776

PATHOPHYSIOLOGICAL AND PROGNOSTICSIGNIFICANCE OF CEREBROSPINAL-FLUID

LACTATE IN CEREBRAL MALARIA

NICHOLAS J. WHITESORNCHAI LOOAREESUWAN

RODNEY E. PHILLIPS

DAVID A. WARRELLPORNTHEP CHANTHAVANICH

PRANEET PONGPAEW

Faculty of Tropical Medicine, Mahidol University, Bangkok,Thailand; Tropical Medicine Unit, Nuffield Department of Clinical

Medicine, University of Oxford; and Department of TropicalMedicine, Liverpool School of Tropical Medicine, Liverpool

Summary Cerebrospinal-fluid (CSF) lactate concen-trations were elevated in all but 1 of 45

patients with cerebral malaria. They were significantlyhigher in patients who died (9·0±5·3 mmol/l, mean±SD)than in survivors (3·4±1·1 mmol/l, p=0·0002) and hadreturned to normal values in each of 9 patients studied afterrecovery of consciousness. There was a significant negativecorrelation between CSF lactate and CSF glucose. All 11patients with CSF lactate concentrations above 6 mmol/ldied. CSF lactate is thus an important prognostic indicator incerebral malaria and these findings suggest that hypoxiacontributes to the pathogenesis of this disorder.

Introduction

CEREBRAL malaria is the commonest clinical manifestationof severe falciparum malaria and is fatal in approximately20% of cases.’ Death results from several pathologicalprocesses which commonly coexist in the same individual.These include acute pulmonary oedema, acute renal failure,bacterial superinfection, and metabolic acidosis. The

pathogenesis of coma in severe falciparum malaria isunknown and, although a variety of hypotheses have beenproposed, none has been confirmed in man. Cerebral oedemaresulting from a diffuse increase in cerebral capillarypermeability is no longer accepted as a consistent finding antemortem.2 Disturbances in cerebral microcirculatory flowcaused by mechanical changes in parasitised red cells3 andadhesion to capillary endothelium4 may well be important,but this has not been studied in man. Reduction in cerebral

perfusion would obviously have important physiologicalconsequences because of the brain’s high energy require-ments. We have measured the concentrations of lactate in the

cerebrospinal fluid (CSF) of patients with cerebral malaria asan indication of anaerobic cerebral glycolysis and have corre-lated these observations with clinical state and outcome.

Methods

Patients with severe falciparum malaria were admitted to theintensive-care unit of Pra Pokklao hospital, Chantaburi, EasternThailand. Cerebral malaria was diagnosed in patients with provedfalciparum malaria who were unrousably comatose, had not had anepileptic seizure in the previous 6 h, and in whom other causes ofcoma had been excluded. Children under 6 were not included. The

investigation and management of such patients has been describedpreviously. 1 All were treated with intravenous quininedihydrochlonde as soon as possible: 20 mg/kg loading dose (16. 7mg/kg base) infused over 4 h, followed by 10 mg/kg at 8 h intervals. 5Lumbar puncture was performed within 6 h of admission. CSFopening pressure was measured with the patient lying on his side.CSF was taken into plain plastic tubes for culture, microscopy,protein, and lactate estimation. A sample for glucose measurementwas taken into a fluoride oxalate tube. Simultaneous venous blood

samples were taken for glucose estimation. Measurement of arterial

blood gases and pH was possible in only a few cases. Samples forglucose and lactate estimation were frozen rapidly and stored at- 70°C until analysed. 9 patients consented to repeat lumbarpuncture after they had fully regained consciousness and wereconvalescing. Standard enzymatic methods were used to measurelactate (lactate dehydrogenase reaction, Boehringer MannheimDiagnostica) and glucose (hexokinase reaction, Roche Diagnostica).The samples were analysed without knowledge of the clinical

details. Student’s two-tailed t test was used to compare normallydistributed data.

Results

Patients

45 patients (29 male, 16 female) aged between 8 and 60years (mean 23 - 2 years) were studied. 15 patients died. Deathwas principally attributed to pulmonary oedema (5),pneumonia (4), acute renal failure (3), septicaemia (1), severemetabolic acidosis (1), and unknown causes (1). Smears madefrom post-mortem needle-biopsy specimens of the brainshowed the classical appearances of cerebral malaria in all 15cases. The duration of coma before lumbar puncture was1-70 h (mean 16. 2 h). Coma was graded as describedpreviously;! 39 patients were grade 2 (non-purposive, non-localising response to pain), 5 were grade 3 (no response topain), and 1 was grade 4 (totally unresponsive and areflexic).21 patients had convulsions, before lumbar puncture in 14cases and after lumbar puncture in 7. 5 patients had morethan one seizure, and in 1 case seizures were sustained and

repeated. The time between lumbar puncture and death inthe 15 fatal cases ranged from 2 to 63 h (median 11 h).

Laboratory Results

Parasite counts varied between 112 and 1071 520

parasitised erythrocytes/1 blood (geometric mean 16 109/µl).There was a significant difference between the geometricmean parasite counts of patients who died and survivors(75 992/lll and 10 189/µl respectively, p=0-05). The overallhaematocrit on admission was 26.3±8.5% (mean±SD);there was no difference between patients who died

(24 .3±9 . 7%) and survivors (27 - .3±7 . 8%). Serum creatininevalues on admission were significantly higher in patients whodied than in survivors (3.75±2.52 mg/dl and 2-12±2-08mg/dl respectively [331±223 µmol/l and 187±184 µmol/l],p = 0 - 026). Arterial blood gases were measured in 5 patientsand were normal; p02 ranged from 77 to 100 mm Hg (10. 0and 13.3 3 kPa), and pC02 varied between 33 and 42 mm Hg(3 - 3 and 5 - 6 kPa) after correction for body temperature.Arterial pH was abnormal (7-22) in 1 of these patients whowas in renal failure but was in the normal range in the other 4.Arterial lactate and glucose concentrations were measured atthe time of lumbar puncture in 4 of the survivors and 2

patients who died; lactate concentrations were below 4-0 0mmol/I in each of the survivors and glucose values variedbetween 5 and 8 mmol/l. Arterial lactate concentrations in the2 patients who later died were 4. 6 and 7. 6 mmol/l, andarterial glucose concentrations were 7. 8 and 12 . 3 mmolllrespectively. Hypoglycaemia developed in 5 patients, 3 ofwhom died, but venous blood glucose values were low in only1 patient at the time of lumbar puncture (CSF/venous bloodglucose was 0-42/2-12 mmol/1, and CSF lactate was 6.3 3mmol/1).

CSF Findings

Cerebrospinal-fluid opening pressures were below 200 mmHg (the upper limit ofnormal6) in 37 of the 45 patients. There

777

CSF lactate concentrations in cerebral malaria.

9 patients were studied again after recovery of consciousness.

were more than 10 red cells/µl ofCSF in 6 of the patients, butno sample contained more than 10 lymphocytes/1;polymorphonuclear leucocytes were not seen. CSF lactateconcentrations were above the normal range ( 1 . 1 to 2 - 2

mmol/1)6 in all but 1 of the 45 patients with cerebral malaria(see figure). They were significantly higher in patients whodied than in survivors (9 - 0±5 - 3 mmol/1 and 3 - 4± 1’ 1 mmol/1respectively, p = 0 - 0002). All 11 patients with CSF lactateconcentrations above 6 mmol/1 died. In all 9 patients whowere studied again after recovery, CSF lactate concentrationshad returned to the normal range (1 . 54±0. 27 mmol/1)(significance of change, p=0-001). CSF glucoseconcentrations ranged from 0 - 4 to 7 mmol/1 and were lowerin patients who died than survivors (2 - 92--t 1 - 54 mmol/1 and4.32± 1.16 mmol/1 respectively, p = 0 - 003). There was asignificant negative correlation between CSF lactate and CSFglucose (r=-0-54, p=0-0001). All convalescent CSF

glucose values were within the normal range.

Discussion

CSF lactate concentrations were raised in all but 1 of thepatients with cerebral malaria, were significantly higher inthose who died, and had returned to normal in all patientsstudied again after recovery of consciousness. Thus CSFlactate appears to reflect the severity of the pathologicalprocesses involved in cerebral malaria and is an indicator ofprognostic significance. In normal circumstances the brainuses glucose as its sole substrate and oxidises over 95% tocarbon dioxide and water; very little pyruvic or lactic acid is

produced aerobically.’ Small amounts of lactic acid do crossthe blood-brain barrier by a specific carrier-mediated

transport mechanism shared with some other organic acids.8In man the lack of relationship between blood and CSFlactate concentrations9 suggests that CSF concentrationsreflect metabolic events within the central nervous systemrather than the blood. Lactate accumulates in the brain whenthere is an increase in the NADH/NAD + ratio or a rise in

pyruvate concentration. Pyruvate rises when insufficient

oxygen is available for full aerobic metabolism in the

tricarboxylic acid cycle in conditions such as ischaemia andhypoxia.10 Ischaemia is associated with an enhanced rate ofanaerobic glycolysis, and the amount of lactic acid so formedis determined by the available concentrations of glucose(cerebral glycogen stores are very small). There is increasingevidence that intracellular acidosis is the primary cause of celldeath in hypoxia and ischaemia.l’,’2 Elevated CSF lactateconcentrations have been found in several conditions inwhich cerebral hypoxia is thought to exist and are usuallyassociated with low CSF glucose concentrations.’ 3,14 Inbacterial meningoencephalitis the elevated CSF lactateconcentration provides a diagnostically useful distinctionfrom viral meningoencephalitis and is thought to reflectcerebral hypoxia rather than bacterial lactate production.15,16CSF lactate concentrations remain raised for up to a weekafter starting antimicrobial treatment, whereas glucoseconcentrations are usually normal by the third day. Lactateaccumulation in the CSF in cerebral malaria also presumablyreflects cerebral hypoxia, but some contribution from

parasite lactate production remains possible.Plasmodia have an enormous requirement for glucose, and

those which parasitise mammals metabolise it anaerobicallyto lactate." Parasitised red cells consume 20 to 30 times moreglucose than unparasitised ones. 17,18 Glucose metabolic rateand consequently lactate production is proportional to thestage of parasite development, mature trophozoites andschizonts producing most lactate. 19 These large quantities oflactate must be cleared by the host. Hypoglycaemia and lacticacidosis occur in lethal animal malarias and in severe

falciparum malaria in man. 21 In cerebral malaria red cellscontaining mature forms of the parasite are sequestered in thecerebral capillaries.21 In these capillaries the result ofabundant lactate production by parasites and reduced flow ofparasitised erythrocytes causing impaired lactate clearancewould be local lactic acidaemia. The permeability of theblood-brain barrier to larger-molecular-weight species isnormal in cerebral malaria,22 but it remains possible thatthere is an increased flux of smaller-molecular-weightcompounds and some contribution from the blood to the CSFlactate concentration. Current evidence suggest that this isnot as important as the contribution from the brain.23Lactate accumulates in the CSF in hypocapnia. 24 Blood gas

tensions were measured in only 5 of these patients, and pC02and p02 values (corrected for temperature) were normal. Wehave also measured blood gas tensions repeatedly in 11

patients with cerebral malaria both in coma and followingrecovery; there were no differences in p02 or pC02 values.Pulmonary oedema, aspiration pneumonia, and septicaemiacommonly complicate severe falciparum malaria and causehypoxaemia, but the majority of patients included in thepresent study did not have these conditions at the time of theiradmission lumbar puncture and none was clinically cyanosedor in respiratory distress. Systemic hypoxaemia or

hypocapnia cannot therefore explain their elevated CSFlactate concentrations. Epileptic seizures occur in 50% ofpatients with cerebral malaria. Enhanced glycolysis and

778

cerebral lactate accumulation occur with seizures,25 but thisdoes not persist unless the seizures are repetitive or

continuous or cerebral oxygen supply is reduced. Lumbarpuncture was not performed within 6 h of a convulsion in thisseries, so it is unlikely that seizures were a major contributionto CSF lactate accumulation.

The most likely explanation for the elevation of CSF lactateand the inverse correlation between CSF lactate and glucosein this study is that there is cerebral hypoxia in cerebralmalaria, probably related to sequestration of parasitised redcells in the cerebral microcirculation, and this leads to

anaerobic glycosis in the brain. Persisting neurological deficitis not a feature of cerebral malaria, which suggests that insurvivors, at least, the brain is protected from excessiveintracellular acidosis. In the mouse, fulminant P yeoliiinfections are not associated with a fall in cerebral pH despitesevere acidaemia.26 Local or systemic hypoglycaemia may beprotective in this respect.27 CSF lactate reflected the severityof cerebral malaria and predicted the prognosis with greaterprecision than any other single clinical sign or laboratoryresult. All patients with a CSF lactate concentration over6 mmol/1 died. This measurement may therefore be of valuein the assessment and planning of treatment of patients withcerebral malaria.

We thank the director of Pra Pokklao hospital, Dr Chaisit Dharakul, and hisstaff for their help, Mrs Kamolrat Silamut and Mrs Vanaporn Wutiekanun forexcellent technical help, Dr Sanjeev Krishna for his advice, and Miss EuniceBerry for preparing the manuscript. This study was part of the Wellcome-Mahidol University, Oxford Tropical Medicine research programme fundedby the Wellcome Trust.

Correspondence should be addressed to D. A. W., Faculty of TropicalMedicine, Mahidol University, 420/6 Rajvithi Rd, Bangkok 10400, Thailand.

REFERENCES

1. Warrell DA, Looareesuwan S, Warrell MJ, Intaraprasert R, Kasemsarn B, Bunnag D,Harinasuta T. Dexamethasone proves deleterious in cerebral malaria: Double blindtrial in 100 comatose patients. N Engl J Med 1982; 306: 313-19.

2. Looareesuwan S, Warrell DA, White NJ, et al. Do patients with cerebral malaria havecerebral oedema? Lancet 1983; i: 434-47.

3. Miller LH, Usami S, Chien S. Alterations in the rheologic properties of Plasmodiumknowlesi infected red cells. A possible mechanism for capillary obstruction. J ClinInvest 1971; 50: 1451-55.

4. Udeinya IJ, Schmidt JA, Aikawa M, Miller LH, Green I. Falciparum malaria infectederythrocytes specifically bind to cultured human endothelial cells. Science 1981;213: 555-57.

5. White NJ, Warrell DA, Looareesuwan S, Chanthavanich P, Bunnag D, Harinasuta T.Quinine loading dose in cerebral malaria. Am J Trop Med Hyg 1983; 32: 1-5.

6. Fishman RA. Cerebrospinal fluid in diseases of the nervous system. Philadelphia: WBSaunders, 1980.

7 Siesjö B Brain energy metabolism. Chichester: John Wiley, 1978.8. Pardridge WM, Oldenorf WH Transport of metabolic substrates through the blood-

brain barrier. J Neurochem 1977; 28: 512-.9. Posner JB, Plum F. Independence of blood and cerebrospinal fluid lactate. Arch Neurol

1967; 16: 492-96.10. Siesjö B Lactic acidosis in the brain: Occurrence, triggering mechanisms, and

pathophysiological importance. In: Porter R, Lawrenson G, eds. Metabolic acidosis(Ciba Foundation symposium no 87). London: Pitman Books, 1982: 77-100.

1 1. Myers RE. Lactic acid accumulation as a cause of brain oedema and cerebral necrosisresulting from oxygen deprivation. In: Korobkin R, Guilleminault G, eds.Advances in perinatal neurology. New York: Spectrum, 1979: 85-114

12 Rehncrona S, Rosen I, Siesjö B Excessive cellular acidosis: an important mechanism ofneuronal damage in the brain. Acta Physiol Scand 1981; 110: 435-47.

13. Edgren E, Terent A, Hedstrand U, Ronquist G. Cerebrospinal fluid markers in relationto outcome in patients with global cerebral ischaemia. Crit Care Med 1983; 11: 4-6.

14. Jordan GW, Statland B, Halsted C. CSF Lactate in diseases of the CNS. Arch InternMed 1983, 143: 85-87.

15. Brook I, Bricknell KS, Overturf GD, Finegold SM. Measurement of lactic acid incerebrospinal fluid of patients with infections of the central nervous system. J InfectDis 1978; 137: 384-90.

16 Menkes JH. Improving the long term outlook in bacterial meningitis. Lancet 1979; ii:559-60.

17. Sherman IW. Biochemistry of Plasmodium (malaria parasites). Microbiol Rev 1979; 43:453-95.

18. Jensen MD, Conley M, Helstowski LD. Culture of Plasmodium falciparum: The role ofpH, glucose, and lactate J Parasitol 1983; 69: 1060-67.

19 Pfaller MA, Krogstad DJ, Parquette AR, Nguyen-Dinh P. Plasmodium falciparum:Stage-specific lactate production in synchronized cultures. Exp Parasitol 1982; 54:391-96

20. White NJ, Warrell DA, Chanthavanich P et al. Severe hypoglycaemia and

hyperinsulinaemia in falciparum malaria. N Engl J Med 1983, 309: 61-6621. MacPherson G, Warrell MJ, White NJ, Looareesuwan S, Warrell DA. Human cerebral

malaria. a quantitative ultrastructural analysis of parasitised erythrocytesequestration. Am J Pathol (in press).

22 Warrell DA, Looareesuwan S, White NJ, et al. Permeability ofthe blood-CSF-bartierin human cerebral malaria. Abstract. XI Int Cong Trop Med Malaria. Calgary,Canada, Sept 16-22, 1984.

23. Weyne J, Leusen I. Lactate in CSF in relation to brain and blood. In: Cserr HF, ed.Fluid environment of the brain. New York: Academic Press, 1975: 255-76

24 Plum F, Posner JB. Blood and cerebrospinal fluid during hyperventilation Am JPhysiol 1967; 212: 864-70.

25. Chapman AG, Meldrum BS, Siesjö B. Cerebral metabolic changes during prolongedepileptic seizures in rats. J Neurochem 1977; 28: 1025-35.

26. Krishna S, Shoubridge EA, White NJ, Weatherall DJ, Radda GK. Plasmodium yoeliiBlood oxygen and brain function in the infected mouse. Exp Parasitol 1983; 56:391-96.

27. Gardiner M, Smith ML, Kagstrom ML, Shohami E, Siesjö BK. Influence of the bloodglucose concentration on brain lactate accumulation during severe hypoxia andsubsequent recovery of brain energy metabolism. J Cereb Blood Flow Metab 1982, 2:429-38.

TRANSFER OF IgA DEFICIENCY TO ABONE-MARROW-GRAFTED PATIENT WITH

APLASTIC ANAEMIA

LENNART HAMMARSTRÖMOLLE RINGDÉN

BERIT LÖNNQVISTC. I. EDVARD SMITH

THOMAS WIEBE

Departments of Clinical Immunology, Medicine, andTransplantation Surgery, Huddinge Hospital, Huddinge, Sweden;Department of Immunology, Wallenberglaboratory, Lilla Frescati,

Stockholm; and Department of Paediatrics, Lunds UniversityHospital, Lund, Sweden

Summary IgA deficiency developed in a 2-year-old boywith aplastic anaemia who received a bone-

marrow graft from his HLA-identical, 6-year-old, IgA-deficient sister. Southern blot analysis revealed the presenceof &agr;-genes in both children, thus suggesting a defect of

lymphocyte stem-cell differentiation as a cause of IgAdeficiency. Tissue typing showed homozygosity of HLA A1,B8, DR3, the haplotype associated with IgA deficiency inhealthy people. Despite normal serum levels of IgGsubclasses in both donor and recipient, both children showeda relative lack of specific IgG2 anticarbohydrate antibodies.This suggests that their IgA deficiency is part of a morefundamental aberration of immunoglobulin class andsubclass distribution.

Introduction

SELECTIVE IgA deficiency (serum IgA <0’ 05 g/1) is thecommonest immunodeficiency in man. Genetic factors

clearly influence the serum levels of IgA, and deficiencyappears to be associated with genes within the HLA locus.2,3In a number of cases lack of IgA is associated withconcomitant IgG2 deficiency,4 a prognostic marker for

permanent lack of IgA.’IgA synthesis is highly dependent on thymus-derived

cells,6 but it is still unclear whether the impaired capacityseen for differentiation in IgA-deficient individuals is due toaltered thymic/bursal microenvironment or reflects a

genetically determined defect in the maturation of T or Bcells and/or a structural defect of IgA heavychain constantgenes.Bone-marrow transplantation is a well-established method

for treatment of patients with haematological malignancies,severe aplastic anaemia, and some inborn errors ofmetabolism. We describe a patient with aplastic anaemia whoreceived a bone-marrow graft from his IgA-deficient sister.