08wait 935 (ds) 22/8/02 11:31 am Page 294

Tuberculosis Meningitis and Attention Deficit Hyperactivity

Disorder in Children

by J. W. V. Wait,a L. Stanton,a and J. F. Schoemanb

aDepartment of Psychology, University of Stellenbosch, Republic of South Africa
bDepartment of Paediatrics and Child Health, Tygerberg Hospital and University of Stellenbosch, Republic of South Africa

Summary
The purpose of this study was to investigate the prevalence of attention deficit hyperactivity disorder
(ADHD) in children who recovered from tuberculous meningitis (TBM) as part of an ongoing TBM
research project. During this study, each TBM group subject underwent a thorough clinical–
neurological examination, and a test battery which included the child behaviour check list (CBCL)
Teacher’s Report Form and Conners Rating Scale. The parents and teachers of each of the 21 TBM
group and 21 control group subjects completed the above-mentioned questionnaires. All 21 TBM
group subjects displayed symptoms of ADHD. The TBM group was significantly more hyperactive
and unable to sustain attention than the control group. Furthermore, TBM group subjects were
perceived as being significantly more unpopular, obsessive, compulsive and aggressive than the
control group subjects. With regard to the frequency of externalizing behaviour, the TBM group
subjects displayed significantly more externalizing behaviours as well as symptoms of attention
deficit and hyperactivity. No significant differences between parents’ and teachers’ ratings were
found. We conclude that ADHD is a common long-term complication of TBM.

Introduction brain most commonly affected by ischaemia are the

Infectious diseases are the main killers of young basal ganglia, thalami and brainstem. Infarcts of the
people in developing countries.1 Lungs are the deep grey matter are readily demonstrated by
primary target organs for infection by Mycobac- computerized tomographic (CT) Scan in TBM.
terium tuberculosis. Various other organ systems can Brainstem ischaemia is much more difficult to prove
be secondarily affected as a result of the lymphoid but can sometimes be seen on magnetic resonance
haematogenous spread of the infection. Because of imaging (MRI). Some degree of brainstem ischaemia
the high morbidity and mortality, infection of the is almost without exception demonstrated at
central nervous system by M. tuberculosis is the most autopsy. In a phase of rapid brain growth, the brain
dangerous complication of pulmonary tuberculosis. of affected children is particularly vulnerable and
Tuberculosis meningitis (TBM) develops within 6 extremely susceptible to permanent damage. Conse-
months of primary infection. If the immune system of quently, the after-effects of TBM before the age of 2
the body fails to prevent the development of TBM, it years are more serious than when infection occurs at
penetrates into the subarachnoid space. The subse- a later stage.5–7
quent inflammatory response leads to the develop- The aetiology of attention deficit hyperactivity
ment of TBM.2 Approximately 5 per cent of children disorder (ADHD) is multi-factorial. Biological,
diagnosed with tuberculosis in the Western Cape, psychological and behavioural explanations have
South Africa develop TBM. Despite early diagnosis been offered.8 As far as the biological explanations
and optimal treatment, 8 per cent of those affected are concerned, children who display symptoms of
die,3 and approximately 80 per cent of the survivors ADHD reveal a neuro-anatomical abnormality in
will have neurological sequelae. TBM occurs most the area of brain that regulates attention and motor
frequently during infancy and early childhood and is behaviour. Brain damage resulting from cerebral
the most common cause of death from tuberculosis in infections, trauma or birth injuries has been
children.2,4 Brain damage in TBM is due to the regarded as the primary cause of ADHD. However,
combined effects of raised intracranial pressure due researchers suggested that fewer than 5 per cent of
to obstructive hydrocephalus and ischaemic brain the children reflecting symptoms of ADHD had
injury as a result of peri-arteritis. The areas of the neurological disorders that satisfied this aetio-
logical explanation.9 There is, moreover, no satis-
factory proof that children who have had meningitis
Correspondence: Dr J. W. V. Wait, Department of Psychology, are at greater risk to develop behavioural
University of Stellenbosch, Republic of South Africa. Tel. 27 21 disorders.5
8083456; Fax: 27 21 8083584. E-mail <[email protected]> Three anatomically integrated and interdependent

294 Journal of Tropical Pediatrics Vol. 48 October 2002  Oxford University Press 2002
08wait 935 (ds) 22/8/02 11:31 am Page 295

J. W. V. WAIT ET AL.

systems regulate attention processes: the brainstem, Measuring instruments

diffuse thalamic system and thalamo-frontal system.9 The measuring instruments consisted of two ques-
Recently researchers have demonstrated that in tionnaires. The Teacher’s Report Form (TRF) of the
more than 50 per cent of TBM cases, brainstem Child Behavior Checklist12 was completed by both
responses have been impaired.10 Although neuro- the parents and the class teacher of each subject. The
anatomical studies of ADHD did not always show two versions of the Conners Rating Scale (CRS), the
consistent results, the majority of research supports parents’13 and the teachers’ questionnaire,14 were
the hypothesis that a central nervous system mechan- administered to parents and teachers, respectively, to
ism contributes to the development of some cases of measure the prevalence of attention deficit and
ADHD.8,9,11 Unfortunately, no research could be hyperactivity behaviour among the subjects.15
found on the association between TBM and the
prevalence and frequency of ADHD. The relation- Statistical analyses
ship between diffuse cerebral damage, as in the case The statistical processing of data was done by means
of TBM, and less specific neuropsychological and of the SPSS computer package. The Wilcoxon
behavioural manifestations, such as ADHD, has not Signed Ranks Test for Paired Data was used to deter-
been well documented and there is very little litera- mine statistically significant differences between the
ture on this subject. TBM and control group scores on the subscales of
The specific aims of the study were to establish the TRF and CRS total scores and to determine the
(a) the frequency of behavioural manifestations significance of differences between the scores of
typical of ADHD among children previously parents and teachers.
treated for TBM; (b) whether the difference in
prevalence and frequency of ADHD among TBM
and control subjects was significant; and (c) Results
whether the occurrence of ADHD symptoms in
affected patients manifested more often at home CBCL Teacher’s Report Form
than at school. Figure 1 provides a graphic representation of the
mean scores of TBM and control group members on
the TRF as presented by the parents. The signifi-
Patients and Methods cance of the differences between parents’ assessment
Details of the study have been published elsewhere.7 of the TBM and control groups, as determined by
Subjects were admitted to the Tygerberg Hospital means of the Wilcoxon Signed Ranks Test for Paired
between 1989 and 1994 and diagnosed with and Data, appears in Table 2. Significant differences were
treated for TBM. The first 21 patients who were seen found between the TBM and control groups on the
as part of the long-term follow-up study and who had following subscales of the CBCL: unpopularity (z =
a sibling who fulfilled the criteria for inclusion in the –2.95, p = 0.003); obsessive-compulsiveness (z =
control group were included in the current study and –3.28, p = 0.001); attention deficit (z = –4.02, p =
constituted the TBM group (n = 21). A control group 0.0001); nervousness-hyperactivity (z = –3.95, p =
consisted of siblings of the TBM group (n = 21). A 0.0001); and aggressiveness (z = –2.28, p = 0.023).
subject was selected for the control group if he or she TBM group members had significantly more
was a biological brother or sister of a TBM patient, symptoms of ADHD than their siblings. This is
had never had TBM, was at school, and lived in the evident from their parents’ indication that they have
same house as the patient. revealed a significantly greater attention deficit and
The clinical data of the TBM group and control were more hyperactive than the control group
group are shown in Table 1. The age distribution of members. Figure 2 provides a graphic representation
both groups varied from 7 to 16 years, with an of the mean scores of TBM and control group
average age of 12 years 3 months in the control group members on the CBCL TRF as presented by the
and an average age of 10 years 6 months in the TBM teachers.
group. The control group consisted of 13 boys and 8 The Wilcoxon Signed Ranks Test for Paired Data
girls, and the TBM group of 9 boys and 12 girls. The was used. The results of the teachers’ assessment of
average total IQ of the TBM group was 71.67 (SD = subjects’ behaviour appear in Table 3. Significant
15.36). The average total IQ of the control group was differences were found between the TBM and control
79.14 (SD = 13.81). The Wilcoxon Signed Ranks Test groups on the following subscales of the TRF: unpop-
for Paired Data was used to establish whether this ularity (z = –3.35, p = 0.0008); obsessive-compulsive-
difference had statistical significance. The IQs of ness (z = –3.16, p = 0.001); attention deficit (z = –4.02,
TBM and control groups were not significant (z = p= 0.0001); nervousness-hyperactivity (z = –3.94, p =
–1.93, p = 0.053). 0.0001); and aggressiveness (z = –2.14, p = 0.032). The
The socio-economic conditions of all the subjects teachers’ assessment shows that TBM group
can be described as poor, and most of them lived in members displayed more symptoms of ADHD than
squatter camps in the Cape Peninsula. their siblings.

Journal of Tropical Pediatrics Vol. 48 October 2002 295

08wait 935 (ds) 22/8/02 11:31 am Page 296

J. W. V. WAIT ET AL.

TABLE 1
Clinical data of TBM and control groups

TBM group Control group

Subject Age Sex Neurological status Hospitalization Age Sex

(months) ––––––––––––––––––––––––– (months)
Age* TBM
Stage

1 134 F Normal 48 II 171 F

2 126 M Clumsy 12 III 170 F
3 107 F R-hemiplegia 24 II 87 M
4 132 F Normal 36 II 104 M
5 91 M Normal 24 II 132 F
6 142 M Microcephaly 24 II 192 F
7 83 F L-hemiplegia 24 III 173 F
8 97 F Normal 6 II 168 F
9 126 F Normal 96 II 100 M
10 196 M Normal 12 II 143 M
11 104 F Gen. low muscle tone 24 III 89 M
12 112 F Normal 12 II 164 M
13 153 F Normal 48 II 175 F
14 192 M Normal 72 II 179 F
15 125 F L-hemi; slight R- 12 III 199 M
hemiplegia
16 135 M Slight L-hemiplegia 12 II 189 M
17 150 F Normal 72 II 108 F
18 94 F Normal 48 II 138 M
19 113 M Normal 24 III 119 M
20 115 M Normal 48 II 142 F
21 101 M Normal 12 II 152 M

M 125.14 36 147.33
SD 29.90 24.0 35.01

*Age (in months) at admission.

FIG. 1. CBCL profile of TBM, () and control () groups—parent format.

296 Journal of Tropical Pediatrics Vol. 48 October 2002

08wait 935 (ds) 22/8/02 11:31 am Page 297

J. W. V. WAIT ET AL.

FIG. 2. CBCL profiles of TBM () and control () groups—teacher format.

TABLE 2 TABLE 3
Results of the Wilcoxon Signed Ranks Test for Results of the Wilcoxon Signed Ranks Test for
Paired Data with CBCL scores of TBM and control paired data with CBCL scores of TBM and control
groups—parent format (n = 42) groups—teacher format (n = 42)

Subscales z Significance Subscales z Significance

Anxiety –0.64 Anxiety –0.04

Social withdrawal –1.74 Social withdrawal –1.65
Unpopularity –2.95 ** Unpopularity –3.35 ***
Self-destructiveness –1.24 Self-destructiveness –0.60
Obsessive-compulsiveness –3.28 *** Obsessive-compulsiveness –3.16 ***
Attention deficit –4.02 *** Attention deficit –4.02 ***
Nervousness-hyperactivity –3.95 *** Nervousness-hyperactivity –3.94 ***
Aggressiveness –2.28 * Aggressiveness –2.14 *
Other problems –1.73 Other problems –1.28

* p < 0.05; ** p < 0.01; *** p < 0.001. * p < 0.05; ** p < 0.01; *** p < 0.001.

Conners Rating Scale (CRS) fewer behavioural problems at school than TBM
Descriptive statistics of CRS total scores of TBM and group members (z = –3.91, p = 0.000). Thus, accord-
control groups for both parent format and teacher ing to the teachers as well as the parents, children
format are presented in Table 4. TBM group who have had TBM have a higher frequency of
members had higher CRS total scores than the externalizing behaviour, attention deficit and hyper-
control group (as judged by the parents). The activity in comparison with control group members.
Wilcoxon Signed Ranks Test for Paired Data was The Wilcoxon Signed Ranks Test for Paired Data
used to establish whether this difference had statisti- was also used to examine objective 3. According to
cal significance. The Wilcoxon T-scores appear in the Conners total scores, the TBM group displayed
Table 5. Control group members displayed signifi- more behavioural problems at school than at home
cantly fewer behavioural problems at home than (z = –2.86, p = 0.004, Table 5). However, no signifi-
TBM group members (z = –3.76, p = 0.000), and also cant differences were found between the parents’

Journal of Tropical Pediatrics Vol. 48 October 2002 297

08wait 935 (ds) 22/8/02 11:31 am Page 298

J. W. V. WAIT ET AL.

TABLE 4 hyperactive behaviour at school than at home. The

Descriptive statistics of CRS total scores of TBM and attention deficit and hyperactive behaviour of the
control groups (n = 42) TBM group occurred both at home and at school.

Conners Rating Scales Mean SD

Discussion
TBM group (n = 21) The results obtained by this study show that all the
Parent format 27.19 6.49
children in the TBM group displayed symptoms of
Teacher format 32.76 6.56
Control group (n = 21)
ADHD. The statistically significant results of TBM
Parent format 14.76 8.98 group members’ hyperactive behaviour and their
Teacher format 14.10 8.30 inability to pay and sustain attention, confirm the
supposition that ADHD can be caused by damage to
the brain areas that control attention and behaviour
and by acute and serious brain infections. These
TABLE 5 findings correspond with the views of Ballard, et al.11,
Results of the Wilcoxon Signed Ranks Test for Barkley9 and Trexler and Zappala.16 According to
Paired Data with CRS total scores of TBM and Barkley9 and Schaffer,5 however, fewer than 5 per
control groups cent of children who display symptoms of ADHD
(n = 42) have neurological disorders. Although this study
focused mainly on the prevalence of ADHD-type
Scales z p Significance symptoms in TBM the additional finding of a high
incidence of disruptive behavioural disorders in
TBM group (n = 21)
Parent scores and teacher
TBM deserves further study.
scores –2.86 0.004 ** Although we showed dramatic differences in
Control group (n = 21) behaviour of TBM survivors and siblings, the results
Parent scores and teacher of this study should be seen in the light of the small
scores –0.26 0.79 sample size. It is also important to consider other
Parent format variables apart from TBM which could possibly have
TBM group and control affected the outcome, such as the role of family func-
group (n = 42) –3.76 0.000 *** tioning (e.g. behavioural control), and a family
Teacher format
history of psychiatric disorder. The possibility that
TBM group and control
group (n = 42) –3.91 0.000 ***
the TBM patients were premorbidly, genetically
predisposed to develop ADHD could also have
* p < 0.05; ** p < 0.01; *** p < 0.001.
played a role.

References
TABLE 6 1. Ebrahim GJ. Multi-drug resistant tuberculosis. J Trop Pediatr
Results of the Wilcoxon Signed Ranks Test for 2001; 46: 320–21.
paired data with CBCL scores of TBM group— 2. Donald PR. Aids to the diagnosis of meningitis in a population
parent and teacher format (n = 21) with a high prevalence of tuberculosis. MD dissertation,
Stellenbosch University, 1985.
Subscales z Significance 3. Department of Health. Health statistics report and health
services evaluation (Western Cape Province), 2001.
Anxiety –0.86 NS 4. Deeny JE, Walker MJ, Kibel MA, Molteno CD, Arens LJ.
Social withdrawal –0.73 NS Tuberculous meningitis in children in the Western Cape. S Afr
Unpopularity –0.46 NS Med J 1985; 68: 75–8.
Self-destructiveness –0.80 NS 5. Schaffer D. Brain damage. In: Rutter M, Hersov L (eds), Child
Obsessive-compulsiveness –0.10 NS and Adolescent Psychiatry. Blackwell Scientific Publications,
Attention deficit –1.07 NS London, 1985; 129–45.
Nervousness-hyperactivity –0.27 NS 6. Schoeman JF. The role of continuous intracranial pressure
Aggressiveness –1.120 NS monitoring in the management of children with tuberculous
Other problems –1.150 NS meningitis. MD dissertation, Stellenbosch University, 1987.
7. Ravenscroft A, Schoeman JF, Donald, PR. Tuberculous granu-
lomas in childhood tuberculous meningitis: Radiological
features and course. J Trop Pediatr 2001; 47: 5–12.
8. Kronenberger WG, Meyer RG. The Child Clinician’s
and teachers’ assessment of TBM group members on Handbook. Allyn & Bacon, Massachusetts, 1996.
the CBCL (Table 6). It could not be said whether 9. Barkley RA. Diagnosis and assessment of attention deficit
TBM had had a more severe impact on the hyperactivity disorder. Compr Ment Health Care 1991; 1: 27–43.
prevalence and frequency of attention deficit and 10. Kapoor RK, Makharia A, Shukla R, Misra PK, Sharma B.

298 Journal of Tropical Pediatrics Vol. 48 October 2002

08wait 935 (ds) 22/8/02 11:31 am Page 299

J. W. V. WAIT ET AL.

Brainstem auditory evoked response in tuberculous meningitis. 14. Conners CK. Rating scales for use in drug studies with children.
Indian J Pediatr 1997; 64399–407. Psychopharmacol Bull 1973; 24–84.
11. Ballard S, Bolan M, Snyder S, Pasterczyk-Seabolt C, Martin D. 15. Moehle KA, Fitzhugh-Bell KB. Factor analysis of the Conners
The neurological basis of attention deficit hyperactivity Teacher Rating Scale with brain-damaged and learning disabled
disorder. Adolescence 1997; 32: 855–62. children. Psychol Schools 1989; 26: 113–25.
12. Achenbach TM, Edelbrock C. Manual of the Teacher’s Report 16. Trexler LE, Zappala G. Neuropathological determinants of
Form and Teacher Version of the Child Behaviour Profile. Allyn acquired attention disorders in traumatic brain injury. Brain
& Bacon, Massachusetts, 1986. Cogn 1988; 8: 291–302.
13. Conners CK. A teacher rating scale for use in drug studies with
children. Am J Psychiatry 1969; 126: 884–85.

Journal of Tropical Pediatrics Vol. 48 October 2002 299