454
Specificity of hypoglycaemia for cerebral malaria
in
children
Glycaemic status on hospital admission was compared in 97 children with severe falciparum malaria (36 with cerebral malaria) and 89 children with other serious illnesses (32 in coma; 57 with acute pneumonia, not in coma). The frequency of hypoglycaemia (blood glucose below 2·2 mmol/l) did not differ significantly between malarial and control patients (5·2% vs 11·2%) nor between the comatose (11·1% vs 18·8%) and conscious (1·6% vs 7·0%) malarial and control subgroups. Compared with normoglycaemic patients, hypoglycaemic patients had appropriately low serum insulin (3·0 vs 8·2 mU/I) and C-peptide (0·13 vs 0·42 mmol/l) and high plasma non-esterified fatty acids (1·42 vs 0·83 mmol/l). Hypoglycaemia, the level of consciousness, and death were all significantly associated with the time since the last meal. Hypoglycaemia is not a specific complication of malaria but is found in severely ill fasted children,
resulting from glycogen depletion and perhaps impaired hepatic gluconeogenesis. It should be sought in all severely sick children. A single bolus dose of glucose may not be enough to correct it. Introduction There
many reports of
hypoglycaemia in severe malaria.l-6 The falciparum hypoglycaemia has been to attributed greater glucose utilisation by the parasitised cellss°’8and/or to quinine therapy.9-12 However, all reported studies have lacked a control group, and it is not clear whether the hypoglycaemia is a specific complication of malaria. To determine the specificity of hypoglycaemia for severe malaria, we investigated glycaemic status, clinical presentation, and outcome of 97 children with severe and complicated malaria and a control group of 89 children admitted with other serious illnesses. are
Patients and methods The study was carried out in children aged 0-7 years admitted to the paediatric wards of Muhimbili Medical Centre, Dar es Salaam, Tanzania. Patients with severe illnesses are usually referred to this centre from hospitals and other health units in and around Dar es Salaam. The study group consisted of 97 children with severe and complicated malaria as defined by the World Health Organisation.13 This group was subdivided into comatose patients with cerebral malaria (n = 36) and children with other forms of severe and complicated malaria (n 61). The control group for cerebral malaria patients consisted of 32 children in coma due to other causes (7 pyogenic meningitis; 14 possible aseptic meningitis; 5 encephalopathy after ingestion of traditional medicines; 6 unknown cause, died before lumbar puncture). The control group for patients with severe non-cerebral malaria consisted of 57 conscious patients with acute lower =
respiratory tract infection. Coma was defmed as a score of 4 or less on the modified Glasgow Coma Scale for children.14 Children were recruited into the study only if the nutritional status was normal according to the Wellcome Trust International Working Party. 15 Other reasons for exclusion were quinine treatment or a glucose infusion before admission. All children admitted between May 1 and Nov 30,1989, were included if they met the above criteria. After informed consent had been given by parent or guardian, a detailed history was taken and a thorough physical examination carried out. The history included duration of illness, time since the last meal, and drugs taken before admission. Blood for examination for malaria parasites and estimation of blood glucose and intermediary metabolite (3-hydroxybutyrate, lactate, alanine, and glycerol) concentrations, serum insulin and C-peptide, and plasma non-esterified fatty acids was taken on admission and before treatment was started. Two thick and two thin blood smears were taken from each patient and stained for malaria parasites with giemsa stain. Malaria parasite densities were expressed as the number of trophozoites per 200 white blood cells. Lumbar puncture was done on all comatose or convulsing children. Management of the children with severe malaria was as described by Msengi and Y ohani.16 All children were given intravenous quinine 10 mg/kg in 5 % dextrose 10 ml/kg every 8 h for at least 3 days. Hypoglycaemia, defmed as a blood glucose concentration below 2-2 mmol/1, was treated with an intravenous bolus of 50% glucose (1 ml/kg). Venous whole blood glucose was estimated by the glucose oxidase method with a Yellow Springs glucose analyser (Ohio, USA). Blood samples for hormones and metabolite assays, after centrifugation and separation, were stored at - 70°C and flown still frozen to Newcastle upon Tyne, UK. Insulin was measured by double-antibody radioimmunoassay 17 and C-peptide by means of the Novo kit.18 Plasma non-esterified fatty acids and metabolites" were assayed enzymically with a ’Cobas Bio’ centrifugal analyser
(Roche). The results were analysed and compared by means of the chi-square test, Kendall’s correlation coefficient, the MannWhitney test, and discriminant analysis, with the Statistical Package for the Social Sciences.2o A stepwise discriminant analysis was used to define factors associated with hypoglycaemia, coma, and survival. All the factors listed in table i were entered into the analysis. For age, duration of illness, time since the last meal, serum bilirubin, blood lactate, blood alanine, blood glycerol, blood 3-hydroxybutyrate, plasma non-esterified fatty acids, temperature, and haemoglobin, natural logarithms were used in the analysis, and for serum alanine aminotransferase activity and weight (as percentage of the median for height, by US standards in the CASP anthropometric software) reciprocals were used. The actual values of these variables were not
ADDRESSES. Departments of Paediatrics (N G. Kawo, MD, Prof A. E Msengi, MMed), Medicine (A B M Swai, MMed, Prof D G McLarty, FRCP), and Clinical Chemistry (L. M. Chuwa, MD), Muhimbili Medical Centre, University of Dar es Salaam, Tanzania, and Department of Medicine, University of Newcastle upon Tyne, UK (Prof K G M M Alberti, FRCP) Correspondence to Prof K. G M. M Alberti, Department of Medicine, University of Newcastle upon Tyne, Floor 4, William Leech Building, Framlington Place, Newcastle upon Tyne NE2 4HH, UK
455
TABLE I-CLINICAL AND BIOCHEMICAL DATA
FA =fatty acids. By Mann-Whitney test difference between
malaria and control groups
significant at *p < 0 001, tp
Specificity of hypoglycaemia for cerebral malaria
in
children
Glycaemic status on hospital admission was compared in 97 children with severe falciparum malaria (36 with cerebral malaria) and 89 children with other serious illnesses (32 in coma; 57 with acute pneumonia, not in coma). The frequency of hypoglycaemia (blood glucose below 2·2 mmol/l) did not differ significantly between malarial and control patients (5·2% vs 11·2%) nor between the comatose (11·1% vs 18·8%) and conscious (1·6% vs 7·0%) malarial and control subgroups. Compared with normoglycaemic patients, hypoglycaemic patients had appropriately low serum insulin (3·0 vs 8·2 mU/I) and C-peptide (0·13 vs 0·42 mmol/l) and high plasma non-esterified fatty acids (1·42 vs 0·83 mmol/l). Hypoglycaemia, the level of consciousness, and death were all significantly associated with the time since the last meal. Hypoglycaemia is not a specific complication of malaria but is found in severely ill fasted children,
resulting from glycogen depletion and perhaps impaired hepatic gluconeogenesis. It should be sought in all severely sick children. A single bolus dose of glucose may not be enough to correct it. Introduction There
many reports of
hypoglycaemia in severe malaria.l-6 The falciparum hypoglycaemia has been to attributed greater glucose utilisation by the parasitised cellss°’8and/or to quinine therapy.9-12 However, all reported studies have lacked a control group, and it is not clear whether the hypoglycaemia is a specific complication of malaria. To determine the specificity of hypoglycaemia for severe malaria, we investigated glycaemic status, clinical presentation, and outcome of 97 children with severe and complicated malaria and a control group of 89 children admitted with other serious illnesses. are
Patients and methods The study was carried out in children aged 0-7 years admitted to the paediatric wards of Muhimbili Medical Centre, Dar es Salaam, Tanzania. Patients with severe illnesses are usually referred to this centre from hospitals and other health units in and around Dar es Salaam. The study group consisted of 97 children with severe and complicated malaria as defined by the World Health Organisation.13 This group was subdivided into comatose patients with cerebral malaria (n = 36) and children with other forms of severe and complicated malaria (n 61). The control group for cerebral malaria patients consisted of 32 children in coma due to other causes (7 pyogenic meningitis; 14 possible aseptic meningitis; 5 encephalopathy after ingestion of traditional medicines; 6 unknown cause, died before lumbar puncture). The control group for patients with severe non-cerebral malaria consisted of 57 conscious patients with acute lower =
respiratory tract infection. Coma was defmed as a score of 4 or less on the modified Glasgow Coma Scale for children.14 Children were recruited into the study only if the nutritional status was normal according to the Wellcome Trust International Working Party. 15 Other reasons for exclusion were quinine treatment or a glucose infusion before admission. All children admitted between May 1 and Nov 30,1989, were included if they met the above criteria. After informed consent had been given by parent or guardian, a detailed history was taken and a thorough physical examination carried out. The history included duration of illness, time since the last meal, and drugs taken before admission. Blood for examination for malaria parasites and estimation of blood glucose and intermediary metabolite (3-hydroxybutyrate, lactate, alanine, and glycerol) concentrations, serum insulin and C-peptide, and plasma non-esterified fatty acids was taken on admission and before treatment was started. Two thick and two thin blood smears were taken from each patient and stained for malaria parasites with giemsa stain. Malaria parasite densities were expressed as the number of trophozoites per 200 white blood cells. Lumbar puncture was done on all comatose or convulsing children. Management of the children with severe malaria was as described by Msengi and Y ohani.16 All children were given intravenous quinine 10 mg/kg in 5 % dextrose 10 ml/kg every 8 h for at least 3 days. Hypoglycaemia, defmed as a blood glucose concentration below 2-2 mmol/1, was treated with an intravenous bolus of 50% glucose (1 ml/kg). Venous whole blood glucose was estimated by the glucose oxidase method with a Yellow Springs glucose analyser (Ohio, USA). Blood samples for hormones and metabolite assays, after centrifugation and separation, were stored at - 70°C and flown still frozen to Newcastle upon Tyne, UK. Insulin was measured by double-antibody radioimmunoassay 17 and C-peptide by means of the Novo kit.18 Plasma non-esterified fatty acids and metabolites" were assayed enzymically with a ’Cobas Bio’ centrifugal analyser
(Roche). The results were analysed and compared by means of the chi-square test, Kendall’s correlation coefficient, the MannWhitney test, and discriminant analysis, with the Statistical Package for the Social Sciences.2o A stepwise discriminant analysis was used to define factors associated with hypoglycaemia, coma, and survival. All the factors listed in table i were entered into the analysis. For age, duration of illness, time since the last meal, serum bilirubin, blood lactate, blood alanine, blood glycerol, blood 3-hydroxybutyrate, plasma non-esterified fatty acids, temperature, and haemoglobin, natural logarithms were used in the analysis, and for serum alanine aminotransferase activity and weight (as percentage of the median for height, by US standards in the CASP anthropometric software) reciprocals were used. The actual values of these variables were not
ADDRESSES. Departments of Paediatrics (N G. Kawo, MD, Prof A. E Msengi, MMed), Medicine (A B M Swai, MMed, Prof D G McLarty, FRCP), and Clinical Chemistry (L. M. Chuwa, MD), Muhimbili Medical Centre, University of Dar es Salaam, Tanzania, and Department of Medicine, University of Newcastle upon Tyne, UK (Prof K G M M Alberti, FRCP) Correspondence to Prof K. G M. M Alberti, Department of Medicine, University of Newcastle upon Tyne, Floor 4, William Leech Building, Framlington Place, Newcastle upon Tyne NE2 4HH, UK
455
TABLE I-CLINICAL AND BIOCHEMICAL DATA
FA =fatty acids. By Mann-Whitney test difference between
malaria and control groups
significant at *p < 0 001, tp
