Cognitive sequelae of severe malaria with impaired consciousness

TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1999)93,529-534

Cognitive

sequelae

of severe malaria with

impaired

consciousness

P. A. Holding’,3, J. Stevenso&, N. Peshu’ and K. Marsh1g3 ‘KEMRI CRC Kilifi Unit, I? 0. Box 230, Kilifi, Kenya; ‘Depanment of Psychology, University of Southampton, Highfield, Southampton SO1 7 1BJJ UK; 3Nuj5eld Department of Clinical Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK Abstract Although cerebral malaria is the most common acute encephalopathy arising in children in Africa little is known of its effect upon the longer-term cognitive development of survivors. In Kenya, we compared the performance of 87 survivors of severe malaria with impaired consciousness to matched community controls on a wide range of tasks, not less than 42 months post illness episode. The presence of cognitive impairment was then related to both the pattern of symptoms at the time of the acute illness and the presence of gross neurological impairment on discharge. Significant group differences were found in areas of cognitive functioning suggestive ofwidespread impairment in the development ofthe ability to initiate, plan and carry out tasks (the executive functions). On tasks of more discrete cognitive skills (information processing) there were no significant group differences, although impaired performance was found more frequently in the severe malaria group. The odds ratio associated with the development of cognitive impairment following severe malaria with impaired consciousness was found to be 4.48 (95% CI 1.22, 16.47). A combination of 4 signs (coma, hypoglycaemia, seizures, and absence of hyperpyrexia) proved to have greater accuracy than the presence of gross neurological sequelae in predicting cognitive impairment (95% vs 93% specificity, 67% vs 58% sensitivity). Keywords:

malaria, cognitive development, cognitive sequelae, childhood encephalopathy, Kenya

Introduction Plasmodium falciparum is a major cause of diffuse encephalopathy in African children (MARSH, 1992; World Bank, 1993). This encephalopathy has a mortality rate in hospital of lo-40% (GREENWOOD et al., 1987; MOLYNEUX et al., 1989; WHO, 1990; MARSH et al., 1995). Although the majority of survivors of this encephalopathy appear to make a full recovery, in studies across Africa 5-20% of children have been found to have residual neurological sequelae up to 1 year post discharge (MOLYNEKX et al., 1989; BREWSTER et al., 1990; BONDI, 1992; VAN HENSBROEK et al., 1997). Common sequelae include cortical blindness, aphasia, and hemiplegia. As studies to date have largely been short term, and concerned with gross neurological impairment, few data are available to indicate the possible longer-term educational implication of these impairments, nor to indicate the prevalence of more subtle cognitive and behavioural impairments. The most likely pattern of effect was envisaged as being related to a ‘critical level’ of damage, where cognitive function would be unaffected unless the cerebral insult were sufficient to manifest itself as gross neurological sequelae (CHADWICK et al., 198 1; FINGER & ALMLI, 1988; MUNTENDAM et al., 1996). The main objectives of this study were to determine whether survivors of severe malaria with impaired consciousness experience significant cognitive impairment, and to determine the attendant risk factors. Materials and Methods Sample A power calculation for a multivariate analysis of variance, with 2 groups and 30 dependent variables, was computed (COHEN, 1992). The alpha level was set at 0.05, and the power at 0.80. In order to detect an effect size of 0.4, comparable to the effect size found in previous research on meningitis, 2 equal size samples of 78 would be required (TAYLOR et al., 1990). Hospital records provided a pool of 106 appropriate subjects, 89 of which were available for recruitment. This was sufficient to meet the requirements of the above power calculation. Index cases were children of a single ethnic group (Mijikenda) with severe malaria and impaired consciousness (Blantyre score of ~4, cannot localize pain; MOLYNEUX et al., 1989) admitted to the KEMRI clinical Address for correspondence: P. A. Holding, KEMRI Centre for Geographic - _ Medicine Research. Coast. P.O. Box 230. Kilifi. . Kenya; fax +254 (0) 125 22390,e-mail [email protected]

research ward at Kilifi District Hospital (Kenya). Inclusion criteria were: age >9 months at admission, and in their 7th year at assessment (to provide a period for spontaneous recovery, and a logistically manageable focus for the development of the test materials). Exclusion criteria were: evidence of a physical disability which could interfere with the completion of the assessment tasks; primary residence of more than 1 day’s return travel to Kilifi District Hospital; or failure to obtain parental consent. A matched pairs design was used. Controls were recruited from a pool of children randomly selected in the community (full details are published in SNOW et al., 1998). Cases and controls were matched for age (f3 months), gender and mother tongue. In addition matching was also made on the maximum number possible of 7 measures taken at the time of the original illness. These were selected because they were either significant risk factors for the development of cerebral malaria or indicators of nutritional and socioeconomic status, and thus potentially related to cognitive outcome. These measures were placed in a prioritized order for matching, based upon the size of the odds ratios calculated as an estimate of relative risk for severe malaria (K. Marsh, personal communication). Each child was assigned membership of the risk or no-risk group (in parentheses is the definition for risk group status): nutritional status (wasted/wasted and stunted), mother’s ability to speak English (no), mean corpuscular volume (c69.4 fl), mother’s occupation (housewife), mother’s level of schooling (none), hookworm density (>5 eggs/mg), and mid-upper arm circumference (c13.5 cm). Overall a mean of 8 matched variables (SD 1.12) was achieved. Procedures Each child was seen on 3 separate occasions, by assessors blind to their group status, who were fluent in the appropriate languages, and trained in the assessments to be used. During the 1st appointment children were screened for current nutritional status (mid-upper arm circumference), sight (acuity, pass at 6/18, and visual field), hearing (pass, 30 dB at 500,1000,2000 and 4000 Hz both ears), and general health. Children who failed any of these steps were referred to medical personnel for review, and seen for assessment following appropriate management. The administration of the Kilifi Assessment Battery (HOLDING, 1998) then took place over 2 assessment sessions. The battery included tests of informationprocessing skills (comparable to those included in IQ

530

P.A.HOLJXNGETAL.

tests), achievement, visuo-motor speed, attentiomplanning, language development, observations of test session behaviour, and a parental assessment of the child’s behaviour at home. The content and procedures of all sub-tests were adapted to suit the current assessment context, and had undergone piloting and reliability studies on a local population. Two 20-min observations of the test session behaviour occurred for each child and were carried out by the principal investigator. (See Table 1 for a more detailed description of sub-tests.) Statistical analysis Scores and ratings were analysed using SPSS for Windows (NORUSIS,1992). In view of the multiple measures being taken, the main core of the battery was analysed using a multivariate analysis ofvariance (MANOVA) procedure. A non-parametric technique (Wilcoxon) was applied to other tasks where appropriate. Impaired performance was defined as a score falling at 2 or more standard deviations below the mean for the control group (SAT& 1993). An impairment score was then calculated from performance on the main core of the battery, i.e., the information-processing tasks, indicating the number of sub-tests in which an impaired performance was recorded. Only those children with an impairment score of 32 were designated as being ‘cognitively impaired’. Logistic regression analyses were conducted to assess the associations of disease-related factors with cognitive impairment for the case children. These factors included: neurological complications at discharge, age at admission, rectal temperature, and laboratory findings (haemoglobin, parasite count, blood glucose). To facilitate subsequent- interpretation these variables were dichotomized. The relative risk ofmembershin ofthe imnaired group was estimated for each individual variable, as an odds ratio with a 95% confidence interval (CI). The inclusion criterion for the logistic regression was a univariate P value ~0.2 in the cognitively impaired vs non-impaired comparisons. For retention in the model a Pvalue GO.05 in the logistic regression was required. A Table 1. Description of cognitive impaired consciousness

Hand movementsa Matrix analogies” Magic windowa Number recall” Word ordera

Risk factors for cognitive impairment Table 4 describes the univariate analvsis of the clinical variables comparing the cognitively impaired group of cases with the remainder of the cases. The individual

measures

used to assess sequelae

of severe malaria

with

Items measured early mathematical skills including counting, and computation (adapted from the K-ABC”, and the BAS (ELLIOT et al., 1983) Moving pegs-scores included relative hand skill, and overall speed Visual search paper and pencil cancellation task (adapted from BADDELEY et al., 1995)

Arithmetic Pegboard Attention Language Verbal comprehension Fluency Syntax, semantics, articulation

“AdaptedfromtheK-ABC

Developmental outcome Information processing. In all sub-tests, with the exception of Magic window, the mean score of cases was inferior to that of controls, and in each there was a greater number of cases showing an impaired performance. Thus, although there was no overall significant difference (F = 1.06 P > 0.1, df = 8,78), there was a signiticant difference in the ‘impairment score’ between cases and controls (Wilcoxon Z = -250, 2-tailed P = 0.01; Table 2). Cerebral malaria was associated with an odds ratio of 4.48 (95% CI 1.22, 16.47) for cognitive impairment (Figure). There were significant differences in mean scores between cases and controls in aspects of language development (syntax and articulation), attentiomplanning (visual search), and behaviour (parent questionnaire) (see Table 3).

Spatial ability and visuo-motor co-ordination (adapted from RUTTERet al., 1970) Short-term memory of faces Visuo-perceptual organization; the identification of partially complete silhouette drawings Memory for a visual sequence Selection of a design which best completes a 2 by 2 visual analogy Recognition and integration of components of a picture, presented in segments Memory for digit sequences, reproduced forwards Memory of verbal sequences, reproduced through pointing at a sequence of silhouette pictures

Face recognition” Gestalt closure”

Parent questionnaire”

Results Overall, 87 pairs were available for assessment in 1995-96 (Figure), with a mean age at assessment of 79.7 months (SD 3.7). The cases had been treated for severe malaria with impaired consciousness at a mean age of 26 months (SD 7.2; range lo-42 months), thus ensuring a minimum of 42 months since the illness episode (range 42-70 months). Thirty seven of the pairs were female. There were no significant differences between the control and study groups on a range of family and background measures, including: income source, family size, schooling, and parental report of early developmental history. There were also no significant differences in the number of children requiring attention following sight and hearing screening.

Description

Sub-test Information processing Construction

Behaviour observation

and behaviour

best-fit model was sought through exploratory analysis of different cut-off points for the disease-related variables.

Selection of appropriate line drawing to match word provided (based upon SIGMAN et aZ.‘s (1989) adaptation of the Peabody picture vocabulary test) Questions, riddles, and picture cards, used to stimulate free speech to assess pragmatic errors (DAME0 & OLLER, 1980) Ratings made on transcripts Observations (2 X 20 min) of on-off task behaviour, and response to stress (schedule developed in Kilifi) Schedule developed in Kilifi to assess behaviour difficulties in the home

(KMJFh4AN&KAUFMAN,1983).

COGNITIVE

SEQUELAE

531

OF SEVERE MALARIA

Cases discharged

1

c

I 80

I

1 (A)

DEAD at follow-up “’

i

Neurological Y( (9%) Cognitively unimpaired

Figure. Selection from hospital records and cognitive outcome of 87 children in Kenya who had had severe malaria with impaired consciousness. “Discharge record incomplete. ‘8 moved away/untraceable, 1 parental refusal. ‘2 excluded, 1 because the case child was too physically disabled to manipulate the materials, and the other because of a similar problem with his/her matched control.

Table 2. Levels severe malaria

of cognitive

impairment

found

Cases No. of testsa

2+ 3+ 4+

No. of subjects 12 11 10

“In which a subject’s performance bMcNemar’s binomial.

in children

who had had

Controls

(%)

No. of subjects

(%I

I::;

3

8;;

(12)

fi

(0)

2-tailed P valueb 0.02 0.01 0.002

was designated as impaired.

variables making a significant contribution to the best-fit model were the measures of coma (Blantyre score 6 2), hypoglycaemia (S 2.6 mmol/L), seizures (3+), and absence of hyperpyrexia (<4O”C). This combination resulted in 8 of the cognitively impaired group being correctly predicted, 4 children were missed and 4 incorrectly selected (95% specificity, 67% sensitivity). The odds ratio for the development of cognitive impairment associated with gross neurological impairment at discharge was 19.6 (95% CI 4.64, 84.65). As compared to the multiple factor model (severity of clinical disease), using neurological sequelae as a predictor of future cognitive impairment resulted in a lower sensitivity (58% vs 67%), with a comparable specificity (93% vs 95%), such that 5 (42%) of the children subsequently found to be cognitively impaired would not have been identified through neurological sequelae alone. As the overlap between the 2 approaches was only 50% ofchildren correctly identified, a combination ofthe

2 approaches was calculated. This culminated in an increase in sensitivity to 75%, but there was an overinclusion of children identified as at risk (specificity reduced to 88%). Three children failed to be selected by either approach. Discussion A proportion of survivors of severe malaria with impaired consciousness are left with sequelae. However, with 1 exception, studies leading to this view have been short term, and focusing on gross sequelae. The exception, carried out in the Gambia, found no significant effect upon cognitive performance in a population in which gross neurological sequelae at discharge are also absent (MUNTENDAM et al., 1996). The pattern of differences in performance on the information-processing sub-tests found between cases and controls in the current study was characterized by impairments in only a small number of survivors. Con-

532

P. A. HOLDING ETAL.

Table 3. Comparison control children

on the cognitive

n (pairs)

Sub-test (max./best score) Information

and behaviour

measures

of children

who had had severe malaria

Cases

Controls

(mean f SD)

(mean & SD)

and

2-tailed P value

Value

processing

Construction (28) Face recognition (15) Gestalt closure (16) Hand movements (20) Matrix analogies (14) Magic window (14) Number recall (18) Word order (12)

87 87 87 87 86 87 87 87

15.46 6.51 5.56 12.11 5.60 6.29 10.74 6.52

7.27 2.86 3.73 3.66 3.03 3.06 3.62 2.77

17.20 6.64 5,66 12.61 5.73 6.24 11.46 6.84

5.68 1.99 2.82 2.88 2.50 2.62 2.37 2.28

F F F F F F F F

3.63 0.18 0.01 1.12 0.16 0.03 2.73 0.74

0.06 0.67 0.92 0.29 O-69 0.87 0.10 0.39

Arithmetic (45) Pegboard” (12) Attention/visual search (48)

29.24 23.49 42.47

9.42 24.74 10.81

31.15 17.68 45.62

6.66 2.78 2.71

t = -1.53

iii:: 65

Z = -1.43 Z = -1.95

0.13 0.15 0.03

87 87 87 87 87

20.83 11.68 3.74 3.05 2.71

7.24 21.99 0.69 0.53 0.63

21.63 7.67 3.92 3.17 2.90

5.75 12.23 0.27 0.41 0.34

F = Z= Z = z = Z =

0.38 0.24 0.02 0.08 0.02

87 87 87

0.53 0.50 20.98

5.49 4.39 3.00

-0.53 -0.50 18.92

1.59 1.82 2.00

and cognitive

impairment

= = = = = = = =

Language Verbal comprehension (40) Fluency (pragmatic errors)a (0) Syntax (4) Semantics (4) Articulation(?)

0.77 -1.18 -2.07 -1.47 -2.25

Behaviour Observation Off-task” (0) Response to stressa (0) Parent questionnaire” (15)

z = -0.95 z = -1.77 t = 5.35

0.36 0.08
“A lower acore reflects a superior performance.

Table 4. The relationship severe malaria

between

clinical

Variable Age at admission (~24 months) Coma score (C2) Time to localization of pain (224 h) Hypoglycaemia (s2.6 mmol/L) Total seizures (23) Haemoglobin (~5 g/dL) Respiratory distress (% with 1+ episode) Hyperparasitaemia (>500 OOO/100 RBC) Hyperpyrexia (340°C)

variables

Fisher’s exact 2-tailed P 0.76 0.007 0.07 0.02 0.17 0.75 0.31 0.68 0.11

Odds ratio (associated with cognitive impairment) 0.71 Undefined 3.41 5.40 5.18 1.27 0.51 0.48 Undefined

in the children

who had had

95% CI 0.20-2.57 All ‘cognitively 0.97-l 1.96 1.46-19.91 0.63-42.43 0.37-4-39 0.15-1.79 0.06-4.05 All ‘cognitively <4O”C

impaired’

<2

impaired’

CI, confidence interval; RBC, red blood cells.

trary to the inference made by the authors of the Gambian study this sub-group was not adequately predicted by gross neurological sequelae manifest on discharge. This, and other differences, discussed below, found between the 2 studies may largely be explained by differences in the selection of cognitive functions investigated, and in particular by the power of a larger sample size to detect a significant difference. In contrast to the Gambian study significant group differences were identified in aspects of language development, behaviour difficulties and attention. This pattern of impairment suggests an immaturity in the development of affected children, when measured at 3-5 years post insult. This may be a consequence of differences in the experiences and stimulation provided to children following a severe, life-threatening episode of illness. If so, then performance would be expected to improve following appropriate educational input,

although behavioural differences may be more persistent unless changes also occur in parental attitude and approach (CAMPBELL, 1995). Even if the differences identified are primarily attributable to brain damage due to the malaria infection, stimulation and appropriate educational opportunities could still diminish the difference between the case children and their peers. However if the effects, although subtle, are irreversible, then the difference between the 2 groups would be expected not only to persist but also to become more marked as development progresses, and as the tasks that are required to be tackled become more complicated (RUTTER, 1984). Had extensive brain damage been widespread amongst the cases then we might have expected qualitative differences in cognitive performance (VYGOTSKY, 192911993; LuruA, 1973). These were not apparent either in terms of a different pattern of strengths, or in

COGNITNE

533

SEQUELAE OF SEVERE MALARIA

differences in the pattern of errors for the majority of the case group. Nonetheless, given the general trend of lower means among the case group, we cannot exclude the possibility of an impairment of a function not evident at age 6 years, but which could lead to an increasing difference in performance levels. It will clearly be important to examine the performance of these children on more complex tasks as they mature. The performance of the sub-group of survivors with definite cognitive impairment was more consistent with irreversible brain damage. On the visuo-motor task the cognitively impaired group showed a significant imbalance in relative proficiency. An imbalance in information-processing skills was also a feature of this group. The majority of this group was identifiable by the combination of depth of coma, and 3 independent clinical factors. Hypoglycaemia is reported in 13134% of cases of severe malaria (BREWSTER etal.. 1990: BONDI. 1992: MARSH et al., 1995‘), and has been associated with sequelae in other encephalopathies (LETSON et&, 1992). Similarly multiple seizures may lead to cognitive impairment in other circumstances and occur in ip to 85% of cases of cerebral malaria (O’LEARY et al., 1981: AICARDI & CHEVRIE. 1983; W~RUIRU et al., 1956). ithough association doe; not prove causality, and both hypoglycaemia and multiple seizures may simply reflect the severity of underlying damage, these data suggest the importance of closely investigating the replicability of the associations found, and the effects of intervention to manage these risk factors. As hyperpyrexia has been associated with poor outcome in other contexts we were surprised at the apparent protective effect in this sample (GINSBERG et al., 1992; VARNEY et aZ., 1994). Although this may be a chance finding, the recent suggestion that measures to lower temperature in malaria may not necessarily be beneficial suggests that this should be examined further (BRANDTS et al., 1997). To conclude, survivors of severe malaria with impaired consciousness are at risk of developing cognitive or behavioural sequelae. When using a definition of impairment focusing on information-processing skills we detected impairment in 14% of our sample. Among these were children who, with support, could be expected to cope with the mainstream school system. The remainder displayed a performance that was both quantitatively and qualitatively different from their peers, and are likely to require more intensive and specialized educational support. Gross neurological impairment on discharge alone was not sufficient to detect a sizeable minority of these children. We identified hypoglycaemia and multiple seizures as potentially important areas where changes in clinical management might result in the reduction of the risk of development of cognitive sequelae. A greater understanding of the relationship between other influences upon a child’s general development and the disease episode is required to identify with greater accuracy children at risk of a poor outcome. J

,

Acknowledgements

This study was supported by KEMRI, the Wellcome Trust, and the Directors Initiative Fund of the UNDP/World BanW WHO Special Programme for Researchand Training in Tropical Diseases (TDR). We thank the assessmentteam, K. Katana, T. Mkala, J. Gona, B. Mwamuye, L. Mbonani, J. Magongo, and especially Sidi Kazungu who produced the materials. Thanks go too to the staff of the KEMRI Unit for their clinical support, and Dr D. Smith for his statistical advice. We are grateful to Dr J. Crawley, Dr C. Newton and Professor K. Connolly for their comments on this article, and particularly to Tore Godal for his support in launching this project. K. Marsh is a Wellcome Trust Senior Research Fellow in Clinical Science (431342). This paperispublishedwiththepermissionoftheDirectorofKEMRI. References

Aicardi, J. & Chevrie, J. J. (1983). Consequences of status epilepticus in infants and children. Advances in Neurology, 34, 115-125.

Baddeley, A., Gardner, J. M. & Grantham-McGregor, S. G.

(1995). Cross-cultural coanition: develouing tests for developingcountries. Applied cognitive Psycho-%& 9, 173- 195. Brandts, C. N., Ndjave, M., Graninger, W. & Kremsner, P. G. (1997). Effect of paracetamol on-parasite clearance time in Plasmodium falciaarum malaria. Lancet. 35. 704-709.

Brewster, D. k., ‘Kwiatkowski, D. & ‘cite, N. J. (1990). Neurological sequelae of cerebral malaria in children. Lancet, 36, 1039-1043:

Bondi, F. S. (1992). The incidence and outcome of neurological abnormalities in childhood cerebral malaria: a long term follow-up of 62 survivors. Transactions of the Royal Society of Tropical Medicine and Hygiene, 86, 17 - 19. Campbell, S. B. (1995). Behaviour problems in pre-school children: a review of recent research. Youma of Child Psvchol* ogy and Psychiatty, 36, 113-149. ” Chadwick, O., Rutter, M., Brown, G., Shaffer, D. & Traub, M. (198 1). A prospective study of children with head injuries: II. Cognitive sequelae. Psychological Medicine, 11,49-61. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159. Damico, J. & Oller, J. W. (1980). Pragmatic versus morphological syntactic criteria for language referral. Language, Speech, and Hearing Services in Schools X1.8,85-94. Elliot, C., Murray, D. J. &Pearson, L. S. (1983). British Ability Scales Revised. Windsor: N.F.E.R. Finger, S. & Almli, C. R. (1988). Margaret Kennard and her ‘principle’ in historical perspective. In: Brain Injuy and Recovery, Theoretical and Controvenial Issues, Finger, S., Levere, T. E., Almli, C. R. & Stein, 0. G. (editors). New York, London: Plenum Press. pp. 117- 13 1. Ginsberg, M. D., Sternau, L. L., Globus, M. Y., Dietrich, W. D. & Busto, R. (1992). Therapeutic modulation of brain temperature: relevance to ischemic brain injury. Cerebrovascular Brain Metabolism Review, 4, 189-225. Greenwood, B., Bradley, A. K., Greenwood, A. M., Byass, P., Jammeh, K., Marsh, K., Tulloch, S., Oldfield, F. S. J. & Hayes, R. (1987). Mortality and morbidity from malaria among children in a rural area of The Gambia, West Africa. Transactions of the Royal Society of Tropical Medicine and Hygiene, 81,478-486. Holding, P. A. (1998). Doescerebral malaria constitute a risk factor e;xcial educational needs?PhD thesis, University of London,

Kaufman, A. S. &Kaufman, N. L. (1983). Kaufman Assessment Batteryfor Children. CirclePines: kneriianG;idance Service. Letson, W. G., Gellin, B. G., Bulhow, L. R., Parks, D. J. & Ward, J. I. (1992). Severity and frequency of sequelae of bacterial meningitis in Alaska native infants. AmericanJournal of the Disabled Child, 146.560-566. L&a, A. R. (1973): The- Working Brain. An Introduction to Neuropsychology. Harmondsworth, UK: Penguin. Marsh, K. (1992). Malaria-a neglected disease? Parasitology, 104, S53-S69. Marsh, K., Forster, D., Waruiru, C., Mwangi, I., Winstanley, M., Marsh, V., Newton, C., Winstanley, P., Warn, P., Peshu, N., Pasvol, G. & Snow, R. (1995). Indicators of life threatening malaria in African children. New England Journal of Medicine, 332, 1399-1404. Molyneux, M. E., Taylor, T. E., Wirima, J. J. & Borgenstein, A.

(1989). Clinical features and prognostic indicators in paediatric cerebral malaria: a study of 131 comatose Malawian children. Quarterly Journal of Medicine, 71, 441-459. Muntendam, A. H., Jaffar, S., Bleichrodt, N. &van Hensbroek, M. B. (1996). Absence of neuropsychological sequelae following cerebral malaria in Gambian children. Transactions of the Royal Society of Tropical Medicine and Hygiene, 90, 391-394. N&&is, M. J. (1992). SPSSfor Windows. Chicago: SPSS Inc. O’Leary, D. S., Seidenberp, M.. Berent. S. & Boll. T. I. (198 1). Effe& of agi of onset oTton;c-clo& seizures bn ie;rops$chological performance in children. Epilepsia, 22, 197-204. Rutter, M. (1984). Developmental Neuropsychiaty. Edinburgh: Churchill Livingstone. Rutter, M., Graham, P. & Yule, W. (1970). A Neuropsychiatric Study ofchildhood. London: Heineman. Satz, P. (1993). Brain reserve capacity on symptom onset after brain injury: a formulation and review of evidence for threshold theory. Neuropsychologv, 7,273-295. Sigman, M., Neuman, C., Jansen, A. A. J. & Bwibo, N. (1989). Cognitive abilities of Kenyan children in relation to nutrition, family characteristics and education. Child Development, 60, 1463-1474. Snow, R. W., Peshu, N., Forster, D., Bomu, G., Mitsanze, E., Ngumbao, E., Chisengwa, R., Armstrong Schellenberg, J. R. M., Hayes, R. J., Newbold, C. I. & Marsh, K. (1998).

534

Environmental and entomological risk factors for the development of clinical malaria among children on the Kenyan coast. Transactions of the Royal Society of Tropical Medicine and Hygiene, 92,381-385.

Taylor, H. G., Mills, E. L., Ciampi, A., du Berger, R., Watters, G. V., Gold, R., MacDonald, N. & Michaels, R. H. (1990). The sequelae of Haemophilus influenzae meningitis in school age children. New England Journal of Medicine, 323, 1657-1663. van Hensbroek, M. B., Palmar, A., Jaffar, S., Schneider, G. & Kwiatkowski, D. (1997). Residual neurologic sequelae after childhood cerebral malaria. 3ourd of Pediatrics, 131, 125-129.

Vamey, N. R., Campbell, D. & Roberts, R. J. (1994). Long-term neuropsychological sequelae of fever associated with amnesia. Archives of Clinical Neuropsycholgy, 9,347-352. Vygotsky, L. S. (1993; originally published in Russian in 1929).

Clinical Infectious Diseases: a Practical Approach. R. K. Root (editor-in-chief), F. Waldvogel, L. Corey, W. E. Stamm (editors). New York and Oxford: Oxford University Press, 1999. xxiv + 1014 pp. Price E99.50. ISBN 0-19-508103-X. This new multi-author infectious diseases textbook is aimed primarily at the practising clinician, but perhaps not specifically the infectious disease specialist. As the preface states, the intention was to produce a book at a level ‘somewhere between general medical texts and more encyclopaedic infectious disease texts’. The 150 chapter authors are predominantly from North America, although there are several from Europe, but few from elsewhere in the world. The first major section of the book is devoted to the essential scientific background of infectious diseases, including the pathophysiology and epidemiology, and also basic details of diagnostic methods and the use of antibiotics. Modern techniques are discussed in detail, and the sections on imaging technology and summarizing molecular diagnostic methods are particularly useful. However, basic bacteriology, virology, mycology and parasitology diagnostic approaches are compressed into 4 very concise chapters. This is not a technical manual, and the approach to the diagnosis of infection is geared towards areas of the world where newer technology is available. The main body of the book is devoted to infectious disease syndromes, with each chapter addressing infection of a particular system or presentation, rather than infection due to specific organisms. Therefore, there is a chapter on infections with rash, infectious diseases with lymphadenopathy, meningitis, acute pneu-

P. A. HOLDING ET-AL.

The fundamentals of defectology. In: The Collected Works of L. S. &‘ygottky, vol. 2, Rieber, R. W. & Carton, A. S. (editors). New York & London: Plenum Press. Waruiru, C. M., Newton, C. R. J. C., Forster, D., New, L., Winstanley, P., Mwangi, I., Marsh, V., Winstanley, M., Snow, R. W. & Marsh, K. (1996). Epileptic seizures and malaria in Kenyan children. Transactions of the Royal Sociey of Tropical Medicine and Hygiene, 90, 152- 155.

WHO (1990). Severe and complicated malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene, 84 (supplement 2). World Bank (1993). Investing in Health. World Development Report. Oxford: Oxford University Press. Received 15 March 1999; revised 14 May i 999; acceptedfor publication 18 May 1999

monia, and so on. There is also a section on infection in specific groups of patients, including transplant recipients, burns and patients infected with HIV, amongst others. This method of organization clearly has the advantage of reflecting how patients present in clinical practice, but makes it difficult to look up an individual organism, as it may be necessary to refer to several different chapters. The sections on fever in tropical countries refer specifically to tropical diseases as they would present in a traveller returning home, and would not be of particular use in countries where these infections are endemic. I enjoyed looking at this book. The layout is excellent, and the text is easy to read. The diagrams are very clear, and there is extensive use of tables to summarize information. Core references are given with annotations from the chapter author. The text and references appeared to be relatively up to date (references from 1997 were the most recent I could find). The editors of this book have achieved their aim of producing a book that will sit in the middle ground between the simple and comprehensive textbooks on infectious diseases, but in some respects this makes it difficult to decide who should buy it. With 1000 pages it is probably too detailed and costly for individual medical students or junior doctors, and it is not detailed enough for the specialist clinician treating infectious diseases every day. I would use it in conjunction with other infection-related textbooks, primarily for the excellent summary tables and diagrams. Nicholas M. Brown Consultant Medical Microbiologist Box 236 Addenbrooke’s NHS Trust Hills Road Cambridge CB2 2Q W, UK