Predictors of stroke in patients of tuberculous meningitisand its effect on the outcome

H.K. ANURADHA, R.K. GARG, A. AGARWAL, M.K. SINHA, R. VERMA,M.K. SINGH and R. SHUKLA

From the Department of Neurology, Chhatrapati Shahuji Maharaj Medical University,

Lucknow 226 003, UP, India

Address correspondence to R.K. Garg, Department of Neurology, Chhatrapati Shahuji Maharaj MedicalUniversity, Lucknow 226 003, UP, India email: garg50@yahoo.com

Received 8 November 2009 and in revised form 15 April 2010

Summary

Background: Stroke is a devastating complication oftuberculous meningitis and is an important deter-minant of its outcome.Aim: To prospectively evaluate the predictive factorsfor stroke in patients with tuberculous meningitisand to assess the impact of stroke on the overallprognosis and outcome.Methods: We evaluated and followed 100 patientsof tuberculous meningitis for 6 months. Magneticresonance imaging was performed at inclusion andafter 6 months. We evaluated the predictors ofstroke and also assessed the effect of stroke on theoutcome. Outcome was defined with the help ofmodified Rankin scale.Results: Of the 100 patients, 6 lost to follow-up.Thirty patients had stroke, 27 of them had stroke atinclusion. Three patients developed stroke duringfollow-up. In most of the patients, stroke was amanifestation of advanced stages of tuberculousmeningitis. Internal capsule/basal ganglia were the

most frequently involved sites. Infarcts commonlyinvolved the middle cerebral arterial territory.On univariate analysis, predictors of stroke wereaged >25 years (P25 years was found asignificant predictor of stroke. Strokes in patientswith tuberculous meningitis were associated withpoor prognosis.Conclusions: Stroke occurred in 30% of caseswith tuberculous meningitis. Advanced stage oftuberculous meningitis, basal exudates, optochias-matic arachnoiditis and vision impairment weresignificant predictors of stroke. Stroke independentlypredicted the poor outcome of tuberculousmeningitis.

Introduction

Stroke is a common complication of tuberculous

meningitis. Several studies in the past have observed

that 20% of the patients with tuberculous menin-gitis develop a stroke in the course of the illness.1,2

Infarcts are one of the characteristic imaging

abnormalities of tuberculous meningitis. Infracts

can either be asymptomatic or symptomatic.

Symptomatic strokes in tuberculous meningitis

often present with dense hemiplegia. Tuberculous

meningitis, in patients with infarcts, is reported to

be fatal up to three times more often than in those

without infarcts. In survivors, the extent of ischemic

cerebral damage is an important determinant of dis-

ability.35 Most of the past observations were based

on retrospective analysis. Hence, we designed

the present study to prospectively evaluate the

! The Author 2010. Published by Oxford University Press on behalf of the Association of Physicians.All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Q J Med 2010; 103:671678doi:10.1093/qjmed/hcq103 Advance Access Publication 29 June 2010

by guest on May 31, 2015

Dow

nloaded from

predictive factors for stroke in patients with tubercu-

lous meningitis. We also assessed the impact of

stroke on the overall prognosis and outcome.

Material and methods

Patients (>14 years of age) with a clinical syndromeof meningitis attending outpatient and inpatient

facilities of Chhatrapati Shahuji Maharaj Medical

University, Lucknow, Uttar Pradesh, India were

included in the study. The patients were enrolled

between March 2008 and March 2009.

Institutional Ethics Committee of our University

approved the study protocol. Written informed con-

sent to participate in the study was obtained from all

patients or their legal guardians.

Diagnostic criteria of tuberculousmeningitis

The diagnosis of tuberculous meningitis was based

on clinical, cerebrospinal fluid (CSF) and radiologic-

al criteria. The essential criteria were the presence of

meningitic syndrome comprising headache, vomit-

ing and fever for 2 weeks, along with characteristicCSF abnormalities (dominant lymphocytic pleocyto-

sis, with raised protein). Other criteria were pres-

ence of exudates, infarcts, hydrocephalus and

tuberculoma singly or in various combinations on

neuroimaging, evidence of tuberculosis at extra-

central nervous system sites and response to antitu-

berculous therapy.Tuberculous meningitis was considered definite

if acid-fast bacilli were demonstrated in the CSF. It

was considered probable in patients with one or

more of the following: suspected active pulmonary

tuberculosis on chest radiography, acid-fast bacilli

found in any specimen other than the CSF and clin-

ical evidence of other extra-pulmonary tuberculosis.

Tuberculous meningitis was considered possible in

patients with at least four of the following: a history

of tuberculosis, predominance of lymphocytes in the

CSF, duration of illness of >5 days, a ratio of CSFglucose to plasma glucose of

and T1-weighted contrast sequences were ob-tained at the time of inclusion and at the end of6 months.Cerebral infarctions were categorized according

to site, number and size. The locations, where aninfarct was looked for, included basal ganglion, thal-amus, internal capsule, cerebellum, brainstem andterritory of main cerebral arteries (posterior, anteriorand middle cerebral artery). Lacunar infarct wasdefined as a cerebral infarct (by the above criteria)with a maximum diameter of 30% and the size of one or bothtemporal horns >2mm. Communicating hydro-cephalus was defined as enlargement of the ven-tricles, without evidence of an obstructing lesionalong the intraventricular CSF pathways down tothe level of first cervical spinal segment, includingthe fourth ventricular outflow tracts and the cerebralaqueduct. Tuberculomas were defined as discrete orcoalescing lesions which were hypo- or isointenseon T1-weighted images and hypointense onT2-weighted images showing homogeneous nodularcontrast enhancement (non-caseating tuberculoma);or hypoisointense in T1-weighted images and isohypointense on T2-weighted images showing ringenhancement (caseating tuberculoma).

Treatment

After evaluation, antituberculous treatment was im-mediately started. Antituberculous regimen included2 months intensive therapy comprising isoniazid(5mg/kg of body weight; maximum, 300mg), rifam-picin (10mg/kg; maximum, 600mg), pyrazinamide(25mg/kg; maximum, 2 g/day) and ethambutol(20mg/kg; maximum, 1200mg), followed bya continuation phase comprising of isoniazid

and rifampicin for 6 months.11 All patients

received intravenous dexamethasone for 4 weeks

(0.4mg/kg/day for Week 1, 0.3mg/kg/day for

Week 2, 0.2mg/kg/day for Week 3 and

0.1mg/kg/day for Week 4) and then oral treatment

for 4 weeks, starting at a total of 4mg/day and

decreasing by 1mg each week.12 Patients were

also provided appropriate symptomatic treatment

(intravenous fluids, dexamethasone, mannitol, anti-

epileptic drugs and/analgesics if required).

Pyridoxine was given orally 2040mg/day.

Follow-up and assessment of outcome

Patients were followed up after 1st, 2nd and

6th month after inclusion in the study. During

follow-up, the occurrence of acute stroke was noted.Final outcomes (death or disability), at the end of

6 months of treatment, were assessed on the basis of

modified Rankin scale. A score of 0 indicated no

symptoms; 1 indicated minor symptoms not interfer-

ing with lifestyle; 2 indicated symptoms that might

restrict lifestyle, but patients could look after them-

selves; 3 indicated symptoms that restricted lifestyle

and prevented independent living; 4 indicated

symptoms that prevented independent living, al-

though constant care and attention were not

required; and 5 indicated total dependence on

others, requiring help day and night. The classifica-

tion of outcomes as good (a score of 0), intermedi-

ate (scores of 1 or 2) or severe disability (scores of

3, 4 or 5) was defined before the start of the study.Improvement in stroke was defined as reduction

of NIHSS score by 1 or >1 point from baseline.Similarly, deterioration was defined as increase in

NIHSS score by 1 or >1 point.

Statistical analysis

All data were recorded prospectively and entered

into an electronic database (Microsoft Excel soft-

ware), and checked before analysis. Continuous

variables were compared by the Students t-test nor-

mally distributed and the MannWhitney U-test if

not normally distributed. Categorical variables

were compared by the Chi-square test. A 5% level

of significance was used in all analyses. A cut-off

value was determined for each non-parametric

variable to perform univariate analysis. For multi-

variate analysis, Sigma Stat 2.0 software was used

and the multiple logistic regression method was per-

formed for those variables having a significance of

0.1 on univariate analysis. The analysis was under-

taken using SPSS software (version 15; SPSS,

Chicago, IL, USA).

Predictors of stroke in tuberculous meningitis 673

by guest on May 31, 2015

Dow

nloaded from

Results

During the study period, among 117 patientsfulfilling our diagnostic criteria of tuberculous men-ingitis, 100 patients were enrolled in the study.Reasons for exclusion of 17 patients have beengiven in Figure 1. Six patients were lost to follow-up.Their last observations were carried forward to6 months, and all 100 patients were included inthe final analysis.

Baseline characteristics

Mean age of the tuberculous meningitis patients was30 13 years. Fifty-nine of them were men. None ofthe patients tested positive for HIV infection. Noneof our patients had evidence of previous stroke.Other epidemiological and clinical details ofincluded patients are provided in Table 1. MRIabnormalities seen at the time of inclusion are alsoprovided in Table 1.

Incidence of stroke

At inclusion, focal neurological deficit was presentin 33 patients. All patients had hemiplegia exceptone who had hemichorea. Of these 33 patients,27 patients neuroimaging showed a definite infarct;

hemorrhagic transformation of infarct was seen inone patient. In six cases intracerebral tuberculomaswere the cause of focal neurologic deficits. Later,six patients were lost to follow-up. Three patientsdeveloped infarct during follow-up. In our series,totally 30 (30%) patients had definite stroke.Most of the patients (40%) with cerebral infarcts

had lesions in the region of internal capsule andbasal ganglia. In 12 (40%) patients neuroimagingrevealed multiple infarcts. The location of infarcts,in rest of the patients, was thalamus in one, brain-stem in two and cortical in three patients. Amongthree new strokes, which happened duringfollow-up, two were in internal capsule and onewas in pons. In most of them, infarcts were locatedin the middle cerebral artery territory (Figures 2and 3).

Predictors of stroke

Since only three new patients developed strokeduring follow-up, statistically significant predictorsof stroke in these patients could not be assessed.All 30 patients with stroke were analyzed together.Predictors of stroke, on univariate analysis, were

aged >25 years (P< 0.001), cranial nerve involve-ment (P< 0.001), presence of exudates or

Figure 1. Study design. The observations of patients who were lost to follow-up were carried forward to 6 months, andall 100 patients enrolled were included in the analysis (TBM: tuberculous meningitis; NIHS: National Institute of Health

Stroke Scale).

674 H.K. Anuradha et al.

by guest on May 31, 2015

Dow

nloaded from

meningeal enhancement in and around sylvian fis-sure (P=0.026), posterior fossa (P=0.016) and opticchiasma (P=0.04). Poor visual acuity (P=0.004)and Stage III tuberculous meningitis (P25 years(P=0.012) were found to be significantly associatedwith the stroke.

Prognosis among patients with stroke

During the study period, 13 patients (12.9%) expired(median survival: 42 days, range: 1290 days).Among 27 patients who had stroke at baseline,four patients died, four of them deteriorated, eightremained unchanged and 11 improved.The significant poor prognostic indicators (for pa-

tients who either deteriorated or died) among pa-tients having stroke were presence of cranial nervedeficit (P=0.015), internal capsule infarct (P=0.05),centrum semiovale (P=0.001) and brainstem(P=0.00) infarcts. Infarcts in posterior cerebral

Table 1 Epidemiological, clinical and neuroimagingcharacteristics of patients with tuberculous meningitis

(n=100)

Ageyears (mean SD) (range) 30 13 (1485)Sex

Male 59 (59)

Female 41 (41)

HIV positive 0

Duration of illness days

(mean SD) (range)51 52 (6240)

BMRC staging

Stage I 20 (20)

Stage II 39 (39)

Stage III 41 (41)

Clinical features

Fever, weight loss, headache and/or

vomiting

100 (100)

Seizures 21 (21)

Glasgow coma score (mean SD)(range)

13 2 (515)

Meningeal signs 76 (76)

Hemiparesis 32 (32)

Cranial nerve palsy 52 (52)

Vision acuity

Normal (>6/18) 74 (74)Low vision (between 6/18

and 3/60)

20 (20)

Blindness(

artery territory (P=0.00) and multiple territory in-

farcts (P=0.003) were other poor prognostic indica-

tors. (Table 3 and Figure 4)

Overall prognosis after 6 monthsof follow-up

After 6 months, in total, 27 patients (27%) either

died or survived with severe disability; 38 (38%)

had intermediate and 35 (35%) had good outcome.On univariate analysis, the vascular events that

were associated with death and disability were in-

ternal capsule infarct (P=0.006), basal ganglia in-

farct (P=0.025), middle cerebral artery territory

involvement (P=0.001) independently as well as

in association with Stage III tuberculous meningitis

(P25 years* 1.78 (1.382.28) 0.000Cranial nerve deficit 1.53 (1.172.00) 0.001

Vision impairment 1.07 (1.203.46) 0.023

Focal deficit 1.48 (1.012.17) 0.016

BMRC Stage III of TBM 1.51 (1.112.06) 0.001

MRI abnormalities

Basal exudates 1.3 (1.011.67) 0.042

Sylvian fissure exudates 1.29 (1.001.67) 0.026

Optochiasmatic exudates 1.51 (0.982.34) 0.020

BMRC: British Medical Research Council; CI: confidence

interval; TBM: Tuberculous meningitis; MRI: magnetic res-

onance imaging.

*P=0.012 on multivariate analysis.

676 H.K. Anuradha et al.

by guest on May 31, 2015

Dow

nloaded from

age >25 years, cranial nerve deficits, vision impair-ment, advanced stage of tuberculous meningitis and

presence of exudates (posterior fossa, optic chiasma

and sylvian fissure) but none of these variables

except age >25 years was found significant onmultivariate analysis. Our study like many other

similar studies suggests that vascular complications

in patients with tuberculous meningitis are because

of basal exudates and meningeal inflammation.

Vascular complications can paradoxically develop

even after treatment with antituberculous drugs, per-

haps suggesting an immune mechanism.20

Presence of stroke in tuberculous meningitis is

often considered a factor responsible for poor out-

come. In a study by Kalita and coworkers,17 stroke

predicted poor outcome after 3 months of treatment.

In children with tuberculous meningitis, stroke is

associated with predictors of neurodevelopmental

outcome at 6 months.21 In our study, stroke, espe-

cially the middle cerebral artery territory infarct, was

associated with poor prognosis at 6 months.

However, outcome of tuberculous meningitis in

our study was also found to be associated with

many other factors such as cranial nerve and focal

neurologic deficits, vision impairment, meningeal

enhancement, advanced stage of tuberculous men-

ingitis, low Glasgow coma scale score, baseline-

modified Rankin scale and high protein content

of CSF.Inflammatory changes in the vessels of the circle

of Willis are the most possible reason for vascular

complications in tuberculous meningitis. Exudative

basal meningitis results in strangulation, vasospasm,

constriction, periarteritis and even necrotizing

panarteritis of vessels with or without secondary

thrombosis. These changes jeopardize arterial

blood flow causing ischemia and infarct.20

In a recent study, it was observed that corticoster-

oids may affect outcome from tuberculous meningi-

tis by reducing hydrocephalus and preventing

infarction.22 Our study suggests that the mainstay

of treatment, antituberculous chemotherapy and

corticosteroids, is ineffective in halting the progres-

sion of cerebrovascular complications, as three pa-

tients developed stroke despite administration of

antituberculous therapy and dexamethasone. Our

study was focused, particularly, on predictors of

stroke and outcome in patients of tuberculous men-

ingitis, in the hope that this may prompt further re-

search toward rational and targeted intervention in

preventing development and progression of stroke in

persons with tuberculous meningitis.In conclusion, the present study documents that

stroke in patients with tuberculous meningitis con-

tinues to be a serious complication. We report that

in patients with tuberculous meningitis important

predictors of occurrence of stroke are advanced

stage of illness; presence of exudates, basal and syl-

vian fissure; meningeal inflammation and vision im-

pairment. Stroke, independently as well as along

with other factors, determines the poor outcome in

tuberculous meningitis.

Conflict of interest: None declared.

Figure 4. Bar diagram showing significance of infarcts onoutcome (Disability and death) at 6 months (ACA: anterior

cerebral artery; BG: Basal ganglia; IC: internal capsule;

MCA: middle cerebral artery; PCA: posterior cerebral

artery; MRS: modified Rankin scale).

Table 3 Significant predictors of deterioration in patientswith stroke at 6 months

Parameter Relative risk (95% CI) P

Cranial nerve deficit 1.083 (1.0021.172) 0.015

Infarcts 1.11 (0.971.27) 0.027

Internal capsule 1.12 (0.921.35) 0.05

Centrum semiovale 1.94 (0.497.76) 0.001

Brainstem 1.73 (0.913.29) 0.000

PCA territory 1.32 (0.951.83) 0.000

Multiple territory 1.26 (0.891.79) 0.003

PCA: posterior cerebral artery; CI: confidence interval.

Table 4 Prognostic significance of infarcts on outcome(disability and death) after completion of 6 months of

follow up

Location of infarct Relative risk (95% CI) P

Infarct 1.17 (0.851.61) 0.009

Internal capsule infarct 1.31 (0.8362.06) 0.006

Basal ganglia infarct 1.61 (0.932.81) 0.025

MCA territory infarct* 1.33 (0.911.95) 0.001

*P=0.024 on multivariate analysis.

CI: confidence interval; MCA: Middle cerebral artery.

Predictors of stroke in tuberculous meningitis 677

by guest on May 31, 2015

Dow

nloaded from

References1. Thomas MD, Chopra JS, Walia BN. Tuberculous meningitis

(T.B.M.) (a clinical study of 232 cases). J Assoc Physicians

India 1977; 25:6339.

2. Dalal PM. Observations on the involvement of cerebral ves-

sels in tuberculous meningitis in adults. Adv Neurol 1979;

25:14959.

3. Winter WJ. The effect of streptomycin upon the pathology of

tuberculous meningitis. Am Rev Tuberc 1950; 61:17184.

4. Dastur DK. Neurosurgically relevant aspects of pathology

and pathogenesis of intracranial and intraspinal tuberculosis.

Neurosurg Rev 1983; 6:10310.

5. Nourse P, Schoeman JF, van der Merwe PL. Does cerebral

perfusion pressure influence outcome in children with tuber-

culous meningitis? Dev Med Child Neurol 2004; 46:35760.

6. Thwaites GE, Chau TTH, Stepniewska K, Phu NH,

Chuong LV, Sinh DX, et al. Diagnosis of adult tuberculous

meningitis by use of clinical and laboratory features. Lancet

2002; 360:128792.

7. British Medical Research Council. Streptomycin treatment of

tuberculous meningitis. BMJ 1948; 1:582597.

8. WHO. international Statistical Classification of Diseases,

injuries and Causes of Death. Geneva, WHO, 10th

Revision, 1993.

9. Adams HP Jr, Adams RJ, Brott T, del Zoppo GJ, Furlan A,

Goldstein LB, et al. Guidelines for the early management of

patients with ischemic stroke: a scientific statement from the

Stroke Council of the American Stroke Association. Stroke

2003; 34:105683.

10. Longstreth WT Jr, Dulberg C, Manolio TA, Lewis MR,

Beauchamp NJ Jr, OLeary D, et al. Incidence,

manifestations, and predictors of brain infarcts defined

by serial cranial magnetic resonance imaging in the

elderly: the Cardiovascular Health Study. Stroke 2002;

33:237682.

11. World Health Organization. Treatment of Tuberculosis:

Guidelines for National Programmes. 3rd edn. Geneva,

Switzerland, World Health Organization, 2002.

12. Thwaites GE, Band ND, Dung NH, Quy HT, Oanh DTT,

Thoa NTC, et al. Dexamethasone for the treatment of tuber-

culous meningitis in adolescents and adults. N Engl J Med

2004; 351:174151.

13. Lan SH, Chang WN, Lu CH, Lui CC, Chang HW. Cerebral

infarction in chronic meningitis: a comparison of tuberculous

meningitis and cryptococcal meningitis. Q J Med 2001;

94:24753.

14. Chan KH, Cheung RT, Lee R, Mak W, Ho SL. Cerebral

infarcts complicating tuberculous meningitis. Cerebrovasc

Dis 2005; 19:3915.

15. Koh SB, Kim BJ, Park MH, Yu SW, Park KW, Lee DH. Clinical

and laboratory characteristics of cerebral infarction in tuber-

culous meningitis: a comparative study. J Clin Neurosci

2007; 14:10737.

16. Shukla R, Abbas A, Kumar P, Gupta RK, Jha S, Prasad KN.

Evaluation of cerebral infarction in tuberculous meningitis by

diffusion- weighted imaging. J Infect 2008; 57:298306.

17. Kalita J, Misra UK. Nair PP. Predictors of stroke and its sig-

nificance in the outcome of tuberculous meningitis. J Stroke

Cerebrovasc Dis 2009; 18:2518.

18. Nair PP, Kalita J, Kumar S, Misra UK. MRI pattern of infarcts

in basal ganglia region in patients with tuberculous meningi-

tis. Neuroradiology 2009; 51:2215.

19. Misra UK, Kalita J. Motor evoked potentials in ischaemic

stroke depend on stroke location. J Neurol Sci 1995;

134:6772.

20. Lammie GA, Hewlett RH, Schoeman JF, Donald PR.

Tuberculous cerebrovascular disease: a review. J Infect

2009; 59:15666.

21. Springer P, Swanevelder S, van Toorn R, van Rensburg AJ,

Schoeman J. Cerebral infarction and neurodevelopmental

outcome in childhood tuberculous meningitis. Eur J

Paediatr Neurol 2009; 13:3439.

22. Thwaites GE, Macmullen-Price J, Tran TH, Pham PM,

Nguyen TD, Simmons CP, et al. Serial MRI to determine

the effect of dexamethasone on the cerebral pathology of

tuberculous meningitis: an observational study. Lancet

Neurol 2007; 6:2306.

678 H.K. Anuradha et al.

by guest on May 31, 2015

Dow

nloaded from