoc 2014;113:764e5.
27. McLain RF, Isada C. Spinal tuberculosis deserves a place on the
radar screen. Cleve Clin J Med 2004;71:543e9.
28. Sternbach G. Percivall Pott: tuberculous spondylitis. J Emerg
Med 1996;14:79e83.
29. Nussbaum ES, Rockswold GL, Bergman TA, Erickson DL,
Seljeskog EL. Spinal tuberculosis: a diagnostic and manage-
ment challenge. J Neurosurg 1995;83:243e7.
30. Kotil K, Alan MS, Bilge T. Medical management of Pott disease
in the thoracic and lumbar spine: a prospective clinical study. J
Neurosurg Spine 2007;6:222e8.
31. Garg RK, Somvanshi DS. Spinal tuberculosis: a review. J Spinal
Cord Med 2011;34:440e54.
32. McAdam AJ, Onderdonk AB. Laboratory diagnosis of infectious
diseases. In: Kasper D, Fauci A, Hauser S, Longo D, Jameson J,
Loscalzo J, editors. Harrison’s principles of internal medicine.
19th ed. New York, NY: McGraw-Hill; 2015. http://
accessmedicine.mhmedical.com/content.aspx?bookidZ1130&
SectionidZ63652694 [Accessed March 4, 2016].
33. Gautam MP, Karki P, Rijal S, Singh R. Pott’s spine and para-
plegia. J Nepal Med Assoc 2005;44:106e15.
34. Alli OA, Ogbolu OD, Alaka OO. Direct molecular detection of
Mycobacterium tuberculosis complex from clinical sam-
plesdan adjunct to cultural method of laboratory diagnosis of
tuberculosis. N Am J Med Sci 2011;3:281e8.
35. Watts HG, Lifeso RM. Current concepts review. Tuberculosis of
bones and joints. J Bone Joint Surg Am 1996;78:288e99.
36. Dass B, Puet TA, Watanakunakorn C. Tuberculosis of the spine
(Pott’s disease) presenting as ‘compression fractures’. Spinal
Cord 2002;40:604e8.
37. Alvi AA, Raees A, khan Rehmani MA, Aslam HM, Saleem S,
Ashraf J. Magnetic resonance image findings of spinal tuber-
closis at first presentation. Int Arch Med 2014;7:12.
38. Chauhan A, Gupta BB. Spinal tuberculosis. Indian Acad Clin
Med 2007;8:110e3.
39. World Health Organization. The global tuberculosis control.
http://www.who.int/tb/publications/global_report/en/
[Accessed October 27, 2015].
835
40. Jain AK. Tuberculosis of the spine: a fresh look at an old dis-
ease. J Bone Joint Surg Br 2010;92:905e13.
41. Jung NY, Jee WH, Ha KY, Park CK, Byun JY. Discrimination of
tuberculous spondylitis from pyogenic spondylitis on MRI. AJR
Am J Roentgenol 2004;182:1405e10.
42. Harada Y, Tokuda O, Matsunaga N. Magnetic resonance imaging
characteristics of tuberculous spondylitis vs. pyogenic spon-
dylitis. Clin Imaging 2008;32:303e9.
43. Mohammadreza E, Fariborz S, Gholamreza B. Pott’s disease: a
review of 58 cases. Med J Islamic Republic Iran 2010;23:
200e6.
44. Pertuiset E, Beaudreuil J, Lioté F, Horusitzky A, Kemiche F,
Richette P, et al. Spinal tuberculosis in adults. A study of 103
cases in a developed country, 1980-1994. Medicine 1999;78:
309e20.
45. Pappou IP, Papadopoulos EC, Swanson AN, Mermer MJ,
Fantini GA, Urban MK, et al. Pott’s disease in the thor-
acolumbar spine with marked kyphosis and progressive
paraplegia necessitating posterior vertebral column resection
and anterior reconstruction with a cage. Spine 2006;31:
E123e7.
46. Kumar R, Srivastava AK, Tiwari RK. Surgical management of
Pott’s disease of the spine in pediatric patients: a single sur-
geon’s experience of 8 years in a tertiary care center. J Pediatr
Neurosci 2011;6(Suppl. 1):S101e8.
47. Jain AK, Srivastava A, Saini NS, Dhammi IK, Sreenivasan R,
Kumar S. Efficacy of extended DOTS category I chemotherapy
in spinal tuberculosis based on MRI-based healed status. Indian
J Orthop 2012;46:633e9.
48. Jutte PC, van Loenhout-Rooyackers JH, Borgdorff MW, Van
Horn JR. Increase of bone and joint tuberculosis in The
Netherlands. J Bone Joint Surg Br 2004;86:901e4.
49. Dinnes J, Deeks J, Kunst H, Gibson A, Cummins E, Waugh N,
et al. A systematic review of rapid diagnostic tests for the
detection of tuberculosis infection. Health Technol Assess
2007;11:1e196.
50. Demkow U, Zielonka TM, Filewska M, Białas-Chromiec B,
Michałowska-Mitczuk D, Załeska J, et al. Value of the immu-
nochromatographic assay for detecting IgG antibodies against
38 kDa mycobacterial antigen in diagnosis of tuberculosis.
Pneumonol Alergol Pol 2000;68:355e62.
51. Deng S, Yuan T, Xia J, Huang H, Cheng X, Chen M. Clinical
utility of a combination of lipoarabinomannan, 38-kDa, and 16-
kDa antigens as a diagnosis tool for tuberculosis. Diagn
Microbiol Infect Dis 2011;71:46e50.
52. Feng Y, Diao N, Shao L, Wu J, Zhang S, Jin J, et al. Interferon-
gamma release assay performance in pulmonary and extrap-
ulmonary tuberculosis. PLoS One 2012;7:e32652.
53. Yuan K, Zhong ZM, Zhang Q, Xu SC, Chen JT. Evaluation of an
enzyme-linked immunospot assay for the immunodiagnosis of
atypical spinal tuberculosis (atypical clinical pre-
sentation/atypical radiographic presentation) in China. Braz J
Infect Dis 2013;17:529e37.
54. Yuan K, Liang D, Wu XQ, Yao ZS, Jin DX, Yang ZD, et al. Diag-
nostic value of enzyme-linked immunospot assay using
CFP10/ESAT6 fusion protein as antigen in spinal tuberculosis.
Zhongguo Yi Xue Ke Xue Yuan Xue Bao 2015;37:44e9.
55. Lai CC, Tan CK, Lin SH, Liao CH, Huang YT, Hsueh PR. Diag-
nostic performance of whole-blood interferon-g assay and
enzyme-linked immunospot assay for active tuberculosis.
Diagn Microbiol Infect Dis 2011;71:139e43.
56. Van der Spoel VDA, McLeod A, Botha PL, Shipley JA,
Kapnoudhis MA, Beukes CA. The diagnosis of skeletal tuber-
culosis by polymerase chain reaction. Cent Afr J Med 2000;46:
144e9.
57. Berk RH, Yazici M, Atabey N, Ozdamar OS, Pabuccuoglu U,
Alici E. Detection of Mycobacterium tuberculosis in formalde-
hyde solution-fixed, paraffin-embedded tissue by polymerase
836
chain reaction in Pott’s disease. Spine (Phila Pa 1976) 1996;21:
1991e5.
58. Vadwai V, Boehme C, Nabeta P, Shetty A, Alland D, Rodrigues C.
Xpert MTB/RIF: a new pillar in diagnosis of extrapulmonary
tuberculosis? J Clin Microbiol 2011;49:2540e5.
59. Hillemann D, Rüsch-Gerdes S, Boehme C, Richter E. Rapid
molecular detection of extrapulmonary tuberculosis by the
automated GeneXpert MTB/RIF system. J Clin Microbiol 2011;
49:1202e5.
60. Lee IS, Lee JS, Kim SJ, Jun S, Suh KT. Fluorine-18-fluorodeox-
yglucose positron emission tomography/computed tomography
imaging in pyogenic and tuberculous spondylitis: preliminary
study. J Comput Assist Tomogr 2009;33:587e92.
61. Sørensen AL, Nagai S, Houen G, Andersen P, Andersen AB.
Purification and characterization of a low-molecular-mass T-
cell antigen secreted by Mycobacterium tuberculosis. Infect
Immun 1995;63:1710e7.
62. Pai M, Riley LW, Colford Jr JM. Interferon-gamma assays in the
immunodiagnosis of tuberculosis: a systematic review. Lancet
Infect Dis 2004;4:761e76.
63. Porsa E, Cheng L, Graviss EA. Comparison of an ESAT-6/CFP-10
peptide-based enzyme-linked immunospot assay to a tuber-
culin skin test for screening of a population at moderate risk of
contracting tuberculosis. Clin Vaccine Immunol 2007;14:
714e9.
64. Cho OH, Park KH, Kim SM, Park SJ, Moon SM, Chong YP, et al.
Diagnostic performance of T-SPOT. TB for extrapulmonary
tuberculosis according to the site of infection. J Infect 2011;
63:362e9.
65. Lai CC, Wang HC. Diagnostic performance of enzyme-linked
immunospot assay and whole-blood interferon-g assay for the
diagnosis of extrapulmonary tuberculosis. J Microbiol Immunol
Infect 2011;44:406e7.
66. Fan L, Chen Z, Hao XH, Hu ZY, Xiao HP. Interferon-gamma
release assays for the diagnosis of extrapulmonary tubercu-
losis: a systematic review and meta-analysis. FEMS Immunol
Med Microbiol 2012;65:456e66.
67. Rangaka MX, Wilkinson KA, Glynn JR, Ling D, Menzies D, Mwansa-
Kambafwile J, et al. Predictive value of interferon-gamma
release assays for incident active tuberculosis: a systematic
review and meta-analysis. Lancet Infect Dis 2012;12:45e55.
68. Negi SS, Gupta S, Khare S, Lal S. Comparison of various
microbiological tests including polymerase chain reaction for
the diagnosis of osteoarticular tuberculosis. Indian J Med
Microbiol 2005;23:245e8.
69. Monni T, Visser A, Visser HF, Motsitsi SN. Clinical utility of tis-
sue polymerase chain reaction in the diagnosis of spinal
tuberculosis. SA Orthop J 2012;11:23e7.
70. Pai M, Flores LL, Pai N, Hubbard A, Riley LW, Colford Jr JM.
Diagnostic accuracy of nucleic acid amplification tests for
C.-H. Chen et al.
tuberculous meningitis: a systematic review and meta-anal-
ysis. Lancet Infect Dis 2003;3:633e43.
71. Pai M, Flores LL, Hubbard A, Riley LW, Colford Jr JM. Nucleic
acid amplifiaction tests in the diagnosis of tuberculous pleur-
itis: a systematic review and meta-analysis. BMC Infect Dis
2004;4:6.
72. Tortoli E, Russo C, Piersimoni C, Mazzola E, Dal Monte P,
Pascarella M, et al. Clinical validation of Xpert MTB/RIF for the
diagnosis of extrapulmonary tuberculosis. Eur Respir J 2012;
40:442e7.
73. Desai SS. Early diagnosis of spinal tuberculosis by MRI. J Bone
Joint Surg Br 1994;76:863e9.
74. Weng CY, Chi CY, Shih PJ, Ho CM, Lin PC, Chou CH, et al. Spinal
tuberculosis in non-HIV-infected patients: 10 year experience
of a medical center in central Taiwan. J Microbiol Immunol
Infect 2010;43:464e9.
75. Yoon YK, Jo YM, Kwon HH, Yoon HJ, Lee EJ, Park SY, et al.
Differential diagnosis between tuberculous spondylodiscitis
and pyogenic spontaneous spondylodiscitis: a multicenter
descriptive and comparative study. Spine J 2015;15:
1764e71.
76. Teegala R, Kumar P, Kale SS, Sharma BS. Craniovertebral
junction tuberculosis: a new comprehensive therapeutic
strategy. Neurosurgery 2008;63:946e55.
77. Wares F, Balasubramanian R, Mohan A, Sharma SK. Extrap-
ulmonary tuberculosis: management and control. In: Tuber-
culosis control in India. New Delhi: Ministry of Health and
Family Welfare; 2005. p. 95e114.
78. Jain AK, Sinha S. Evaluation of systems of grading of neuro-
logical deficit in tuberculosis of spine. Spinal Cord 2005;43:
375e80.
79. Shikhare SN, Singh DR, Shimpi TR, Peh WC. Tuberculous oste-
omyelitis and spondylodiscitis. Semin Musculoskelet Radiol
2011;15:446e58.
80. Kuo CH, Lin SM, Lee KY, Chung FT, Feng PH, Hsiung TC, et al.
Algorithmic approach by endobronchial ultrasound-guided
transbronchial needle aspiration for isolated intrathoracic
lymphadenopathy: a study in a tuberculosis-endemic country.
J Formos Med Assoc 2014;113:527e34.
81. Richter C, Ndosi B, Mwammy AS, Mbwambo RK. Extrapulmonary
tuberculosisea simple diagnosis? A retrospective study at Dar
es Salaam, Tanzania. Trop Geogr Med 1991;43:375e8.
82. Lin HC, Yu MC, Liu HJ, Bai KJ. Impact of food intake on the
pharmacokinetics of first-line antituberculosis drugs in Taiwa-
nese tuberculosis patients. J Formos Med Assoc 2014;113:
291e7.
83. Dou HY, Chen YY, Kou SC, Su IJ. Prevalence of Mycobacterium
tuberculosis strain genotypes in Taiwan reveals a close link to
ethnic and population migration. J Formos Med Assoc 2015;
114:484e8.