TRANSACTIONSOFTHEROYALSOCIETYOFTROPICALMEDICINEANDHYGIENE(1999)93,616-618
Plasma glutamine
levels and falciparum
malaria
G. Cowan’, T. Planche2, T. Agbenyega3, G. Bedu-Addo3, A. Owusu-OforP, J. Adebe-Appiah3, D. Agranoff 2, C. Woodrow*, L. Castell“, B. Elford’ and S. Krishna2 ‘Department of Molecular Parasitology, Institute of Molecular Medicine, John Radclize Hospital, Oxford OX3 9DU, UK; 2Department of Infectious Diseases, St George’s Hospital Medical Teaching Hospital, Kumasi, School, Cranmer Terrace, London S W17 ORE, UK; 3Department of Child Health, Komfo-Anokye Ghana; 4Department of Biochemisty, University of Oxford, Oxford, UK
Abstract Glutamine deficiency is associated with increased rates of sepsis and mortality, which can be prevented by glutamine supplementation. Changes in glutamine concentration were examined in Ghanaian children with acute falciparum malaria and control cases. The mean (SD) plasma glutamine concentration was lower in patients with acute malaria (401 (82) prnol/L, n = 50) than in control patients (623 (67) pmol/L, n = 7; P < 0.001). Plasma glutamine concentrations all rose in convalescence. The mean (SD) increase in plasma glutamine was 202 (123) pmol/L (n = 18; P < 0.001) compared with acute infection. We conclude that acute falciparum malaria is associated with large decreases in plasma glutamine and these falls may increase susceptibility to sepsis and dyserythropoeisis. Keywords:
malaria, l’lmmodium falciparum, children, glutamine, Ghana
Introduction Glutamine is the most abundant circulating amino acid and becomes conditionally essential in catabolic states. It is an essential substrate for rapidly dividing cells, leucocytes and reticulocytes (CASTELL et al., 1994). In recent studies, glutamine supplementation in critically ill patients with low plasma glutamine concentrations decreased the incidence of sepsis and late mortality (GRIFFITHS, 1997; NEU et al., 1997). Although metabolic complications are increasingly recognized as being important factors in the pathophysiology of severe and complicated malaria infections (NEMTON & KRISHNA, 1998), plasma glutamine concentrations have not been measured in acute malaria infections. We therefore designed this prospective study to test the hypotheses that plasma glutamine concentrations fall significantly during acute malaria and that these falls may be associated with other complications of infection such as hyperlactataemia and anaemia. Methods Parents or guardians gave informed consent before children were Recruited. The study was approved by the Ethical Review Committee of the School of Medical Sciences, University of Science and Technology, Kumasi, Ghana. Patients
Children (aged l-10 years) admitted between September and November 1996 to the Department of Child Health, Komfo-Anokye Teaching Hospital, Kumasi, Ghana, with a suspected diagnosis of malaria were screened by history, examination and measurement of blood lactate, glucose, haematocrit and blood-film examination. Severe malaria was defined as > 10 OOO/pL of asexual stages of Plasmodium falciparum and 1 or more of the following: Blantyre Coma Score (BCS) 10 OOO/& without any features of severity. Children with severe malaria were treated with parenteral quinine and those with nonsevere malaria were treated with artesunate. Patients were discharged when clinically well and aparasitaemic, and attended for follow-up between 6 and 16 days after admission when blood was taken for glutamine assay, haematocrit and blood-film examination. In patients Address for correspondence: Dr Sanjeev Krishna, Department of Infectious Diseases, St George’s Hospital Medical School, Cranmer Terrace, London SW 17 ORE, UK; phone $44 (0) 18 1 725 5836, fax +44 (0)181725 3487, e-mail
[email protected]
who attended follow-up more than once, the final value ofhaematocrit was used in calculations. Control patients were malaria parasite-negative children (aged 1- 10 years) awaiting routine minor surgery (e.g., repair of inguinal hernia) during the time of the study. Laborato y analyses
Heparinized blood was collected on ice, and plasma was separated within 30 min, stored at -80°C and transported on dry ice. Glucose and lactate were measured on site (YSI Instrument Co. Inc., Yellow Springs, OH, USA). Plasma samples (300 a) were deproteinized with perchloric acid (300 pL) and the supernatants adjusted to pH 7 with 20% KOH. Glutamine levels were measured spectrophotometrically (WINDMUELLER & SPAETH, 1974) by mixing plasma extract (100 ,ttL), notassium uhosnhate uH 8 (50 umoles). NADH TO.16 umoles), givcerol ?7.5%), BSA 0.05%, L-glutamate ’ dehydrogenase (10 U/assay), 2-oxolglutarate (3.75 umoles) in a total volume of 1.1 mL. Glutamine was determined from decrease in absorbance at 340 nm after the addition of asparaginase (2 U/assay) and changes in absorbance were corrected for parallel changes in a reagent blank. The intraassay coefficients of variation at 300~U,M, 400~us and 600~UM elutamine were respectively 2:5%, 1.3% and 1.2% and mterassay coefficients of variation at 300~~JV and 40O-p,~ glutamine were respectively 3.5% and 3.6%. analyses Statistical analyses were carried out using SYSTAT (version 5.2, Systat) or Stata Statistical Software (Release 5.0 College Station). After checking for distribution assumptions by the Shapiro-Wilks W test and transforming the data logarithmically or by Box-Cox transformations, as appropriate, we analysed normally distributed data by 2-tailed Student’s t test and nonU statistic. parametric data with the Mann-Whitney Proportions were compared with Fisher’s exact test and correlations were assessed by analysis of Pearson’s product moment coefficient. Statistical
Results Of the 160 children screened during this period, 106 had malaria and 54 had other acute illnesses. Glutamine was assayed in 50 parasitaemic children and in 18 children who attended for follow-up, as well as in 7 control subjects. Ten children with severe malaria were studied (there were no fatalities). The baseline characteristics of the patients are given in the Table. The children who were screened for malaria and the subset admitted to this study were comparable for demographic, clinical and laboratory variables. The baseline characteristics of children who subsequently attended for
PLASMA GLUTAMINE
Table.
Clinical
IN FALCIPARUM
and laboratory
MALARIA
617
characteristics
Variable Demographic Age (months) Male:female Weight (kg) Height (cm) Clinical Pulse (per min) Arterial pressure, mean (SD) (mmHg) Respirations (breaths per min) Temperature (“C) Spleen (cm below costal margin) Blantyre coma score ~2 (%) Laboratory Haematocrit (%) Parasitaemia, geometric mean (/pL) Lactate (mmol/L) Glucose (mmol/L)
at baseline
All patients screened with malaria n= 106 45.5 (31-63) 61:45 13 (11.4-17.0) [?Z= 851 97(89-106) [n=92] 140(126-148) 67 (7.9) [n = 791 38(32-48) 38.5 (37.9-39.2) 0 (O-3) 4.7% 30(25-35)
108 930 2.8 (2.1-4.0) 5.3 (4.1-6.3)
[n=95] [n=95]
of the Ghanaian
children
with malaria
Study patients All study
Subset with
patients n = 50
severe malaria n= 10
48(34-66) 26~24 14.25 (11.4-17.0) 99(90-111)
37(36-60) 5:5
12.5 (11.8-17) 96.5 (93-108)
140(128-150) 70 (8.3) [x=45] 38(30-46) 38.5 (38.0-39.3)
1 (O-3) 8%
Subset who later
attended follow-up It= 18 50.5 (36-72) 9:9 14.75 (12-175) 103(92-114)
135(128S152) 73 (9.1) 32(28-401 38.5 (+6.2-3i.8) 3 (O-4) 40%
134(130-148) 73 (8.5) 33(3biO) 38.5 (38-39.6) 0 (O-2)
25(21-35) 136900 5.15 (3.3-5.15) 5.55 (4.4-7.1)
30 (27.5-33) 95780 2.9 (2.4-4.3) 5.3 (4.2-7.5)
30(24-34) 132390 3.0 (2.2-4.1) 5.5 (4.4-6.8)
0%
All values median (IQR) unless otherwise indicated.
follow-up were also comparable to the study group as a whole (Table). The 5 girls and 2 boys comprising the control group, with a median (range) age of 72 months (l&96), had a median (range) haematocrit value of 35% (25-42). Plasma glutamine concentrations and malaria Figure 1 displays plasma glutamine concentrations in children with acute malaria, convalescent children and surgical control cases. The mean (SD) plasma glutamine concentration was low in acute malaria (401 and rose in convalescence (585 (82) POW (104) ,umol/L) (a 46% increase). There was no relationship between the day number when follow-up samples were obtained and the magnitude of rise in plasma glutamine, suggesting that glutamine levels during convalescence were representative of levels in health. Paired convalescent values were all higher than those measured during acute malaria with a mean (SD) rise in plasma glutamineof202(123) wol/L(n= 18;P
