Journal of Infection and Public Health (2010) 3, 196200

Tuberculous spondylitis: Risk factors and

clinical/paraclinical aspects in the
south west of Iran
S.M. Alavi a,, M. Shari b

a Infectious Disease Ward, Jundishapur Infectious and Tropical Diseases Research Center, Razi Hospital of
Jundishapur University of Medical Sciences, No. 52, West 11 Avenue, Kianabad, Ahvaz, Iran
b Ahvaz Health Center, Iran

Received 16 April 2010 ; received in revised form 7 September 2010; accepted 10 September 2010

KEYWORDS Summary
Tuberculosis spondilitis; Background and objectives: Tuberculous spondylitis (TS) is both the most common
Risk factors; and the most dangerous form of TB infection. Delay in diagnosis and management
Clinical aspects; causes spinal cord compression and spinal deformity. The aim of this study was to
Paraclinic;
identify the clinical and paraclinical aspects and also to describe its risk factors in
Khuzestan, a province located in the south west of Iran.
Khuzestan
Method: In this medical record-based retrospective study 69 cases of TS registered
in Khuzestan Health Center from 1999 to 2008, were reviewed. For each TS case two
extra pulmonary TB cases (without spinal involvement) were randomly selected as
control. Related data in patients were analyzed in SPSS software (version 16, USA)
using chi square and Fishers exact test. Differences with P < 0.05 were considered
signicant.
Results: The mean age of patients was 43.7 18.3 years, and 60.8% were males. The
mean time of delay: for patients delay, doctors delays and from diagnosis to initi-
ation of treatment was 1.8 1.1, 6.8 4.3 and 1.3 1.2 months, respectively. In 56
cases (81.1%), TS was diagnosed on lumbosacral radiograph, in remaining cases by
lumbosacral MRI. Twenty patients (30.4%) had a previous TB history, 30.4% had under-
lying medical disorders such as diabetes mellitus (30%), steroid use (45%), chronic
renal failure (50%). The most common clinical ndings were: backache 98.5%, fever
26.1%, spinal tenderness 84.1%, paraparesis 26.1%, and kyphosis 28.9%. Labora-
tory results were: elevated sedimentation rate (ESR) 92.8% and positive CRP 86.9%.
There were statistically signicant differences in age, gender, CRF, imprisonment
and previous TB infection between the two groups.

Corresponding author. Tel.: +98 611 3387724; +98 9161184916.

E-mail address: [email protected] (S.M. Alavi).

1876-0341/$ see front matter 2010 King Saud Bin Abdulaziz University for Health Sciences. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jiph.2010.09.005
Tuberculous spondylitis risk factors and its clinical/paraclinical aspects in the south west of Iran 197

Conclusion: TS is therefore a diagnosis that attending doctors need to consider early.

Patients in endemic area of TB, presenting with back pain in association with con-
stitutional symptoms and elevated ESR and or positive CRP should be investigated to
exclude TS. Older age, male gender, chronic peritoneal dialysis, imprisonment and
previous TB infection may be the main risk factors of TS.
2010 King Saud Bin Abdulaziz University for Health Sciences. Published by Elsevier
Ltd. All rights reserved.

Introduction ferent countries are reported [518]. Since data

on TS from Iran is scanty despite the endemicity of
Tuberculous spondylitis (TS) or spinal tuberculosis is disease, we conducted this study aiming to identify
usually secondary to pulmonary or intestinal tuber- the clinical and paraclinical aspects of TS and also
culosis and may also be the rst manifestation of to describe probable risk factors of the disease in
tuberculosis (TB). Spinal TB (often called Potts Khuzestan, a province located in the south west of
disease) is by denition, an advanced disease, Iran.
requiring meticulous assessment and aggressive
systemic therapy [1]. In addition to systemic ther-
apy of TS, surgical intervention may be indicated.
Surgery is indicated for diagnostic dilemma, neu- Methods and patients
rological complication and prevention of kyphosis
progression [2]. This was a medical record-based retrospective
TS is both common and most dangerous form of study in which patients les with the diagnosis
TB infection. Delay in establishing diagnosis and of TS were reviewed. All cases of TS registered in
management causes spinal cord compression and Khuzestan Health Center (KHC) afliated to Ahvaz
spinal deformity [3]. Jundishapur University of Medical Sciences over the
Physicians should keep the diagnosis in mind, 10-year period from 1999 to 2008, were reviewed.
especially in a patient from a group with a high According to Iranian Health System, TB is an obliga-
rate of tuberculosis infection. In endemic countries tory reportable disease and all TB cases around the
TS usually occurs in young adults, but in developed country, in villages (Khane-e-Behdasht) and cities
countries it has become a disease of older persons (hospitals) should be reported to health centers
[4,5]. (provinces, medical universities) under National
Symptoms of spinal tuberculosis are back pain, TB Program. By tracing the TB epidemiological
weakness, weight loss, fever, fatigue, and malaise forms in KHC we listed the hospitals where the
[68]. Paraspinal cold abscesses develop in 80% or patients were diagnosed as TS by neurosurgeons or
more [6]. It is much more prone to develop neuro- infectious disease subspecialist. All medical charts
logical manifestation, paraplegia of varying degrees of TB patients with spinal involvement in above-
[9]. mentioned hospitals were reviewed. Only those
The palpation of spinous process in routine clin- with histological or microbiological conrmation
ical examination is the most rewarding clinical or those who responded to antituberculosis ther-
method and is an invaluable measure for early apy were included [6]. For each TS case two
recognition [1,4]. Diagnosis of TS is made by typical extra pulmonary TB cases (without spinal involve-
sign and symptoms along with the evidence of past ment) randomly selected as control. The following
exposure to tuberculosis or concomitant visceral information was obtained: age, sex, symptoms
tuberculosis, imaging showing abscess formation and signs at presentation, imaging techniques,
and spinal cord compression and by appropriate mycobacterium cultures and smear, histopathology
clinical response to anti-TB treatment. study, treatment and clinical outcome. Additional
The diagnosis is conrmed by microbiological data about addiction, imprisonment, underlying
and histopathological examination [1,4,6]. Microbi- diseases (e.g. diabetes, renal or liver disease, and
ological examination should comprise microscopy, HIV/AIDS), drug side effects and duration between
culture and PCR analysis for rapid detection of TS disease occurrence and initiation of treatment also
[10]. Various and dissimilar data about epidemio- were recorded.
logical, clinical, laboratory results, diagnosis and Data were analyzed in SPSS software (version
treatment outcome from various hospitals and dif- 16, USA) using chi square and Fishers exact test.
198 S.M. Alavi, M. Shari

Differences with P < 0.05 were considered signi- Table 1 Frequency of clinical, laboratory and imag-
cant. ing ndings among studied patients with spinal
tuberculosis.
Variables Number Percentage
Results Clinical
Fever 18 26.1
During the period of study, a total of 15,678 cases of Night sweating 12 17.4
Weight loss 10 14.5
pulmonary and extrapulmonary TB were registered
Back pain 68 98.5
in KHC, 213 patients were diagnosed as TS. We only Local tenderness 58 84.1
could nd 69 medical records (having needed data Paraparesis 18 26.1
for study) of TS in our studied hospitals. Kyphosis 20 28.9
The mean age SD of patients was 43.7 18.3 Laboratory
years, and 42 (60.8%) of the patients were male. Leukocytosis 10 14.5
The mean intervals from onset of symptoms to ESR >20 64 92.8
attend the health center or private physician Positive CRP 60 86.9
(patients delay), from doctors visit to diagnosis Imaging
(doctors delay) and from diagnosis to the initi- Chest X-ray 18 26.1
ation of treatment were 1.8 1.1, 6.8 4.3 and Spinal X-ray 56 81.1
Spinal MRI 13 18.8
1.3 1.2 months, respectively. In 56 cases (81.1%),
Bone scanning 10 14.5
TS was suspected to exist after observing wedge
pattern on spinal radiograph, in remaining cases ESR: erythrocyte sedimentation rate, CRP: C-reactive
protein, MRI: magnetic resonance imaging.
spinal MRI was the initial diagnostic tool in diag-
nosis. Twenty patients (30.4%) had a previous TB
history. Cavitary lesion was observed in 12 patients
(17.4%) and the upper lobes were predominantly cases (68.1%), whereas 22 patients (31.9%) required
involved on chest radiograph. Twenty-one patients additional surgical intervention, mainly those with
(30.4%) had underlying medical disorders such as spinal cord compression, spinal deformity, or risk
diabetes mellitus in 6 (30%), steroid use in 9 (45%), of spinal instability. There was denite improve-
chronic renal failure (CRF) in 10 (50%) and liver dis- ment in 53 cases (76.8%). There was no serious
eases in 3 (15%). About 80% of these CRF patients drug side effect requiring drug discontinuation. The
were undergoing peritoneal dialysis and were reg- best outcome was in those patients diagnosed in
istered as intestinal TB. More than one underlying early phase before the occurrence of spinal defor-
diseases were detected in 7 of the patients. Fif- mity or neurological symptoms. By comparing data
teen patients (21.7%) had history of stay in prison, between patients with spinal tuberculosis (TS) and
among them 12 (80%) were intravenous drug users. patients without spinal involvement (NTS), statis-
Twelve patients (17.4%) had concurrent HIV infec- tically signicant differences in age, gender, CRF,
tion. TS concurrent with TB of other sites was imprisonment and previous TB between two groups
as follows: cervical lymph node 3 (4.3%), intesti- (Table 2) were noted.
nal TB 11 (15.9%) and genitourinary TB 5 (7.2%).
As shown in Table 1 the most frequent symptoms
were backache 68 (98.5%) and fever 18 (26.1%), Discussion
and the most frequent sign was spinal tenderness
58 (84.1%). Other clinical ndings were: paraparesis This study showed that TS is more prevalent in older
18 (26.1%), kyphosis 20 (28.9%). Laboratory results population than the younger population. Previous
were: elevated sedimentation rate ESR 64 (92.8%), studies have reported that in endemic TB area,
CRP 60 (86.9%), other laboratory workup was not skeletal TB complications occurred most commonly
helpful. Although spinal X-ray in some cases was in children and young adults [1,4]. Mean age of TS
helpful in diagnosis, but MRI of the spine using patients in Fennira et al. study was 48.6 years and in
contrast medium was the best diagnostic imaging Garca-Lechuz study was 58 years [15,16]. Our nd-
tool for detecting TS. Ten patients (14.5%) had tho- ing is consistent with the results of investigations in
racic spine involvement and 33 patients (47.8%) had developed countries where the occurrence of pedi-
paraspinal abscesses. Tissue aspirates had a yield of atric TS is extremely rare [4,5]. Iran, although is an
27 (39.1%), 15 (21.7%), and 6 (8.7%) for granulomas, endemic area for TB, but we believe that mass vac-
acid-fast bacilli on smear examination, and culture, cination of BacilleCalmetteGuerin (BCG) started
respectively. Medical therapy alone was given in 47 in 1984 and continuing through the integrated
Tuberculous spondylitis risk factors and its clinical/paraclinical aspects in the south west of Iran 199

Table 2 Comparing the risk factors for spinal tuberculosis between patients with and without spinal involvement.
Variables TS-group (n = 69) NTS-group (n = 138) P-value OR 95% CI
Age
>35 years 58(84.1) 73(52.9) <0.0001 4.7: 2.39.7a
<35 years 11(15.9) 65(47.1)
Sex
Male 42(60.9) 62(44.9) 0.03 1.9: 1.13.4a
Female 27(39.1) 76(55.1)
Diabetes mellitus 6(8.7) 9(6.5) 0.57 1.4: 0.54
Chronic renal failure 10(14.5) 8(5.8) 0.04 2.7: 1.17.3a
IVDU 12(17.4) 18(13) 0.41 1.4: 0.63.1
History of imprisonment 15(21.7) 12(8.7) 0.01 2.9: 1.36.6a
History of corticosteroid 9(13) 12(8.7) 0.33 1.6: 0.63.9
History of previous TB 20(28.9) 24(17.4) 0.04 1.9: 1.53.9a
HIV co-infection 12(17.4) 19(13.8) 0.53 1.3: 0.62.9
IVDU: intravenous drug user, TB: tuberculosis, HIV: human immunodeciency syndrome.
a Risk factor for spinal tuberculosis, TS: patients with spinal tuberculosis, NTS: patients without spinal involvement.

Expanded Program of Immunization (EPI) since 1993 pain, tenderness and paraparesis [4,6,8]. Our nd-
resulted in signicant decreases in TB complications ing regardless of their frequency is in agreement
in children and young adults. with these studies.
According to the present study TS occurred In our study a history of previous TB was observed
more frequently in males than in females (60.8% in only 30.4%. This rate in published studies ranges
vs. 39.2%). Mulleman et al. in their work have from 7.2% to 100% [6,7,9,12,13]. These differences
reported female predominance [13] but in some reect the variation of TB prevalence among differ-
other reports prevalence of TS in both males and ent communities with different socioeconomic and
females was equal [68,14]. These differences are public health status.
not clear for us; we think that in our study males We found elevated ESR in 92.8% and positive
due to their life style and underlying diseases may CRP in 86.9% patients. Diwakar et al., Alothman
be at a higher risk of TB skeletal complication. et al and Rodriguez-Gomez have described ele-
In the most endemic areas for TB, TB with its vated ESR in their studies in more than 90% of
complication is a neglected disease and there is their cases [6,7,12]. CRP was positive in only 26%
a considerable gap between occurrence of illness cases in Maeda et als report. We believe the dif-
and the time of diagnosis [7]. In our study this ference between our result and the result of Maeda
gap was far away from the previous studies (8.6 is contributed to aged patients in Maedas study.
months vs. 4 months). This delay is due to the negli- Overall, we think CRP and ESR are useful tests
gence of patients and lack of awareness of primary in diagnosing TS in kyphotic cases in endemic TB
physician[12]. We suppose that low level of pub- area.
lic awareness about TB and its complication in the As mentioned above, if there is any suspicion of
region is the main reason for these phenomena. spinal TB, MRI of the spine using contrast medium
We believe that the reasons for delay 1.3 months should be implemented as diagnostic means. After
from diagnosis to the initiation of treatment may the collection of spinal and/or paraspinal speci-
be as follows: (1) Patients delay from the time of men, microbiological examination should comprise
referral by practitioner to health center to initi- microscopy, culture and PCR to avoid any delay to
ate anti-TB drugs. (2) Primary denial of the patient establish the correct diagnosis and specic treat-
having TB, because TB is still considered a stigma ment.
in public thought in the region. (3) Some patients In this study most patients were cured by anti-TB
low opinion on health care provider comparing with drugs. Neurosurgery was performed in only 31.9%
experienced aged physicians (mostly not familiar patients that is similar to the study of Rodriguez-
with the health policy of TB national program). (4) Gomez (32%). Operations for spinal tuberculosis are
Lack of some specic anti-TB drugs in rural health now indicated less for the control of disease than
centers while the patient needs in the rst visit. for complications, including nonresponding neural
It is documented that vertebral involvement is a decit (nearly 40% of neural complications), pre-
chronic process in TB and the chief complaints of vention or correction of severe kyphotic deformity,
the patients are local manifestations such as back and for tissue diagnosis [18].
200 S.M. Alavi, M. Shari

The present study revealed that older age, male References

gender, chronic peritoneal dialysis, imprisonment
and previous TB are associated with higher risk [1] Gautam MP, Karki P, Rijal S, Singh R. Potts disease and
of TS. We believe that these mentioned variables paraplegia. JNMA J Nepal Med Assoc 2005;44(159):10615.
[2] Jain AK, Dhammi IK. Tuberculosis of the spine: a review.
may be the main risk factors of TS. In published
Clin Orthop Relate Res 2007;460:3949.
studies, aging, intestinal TB, alcoholism, Diabetes [3] Turgut M. Spinal tuberculosis (Potts disease): its clinical
mellitus, corticosteroid and previous TB have been presentation, surgical management, and outcome. A survey
considered as spinal TB risk factors [4,13]. Con- study on 694 patients. Neurosurg Rev 2001;24(1):813.
sidering the history of imprisonment as risk factor [4] Fitzgerald DW, Sterling TR, Haas DW. Mycobacterium tuber-
culosis. In: Mandell GL, Bennett JE, Dolin R, editors.
may be doubtful. Stay in prison puts patients at
Principle and practice of infectious diseases. 7th ed.
a higher risk of TB infection, but progress of the Philadelphia: Churchill Livingstone; 2010. p. 312963.
infection to spinal involvement is not clear to [5] Hosalkar HS, Agrawal N, Reddy S, Sehgal K, Fox EJ, Hill RA.
us. We think further investigation is needed to Skeletal tuberculosis in children in the Western world: 18
conclude the association between TS and impris- new cases with a review of the literature. J Child Orthop
2009;3(August (4)):31924.
onment.
[6] Alothman A, Memish ZA, Awada A, Al-Mahmood S, Al-Sadoon
S, Rahman MM, et al. Tuberculous spondylitis: analysis
of 69 cases from Saudi Arabia. Spine 2001;26(December
Conclusion (24)):56570.
[7] Rodriguez-Gomez M, Willisch A, Fernandez-Dominguez L,
Lopez-Barros G, Garca-Porra C, Gonzalez-Gay MA. Tuber-
Spinal tuberculosis is therefore a diagnosis that culous spondylitis: epidemiologic and clinical study in
attending doctors need to consider early. Patients non-HIV patients from northwest Spain. Clin Exp Rheumatol
from countries with a high prevalence of tuber- 2002;20(MayJune (3)):32733.
culosis, presenting with back pain in association [8] Sinan T, Al-Khawari H, Ismail M, Ben-Nakhi A, Sheikh M.
Spinal tuberculosis: CT and MRI feature. Ann Saudi Med
with constitutional symptoms and a high ery-
2004;24(NovemberDecember (6)):43741.
throcyte sedimentation rate and/or positive CRP [9] Godlwana L, Gounden P, Ngubo P, Nsibande T, Nyawo K,
should be investigated to exclude tuberculo- Puckree T. Incidence and prole of spinal tuberculosis in
sis. In the context of a normal spinal x ray patients at the only public hospital admitting such patients
and a high clinical suspicion, further imaging in KwaZulu-Natal. Spinal Cord 2008;46(May (5)):3724.
[10] Drohniewski FA, Caws M, Gibson A, Young D. Modern
such as MRI using contrast medium should be
laboratory diagnosis of tuberculosis. Lancet infect Dis
warranted. 2003;3:1417.
[11] Khorvash F, Javadi AA, Izadi M, Jonaidi Jafari N, Ranjbar R.
Spinal tuberculosis: a major public health hazard in Isfahan.
Pak J Biol Sci 2007;10(October (19)):34004.
Study limitations [12] Diwakar L, Logan S, Ghaffar N, Hare A, Lynn W, Ash S.
Low back pain: think of tuberculosis. BMJ 2006;333(July
As this was a retrospective study, some of the infor- (7560)):201.
mation required was incomplete. Sample size of 69 [13] Mulleman D, Mammou S, Griffoul I, Avimadje A, Goupille P,
patients was small for calculating the appropriate Valat JP. Characteristics of patients with spinal tuberculosis
in a French teaching hospital. Joint Bone Spine 2006;73(July
rates or comparing the variables. Future prospec-
(4)):4247.
tive study in larger scale is recommended. Case [14] Solagberu BA, Ayorinde RO. Tuberculosis of the spine in
denition of TS was dependent on the diagnosis reg- Ilorin, Nigeria. East Afr Med J 2001;78(April (4)):1979.
istered in the medical le. Better result should be [15] Garca-Lechuz JM, Julve R, Alcal L, Ruiz-Serrano MJ,
found by culture conrmation. Munoz P. Tuberculous spondylodiskitis (Potts disease):
experience in a general hospital. Enferm Infect Microbiol
Clin 2002;20(January (1)):59.
[16] Fennira H, Bourguiba M, Ben Rejeb N, Ben Miled K, Cher-
Conict of interest miti F, Maalej S, et al. Vertebral tuberculosis revealed by
thoracic manifestations. A study of ve cases. Tunis Med
2006;84(December (12)):8115.
There is no conict of interest in this study.
[17] Akhmetov AA, Amanzholova LK, Tutkyshbaev SO. Tubercu-
Funding: This work is funded by Jundishapur lous spondylitis complicated by spinal disorders: diagnosis
Infectious Diseases Research Center. and treatment. Probl Tuberk 2000;(3):357.
Ethical approval: This work is supported by eth- [18] Tuli SM. Tuberculosis of the spine: a historical review. Clin
ical committee of JIDRC. Orthop Relat Res 2007;460(July):2938.

Available online at www.sciencedirect.com