Predictors of Stroke and Its Significance in the Outcome of

Tuberculous Meningitis

Jayantee Kalita, DM, Usha Kant Misra, DM, and Pradeep Pankajakshan Nair, DM

Background and Aim: We sought to study the frequency and predictors of stroke in
tuberculous meningitis (TBM) and its prognostic significance. Design: This was an
observational study in a tertiary care teaching hospital. Methods: In all, 122 patients
with TBM aged 4 to 82 years diagnosed on the basis of clinical, cerebrospinal fluid,
and magnetic resonance imaging criteria were prospectively evaluated. Severity of
meningitis was graded into stage I to III. Magnetic resonance imaging was done at
admission and 3 months after treatment. Outcome was defined at 3 and 6 months as
complete, partial, or poor. Predictors of stroke and its significance in long- and short-
term outcome were evaluated. Findings: A total of 55 patients had stroke; 42 at
admission and 13 developed within 3 months of 4 drug antitubercular treatment.
Strokes were ischemic in 54 (hemorrhagic transformation in 7) and hemorrhagic
in one. Basal ganglia infarctions were present in 30, thalamic in 9, brainstem in 10,
cortical in 27, and cerebellar in 4 patients. Stoke was multiple in 29 patients. In all,
38 patients had infarctions in anterior circulation, 7 in posterior, and 10 in both.
Stroke was significantly related to stage of meningitis, hydrocephalus, exudate,
and hypertension. No difference was found in clinical or laboratory parameters in
early and late strokes. At 6 months, 28 patients died. At 3 months there were 21
patients lost to follow up and at 6 months there were 30 patients lost to followup.
Outcome is based on the rest of the patients, ie. 101 patients at 3 months and 92
patients at 6 months. Conclusion: Stroke occurs in 45% of patients with TBM
both in early and later stage, mostly in basal ganglia region, and predicts poor
outcome at 3 months. Key Words: Tuberculous meningitis—vasculitis—magnetic
resonance imaging—computed tomography—stroke—prognosis—outcome.
! 2009 by National Stroke Association

Tuberculous meningitis (TBM) is the commonest cause
of subacute or chronic meningitis in the developing world
and is estimated to constitute about 10% of cases of tuber-
culosis (TB).1 It is the most devastating manifestation of
Mycobacterium TB. With the development of acquired
immunodeficiency syndrome, it has become important
even in those areas where it was hitherto not important.
From the Department of Neurology, Sanjay Gandhi Post Graduate
Institute of Medical Sciences, Lucknow, India.
Received September 6, 2008; revision received October 29, 2008;
accepted November 3, 2008.
Address correspondence to Jayantee Kalita, DM, Department of
Neurology, Sanjay Gandhi Post Graduate Institute of Medical Sci-
ences, Lucknow 226014, India. E-mail: jayanteek@yahoo.com.
1052-3057/$—see front matter
! 2009 by National Stroke Association
doi:10.1016/j.jstrokecerebrovasdis.2008.11.007
TBM is associated with basal exudates, tuberculoma,
hydrocephalus, and vasculitis, which determine the
clinical picture and outcome. Small and medium vessels
traversing the basal exudates may show vasculitis, which
results in infarctions.2 The incidence of vasculitis in TBM
was variably reported. In autopsy study on TBM, vascu-
litis was reported in 41%.3 On angiographic studies in
TBM, irregular beaded narrowing in supraclinoid portion
of internal carotid artery, widely sweeping pericallosal
artery, and thalamostriate vein and delayed circulation
of middle cerebral vein with scanty collateral circulation
were reported.4 Computed tomography (CT) scan studies
show infarctions in 17% to 63% of patients.5,6 In TBM, the
basal ganglionic infarcts are the commonest because of
involvement of perforating vessels. In a CT scan study,
the infarctions were located in the TB zone in 75% of
patients, which comprised of head of caudate,

Journal of Stroke and Cerebrovascular Diseases, Vol. 18, No. 4 (July-August), 2009: pp 251-258 251
252
anteromedial thalami, anterior limb and genu of internal
capsule due to involvement of medial striate, thalamotub-
eral, and thalamoperforate arteries.7 On magnetic reso-
nance imaging (MRI), infarctions were reported in 10 of
20 patients with TBM and confirmed the reported distri-
bution of vascular involvement.8 According to some
authors, TBM distinguishes itself by paucity of acute
stroke and focal signs in TBM develop insidiously. Tran-
sient ischemic attack or reversible ischemic neurologic
deficit and stuttering course of stroke are not seen.9 It is
rare to have aphasia, agnosia, and vertebrobasilar terri-
tory infarction in TBM10; this view, however, was contra-
dicted in recent studies and TBM is included as a cause of
stroke in young people.8,10,11 Infarct has been reported as
a poor prognostic predictor in children with TBM.12 There
is no comprehensive study in adults evaluating the
predictors of infarct and its role in short- and long-term
outcome of patients with TBM. In this study, we report
the frequency and predictors of stroke and its significance
in long- and short-term outcome of TBM.
Subject and Methods
Inclusion Criteria
Patients with TBM diagnosed on the basis of clinical, ce-
rebrospinal fluid (CSF), and radiologic criteria were
included. Investigations done in this study are routine in
the management of TBM and are approved by our
institutional hospital information service. The diagnosis
of TBM was based on clinical, CTor MRI scan, and CSF cri-
teria. The essential criteria included presence of meningitic
symptoms comprising fever, headache, and vomiting for 2
weeks or more where malaria and septic and fungal men-
ingitides were excluded. The supportive criteria included:
(1) CSF cells 0.2 3 109/L or more with predominant lym-
phocytes, protein more than 1 g/L, and sterile bacterial
and fungal culture; (2) CT or MRI scan evidences of
exudates, infarctions, hydrocephalus, and tuberculoma in
isolation or in various combinations; (3) evidence of extra-
central nervous system (CNS) TB; and (4) response to
antitubercular therapy. Presence of essential and 3 of 4 sup-
portive criteria was considered highly probable and two
supportive criteria as probable TBM. Presence of Acid
Fast Bacillus (AFB) in CSF smear or culture, positive poly-
merase chain reaction for Mycobacterium TB, or IgM
enzyme-linked immunosorbent assay (ELISA) in CSF was
considered definitive evidence of TBM.13
Evaluation
Detailed clinical examination was performed. Con-
sciousness was assessed by Glasgow Coma Scale. Focal
neurologic deficit (monoplegia, hemiplegia, quadriplegia,
or paraplegia) and cranial nerve palsy were noted. Mus-
cle power was graded as normal, partial, and complete
weakness. Muscle tone and reflexes were recorded as
J. KALITA ET AL.
increased, normal, or decreased. Evidence of TB outside
the CNS such as lung, lymph node, bone, and joint was
noted.
Severity of stroke was defined by Canadian Neurolog-
ical Scale.14 The severity of meningitis was graded into:
stage I, meningitis only; stage II, meningitis with focal
signs without alteration in sensorium; and stage III,
meningitis with altered sensorium.15
Investigations
The investigations included blood counts, serum chem-
istry, erythrocyte sedimentation rate, human immunode-
ficiency virus serology, and chest radiograph. CSF was
examined for opening pressure, protein, cells, glucose,
bacteria, fungi, and AFB and culture by BACTEC. CSF
was also examined for polymerase chain reaction and
IgM enzyme-linked immunosorbent assay for Mycobacter-
ium TB. Cranial MRI was done using 1.5-T system (Signa,
GE Medical Systems, Milwaukee, Wis). T1, T2, fluid-
attenuated inversion recovery, diffusion-weighted imaging,
and T1 contrast sequences were obtained. Abnormalities
such as exudates, tuberculoma (number, location, and
size), infarctions, and hemorrhage (location and size)
were noted. Presence of hydrocephalus and periventricu-
lar signal changes were recorded.
The patients with TBM were categorized into with stroke
and without. The diagnosis of stroke was based on clinical
and/or MRI evidences of acute stroke on diffusion-
weighted imaging. The occurrence of acute stroke on
repeated MRI was also noted.
Treatment and Follow-up
Patients were treated with a 4-drug antitubercular daily
regimen (rifampicin [10 mg/kg], isoniazid [5 mg/kg],
pyrazinamide [25 mg/kg], and ethambutol [15 mg/kg]).
All 4 drugs were continued for 8 months followed by 3
drugs (isoniazid, rifampicin, and ethambutol) for 12
months after which two drugs (isoniazid and ethambutol)
were continued for a total duration of 18 months. Prednis-
olone (0.5-1 mg/kg) was prescribed only to the patients
with encephalopathy, increased intracranial pressure,
and impending visual failure for a period of 1 month
followed by taper in the next month. Ventriculoperitoneal
shunt was carried out if there was increased intracranial
pressure with features of herniation as a result of obstruc-
tive hydrocephalous. In patients with communicating
hydrocephalous with features of increased intracranial
pressure, repeated CSF drainage by lumbar puncture
was tried before subjecting them to shunt surgery.
Outcome was assessed at the end of 3 and 6 months of
treatment on the basis of Barthel index (BI) score as
complete (BI 5 20), partial (BI 12-19), or poor (BI , 12).
Statistical Analysis
The categorical variables of clinical, laboratory, and
radiologic features of the patients with and without
PREDICTORS OF STROKE AND ITS SIGNIFICANCE IN TBM
evidence of stroke were compared by Chi-square and
Student’s t tests and continuous variables by independent
t test or Mann-Whitney U test using software (SPSS,
Version 12, SPSS, Chicago, IL). Multiple comparisons
were done if needed. For analysis of the predictors of
stroke the variables having significance of .1 on univariate
were taken for binary logistic regression analysis for de-
riving the best set of predictors of stroke in TBM.
Results
There were 122 patients with TBM whose median age
was 32 (range 4-82) years; 7 were children (,12 years)
and 41 were older than 40 years. In all, 58 were female.
The median duration of illness on admission to our insti-
tute was 8 months. Before admission to our institute, 76
patients were receiving antitubercular treatment for a me-
dian duration of 30 (2-120) days. A total of 55 patients had
stroke during the course of illness and 67 did not. Sixteen
patients developed infarct within the first month of illness
and 39 later. On admission to our institute, 42 patients had
stroke and another 13 developed it during follow-up. Of
the 42 patients with infarcts on admission, 9 developed
additional infarcts during our follow-up. The severity of
meningitis was stage I in 40, stage II in 68, and stage III
in 14 patients. In our study, 27 patients had definite, 53
highly probable, and 42 probable TBM. The evidence of
extra-CNS TB was present in 32 patients: pulmonary in
27 (miliary in 12), bone and joint in 3, and lymph node
in two. Associated diabetes mellitus was present in 11
patients and two were younger than 40 years. Seven of
them had stroke; 3 had clinical stroke confirmed on MRI
and in 4 patients infarcts were detected on MRI only. Six
patients had hypertension, all were older than 40 years
and had stroke; two patients had clinical stroke confirmed
on MRI and in 4 patients acute infarcts were found on MRI.
Description of Stroke
Stroke was ischemic in 54 and hemorrhagic in one
patient. Seven patients had hemorrhagic transformation
Figure 1. Cranial MRI of patient with stage III
Tuberculosis meningitis (TBM) shows cortical and
subcortical infarcts in Diffusion weighted Imaging
(DW1) (A) and T2 images (B).
253
of ischemic stroke. History of acute neurologic deficit
suggesting clinical diagnosis of stroke was present in 23
and in the remaining 32, neurologic deficit either was
gradual or not detected but MRI revealed evidence of
acute stroke.
The majority of patients had basal ganglia infarcts with
or without other areas of involvement. The location of
infarction was cortical in 27, subcortical white matter in
12, basal ganglia in 30, thalamus in 9, brainstem in 10,
and cerebellar in 4 patients. According to vascular terri-
tory, infarcts were located in anterior circulation in 38,
posterior circulation in 7, and both anterior and posterior
circulations in 10 patients. Single infarct was present in 26
and multiple in 29 patients (Figs 1 to 3).
Predictors of Stroke
Hypertension (P 5 .007), stage of meningitis (P 5 .001),
and presence of hydrocephalus (P 5 .002) and exudate
(P 5 .007) were significantly related to stroke in TBM.
Other variables such as age (P 5 .70), sex (P 5 .43), level
of consciousness (P 5 .06), diabetes (P 5 .22), human im-
munodeficiency virus status (P 5 1.00), CSF findings, and
presence of tuberculoma were not significantly related to
occurrence of stroke. Details are presented in Table 1. On
regression analysis none of these variables was, however,
related to stroke.
On comparing stroke within 1 month of illness (16)
with those developed later (39), no significant difference
was revealed in age (P 5 .24), sex (P 5 .765), stage of men-
ingitis (P 5 1.00), Glasgow Coma Scale score (P 5 .42),
hydrocephalus (P 5 .51), tuberculoma (P 5 .49), exudates
(P 5 .49), diabetes (P 5 .41), or hypertension (P 5 1.00).
Details are summarized in Table 2.
Outcome
In our study 25 patients died at 3 months and 3 more
patients died between 3 to 6 months of follow-up; two
underwent shunt surgery. At 3 months, 37 patients had
complete, 24 partial, and 15 poor recovery. At 6 months,
254 J. KALITA ET AL.

Figure 2. MRI of patient with stage II TBM

shows hydrocephalous, exudates, and granuloma
T1 contrast (T1C) and infarctions in posterior cir-
culation Diffusion Weighted Imaging (DW1).
FLAIR, Fluid-attenuated inversion recovery.

41 patients had complete, 16 partial, and 7 poor recovery. Patients with cortical location of stroke were associated
Presence of stroke was associated with poor outcome at 3 with higher frequency of deaths (P 5.02). Details are sum-
months (P 5 .002) but not at 6 months (P 5 .06) (Fig 4). marized in Table 3.

Figure 3. MRI of patient with stage I TBM shows hydrocephalous and hemorrhagic infarction in caudate (both T1 and T2 hyperintense).
PREDICTORS OF STROKE AND ITS SIGNIFICANCE IN TBM 255

Table 1. Comparison of patients with tuberculous meningitis with and without stroke

Variable Stroke (N 5 55) No stroke (N 5 67) P value

Age (y)
,12 4 3 .70
.12 51 64
Sex
Male 31 33 .43
Female 24 34
Diabetes mellitus
Present 7 4 .22
Absent 48 63
Hypertension
Present 6 0 .007
Absent 49 67
GCS score
3-7 9 5 .24
8-10 8 8
11-15 38 54
Stage of TBM
I 9 31 .001
II 37 31
III 9 5
HIV
Present 2 3 1.00
Absent 53 64
Radiology
Hydrocephalus 30 17 .001
Exudates 21 11 .008
Tuberculoma 28 32 .729
Duration (wk) 12.15 6 16.61 11.55 6 12.05 .82
CSF
Cell/mL 195.54 6 209.67 192.20 6 245.38 .93
Protein mg% 119.24 6 89.07 117.1 6 88.15 .90
Sugar mg% 52.13 6 39.35 48.0 6 31.50 .53
Recovery
Complete 8 29 .002
Partial 12 12
Poor/death 24 16

Abbreviations: CSF, cerebrospinal fluid; GCS, Glasgow Coma Scale; HIV, human immunodeficiency virus; TBM, tuberculous meningitis.

Discussion weakness. It is possible that in the presence of more strik-

In this study, 45% of patients with TBM had stroke, and
stroke was associated with poor outcome at 3 months but
not at 6 months. The frequency of stroke in TBM was
reported in 28% to 50%.8,13,16-18 This wide variation in
the frequency of infarction in TBM may be caused by dif-
ferences in patient population and methodology. In our
study, the diagnosis of stroke was based on MRI findings
as the focal weakness may have insidious course in the
background of protean manifestations of TBM. Transient
ischemic attack, reversible ischemic neurologic deficit,
and stuttering progression of cerebrovascular accidents
are reported to be rare in TBM.9 In our study, however,
23 patients reported acute hemiplegia or hemiparesis
(,24 hours); whereas 32 had slower or unreported focal
ing feature of altered sensorium, headache, and vomiting,
the mild or even significant weakness may escape the
attention of patients or relatives. In TBM, the small vessels
traversing the basal exudates may develop panarteritis
whereas in larger vessels, periarteritis. It has been
observed in TBM that the vascular involvement starts in
adventitia and progressively encroaches to involve the
entire vessel wall constituting panarteritis tuberculosa.19
In addition to vasculitis and thrombosis, vasospasm has
also been reported in TBM. In late stage, organization of
basal exudates may strangulate the vessels leading to vas-
cular narrowing and focal weakness. TBM usually results
in ischemic stroke but rarely it may result in hemorrhagic
infarcts that are attributed to both arterial and venous
thrombosis.8,20,21 Vasculitis may result in intimal fibrinoid
256 J. KALITA ET AL.

Table 2. Relationship of demographic and clinical variables patients, MR angiography was not done as its limitation
in patients with tuberculous meningitis with stroke within 1 in detecting small vessel pathology is well known.
month of illness and later Basal ganglia infarctions are the commonest in TBM
and are attributed to the involvement of lenticulostriate
Early stroke Late stroke vessels. The majority of our patients with basal ganglia
Variable (n 5 16) (n 5 39) P value infarct had anterior involvement. Among those 4 patients
Age (y) with posterior involvement, 3 had diabetes mellitus,
,40 11 18 .127 which is a major risk factor for ischemic stroke. This
.40 5 21 observation is consistent with a previous report that
Sex TBM preferentially involves anteromedial basal ganglia.7
Male 10 21 .557 Terminal portions of internal carotid and proximal
Female 6 18 middle cerebral artery are also commonly involved in
Diabetes mellitus TBM. In angiographic studies, irregular narrowing is
Present 3 4 .66 commonly reported. Curiously enough, the vertebrobasi-
Absent 13 35 lar system, which is equally bathed in exudates, usually
Hypertension
escapes. Arteries traversing the sulci of the vertex are
Present 2 4 1.00
also unaffected.9 We found basal ganglia infarct in 30
Absent 14 35
GCS score and thalamic in 9 patients. The strokes were in anterior
3-7 4 5 .657 circulation in 38, posterior in 7, and both in 10 patients.
8-10 2 6 Moreover, 10 patients had brainstem and 4 cerebellar
11-15 10 28 infarctions suggesting that all the vessels can be affected
Stage of TBM to a variable frequency. Similar observations were
I 2 7 .917 reported in MR angiography8 and CT scan study.16
II 11 26 In our study, stroke was significantly related to stage of
III 3 6 meningitis, hypertension, hydrocephalus, and exudate on
HIV univariate analysis but none was found to be related on
Present 0 2 .397
binary regression analysis. This suggests that stroke may
Absent 16 35
not be related to a single independent factor; rather, there
Steroid
Received 5 14 .742 may be many factors that simultaneously play a role in its
Not received 11 25 occurrence. A similar observation was reported in a recent
Radiology study on 38 patients with TBM where none of the clinical
Hydrocephalus 10 20 .448 and laboratory variables were related to infarct.22
Exudates 5 16 .89 The frequency of stroke and hydrocephalous has been
Tuberculoma 7 21 .496 shown to reduce after corticosteroid therapy23 although
in experimental studies inflammatory markers were not
Abbreviations: GCS, Glasgow Coma Scale; HIV, human immu-
affected by corticosteroid.24 In our study corticosteroid
nodeficiency virus; TBM, tuberculous meningitis.
did not influence the occurrence of infarct. This may be
because of lower dose of corticosteroid used in our study
degeneration leading to rupture of vessel walls and or vascular involvement in TBM may be a mechanical
hemorrhagic lesions.3 In our study, 7 patients had hemor- rather than immunologic process. The role of corticoste-
rhagic infarctions and one intracerebral hemorrhage. All roid, however, can not be commented on much from
were normotensive and their MR angiography and this study as this study was not designed as a randomized
venography findings were normal. In the remaining control trial to evaluate its role.

Figure 4. Percentage of patients with

complete, partial, and poor (including
death) recovery at 3 and 6 months with
and without strokes.
PREDICTORS OF STROKE AND ITS SIGNIFICANCE IN TBM 257

Table 3. Significance of infarcts in different locations in 6-month outcome of tuberculous meningitis

Location of infarct Complete recovery, N 5 41 Partial recovery, N 5 16 Poor recovery, N 5 7 Dead, N 5 28 P value

Cortical Present 5 (21.7%) 8 (39.1%) 1 (4.3%) 8 (34.7%) .02

Absent 36 (51.4%) 8 (11.4%) 6 ((8.6%) 20 (28.6%)
Subcortical Present 3 (33.3%) 1 (11.1%) 1 (11.1%) 4 (44.4%) .73
Absent 38 (45.8%) 15 (18.1%) 6 (7.2%) 24 (28.9%)
Basal ganglia Ul 4 (36.4%) 3 (27.3%) 2 (18.2%) 2 (18.2%) .565
Bl 4 (36.4%) 1 (9.1%) 1 (9.1%) 5 (45.4%)
None 33 (47.1%) 12 (17.1%) 4 (5.7%) 21 (28.6%)
Brainstem Ul 2 (33.3%) 0 (0%) 1 (16.7%) 3 (50%) .52
Bl 1 (33.3%) 0 (0%) 0 (0%) 2 (66.7%)
None 38 (45.8%) 16 (19.3%) 6 (7.2%) 23 (27.7%)
Thalamus Ul 3 (50%) 0 (0%) 0 (0%) 3 (50%) .62
Bl 1 (100%) 0 (0%) 0 (0%) 0 (0%)
None 37 (43.5%) 16 (18.8%) 7 (8.3%) 25 (29.4%)
Arterial territory Anterior 10 (35.7%) 9 (32.1%) 3 (10.7%) 6 (21.4%) .05
Posterior 0 (0%) 0 (0%) 0 (0%) 4 (100%)
Both 3 (37.5%) 1 (12.5%) 1 (12.5%) 3 (37.5%)
None 37 (60.6%) 6 (9.8%) 3 (4.9%) 15 (24.6%)
No. of infarcts Single 5 (27.8%) 8 (44.4%) 1 (5.5%) 4 (22.2%) .42
Multiple 8 (32%) 4 (16%) 3 (12%) 10 (40%)
None 28 (57.1%) 4 (8.2%) 3 (6.1%) 14 (28.6%)

Abbreviations: Bl, bilateral; Ul, unilateral.

Presence of stroke in TBM has variably related to poor
outcome.12,13,25,26 In children with TBM, stroke at any
location except hemispheric involvement was related to
poor outcome at 6 months.12 In the study by Leiguarda
et al17 none of the patients recovered completely. In an
outcome study, stroke predicted poor outcome at 3
months on univariate analysis but on multivariate analy-
sis it did not achieve statistical significance.13 Similarly
neurologic sequelae at 1 year was also not related to
infarction but was related to focal motor deficit at admis-
sion.26 In the current study, stroke on MRI, although
related to poor prognosis at 3 months, did not achieve
statistical significance for 6-month outcome. Infarctions
in TBM usually are small, multiple, and located in the
basal ganglia or deep white matter reflecting the outcome
of lacunar stroke, which is usually good. Moreover,
outcome of TBM is dependent on multiple factors such
as duration of illness, severity of meningitis, focal deficit,
hydrocephalous, tuberculoma, and exudates.
Our study reveals that stroke in TBM occurs in 45% of
patients and can have a wide spectrum, with respect to
onset, course, and topography. Stroke predicts poor
outcome at 3 months but 6-month outcome is indepen-
dent of stroke.
References
1. Dye C, Scheele S, Dolin P, et al. Consensus statement.
Global burden of tuberculosis: Estimated incidence,
prevalence and mortality by country. World Health
Organization surveillance and monitoring project.
JAMA 1999;282:677-686.
2. Dastur DK, Dave UP. Further observations on the fine
structure of blood vessels in neurotuberculosis. In:
Cervos-Navarro J, ed. Advances in neurology. New
York: Raven Press, 1978:577-589.
3. Dastur DK, Lalitha VS, Udani PM, et al. The brain and
meninges in tuberculous meningitis–gross pathology in
100 cases and pathogenesis. Neurol India 1970;8:86-100.
4. Lehrer H. The angiographic triad in tuberculous menin-
gitis. Radiology 1966;87:829-835.
5. Bullock MR, Welchman JM. Diagnostic and prognostic
features of tuberculous meningitis on CT scanning.
J Neurol Neurosurg Psychiatry 1982;45:1098-1101.
6. Kingsley DP, Hendrickse WA, Kendall BE, et al. Tubercu-
lous meningitis: Role of CT in management and progno-
sis. J Neurol Neurosurg Psychiatry 1987;50:30-36.
7. Hsieh FY, Chia LG, Shen WC. Locations of cerebral infarc-
tions in tuberculous meningitis. Neuroradiology 1992;
34:197-199.
8. Gupta RK, Gupta S, Singh D, et al. MR imaging and angi-
ography in tuberculous meningitis. Neuroradiology
1994;36:87-92.
9. Dalal PM. Observations on the involvement of cerebral ves-
sels in tuberculous meningitis in adults. In: Goldstein M,
ed. Advances in neurology. Vol 25. New York: Raven Press,
1979a:149-159.
10. Dalal PM. Stroke in the young in west central India. In:
Goldstein M, ed. Advances in neurology. Vol 25. New
York: Raven Press, 1979b:339-348.
11. Thakur SK, Gopinathan VP, Bhalla IP, et al. Hemiplegia in
the young. J Assoc Physicians India 1987;35:147-148.
12. Andronikou S, Wilmshurst J, Hatherill M, et al. Distribu-
tion of brain infarction in children with tuberculous
meningitis and correlation with outcome score at 6
months. Pediatr Radiol 2006;1289-1294.
13. Misra UK, Kalita J, Srivastava M, et al. Prognosis of tuber-
culous meningitis: A multivariate analysis. J Neurol Sci
1996;137:57-61.
258
14. Cote R, Hachinski VC, Shurvell BL, et al. The Canadian
neurological scale: A preliminary study in acute stroke.
Stroke 1986;17:731-737.
15. British Medical Research Council. Streptomycin in tuber-
culosis trials committee: Streptomycin treatment of tuber-
culous meningitis. Lancet 1948;1:582–596.
16. Bhargava S, Gupta AK, Tandon PN. Tuberculous menin-
gitis–a CT study. Br J Radiol 1982;55:189-196.
17. Leiguarda R, Berthier M, Starkstein S, et al. Ischemic in-
farction in 25 children with tuberculous meningitis.
Stroke 1988;19:200-204.
18. Lamprecht D, Schoeman J, Donald P, et al. Ventriculoper-
itoneal shunting in childhood tuberculous meningitis. Br
J Neurosurg 2001;15:119-125.
19. Rich AR, McCordock HA. Pathogenesis of tubercular
meningitis. Bull John Hopkins Hosp 1933;52:5-13.
20. Modilevsky T, Arlazoroff A, Goldenberg E, et al. Sponta-
neous subarachnoid bleeding and stroke syndrome in tu-
berculous meningoencephalitis. Harefuah 1979;97:68-70.
21. Yeh ST, Lee WJ, Lin HJ, et al. Nonaneurysmal
subarachnoid hemorrhage secondary to tuberculous
J. KALITA ET AL.
meningitis: Report of two cases. J Emerg Med 2003;
25:265-270.
22. Koh SB, Kim BJ, Park MH, et al. Clinical and laboratory
characteristics of cerebral infarction in tuberculous men-
ingitis: A comparative study. J Clin Neurosci 2007;
14:1073-1077.
23. Thwaites GE, Macmullen-Price J, Tran TH, et al.
Serial MRI to determine the effect of dexamethasone
on the cerebral pathology of tuberculous meningitis:
An observational study. Lancet Neurol 2007;6:230-236.
24. Simmons CP, Thwaites GE, Quyen NT, et al. The clinical
benefit of adjunctive dexamethasone in tuberculous men-
ingitis is not associated with measurable attenuation of
peripheral or local immune responses. J Immunol 2005;
175:579-590.
25. Chan KH, Cheung RT, Lee R, et al. Cerebral infarcts com-
plicating tuberculous meningitis. Cerebrovasc Dis 2005;
19:391-395.
26. Kalita J, Misra UK, Ranjan P. Predictors of long-term neu-
rological sequelae of tuberculous meningitis: A multivar-
iate analysis. Eur J Neurol 2007;14:33-37.