European Review for Medical and Pharmacological Sciences 2012; 16(Suppl 2): 58-72

Tuberculous spondylodiscitis: epidemiology,

clinical features, treatment, and outcome
E.M. TRECARICHI, E. DI MECO, V. MAZZOTTA, M. FANTONI
Institute of Infectious Diseases, School of Medicine, Catholic University of the Sacred Heart, Rome
(Italy)

Abstract. – Background: Tuberculous The spine is the most common site for osseous
spondylodiscitis (TS) is a rare but serious clini- involvement by tuberculosis (TB). TS has been
cal condition which may lead to severe deformi- reported to accounts for 1-5% of all TB cases
ty and early or late neurological complications.
Aim: To discuss certain aspects of the ap-
from many reports2-8, and for about 50% of the
proach to TSs, focusing upon epidemiology, di- cases of articulo-skeletal TB infections2,9-11.
agnosis, and treatment outcome. Comparative studies of spontaneous spinal in-
Materials and Methods: For the purpose of fections performed in developed countries show
this review, a literature search was performed that Mycobacterium (M.) tuberculosis is the
using the Pubmed database through to 19th Oc- causative agent of spinal infections with a fre-
tober 2011 to identify studies published in the quency ranging from 17% to 39%, thus repre-
last 20 years, concerned in epidemiological, clin-
ical, diagnostic, and therapeutical aspects of TS senting an important issue even in a contest of
in adults. Only studies drafted in English lan- low endemicity for TB infection2,12-15.
guage and reporting case series of more than 20 Many observational studies have been pub-
patients have been included. lished in the last 20 years in order to identify the
Results: TS has been reported to accounts clinical, microbiological, and radiological fea-
for 1-5% of all TB cases, and for about 50% of tures of patients with TS and to assess the cor-
the cases of articulo-skeletal TB infections. De-
spite the actual availability of more effective di-
rect management in terms of diagnosis and treat-
agnostic tools, early recognition of TS remains ment3,5,6,8,9,12-14,16-36. Most of these studies have
difficult and a high index of suspicion is needed been conducted in developing nations, where the
due to the chronic nature of the disease and its incidence of tuberculosis is higher and the aver-
insidious and variable clinical presentation. A age age of patients at presentation is low-
prompt diagnosis is required to improve long er16,19,20,22,24,28,29,34. Aim of this review is to dis-
term outcome, and a microbiological confirma- cuss certain aspects of the approach to TSs, fo-
tion is recommended to enable appropriate
choice of anti-mycobacterial agents. Surgery cusing upon epidemiology, diagnosis, and treat-
has an important role in alleviating pain, correct- ment outcome.
ing deformities and neurological impairment,
and restoring function.
Conclusions: Further studies are required to
assess the appropriate duration of anti-microbial
treatment, also in regarding of a combined surgi- Epidemiology
cal approach.

Key words: Tuberculosis remains the most common cause

of death due to an infectious disease worldwide:
Tuberculosis, Spondylodiscitis, Pott’s disease, Epi- according to the World Health Organization’s
demiology, Treatment, Outcome. Global TB Report 2010, in 2009, there were 9.4
million estimated incident cases (range, 8.9 mil-
lion-9.9 million) of TB globally (equivalent to
Introduction 137 cases per 100 000 population); 1.3 million
people died among Human Immunodeficiency
Tuberculous spondylodiscitis (TS), known al- Virus (HIV)-negative people (range, 1.2 million-
so as Pott’s disease, was first described in 1779 1.5 million) and 0.38 million died among HIV-
by Percival Pott, one of the leading surgeons in positive people (range, 0.32 million-0.45 million)
London in the eighteenth century1. because of tuberculosis37.

58 Corresponding Author: Enrico Maria Trecarichi, MD; [email protected]

 Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome

Most of the estimated number of cases of TB more common in patients aged under 40 com-
occurred in Asia (55%) and Africa (30%), and pared to those over 4050.
smaller proportions were registered in Eastern However, a recent Japanese epidemiological
Mediterranean Region (7%), European Region survey reported that the proportion of TB infec-
(4%), and Region of the Americas (3%). Of the tion among patients aged more than 70 years was
9.4 million incident cases in 2009, about 1.0-1.2 31.2% of the total case of TSs in 1994, and by
million (11-13%) were HIV-positive37. 2002 had increased to 41.5%; the same study re-
Data suggest that tuberculosis is still a major ported a similar trend for spinal TB26. Increasing
problem of public health not only in developing life expectancy deals with the occurrence of a se-
countries but also in the western world37, where ries of concatenating events: malnutrition, under-
the highest burden of disease involves immi- lying acute or chronic diseases, and the biologi-
grants and foreign-born patients. cal changes with aging, all contributing to the ex-
In Europe, 329 391 new episodes of TB and pected age-associated decline in cellular immune
46 241 deaths due to TB have been reported in responses to infecting agents such as M. tubercu-
2009; the estimated percentage of extra-pul- losis51,52.
monary cases in European countries was about Risk factors for TS has been largely investigat-
14%38. ed in several studies: underling diseases such as
In United States, even if the number of TB diabetes mellitus and chronic renal failure have
cases reported annually has decreased by approx- been found in 5% to 25% and 2% to 31% of pa-
imately 57% since 1992, with a decline of 10.5% tients, respectively3,6,12,13,16,17,27,31,35, whereas pro-
in 2009 compared to 2008, the proportion of total longed corticosteroid therapy has been reported
cases occurring in foreign-born persons has in- in 3% to 13% of patients3,6,16,17,27,35. TB is the
creased every year from 1993 to 2008, and in most common and virulent opportunistic infec-
2009 59% of TB cases occurred in foreign-born tion associated with HIV disease53, and skeletal
persons39. tuberculosis is more frequent in HIV-positive pa-
Large migratory movements from areas tients than in HIV-negative54. Godlwana et al19,
where TB is endemic, the accumulation in large in their study performed in South Africa, re-
cities of enormous pockets of poverty, unem- ported a rate of HIV seropositive of 28% of to-
ployment, poor nutrition, and poor living facili- tal TS cases, and similar results were found by
ties have been recognized as major elements both Leibert and Rezai in USA (27% and 25%
that play a role in the resurgence in TB in devel- respectively)25,33. Finally, in a patient presenting
oped countries. In addition, the HIV epidemic, with chronic back-pain, high suspicion of TS
the emergence of multidrug-resistant strains of should be evocated by a previous history of TB,
M. tuberculosis, and the immunity deterioration which is reported in a proportion ranging from
due to aging, all contribute to made TB an in- 5% to 100% of patients diagnosed with
creasingly common problem, especially among TS3,6,13,16,19,27,31,32,35.
ethnic minorities 40-47 , and extra-pulmonary
forms of TB have been reported to be more fre-
quent among immigrants in developed
countries48,49. In most of the studies performed
in western countries (United Kingdom, United Pathogenesis
States, France, Switzerland) immigrant patients
represented more than 50% of patients diag- In a significant percentage of TS cases there is
nosed with TS5,23,31-33. no evidence of primary infection: concurrent local-
Demographic characteristics and principal risk izations of TB in other sites are reported in 3% to
factors reported in 29 different observational 65% of cases, with a rate of pulmonary involve-
studies are shown in Table I. ment that ranges from 1% to 67%, mostly account-
The mean age of presentation of TS is report- ing for more than 20% 3,5,6,8,12,13,16,18,24, 26,29,31,33-36
ed to range between 30 to 40 years5,18,20,22,28,29,34,36. (Table II).
In the retrospective review performed by Turgut As for pyogenic spondylodiscitis, TSs can re-
et al36 and including all cases of TS reported in sult from arterial haematogenous seeding of the
Turkey from 1985 to 1996, the mean age was 32 M. tuberculosis starting from a quiescent or ac-
years, whereas in a large French epidemiological tive pulmonary focus, or can be due to contigu-
study on spondylodiscitis, TS was significantly ous or lymphatic spread from pleural disease55.

59
60
Table I. Country, number of patients, demographic characteristics, and risks factors in 29 different observational studies.

Author Study Country Patients Male Age Foreign HIV IDU DM CRF Cortico- Previous
(reference) period (n°) (mean) born steroids TB

Alothman et al (3) 1985-1998 Saudi Arabia 69 53.6 53 - - - 10 - 3 7.2

Alavi et al (16) 1999-2008 Iran 69 60.8 44 - 17.4 17.4 8.7 14.5 13 28.9
Colmenero et al (12) 1983-1995 Spain 42 50 - - - 7.1 4.9 - - -
Colmenero et al [2] (17) 1983-2002 Spain 78 51.3 49 5.1 7.7 11.5 11.5 - 5.1 -
Cormican et al (5) 1999-2004 UK 21 61.9 35 90.5 4.8 4.8 - - - -
Dharmalingam (18) 2000-2002 Malaysia 33 72.7 36 9 - - - - - -
Goldwana et al (19) 2005-2006 South Africa 104 46 - - 28 - - - - 100
Hadadi et al (20) 2003-2005 Iran 22 56.5 40 43.5 - - - - - -
Hayes et al (21) 1985-1992 UK 21 52.4 38 - - - - - - -
Jalle et al (22) 1985-1989 Malaysia 31 61.3 35.4 - - - - - - -
Janssens et al (23) 1976-1986 Switzerland 26 61.5 48 53.8 - - - - - -
Khorvash et al (24) - Iran 100 58 - - - - - - - -
Kim et al (13) 2003-2007 South Korea 47 36.2 55 - - - 17 0 - 12.8
Leibert et al (25) 1988-1995 USA 26 69.2 44 46.1 26.9 26.9 - - - -
Luzzati et al (14) 1995-2005 Italy 27 - - - - - - - - -
Maeda et al (26) 1990-2002 Japan 23 43.5 76 - - - - - - -
Mulleman et al (27) 1986-2003 France 24 37 61 37.5 0 - 8.3 - 8.3 25
Mwachaka et al (28) 2004-2009 Kenia 129 52.7 33 - 0 - - - - -
Nene et al (29) 1998-2000 India 70 38.5 37 - 0 - - - - -
Nussbaum et al (8) 1973-1993 USA 29 41.4 - 58.6 - - - - - -
Park et al (30) 1994-2003 South Korea 137 50.4 44 - - - - - - -
Pertuiset et al (31) 1980-1994 France 103 66 41 68 0 - 1 1.9 1 18
Ramachandran et al (32) 1998-2002 UK 34 54 47 86.2 - - - - - 18.4
Rezai et al (33) 1988-1993 USA 20 70 49 50 25 25 - - - -
Solagberu et al (34) 1994-1999 Nigeria 27 50 27 - - - - - - -
E.M. Trecarichi, E. Di Meco, V. Mazzotta, M. Fantoni

Su et al (35) 2002-2008 Taiwan 48 50 64 - 2.1 6.3 25 31.3 4.2 12.5

Talbot et al (9) 1999-2004 UK 29 - - - - - - - - -
Turgut et al (36) 1985-1996 Turkey 694 50 32 - - - - - - -
Weng et al (6) 1998-2007 Taiwan 38 60.5 68 - - - 18.4 18.4 5.3 5.3

All values are presented as % of patients except as noted otherwise in line 1. Abbreviations: IDU: intravenous drug user; DM: diabetes mellituts; CRF: chronic renal failure; TB:
tuberculosis. Where absent, data were not available.
 Table II. Localization of TS lesions, disease extension, and presence of other foci of TB infection in 29 different observational studies.

Author Involved Cervical Thoracic Thoraco- Lumbar Lumbo- Disc Paraspinal Epidural Posterior Other Pulmonary
(reference) vertebre lumbar locali- sacral involv- abscess compres- involve- foci of TB
(mean) localiza-tion zation localization ment sion ment TB

Alothman et al (3) - - 55 - 36 - - 80 33 - 2.9 1.4

Alavi et al (16) - - 14.5 - - - - 47.8 - - 44.9 17.4
Colmenero et al (12) 2.6 0 64.3 - 33.3 - - 78.3 - - 33.3 -
Colmenero et al[2](17) 2.6 3.9 41 14.1 32.1 6.4 97.4 73.1 65.4 76.9 19.2 11.5
Cormican et al (5) 2.6 - - - - - 95.3 - 65 - 33 19
Dharmalingam (18) - 18.2 30.3 6.1 27.2 6.1 - - - - - 66.6
Goldwana et al (19) - 11 42 10 30 5 - - - - - -
Hadadi et al (20) - - 50 - 22.7 - 53.3 29.6 - - - -
Hayes et al (21) - 14.3 66.7 - 19 - - - - - - 23.8
Jalle et al (22) - - - - - - - - - - - -
Janssens et al (23) 2.2 53.8 - 7.7 38.5 - - 57 - - 38.5 -
Khorvash et al (24) - 5 11 29 - - - 20 - - - 18
Kim et al (13) - 8.5 38.3 - 53.2 - - 59.6 63.8 - 31.9 25.5
Leibert et al (25) - - - - - - - - - - - -
Luzzati et al (14) - 7.4 37 44.4 11.1 0 - - - - - -
Maeda et al (26) 2.3 4.3 47.8 30.4 21.7 - 73.9 86.9 - 17.4 4 4
Mulleman et al (27) - 8.3 45.8 - 50 - 58.3 - - - - -
Mwachaka et al (28) 1.8 6.2 43.4 8.8 37.2 2.7 - - - - - -
Nene et al (29) 2.5 - 84.3 15.7 - - 71.4 32.8 30 2.8 - 2.8
Nussbaum et al (8) - 3 55 10 28 - - 45 31 - 52 10
Park et al (30) 2.2 - 31.3 9.7 44.8 - - 58.6 28.6 - 15.5 -
Pertuiset et al (31) - 12 47 - 66 - 49.5 - - 5.8 27 15.5
Ramachandran et al (32) - 13.1 39.5 - 26.3 - - - - - 31.6 -
Rezai et al (33) 2.5 10 65 10 20 - 85 90 65 - 65 65
Solagberu et al (34) - 7.4 22.2 22.2 40.7 - - - - - 11.1 11.1
Su et al (35) 2.5 4.2 33.3 16.7 41.7 4.2 - 77.1 37.5 - 25 22.9
Talbot et al (9) - 6.9 27.5 13.8 34.5 - - - - - - -
Turgut et al (36) - 4.2 55.8 16.9 22.8 - - - - 4.7 5 2.7
Weng et al (6) 3.5 2 37 8 39 11 84 50 71 - 42.1 39.5

All values are presented as % of patients except as noted otherwise in line 1. Abbreviations: TB, tuberculosis. Where absent, data were not available.
Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome

61
 E.M. Trecarichi, E. Di Meco, V. Mazzotta, M. Fantoni

Since the intervertebral disc does not have a di- ized countries, although the reasons for this re-
rect blood supply in adults, most haematoge- main unclear.
nous infections of the disc space are the result TSs are frequently complicated by forma-
of dissemination from the adjacent bone. The tion of large paravertebral abscesses, which
natural evolution of the infection is the forma- have been reported in 20% to 90% of cas-
tion of a granuloma, whose centre tends to es 3,6,8,12,13,16,17,20,23,24,26,29,30,33,35. Anterior epidural ab-
caseate and to become necrotic. The infection scesses in TSs have generally similar character-
can then progress to destroy the bone, causing istics to those of pyogenic spondylodiscitis55,
pain and leading to the collapse of the vertebral and epidural compression occurs frequently in
bodies producing the classic roentgenographic TSs, with percentages up to 50% of cases in
picture of anterior wedging of two adjacent ver- some studies5,6,13,17,33 (Table II). In addition, ab-
tebral bodies with destruction of the interverte- scesses may be found in sites far from the infec-
bral disk. Physical findings of a tender spine, tious focus, particularly in the sheaths of the
prominence or gibbus are common clinical pre- psoas muscles55-57.
sentation55-57. The vertebral body collapse has TSs have in general a more indolent and less
been reported with a frequency ranging from painful clinical evolution than pyogenic infec-
13% to 87%3,8,26,35. tions, because of microbiological characteristics
The thoracic segment and the thoraco-lumbar of M. tuberculosis consisting in slow growth,
hinge represent the most frequent localizations of propensity for an oxygen rich environment, and
TS, followed by lumbar and cervical segments absence of proteolytic enzymes. Consequently,
(Table II). In the largest case series published by neurological complications due to spinal cord
Turgut, the thoracic spine was involved in 55.8% compression are frequent. Nerve roots may be
of cases, the lumbar in 22.8%, and the cervical in compressed with consequent pain or radiculopa-
4.2%36. However, Park et al30 reported a more thy, but more commonly compression on spinal
frequent involvement of the lumbar segment cord or cauda equina gives rise to myelopathy or
(44.8%), followed by thoracic (31.3%), and tho- paraplegia. These neurological complications can
raco-lumbar hinge (9.7%). occur early in active disease due to inflammatory
The predominant localization in the thoracic tissues, epidural abscess, protruded disc, pachy-
segment could be related to the frequent involve- mengitis, or spinal subluxation, but also after
ment of mediastinal lymph nodes and the pleura years from TS event due to severe kyphosis with
in pulmonary TB, from where microorganisms chronic spinal cord compression and atrophy,
can reach the vertebral bone through the lym- with or without reactivation of the infection (late-
phatic route as mentioned above55. onset paraplegia)54-56,61.
Tuberculous lesions are more likely to involve
more than two vertebrae compared to pyogenic
cases13, ranging the mean number of vertebrae in- Clinical presentation
volved from 1.8 to 3.55,6,12,17,23,26,28-30,33,35 (Table II).
Compared to pyogenic haematogenous Principal signs and symptoms reported in 29
spondilodiscitis, in TSs the posterior elements of different observational studies are shown in
the vertebrae (pedicles, transverse processes, Table III. The clinical manifestations of TSs
laminae, and posterior spinous processes) could are related to both systemic illness and/or local
be involved more frequently58,59. Maeda et al26 re- infection. As stated above, evidence of other
ported a posterior involvement in 17% of TS cas- foci of tuberculosis and systemic symptoms
es; in these patients severe neurological deficits have been reported with rates ranging from 3 to
are generally present60,61. The involvement of the 65%3,5,6,8,12,13,16,18,24,26,29,31,33-36 (Table II). However,
intervertebral disc space has been frequently re- Turgut36 found that only 35 patients out of a to-
ported, with a proportion that varies from 50% to tal of 694 cases reviewed (5%) had evidence of
97.5%5,6,17,20,26-30,33. Pertuiset at al31 identified two other foci of infection and only 16 (2.7%) a pul-
distinct patterns of TS: the classic form of monary involvement.
spondylodiscitis, and an increasingly common Fever is often absent, reported in less than 40%
atypical form characterized by spondylitis with- of cases in most of the studies3,5,6,8,12,16,17,26,27,31,35,
out disk involvement; their data suggest that the whereas weight loss, night sweats, and malaise
atypical form could now be the most common are manifestations that occur generally in less
form of TS in foreign-born subjects in industrial- than 30% of patients3,5,16,27,28,35,36.

62
 Table III. Clinical characteristics of TS in 29 different observational studies.

Author Back pain LEW Constitutional Fever Weight Night Neurological Diagnostic
(reference) symptoms loss sweats symptoms delay (months)

Alothman et al (3) 84 28 - 32 27.5 22 28 -

Alavi et al (16) 98.5 - - 26.1 14.5 17.4 26.1 6.8
Colmenero et al (12) 90 - 45 32 - - 76 5.7
Colmenero et al [2] (17) 83.3 - 35.9 34.6 - - 44.9 6.1
Cormican et al (5) 100 9.5 47.6 24 24 29 29 11
Dharmalingam (18) - - - - - - 51 -
Goldwana et al (19) - - - - - - 56 -
Hadadi et al (20) - - - - - - 68 18
Hayes et al (21) 95.2 - 71.4 - - - 47 3.7
Jalle et al (22) 90.3 - - - - - 30.9 -
Janssens et al (23) - - - - - - 46 4
Khorvash et al (24) 99 - 80 - - - 34 2-3
Kim et al (13) 87.2 - - 17 - - 27.7 3
Leibert et al (25) 100 - - 57.7 57.7 - 26.7 3
Luzzati et al (14) 96.3 - - 14.8 - - 63 4
Maeda et al (26) 82.6 - - 30.4 - - 56.5 4.9
Mulleman et al (27) 100 - 33.3 37.5 41.7 - - 4.3
Mwachaka et al (28) 77.5 72.9 - - 7 20.9 72.9 -
Nene et al (29) - - - - - - 15.7 -
Nussbaum et al (8) 79 - - 45 - - 76 -
Park et al (30) - - - - - - - -
Pertuiset et al (31) 97 - 56 31 48 18 50 4
Ramachandran et al (32) - - - - - - 36.8 -
Rezai et al (33) 100 40 - - 35 - 50 3
Solagberu et al (34) - - - - - - 88.9 -
Su et al (35) 95.8 62.5 - 39.6 8.3 - 77.1 6.9
Talbot et al (9) - - - - - - 20.7 -
Turgut et al (36) 21.3 69.2 - 1.6 - - - -
Weng et al (6) 100 53 8 16 - - 26 2

All values are presented as % of patients except as noted otherwise in line 1. Abbreviations: LEW: lower extremities weakness. Where absent, data were not available
Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome

63
 E.M. Trecarichi, E. Di Meco, V. Mazzotta, M. Fantoni

Back pain, often associated with spine stiffness idence of vertebral collapse on magnetic reso-
and spasms of the paravertebral muscles, is the nance imaging (MRI) in 29 different observation-
most common reported symptom, occurring in al studies are reported in Table IV.
more than 80% of patients3,5,6,12-14,16,17,21,22,24-27,33,35, Among non-invasive diagnostic tools, labora-
followed by lower extremities weakness, found in tory studies can be suggestive of chronic infec-
up to 73% of patients28,36. Torticollis, neck pain, tion (anaemia, hypoproteinemia), and an elevated
stiffness, dysphagia, and/or inspiratory stridor are erythrocyte sedimentation rate (ESR) is usually
the most frequent clinical manifestations in cervi- found8,12,20,21,35,62. Interestingly, the C reactive pro-
cal TSs63. When diagnosis is made in advanced tein (CRP) has been usually found to be normal
stages, soft tissue swelling, draining sinus, and or slightly elevated, whereas white blood cells
spinal tenderness could be evident at physical ex- count is normal in most of cases6,26.
amination16. Kyphotic deformity of the spine oc- Tuberculin skin test and interferon gamma-re-
curs as a consequence of collapse of the anterior lease assays (IGRA) are approved as indirect
spinal elements and is more frequent in thoracic tests for diagnosis of M. tuberculosis infections.
TSs than in other spine localizations64; Colmenero However, these tests do not allow to discriminate
et al12 reported a rate of spinal deformity of 41%. active from latent infections, but when integrated
Neurological involvement has been reported in with clinical and radiological findings, they can
16-89% of TS cases3,5,6,8,12-14,16,17,19-22,24,26,28,29,31-35. give to clinicians an important diagnostic orienta-
Neurological deficits occur because of kyphotic tion66. The percentage of positive tuberculin skin
deformity, spinal abscess, and/or granulation tis- test ranges from 10% to 83% of patients in the
sue compressing the spinal cord or cauda equina64. studies considered8,17,20,22,24,25,27,28,31,33,34, whereas
In the study by Colmenero et al17, neurological no data have been specifically reported to this
impairment has been reported in 45% of TS cases, date about the role of IGRA in TSs.
and paraplegia was present in 6%; these patients Imaging studies are essential for diagnosis and
were more likely to have epidural, paraspinal or management of TSs, and could allow the differ-
psoas abscesses. The most common signs of neu- ential diagnosis by assessing features that are
rological impairment are numbness, lower limb characteristic of certain infectious etiologies, in-
weakness, and urinary disorders20,28,33,34. flammatory lesions, and malignancies. Soft tis-
sue mass with calcification or bony fragments,
vertebral collapse with relative preservation of
Diagnosis the intervertebral disc, gibbus deformity, and
presence of large paravertebral mass or abscesses
The large spread in symptoms duration before have been considered characteristics of TS7,67.
diagnosis reflects both the variable and chronic Plain radiography, usually employed as a screen-
nature of the disease both the difficulty in recog- ing test, characteristically shows a destructive
nising spinal TB. This can explain why the mean process of the thoracic or lumbar vertebrae with in-
duration of symptoms before a correct diagnosis volvement of the adjacent disc space which is usu-
is formulated could be so long: it ranges from 2 ally evident later in the course of the disease and is
months to 4 years6,12-14,16,17,20,24,25,27,33,35. In addition, less pronounced than in pyogenic infections68.
in the elderly diagnosis of TSs could be more dif- Computed tomography (CT) scanning in the
ficult, and delays may result: the common presen- axial plane with bone windows can be utilized to
tation with a persistent, localized pain in the back, define the precise extent of bone involvement
in absence of other systemic symptoms (e.g. and to identify a calcified paraspinal mass, com-
fever), is often misdiagnosed because the majori- mon in TSs68,69 but rare in pyogenic abscesses70.
ty of elderly have some degree of backache due to MRI has become the method of choice for di-
degenerative changes or osteoporosis of the spine. agnosis of TS, due to its capability to provide in-
Of note, spinal metastasis and other causes of formation about the epidural space and spinal
pathological fractures should be considered in the cord71-73. A proportion from 20% to 100% of pa-
differential diagnosis of TSs in elderly26. tients underwent a MR in the studies ana-
For these reasons physicians must exercise a lyzed3,5,6,8,17,26-28,31,33,35. The major MRI findings
high index of suspicion to achieve early diagno- reported by Maeda et al 26 were osteolytic
sis of TSs. Diagnostic tools for TSs can be divid- changes (86%), narrowing of disc space (73%),
ed into two major groups: invasive and non-inva- loss of vertebral body height (69%), erosion of
sive procedures. Diagnostic tools utilized and ev- the vertebral endplates (56%).

64
 Table IV. Radiological and microbiological diagnostic tools and evidence of vertebral collapse on Magnetic Resonance Imaging in 29 different observational studies.

Author (reference) TST° Biopsy* Surgical§ RGNA§ Smear+ Coltural+ Histology+ MRI^ Vertebral Collaps^

Alothman et al (3) - 72.4 54 46 30 50 70 49.3 87

Alavi et al (16) - - - - 21.7 8.7 39.1 18.8 -
Colmenero et al (12) - 80.9 - - - 47 - - -
Colmenero et al[2] (17) 83.1 79.4 60 40 - 56.4 - 56.4 -
Cormican et al (5) - 90.5 21 79 75 76 - 95.2 -
Dharmalingam (18) - 48.5 - - - - - - -
Goldwana et al (19) - - - - - - - - -
Hadadi et al (20) 36.3 - - - - - - - -
Hayes et al (21) - - - - - - - - -
Jalle et al (22) 70.9 - - - - - - - -
Janssens et al (23) - 53.8 - - 14 21 76 - -
Khorvash et al (24) 68 40 - - 25 62.5 92.5 100 -
Kim et al (13) - - - - - 91.1 - - -
Leibert et al (25) 80.8 - - - - - - - -
Luzzati et al (14) - - - - - 81.5 18.5 - -
Maeda et al (26) - 100 - - - - 91.3 100 13
Mulleman et al (27) 58.3 87.5 4.7 95.2 - 65 40 41.7 -
Mwachaka et al (28) 10 25.6 100 0 - 10 - 45 -
Nene et al (29) - 15.7 - 100 - 27.3 45.5 - -
Nussbaum et al (8) 62 100 59 41 - - - 28 31
Park et al (30) - - - - - - - - -
Pertuiset et al (31) 64 90.2 17 83 32 83 70 42.7 -
Ramachandran et al (32) - - - - - - - - -
Rezai et al (33) 95 90 - 89.9 - - - 60 -
Solagberu et al (34) 66.7 - - - - - - - -
Su et al (35) - 93.7 - 60 22.2 40 53.3 100 68.8
Talbot et al (9) - - - 100 - - - - -
Turgut et al (36) - - - - - - - - -
Weng et al (6) - 100 84 16 58 63 84 92 -

All values are presented as % of patients except as noted otherwise in line 1. Abbreviations: TST: Tuberculin Stimulation Test ; RGNA: Radiologically Guided Needle Aspiration;
MRI: Magnetic Resonance Imaging, performed.
°% of TST-positive on total of patients;
*% of biopsies performed on total of patients;
§
% on total of biopsies performed;
+
% of positive results on biopsies performed;
Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome

^
% on total of patients.

65
 E.M. Trecarichi, E. Di Meco, V. Mazzotta, M. Fantoni

In the study conducted by Colmenero et al.17 species belonging to M. tuberculosis complex.

findings compatible with vertebral osteomyelitis However, they require advanced laboratory tech-
were evident in 95.5% of MR scans vs. 85.7% of niques and cannot distinguish dead forms from
plain radiographs, confirming the limitations of viable organisms55.
conventional radiological studies for the diagno- The diagnostic yield of the two bioptic proce-
sis of TS. dures has been evaluated in different studies, and
Nevertheless, radiological investigations could no significant differences have been reported be-
be suggestive for diagnosis of TS, performing a tween RGNA and surgical biopsy in terms of
vertebral biopsy whenever possible is mandatory sensitivity for detection of granuloma, acid-fast
to confirm diagnosis by isolation of M. tubercu- bacillus smear on Ziehl-Neelsen’s stain and/or
losis and for the purpose of testing its antimicro- culture, with rates of diagnostic success ranging
bial sensitivity. from 20% to 90% for both the proce-
Radiologically guided needle aspiration dures3,6,13,14,23,31,74. According to these findings,
(RGNA) is a simple, reliable, and practical ap- and due to the low associated morbidity, RGNA
proach for diagnosing TS, since it provides use- should be recommended for diagnosis in patients
ful material for histopathological and microbio- with no indications for immediate surgery.
logical studies74,75. RGNA is particularly indicat-
ed for patients who are candidates to chemother-
apy alone, and it may also be used for drainage
of paraspinal abscesses33. Otherwise, open surgi- Treatment and outcome
cal biopsy may be necessary if needle biopsy is
unsuccessful, or in the cases in which surgical in- Goals of management of TSs are to eradicate
tervention is indicated for treatment62. the infection, to prevent or treat neurological
More than 70% TS cases underwent a biop- deficits, to correct or avoid the occurrence of
tic procedure in most of published stud- spinal deformities and to achieve normalization
ies3,5,6,8,12,17,26,27,31,33,35, with a variable proportion of patient’s daily activities as soon as possible.
between the surgical and radiologically guided On the basis of current evidences, TSs are a med-
techniques: although RGNAs have been more ical condition, unless accompanied by surgical
frequently used 5,9,27,29,31,33, surgical biopsy has complications57,62,76,77. Medical treatment, dura-
been performed in more than 60% of cases in tion of anti-mycobacterial regimens, surgical
some reports6,17,28. treatment and outcome in 29 different observa-
Histological examination could reveal epithe- tional studies are reported in Table V.
lioid cell granulomas, granular necrotic back- Indications for surgical intervention are pres-
ground with lymphocytic infiltration and, scat- ence of neurological deficits caused by spinal
tered multinucleated Langhans giant cells74,75; the cord compression, spinal deformity with instabil-
proportion of suggestive histology is usually ity, severe or progressive kyphosis, no response
high, up to 90% of cases24,26. to or failure of anti-TB therapy, large paraspinal
Microbiological investigation should include abscesses and non-diagnostic biopsies, even if re-
bacterial, fungal, and mycobacterial stains and peated8,10,76,78.
culture. The definitive diagnosis of TS is ob- A Cochrane review of randomized controlled
tained by culturing M. tuberculosis from patho- trials regarding the use of routine surgery in ad-
logical specimens (bioptic specimen, pus ob- dition to chemotherapy for TS has been con-
tained from the drainage of paraspinal abscesses, ducted by Jutte et al79 in 2006, and no signifi-
and/or sputum for cases in which a concomitant cant benefits have been demonstrated for any of
active pulmonary infection is present). Since cul- the outcomes measured: kyphosis angle, neuro-
tures on solid media require 3 to 8 week to show logical deficits, bony fusion, death from any
a visible growth, there are other techniques en- cause, activity level regained, change of allocat-
abling faster results: acid-fast stain of specimen’s ed treatment or bone loss. However, partially,
smears could give a same-day result with the lim- these results could be due to a less severe dis-
itation to be less sensitive than culture and un- ease of patients included in the review. Several
able to distinguish M. tuberculosis from other studies suggest that the risk of deformity, insta-
Mycobacteria; nucleic acid amplification assays bility, and progression to neurologic lesions or
have a sensibility intermediate between acid-fast deficits correlates with the size of the vertebral
stain and culture and can identify Mycobacterium lesions80-82.

66
 Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome

Table V. Medical treatment, duration of anti-mycobacterial regimens, surgical treatment, and outcome in 29 different obser-
vational studies.

Author Medical Duration MDR TB Surgery Improve- Relapse Neurological

(reference) treatment (months) ment sequelae

Alothman et al (3) 100 - - 46 91.3 - 27.5

Alavi et al (16) 100 - - 31.9 76.8 - -
Colmenero et al (12) 100 10 - 76.2 76.2 7.3 49.3
Colmenero et al [2](17) 100 - 1.3 70.5 - 5.1 -
Cormican et al (5) 100 13 6 28.7 52.4 - 14.2
Dharmalingam (18) 100 - - 61 - - -
Goldwana et al (19) - - - - - - -
Hadadi et al (20) - - - - - - -
Hayes et al (21) 100 12-24 - 47.6 91 9.5 -
Jalle et al (22) - - - 41.9 - - -
Janssens et al (23) 100 - - 34.6 - - 26.9
Khorvash et al (24) 100 12 - 50 - - 11
Kim et al (13) 100 7-10 - 85.1 - - -
Leibert et al (25) - 12 0 - - - -
Luzzati et al (14) 100 9 - 40.7 30.8 0 0
Maeda et al (26) 100 12 - 91.3 - - -
Mulleman et al (27) - - - - - - -
Mwachaka et al (28) 100 9 - 25.6 70 - 18.4
Nene et al (29) 100 12-27 4.3 1.4 98.5 - -
Nussbaum et al (8) 100 - - 86.2 - - 34
Park et al (30) 100 - - 71.5 81 - 1.7
Pertuiset et al (31) 100 14 0 24 70.8 1 -
Ramachandran et al (32) 100 - - 55.5 - 11.7 5.9
Rezai et al (33) 100 18-12 5 55 95 0 5
Solagberu et al (34) 85.1 5.7 - 0 - - -
Su et al (35) 100 12 5.3 62.5 70 - -
Talbot et al (9) 100 - - 13.8 - - -
Turgut et al (36) 99.4 - - 98.4 44.5 - -
Weng et al (6) 100 11 2.6 84 71 0 26

All values are presented as % of patients except as noted otherwise in line 1. Abbreviations MDR: multi drug resistant; TB: tu-
berculosis. Where absent, data were not available.

Conservative management of TSs consists of men should be based on susceptibility tests which
antimicrobial therapy and non-pharmacological should be always performed on initial isolate of
treatments such as immobilization with ortho- M. tuberculosis; patients harboring strains of M.
pedic corsets, which is required when pain is tuberculosis resistant to first line antimicrobial
significant or there is the risk of spinal instabil- compounds (mostly INH and/or RIF) are at high
ity83,84. risk for treatment failure and further acquired re-
The basic principles underling the treatment of sistance, and they must be referred to a specialist
pulmonary TB also apply to TSs. When the diag- in infectious diseases85,86 (in Table VII second line
nosis of TS is correctly formulated, anti-my- anti-mycobacterial drugs are reported).
cobacterial therapy should be initiated, with a 2- The duration of treatment remains controver-
months initial phase of combination of four first sial. The American Thoracic Society (ATS) and
line anti-mycobacterial agents (i.e. isoniazid the Center for Disease Control (CDC) advocate
(INH), rifampin (RIF), pyrazinamide, and etham- short course treatment: 6-9 months for adults and
butol), followed by the continuation phase of ei- 12 months for children with uncomplicated TS
ther 4 or 7 months with isoniazid and rifampin85,86 caused by a fully sensitive M. tuberculosis
(in Table VI drug dosage and administration are isolate 85,88. Medical Research Council (MRC)
reported). First line anti-mycobacterial agents studies showed that a 6-month regimen for TS,
have been demonstrated to have a good bone pen- when combined with surgery, is as effective as a
etration87. 9-month regimen89-91. Other studies have reported
The issue of drug resistance has to be taken in- similar findings in adults, regardless combination
to account, and definitive anti-mycobacterial regi- with surgical treatment30,35,92-94.

67
 E.M. Trecarichi, E. Di Meco, V. Mazzotta, M. Fantoni

Table VI. First line anti-mycobacterial drugs: preparation and daily doses. Modified from CDC, Treatment of Tuberculosis, 200386.

Drug Preparation Daily doses

Isoniazid Tablet, elixir, aqueous solution Adult (max) 5 mg/kg (300 mg)
for i.v. or i.m. injection Children (max) 10-15 mg/kg (300 mg)
Rifampin Capsule, powder for o.a., aqueous Adult (max) 10 mg/kg (600 mg)
solution for i.v injection Children (max) 10–20 mg/kg (600 mg)
Pyrazinamide Tablet Adult* (mg/kg) 1,000 (18.2–25.0)a; 1,500
(20.0–26.8)b; 2,000† (22.2–26.3)c
Children (max) 15–30 mg/kg (2.0 g)
Ethambutol Tablet Adult* (mg/kg) 800 (14.5–20.0)a; 1,200 (16.0–21.4)b;
1,600† (17.8–21.1)c
Children (max) 15–20 mg/kg (1.0 g)
Rifabutin Capsule Adult (max) 5 mg/kg (300 mg)
Children (max) Appropriate dosing for children
is unknown
i.v.: intra venous; i.m.: intra muscular; o.a.: oral administration; *Based on estimated mean body weight: a40-55 kg, b56-75 kg,
c
76-90 kg; †Maximum dose regardless of weight.

In contrast, a longer course of anti-mycobacte- ly people and immigrants, whereas in many de-
rial therapy for TS has been administered in oth- veloping nations, where patients in a young age
er studies3,25,31,33. In addition, in the retrospective are the most affected, it is still a source of clinical
study performed by Ramachandran et al32, they and socio-economic problems, also considering
found an alarming rate of relapse with the 6- the lack of instrumentation, surgical expertise,
months regimen, thus suggesting that treatment and economical resources. Despite the actual
should be continued for at least 9 months. In ad- availability of more effective diagnostic tools,
dition, Cormican et al 5 have suggested that a early recognition of TS remains difficult and a
longer duration of treatment could be required in high index of suspicion is needed due to the
cases with persistent abnormal MR scans, even in chronic nature of the disease and its insidious and
presence of apparent clinical disease resolution. variable clinical presentation. Patients presenting
In general, younger age3,30,31,35 and early diag- with chronic back pain and neurological symp-
nosis have been reported as factors associated toms, with or without a story of previous or active
with a good outcome, whereas presence of para- TB infection, should be investigated in order to
plegia at presentation has been recognised as the rule out TS. A prompt diagnosis, indeed, is re-
most negative prognostic factor3. quired to ensure improved long term outcome,
and a microbiological confirmation is recom-
mended to enable appropriate choice of anti-my-
cobacterial agents. Although TS is essentially a
Conclusions medical condition, surgery has an important role
in alleviating pain, correcting deformities and
TS is a rare but serious clinical condition neurological impairment, and restoring function,
which may lead to severe deformity and early or and it is indicated in selected group of patients.
late neurological complications. In western coun- Further studies are required to assess the more ap-
tries TS is a health issue mostly concerning elder- propriate duration of anti-microbial treatment, al-
so in regarding of a combined surgical approach.
Table VII. Second line anti-mycobacterial drugs.

• Cycloserine
• Ethionamide References
• Streptomycin
• Amikacin/Kanamycin 1) FLAMM ES. PERCIVALL POTT: an 18th century neu-
• Capreomycin rosurgeon. J Neurosurg 1992; 76: 319-326.
• P-Aminosalicylic-Acid (PAS)
• Levofloxacin 2) TURUNC T, DEMIROGLU YZ, UNCU H, COLAKOGLU S,
• Moxifloxacin ARSLAN H. A comparative analysis of tuberculous,
• Gatifloxacin brucellar and pyogenic spontaneous spondy-
lodiscitis patients. J Infect 2007; 55: 158-163.

68
 Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome

3) ALOTHMAN A, MEMISH ZA, AWADA A, AL-MAHMOOD 17) COLMENERO JD, JIMÉNEZ-MEJÍAS ME, REGUERA JM,
S, AL-SADOON S, RAHMAN MM, KHAN MY. Tuber- PALOMINO-NICÁS J, RUIZ-MESA JD, MÁRQUEZ-RIVAS
culous spondylitis: analysis of 69 cases from J, LOZANO A, PACHÓN J. Tuberculous vertebral os-
Saudi Arabia. Spine (Phila Pa 1976) 2001; 26: teomyelitis in the new millennium: still a diagnos-
565-570. tic and therapeutic challenge. Eur J Clin Microbiol
4) SCHLESINGER N, LARDIZABAL A, RAO J, RAO J, MC- Infect Dis 2004; 23: 477-483.
DONALD R. Tuberculosis of the spine: experience 18) DHARMALINGAM M. Tuberculosis of the spine-the
in an inner city hospital. J Clin Rheumatol 2005; Sabah experience. Epidemiology, treatment and
11: 17-20. results. Tuberculosis (Edinb) 2004; 84: 24-28.
5) CORMICAN L, HAMMAL R, MESSENGER J, MILBURN 19) GODLWANA L, GOUNDEN P, NGUBO P, NSIBANDE T,
HJ. Current difficulties in the diagnosis and man- N YAWO K, P UCKREE T. Incidence and profile of
agement of spinal tuberculosis. Postgrad Med J spinal tuberculosis in patients at the only public
2006; 82: 46-51. hospital admitting such patients in KwaZulu-Na-
6) WENG CY, CHI CY, SHIH PJ, HO CM, LIN PC, CHOU tal. Spinal Cord 2008; 46: 372-374.
CH, WANG JH, H O MW. Spinal tuberculosis in 20) HADADI A, RASOULINEJAD M, KHASHAYAR P, MOSAVI
non-HIV-infected patients: 10 year experience of M, MAGHIGHI MORAD M. Osteoarticular tuberculo-
a medical center in central Taiwan. J Microbiol Im- sis in Tehran, Iran: a 2-year study. Clin Microbiol
munol Infect 2010; 43: 464-469. Infect 2010; 16: 1270-1273.
7) SHARIF HS, MORGAN JL, AL SHAHED MS, AL THAGAFI 21) HAYES AJ, CHOKSEY M, BARNES N, SPARROW OC.
MY. Role of CT and MR imaging in the manage- Spinal tuberculosis in developed countries: diffi-
ment of tuberculous spondylitis. Radiol Clin North culties in diagnosis. J R Coll Surg Edinb 1996; 41:
Am 1995; 33: 787-804. 192-196.
8) NUSSBAUM ES, ROCKSWOLD GL, BERGMAN TA, ER- 22) JALLEH RD, KUPPUSAMY I, MAHAYIDDIN AA, YAACOB
ICKSON DL, SELJESKOG EL. Spinal tuberculosis: a MF, YUSUF NA, MOKHTAR A. Spinal tuberculosis: a
diagnostic and management challenge. J Neuro- five-year review of cases at the National Tuber-
surg 1995; 83: 243-247. culosis Centre. Med J Malaysia 1991; 46: 269-
273.
9) TALBOT JC, BISMIL Q, SARALAYA D, NEWTON DA,
FRIZZEL RM, SHAW DL. Musculoskeletal tuberculo- 23) JANSSENS JP, DE HALLER R. Spinal tuberculosis in
sis in Bradford–a 6-year review. Ann R Coll Surg a developed country. A review of 26 cases with
Engl 2007; 89: 405-409. special emphasis on abscesses and neurologic
complications. Clin Orthop Relat Res 1990; 257:
10) WATTS HG, LIFESO RM. Tuberculosis of bones and
67-75.
joints. Current concepts. J Bone Joint Surg Am
1996; 78: 288-298. 24) KHORVASH F, JAVADI AA, IZADI M, JONAIDI JAFARI N,
RANJBAR R. Spinal tuberculosis: a major public
11) MARTINI M, QUAHES M. Bone and joint tuberculo-
health hazard in Isfahan. Pak J Biol Sci 2007; 10:
sis: a review of 652 cases. Orthopedics 1988; 2:
3400-3404.
861-866.
25) L EIBERT E, S CHLUGER NW, B ONK S, R OM WN.
12) COLMENERO JD, JIMÉNEZ-MEJÍAS ME, SÁNCHEZ-LO-
Spinal tuberculosis in patients with human im-
RA FJ, REGUERA JM, PALOMINO-NICÁS J, MARTOS F,
munodeficiency virus infection: clinical presenta-
GARCÍA DE LAS HERAS J, PACHÓN J. Pyogenic, tu-
tion, therapy and outcome. Tuberc Lung Dis 1996;
berculous, and brucellar vertebral osteomyelitis: a
77: 329-334.
descriptive and comparative study of 219 cases.
Ann Rheum Dis 1997; 56: 709-715. 26) MAEDA Y, IZAWA K, NABESHIMA T, YONENOBU K. Tu-
berculous spondylitis in elderly Japanese pa-
13) KIM CJ, SONG KH, JEON JH, PARK WB, PARK SW,
tients. J Orthop Sci 2008; 13: 16-20.
KIM HB, OH MD, CHOE KW, KIM NJ. A compara-
tive study of pyogenic and tuberculous spondy- 27) M ULLEMAN D, M AMMOU S, G RIFFOUL I, AVIMADJE
lodiscitis. Spine (Phila Pa 1976) 2010; 35: 1096- A, GOUPILLE P, VALAT JP. Characteristics of pa-
1100. tients with spinal tuberculosis in a French
teaching hospital. Joint Bone Spine 2006; 73:
14) LUZZATI R, GIACOMAZZI D, DANZI MC, TACCONI L,
424-427
CONCIA E, VENTO S. Diagnosis,management and
outcome of clinically-suspected spinal infection. J 28) M WACHAKA PM, R ANKETI SS, N CHAFATSO OG,
Infect 2009; 58: 259-265. KASYOKA BM, KIBOI JG. Spinal tuberculosis among
human immunodeficiency virus-negative patients
15) PERRONNE C, SABA J, BEHLOUL Z, SALMON-CÉRON
in a Kenyan tertiary hospital: a 5-year synopsis.
D, LEPORT C, VILDÉ JL, KAHN MF. Pyogenic and tu-
Spine J 2011; 11: 265-269.
berculous spondylodiskitis (ver tebral os-
teomyelitis) in 80 adult patients. Clin Infect Dis 29) NENE A, BHOJRAJ S. Results of nonsurgical treat-
1994; 19: 746–750. ment of thoracic spinal tuberculosis in adults.
Spine J 2005; 5: 79-84.
16) ALAVI SM, SHARIFI M. Tuberculous spondylitis: risk
factors and clinical/paraclinical aspects in the 30) PARK DW, SOHN JW, KIM EH, CHO DI, LEE JH, KIM
south west of Iran. J Infect Public Health 2010; 3: KT, HA KY, JEON CH, SHIM DM, LEE JS, LEE JB,
196-200. CHUN BC, KIM MJ. Outcome and management of

69
 E.M. Trecarichi, E. Di Meco, V. Mazzotta, M. Fantoni

spinal tuberculosis according to the severity of 44) SNIDER DE JR, ROPER WL: The new tuberculosis.
disease: a retrospective study of 137 adult pa- N Engl J Med 1992; 326: 703-705.
tients at Korean teaching hospitals. Spine (Phila 45) O DONE A, R ICCÒ M, M ORANDI M, B ORRINI BM,
Pa 1976) 2007; 32: 130-135. PASQUARELLA C, SIGNORELLI C. Epidemiology of tu-
31) PERTUISET E, BEAUDREUIL J, LIOTÉ F, HORUSITZKY A, berculosis in a low-incidence Italian region with
K EMICHE F, R ICHETTE P, C LERC -W YEL D, C ERF - high immigration rates: differences between not
PAYRASTRE I, DORFMANN H, GLOWINSKI J, CROUZET Italy-born and Italy-born TB cases. BMC Public
J, BARDIN T, MEYER O, DRYLL A, ZIZA JM, KAHN MF, Health 2011; 23: 376.
KUNTZ D. Spinal tuberculosis in adults. A study of
46) AILINGER RL, MARTYN D, LASUS H, LIMA GARCIA N.
103 cases in a developed country, 1980-1994.
The effect of a cultural intervention on adherence
Medicine (Baltimore) 1999; 78: 309-320.
to latent tuberculosis infection therapy in Latino
32) RAMACHANDRAN S, CLIFTON IJ, COLLYNS TA, WATSON immigrants. Public Health Nurs 2010; 27: 115-
JP, PEARSON SB. The treatment of spinal tubercu- 120.
losis: a retrospective study. Int J Tuberc Lung Dis
2005; 9: 541-544. 47) RAJAGOPALAN S. Tuberculosis and aging: a global
health problem. Clin Infect Dis 2001; 33: 1034-
33) REZAI AR, LEE M, COOPER PR, ERRICO TJ, KOSLOW 1039.
M. Modern management of spinal tuberculosis.
Neurosurgery 1995; 36: 87-97; discussion 97-8. 48) MEDICAL RESEARCH COUNCIL. Medical Research
Council National Survey of Tuberculosis Notifica-
34) SOLAGBERU BA, AYORINDE RO. Tuberculosis of the tions in England and Wales in 1983: characteris-
spine in Ilorin, Nigeria. East Afr Med J 2001; 78: tics of disease Tubercle; 68: 19-32.
197-199
49) HESSELINK DA, YOO SM, VERHOEVEN GT, BROUW-
35) SU SH, TSAI WC, LIN CY, LIN WR, CHEN TC, LU ERS JW, SMIT FJ, VAN SAASE JL. A high prevalence
PL, H UANG PM, T SAI JR, WANG YL, F ENG MC, of culture-positive extrapulmonary tuberculosis in
WANG TP, CHEN YH. Clinical features and out- a large Dutch teaching hospital. Neth J Med
comes of spinal tuberculosis in southern Tai- 2003; 61: 65-70.
wan. J Microbiol Immunol Infect 2010; 43: 291-
300. 50) GRAMMATICO L, BARON S, RUSCH E, LEPAGE B, SUR-
ER N, DESENCLOS JC, BESNIER JM. Epidemiology
36) TURGUT M. Spinal tuberculosis (Pott’s disease): its
of vertebral osteomyelitis (VO) in France: analysis
clinical presentation, surgical management, and
of hospital-discharge data 2002-2003. Epidemiol
outcome A survey study on 694 patients. Neuro-
Infect 2008; 136: 653-660.
surg Rev 200; 24: 8-13.
37) WORLD HEALTH ORGANIZATION (WHO). Global Tu- 51) MORI T, LEUNG CC. Tuberculosis in the global ag-
berculosis control. WHO report 2010. Geneva, ing population. Infect Dis Clin North Am 2010; 24:
Switzerland: World Health Organization press. 751-768.
38) EUROPEAN CENTRE FOR DISEASE PREVENTION AND 52) RAJAGOPALAN S, YOSHIKAWA TT. Tuberculosis in the
CONTROL/WHO REGIONAL OFFICE FOR EUROPE. Tu- elderly. Z Gerontol Geriatr 2000; 33: 374-380.
berculosis surveillance in Europe 2009. Stock- 53) Mortality Among Patients with Tuberculosis and
holm: European Centre for Disease Prevention Associations with HIV Status–United States,
and Control, 2011. 1993-2008. MMWR Morb Mortal Wkly Rep 2010;
39) CDC. Reported Tuberculosis in the United States, 59: 46.
2009. Atlanta, GA: U.S. Department of Health and 54) W EINSTEIN MA, E ISMONT FJ. Infections of the
Human Ser vices, CDC, October 2010. spine in patients with human immunodeficien-
http://www.cdc.gov/tb. Accessed September, 2011. cy virus. J Bone Joint Surg Am 2005; 87: 604-
40) Prevention and control of tuberculosis in US com- 609.
munities with at risk minority population: Recom- 55) Mandell, Douglas and Bennett’s principles and
mendation of the Advisory Council for the Elimi- practice of infectious diseases. Seventh edition.
nation of Tuberculosis. MMWR Morb Mortal Wkly Churchill Livingstone Elsevier, 2010.
Rep 1992; 41: 1-11.
56) ALMEIDA A. Tuberculosis of the spine and spinal
41) SELWYN PA, HARTEL D, LEWIS VA, SCHOENBAUM EF, cord. Eur J Radiol 2005; 55: 193-201.
V ERMUND SH, K LEIN RS, WALKER AT, F RIEDMAN
GH. A prospective study of the risk of tuberculo- 57) SKAF GS, KANAFANI ZA, ARAJ GF, KANJ SS. Non-
sis among intravenous drug users with human im- pyogenic infections of the spine. Int J Antimicrob
munodeficiency virus infection. N Engl J Med Agents 2010; 36: 99-105.
1989; 320: 545-550. 58) BABINCHAK TJ, RILEY DK, ROTHERAM EB JR. Pyo-
42) SCHIEFFELBEIN CW JR, SNIDER DE JR. Tuberculosis genic vertebral osteomyelitis of the posterior el-
control among homeless populations. JAMA ements. Clin Infect Dis 1997; 25: 221-224.
1988; 148: 1843-1846. 59) MAIURI F, IACONETTA G, GALLICCHIO B, MANTO A,
43) Plan to Combat Extensively Drug-Resistant Tu- BRIGANTI F. Spondylodiscitis. Clinical and magnet-
berculosis. MMWR Morb Mortal Wkly Rep 2009; ic resonance diagnosis. Spine (Phila Pa 1976)
58: 1-48. 1997; 22: 1741–1761.

70
 Tuberculous spondylodiscitis: epidemiology, clinical features, treatment, and outcome

60) KUMAR K. A clinical study and classification of 77) PERTUISET E. Medical therapy of bone and joint tu-
posterior spinal tuberculosis. Int Orthop 1985; 9: berculosis in 1998. Rev Rheum Engl Ed 1999; 66:
147-152. 152-157.
61) RAHMAN N. Atypical forms of tuberculosis. J Bone 78) CHEN WJ, CHEN CH, SHIH CH. Surgical treatment
Joint Surg Br 1980; 62: 162–165. of tuberculous spondylitis. 50 patients followed for
62) CHEUNG WY, LUK DK. Tuberculosis of the spine. 2-8 years. Acta Orthop Scand 1995; 66: 137-142.
Orthopaedics Trauma 2011; 25: 161-166. 79) JUTTE PC, VAN LOENHOUT-ROOYACKERS JH. Routine
surgery in addition to chemotherapy for treating
63) LIFESO R. Atlanto-axial tuberculosis in adults. J
spinal tuberculosis. Cochrane Database Syst Rev
Bone Joint Surg Br 1987; 69: 183-187
2006.
64) JUTTE P, WUITE S, THE B, VAN ALTENA R, VELD-
80) KORKUSUZ F, ISLAM C, KORKUSUZ Z. Prevention of
HUIZEN A. Prediction of deformity in spinal tuber-
postoperative late kyphosis in Pott’s disease by
culosis. Clin Orthop Relat Res 2007; 455: 196-
anterior decompression and intervertebral graft-
201
ing. World J Surg 1997; 21: 524-528.
65) JAIN AK, AGGARWAL A, MEHROTRA G. Correlation of
81) R AJASEKARAN S. The problem of deformity in
canal encroachment with neurological deficit in
spinal tuberculosis. Clin Orthop Relat Res 2002;
tuberculosis of the spine. Int Orthop 1999; 23: 85-
398: 85-92.
86.
82) RAJASEKARAN S, SHANMUGASUNDARAM TK. Predic-
66) CENTERS FOR DISEASE CONTROL AND PREVENTION. tion of the angle of gibbus deformity in tuberculo-
Updated Guidelines for Using Interferon Gamma sis of the spine. J Bone Joint Surg Am 1987; 69:
Release Assays to Detect Mycobacterium tuber- 503-509.
culosis Infection–United States, 2010. Morb Mor-
tal Wkly Rep 2010; 59: RR-5. 83) G OULIOURIS T, A LIYU SH, B ROWN NM. Spondy-
lodiscitis: update on diagnosis and management.
67) SHARIF HS, AIDEYAN OA, CLARK DC, MADKOUR MM, J Antimicrob Chemother 2010; 65: 11-24.
AABED MY, MATTSSON TA, AL-DEEB SM, MOUTAERY
KR. Brucella and tuberculous spondylitis: compar- 84) QUINONES-HINOJOSA A, JUN P, JACOBS R, ROSEN-
ative imaging features. Radiology 1989; 171: 419- BERG WS, WEINSTEIN PR. General principles in the
425 medical and surgical management of spinal infec-
tions: a multidisciplinary approach. Neurosurg Fo-
68) JEVTIC V. Vertebral infection. Eur Radiol 2004; 14: cus 2004; 17: E1.
43-52.
85) CENTERS FOR DISEASE CONTROL AND PREVENTION.
69) GROPPER GR, ACKER JD, ROBERTSON JH. Comput- Treatment of Tuberculosis, American Thoracic
er tomography in Pott’s disease. Neurosurgery Society, CDC, and Infectious Diseases Society of
1982; 10: 506-508. America. Morb Mortal Wkly Rep 2003; 52: 1-88.
70) LAROCCA H. Spinal sepsis, in Rothman RH, Sime- 86) WHO (2010) Treatment of tuberculosis: Guide-
one FA (eds). The Spine. Philadelphia, 1982; pp. lines for national programmes. Available:
757-774. http://whqlibdoc.who.int/publications/2010/978924
71) KIM NH, LEE HM, SUH JC. Magnetic resonance 1547833_eng.pdf. [Accessed November 2011]
imaging for the diagnosis of tuberculous spondyli- 87) KUMAR K. The penetration of drugs into the le-
tis. Spine 1994; 19: 2451-2455 sions of spinal tuberculosis. Int Orthop 1992; 16:
72) MODIC MT, FEIGLIN DH, PIRAINO DW, BOUMPHREY F, 67-68.
WEINSTEIN MA, DUCHESNEAU PM, REHM S. Verte- 88) BLUMBERG HM, BURMAN WJ, CHAISSON RE, DALEY
bral osteomyelitis: assessment using MR. Radiol- CL, ETKIND SC, FRIEDMAN LN, FUJIWARA P, GRZEM-
ogy 1985; 157: 157-166. SKA M, H OPEWELL PC, I SEMAN MD, JASMER RM,
73) LEDERMANN HP, SCHWEITZER ME, MORRISONWB, KOPPAKA V, MENZIES RI, O’BRIEN RJ, REVES RR,
CARRINO JA. MR imaging findings in spinal infec- R EICHMAN LB, S IMONE PM, S TARKE JR, V ERNON
tions: rules or myths? Radiology 2003; 228: 506- AA; American Thoracic Society, Centers for Dis-
514. ease Control and Prevention and the Infectious
Diseases Society. American Thoracic Soci-
74) MONDAL A. Cytological diagnosis of vertebral tu- ety/Centers for Disease Control and Preven-
berculosis with fine-needle aspiration biopsy. J tion/Infectious Diseases Society of America: treat-
Bone Joint Surg Am 1994; 76: 181-184. ment of tuberculosis. Am J Respir Crit Care Med
75) FRANCIS IM, DAS DK, LUTHRA UK, SHEIKH Z, SHEIKH 2003; 167: 603-662.
M, BASHIR M. Value of radiologically guided fine 89) MEDICAL RESEARCH COUNCIL WORKING PARTY ON
needle aspiration cytology (FNAC) in the diagno- TUBERCULOSIS OF THE SPINE. Five-year assessment
sis of spinal tuberculosis: a study of 29 cases. Cy- of controlled trials of short-course chemotherapy
topathology 1999; 10: 390-401. regimens of 6, 9 or 18 months’ duration for spinal
76) MOON MS. Tuberculosis of the spine. Controver- tuberculosis in patients ambulatory from the start
sies and a new challenge. Spine (Phila Pa 1976) or undergoing radical surgery. Int Orthop 1999;
1997; 22: 1791-1797. 23: 73-81.

71
 E.M. Trecarichi, E. Di Meco, V. Mazzotta, M. Fantoni

90) MEDICAL RESEARCH COUNCIL WORKING PARTY ON spinal tuberculosis. Int J Tuberc Lung Dis 2002;
T UBERCULOSIS OF THE S PINE . Controlled trial of 6: 259-265.
short-course regimens of chemotherapy in the 93) UPADHYAY SS, SAJI MJ, YAU AC. Duration of anti-
ambulatory treatment of spinal tuberculosis: re- tuberculosis chemotherapy in conjunction with
sults at three years of a study in Korea. J Bone radical surgery in the management of spinal tu-
Joint Surg Br 1993; 75: 240-248. berculosis. Spine (Phila Pa 1976) 1996; 21:
91) MEDICAL RESEARCH COUNCIL WORKING PARTY ON 1898-1903.
TUBERCULOSIS OF THE SPINE. A controlled trial of 94) PARTHASARATHY R, S RIRAM K, S ANTHA T, P RAB -
six-month and nine-month regimens of HAKAR R, S OMASUNDARAM PR, S IVASUBRAMANIAN
chemotherapy in patients undergoing radical S. Short-course chemotherapy for tuberculo-
surgery for tuberculosis of the spine in Hong sis of the spine. A comparison between am-
Kong. Tubercle 1986; 67: 243-259. bulant treatment and radical surgery–ten-year
92) VAN L OENHOUT-R OOYACKERS JH, V ERBEEK AL, report. J Bone Joint Surg Br 1999; 81: 464-
J U T T E PC. Chemotherapeutic treatment for 471.

72