1039
REVIEW ARTICLE Neurological sequelae of cerebral
Out of 604 Gambian children admitted with falciparum malaria to one hospital between September and December, 1988, 308 had cerebral malaria and 203 were severely anaemic
(haemoglobin < 60 g/)). 14% of those with cerebral malaria died, as did 78% of those with severe anaemia. 32 (12%) of children surviving cerebral malaria had residual neurological deficit. 69 other children were admitted with clinical features strongly suggestive of cerebral malaria but with negative blood films; 16 of these died and 3 had residual neurological deficits. The commonest sequelae of cerebral malaria were hemiplegia (23 cases), cortical blindess (11), aphasia (9), and ataxia (6). Factors predisposing to sequelae included prolonged coma, protracted convulsions, severe anaemia, and a biphasic clinical course characterised by recovery of consciousness followed by recurrent convulsions and coma. At follow up 1-6 months later over half these children had made a full recovery, but a quarter were left with a major residual neurological deficit. Cerebral malaria in childhood may be an important cause of neurological handicap in the tropics. Lancet 1990; 336: 1039-43, Introduction
Falciparum malaria is a major cause of childhood morbidity and mortality in the tropics. In The Gambia, West Africa, falciparum malaria causes approximately a quarter of all deaths between the ages of 1 and 4 years1,2 and is responsible for over half the paediatric hospital admissions during the rainy season. Malaria transmission is becoming unstable in The Gambia and the commonest clinical presentation of severe Plasmodiumfalciparum infection is cerebral malaria in childhood. In contrast to other causes of coma in infection, cerebral malaria carries a good neurological prognosis for survivors; a recent book on malaria carries the statement that and mental defects are very rare" The impression that residual neurological deficit rarely follows recovery of consciousness in cerebral malaria derives largely from observations in adults. Although neurological sequelae have been noted to follow childhood cerebral malaria,4-18 in most series details of long-term outcome have been lacking. This is an important issue because cerebral malaria is such a common disease in the tropics, and neurological handicap imposes a
"neurological sequelae following recovery.33
malaria in children
considerable burden on affected individuals, their families, and the community. We describe a prospective hospitalbased study on the incidence and outcome of residual neurological deficit following cerebral malaria in Gambian children.
Patients and methods
Study site This study was conducted in the 75-bed paediatric wing of the Royal Victoria Hospital, which is located in the capital Banjul and functions as a provincial hospital for Western Gambia, with a catchment population of 128 000 children aged under 10 years (approximately half the population of the country).19 There are fifteen health centres or clinics within this region. Approximately 95% of paediatric admissions come from this catchment area. Children with febrile illnesses are given chloroquine by village health workers or staff at dispensaries or clinics. Patients with severe malaria are then referred to hospital, but in some cases parasitaemia may have cleared by the time of admission. The data reported here were collected prospectively by the paediatrician (D.B.) during the rainy season, between Sept 1 and Dec 31, 1988. This study was not a funded research project, so the information was obtained within the usual constraints of a busy routine clinical service in Africa.
Patients Cerebral malaria was diagnosed clinically on the basis of a febrile illness with convulsions and unrousable coma lasting more than 30 min after seizures had stopped, other causes having been excluded. Thus children with febrile convulsions associated with malaria were not included as cerebral malaria. Malaria was deemed to be "definite" if asexual forms of P falciparum were found on the peripheral blood smear, and the clinical presentation was compatible with the diagnosis (P falciparum accounts for over 99% of malaria admissions in this hospital). Malaria was classified as probable if, despite the absence of asexual stages in the peripheral blood smear, the clinical features were suggestive and there was a clear history of previous antimalarial treatment within the previous 7 days (malarial pigment was often seen in peripheral blood monocytes and polymorphonuclear leucocytes in these cases). Malaria parasitaemia on thick Giemsa stained blood films was graded according to a standardised scale20 whereby 1 + represents 1-10 malaria parasites per/ 100 high power fields (HPF; x 1000) and ADDRESSES. Royal Victoria Hospital, Banjul, The Gambia (D. R Brewster, FRACP); Medical Research Council Laboratories, Fajara, The Gambia (D Kwiatkowski, MRCP); and Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand, and Nuffield Department of Clinical Medicine, University of Oxford, UK (N J White, FRCP). Correspondence to Dr David Brewster, PO Box 345, Banjul, The Gambia.
1040
Ayt;
Fig
III
ytjClI
1-Age distribution of cerebral malaria and
4 + refers
to more
than 10
severe
anaemia.
parasites/HPF. (1parasite/HPF
=
500
parasites/ul.) Treatment All patients were treated initially with subcutaneous chloroquine phosphate 4 mg base/kg 6 hourly. This was changed to 10 mg base/kg daily orally once the child regained conciousness. The total was at least 25 mg base/kg. Convulsions treated with paraldehyde 015 ml/kg by intramuscular injection or diazepam 0’4mg/kg rectally. Intravenous fluids were used only sparingly because of nursing staff shortages, so many comatose children received small volumes of fluid by nasogastric tube until they were able to drink. If the blood glucose was 22 mmol/1 or less on glucose
dose
were
oxidase stick testing (BM test: Boehringer Mannheim), intravenous 50% dextrose was given (O’5-l ml/kg) and any response noted. Antibiotics (ampicillin or chloramphenicol 100 mg/kg/day) were reserved for children with abnormal chest signs suggestive of aspiration pneumonia, or a relapse of fever after parasite clearance (which occurred in less than 5% of cases).
Assessment The level of consciousness was assessed according to the modified Glasgow coma scale for children.18,21 The severity of convulsions was graded on a 0-5 scale based on the number of anticonvulsant doses administered. Neurological sequelae were defmed as gross motor defects, large visual field defects, or inability to talk upon regaining consciousness lasting for at least 24 h. Transient or subtle
defects were not included and would not have been detected in all cases since children with cerebral malaria were usually discharged soon after regaining consciousness. Sickle-cell anaemia was excluded in all cases with neurological sequelae by thin blood film examination and sickle test. The mothers of all children with residual neurological abnormalities were asked to bring their children back to the hospital for outpatient follow-up. Children who did not turn up for follow-up were visited at home if they lived within a 40 km radius of Banjul. As part of a separate study on severe malaria, blood samples were taken for full blood count, thick film, urea, creatinine, bilirubin, and blood glucose in 193 children (62%) with cerebral malaria.
admissions. 308 children were judged to have definite (slide-positive) malaria and 69 to have "probable" (treated, slide-negative) cerebral malaria. Thus cerebral malaria accounted for 27% of all paediatric admissions and 51 % of all malaria cases admitted during the study period. Severe anaemia (Hb < 60 g/1) was recorded in 248 (33%) of the malaria cases and included 56 children with cerebral malaria. The mean age of children with severe anaemia was approximately half that of children with cerebral malaria (p
REVIEW ARTICLE Neurological sequelae of cerebral
Out of 604 Gambian children admitted with falciparum malaria to one hospital between September and December, 1988, 308 had cerebral malaria and 203 were severely anaemic
(haemoglobin < 60 g/)). 14% of those with cerebral malaria died, as did 78% of those with severe anaemia. 32 (12%) of children surviving cerebral malaria had residual neurological deficit. 69 other children were admitted with clinical features strongly suggestive of cerebral malaria but with negative blood films; 16 of these died and 3 had residual neurological deficits. The commonest sequelae of cerebral malaria were hemiplegia (23 cases), cortical blindess (11), aphasia (9), and ataxia (6). Factors predisposing to sequelae included prolonged coma, protracted convulsions, severe anaemia, and a biphasic clinical course characterised by recovery of consciousness followed by recurrent convulsions and coma. At follow up 1-6 months later over half these children had made a full recovery, but a quarter were left with a major residual neurological deficit. Cerebral malaria in childhood may be an important cause of neurological handicap in the tropics. Lancet 1990; 336: 1039-43, Introduction
Falciparum malaria is a major cause of childhood morbidity and mortality in the tropics. In The Gambia, West Africa, falciparum malaria causes approximately a quarter of all deaths between the ages of 1 and 4 years1,2 and is responsible for over half the paediatric hospital admissions during the rainy season. Malaria transmission is becoming unstable in The Gambia and the commonest clinical presentation of severe Plasmodiumfalciparum infection is cerebral malaria in childhood. In contrast to other causes of coma in infection, cerebral malaria carries a good neurological prognosis for survivors; a recent book on malaria carries the statement that and mental defects are very rare" The impression that residual neurological deficit rarely follows recovery of consciousness in cerebral malaria derives largely from observations in adults. Although neurological sequelae have been noted to follow childhood cerebral malaria,4-18 in most series details of long-term outcome have been lacking. This is an important issue because cerebral malaria is such a common disease in the tropics, and neurological handicap imposes a
"neurological sequelae following recovery.33
malaria in children
considerable burden on affected individuals, their families, and the community. We describe a prospective hospitalbased study on the incidence and outcome of residual neurological deficit following cerebral malaria in Gambian children.
Patients and methods
Study site This study was conducted in the 75-bed paediatric wing of the Royal Victoria Hospital, which is located in the capital Banjul and functions as a provincial hospital for Western Gambia, with a catchment population of 128 000 children aged under 10 years (approximately half the population of the country).19 There are fifteen health centres or clinics within this region. Approximately 95% of paediatric admissions come from this catchment area. Children with febrile illnesses are given chloroquine by village health workers or staff at dispensaries or clinics. Patients with severe malaria are then referred to hospital, but in some cases parasitaemia may have cleared by the time of admission. The data reported here were collected prospectively by the paediatrician (D.B.) during the rainy season, between Sept 1 and Dec 31, 1988. This study was not a funded research project, so the information was obtained within the usual constraints of a busy routine clinical service in Africa.
Patients Cerebral malaria was diagnosed clinically on the basis of a febrile illness with convulsions and unrousable coma lasting more than 30 min after seizures had stopped, other causes having been excluded. Thus children with febrile convulsions associated with malaria were not included as cerebral malaria. Malaria was deemed to be "definite" if asexual forms of P falciparum were found on the peripheral blood smear, and the clinical presentation was compatible with the diagnosis (P falciparum accounts for over 99% of malaria admissions in this hospital). Malaria was classified as probable if, despite the absence of asexual stages in the peripheral blood smear, the clinical features were suggestive and there was a clear history of previous antimalarial treatment within the previous 7 days (malarial pigment was often seen in peripheral blood monocytes and polymorphonuclear leucocytes in these cases). Malaria parasitaemia on thick Giemsa stained blood films was graded according to a standardised scale20 whereby 1 + represents 1-10 malaria parasites per/ 100 high power fields (HPF; x 1000) and ADDRESSES. Royal Victoria Hospital, Banjul, The Gambia (D. R Brewster, FRACP); Medical Research Council Laboratories, Fajara, The Gambia (D Kwiatkowski, MRCP); and Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand, and Nuffield Department of Clinical Medicine, University of Oxford, UK (N J White, FRCP). Correspondence to Dr David Brewster, PO Box 345, Banjul, The Gambia.
1040
Ayt;
Fig
III
ytjClI
1-Age distribution of cerebral malaria and
4 + refers
to more
than 10
severe
anaemia.
parasites/HPF. (1parasite/HPF
=
500
parasites/ul.) Treatment All patients were treated initially with subcutaneous chloroquine phosphate 4 mg base/kg 6 hourly. This was changed to 10 mg base/kg daily orally once the child regained conciousness. The total was at least 25 mg base/kg. Convulsions treated with paraldehyde 015 ml/kg by intramuscular injection or diazepam 0’4mg/kg rectally. Intravenous fluids were used only sparingly because of nursing staff shortages, so many comatose children received small volumes of fluid by nasogastric tube until they were able to drink. If the blood glucose was 22 mmol/1 or less on glucose
dose
were
oxidase stick testing (BM test: Boehringer Mannheim), intravenous 50% dextrose was given (O’5-l ml/kg) and any response noted. Antibiotics (ampicillin or chloramphenicol 100 mg/kg/day) were reserved for children with abnormal chest signs suggestive of aspiration pneumonia, or a relapse of fever after parasite clearance (which occurred in less than 5% of cases).
Assessment The level of consciousness was assessed according to the modified Glasgow coma scale for children.18,21 The severity of convulsions was graded on a 0-5 scale based on the number of anticonvulsant doses administered. Neurological sequelae were defmed as gross motor defects, large visual field defects, or inability to talk upon regaining consciousness lasting for at least 24 h. Transient or subtle
defects were not included and would not have been detected in all cases since children with cerebral malaria were usually discharged soon after regaining consciousness. Sickle-cell anaemia was excluded in all cases with neurological sequelae by thin blood film examination and sickle test. The mothers of all children with residual neurological abnormalities were asked to bring their children back to the hospital for outpatient follow-up. Children who did not turn up for follow-up were visited at home if they lived within a 40 km radius of Banjul. As part of a separate study on severe malaria, blood samples were taken for full blood count, thick film, urea, creatinine, bilirubin, and blood glucose in 193 children (62%) with cerebral malaria.
admissions. 308 children were judged to have definite (slide-positive) malaria and 69 to have "probable" (treated, slide-negative) cerebral malaria. Thus cerebral malaria accounted for 27% of all paediatric admissions and 51 % of all malaria cases admitted during the study period. Severe anaemia (Hb < 60 g/1) was recorded in 248 (33%) of the malaria cases and included 56 children with cerebral malaria. The mean age of children with severe anaemia was approximately half that of children with cerebral malaria (p
