Brainstem auditory evoked potentials in tubercular meningitis and their correlation with radiological findings.

J Kalita, UK Misra 
 Department of Neurology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow - 226014, India., India

Correspondence Address:
J Kalita
Department of Neurology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow - 226014, India.
India

Full Text

::   Introduction

Tubercular meningitis (TBM) differs from pyogenic

and viral meningitis by predominant basal

involvement, associated granuloma formation,

vasculitis, infarction and hydrocephalus.[1],[2] These

associated pathologies may affect the brainstem

functions. Brainstem auditory evoked potential

(BAEP) studies may provide objective noninvasive

documentation of these functions. BAEP studies have

been extensively conducted in pyogenic meningitis

for monitoring hearing loss.[3],[4],[5],[6] We could not get any

report of BAEP changes in TBM in the literature. In

the present communication, results of BAEP have

been correlated with clinical and radiological

findings.

::   Material and methods

Patients with TBM seen between 1996 and 1998 were

included in this study. All the patients underwent a

detailed neurological evaluation. Consciousness was

assessed by Glasgow coma scale (GCS), muscle

power by Medical Research Council scale and muscle

tone by Ashworth scale.[7] In comatose patients,

oculocephalic reflex was noted.

Plain and contrast enhanced cranial CT with 10 mm

cuts was done in all cases. Cranial MRI was carried

out on 2T scanner operating at 1.5T, (Magnetome SP

Siemens, Germany). T1WI, proton density and T2

weighted spin echo sequences were obtained.

Presence of hydrocephalus, infarction, exudate and

tuberculoma were noted. The diagnosis of TBM was

based on clinical, CT scan and CSF criteria.7 The

clinical criteria included fever, headache and neck

stiffness for more than 2 weeks. The supporting

evidences were obtained from CSF cells (0.02x109/L

or more, with lymphocytic predominance), proteins

(more than 1 gm/L), sterile bacterial and fungal

culture and CT scan findings for the presence of

hydrocephalus and exudate. Evidence of tuberculosis

outside the central nervous system and response to

antitubercular therapy were noted. In the CT scan,

presence of tuberculoma and infarction were

recorded. The diagnostic categories included (i)

highly probable: clinical and 3 supportive criteria;

(ii) probable: clinical and 2 supportive criteria and

(iii) possible: clinical and one supportive criteria.[8] The

severity of TB meningitis was graded into stage I :

meningitis only, stage II : meningitis and neurological

signs and stage Ill : meningitis, neurological signs

with altered sensorium.[9]

Brainstem auditory evoked potentials (BAEP) : BAEP

were recorded with 0.1 ms rarefaction click

stimulation delivered monaurally at 95 dB. A uniform

stimulus strength was used in all the patients because

most of our patients had altered sensorium. The

stimuli were delivered at a rate of 10 Hz. 2048

responses were twice averaged with a bandpass filter

of 100Hz-3kHz.[10] The latencies of different waves

and interpeak latencies (IPL) of I-III, I-V and III-V

were measured. The results were compared with

control values obtained from 30 healthy volunteers,

whose age ranged between 15 and 68 years. The upper

limit of normal was defined as mean+2.5 SD of

control. The amplitude ratio of wave V/I was also

calculated and considered abnormal if it was below 1.

The clinical and radiological parameters were

correlated with BAEP findings. The recovery was

defined on the basis of 3 month Barthel index score

(BI) into good (BI>12) or poor (BI< 12).[11]

::   Results

Twenty four patients of TBM were included in this

study. Their mean age was 26.4+14.[9] (range 10-62)

years. Eight of them were females. Sixteen patients

were in stage Ill, 5 in stage II and 3 in stage I

meningitis. Their mean GCS score was 10.8+4.2.

Cranial nerves were involved in 12 patients; optic

nerve in 5 and partial or complete external

ophthalmoplegia in 7 patients. In patients with stage I

and II meningitis, bedside hearing tests were normal.

Focal weakness was present in 17 patients which

included quadriplegia in 10, hemiplegia and

paraplegia in 3 each and monoplegia in one patient.

Cranial CT scan was carried out in all and MRI in 15

patients. Cranial CT scan revealed exudate in 9,

hydrocephalus in 16, infarction in 12 and tuberculoma

in 3 patients. The infarctions were located mostly in

basal ganglia. Brainstem infarction was present in 2

patients only; crus cerebri was involved in 1 and right

pontomedullary region in the other. MRI scan was

done in 15 patients and revealed additional findings in

8, which included a midbrain granuloma in 1. The

MRI in remaining patients revealed supratentorial

abnormalities.

Brainstem auditory evoked potentials were

unrecordable in one and recordable in 23 patients.

These were abnormal in 15 of them. The absolute

latency of BAEP waveforms was normal in most

patients [Figure 1] except in 3 sides where wave III

latency was prolonged. Interpeak latency of I-III was

prolonged in 3 sides. Wave V and III were not

recordable in 1 and 3 sides respectively. The group

difference between absolute latency of waveforms and

IPL between TBM patients and controls was not

significant. The mean and SD of various BAEP

waveforms and IPLs in TBM patients and controls are

shown in [Table I]. The V/I amplitude ratio was

abnormal in 12 sides, however, the group differences

were not significant (Z=-1.43).

BAEP and clinicoradiological correlation : The

BAEP abnormalities in 15 patients, did not correlate

with stage of meningitis (X2=0.04, df=2 NS) and level

of consciousness (X2=0.94, df=2 NS). Out of 3

patients in stage I meningitis, BAEP was abnormal in

2, whereas out of 16 patients with stage III meningitis,

it was abnormal in 10 patients. The BAEP

abnormalities also did not correlate with 3 month

outcome (X2=0.10, df=1 NS). At 3 month followup, 6

patients had good and 14 had poor outcome. CT scan

abnormalities also did not significantly correlate with

BAEP abnormalities. Hydrocephalus was present in

16 patients, 10 of these patients had abnormal BAEP

and 8 of whom had abnormal V/I amplitude ratio. The

other abnormalities in the patients with hydrocephalus

included abnormal wave II in 1, wave III in 6, wave V

in 1, I-Ill IPL in 4, III-V IPL in 2 and I-V IPL in 1

patient. Presence of hydrocephalus, however, did not

correlate with BAEP abnormality (X2=0.20, df=1,

NS). One patient had lateral pontomedullary

infarction in whom wave III and V were unrecordable.

The other patient had infarction of midbrain and crus

and his wave V/I amplitude ratio was abnormal. The

distribution of radiological changes in patients with

abnormal BAEP are summarised in [Table II].

::   Discussion

In this study, BAEP was abnormal in 16 out of 24

patients with TBM. The abnormalities included

reduction of wave V/I amplitude ratio, prolongation of

I-III IPL; absence of wave III and V; however, the

group difference was not significant. The reduction of

wave V/I amplitude ratio was the commonest

abnormality, and was present in 11 patients. The

amplitude reduction may be due to raised intracranial

tension (ICT) which often accompanies TBM.

Presence of hydrocephalus in 8 patients with V/I

amplitude reduction further supports the possibility of

raised ICT, although intracerebral pressure monitoring

was not undertaken. Reduction of wave V/I amplitude

ratio has also been reported in supratentorial space

occupying lesion resulting in upper brain stem

compression in cats.[12] It has been shown that

mechanical compression of upper brainstem causes

interference with brainstem circulation. In our study,

2 patients had brainstem infarction. BAEP was

abnormal in both. Wave V/I amplitude ratio reduction

was seen in 1 and unrecordable wave Ill and V in the

other. BAEP has been reported to be useful in

monitoring brainstem ischaemia and impending

brainstem stroke respectively in patients presenting

with vertigo.[13],[14],[15] The BAEP abnormalities are

common in the acute stage compared to recovery

period.13 Vasculitis, although common in TBM, may

be seen in 20.5-55% patients depending on the

method of evaluation,[15],[16] but is reported to be rare in

vertebrobasilar territory.[1],[17] The infarctions are

usually present in supratentorial perforating vessels

affecting basal ganglia, thalamus and internal capsular

region. In our study, basal ganglionic and capsular

infarctions were present in 10 and brainstem stroke in

2 patients only.

Prolongation of wave I-III IPL was noted in 3 patients.

It suggests abnormality in proximal part of VIII nerve

to pontomedullary junction or lower pons around

superior olive or trapezoid bodies.[18] The midbrain

granuloma was seen in one of our patients in whom

wave III was unrecordable. On analysing the CT scan

with prolonged I-III IPL, miliary granuloma,

hydrocephalus with thalamic and capsular infarctions

and hydrocephalus with supratentorial granuloma was

present in one patient each. Presence of basal exudate

may also result in I-III IPL prolongation. The diversity

of BAEP abnormalities and lack of specific pattern

may be due to diversity of pathophysiological

mechanisms in TBM e.g. in the same patient

hydrocephalus, infarctions, tuberculoma and varying

degree of raised intracranial pressure (ICP) may be

present in various combination at different stages of

meningitis. Most of our patients were in stage III,

therefore, the correlation of hearing deficit with BAEP

changes was not possible. In bacterial meningitis,

BAEP studies have been used to monitor the hearing

deficit.[3],[5]

From this study, it can be concluded that although

BAEP abnormalities are frequent in TBM but are

probably nonspecific and mostly in the form of

reduced V/I amplitude ratio. Further studies are

needed including CSF pressure monitoring to evaluate

the role of ICP on BAEP changes and possible

reversibility of these changes after manoeuvers

reducing ICP in TBM.

References