The Pathophysiology of Malaria

The Pathophysiology of Malaria

The Pathophysiology of Malaria NICHOLAS J. WHITE Wellcome-Mahidol University, Oxford Tropical Medicine Research Programme, Faculty of Tropical Medici...

5MB Sizes 0 Downloads 5 Views

The Pathophysiology of Malaria NICHOLAS J. WHITE

Wellcome-Mahidol University, Oxford Tropical Medicine Research Programme, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand; Wellcome Trust Clinical Research Unit, Centre for Tropical Diseases, Cho Quan Hospital, Ho Chi Minh City, Vietnam; NuBeld Department of Clinical Medicine, John Radclifle Hospital, University of Oxford, Oxford, UK and MAY HO

Wellcome-Mahidol University, Oxford Tropical Medicine Research Programme, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand; Department of Microbiology and Infectious Diseases, Health Sciences Centre, University of Calgary, Calgary, Alberta, Canada I. Introduction ............................................ 11. Animal Models ...................................... 111. Human Malaria ........................ ............... A. Clinical features ........................................ B. Causes of death and permanent sequelae .............................. IV. Pathogenesis .......................... A. Sequestration ...................................................... B. Reduced red cell deformability C. Cytoadherence .................................. D. Putative endothelial cytoadherence receptors .......................... E. Parasite cytoadherence ligands . . . . . . . . . . ...............

.................................. V. Parasite Virulence Factors ................................... A. Multiplication ........................... ............. B. Synchronicity .................................... A.


................................................ ................. Human studies .............................................

Animal studies


84 85 86 86 88 93 93 95 96 98 101

I02 105 105


110 110 110 112 112

Cnpyrighr 0 1992 Academic Press Limired All rights o/reprodurrion in any /orm reserved

84 VII.


Pathophysiology of Vital Organ Dysfunction A. Cardiovascular abnormalities . . . . . . . . B. Algid malaria and septicaemia ....................................... C. Pulmonary oedema ............................................ D. Blood flow and metabolism ..................................... E. Cerebral malaria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Capillary permeability . . . G. Fluid space changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Electrolyte changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Endocrine dysfunction J. Renal impairment . . . . K. Gastrointestinal dysfunction ......................................... L. Liver dysfunction .................................... M. Hypoglycaemia . . N. Lactic acidosis . . . . . . . . . . . . . . . . 0. Skeletal muscle abnormalities ........................................ P. Anaemia.. ......................................................... Q. Bone marrow function R. Blackwater fever . . . . . . S. Thrombocytopenia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . T. Coagulation.. ...................................................... U. Splenic function . . . . . . . . . . . . . . . . .

1 I4

W. Immune dysfunction ................................................ X. Complement . . . . . Y. Pregnancy . . . . . . . VIII. Conclusion ............................




1 I4

I15 119 122 125 126 129 130 135 136 139 139



Plasmodium vivax, P. malariae and P. ovale are phylogenetically associated, and lie relatively close to the simian malarias ( P . knowlesi, P. fragile) on the evolutionary tree. They very rarely cause fatal disease in man. P. falciparum, which may cause malignant tertian malaria, is a major cause of death in the tropics. This parasite is closest phylogenetically to the malaria parasites of birds ( P . gallinaceum, P. lophurae), and is far from the benign human malarias (Waters, A. P., et al., 1991). It is thought to be a recent evolutionary acquisition of man. The propensity of P . falciparum to kill its host has been considered evidence of a lack of parasitic sophistication-a debatable point. In fact, P . falciparum is an extremely successful parasite infecting approximately 200 million people, of whom the vast majority are unaware of its presence in their bloodstream. Overall, falciparum malaria is the most important parasitic disease of man (Wernsdorfer and McGregor, 1988). The resurgence of malaria in the past two decades has stimulated a



considerable amount of scientific and medical research. Since the review by Maegraith and Fletcher in 1972 our understanding of pathophysiological mechanisms in malaria has advanced considerably in areas such as the pathogenesis of metabolic dysfunction, the molecular processes involved in cytoadherence, and the causes of anaemia. However, in other areas progress has been slow. We do not know why coma occurs, or what causes pulmonary oedema in severe malaria, and we do not fully understand blackwater fever. There is still no good in vitro correlate of immunity. Much of the recent research has been conducted either in animal models or with cultured P. fakiparum parasites. The relevance of the observations, and the hypotheses they generate, to disease in man still needs to be established in many cases.

11. ANIMAL MODELS Certain features of severe malaria are common to all infected species (Olsson, 1967). The metabolic consequences of hypoglycaemia and lactic acidosis occur in the terminal phase of avian, rodent, simian and human malarias (Sadun et af., 1965; White, 1985). It appears reasonable, therefore, to extrapolate from observations in the animal to the disease in man (Ehrich et al., 1984). Unfortunately, there is no good animal model for cerebral malaria (Yoeli, 1976). Although sequestration occurs in several of the animal malarias (e.g. P.fragile and P . coatneyi in rhesus monkeys), the hosts do not become unconscious at low peripheral parasitaemias. Cerebral symptoms are not a feature of rhesus monkeys infected with P . knowlesi, or Aotus or Saimiri monkeys infected with P. falciparum (two commonly used models). Nervous system dysfunction occurs in certain rodent models, but the clinical manifestations (focal signs, coma only in the terminal stages) and the main pathological findings (vasculitis with mononuclear cell aggregation and infiltration, little or no sequestration) are different from those of human cerebral malaria (Polder et af., 1983; Thumwood et al., 1988). Several hypotheses concerning the pathophysiology of coma in falciparum malaria and, unfortunately, several suggested treatments for severe malaria, have been based on animal models of dubious relevance to human disease. For example, the use of steroids in cerebral malaria derived largely from observations on rhesus monkeys infected with P . knowlesi (Maegraith and Fletcher, 1972), and the use of cyclosporin was based on observations on rodents infected with P. berghei (Grau et af., 1987a). Both treatments were subsequently found to be harmful in man.


N. J.



Immunity to malaria develops slowly. In areas of intense transmission of P. falciparum, the infant is inoculated repeatedly with sporozoites in the first months of life, but, for a variety of reasons which include the passive transfer of malarial immunity and a less favourable intra-erythrocytic environment (a high haemoglobin F content), severe disease is rare at this age (Bruce-Chwatt, 1952; Pasvol et al., 1976). Most deaths occur between the ages of one and four years (Greenwood et al., 1987). Thereafter the infection becomes progressively less severe, and by the time adulthood is reached falciparum malaria is largely asymptomatic (premunition). The rate at which this process occurs is determined by the intensity of malaria transmission and thus the number of infections. Age may be an independent factor; i.e., for the same exposure adults develop immunity more rapidly than do children (Baird et al., 1991). In areas where transmission is low, uneven or highly seasonal, then symptomatic disease is seen at all ages. The principal manifestation of severe malaria in these circumstances is cerebral malaria (Brewster et al., 1990), whereas severe anaemia in young children is more prominent in areas of intense stable transmission. Several factors are responsible for the slow acquisition of immunity to malaria. Malaria parasites exhibit considerable antigenic diversity, and readily undergo antigenic variation. Most, or possibly all, naturally acquired infections consist of several discrete genotypic and antigenic parasite clones. In addition to this diverse antigenic array, the host immune system is activated in a non-specific manner, while malaria antigen-specific immune responses are suppressed (Ho et al., 1986Fan “immunological smoke screen”. The precise immunological processes and other host defence mechanisms responsible for the control of falciparum malaria infections have not been elucidated, but their failure contributes to the development of severe disease with heavy parasite burdens and potentially lethal vital organ dysfunction. Why some infections are more severe than others is at present largely unknown. A list of possible factors is given in Table 1. It should be noted that although a variety of cellular immune defects has been documented, it is not yet known whether these are a cause or an effect of severe malaria. Indeed nearly all the factors this table are controversial. A.


Uncomplicated malaria is a febrile illness associated with headache, muscular discomfort, weakness and malaise. These features are non-specific, resembling influenza, and are common to the four human malarias caused by P . falciparum, P. vivax, P . malariae and P . ovale. As the untreated



TABLE1 Possible factors determining the severity of falciparum malaria Factor

Genetic Protective erythrocyte abnormalities (surface receptors, cytoskeleton, haemoglobin type, enzymes, availability of reticulocytes), HLA type Age Haemoglobin F Transfer of maternal immunity Pregnancy Nutrition Protective effects Malnutrition Low iron Deficiencies of riboflavin, a-tocopherol, essential amino acids Host defence Humoral immunity Cellular immunity Phagocytic function Failure to augment splenic function rapidly?


Miller (1988) Hill et al. (199 Pasvol et al. (1 976) McGregor and Wilson (1988) WHO (1990)

Edington ( 1967) Murray et al. (1975) Thurnham et al. (1983) Tharavanij et al. ( 1984) Druilhe et al. (1983) Ward et al. (1984)

Other diseases/conditions Concomitant or previous infection impairing host defence? Anaemia Drug addiction Parasite factors Cytoadherence Rosetting Multiplication capacity? Antigenic variation? Sporozoite inoculum/viability Environmental/social factors Location Availability and use of antimalarial drugs Family respanse to illness Competence and resources of primary health care worker Competence and resources of referral centre

Ho et al. (1991a) Carlson et al. (1990b)



infection becomes synchronized, the fever becomes periodic with pyrexial spikes every one, two or three days associated with chills or rigors. Synchronization occurs earlier with P. vivax and P . ovule than with P . falciparum. The untreated infection continues for weeks or months in non-immune patients, but only P. falciparum produces fulminant disease. Severe falciparum malaria is a multi-system disease. The clinical features reflect the pattern of vital organ involvement. Coma is the most prominent manifestation (Khan, 1945) (Fig. 1). Loss of consciousness may be sudden following a seizure, or gradual (White and Looareesuwan, 1987bi.e. over a period of hours, often preceded by a period of delirium. The neurological signs are usually symmetrical and suggest diffuse encephalopathy. Retinal haemorrhages can be seen by direct ophthalmoscopy in approximately 15% of comatose patients (Kayembe et al., 1980; Looareesuwan et al., 1983a; WHO, 1990), but papilloedema is rare ( 1 YO).Convulsions are generalized in most cases and often herald the onset of cerebral malaria; they are particularly common in children. Most patients recover consciousness within 3 days (children recover more quickly than adults). Jaundice is more common in adults than children with severe malaria. Anaemia develops rapidly in all age groups. Shock may occur in severe malaria, but most patients are warm and well perfused with sinus tachycardia and blood pressure at the lower end of the normal range. Abnormal ventilatory patterns may be associated with extensor or flexor posturing. Sustained hyperventilation (Kussmaul’s breathing) is a worrying finding indicating metabolic acidosis if the chest examination is normal, or aspiration pneumonia or pulmonary oedema when there are abnormal chest signs. Vomiting is common, particularly in children. Enlargement of the liver and spleen is usual and the spleen is often palpable, but lymphadenopathy is not a feature of malaria. A bleeding diathesis resulting from disseminated intravascular coagulation occurs in approximately 5% of patients with severe malaria.



1. Fatal malaria in adults

P . vivax, P . malariae and P . ovule kill very rarely during acute infections. Occasional patients debilitated from other diseases may die, and rupture of the enlarged spleen may cause death from haemorrhage if immediate surgery is not available (Covell, 1955). In most adults who die from severe falciparum malaria there is multiple vital organ dysfunction and an adequate explanation for the fatal outcome is apparent to the attending clinician. The principal causes of death are pulmonary oedema, acute renal failure, and metabolic acidosis with circulatory failure (WHO, 1986, 1990). Although



FIG. 1 . (a) Cerebral malaria. Opisthotonus in a comatose Gambian child. (b) Disconjugate gaze in a child with cerebral malaria.



any of these manifestations may occur in isolation, they more commonly coexist, and it is often difficult to apportion causality. Coma is usual, but not invariable. In the majority of adult cases, nervous system dysfunction does not appear to contribute directly to death (White and Looareesuwan, 1987), although fatal aspiration pneumonia may follow convulsions. Spontaneous respiratory arrest may occur with cerebral malaria, commonly following a sequence of hyperventilation, cyclical (Cheyne-Stokes), and then ataxic breathing patterns. This is more common in children. Renal failure during or after acute falciparum malaria is an important cause of death in adults. Patients with severe malaria may die from secondary complications; aspiration pneumonia, spontaneous Gram-negative septicaemia, or infective complications of intensive therapy. Deaths from antimalarial drug toxicity, bleeding, or cardiac dysrhythmias also occur but are rare. Deaths from acute pulmonary oedema and renal failure may happen after several days treatment when parasitaemia has fallen, or even cleared, but most fatalities occur within 3 days of admission to hospital during the acute phase of the disease. Many laboratory measurements have been correlated with outcome. The more readily measurable poor prognostic variables are hypoglycaemia, hyperlactataemia, raised cerebrospinal fluid (CSF) lactate, leucocytosis, acidosis and raised serum concentrations of urea, creatinine and the transaminases. High parasite counts are associated with an increased risk of fatal outcome. In the classic studies from Kuala Lumpur, counts over 100 000 per pl were associated with increased mortality, and 50% of those with counts over 500000 per pl died (Field and Niven, 1937; Field, 1949). However, there is considerable variability. Parasite counts in semi-immune African children can be well over 100 000 per p1 without symptoms or signs of severe disease, whereas non-immune patients may die with very low peripheral parasitaemias. Recent studies have shown that staging of parasite development in the peripheral blood film adds considerably to the prognostic sensitivity and specificity of the parasite count (N. J. White, unpublished observations). 2. Fatal malaria in children Falciparum malaria is a major cause of infant death and morbidity in the tropics (McGregor et al., 1956; Greenwood et al., 1987). Children with severe malaria have a slightly lower overall mortality rate than adults. Definitions of cerebral malaria have varied considerably but, of the interpretable published series in which chloroquine or quinine was used for treatment, the overall mortality rate from cerebral malaria in adults was 21.5% (141/655) compared with 13.8% (230/1383) in children (P= 0.009). The causes of death in children are often different from those in adults. Vital organ dysfunction is less evident. Acute renal failure is very rare in young



children, whereas renal impairment contributes to over half the adult deaths (White et al., 1987a; Molyneux et al., 1989b; WHO, 1990). Pulmonary oedema is also unusual. In contrast, hypoglycaemia and lactic acidosis with terminal circulatory failure and aspiration pneumonia are more common in young children. Many children die suddenly in the acute phase of severe malaria without a clear explanation (unpublished observations). There is primary respiratory arrest without circulatory failure. Of the 26 deaths in a series of 180 Gambian children with severe malaria, no clear explanation was evident in 14 children, four of whom died with primary respiratory arrest (N. J. White and D. Waller, unpublished observation). The possibility that this phenomenon results from brainstem compression- secondary to raised intracranial pressure has been suggested by recent findings that opening pressures at lumbar puncture are usually raised in children with cerebral malaria (Newton et al., 1991; Waller et al., 1991). This is in contrast to the findings in adults where opening pressures are normal in 80% of cases (Fig. 11) and are significantly lower in fatal cases than in survivors (Warrell et al., 1988). Some of the neurological findings in children with cerebral malaria have been attributed to the development of a tentorial pressure cone, but the specificity of these interpretations needs to be confirmed (Section VII.F.2). Nevertheless, the observations are important because of the possibility of providing specific treatment (e.g. with mannitol) which would reduce intracranial pressure. In areas of stable intense transmission (hyper- or holoendemicity), anaemia becomes more common than cerebral malaria as a presentation of severe malaria in children. The case specific mortality rate is lower. In a recent large series of over 600 severely ill children reported from The Gambia, overall mortality was approximately 7% in children with severe anaemia compared with 16% in cerebral malaria (Brewster et al., 1990). The causes of death in children with severe malarial anaemia are also different. Some children have relatively low parasite counts, and in these the pathophysiological processes relate largely to severe anaemia, whereas in others both severe malaria and severe anaemia coexist. High output heart failure is an important and potentially lethal manifestation of severe anaemia. Many children die suddenly before or during blood transfusion. Twenty-four hour electrocardiographic recordings suggest that cardiac arrhythmias are probably not a cause of “unexplained deaths” in acute cerebral malaria (F. Nosten and N. J. White, unpublished observations), but this has not been studied in severe malarial anaemia.

3. Permanent sequelae Approximately 10% of children, but less than 5% of adults, surviving cerebral malaria have obvious permanent residual neurological sequelae



(Molyneux et al., 1989b; Brewster et al., 1990). In over half the cases there is hemiparesis (Fig. 2), but cortical blindness and clinical evidence of more diffuse brain damage are also common. In half the children discharged from hospital with a neurological deficit, there is full recovery within 6 months (Brewster et al., 1990). The possibility that cerebral malaria causes more subtle permanent defects such as slight motor impairment or mild intellectual retardation in survivors has not been explored. (a)

FIG.2. (a) Right hemiparesis following cerebral malaria (courtesy of Dr J. Crawley). (b) Diffuse cortical damage 6 months after cerebral malaria.



Carotid angiography has been performed in 1 1 cases of post-cerebral malaria stroke. Four showed large vessel obstruction and one segmental narrowing, but the remaining six were normal (Collomb et al., 1967; Sanohko et al., 1968; Omanga et al., 1983). Persistent neurological deficit is associated with protracted hypoglycaemia, seizures, prolonged and profound coma, and coexistent severe anaemia during the acute phase of cerebral malaria. There is no evidence that large intracerebral haemorrhages occur, so presumably cerebral infarction is responsible for the neurological deficit, but the mechanisms responsible have not been elucidated. Obviously the increased cerebral metabolic demands associated with seizures, hypoglycaemia, reduced cerebral perfusion pressure and cerebral arterial oxygen transport may all combine to compromise the fragile balance between blood supply and metabolic demands, particularly in vulnerable “watershed” areas of limited vascular reserve such as the parieto-occipital region (Sections VII.D.1 and VI1.T).

IV. PATHOGENESIS The pathology of uncomplicated malaria is determined by the destruction of parasitized erythrocytes, the intravascular liberation of parasite and host products at merogony (schizogony)*, and the host reaction to this process. In severe falciparum malaria there is a greater parasite burden and sequestration in the microcirculation of the vital organs. These two factors account for the lethal potential of this parasite. A.


All stages of parasite development are seen in blood smears taken during infections with P. vivax, P . malariae and P . ovale, but the peripheral blood in P. fakiparum malaria rarely contains pigmented trophozoites or meronts (schizonts).* The intravascular sequestration of erythrocytes containing these mature forms of the parasite is an essential pathophysiological feature of falciparum malaria (Luse and Miller, 1971). The degree of vascular sequestration varies between organs, being greatest in the brain in patients with cerebral malaria and least in the skin (Li et al., 1983; MacPherson et al., 1985). In patients who die without developing cerebral malaria, sequestration is significantly less in the brain (Pongponratn et al., 1991). These findings suggest a relationship between the organ distribution of sequestration and pathology. * Throughout this review, the etymologically more consistent terms meront and merogony have been used instead of schizont and schizogony (eds).



FIG.3. Cerebral venule packed with parasitized erythrocytes and pigment in a fatal case of cerebral malaria (courtesy of Professor M.Aikawa).

Within a few years of Laveran’s discovery of the malaria parasite in the blood of a febrile patient in Algeria, pathological observations from Italy (Marchiafava and Bignami, 1894) recorded the extraordinary discrepancy between the microscopical appearance of the peripheral blood and that in the cerebral vessels of patients dying from cerebral malaria. The capillaries and venules in the brain were packed with erythrocytes containing mature forms of the parasite and abundant brown-black pigment, which were not seen in ante-mortem blood samples (Fig. 3). These findings were confirmed in pathological reports on the soldiers dying from falciparum malaria in Macedonia during the First World War of 1914-1918 (Dudgeon and Clarke, 1917, 1918; Gaskell and Miller, 1920). It was suggested that the parasitized erythrocytes had difficulty traversing the capillary bed and, as a result, blood flow was obstructed. Initially it was thought that thrombus formation occurred (Dudgeon and Clarke, 1917), altho.ugh the authors also conceded that microvascular obstruction by parasitized erythrocytes might be reversible (Dudgeon and Clarke, 1918). Later pathological studies have concluded that widespread thrombus does not occur in fatal cerebral malaria, and have favoured the concept of “plugging” (i.e. obstruction) of small vessels by masses of parasitized erythrocytes (Gaskell and Miller, 1920; Spitz, 1946) (Section VI1.T). The mechanism of microvascular obstruction was investigated in a series



of studies by Knisely and colleagues (1941). They studied rhesus monkeys infected with the lethal kra monkey parasite, P. knowlesi (which does not sequester), and also recorded observations in patients with falciparum malaria. The microcirculation was directly visualized and cinematographic recordings were made of flow in the mesenteric vessels of the monkeys and the bulbar conjunctiva in man. During infection the erythrocytes were seen to agglutinate and eventually to form what was delicately described as a “thick muck-like sludge”. Pathological events in severe malaria were interpreted as resulting from ischaemia, hypoxia, and any subsequent toxic effects resulting from release of unidentified materials from the “sludge”. Although microvascular thrombus deposition appears to be unusual, large vessel occlusion may occur rarely in childhood cerebral malaria leading to stroke (Brewster et af., 1990). In the last 20 years investigations have focused first on reduced red cell deformability, and more recently on the specific adherence (cytoadherence) of infected erythrocytes to vascular endothelium. B.


Red cells containing malaria parasites do not pass through micropore filters as easily as unparasitized erythrocytes. This suggests that these infected erythrocytes are less deformable than normal cells, and might therefore not pass as easily through capillary beds (Miller, L. H. et af., 1971, 1972; Lee, M. V. et al., 1982). In normal microcirculatory flow, red cells (diameter 7-8 pm) must undergo considerable deformation in their passage through the capillary (diameter 3 4 p m ) . Capillary blockage does occur when red cells are unusually rigid, as in sickle cell crisis, but the clinical features and organ distribution of vascular obstruction in this condition are most unlike those of severe malaria. The reduced deformability of red cells infected with P. falciparum is directly proportional to the maturity of the parasite (Cranston et al., 1984); the older, and larger, the parasite, the more rigid is the infected cell. There is increased expression of phosphatidylserine and phosphatidylethanolamine and reduced phosphatidylcholine on the outer leaflet of the trophozoite-infected erythrocyte membrane. Cells containing meronts also have reduced sphingomyelin in the outer leaflet (Maguire et al., 1991). These abnormalities of phospholipid distribution, which may result from depletion of adenosine triphosphate, oxidative stress and alterations in the cytoskeleton, influence the surface properties of the infected erythrocyte. They are associated with increased phagocytic clearance and adherence to monocytes and endothelial cells (Section VII.P.4). Several other factors contribute to reduced red cell deformability: increased membrane stiffness, increased cytoplasmic viscosity resulting from changes in membrane permeability



(Dunn, 1969; Kutner et al., 1983), reduced surface area-to-volume ratio (increased sphericity), and principally the rigidity of the parasite itself (Nash et al., 1989). However, reduced erythrocyte deformability alone does not explain the phenomenon of sequestration nor, therefore, the severity of severe falciparum malaria. It does not explain the concentration of parasitized erythrocytes in venules which are downstream from the site of minimum vascular cross-sectional area (i.e. the mid capillary), nor the precise parasite stage specificity of sequestration. If a rigid red cell, or an aggregate of cells such as a “rosette”, irreversibly blocks flow by becoming stuck in a small blood vessel, then the tail of erythrocytes stacked behind the obstructing cell should have a similar proportion of parasitized cells to that in the peripheral blood (White, 1985). This is analogous to a car accident causing a traffic jam. In addition, reduced red cell deformability does not explain preferential sequestration in the cerebral microvasculature, which has similar internal dimensions to the vessels in other organs. C.


The principal event causing sequestration and impeding microcirculatory flow appears to be the cytoadherence of parasitized erythrocytes (PRBC) to vascular endothelium (Luse and Miller, 1971; Raventos-Suarez et al., 1985). Cytoadherence is a specific process, in that it occurs only in capillaries and post-capillary venules, and involves only erythrocytes containing the more mature stages of the parasite, viz. trophozoites and meronts. The cytoadherent properties of P . fakiparum appear to be modulated by the spleen (Hommel et al., 1983). Infected erythrocytes do not cytoadhere in splenectomized saimiri monkeys, but the property is regained after several asexual cycles of a cloned line following transfusion into monkeys with intact spleens (David et al., 1983). It is not known how this modulation takes place. At the ultrastructural level, electron-dense, knob-like protrusions of the erythrocytic membrane are seen at the points of contact between the PRBC and endothelial cells (MacPherson et al., 1985) (Fig. 4). These knobs were considered essential for cytoadherence by facilitating the initial attachment of the infected erythrocyte to the vascular endothelial cell, and by concentrating the parasite ligands at a particular site. In the last few years, the importance of the knobs has been disputed. Two clones of knobless (K -) laboratory-adapted parasites (Biggs et al., 1989; Udomsangpetch ef al., 1989), and one clinical isolate (S. Semoff, B. Singh and M. Hommel, unpublished data), have been shown to cytoadhere in vitro in the absence of knobs, but probably through the same molecular mechanism (see later) as the knobby (K+) parasites (Biggs et al., 1990). Since the laboratory-adapted K - cytoadherent parasites were obtained after many cycles of selection from a K + isolate, it is possible that K + and K - parasites coexist in



isolates from natural infections. Parasitism would continuously favour the selection of K + organisms if they were able to form a more stable union with host cells, and thus evade splenic clearance. The relevance of these experimental findings remains to be determined, since cytoadherence under the static conditions of the assays in vitro may require a less durable interaction than that of infected erythrocytes exposed to the dynamics of the host circulation and immune response in vivo. Raventos-Suarez et al. (1985) have investigated microvascular obstruction by P.falciparum and the role of knobs in cytoadherence using the isolated perfused rat mesocaecum. When human erythrocytes infected with P.falciparum were added to the perfusate, blood flow was seen to slow and then finally stop, as a result of cytoadherence to the vascular endothelium. Only K + erythrocytes cytoadhered and blocked flow. Ultrastructural studies of human tissues from fatal malaria cases have not shown cytoadherence independent of knobs.

FIG.4. Cytoadherence between parasitized erythrocyte (PE) and cerebral vascular endothelial cell (E) showing knobs at the points of attachment (arrowed) (courtesy of Professor M. Aikawa) ( x 38 000).




The stage and host cell specificity of cytoadherence suggest that the interaction between PRBC and endothelial cells involves specific parasite ligands and host receptors. The quest for these proteins has proved more difficult and confusing than expected originally, and has relied heavily on the use of models in vitro. The search began with the observation that PRBC adhere to cultured human umbilical vein endothelial cells (HUVEC) in vitro (Udeinya et al., 1981) with the same stage and host cell specificity as observed with sequestration in vivo. This model proved difficult and unpredictable. A number of normal cell types and continuous cell lines have subsequently been shown to have cytoadherent properties, of which the human amelanotic melanoma cell line C32 (American Type Culture Collection, no. CRL1585) (Schmidt et a/., 1982) has been the most extensively employed. In addition, putative receptor proteins have been purified and studied either immobilized on plastic or in their soluble forms. Although such studies provide definitive information regarding cytoadherence to a particular molecule, it has been argued that the immobilized or free receptor molecule in vitro may not have the same tertiary structure as that expressed on the cell surface in vivo. One approach to circumvent this problem has been to transfect COS cells with plasmids carrying the complementary deoxyribonucleic acid (cDNA) for the receptors. The receptor under investigation is then expressed on the otherwise antigen-free surface of the COS cell. 1. Thrombospondin

The first molecule identified as a potential receptor for cytoadherence was thrombospondin (TSP), an adhesive glycoprotein produced by activated platelets and involved ubiquitously in cell-to-cell interactions (Tandon et al., 1989). Using purified TSP, Roberts et al. (1985) showed that PRBC adhered selectively to TSP in a dose-dependent manner, but not to other adhesive proteins such as fibronectin and von Willebrand’s factor. Cytoadherence was specifically inhibited by anti-TSP monoclonal antibodies and soluble TSP, and occurred under both static and shear-flow conditions (Rock et al., 1988). However, further work has revealed that while TSP may contribute to cytoadherence, it is not sufficient to mediate the process alone. PRBC do not adhere to every melanoma cell line which secretes TSP (Panton et al., 1987) and anti-TSP antibodies neither bind, nor inhibit cytoadherence, to C32 melanoma cells (Ockenhouse et al., 1989a). 2.


The second molecule to be implicated in cytoadherence was the leucocyte differentiation antigen CD36, a membrane glycoprotein (molecular mass



88 kDa). A monoclonal antibody to C36,OKM5, was shown to inhibit and . reverse the cytoadherence of PRBC to a number of target cells in vitro including epithelial cells and C32 melanoma cells (Barnwell et al., 1985; Panton et al., 1987). PRBC have been shown subsequently to adhere selectively to purified CD36 immobilized on plastic. The purified protein specifically and competitively inhibits the cytoadherence of PRBC to the C32 cells and endothelial cells (Barnwell et al., 1989; Ockenhouse et al., 1989a). Furthermore, although binding to purified TSP and purified CD36 are highly correlated (Hasler et al., 1990), PRBC adhere directly to CD36transfected COS cells in the absence of TSP (Oquendo et al., 1989). CD36 is also expressed on the surface of monocytes and platelets, and PRBC have also been shown to cytoadhere to these cells (Ockenhouse and Chulay, 1988; Ockenhouse et al., 1989b). Peripheral blood monocytes are triggered by this process to produce oxidative metabolites which are toxic to intra-erythrocytic parasites (Ockenhouse et al., 1984). The current concept with regard to the role of CD36 and TSP in cytoadherence has been summarized by Barnwell et al. (1989). The two molecules could interact by the association of soluble TSP in plasma with exposed parasite ligands. The TSP-ligand complex would then interact with cell-bound CD36. Alternatively, plasma or cellular TSP may associate with the cell surface membrane receptor before interaction with parasite ligands. There is some evidence that CD36 acts as the natural receptor for TSP (Asch et al., 1987). However, cytoadherence to CD36 may occur completely independently of TSP, as has been shown with CD36-transfected COS cells (Oquendo et al., 1989). This adherence is not inhibited or reversed by antiTSP antibody, and TSP does not adhere to the CD36-transfected COS cells. 3. ICAM-I

A third candidate receptor molecule, the intercellular adhesion molecule 1 (ICAM-1) or CD54, has been identified by Berendt et al. (1989). This glycoprotein (also molecular mass = 88 kDa) has a well established role in mediating cellular immune responses by acting as a ligand for lymphocyte function antigen 1 (LFA-1). ICAM-1 is also the natural receptor for rhinovirus attachment. Binding of parasitized erythrocytes to CD36 or ICAM-1 expressed on transfected COS cells can be inhibited by their respective monoclonal antibodies. A schematic diagram of the possible molecular interactions involved in cytoadherence is shown in Fig. 5. 4. Clinical correlates To determine the relative importance of the three receptor molecules in vivo, the cytoadherence of parasite isolates taken directly from patients with acute

1 00


FIG. 5. Schematic diagram of the molecular interactions involved in cytoadherence between the parasitized erythrocyte (PRBC) and endothelial cells. The knob protrusions could bear specific ligands for thrombospondin (TSP-R, and TSP, respectively), CD36 and ICAM-I. These three ligands could be expressed as separate molecular entities (A) or as a single composite cell surface molecule (B). The endothelial cell is also shown secreting TSP, which is then bound to endothelial cell components (TSP-RE). TSP could also derive from platelets or other cells and act as a bridge between the receptor (TSP-R,) and TSP-RE or CD36. Cytoadherence may also take place independently of TSP with direct interaction between CD36 and/or ICAM-I and their PRBC ligands. (Reproduced with kind permission from Howard and Gilladoga, 1989.)

falciparum malaria has been examined in several studies. Binding to TSP (Sherwood et al., 1987), C32 melanoma cells (Marsh et al., 1988; Ho et al., 1991a) and purified CD36 (Ockenhouse et al., 1991b) was directly proportional to parasitaemia. When parasite isolates were compared at a fixed parasitaemia (i.e. binding was “normalized”), a range of intrinsic cytoadherent capabilities among different isolates was evident. In the case of cytoadherence to C32 melanoma cells and purified CD36, the degree of binding was positively correlated with biochemical indicators of disease severity in adult Thai patients (Ho et al., 1991a; Ockenhouse et al., 1991b), but there was no correlation between cytoadherence and the presence of cerebral symptoms either in this series (Ho et al., 1991a) or in a separate study of Gambian children (Marsh et al., 1988). Indeed, adults with cerebral malaria tended to



have lower melanoma cell binding rates than other patients with severe disease. These findings support the hypothesis that CD36 mediates sequestration in vital organs other than the brain but question the role of CD36 in mediating cerebral sequestration. In immunohistochemical studies of human tissues using the monoclonal antibody OKM5, CD36 can be demonstrated on vascular endothelium in sections of lung, kidneys and liver (Knowles et af., 1984), but not in the brain (A. Berendt, personal communication). However, using a different monoclonal antibody, CD36 was detected on cerebral vascular endothelium (Barnwell et af., 1989). This suggests that a different receptor epitope of CD36 may be expressed in the cerebral microvasculature. In contrast to the results with CD36, cytoadherence of freshly isolated P . ,fakiparum to purified ICAM- I , and to a sub-clone of the C32 melanoma cell line bearing ICAM- I but not CD36, was generally low in one study and bore no quantitative relationship to any clinical manifestations of malaria (Ockenhouse et af., 1991b). When both CD36 and ICAM-I were expressed together, as on the surface of the C32 melanoma cells, there was preferential binding to CD36. The weight of evidence suggests that CD36 is the most important of the candidate receptor molecules thus far identified. However, the true role of these molecules, and others perhaps yet undiscovered, in P . fakiparum sequestration will undoubtedly require more investigation. Further studies must also take into account the distribution and density of the endothelial ligands on different tissues, in order to reconcile the variability in end organ damage seen in patient populations of differing age and background immunity. E.


There is relatively little information on the parasite ligands involved in cytoadherence. Although a number of parasite-derived proteins has been detected on infected erythrocytes (Howard, 1987; Hommel and Semoff, 1988), it has been assumed that only protein(s) which protrude from the surface of the membrane would be likely to mediate cytoadherence. So far only P . falciparum erythrocyte membrane protein (PWMPI) has been shown unequivocally to have this property (Leech et af., 1984). PfEMPl consists of a family of molecules (molecular mass 240-260 kDa) in which only the higher molecular mass variants are associated with cytoadherence (Magowan et af., 1988). A similar molecule of molecular mass = 270 kDa, called sequestrin, has recently been demonstrated on the surface of infected erythrocytes using anti-isotypic antibodies raised against OKM8, a monoclonal antibody specific for the putative endothelial receptor CD36 (Ockenhouse et af.,



1991a). This finding further strengthens the hypothesis that CD36 is the receptor for the parasite ligand on vascular endothelium. The definitive proof of the role of these parasitized red cell surface proteins in cytoadherence awaits the production of specific isotypic antibodies and/or the cloning of the genes encoding these antigens. Two other molecules have been proposed as the cytoadherence parasite ligand, although the evidence supporting them is far less convincing. The 155 kDa ring-infected erythrocyte antigen (RESA) antigen, which is transferred from the merozoite to the erythrocyte membrane during invasion, was thought initially to be entirely submembranous, but recent evidence suggests that part of the molecule is exposed on the exterior of the red cell as the parasite matures. RESA could therefore have a role in cytoadherence. The RESA antigen has been shown to have cross-reactive epitopes with band 3 protein (Holmquist et al., 1988), the human erythrocyte anion transporter, and this too has been implicated as a ligand for cytoadherence (Winograd and Sherman, 1989). Presumably, changes in the erythrocyte cytoskeleton which occur as a result of parasitization expose previously hidden host molecules (neoantigens) on the cell surface. At present there is considerably less evidence to support a role for RESA or modified band 3 in cytoadherence than for PfEMPl, but the situation is far from resolved. Regardless of the eventual identity of the cytoadherent ligand, a conserved component must be present since all P. falciparum parasites causing natural infections cytoadhere. In addition, there must be a strain-variable component since inhibition or reversal of cytoadherence by immune sera occurs in a strain-specific manner (Udeinya et al., 1983; Singh et al., 1988). Indeed, the cytoadherence surface proteins show antigenic variation within cloned parasite lines in a manner analogous to the schizont-infected cell antigen (SICA) of P. knowlesi. The constant and variant components could be either closely associated molecules or different epitopes on the same molecule. This ability of the parasite to vary the surface antigenicity of the cytoadherent protein is obviously important for its survival as it helps to evade host recognition and thus parasite removal. F. ROSETTING

Non-parasitized erythrocytes will agglutinate around red cells containing mature forms of the parasite in vitro (David et al., 1988; Udomsangpetch et al., 1989b). This phenomenon is termed rosetting and may sometimes be seen in fresh blood samples (Ho et al., 1991). It shares many characteristics with the properties of cytoadherence. Rosetting occurs only with species of Plasmodium which also exhibit cytoadherence (Handunnetti et al., 1989). Both phenomena occur with mature stages of P . falciparum and begin after approximately 26 h of intra-erythrocytic development (David et al., 1988).



Rosetting can be reversed by immune sera which also reverse cytoadherence (David et al., 1988). The antigens responsible for rosetting and cytoadherence are both very protease-sensitive (Udomsangpetch et al., 1989b). Both properties are maximal at acidic pH, but rosetting is inhibited by heparin and EGTA*, a calcium chelator, whereas cytoadherence is not (Carlson et al., 1990a). Interestingly, rosetting is inhibited by one anti-CD36 monoclonal antibody (OKM8) but not by another (OKM5). In the perfused rat mesocaecum model, human erythrocytes containing mature P . falciparum parasites of known rosetting lines (K+ R + ) cause more microvascular obstruction than infected cells from cytoadherent but non-rosetting lines (K+ R-). They cause greater resistance to microvascular flow, and the erythrocytes aggregate readily in the larger venules (Kaul et al., 1991). There have been several recent studies of rosetting in parasite isolates obtained from patients with acute falciparum malaria. Rosette formation varied considerably: whereas all fresh isolates cytoadhered to some extent, not all isolates showed rosetting (Wahlgren et al., 1990). Rosetting is associated with cerebral malaria. Parasites obtained from Gambian children with cerebral malaria showed a significantly greater degree of rosetting than those from children with uncomplicated disease (Carlson er al., 1990b). Patients with cerebral malaria also lacked antibodies which could inhibit rosette formation in vitro, whereas these antibodies were present in about 20% of immune sera. No significant correlation between rosetting and biochemical indices of disease severity was seen in a smaller population of adult patients in Thailand (Ho et al., 1991b) but, as in the Gambian series, rosette formation tended to be greater in patients with cerebral malaria. In this latter series there was a significant inverse correlation between rosetting and cytoadherence for a given isolate. Cytoadherence was not increased when rosetting was partially inhibited by heparin, which suggests that the inverse relationship is not the result of steric hindrance. Cytoadherence and rosette formation properties are probably intrinsic to the parasites, with individual parasite isolates having greater propensity for one or the other (Ho et al., 1991b). This is consistent with the observation that isolates from patients with cerebral malaria have increased rosetting properties, but adhere poorly to C32 melanoma cells in vitro (Marsh et al., 1988; Ho et al., 1991a). Whether the inverse relationship between cytoadherence and rosette formation holds true in vivo remains to be determined. However, it is likely that once a parasitized erythrocyte becomes encased in a rosette there would be a considerable physical obstacle to cytoadherence-so in larger venules the two processes must be to some extent mutually exclusive. The evidence so far linking rosette formation to the pathogenesis of cerebral malaria is suggestive but not conclusive. Unlike cytoadherence, * Ethyleneglycol-bis-(P-aminoethylether)-~,,N,llr,K-tetraacetic acid.



rosettes have not been prominent findings in any histological studies of clinical or post-mortem tissues in falciparum malaria. Significant rosette formation on the arterial side of the microcirculation appears unlikely, as arteriolar obstruction is not a pathological feature of severe malaria. Presumably the low pH and shear forces on the venous side tend to favour rosetting (Howard and Gilladoga, 1989), and this can be seen ex vivo in the perfused rat mesocaecum model (Kaul et al., 1991), but if complete vascular obstruction occurs one would still expect the tail of blood into the capillary (a)

FIG.6. (a) Mechanical separation using micropipettes of a red cell containing a mature Plasmodium falciparum parasite (left) and an uninfected erythrocyte (right). These are the adhesive properties that cause rosetting (courtesy of Dr G. Nash). (b) Hypothetical scheme for the role of rosettes in sequestration. At point A the vessel diameter is sufficient to allow passage of the formed rosette. At point B the rosette interacts with the vessel wall. Flow may be obstructed, or the rosette could disrupt flow allowing cytoadherence to take place (C). (Reproduced with kind permission from Howard and Gilladoga, 1989.)



to reflect the distribution of parasitized erythrocytes in the circulation. In fact, the majority of the erythrocytes sequestered in the cerebral vessels in cerebral malaria are parasitized; i.e., sequestration is selective. The cell-tocell adhesive forces involved in rosetting have been shown experimentally to be considerable and capable of withstanding intravascular shear forces (Fig. 6). It may be that the adhesive properties of red cells containing mature parasites (i.e. the processes that lead to rosette formation in vitro) may contribute to a reduction in forward blood flow, which would then initiate or “encourage” cytoadherence as a secondary phenomenon (Fig. 6b). The passage of uninfected red cells squeezing past the sticky cytoadherent infected cells would also be slowed by this process (static hindrance). A circulating rosette would presumably become obstructed in the pulmonary capillary bed or, if it traversed this successfully, it would then be vulnerable to splenic removal. Rosetting is currently a subject of considerable interest. Its pathophysiological role, particularly in relation to cerebral malaria, will, hopefully, become evident over the next few years. VIRULENCEFACTORS V. PARASITE A.


Infection begins with the inoculation of malaria sporozoites from a female anopheline mosquito probing for a blood meal. The sporozoites migrate rapidly via the lymphatics and the bloodstream to the liver. Each sporozoite which successfully invades a hepatocyte will subsequently develop into a meront (schizont) containing many thousands of merozoites. For P. vivax and P . ovule, development will be delayed in a proportion of invaded cells, and in these hepatocytes the parasites lie dormant as hypnozoites. Months or years later they will “awaken” and develop, and cause the relapses of infection characteristic of these species. Each fully developed hepatic meront of P. vivax, P. ovule and P . malariae (i.e. the benign human malaria parasites) will burst to liberate up to 15 000 merozoites into the host’s bloodstream, whereas the hepatic schizonts of the potentially lethal P. fakiparum contain 30 000 or more merozoites (Garnham, 1966). Furthermore, the exoerythrocytic phase of development (liver merogony) of P. falciparum takes an average of 5.5-7 days (and may be shorter), compared with 7 days or longer in the benign malarias. Thus, early amplification of the infection is greater in the potentially lethal falciparum malaria (Garnham, 1988). The current estimates for the median P. falciparum sporozoite inoculum are 8-15 sporozoites-but the distribution is skewed-and on occasions as



many as 100 sporozoites may be inoculated (Rosenberg, R. and Wirtz, 1990; Ponnudurai et al., 1991). If these estimates are correct, and they are very difficult to prove, then between 3 x lo5 and 3 x lo6 merozoites of P. falciparum are released into the bloodstream to start the asexual erythrocytic cycle. This begins the symptomatic phase of the infection. The merozoites are motile ovoid bodies structurally similar to sporozoites. They rapidly invade passing erythrocytes, and then proceed to consume the red cell contents, growing progressively larger through the trophozoite stage, to become multinuclear meronts inthe last 12 h of the cycle. After 48 h of intraerythrocytic development, the P . falciparum meront bursts to release 18-24 daughter merozoites (estimates have ranged from 8 to 36 per red cell), compared with 12-16 for P. vivax and 8 for P. ovale (Garnham, 1988). Asexual multiplication in vivo is approximately 30-50% efficient during the early phase of infection and, in the non-immune host, the parasitaemia rises by roughly an order of magnitude every 2 days (Fairley, 1947; Kitchen, 1949). At this rate the threshold of thick film microscopical detection (ca. 50 parasitized erythrocytes per pl) is reached in three to four asexual cycles, i.e. 12-16 days after hepatic merogony. If multiplication is unchecked, potentially lethal parasite burdens are reached in another three to four cycles (Fig. 7). It is obvious from this that a higher multiplication rate is associated with


10% parasitaemia


1% parasitaernia

1012 :


pbu ”7 :;

I I I 0I

108 -

,i‘: 8

Detection limit (50pL 1)





100 sporozoites ’



e 1


*.-. r



, ,.-.._,

; : ,-.,;;--*.2 ,--.*-

106 104

! !




*a t



1010 -


,.--. ;



,--. ’ *-.‘













.-- :,-.* .-., ,--=.-/ :



1 sporozoite

102 - MOSaUltO blt e C

* I



Erythrocytic cycle















FIG.7. Plasmodium falciparum infection: expansion of the parasite burden following different sporozoite inocula, and at two different multiplication rates in vivo ( x 6 and x 10) for the larger inoculum. (Reproduced from White and Krishna, 1989.)



a shorter duration of illness before reaching dangerous parasite burdens, and that young children begin the asexual phase of parasite development at a higher parasitaemia, as they have a smaller blood volume in which to dilute the merozoites liberated by hepatic merogony (White and Krishna, 1989). For example, if one assumes that all the merozoites released at merogony invade a red cell, then the combination of a single large viable sporozoite inoculum, or multiple inocula (e.g. 100 successful parasites) and highly efficient asexual multiplication (15-fold per cycle) in a young child (blood volume 500 ml) would result in potentially lethal parasitaemias (250 000 per pl) in only four cycles (8 days) from hepatic merogony. It is not surprising, therefore, that the length of history of children who die from cerebral malaria can be very short (Molyneux et al., 1989b). 1. Stabilization of parasitaemia

Most untreated episodes of falciparum malaria are not fatal. The log-linear rise in parasite numbers is checked, and the host reaches an equilibrium with the parasite burden, and later effects its removal from the body. With repeated infections in areas of intense transmission, the level at which parasitaemia stabilizes falls; also, the threshold for symptoms rises (Kitchen, 1949). Eventually parasitaemias become asymptomatic, a state known as premunition. Concurrently the risks of unrestrained parasite multiplication to lethal burdens declines. Several factors converge to limit parasite multiplication. These include increased splenic clearance function, the inhibition of merozoite invasion by antibodies, the humoral immune response with antibody opsonization of red cells and merozoites, various antibody-independent cellular effector mechanisms, exhaustion of more receptive erythrocytes (certain red cells are more susceptible to invasion than others), and non-specific activation of host defences leading to release of cytokines and other toxic species that directly or indirectly damage parasites (White and Krishna, 1989). However, equilibration of the parasite population can occur in the absence of these factors. For example, a simple feedback mathematical model in which fever (temperature >40°C) inhibits merogony (Kwiatkowski, 1989) illustrates this tendency to equilibrate (Kwiatkowski and Nowak, 1991)-but also, as in other biological systems, shows fundamental instability (chaotic dynamics) of the parasite population when multiplication rates are high. Such a system has a tendency to “overshoot”-i.e. the number of parasites might become very large indeed before the feedback factor reduces the population. The overshoot may prove fatal. The benign human malarias have an inbuilt brake on their capacity to multiply; P . vivax and P . ovule have a predilection for young erythrocytes, whereas P . malariae is thought to invade mature cells only (Garnham, 1988). P . falciparum also preferentially invades younger erythrocytes (Pasvol et al.,



1980), but can invade cells of all ages-hence the propensity for untrammelled multiplication to reach lethal parasite burdens (Field, 1949). P . falciparum does not develop well in erythrocytes containing haemoglobin F (Pasvol et al., 1976, 1977), which partly explains why malaria is rarely severe in the first few months of life. B.



Fever patterns

The synchronicity of malaria is such a characteristic feature of the infection that the 2 or 3 day periodicity has been enshrined in the terminology of the disease. If the infection is not treated and reaches a host-parasite equilibrium, fever spikes usually occur every 1 (quotidian), 2 (tertian), or 3 days (quartan) (Kitchen, 1949).* Classically P . malariae infection shows a quartan pattern, and the other three human malarias are tertian. The fever is caused by “sporulation” (merogony). There is some evidence that synchronous sequestration can also cause fever (G. Q. Li, personal communication). Nowadays the classical malaria fever charts are rarely Seen because the diagnosis is made early, and appropriate treatment is given before there is time to document the fever pattern adequately. P . malariae, P . ovule and P . vivax tend to synchronize earlier than P . falciparum and intermittent “paroxysms” with chills and rigors are more common in the initial phases of these infections. Indeed P . falciparum rarely shows regular periodicity of fever early in the development of the infection-the pattern of temperature is more usually erratic. Infections may also segregate into more than one brood (James, S. P. et al., 1932), i.e. bimodal or multimodal parasite age distributions. This may cause daily fever spikes (quotidian fever) with alternating “sporulation”, or more complex fever patterns with multiple broods. Although widely practised once, the validity of extrapolating from the fever chart to the number of infecting parasite “broods” has not been established fully. Overall, the fever pattern is seldom useful in diagnosis. 2 . Rigors Synchronicity is important in the pathophysiology of malaria as many of the symptoms of the disease are related to synchronous rupture of meronts with * The apparent terminological contradiction (tertian for two days and quartan for three) arises from the early malariologists’ adoption of the Roman practice of counting the first day of a series as day 1 rather than day 0.



the release of cellular debris and a variety of parasite proteins, glycoproteins and glycolipids into the circulation. This may explain why, in non-immune subjects, the symptoms of P. vivax malaria (highly synchronous) are often worse than those of the less synchronous P. falciparum early in uncomplicated infections. The rigor characteristic of established untreated infections‘ consists of shaking and shivering, a sensation of coldness, irritability and malaise, accompanied by “gooseflesh”, tachycardia and rapidly rising temperature. This is followed by a “flush phase” in which the patient feels warm, vascular resistance falls and there is often profuse sweating. Rigors are unpleasant, but very seldom lethal. Contrary to conventional wisdom, they are unusual in falciparum malaria. A glycoprotein (glycolipid in murine malaria) with many of the properties of bacterial endotoxin is released at merogony. This induces macrophages to release tumour necrosis factor (TNF), the pivotal agent in the cascade of cytokine release (Kwiatkowski et al., 1989). The rigor or “paroxysm” may result from a pulse of cytokine release which causes systemic toxicity and resetting of the hypothalamic thermostat. 3. Distribution of parasites in the body

The synchronicity of P. falciparum malaria determines the discrepancies between peripheral blood parasitaemia and total body parasite burden which result from deep vascular sequestration (see later) and often perplex, and may mislead, the examining physician (White and Krishna, 1989; Davis et al., 1991). Synchronicity determines the proportion of P. falciparum parasites that are circulating in the bloodstream (i.e. are relatively harmless) compared to those more harmful forms sequestered in the microvasculature-i.e. the relationship between peripheral parasitaemia and the total parasite biomass. Pathophysiological processes induced by sequestration and its metabolic sequelae, or by rupturing meronts (cytokine release), are proportional to the number of these unseen mature parasites (White, 1985), and are most evident in synchronous infections at the time when the majority of parasites are in the pathological stage (i.e. sequestered). Synchronicity is also relevant to the therapeutic response to antimalarial treatment. The antimalarial drugs affect principally large ring and young trophozoite stages of parasite development (i.e. the middle third of the 48 h erythrocytic life cycle) (Yayon et al., 1983; Zhang et al., 1986; F. ter Kuae et al., unpublished observations). Obviously, for drugs where blood concentrations fluctuate widely (e.g. chloroquine, artesunate), the greater the proportion of parasites that are at the vulnerable stage of development at the time of peak blood antimalarial drug concentration, the greater will be the antiparasitic effect.




As with all pathogens, there are important biological differences between different strains of malaria parasites. In the era of malaria therapy of neurosyphilis, it was well known that parasites of the same species obtained from different geographical locations differed in their infectivity to mosquitoes, antimalarial drug sensitivity, and virulence. For example, the European strains of P . falciparum were infectious to the English Anopheles mosquito ( A . atroparvus), whereas African and Indian strains were not. The European strains were also more dangerous, with a tendency to produce severe disease rapidly (Shute and Maryon, 1954). In experimentally infected Aotus monkeys there are also considerable differences in virulence between strains of P . falciparum (Schmidt, 1978). The intrinsic ability of a strain to cytoadhere may be one virulence factor (Ho et al., 1991a). Intrinsic multiplication capacity is almost certainly another (although several factors determine the multiplication rate in vivo). Antimalarial drug resistance is an obvious advantage to the parasite, as is the ability to change rapidly its repertoire of antigens expressed on the red cell surface. Satisfactory typing methods which correlate parasite strains with these putative virulence factors have yet to be developed, and the relative importance of strain differences in the pathophysiology of falciparum malaria remains to be elucidated.


The concept that the symptomatology of severe malaria may be mediated by products of activated macrophages was first proposed by I. A. Clark et al. (1981), based on the similarities between the features of severe malaria and endotoxaemia. Over the subsequent years, much evidence has emerged to support the idea that cytokines, in particular TNF, may indeed play an important role in ctlusing some of the pathological changes that characterize malaria. A.


1. Toxicity to the host Most of the experimental data in support of a role for TNF in the pathogenesis of cerebral malaria have come from studies of P . berghei (ANKA strain) infection in CBA/Ca mice. The neurological signs which develop in these mice result from an immunopathological reaction which is strictly dependent on the presence of activated CD4+ T cells (Grau et al., 1986). The syndrome is associated with elevated levels of circulating TNF, and can be prevented by the administration of polyclonal antibodies to TNF



(Grau et al., 1987b,c). The treatment, however, does not influence the parasitaemia, and protected mice eventually die of anaemia. Furthermore, monoclonal antibodies to certain other cytokines, e.g. the combination of anti-interleukin 3 and anti-granulocyte macrophage-CSF (anti-GM-CSF) (Grau et al., 1988b), as well as anti-y interferon (IFN-y) (Grau et al., 1989d), can also prevent cerebral symptoms through a reduction in TNF production. The administration of recombinant TNF to mice infected with P . berghei, and which are resistant to cerebral symptoms, apparently induces the lethal neurological syndrome (Grau et al., 1989a,b,c). Interleukin (IL) 6 levels are also elevated in this model (Grau et al., 1990), but this cytokine seems less directly involved in pathology; high IL-6 levels also occur in the absence of cerebral pathology in animals infected with non-lethal P . yoelii parasites, and anti-IL-6 antibody does not protect against the development of cerebral symptoms. The following sequence of events is proposed. In genetically susceptible strains of mice, CD4+ T cell activation during acute malaria leads to the production of cytokines which “upregulate” a variety of macrophage functions, one of which is the release of TNF. Elevated TNF levels in the circulation alter the surface properties of endothelial cells and cause the local accumulation of leucocytes. These sequestered leucocytes in turn release more TNF, thus amplifying the cytotoxic effects on endothelial cells, with resultant vascular wall damage and haemorrhagic necrosis. In this murine model, sequestration of leucocytes and monocytes is considered to be more important than that of infected erythrocytes. Cytokines have also been implicated in the pathogenesis of anaemia, hypoglycaemia and pulmonary oedema in rodent models. Bone marrow dysfunction results from TNF-induced dyserythropoiesis and erythrophagocytosis (Clark, I. A. and Chaudhri, 1988a; Miller, K. L. et al., 1989). If recombinant TNF is infused into mice infected with P . vinckei, hypoglycaemia, midzonal liver necrosis and neutrophil adhesion in pulmonary vessels occur (Clark, I. A. et al., 1990). These features are commonly seen in terminal P . vinckei infection in association with high levels of circulating TNF. Administration of TNF has also been shown to cause foetal death in pregnant mice infected with P . vinckei (Clark, I. A. and Chaudhri, 1988b). 2.

Toxicity to the parasite

In contrast to the deleterious effect of high levels of TNF on the host, small doses of parenteral IFN-y and TNF have been shown to reduce parasitaemia in P . chabaudi adami (Clark, I. A. et al., 1987) and P . yoelii (Taverne et al., 1987) infections. More recently, low doses of IL-I were found to protect C57BL/6J mice against cerebral pathology induced by P . berghei and also to reduce parasitaemia, although the two effects were separate (Curfs et al.,



1990). These observations have led to the concept that low levels of cytokines could be beneficial by exerting an indirect antiparasitic effect on blood-stage parasites, but high concentrations of the same cytokines may act in concert to produce toxic damage in the host. The antiparasitic effect is known to require other serum components since recombinant T N F and IFN-y are not directly cytotoxic to intra-erythrocytic parasites in vitro (Taverne et al., 1987), and IL-1 has no effect when given to T cell-deficient animals (Curfs et al., 1990). The additional toxic serum components remain to be identified, although the killing of blood-stage P. falciparum in vitro by serum containing T N F results mainly (> 70%) from the lipid peroxide content (Rockett et al., 1988). These peroxides are formed by the interaction of lipoproteins with reactive oxygen intermediates and are unaffected by antioxidants. Malaria parasites are readily killed by free radicals (Malhotra et al., 1988). Lipid peroxidation has the effect of stabilizing the reactive oxygen groups, and thus creating a more stable cytotoxic molecule than other oxygen radicals. B.


Soluble antigens of P. falciparum, released into culture supernatants and also found in the plasma of patients with acute malaria, have been shown to induce T N F release from macrophages (Bate et al., 1988, 1989; Taverne et al., 1990a,b). Two of these antigens, designated Agl and Ag7, are both glycoproteins of molecular mass 60-70 kDa. Like bacterial lipopolysaccharide (LPS), they react in the Limulus assay, and the activity of Ag7 is reduced by polymyxin B. However, C3H/HeJ mice, which are hyporesponsive to LPS, secrete T N F normally in response to these antigens (Riley et al., 1988), suggesting interaction with a different macrophage receptor from LPS. They appear to act directly on monocyte/macrophages without a requirement for accompanying T cell activation. This is consistent with the observation that supernatant preparations containing the two antigens do not stimulate peripheral blood T lymphocytes during acute malaria (Theander et al., 1986). A similar, T-independent TNF-inducing antigen (a glycolipid) has been identified in rodent malaria parasites. Premunition probably represents “tolerance” to these LPS-like substances-i.e. their release during malaria infection stimulates progressively less cytokine release (Bate et al., 1990; Playfair. et al., 1990). Thus “anti-disease” rather than “anti-infection” immunity is induced (Riley et al., 1991). C.


How relevant are these diverse observations in selected rodent models to severe malaria in man? A positive association between plasma concen-



trations of TNF and mortality in severe falciparum malaria has been observed in three studies (Grau et al., 1988b; Kern er al., 1989; Kwiatkowski et al., 1990), but could not be confirmed in a recent study of children in Zaire (Shaffer et al., 1991). In these studies TNF levels were correlated with several indicators of severity, namely hypoglycaemia, hyperparasitaemia and anaemia. Levels of IL-1 (Kwiatkowski er al., 1990), IL-6 (Kern et al., 1989) and IL-8 (J. Friedland, personal communication) have also been shown to be correlated with disease severity. In all these studies, there was considerable overlap in the range of cytokine levels in the different patient groups (Fig. 8).

FIG.8. Relationship between plasma concentrations of tumour necrosis factor (TNF) and disease severity in Gambian children with acute falciparum malaria. (Reproduced with kind permission from Kwiatkowski et al., 1990.)

There are several problems with the cytokine story in human malaria. First, cytokine levels are high in malaria due to P.vivax as well as that due to P.fakiparum-but P. vivax does not kill. Second, there are differences between the clinical manifestations of severe malaria and lethal septic shock-a condition in which there is good evidence for cytokine-mediated pathology (Parillo et al., 1990). Third, the clinical and pathological features of so-called “cerebral malaria” in the P. berghei-infected mouse model (where much of the evidence supporting a role for cytokines in malaria pathology has been obtained) are quite different from those seen in human disease. In humans, there are no haemorrhagic foci or focal intravascular accumulations of large mononuclear cells, and endothelial cell damage is




minimal. Furthermore, unlike P . fakiparum, P . berghei parasites do not sequester in the brain. Indeed, use of the term “cerebral malaria” to describe this neurological syndrome in rodents is potentially misleading. The pathology in the P.vinckei-infected mouse model is also very different from that in severe falciparum malaria. Third, it is now known that peak TNF production occurs at merogony (Kwiatkowski et al., 1989) and the cytokine is then cleared rapidly from the circulation with a half-life of 5-20 min (Michie et al., 1988). A single measurement in blood is therefore of limited value. Moreover, the commercial enzyme-linked immunosorbent assays used in most studies to date measure both free biologically active TNF, as well as TNF bound to circulating receptors. There is also evidence that the bioactive receptor-binding moiety of TNF is an unstable non-covalently linked oligomer that tends to form inactive mono- and polymers in solution. Both oligo- and polymers are measured in current assays. These factors account for the sometimes large discrepancies between the levels of immunoreactive (usually high) and bioactive (usually absent) TNF (Petersen et al., 1989). Finally, the pathological effects of TNF and other cytokines on the host must also depend on the variable density of receptors on different target organs, and individual variations in sensitivity. TNF obviously plays a central role in cytokine-induced pathology, but other cytokines are almost certainly involved too, and their roles may become apparent only when assay methods become more reliable and more widely available. The importance of cytokines in human malaria remains to be determined. At present it seems that TNF could well be involved in the pathogenesis of fever, hypoglycaemia and haemopoietic suppression in malaria, is of uncertain relevance to pulmonary oedema, abortion and circulatory failure, and is unlikely to cause coma in cerebral malaria. The critical question of whether it kills patients directly remains to be answered. These uncertainties may be resolved partly by a study of the therapeutic effects of administering a monoclonal antibody against TNF to patients with severe falciparum malaria.


In severe malaria there is multiple organ dysfunction. Those organs with an obligatory high metabolic rate are particularly affected. A.


Despite intense sequestration of parasitized erythrocytes in the myocardial microvasculature, pump function appears to be remarkably good in severe


1 I5

falciparum malaria (Sprague, 1946). Echocardiography usually indicates an increased ejection fraction (derived from measurement of fractional shortening) (Charoenpan et al., 1990). Transiently reduced ejection fraction with very high central venous pressure (which responded well to inotropes and loop diuretics) has been reported (Le Camus et al., 1988), but this may have resulted from volume overload. Most patients are admitted with an elevated cardiac index (> 5 1 m-' min- ') with low systemic and pulmonary vascular resistance, and low or normal right- and left-sided filling pressures (James, M. F. M., 1985; White, 1986). Similar haemodynamic profiles are observed in bacterial septicaemia, although hypotension is more prominent, and there is clear evidence of a sepsis-induced depression of left and right ventricular systolic function (Parillo et al., 1990). TNF is one of a number of circulating low molecular weight myocardial depressant substances responsible for these negatively inotropic effects in bacterial septicaemia. A detailed study of ventricular performance in severe falciparum malaria would be of value in differentiating the pathological effects of T N F in the two conditions. Dysrhythmias are rare in severe malaria; 24-h electrocardiographic recordings are usually unremarkable even in ultimately lethal infections (F. Nosten and N. J. White, unpublished observations). Blood pressure is usually at the lower end of the normal range with sinus tachycardia. Orthostatic hypotension is common (Butler and Weber, 1973; Kofi-Ekue et al., 1988) and may be profound, particularly with high fever, even in otherwise uncomplicated malaria. There is an associated failure of reflex cardioacceleration, suggesting autonomic dysfunction. Postural hypotension is worsened by the quinoline antimalarial drugs (W. Supanaranond et al., unpublished observations). In very severe falciparum malaria, cardiac index and blood pressure may fall secondary to metabolic acidosis, hypoxaemia and, in some patients, supervening Gram-negative septicaemia (Bygbjerg and Lanng, 1982). Visceral perfusion is reduced in monkeys infected with P. knowlesi, particularly at the time of merogony, and in some cases this is reversible by a blockade (Skirrow et al., 1964). Vasoconstriction may contribute to the reduction in vital organ perfusion in falciparum malaria, although there has been no study in man to confirm this suggestion (Section VII.D.2). B.


The rapid development of shock in severe malaria is termed algid malaria (Sullivan, 1876; Gage, 1926). This is an enigmatic condition, and appears to have several aetiologies. Shock may result from stress-induced gastrointestinal bleeding (i.e. hypovolaemia), severe hypoxaemia and acidosis, or Gramnegative septicaemia. Spontaneous bacterial septicaemia caused by Enterobacteriaceae or pseudomonads is an important cause of sudden clinical deterioration in severe falciparum malaria (Bygbjerg and Lanng, 1982;



WHO, 1986). Endotoxaemia has been reported in severe malaria by several groups, based on positivity in the Limulus lysate assay (Tubbs, 1980; AungKyaw-Zaw et al., 1988; Usawattanakul et al., 1985), but whether this represents ‘‘leakage’’of endotoxin across the gut and past the hepatic barrier, or measurement of the endotoxin-like malarial glycoprotein (Taverne et al., 1990a,b), is uncertain. Septicaemia caused by Salmonella spp. is particularly common in African children with P.falciparum infections (Mabey et al., 1987), but this more often presents with protracted fever rather than shock. Whether algid malaria exists as a discrete entity, i.e. primary hypotension leading to secondary type A lactic acidosis, remains to be confirmed, but most clinicians managing severe malaria have encountered patients who develop hypotension rapidly and without apparent cause-and in whom blood cultures are later shown to be sterile. The pathogenesis of this condition is unknown, although it could result from pulse release of a large amount of a vasoactive mediator such as T N F (Parillo et al., 1990) (Section V1.C). C.


Early workers considered that malaria caused pneumonia directly (Falconer, 1919), but the “intermittent pneumonia” they described was almost certainly pulmonary oedema. Bacterial pneumonia may complicate severe falciparum malaria (Spitz, 1946), particularly if there has been aspiration by a comatose patient. The two conditions can be very difficult to distinguish clinically. However, in most patients lung function and oxygenation are normal. Adults with severe falciparum malaria may develop acute non-cardiogenic pulmonary oedema at any stage of their illness (Bergin, 1967; Deaton, 1970; Fein et al., 1978; Martell et al., 1979; Blanloeil et al., 1980; James, M. F. M., 1985; Feldman and Singer, 1987; Bernadin et al., 1989; Charoenpan et al., 1990). Pulmonary oedema commonly coexists with other vital organ dysfunction (Brooks et al., 1968). It is a particular problem of severe malaria in pregnancy, but is rare in children. Hypoalbuminaemia, hyperparasitaemia and renal failure are common associations. The development of frank pulmonary oedema in malaria carries a mortality rate of approximately 80% despite treatment. As in other circumstances in which the adult respiratory distress syndrome occurs, the precise cause is not known (Gurman et al., 1988). Whether increased pulmonary capillary permeability occurs more commonly than is clinically manifest in severe malaria remains to be determined. Certainly patients with severe malaria are very vulnerable to fluid overload and have a low threshold for developing iatrogenic pulmonary oedema (Hall et al., 1975). Hypoalbuminaemia is a contributory factor by reducing the oncotic pressure of plasma, and thus lowering the pulmonary capillary pressure threshold at which pulmonary oedema will develop.


1 I7

Ultrastructural studies of the lung have not helped to elucidate the pathophysiology. Hyaline membrane formation in the alveoli suggests leakage of proteinaceous fluid (Charoenpan et af.,1990). The alveolar septa1 walls may also be thickened. Cytoadherent parasitized erythrocytes may be seen if the patient dies rapidly (Duarte et al., 1985; Corbett et al., 1989), and large numbers of leucocytes may be seen both in the vascular lumen and adherent to the vascular wall (Pongponratn et al., 1991). These are mainly pigment-containing macrophages, although neutrophils are also evident (MacPherson et al., 1985; Corbett et al., 1989). It has been suggested that pulmonary capillary damage might be caused by release of toxic mediators from these adherent mononuclear cells. In one ventilated patient, open lung biopsy on the 11th day in hospital showed florid diffuse fibrosing alveolitis (Feldman and Singer, 1987), but this is a common and non-specific sequel to prolonged ventilation and secondary infection. D.



Cerebral bloodflow

Although the cardiac index is usually high, and systemic vascular resistance is low in severe malaria (White, 1985), flow in some vital organs may be reduced. In health, autoregulation ensures that blood flow to the brain provides sufficient oxygen and substrates for metabolic demands. In 12 patients with cerebral malaria, cerebral blood flows were within the range considered normal in healthy adults, but were considered inappropriately low for the arterial oxygen content (i.e. oxygen supply) and the augmented metabolic demands associated with fever and infection (Warrell et al., 1988). The cerebral metabolic production of lactate was increased during coma, but fell to normal on recovery of consciousness. In a separate study, CSF concentrations of lactate were found to be elevated consistently in cerebral malaria (White et af., 1985), and were significantly higher in fatal cases than in survivors (Fig. 9). CSF lactate was inversely correlated with CSF concentrations of glucose, and values also returned to normal with recovery of consciousness. Similar findings have been reported in children (White et al., 1987b; Molyneux et al., 1989b). These observations all indicate anaerobic glycolysis within the brain in cerebral malaria. This results presumably from interference with microcirculatory flow (i.e. ischaemia causing hypoxia), but in addition could reflect a flow-independent shift to anaerobic respiration (e.g. inhibition of citric acid cycle activity by toxic moieties such as cytokines or other secondary products). Obligatory anaerobic glycolysis by the sequestered parasites also contributes to local lactate accumulation. But coma in cerebral malaria cannot be explained simply by hypoxia;



volunteers breathing low oxygen mixtures remain conscious with cerebral metabolic rates for oxygen which are lower than those observed in cerebral malaria (Kety and Schmidt, 1948). Other factors must contribute. MeantSD I4








a a


CSF Lactate m moll1


# 6



a a







FIG.9. Cerebrospinal fluid lactate concentrations in cerebral malaria. (Reproduced from White et al., 1985, with permission.)


Liver and skeletal muscle bloodjlow

In uncomplicated malaria, liver blood flow (LBF) is increased in parallel with the rise in cardiac index, but in severe malaria flows are variably reduced (Molyneux et al., 1989a) (Section VI1.L). In a recent study using indocyanine green clearance as a measure of LBF, the median value was


1 I9

10.7 ml kg- min- in six fatal cases of falciparum malaria compared with 15.6 ml kg- min-’ in survivors. There was also a significant inverse correlation between venous lactate concentrations and LBF at flows less than 15 ml kg- min- This suggests that reduced LBF could contribute towards liver dysfunction and lactic acidosis in life-threatening infections (Pukrittayakamee et al., in press). Recent studies of forearm blood flow and metabolism in severe malaria indicate that a shift to anaerobic glycolysis also occurs in skeletal muscle. However, there is no evidence that muscle blood flow is reduced sufficiently to impair or retard intramuscular drug absorption in severe malaria (White, 1985; Waller et al., 1991). The increase in lactate production in these various organs cannot be explained simply by reduction in measured total organ blood flows, and it is unlikely to be explained solely by parasite glycolytic metabolism. This suggests that either microcirculatory obstruction is variable or “patchy” (i.e. low flow in some capillaries, high flow in adjacent vessels) or that the cytoadherent parasitized erythrocytes interfere with gas and substrate exchange and host metabolic functions but do not cause a large increase in the resistance of the circuit. In either case it is likely that normal erythrocytes can squeeze past the adherent parasitized red cells (White, 1986). Otherwise, permanent ischaemic neuronal damage would be the rule rather than the exception following cerebral malaria.





Vascular and metabolic abnormalities

The cause of coma in severe falciparum malaria is not known. Although there is ample evidence for disturbances of microcirculatory flow and consequent anaerobic glycolysis, it is not clear whether this is sufficient to cause unconsciousness. The importance of locally produced or circulating toxic mediators (such as cytokines) is also unresolved. Ultimately it is likely that a combination of impaired flow and metabolic derangement resulting from the adherent highly metabolically active parasitized cells, and local toxicity leads to a derangement of neuronal function. Whether there are specific abnormalities of neurotransmitter release or sensitivity remains to be determined. However, in the limited studies to date there have been no consistent alterations in the pattern of CSF amino acids, and no therapeutic responses to the use of benzodiazepine or opiate antagonists which would point to involvement of a specific neurotransmitter pathway (N. J. White, unpublished observations). At post-mortem examination, multiple small haemorrhages are evident



throughout the white matter of the brain (Spitz, 1946). Retinal haemorrhages are also commonly seen in cerebral malaria (Looareesuwan et al., 1983a), particularly with the use of indirect ophthalmoscopy and fluorescein angiography (E. Schulenberg, T. M. E. Davis and N. J. White, unpublished observations) (Fig. 10). The microvascular damage in the connecting pathways appears to be related to sequestration rather than thrombus formation and may contribute to nervous system dysfunction, seizures and possibly residual sequelae. 2.


The hypothesis that cerebral malaria results from an immune or allergic vasculopathy has been raised by several pathologists (Rigdon and Fletcher, 1945; Schmid, 1974; Toro and Roman, 1978). The diffuse white matter petechial haemorrhages, the glial reaction to these haemorrhages (Durck’s granuloma) and the perivascular and parenchymatous oedema with subsequent demyelination, have been graced by the terms perivenous allergic encephalopathy and disseminated vasculomyelinopathy. In one study it was even suggested that vascular permeability was so extreme that entire red cells could escape through the capillary walls (00and Than, 1989). In two recent pathological studies of human cerebral malaria (00et al., 1987; Boonpucknavig et al., 1990), P.falciparurn antigens and immunoglobulin G deposits have been demonstrated on the cerebral capillary basement membrane. Although these findings were interpreted as representing immune damage, this conclusion should be treated with caution. In one of the studies (Boonpucknavig et al., 1990), only three of the six fatal cases had parasitized erythrocytes in the cerebral microvasculature (probably because these patients died after being in hospital for many days), and in the other study (00et al., 1987), half the patients had been ill for 9 or more days. There was no evidence of associated vasculitis, and complement deposition was seen in only one of 13 cases studied. In the ultrastructural study of 22 fatal cases of cerebral malaria reported by MacPherson et al. (1985), only one patient had been ill for more than 9 days and there was no evidence of immune complex deposition. It is possible that the immunofluorescence studies showing immunoglobulin and malaria antigen deposition on the vascular basement membrane are documenting terminal leakage rather than a primary immunopathological process (Igarishi et al., 1987). The rapid clinical course of human cerebral malaria, the lack of evidence of inflammation in pathological specimens, and the generally good prognosis in survivors all argue against immune damage. Overall, the evidence that cerebral malaria results from an allergic process is unconvincing.



FIG. 10. (a) Retinal haemorrhages and exudates in cerebral malaria (courtesy of S. Ward and Dr E. Schulenberg). (b) Fluorescein angiography of the retina; disruption of the macular microvasculature in cerebral malaria (courtesy of S. Ward and Dr E. Schulenberg).






Cerebral capillary permeability

On the basis of his autopsy observations, Rigdon (1944) concluded that the primary pathological process in cerebral malaria was anoxia, that hypoxia damaged the cerebral capillary endothelium and that this led to an increase in capillary permeability. Maegraith and Fletcher (1972) later concluded, from a series of studies on rhesus monkeys infected with P . knowlesi, that increased capillary permeability. was the primary pathological process in cerebral malaria. In terminally infected animals there was increased permeability to albumin labelled with ‘’’1 and fluorescein isothiocyanate, and increased penetration of the brain by water-soluble dyes. This was rapidly reversed by corticosteroids and the antimalarial drugs chloroquine and mepacrine (Migasena and Maegraith, 1967; Maegraith and Fletcher, 1972). Extravasation of plasma into the cerebral interstitium was considered to account for cerebral oedema-a common post-mortem finding in cerebral malaria. Reduced microcirculatory flow leading to cerebral hypoxia was considered a secondary phenomenon attributed to local haemoconcentration. It was proposed that the increase in cerebral capillary permeability resulted from the local release of inflammatory mediators, originally thought to be kinins (Onabanjo and Maegraith 1970a,b; Maegraith and Fletcher, 1972; Desowitz, 1987), but more recently suggested to be free oxygen radicals (Clark, I. A. et al., 1986; Migasena and Areekul, 1987). As an extension of this suggestion, the beneficial effects of the quinoline antimalarial drugs in cerebral malaria were thought to result from their antiinflammatory, rather than their antiparasitic, activity (Migasena and Maegraith, 1967). The “permeability theory” was widely accepted, and provided the theoretical basis for the use of corticosteroids and osmotic agents to reduce cerebral oedema in human cerebral malaria. The hypothesis that increased capillary permeability causes cerebral malaria is no longer generally accepted. There are several reasons for this. First, there are criticisms of the experiments on which the theory was based; P . knowlesi infection in the rhesus monkey is clinically and pathologically most unlike cerebral malaria in man. The identification of the vasoactive kinins and their precursors relied on non-specific bioassays and, although a permeability defect was identified, the blood/CSF albumin partition was unchanged. This latter finding necessitated postulating increased bi-directional flux of albumin (i.e. exporting albumin from the CSF compartment against a 100 : 1 concentration gradient), which is most unlikely. Second, there is now considerable evidence from studies of human adults with cerebral malaria (Looareesuwan et al., 1983b; Warrell el al., 1986) that



cerebral oedema is not a consistent or prominent feature of the disease in life. Papilloedema is very uncommon in cerebral malaria (Looareesuwan et al., 1983a). Opening pressures at lumbar puncture in adult cerebral malaria are usually normal (Fitz-Hugh et al., 1944; Khan, 1945; Chipman et al., 1966), and in fatal cases are significantly lower than in survivors (Warrell et al., 1986). If there were significant brain swelling in all patients then, in the absence of spinal block, compression of the subarachnoid space within the rigid confines of the cranium would raise CSF pressure. The “anti-inflammatory” drug arguments cannot be sustained either. Chloroquine has no effect on cerebral malaria caused by chloroquine-resistant P . falciparum, yet reversal of the permeability defect in rhesus monkeys infected with P . knowlesi was rapid (Maegraith and Fletcher, 1972). Furthermore the artemesinin-related compounds (qinghaosu) are rapidly effective in cerebral malaria, yet they are structurally unrelated to the “anti-inflammatory” quinoline drugs. In a large prospective double-blind placebo-controlled study, dexamethasone in moderately high doses had no effect on mortality in cerebral malaria, but was shown to prolong coma, and was associated with an increased incidence of superimposed bacterial infections (Warrell et al., 1982). This was confirmed in a later study using very high doses of corticosteroids (Hoffman et al., 1988). Finally, in a series of studies of blood-brain barrier permeability in cerebral malaria (Warrell et al., 1986), there was no consistent alteration in the partitioning of a variety of marker substances (including 1251-albumin). These observations strongly suggest that cerebral oedema is not the cause of coma in cerebral malaria (Looareesuwan et al., 1983b; Badibanga et al., 1986). Permeability may well be increased in some patients. Brain swelling does occur terminally in adults with cerebral malaria (Looareesuwan et al., 1983b), and is a common (but not universal) finding at autopsy (Kean and Smith, 1944; Rigdon, 1944; Spitz, 1946; Edington, 1954, 1967; Thomas, 1971; Riganti et al., 1990; T. T. Hien, personal communication), but it does not appear to cause coma. It is likely that cerebral oedema is an agonal phenomenon in many fatal cases. 2. Intracranial pressure In children with cerebral malaria, lumbar puncture opening pressures are generally lower than those observed in adults, but the “normal” range for young healthy children is considerably lower than that in adults (Fig. 1 I). Thus, in marked contrast to the findings in adults, the majority (80%) of opening pressures in 40 Gambian children with cerebral malaria in one study (Waller et al., 1991), and all those in a study of 26 Kenyan children (Newton er al., 1991), were above the normal range. In this latter detailed prospective





. .. .... .


. .

.. ... . ..... .. .

... ... . .. .






.. 0-






Fatal cases









FIG. 11. Opening pressures of cerebrospinal fluid (CSF) at lumbar puncture in adults and children with cerebral malaria.

study, the clinical findings suggested rostro-caudal progression of brainstem dysfunction in 9 of 12 fatal cases. This was interpreted as indicating central tentorial herniation. However, herniation was also diagnosed clinically in 35% of surviving children. CSF findings were otherwise similar to those in adults, which suggested that there was no significant increase in cerebral capillary permeability. These findings raise the possibility that in children, whose skulls offer relatively less space than those of adults for cranial expansion, there may be raised intracranial pressure (RICP) without cerebral oedema. This could occur as a result of expansion of the intracerebral blood volume by the sequestered parasitized erythrocytes and reflex vasodilatation to compensate for microvascular obstruction and local ischaemia. When intracranial pressure is increased, blood pressure usually rises as a reflex to maintain perfusion pressure (mean arterial pressure minus intracranial pressure). However, blood pressures in children with cerebral malaria are usually at the lower end of the normal range. A failure to maintain



adequate cerebral perfusion pressure will reduce cerebral blood flow and may cause ischaemic damage. RICP and subsequent tentorial or foramen magnum coning could explain the sudden respiratory arrests which appear to be an important cause of death in childhood cerebral malaria. On the other hand, the observations that lumbar puncture opening pressures are no higher in fatal cases than in survivors (Waller et al., 1991), the frequency with which neurological signs attributed to coning occur in survivors (Newton et af., 1991), the lack of convincing evidence that lumbar puncture causes neurological deterioration, and the excellent recovery in children with very high opening pressures all argue against an important role of RICP in the pathophysiology of cerebral malaria. These questions are unresolved at the present time, but as RICP is a potentially treatable condition (with osmotic agents such as mannitol), further studies are clearly needed. 3 . Systemic capillary permeability

Studies of 251-albumin disappearance and preliminary observations on forearm swelling rates suggest that capillary permeability is increased in falciparum malaria (Migasena and Areekul, 1987; Areekul, 1988), and that this returns to normal on recovery. Approximately one-third of patients with severe malaria have extravasation of albumin-bound fluorescein dye on retinal angiography (although this is not correlated with the development of cerebral malaria) (Fig. 10). Urinary microalbumin excretion (a measure of renal capillary permeability) and the transcapillary escape rate for albumin are increased in proportion to disease severity, but the magnitude of these changes is not great (T. M . E. Davis and N. J. White, unpublished observations). The increase in systemic permeability is seldom sufficient to cause oedema in the acute phase of the disease, or significant albuminuria, and probably has no pathophysiological consequence. It is likely that the cerebral capillaries are more resistant to this process, having particularly “tight” basement membrane junctions. Occasionally patients do develop peripheral oedema following the acute phase of the illness, but this resolves without specific therapy. In summary, there is evidence for a moderate generalized increase in capillary permeability in acute malaria, but this is usually symptomless, and there is no convincing evidence that it causes cerebral oedema, coma or other adverse effects in cerebral malaria, G.


Patients with severe malaria, particularly those who have been comatose with high fever for more than 24h, may be significantly dehydrated on



admission to hospital. However, following rehydration plasma volume is increased in both moderate and severe malaria (Chongsuphajaisiddhi et al., 1971; Davis et al., 1990d). Measurements of total body water and extracellular volume (sulphate space) have been normal (Feldman and Murphy, 1945; Chongsuphajaisiddhi et al., 1971). Plasma renin activity, aldosterone and antidiuretic hormone concentrations are elevated in acute malaria (Malloy et al., 1967). These findings suggest activation of homeostatic mechanisms to maintain an adequate circulating blood volume in the presence of generalized vasodilatation and a falling haematocrit. Brooks et al. (1967) suggested that homeostasis is incomplete, and that there is a reduction in “effective blood volume” relative to the expanded capacitance of the dilated vasculature which contributes to orthostatic hypotension in malaria (Section VI1.A). By contrast, in the simian models used to study circulatory pathophysiology in malaria, i.e. the rhesus monkey infected with P. knowlesi or the sequestering tertian parasite P. coatneyi, plasma volume is contracted in the acute phase of the disease, and this is associated with the development of hypotension (Miller, L. H. et al., 1968). Plasma osmolality may be reduced acutely but, together with the electrolyte abnormalities, this returns to normal with control of the infection. Miller, L. H.’et al. (1967) studied the response to a water load in patients with acute uncomplicated malaria who were mildly hyponatraemic, and had normal glomerular filtration rates (assessed by inulin clearance) and renal plasma flows (assessed by p-aminohippurate clearance). The responses were variable, although urinary sodium concentrations remained low. Inappropriate secretion of antidiuretic hormone was suggested in this and a subsequent study (Ogunye and Gbadebo, 1981). However, in more recent studies the antidiuretic hormone response (measured directly) has been found to be appropriate in acute falciparum malaria (R. E. Phillips ef al., unpublished observations). H.


As in many infections, mild hyponatraemia and mild hypochloraemia are common. More profound reduction in serum sodium may occur in severe infections. This is associated with low urinary sodium and high urinary aldosterone excretion (Brooks et al., 1967; Miller, L. H. et al., 1967). In experimental malaria, sodium entry to the erythrocyte is increased and sodium efflux is reduced (Dunn, 1969). Cation fluxes in other tissues have not been studied. Infected erythrocytes also have elevated concentrations of intracellular calcium, but concentrations in uninfected cells are within the normal range (Krishna and Ng, 1989). Plasma concentrations of potassium are remarkably normal in malaria, balanced between the opposing forces of



active haemolysis and possible tissue injury which tend to raise plasma potassium, and activation of the renin-angiotensin system, and quinineinduced insulin release, which tend to reduce it. 1.


Serum thyroxine (T4) concentrations are reduced in many illnesses and malaria is no exception. Serum concentrations remain low until after parasite and fever clearance (Davis et al., 1990a). Very low levels may be found in severe malaria, and in two patients with lethal infections serum T4 was unmeasurable (Davis et al., 1990a). Basal thyrotropin levels are normal, but the response to thyrotropin releasing hormone is reduced, which suggests depression of both pituitary thyrotroph and thyroid gland function (Davis et af., 1990a). This pattern is commonly termed the “sick euthyroid” syndrome. Sequestration or cytokine-mediated pathology could be responsible, but the finding of elevated plasma concentrations of somatostatin (>25 pmol 1-I) in cerebral malaria may also be relevant (Davis et al., 1990a), as this hormone inhibits a broad range of endocrine functions. P. falciparum has been shown to synthesize a somatostatin-like peptide in amounts that could have a biological effect on the host (Pan et af., 1987). Brooks et a f . (1969) studied pituitary-adrenal function in 29 adults with acute falciparum malaria and concluded, on the basis of plasma and urinary 17 hydroxy and 17 ketosteroid concentrations and the metyrapone test, that pituitary-adrenal function was normal. These are rather imprecise measures and more subtle defects cannot be excluded. Mineral homeostasis is perturbed in acute falciparum malaria (Petithory et al., 1983; Lewis, 1987). Hypocalcaemia occurred in 36% and hypophosphataemia in 43% of patients in a recent series (Davis et al., 1991). Plasma concentrations of magnesium were normal in the majority of cases. Profound hypophosphataemia ( <:0.30 mmol 1- I) was associated with severe disease, and this could be of pathological relevance. Very low phosphate concentrations have been associated with confusion, muscle weakness, depressed reflexes and focal neurological deficits, platelet and leucocyte dysfunction, and reduced red cell deformability and oxygen carriage. The major factor contributing to hypophosphataemia was a lowered renal threshold for phosphate (Davis et al., 1991). Parathormone concentrations were also inappropriately low for the corresponding plasma calcium levels, indicating malaria-associated dysfunction of the parathyroid glands. J.


Acute renal failure is a major cause of death in adults with severe falciparum



malaria. The presentation and recovery phase is that of acute tubular necrosis (Sitprija et al., 1967; Stone et al., 1972; Sitprija, 1988; Lumlertgul et al., 1989). Renal dysfunction is associated with cerebral involvement, high parasitaemia, jaundice and haemoglobinuria (Canfield et al., 1968; WHO, 1986), and a high risk of pulmonary oedema. About half of all adult patients with cerebral malaria will have biochemical evidence of renal impairment (raised blood urea and serum creatinine) (White and Looareesuwan, 1987; T. T. Hien, personal communication). Some of these will have a period of oliguria during the acute phase (Sitprija, 1988), and about one-fifth will become anuric. Over half the latter group will die from fulminant disease with a combination of metabolic acidosis and pulmonary oedema. Renal failure may also occur without an oliguric phase: i.e. urine volumes remain normal or increase with a steadily rising blood urea and creatinine. In areas of unstable endemicity, patients may present either with fulminant disease and multiple organ failure (and a poor prognosis), or with established renal failure following severe malaria or associated with haemoglobinuria. The parasitic infection in these latter patients has been controlled by antimalarial drugs and the subsequent management is that of the renal failure only. The prognosis is then entirely dependent on the facilities available for dialysis. By contrast with adults, renal failure is extremely unusual in young children with severe malaria. Even biochemical evidence of renal dysfunction is rare, and when present usually indicates severe dehydration. Renal impairment, and ultimately acute tubular necrosis, in severe falciparum malaria presumably result from a reduction in renal microvascular blood flow (i.e. ischaemic nephropathy). In some patients dehydration and haemoglobinaemia (Blackburn et al., 1954) contribute to this process. Studies of renal cortical flow using the '33Xe clearance technique (Sitprija et al., 1977), angiography and contrast urography (Arthachinta et al., 1974) indicate a reduction in cortical perfusion, but this is common in acute tubular necrosis from any cause. Increased whole blood viscosity and hypovolaemia were considered to account for these findings. Histopathological examination shows cytoadherence in the glomerular capillaries, but to a lesser degree than in the brain or heart. Several pathological studies (Hartenblower et al., 1972; Bhamarapravati et al., 1973; Futrakul et al., 1974; Boonpucknavig and Sitprija, 1979) have also suggested that there is active glomerulonephritis, which would not be surprising in view of the large amounts of malarial and host red cell neoantigens confronting the glomerular filter. Immunofluorescence studies show immunoglobulin deposition. Mesangial and endothelial cell proliferation can be seen by electron microscopy. However, the clinical and laboratory features of the disease are most unlike acute glomerulonephritis (Sitprija, 1988); hypertension does not occur, the urinary sediment lacks red



cell casts, proteinuria is mild or absent, children are unaffected and in most cases renal impairment is transient. It is possible that occasional patients develop significant glomerular disease following acute falciparum malaria, but this appears to be rare. The relative roles of haemoglobin (Brant et al., 1951) and the large amounts of erythrocyte and parasite debris released at merogony in the pathogenesis of malarial renal failure are unresolved. Massive haemolysis causing “black water” certainly causes renal failure both in acute malaria associated with quinine treatment (Blackie, 1944), and also in some patients with glucose-6-phosphate dehydrogenase deficiency who receive oxidant antimalarial (or other) drugs (Section VI1.R). Originally, it was thought that renal failure resulted from blockage of the renal tubules by haemoglobin and related pigments, but studies of renal histology in humans argue against this hypothesis (Maegraith and Findlay, 1944; Dukes et al., 1968). How quinine and severe Malaria consort to induce massive haernolysis remains to be discovered-but the end result is acute tubular necrosis. The natural history of renal failure is determined initially by the overall severity of the infection (approximately two-thirds of deaths associated with malaria renal failure occur rapidly from multiple organ dysfunction), and then by the ability of the medical facilities to conduct peritoneal or haemodialysis. Provided no complications on dialysis ensue, most patients’ renal function will return to normal over a period of several weeks (T. T. Hien, personal communication). K.


Nausea and vomiting are common in malaria, particularly when fever is high. Antimalarial drugs may be regurgitated and comatose patients may aspirate stomach contents into the lungs with fatal consequences. In cerebral malaria the vomitus often contains altered blood (“coffee grounds”), indicating gastric or duodenal bleeding. This presumably results from stress ulceration. Malabsorption of amino acids, sugars, fats and antimalarial drugs have all been reported (Olsson and Johnston, 1969; Karney and Tong, 1972; Segal et al., 1974; Herzog et al., 1982; Molyneux et al., l989a)although oral absorption of antimalarial drugs is adequate in most patients (White, 1985) except those who are seriously ill (and require parenteral treatment anyway). Both sugars requiring active transport, and those absorbed by passive diffusion, are malabsorbed in acute malaria, suggesting impaired splanchnic perfusion (Molyneux et al., 1989a). Histological studies show parasitized erythrocyte sequestration in the gut microvasculature (Olsson and Johnston, 1969). Taken together these findings suggest reduced perfusion of the intestinal microvasculature in falciparum malaria. Abdomi-



nal pain with watery diarrhoea is a common presenting feature of acute falciparum malaria in some areas, but is not seen in others, suggesting either parasite strain differences or unidentified co-factors. The pathogenesis of this condition has not been studied. L.


Jaundice, elevated transaminases, rapid development of hypoalbuminaemia, prolonged coagulation indices, failure of gluconeogenesis and reduced metabolic clearance of the antimalarial drugs all point to liver dysfunction in malaria (Keys et al., 1950; Deller et al., 1967; Hills, 1971; Joshi, Y.K. et al., 1986; WHO, 1990). However, these laboratory measures reflect multiple pathology. Jaundice, which is much more common in adults, results also from haemolysis, although this component is relatively small. Hepatocyte injury and associated cholestasis are more important. The elevated transaminases may also result from skeletal muscle damage (Miller, K. D. et al., 1989), and prolongation in prothrombin and partial thromboplastin times from associated consumptive coagulopathy. Some adults may be deeply jaundiced with abnormally elevated transaminases and relatively little evidence of other vital organ dysfunction, but the combination of “hepatitic” biochemical findings, renal impairment and coma indicating multi-system failure carries a poor prognosis-as indeed it does in any condition. Acute liver failure causing hepatic encephalopathy does not occur. Perhaps the most sensitive marker of liver dysfunction in malaria is the impaired elimination of drugs such as quinine (White et al., 1982), which are cleared principally by hepatic biotransformation. Reduced metabolic clearance parallels disease severity and returns to normal coincident with clinical recovery. In rhesus monkeys infected with P. knowlesi, angiography reveals vasoconstriction of the portal tree (Skirrow et al., 1964). In severe falciparum malaria, clearance of indocyanine green is variably reduced, whereas galactose clearance is not (Molyneux et al., 1989a; Pukrittayakamee et al., in press). Both are measures of liver blood flow, although the kinetics of hepatic removal are different (Section VII.D.2). In uncomplicated malaria clearance of both markers is normal or high. One interpretation of these liver blood flow findings, as with the observations of cerebral blood flow in cerebral malaria, is that microcirculatory obstruction is not uniform. Flow is obstructed in some vessels, but is high in others. Parasitized red cells do sequester in the portal and hepatic vasculature. There is often sinusoidal dilatation with congestion of the centrilobular capillaries, slight hepatocyte swelling, prominent Kupffer cell hyperplasia and pigment deposition, and variable mononuclear cell infiltration (Srichaikul, 1959). There may be



centrizonal necrosis in some cases (Spitz, 1946; Clark, C. and Tomlinson, 1949). In the unfortunate patients who underwent liver biopsy (usually with uncomplicated malaria), liver histology was remarkably normal (Corcoran et al., 1953; de Brito et al., 1969). M.


Hypoglycaemia, commonly accompanied by lactic acidosis, is now recognized as an important manifestation of falciparum malaria (Fisher, 1983; Migasena, 1983; White et al., 1983; Kiire, 1986; Mbelepe et al., 1986; White et al., 1987a; Currie and Kerau, 1988; Das et al., 1988; Taylor et al., 1988; Molyneux et al., 1989a). Although hypoglycaemia may occur in any severe infection, and is an important problem in malnourished children (Butler et al., 1989; Bennish et al., 1990), it appears to be particularly common in severe falciparum malaria, where it is associated with increased mortality (Fig. 12). Hypoglycaemia arises as a result of several discrete processes. 100


Mortality ('A) 60











Lactate I ~iucomratio



FIG. 12. Relationship between the ratio of admission venous plasma lactate to glucose in 200 adults and children with severe malaria (mean and 95% confidence interval).



1. Iatrogenic hypoglycaemia

Quinine is an extremely potent inhibitor of pancreatic islet cell potassium channel permeability. This effect is shared by glucose and stimulates insulin release (Henquin et al., 1975; Henquin, 1979). Chloroquine inhibits insulin degradation, but this does not alter glucose homeostasis significantly, and there is no convincing evidence that chloroquine or other antimalarial drugs (except for quinine and quinidine) cause hypoglycaemia (Phillips et al., 1986b). Quinine stimulates insulin secretion in vitro and in vivo (White et al., 1983; Okitolonda et al., 1986). In terms of free drug concentration, the dextro-rotatory isomer quinidine is about half as potent a secretagogue as quinine (Davis et al., 1990b). Insulin secretion is increased in all patients receiving quinine treatment for falciparum malaria (White et al., 1983; Taylor et al., 1988; Okitolonda et al., 1987; Das et al., 1988), but in most there is also reduced peripheral tissue sensitivity to insulin (Davis et al., 1990c), and the blood glucose is not reduced. Quinine-induced hypoglycaemia is very unusual on the first day of treatment. However, in quininetreated patients who are severely ill for a protracted period, hypoglycaemia becomes increasingly likely. Pregnant women are particularly vulnerable to hyperinsulinaemic hypoglycaemia because of an accelerated ketogenic response to starvation (Metzger et al., 1982), and an amplified pancreatic islet cell response to the secretory stimulus provided by quinine. They may become hypoglycaemic with otherwise uncomplicated malaria. In African children receiving parenteral quinine for severe malaria, insulin secretion is stimulated, but hypoglycaemia results mainly from other factors (see below) (Taylor er al., 1988). The biochemical features of quinine-induced hypoglycaemia are raised plasma concentrations of insulin, lactate and alanine, but low concentrations of ketone bodies (acetoacetate, P-hydroxybutyrate) (White et al., 1987a).

2. Impaired gluconeogenesis Severe malaria is associated with impairment of gluconeogenesis. Hypoglycaemia occurs in the presence of elevated plasma concentrations of the principal gluconeogenic substrates alanine and lactate (White et al., 1983). In patients receiving chloroquine (which does not stimulate insulin secretion), plasma ketones, principally 3-hydroxybutyrate, are usually elevated and plasma glycerol concentrations are normal. Plasma concentrations of triacylglycerols and non-esterified fatty acids are elevated, as in other severe infections (Angus et al., 1971; Onongbu and Onyeneke, 1983; Kawo et al., 1990). This is the usual biochemical pattern in African children with severe



malaria (White et al., 1987b; Taylor et al., 1988; Molyneux et al., 1989b). Fasting rapidly depletes hepatic glycogen in children (who have an average 12 h worth of stores) even if they are well nourished, and glycogen depletion certainly contributes to hypoglycaemia in some cases, but not all children are ketotic, and other factors are undoubtedly responsible. Counter-regulatory hormone concentrations (i.e. growth hormone and cortisol) are elevated (Taylor et al., 1988) and there is often concomitant lactic acidosis. These biochemical features, together with hypoglycaemia, have been described in other severe childhood infections (Phillips et al., 1988; Kawo et al., 1990), but they are particularly common in severe P. falciparum infections, occurring in approximately one-third of all children with cerebral malaria. In humans, clearance of the monosaccharide galactose is normal in severe malaria (Pukrittayakamee et al., in press). This suggests that there may be a specific defect in glycolysis (rather than a generalized reduction in glycolytic enzyme activity) and that the biochemical locus of impaired hepatic gluconeogenesis may reside outside the small segment of the glycolytic pathway between galactose and glucose. Biochemical studies on isolated liver slices from rats with severe P.berghei infections also suggest that there may be a specific defect in the glycolytic pathway (P. A. H. Holloway and D. H. Williamson, personal communication). Hepatic gluconeogenesis from lactate is particularly impaired, and cannot be explained entirely by the reduction in hepatic adenosine triphosphate concentrations. Hepatic ketogenesis is also reduced, but ketogenesis from endogenous substrates shows a relative increase in hydroxybutyrate synthesis. Elevated plasma concentrations of the cytokine T N F are correlated with hypoglycaemia and death in severe falciparum malaria (Grau et al., 1989a-d; Kwiatkowski et al., 1990) (Section V1.C). T N F is a potent inhibitor of hepatic gluconeogenesis (Evans et al., 1989), inducing many of the metabolic derangements observed in malaria, and this cytokine could well be implicated in the pathogenesis of hypoglycaemia in severe malaria. Taken together, these findings point principally to an impairment of hepatic gluconeogenesis which is proportional to the severity of malaria and might be localized to specific biochemical steps in carbohydrate metabolism. The counter-regulatory hormone response is appropriate and proteolysis and lipolysis are unimpaired. There appears to be primary hepatocyte dysfunction. This is compounded by reduced hepatic perfusion, increased peripheral glucose consumption and lactic acid production, reduced or absent glycogen stores, and-in some patients-uinine-stimulated hyperinsulinaemia. The liver cannot clear the increased quantities of lactate and alanine produced peripherally; hypoglycaemia and lactic acidosis result.



3. Glucose consumption The overall consumption of glucose is increased in malaria (Davis et al., 1988). The host is febrile, hypercatabolic and probably respires anaerobically in vascular territories occluded by parasitized erythrocytes. In addition, there is an accelerated demand for glucose by the parasite which is proportional to the stage of intra-erythrocytic development (Pfaller et al., 1982; Jensen et al., 1983). Cells containing mature forms of the parasite may consume up to 70 times as much glucose as their uninfected counterparts. Most of this glucose is metabolized to lactate as the parasite does not have the enzymatic machinery to complete the citric acid cycle (Sherman, 1979). Glycolysis continues even in acid conditions that inhibit host glucose metabolism (Sander et al., 1982). Thus, hypoglycaemia and lactic acidosis may be much worse in the occluded microcirculation of vital organs than in the systemic circulation overall. In the absence of precise methods for calculating the size and synchronicity of the sequestered parasite biomass (White and Krishna, 1989), the overall metabolic contribution of the sequestered parasites cannot be assessed accurately. Rough calculations suggest that, with heavy synchronous parasite burdens, parasite glycolysis could double overall glucose turnover. This would be accommodated easily by increased gluconeogenesis and glycogenolysis in a healthy host with adequate glycogen stores, but in patients with severe malaria the combination of reduced supply (impaired gluconeogenesis and glycogenolysis)and increased demand for glucose leads to hypoglycaemia. When quininestimulated insulin secretion compounds the problem, hypoglycaemia often proves difficult to correct because glucose replacement stimulates further insulin secretion and a vicious cycle ensues. The disposition of administered glucose in severe malaria is normal (Pukrittayakamee et al., 1991a), although peripheral tissue insulin sensitivity is reduced (Davis et al., 1991). N.


Metabolic acidosis is an important cause of death in severe malaria. Concentrations of lactate in arterial and venous blood and in CSF are elevated in proportion to disease severity (White et al., 1983, 1985, 1987b; Warrell et al., 1988; Molyneux et al., 1989b), and return to normal coincident with clinical recovery. Lactic acidosis is associated with hyperalaninaemia and hypoglycaemia. Measurement of venous or cerebrospinal fluid lactate or the lactate/glucose ratio is a useful prognostic index (White et al., 1986) (Fig. 12). In most patients with severe malaria, the anion gap widens as bicarbonate falls, but hyperlactataemia is buffered. However, arterial pH



falls eventually if hydrogen ion production outstrips the buffering capacity of the blood and the impaired ability of the kidney to excrete hydrogen ions. Chemoreceptor triggering leads to increased respiratory drive. Kussmaul’s breathing is an ominous sign in malaria. Hydrogen ion retention results from increased production and reduced clearance of lactic acid, and in adults this is often compounded by retention of other organic acids because of renal failure. Lactic acid arises from host and parasite anaerobic glycolysis (Zolg et al., 1984). The contribution of the host in the form of L (+) lactate is the more important. Approximately 7% of parasite lactate appears as the laevorotatory enantiomer D (-) lactate (Van der Jagt et al., 1990), but plasma concentrations of this isomer are not elevated significantly in severe malaria (S. Krishna and P. A. H. Holloway, personal communication). The healthy liver and kidney are capable of clearing large quantities of L (+) lactic acid from the blood and converting this either to energy (via the Krebs cycle) or to glucose (via the Cori cycle), but in severe malaria compromised gluconeogenic function and reduced liver blood flow reduce lactate clearance (Section VII.D.2). In certain circumstances (partial ischaemia), exogenous glucose has been shown to fuel intracellular lactic acid production and worsen acidosis (Gardiner et al., 1982), but studies in rodents and humans suggest that glucose administration tends to ameliorate the condition in malaria (Holloway et al., 1991; Pukrittayakamee et al., 1991). This is important because patients with severe malaria are often treated with glucose empirically to prevent hypoglycaemia. In the young rat infected with P. berghei lactic acidosis develops predictably as the infection worsens, and is associated with reduced gluconeogenesis and terminal hypoglycaemia (Holloway et al., 1991). In this model lactic acidosis can be attenuated by dichloroacetate, an activator of the pyruvate dehydrogenase complex. This temporary holding measure is now being evaluated in human patients. 0.


The muscles are not painful in severe malaria, but there is biochemical evidence of muscle damage (elevations in serum concentrations of myoglobin and creatine kinase) which parallel disease severity (Miller, K. D. et al., 1989). On histopathological examination, skeletal muscle is a site of vascular sequestration, but muscle cell abnormalities are usually mild. Rarely, rhabdomyolysis occurs (De Silva et al.,. 1988) and acute myoglobinuric renal failure may develop. The quantitative contribution of skeletal muscle (the largest tissue mass in most bodies) to the development of lactic acidosis remains to be determined.





Anaemia is an inevitable consequence of malaria, resulting from a combination of parasitized erythrocyte destruction at merogony, accelerated removal of unparasitized red cells and ineffective erythropoiesis (Zuckerman, 1966; Abdallah et al., 1980; Weatherall and Abdallah, 1982). 1. Oxygen delivery

Anaemia reduces the oxygen carrying capacity of blood. In order to maintain oxygen delivery, blood flow increases and the haemoglobin oxygen dissociation curve (ODC) is shifted to the right. The right shift in the ODC results from increased intra-erythrocytic synthesis of 2-3-diphosphoglycerate and causes increased unloading of oxygen at the tissues. In malaria both compensatory mechanisms may be perturbed. There is microcirculatory obstruction, and in rodents infected with P . yoelii there is a reduced right shift in the ODC (Krishna et al., 1983). Oxygen carriage in humans has not been studied directly. 2. Red cell destruction Anaemia develops rapidly in acute malaria (Perrin et al., 1982), particularly in severe P . falciparum infections (Fleming and Allan, 1969; Davis et al., 1990d; Looareesuwan et al., 1991). The number of red cells lost by parasite destruction is underestimated from the peripheral blood film because of the unknown number of sequestered cells. Using a simple mathematical model, Davis et al. (1990d) have analysed the differences between observed haematocrit, and that predicted by labelling red cells with 'lCr to derive an estimate of the sequestered red cell haematocrit in severe falciparum malaria, and thus the contribution of these hidden cells to the subsequent development of anaemia. These data gave a predicted sequestered cell haematocrit twice that of the peripheral blood (70% vs. 35%). Plasma volume expansion was estimated to contribute 7.5%, parasitized red cell destruction 6.3%, and non-parasitized cell destruction 8.9% (i.e. approximately equal amounts), to the average fall in haematocrit in the patients studied. In acute malaria there is accelerated destruction of unparasitized erythrocytes associated with increased splenic clearance that persists beyond the time of parasite clearance. Reduced survival of unparasitized red cells has been documented both in acute malaria and following treatment (Charoenlarp et al., 1979; Looareesuwan et al., 1987a,b). Survival is particularly short in severe disease, when the haematocrit can fall precipitously in the acute phase (Canfield, 1969; Phillips et al., 1986a; Looareesuwan et al., 1991). Because of this obligatory red cell destruction, the haematocrit can be



considered as a “clock” of the infection. Thus, high parasitaemias without a reduction in haematocrit in a well-hydrated patient imply rapid expansion of the parasite population (i.e. a fulminant infection). 3. Role of antibody The role of red cell-bound antibody (“Coombs positivity”) in the development of anaemia has been contentious (Adner et al., 1968; Rosenberg, E. B. et al., 1973; Facer et al., 1979; Facer, 1980; Weatherall and Abdallah, 1982; Abdallah, 1986). Studies in Gambian children with falciparum malaria suggested that erythrocytes were sensitized with immunoglobulin G (IgG) and the C3 component of complement (Facer, 1980). But, when the erythrocyte-bound antibody was measured in acute malaria, the distribution was found to be normal with average values less than 200 molecules of immunoglobulin per erythrocyte. There was no relationship between changes in these numbers and the subsequent development of anaemia (Merry et al., 1986). However, it may be argued that the red cells studied, i.e. those remaining in the circulation, obviously represented those cells not removed. Cells with greater antibody binding might have been cleared. Increased splenic removal of immunoglobulin-coated red cells has been demonstrated in malaria, even at relatively low levels of antibody sensitization (Ho et al., 1990a). Thus a non-specifically activated phagocytic system (Newsome, 1984) may remove cells which would otherwise be allowed to continue in the circulation. An analogous process may contribute to shortened platelet survival. This increased splenic clearance function lasts for weeks after resolution of the infection (Lee et al., 1989; Ho et al., 1990a). 4. Membrane abnormalities Both parasitized and unparasitized erythrocytes have abnormal cell membranes in malaria (Section 1V.B). There is increased surface expression of phosphatidyl serine residues suggesting eversion of the inner surface of the lipid bilayer (Joshi, P. et al., 1986). This process is analogous to accelerated ageing, and may provide a target for autoantibodies normally present in healthy subjects. Increased red cell surface expression of phosphatidyl serine and senescent antigens, coupled with enhanced Fc receptor and filtrative clearance processes, provide a plausible mechanism for the accelerated removal of parasitized and unparasitized red cells. Q.


The bone marrow fails initially to respond to acute anaemia in falciparum malaria. Reticulocytosis does not occur for many days, and then the



response is often inadequate. Bone marrow, white cell and platelet production is normal or increased. Preliminary data suggest that the erythropoietin response in patients with normal renal function is usually appropriate (P. M. Cotes, personal communication). Morphologically the bone marrow is dyserythropoietic in both falciparum and vivax malaria (Srichaikul et al., 1967; Abdallah et al., 1980; Knuttgen, 1987; Wickramasinghe et al., 1987, 1989). Iron is plentiful in the marrow, but serum iron is low and serum ferritin is very high (Phillips et al., 1986a). In rodent malarias, cytokines, particularly TNF, appear to contribute to bone marrow dysfunction (Clark, I. A. and Chaudhri, 1988a; Miller, K. L. et al., 1989). In human malaria, local cytokine release could be responsible for dyserythropoiesis, but-as in the other situations where cytokine induced pathology is hypothesizeddefinitive proof is lacking. In P.fakiparum infections there is some sequestration of parasitized erythrocytes in the bone marrow, and it has been suggested that this could cause local hypoxia (Wickramasinghe et al., 1987).



This enigmatic condition (Hamilton-Fairley and Bromfield, 1934; Blackie, 1944; Maegraith, 1948, 1952; Bruce-Chwatt, 1987) remains incompletely understood. Objectively there is haemolysis and black (“coca-cola’’ coloured) urine (Barratt and Yorke, 1909-1910). When severe, the patient is a pale, slate-grey colour. Acute renal failure may supervene. Blackwater may be present in the absence of fever (or even of malaria) when patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency take oxidant antimalarial drugs (Gilles and Ikeme, 1960; Chan et al., 1976). This is a predictable and well-characterized reaction. However, many patients with black urine have normal erythrocyte G6PD levels (WHO, 1990). In those patients quinine (which is not itself an oxidant) and malaria somehow conspire to produce massive haemolysis. In most cases renal failure does not occur, blackwater is transient, and the patient recovers. Renal failure is particularly likely in severe malaria, and in this situation the mortality is high. Quinine does not induce haemolysis with the patients’ red cells in v i t r e a n d there is no evidence for anti-quinine antibodies or a drug-hapten mechanism causing antibody-mediated haemolysis (A. H. Merry, personal communication). The role of quinine metabolites in blackwater fever has not been explored. Renal failure presumably results from acute tubular necrosis. Haemoglobin itself is not nephrotoxic (Brandt et al., 1951; Conn et al., I956), but other erythrocytic proteins and cellular material can induce renal failure, particularly if patients are dehydrated or acidotic (Hamilton-Fairley and Bromfield, 1934; Blackburn et al., 1954).





The platelet count is usually low in malaria. A moderately reduced count (ca. 100 000 PI-') is observed in symptomatic infections with all four species of human Plasmodium (Horstmann et al., 1981; Hill et al., 1964), but much lower counts may be recorded, particularly in severe falciparum malaria. Several hypotheses have been proposed to account for thrombocytopenia; disseminated intravascular coagulation (Devakul er al., 1966; Dennis et al., 1967), antibody-mediated clearance (Kelton et al., 1983; Grau et al., 1988a; Srichaikul et al., 1988), and enhanced aggregation (Essien, 1989). Bone marrow megakaryocyte numbers are normal in malaria, suggesting adequate production of platelets. Increased turnover is suggested by observations of platelet enlargement (Fajardo and Rao, 1971), and studies of the disposition of platelets labelled with 'Cr have confirmed accelerated peripheral platelet destruction and splenic pooling (Sheagren et al., 1970; Skudowitz et al., 1973). There is usually increased coagulation cascade activity (Pukrittayakamee et al., 1989), but full blown disseminated intravascular coagulation is rare and occurs in only 5% of cerebral malaria cases (White and Looareesuwan, 1987; WHO, 1990); it is therefore unlikely to be the sole cause of the reduced platelet count. Increased platelet-associated IgG has been demonstrated in both vivax and falciparum malaria (Kelton et al., 1983; Mohanty et al., 1988). Platelets were shown to have saturable binding sites for malarial antigens, and it was suggested that malaria-specific IgG bound to platelet-absorbed antigen via the Fab portion of the molecule. However, in Thailand no relationship was evident between either free or bound platelet-directed antibodies and either the absolute platelet count or changes in the count (S. Looareesuwan and N. J. White, unpublished observations). Platelets taken from patients with acute malaria have shown increased ability to aggregate (Mohanty et a[., 1988). Some workers have reported increased plasma concentrations of the platelet-specific proteins, pthromboglobulin and platelet factor 4, in uncomplicated malaria (Essien, 1989), but in other studies of severe malaria there was no evidence of platelet activation (Supanaranond et al., in press). Thus there is contradictory evidence on both hypotheses, and no firm conclusion can be drawn on the relative importance of immune mechanisms or activation. Platelet antibody is unlikely to play a significant role as the platelet count rises coincident with the resolution of symptoms in malaria, but the other proposed mechanisms require further study. T.


There is considerable laboratory evidence of disseminated intravascular coagulation in falciparum malaria (Conrad, 1969; Jaroonvesama, 1972; Reid



and Nkrumah, 1972; Sucharit et al., 1975; Goodall, 1981; Horstmann and Dietrich, 1985), but the clinical importance of these laboratory findings has been overemphasized. The coagulation cascade is certainly activated in acute malaria (Pukrittayakamee el al., 1989). Erythrocytes containing mature falciparum malaria parasites promote coagulation (Udeinya and Miller, 1987). There is accelerated fibrinogen catabolism (Devakul et al., 1966) with elevated serum fibrin degradation products, but fibrinogen concentrations are usually raised even in severe malaria as part of the acute phase response. Thus, in most patients, synthesis exceeds consumption (Jaroonvesama, 1972; Looareesuwan et al., 1983b; Pukrittayakamee et al., 1989). Hypofibrinogenaemia indicates significant consumptive coagulopathy and may be associated with a bleeding diathesis. In most patients antigen related to factor VIII and von Willebrand factor concentrations are raised, and are correlated directly with parasitaemia and inversely with the platelet count (Horstmann and Dietrich, 1985). Plasma concentrations of antithrombin I11 (the natural inhibitor of thrombin) are reduced in malaria (Pukrittayakamee et al., 1989), and antithrombin 111-thrombin complexes are increased even in uncomplicated malaria. These changes closely parallel disease activity. Thus acceleration of coagulation cascade activity is proportional to disease severity, but only in a relatively few patients with severe disease does disseminated intravascular coagulopathy with consumptive coagulopathy cause significant bleeding (Borochovitz et al., 1970). An association between disseminated intravascular coagulopathy and the development of acute pulmonary oedema has been suggested (Punyagupta et al., 1974), but the high incidence of bleeding in the patients studied may have been more related to coexistent uraemia and treatment with heparin, low molecular weight dextran and steroids. The importance of thrombus formation in the pathology of fatal malaria is uncertain. The pathological interpretations of autopsy findings have been contradictory. Several authors have reported finding microthromboses with fibrin deposition, particularly in the cerebral vessels of patients who died with cerebral malaria (Dudgeon and Clarke, 1917; Rigdon, 1944; Thomas, 1971; Schmid, 1974; Toro and Roman, 1978; Boonpucknavig et al., 1990), but others have considered true thrombus formation to be relatively unusual (Spitz, 1946; MacPherson et al., 1985) or not to occur at all (Gaskell and Miller, 1920; Edington, 1967; Janota and Doshi, 1979). These latter authors considered that the microvascular obstruction was not caused by thrombus, but by “plugging” with a compressed or agglutinated mass of parasitized cells and pigment (Dhayagude and Puranare, 1943; Kean and Smith, 1944; Arieti, 1946; Spitz, 1946; 00et al., 1987). Occasional fibrin strands may be seen, but electron microscopy reveals a “striking absence of platelets” in the obstructed vessels (MacPherson et al., 1985). Overall these results suggest



FIG. 13. (a) Ultrastructure of the spleen: a parasitized erythrocyte (PE) is adherent to the littoral cells (LC) of the splenic sinusoid. E, uninfected erythrocyte; K, knob; M, mitochondrion of littoral cell; P, parasite. (b) Trapping of a parasitized erythrocyte (PE) between two splenic littoral cells (LC). E, uninfected erythrocyte; P, parasite. (Reproduced with kind permission from Pongponratn et al., 1989.)



that microvascular thrombus formation may occur in cerebral malaria, but it is neither widespread nor common. U.


1. Filtration The spleen plays a central role in the clearance of parasitized erythrocytes (Garnham, 1970), recognizing their loss of deformability (Cranston et al., 1984) and opsonization with antibodies and/or complement components. The mechanical filtration function occurs at the inter-endothelial slits in the walls of the splenic sinuses (Weiss et al., 1986; Weiss, 1990) (Fig. 13). In the mouse infected with P . yoelii, a blood-spleen barrier is formed by contraction of the splenic reticular fibres during the phase of rising infection. At crisis (spontaneous resolution of the infection) the barrier is reversed. This anatomical alteration appears to parallel the changes in clearance of parasitized erythrocytes. In rats infected with P . berghei, splenic filtration function is impaired during the period of rising parasitaemia, but then becomes supernormal as parasitaemia is controlled (Wyler et al., 1981). Similar changes in splenic! filtration have been observed in patients with acute falciparum malaria, using the clearance of heat-damaged erythrocytes as a measure of filtration function (Looareesuwan et al., 1987a). These damaged red cells are rigid and spherocytic, and are removed rapidly by the normal spleen. In acute falciparum malaria removal was increased in patients with splenomegaly. In patients without palpable spleens, removal increased rapidly after antimalarial treatment. 2. Immune mechanisms Antibody-coated erythrocytes are removed by Fc receptor-mediated interactions with splenic macrophages. This function is also increased in acute malaria (Lee et al., 1989; Ho et al, 1990a). Red cells coated with relatively low numbers of antibody molecules (300-500 per cell) were removed rapidly from the circulation and were localized by 51Crlabelling to the spleen (Ho et al., 1990a). Clearance was correlated directly with anaemia (Fig. 14) and inversely with parasitaemia, but (unlike filtration function) it was independent of spleen size. High levels of circulating immune complexes may compromise splenic interactions mediated by Fc receptors, but in these studies there was no correlation between antibody-coated red cell clearance and the level of circulating immune complexes. As the intra-erythrocytic parasite grows, increasing amounts of parasite



and host neoantigens are exposed on the surface of the red cell, and more antibody is bound (Luzzi et al., 1991). Once the cell has sequestered, it cannot be removed by the spleen. Thus, there is a balance between the ability of the spleen to remove the increasingly opsonized and rigid erythrocyte, and the parasite's capacity to export cytoadherence proteins to the red cell surface and effectively remove itself from the circulation. Although phagocytic activity is generally increased in uncomplicated malaria, there is evidence that it is compromised in severe malaria (Ward et al., 1984). In this situation, it is difficult to distinguish cause from effect.


50 0.

40 Hematocrlt




30 -





















,, "



tIn (hours)

FIG. 14. Relationship between haematocrit and clearance half-time (t,,J of erythrocytes coated with immunoglobulin G in acute falciparum malaria. (Reproduced from Ho et al., 1990a.)

In acute infections, failure to augment rapidly both filtration and macrophage clearance functions would allow unrestrained parasite multiplication to occur and severe malaria to develop. It is not clear which of these two functions is the more important. Increased filtration, and particularly Fcmediated clearance function, persist for many weeks after acute infection (Lee et al., 1989; Ho et al., 1990a), and almost certainly contribute to the persistent shortened red cell survival time and the anaemia following acute malaria (Looareesuwan et al., 1987b). There have been few detailed ultrastructural studies of the spleen in human malaria. In one case report (Pongponratn et al., 1989) of a 13-yearold boy who died of acute renal failure 5 days after admission to hospital, cordal macrophage erythrophagocytosis was prominent. The phagocytosed parasitized erythrocytes contained mainly ring forms and young trophozoites. Although red cells containing more mature parasites should be preferentially cleared by the spleen, being both less deformable and more likely to be opsonized, they also sequester and thereby escape splenic removal. This balance between removal of infected cells by the spleen and their cytoadherence to vascular endothelium is critical to the pathogenesis of



malaria, as it determines the multiplication rate in vivo, and therefore, ultimately, disease severity (Section V.A). V.


The macrophage is the major immunological effector cell in malaria (Taliaferro and Mulligan, 1937), but neutrophils too play their part. The low peripheral blood neutrophil counts observed in acute malaria have been attributed to a shift to the marginal pool (Dale and Wolff, 1973). Neutrophil counts may be raised in very severe disease, indeed neutrophilia indicates a poor prognosis (Warrell et al., 1982; Stein, 1987; Molyneux et al., 1989a; WHO, 1990). Neutrophils, like macrophages, can ingest free merozoites and infected erythrocytes and commonly contain malarial pigment. In vitro, neutrophils can be shown to be activated by co-incubation with malaria parasites, and they can kill malaria trophozoites and meronts, probably by singlet oxygen release (Nnalu and Friedman, 1988). Evidence for neutrophil activation in vivo comes from observations that the proportion of circulating neutrophils which give a positive reaction with nitroblue tetrazolium is increased (Andersen, 1971), and that plasma levels of polymorphonuclear leucocyte elastase (a granule protein) are increased in severe malaria (R. Clemens and S . Pukrittayakamee, personal communication). Whether neutrophils cause direct tissue damage, such as that causing acute pulmonary oedema, remains to be determined. Defective neutrophil and monocyte chemotaxis has been reported in malaria (Nielsen et al., 1986) and patients with severe malaria are vulnerable to bacterial urinary tract infections, respiratory infections and septicaemias (Bygbjerg and Lanng, 1982; Kharazmi et al., 1987). In acute malaria the eosinophil count is consistently suppressed (Davis et al., 1991), with a rebound eosinophilia in convalescence. Eosinophil secretory products inhibit P . fakiparum multiplication in vitro (Waters, L. S . et af., 1987), but whether the eosinophil contributes to malarial immunity remains uncertain. W.


Acute malaria is associated with both activation and suppression of the immune system (Brasseur et af., 1983; Druilhe et al., 1983). Several studies have shown alterations in peripheral blood lymphocyte subpopulations, but this has not led to a clearer understanding of malaria immunology. This is probably because the peripheral blood populations may not reflect those in the spleen and liver, where much of the immunological activity takes place. 1.

Immune suppression

In acute falciparum malaria there is impaired T cell responsiveness to



malaria-specific antigens. The immune unresponsiveness is manifested as a defect both in proliferation to malaria-specific antigen in vitro (Ho et al., 1986) and, in some cases, malaria-specific antibody production in vivo (Webster et al., 1987). In severe disease, the immune unresponsiveness extends to unrelated antigens (Brasseur et al., 1983; Druilhe et al., 1983), and there is a tendency to develop supervening bacterial infections. Although autoantibodies have been detected (Greenwood, 1968), it is unlikely that they contribute to the pathology of acute falciparum malaria. Histologically, there is a conspicuous lack of immunological damage in all the organs examined from patients who die of the disease (MacPherson et al., 1985). This does not, however, rule out an important role for soluble mediators causing tissue damage. The mechanism of the immune suppression in acute falciparum malaria is complex. A defect in production of IL-2 and IL-2 receptor expression in response to malaria-specific antigens has been demonstrated (Ho et al., 1988). This might suggest an increase in suppressor T cell activity, but although CD8+ and adherent suppressor cells have been reported in immune individuals (Riley et al., 1989a,b), neither has been demonstrated during an acute infection (Ho et al., 1986, 1988). It is now clear that within the CD4+ subpopulation of T lymphocytes there are helper-inducer and suppressor-inducer cells which are distinguishable by additional cell surface markers. Alterations in the proportions of these CD4 cell populations may contribute to the immune abnormalities in acute falciparum malaria, as they do in visceral leishmaniasis (Cillari et al., 1991). In most patients, T lymphocyte function returns to normal by 4-6 weeks after therapy.


2. A ct ivation In acute malaria there is intense cellular immune activation, with markedly elevated levels of circulating soluble factors such as the IL-2 receptor, CD8 antigen and IFN-)I (Ho and Webster, 1990). This immunological reactivity probably results from non-specific T cell activation, since it fails to control the developing infection, and may result in exaggerated production of cytokines and subsequent immunopathology. Increased numbers of y/6 T cells have been reported in both the peripheral blood (Ho et al., 1990b) and spleen (Bordessoule er al., 1990) of patients with acute falciparum malaria. It is not known whether they contribute to the pathophysiology of the disease. 3. Burkitt 's lymphoma Burkitt's lymphoma, an uncontrolled monoclonal proliferation of B lymphocytes, is associated with both malaria and Epstein-Barr virus (EBV)



infection. In areas of intense malaria transmission, Burkitt’s lymphoma is the most common malignancy in childhood. EBV infection is acquired in infancy and is widespread in the tropics. Progression of the infection is normally controlled by virus-specific cytotoxic T cells. The EBV-specific cytotoxic T cell response is significantly decreased (Whittle er a]., 1984) and there is increased proliferation of EBV-carrying lymphocytes during acute malaria (Lam et al., 1991). The resulting increase in virus-infected B cells predisposes to cytogenetic abnormalities and the consequent development of malignancy. 4. Hyper-reac t ive malarial splenomegaly Hyper-reactive malarial splenomegaly (HMS), previously known as the tropical splenomegaly syndrome, represents another aberrant immunological response to malarial parasites. In this condition, there is hypergammaglobulinaemia and massive enlargement of the spleen which leads to hypersplenism with anaemia, leukopenia and thrombocytopenia (Crane, 198I). There is increased susceptibility to infection and increased mortality. Although malaria parasites are rarely demonstrable in the blood, the spleen shrinks and symptoms resolve with effective antimalarial prophylaxis. The hypergammaglobulinaemia is believed to be the result of polyclonal B cell activation in the absence of adequate numbers of CD8+ suppressor T cells (Hoffman et al., 1984), which are removed by an antibody-dependent cytotoxic mechanism (Piessens et al., 1985). In a sub-group of HMS patients who became resistant to antimalarial treatment, rearrangements of the immunoglobulin gene have recently been demonstrated (Bates et al., 1991). The finding suggests that the clonal lymphoproliferation in these patients may evolve into a malignant proliferative disorder such as chronic lymphocytic leukaemia. X.




Complement activation in malaria is proportional to disease severity (Adam et al., 1981; Kidwai et al., 1986). Classical pathway activation predominates (Petchclai et al., 1977; Phanuphak et a[., 1985). Cryoglobulins and circulating immune complexes are detectable, and in one study peak levels of immune complexes and C, breakdown products (e.g. C,d) were correlated with thrombocytopenia (Adam er al., 1981). Hypocomplementaemia has also been documented during the paroxysms of vivax malaria relapse (i.e. at



the time of merogony) (Neva et al., 1974). In both falciparum and vivax malaria, concentrations of the later components of the complement pathway were not reduced (Neva et al., 1974; Petchclai et al., 1977). Despite these associations, there is no convincing evidence that complement activation or immune complexes have a pathological role in acute malaria. In murine malaria, immune complexes inhibit Fc receptor-mediated phagocytosis, but there is no evidence for this in humans (Ho et al., 1990a). 2. Quartan nephropathy Chronic immune complex deposition may cause the glomerulonephritis associated with P . mafariae infections (quartan nephropathy) (Gilles and Hendrickse, 1963; Allison et al., 1969; Hendrickse et al., 1977). In this condition, children living in an endemic area present with nephrotic syndrome which progresses remorselessly to chronic renal failure. Antibodies to P . malariae, and in some cases malaria antigen, have been detected by renal biopsy (Houba, 1975). The histopathological findings vary considerably, but there is some evidence that a coarse granular pattern of immunofluorescent staining in the glomeruli is associated with response to cytotoxic therapy, whereas patients with a fine granular pattern or linear staining do not respond (Hendrickse et a f . , 1971). Why quartan nephropathy develops in a few children, whereas the vast majority of those infected have no renal pathology, is not known. Y.


Falciparum malaria adversely affects pregnancy (Archibald, 1956; Brabin, 1983; Nosten et al., 1991; Brabin and Brabin, 1992). In areas of intense transmission, where symptomatic disease in adults is rare, these adverse effects are largely confined to the foetus of the first pregnancy (Jelliffe, 1968; McGregor et al., 1983). There is intra-uterine growth retardation and a mean reduction in the weight of babies born to primigravidae of approximately 170g (Clark, H. C., 1915; Brabin, 1983). The placenta is a site of preferential parasite sequestration (Bray and Sinden, 1979) and commonly contains large numbers of mature parasites and abundant pigment (and is sometimes black in gross appearance), even when the peripheral blood film is negative. There may also be trophoblastic thickening (Galbraith et al., 1980; Walter et al., 1982), macrophage infiltration and perivillous fibrin deposition (Philippe and Walter, 1985; Anagnos et al., 1986), which presumably interfere with transplacental exchange of substrates and metabolites and reduce foetal growth. By electron microscopy, erythrocytes containing mature parasites are seen free in the vascular spaces and not adherent to the



endothelial surfaces (M. Aikawa, personal communication). How sequestration occurs in the absence of cytoadherence is unclear. Placental parasitization is presumably present for many weeks or months in immune women living in areas of intense malaria transmission. In holo- or hyperendemic areas, there is an increased incidence of maternal anaemia associated with falciparum malaria (Fleming et al., 1969; Gilles et al., 1969; Greenwood et al., 1989), which on occasions may be severe (Fleming, 1989), but most women are asymptomatic. This contrasts sharply with observations in areas of unstable transmission (Menon, 1972; Nosten et al., 1991), where infections with P . falciparum in non-immune pregnant women tend to be severe (Bray and Anderson, 1979; Looareesuwan et al., 1985). Pregnant women with cerebral malaria have a casespecific mortality rate over twice that of non-pregnant adults. Premature labour and foetal death are common. In mesoendemic areas even transient exposure to malaria (i.e. promptly treated disease) is associated with a significant reduction in birthweight (150-180 g) in the first, second and third pregnancies (Nosten et al., 1991). The natural immunosuppression of pregnancy and the pre-partum level of immunity are important determinants of disease severity. There is also evidence of malaria antigen-specific cellular unresponsiveness in pregnancy (Riley et al., 1989b). The notion that the placenta is an “immunologically privileged” site (like the eye, brain and testes) has been used to account for the localization of parasites there; i.e. effector mechanisms operating elsewhere in the body which kill parasites do not operate in the placenta (Riley et al., 1989b). Abnormal steroid metabolism, notably increases in plasmafree cortisol, have been documented in Tanzanian women with malaria in pregnancy (Vleugels et al., 1987), but whether this accounts for pregnancy immunodepression is uncertain. In murine models, abnormalities in steroid metabolism are correlated with immunodepression (Van Zon et al., 1982, 1985), and cytokines appear to be important in the pathogenesis of malariainduced abortion, but whether these observations can be extrapolated to humans also remains to be determined. Pregnant women with severe malaria have an increased risk of pulmonary oedema and hypoglycaemia (White et al., 1983; Looareesuwan et al., 1985). There is an accelerated metabolic response to starvation (Metzger et al., 1982), and the pancreatic P-cell response to secretagogues such as glucose (Spellacy and Goetz, 1963) and quinine is increased (Looareesuwan et ul., 1985; T. M. E. Davis and N. J. White, unpublished observation). As a consequence, pregnant women are particularly vulnerable to quininestimulated hyperinsulinaemic hypoglycaemia (White et al., 1983).



VIII. CONCLUSIQN In the past decade detailed studies of human malaria infections in different age groups and geographical settings have reformulated our ideas on the pathophysiology of the disease. The roles of cytoadherence, rosetting and cytokine release have come to the fore, whereas the parts played by immune damage, intravascular coagulation and increased vascular permeability have receded. Clinical investigation has taken some of the mystery out of malaria, but we still know relatively little for certain. The next challenge is to translate these advances in our understanding of pathophysiology into improved treatments. REFERENCES Abdallah, S. H. (1986). Red cell associated IgG in patients suffering from Plasmodium falciparum malaria. British Journal of Haematology 62, 13-19. Abdallah, S., Weatherall, D. J., Wickramasinghe, S. N. and Hughes, M. (1980). The anaemia of P. falciparum malaria. British Journal of Haematology 46, 171-183. Adam, C., Geniteau, M., Gougerot-Pocidalo, M., Verroust, P., Lebras, J., Gibert, C. and Morel-Maroger, L. (198 I). Cryoglobulins, circulating immune complexes and complement activation in cerebral malaria. Infection and Immunity 31, 53&535. Adner, M. M., Alsratt, C. B. and Conrad, M. E. (1968). Coombs positive haemolytic disease in malaria. Annals of Internal Medicine 68, 33-38. Allison, A. C., Houba, V., Hendrickse, R. G., de Petris, S., Edington, G. M. and Adeniyi, A. (1969). Immune complexes in the nephrotic syndrome of African children. Lancet ii, 1232-1237. Anagnos, D., Lanoie, L. O., Palmieri, J. R., Ziefer, A. and Connor, D. H. (1986). Effects of placental malaria on mothers and neonates from Zaire. Zeitschrift fur Parasitenkunde 72, 57-64. Andersen, B. A. (1971). The NBT test in malaria. Lancet ii, 317-319. Angus, M. G. N., Fletcher, K. A. and Maegraith, B. G. (1971). Studies on the lipids of Plasmodium knowlesi-infected rhesus monkeys (Macaca mulatta). 111. Changes in liver lipids. Annals of Tropical Medicine and Parasitology 65, 419427. Archibald, H. M. (1956). The influence of malaria infection of the placenta on the incidence of prematurity. Bulletin of the World Health Organization, 15, 842-845. Areekul, S. (1988). Transcapillary escape rate and capillary permeability to albumin in patients with Plasmodium falciparum malaria. Annals of Tropical Medicine and Parasitology 82, 135-140. Arieti, S. (1946). Histopathologic changes in cerebral malaria and their relation to psychotic sequelae. Archives of Neurology and Psychiatry 56, 79-104. Arthachinta, S., Sitprija, V. and Kashemsant, U. (1974). Selective renal angiography in renal failure due to infection. Australian Journal of Radiology 18, 446452. Asch, A. S., Barnwell, J., Silverstein, R. L. and Nachman, R. L. (1987). Isolation of the thrombospondin membrane receptor. Journal of Clinical Investigation 79, 1054106 1.



Aung-Kyaw-Zaw, Khin-Maung-U and Myo-Thwe (1988). Endotoxaemia in complicated falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 82, 5 13-5 14. Badibanga, B., Dayal, R., Depierreux, M., Pidival, G. and Lambert, P. H. (1986). Etude des principaux facteurs immunologiques et de la barriere hemato-meninges au cours de la malaria drkbrale chez l’enfant en pays d’endkmie (Zaire). Annales de la SociPtP Belge de MPdecine Tropicale 66, 23-27. Baird, J. K., Jones, T. R., Danudirgo, E. W., Annis, B. A., Bangs, M. J., Basri, H., Purnomo and Masbar, S. (1991). Age dependent acquired protection against Plasmodium falciparum in people having two years exposure to hyperendemic malaria. American Journal of Tropical Medicine and Hygiene 45, 65-76. Barnwell, J. W., Ockenhouse, C. F. and Knowles, D. M. (1985). Monoclonal antibody OKM5 inhibits the in-vitro binding of Plasmodium falciparum infected erythrocytes to monocytes, endothelial and C32 melanoma cells. Journal of Immunology 135, 3494-3497. Barnwell, J. W., Asch, A. S.,,Nachman, R. L., Yamaya, M., Aikawa, M. and Ingravallo, P. (1989). A human 88-kD membrane glycoprotein (CD36) functions in-vitro as a receptor for a cytoadherence ligand on Plasmodium falciparum infected erythrocytes. Journal of Clinical Investigation 84, 765-772. Barratt, J. 0. W. and Yorke, W. (1909-1910). An investigation into the mechanism of production of blackwater. Annals of Tropical Medicine and Parasitology 3, 1-256. Bate, C. A. W., Taverne, J. and Playfair, J. H. L. (1988). Malaria parasites induce TNF production by macrophages. Immunology 64, 227-23 I . Bate, C. A. W., Tqverne, J. and Playfair, J. H. L. (1989). Soluble malarial antigens are toxic and induce the production of tumour necrosis factor in vivo. Immunology 66,600-605. Bate, C . A. W., Taverne, J., Dave, A. and Playfair, J. H. (1990). Malaria exoantigens induce T-independent antibody that blocks their ability to induce TNF. Immunology 70, 3 15-320. Bates, I., Bedu-Addo, G., Bevan, D. H. and Rutherford, T. R. (1991). Use of immunoglobulin gene rearrangements to show clonal lymphoproliferation in hyper-reactive malarial splenomegaly. Lancet 337, 505-507. Bennish, M. L., Azad, A. K., Rahman, 0. and Phillips, R. E. (1990). Hypoglycemia during diarrhoea in childhood. New England Journal of Medicine 332, 1357-1360. Berendt, A. R.,Simmons, D. L., Tansey, J., Newbold, C. I. and Marsh, K. (1989). Intracellular adhesion molecule-1 is an endothelial cell adhesion receptor for Plasmodium falciparum. Nature 341, 57-59. Bergin, J. J. (1967). Malaria and the lung. Military Medicine 132, 522-526. Bernadin, G., Fournier, J. P., Quaranta, J. F., Marty, P., Dellamonica, P. and Mattei, M. (1989). Syndrome de detresse respiratoire aigue d’kvolution fatale aprks exsanguinotransfusion au cours d‘un acces pernicieux palustre. La Presse MPdicale 18, 13941395. Bhamarapravati, N., Boonpucknavig, S., Boonpucknavig, V. and Yaemboonruang, C. (1973). Glomerular changes in acute Plasmodium falciparum infection. Archives of Pathology 96, 289-293. Biggs, B. A., Culvenor, J. G., Ng, J. S., Kemp, D. J. and Brown, G. V. (1989). Plasmodium falciparum: cytoadherence of a knobless clone. Experimental Parasitology 69, 189-197.



Biggs, B. A., Gooze, L., Wycherley, K., Wilkinson, D., Boyd, A. W., Forsyth, K. P., Edelman, L., Brown, G. V. and Leech, J. H. (1990). Knob-independent cytoadherence of Plasmodium falciparum to the leucocyte differentiation antigen CD36. Journal of Experimental Medicine 171, 1983-1992. Blackburn, C. R. B., Hensley, W. J., Grant, D. K. and Wright, F. B. (1954). Studies on intravascular hemolysis in man. Pathogenesis of initial stages of acute renal failure. Journal of Clinical Investigation 33, 825-834. Blackie, W. K. (1944). Blackwater fever. Clinical Proceedings 3, 272-312. Blanloeil, D., Baron, D., de Lajartre, A. Y. and Nicolas, F. (1980). Syndrome de detresse respiratoire aigue de I’adulte au cours de I’accks pernicieux palustre. A propos d’un cas. Skminaire de I’H6pital de Paris 56, 1088-1090. Boonpucknavig, V. and Sitprija, V. (1979). Renal disease in acute Plasmodium falciparum infection in man. Kidney International 16,44-52. Boonpucknavig, V., Boonpucknavig, S., Udomsangpetch, R. and Nitiyanant, P. (1990). An immunofluorescent study of cerebral malaria. Archives of Pathology and Laboratory Medicine 114, 1028-1034. Bordessoule, D., Ganland, P. and Mason, D. Y. (1990). Preferential localisation of human lymphocytes bearing gamma delta T cell receptors to the red pulp of the spleen. Journal of Clinical Pathology 43,461464. Borochovitz, D., Crosley, A. and Metz, J. (1970). Intravascular coagulation with fatal haemorrhage in cerebral malaria. British Medical Journal ii, 710. Brabin, B. J. (1983). Analysis of malaria in pregnancy in Africa. Bulletin of the World Health Organization 61, 1005-1016. Brabin, L. and Brabin, B. J. (1992). Parasitic infections in women and their consequences. Advances in Parasitology 31, 1-81. Brandt, J. L., Frank, N. R. and Lichtman, H. C. (1951). Effects of hemoglobin solution on renal functions in man. Blood 6, 1152-1 158. Brasseur, P., Agrapart, M., Ballett, J. J., Druilhe, P., Warrell, M. J. and Tharavanij, S. (1983). Impaired cell mediated immunity in Plasmodium falciparum infected patients with high parasitaemia and cerebral malaria. Clinical Immunology and Immunopathology 27, 38-50. Bray, R. S. and Anderson, M. J. (1979). Falciparum malaria and pregnancy. Transactions of the Royal Society of Tropical Medicine and Hygiene 73, 427431. Bray, R. S. and Sinden, R. E. (1979). The sequestration of Plasmodium falciparum infected erythrocytes in the placenta. Transactions of the Royal Society of Tropical Medicine and Hygiene 73, 7 16-7 19. Brewster, D., Kwiatkowski, D. and White, N. J. (1990). Neurological sequelae of cerebral malaria in childhood. Lancet 336, 1039-1043. Brooks, M. H., Malloy, J. P., Bartelloni, P. J., Tigertt, W.D., Sheehy, T. W. and Barry, K. G. (1967). Pathophysiology of acute falciparum malaria. Correlation of clinical and biochemical abnormalities. American Journal of Medicine 43, 735744. Brooks, M. H., Kiel, F. W., Sheehy, T. W. and Barry, K. G. (1968). Acute pulmonary edema in falciparum malaria. New England Journal of Medicine 279, 732-737. Brooks, M. H., Barry, K. G., Cirksena, W. J., Malloy, J. P., Bruton, J. and Gilliland, P. F. (1969). Pituitary-adrenal function in acute falciparum malaria. American Journal of Tropical Medicine and Hygiene 18, 872-877. Bruce-Chwatt, L. J. (1952). Malaria in African infants and children in southern Nigeria. Annals of Tropical Medicine and Parasitology 46, 173-200.



Bruce-Chwatt, L. J. (1987). Quinine and the mystery of blackwater fever. Acta Leidensia 55, 18I - 196. Butler, T. and Weber, D. M. (1973). On the nature of orthostatic hypotension in acute malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 22,43942. Butler, T., Arnold, M. and Islam, M. (1989). Depletion of hepatic glycogen in the hypoglycaemia of fatal childhood diarrhoea1 illnesses. Transactions of the Royal Society of Tropical Medicine and Hygiene 83, 839-843. Bygbjerg, I. C. and Lanng, C. (1982). Septicaemia as a complication of falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 76, 705. Canfield, C. J. (1969). Renal and hematologic complications of acute falciparum malaria in Vietnam. Bulletin of the New York Academy of Medicine 45, 10431057. Canfield, C. J., Miller, L. H., Bartellino, P. J., Eichler, P. and Barry, K. G. (1968). Acute renal failure in Plasmodium falciparum malaria. Treatment by peritoneal dialysis. Archives of Internal Medicine 122, 199-203. Carlson, J. C., Holmquist, G., Taylor, D. W., Perlmann, P. and Wahlgren, M. ( I990a). Antibodies to a histidine-rich protein (PMRPI) disrupt spontaneously formed Plasmodium falciparum erythrocyte rosettes. Proceedings of the National Academy of Sciences of the USA 87, 251 1-2515. Carlson, J., Helmby, H., Hill, A. V. S., Brewster, D., Greenwood, B. M. and Wahlgren, M. (1 990b). Human cerebral malaria: association with erythrocyte rosetting and lack of anti-rosetting antibodies. Lancet 336, 1457-1460. Chan, T. K., Todd, D. and Tsao, S. C. (1976). Drug-induced haemolysis in glucose6-phosphate dehydrogenase deficiency. British Medical Journal ii, 1227-1 229. Charoenlarp, P., Vanijanonta, S. and Chat-Panyaporn, P. (1979). The effect of prednisolone on red cell survival in patients with falciparum malaria. South-East Asian Journal of Tropical Medicine and Public Health 10, 127-131. Charoenpan, P., Indraprasit, S., Kiatboonsri, S., Surachittanont, 0. and Tanomsup, S. (1990). Pulmonary edema in severe falciparum malaria. Hemodynamic study and clinicophysiologic correlation. Chest 97, 1 190-1197. Chipman, M., Ladigan, F. C. and Benjapongse, W. (1966). Malaria and the nervous system. Clinical experience with a hospital population in an endemic area of Thailand. S E A T 0 Reports, pp. 180-184. Chongsuphajaisiddhi, T., Kasemuth, R., Tejavanija, S. and Harinasuta, T. (1971). Changes in blood volume in falciparum malaria. South-East Asian Journal of Tropical Medicine and Public Health 2, 344-350. Clark, C. and Tomlinson, W. J. (1949). Pathological anatomy of malaria. In “Malariology” (M. F. Boyd, ed.), pp. 874-883. W. B. Saunders, Philadelphia. Clark, H. C. (1915). The diagnostic value of the placental blood film in aestivoautumnal malaria. Journal of Experimental Medicine 2 2 , 4 2 7 4 . Clark, I. A. (1987). Does endotoxin cause both the disease and parasite death in acute ,malaria and babesiosis? Lancet ii, 75-77. Clark, I. A. and Chaudhri, G. (1988a). Tumour necrosis factor may contribute to the anaemia of malaria by causing dyserythropoiesis and erythrophagocytosis. British Journal of Haematology 70,99-103. Clark, I. A. and Chaudhri, G. (1988b). Tumor necrosis factor in malaria-induced abortion. American Journal of Tropical Medicine and Hygiene 39,246249.



Clark, I. A., Virelizier J.-L., Carswell, E. A. and Wood, P. R. (1981). Possible importance of macrophage-derived mediators in acute malaria. Infection and Immunity 32, 1058-1066. Clark, I. A., Mackie, E. J. and Cowden, W. B. (1986). Injection of free radical generators causes premature onset of tissue damage in malaria-infected mice. Journal of Pathology 148, 301-305. Clark, I. A., Hunt, N. H., Butcher, G. A. and Cowden W. B. (1987). Inhibition of murine malaria (Plasmodium chabaudi) in vivo by recombinant y interferon or tumor necrosis factor, and its enhancement by butylated hydroxyanisole. Journal of Immunology 139, 3493-3496. Clark, I. A., Ilschner, S., MacMicking, J. D. and Cowden, W. B. (1990). TNF in Plasmodium berghei Anka-induced cerebral malaria. Immunology Letters 25, 195198. Collomb, H., Rey, M., Dumas, M., Nouhayi, A. and Petit, M. (1967). Les hemiplegies au cours du paludisme aigiie. Bulletin de la SociPtk Mkdicale d’Afrique Noire Langue Francaise 12, 791-795. Conn, H. L., Wood, J. C. and Rose, J. C. (1956). Circulatory and renal effects following transfusion of human blood and its components to dogs. Circulation Research 4, 18-24. Conrad, M. E. (1969). Pathophysiology of malaria. Hematologic observations in human and animal studies. Annals of Internal Medicine 70, 134-140. Corbett, C. E. P., Duarte, M. I. S., Lancellotti, C. L. P., Silva, M. A. L. G. and Andrade, H. F. (1989). Cytoadherence in human falciparum malaria as a cause of respiratory distress. Journal of Tropical Medicine and Hygiene 92, 112-120. Corcoran, T. E., Hegstrom, G. J., Zoeckler, S. J. and Keil, P. G. (1953). Liver structure in non fatal malaria. Gastroenterology 24, 5 3 4 2 . Covell, G. (1955). Spontaneous rupture of the spleen. Tropical Diseases Bulletin 52, 705-723. Crane, G. (198 I). Tropical splenomegaly: Part 11. Oceania. Clinics in Haematology 10,976-982. Cranston, H. A., Boylan, C. W., Carroll, G. L., Sutera, S., Williamson, J. R., Gluzman, I. Y. and Krogstad, D. J. (1984). Plasmodium falciparum maturation abolishes physiologic red cell deformability. Science 223, 4 0 W 0 2 . Curfs, J. H. A. J., van der Meer, J. M., Saueerwein, R.W. and Eling, W. M. (1990). Low dosages of interleukin 1 protect mice against lethal cerebral malaria. Journal of Experimental Medicine 172, 1287-1291. Currie, B. and Kevau, I. (1988). Hypoglycaemia in malaria. Papua New Guinea Medical Journal 31, 6 5 4 7 . Dale, D. C. and Wolff, S. M. (1973). Studies of the neutropenia of acute malaria. Blood 41, 205. Das, B. A., Satpathy, S. K., Mohanty, D., Mohanty, S., Mishra, S. K., Satapathy, P. C., Patnaik, J. K. and Bose, T. K. (1988). Hypoglycaemia in severe falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 82, 197-20 I . David, P. H., Hommel, M., Miller, L. H., Udeinya, I. J. and Oligino, L. D. (1983). Parasite sequestration in Plasmodium falciparum malaria: spleen and antibody modulation of cytoadherence of infected erythrocytes. Proceedings of the National Academy of Sciences of the USA 80, 5075-5079.



David, P. H., Handunnetti, S. M., Leech, J. H., Gamage, P. and Mendis, K. N. (1988). Rosetting: a new cytoadherence property of malaria-infected erythrocytes. American Journal of Tropical Medicine and Hygiene 38, 289-297. Davis, T. M. E., Looareesuwan, S., Pukrittayakamee, S., Levy, J., Krishna, S., Vilaiwanna, N., Roche, J., Turner, R. and White, N. J. (1988). Glucose turnover in severe falciparum malaria. X I I International Congress for Tropical Medicine and Malaria, Amsterdam, The Netherlands, 18-23 September 1988, p. 87. [Abstract.] Davis, T. M. E., Supanaranond, W., Pukrittayakamee, S., Krishna, S., Looareesuwan, S., Hart, G., Burrin, J. M. and White, N. J. (1990a). The pituitary-thyroid axis in severe falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 84, 330-335. Davis, T. M. E., Karbwang, J:, Looareesuwan, S., Turner, R. C. and White, N. J. (1990b). Comparative effects of quinine and quinidine on glucose metabolism in normal man. British Journal of Clinical Pharmacology 30, 397403. Davis, T. M. E., Pukrittayakamee, S., Supanaranond, W., Looareesuwan, S., Krishna, S., Nagachinta, B., Turner, R. C . and White, N. J. (1990~).Glucose metabolism in quinine-treated patients with uncomplicated falciparum malaria. Clinical Endocrinology 33, 739-749. Davis, T. M. E., Krishna, S., Looareesuwan, S., Supanaranond, W., Pukrittayakamee, S., Attatamsoonthorn, K. and White, N. J. (1990d). Erythrocyte sequestration and anemia in severe falciparum malaria. Analysis of acute changes in venous hematocrit using a simple mathematical model. Journal of Clinical Investigation 86, 793-800. Davis, T. M. E., Ho, M., Supanaranond, W., Looareesuwan, S., Pukrittayakamee, S. and White, N. J. (1991). Changes in the peripheral blood eosinophil count in falciparum malaria. Acta Tropica 48,243-245. Deaton, J. G. (1970). Fatal pulmonary oedema as a complication of acute falciparum malaria. American Journal of Tropical Medicine and Hygiene 19, 196201. De Brito, T., Barone, A. A. and Earia, R. M. (1969). Human liver biopsy in P. falciparum and P. vivax malaria. A light and electron microscopy study. Virchows Archiv 348, 220-229. Deller, J. J., Cifarelli, P. S., Berque, S. and Buchanan, R. (1967). Malaria hepatitis. Military Medicine 132, 614620. Dennis, L. H., Eichelbeger, J. W., Inman, M . M. and Conrad, M. E. (1967). Depletion of coagulation factors in drug resistant Plasmodium fakiparum malaria. Blood 29, 7 13-72 I . De Silva, H. J., Goonetilleke, A. K. E., Senaratna, N., Remesh, N., Jayawickrama, U. S., Jayasinghe, K. S. A. and Amarasekera, L. R. (1988). Skeletal muscle necrosis in severe falciparum malaria. British Medical Journal 296, 1039. Desowitz, R. S. (1987). The pathophysiology of malaria after Maegraith. Annals of Tropical Medicine and Parasitology 81, 599-606. Devakul, K., Harinasuta, T. and Reid, H. A. (1966). lZSI-labelled fibrinogen in cerebral malaria. Lancet ii, 886888. Dhayagude, R. G. and Puranare, N. M. (1943). Autopsy study of cerebral malaria with special reference to malarial granuloma. Archives of Pathology 36, 550-558. Druilhe, P., Brasseur, P., Agrapart, M., Ballet, J. J., Chanthavanich, P., Looareesuwan, S., White, N. J., Warrell, M. J., Warrell, D. A., Tharavanij, S. and Gentilini, M. (1983). T-cell responsiveness in severe Plasmodium fakiparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 77, 671-672.



Duarte, M. I. S., Corbett, C. E. P., Boulos, M. and Amata Neto, V. (1985). Ultrastructure of the lung in falciparum malaria. American Journal of Tropical Medicine and Hygiene 34, 31-35. Dudgeon, L. S. and Clarke, C. (1917). A contribution to the microscopical histology of malaria. Lancet ii, 153-1 56. Dudgeon, L. S. and Clarke, C. (1918). An investigation on fatal cases of pernicious malaria caused by Plasmodium falciparum in Macedonia. Quarterly Journal of Medicine 12, 372-390. Dukes, D. C., Sealey, B. J. and Forbes, J. I. (1968). Oliguric renal failure in blackwater fever. American Journal of Medicine 45, 899-903. Dunn, M. J. (1969). Alterations of red blood cell sodium transport during malarial infection. Journal of Clinical Investigation 48, 674-684. Edington, G. M. (1954). Cerebral malaria in the Gold Coast African. Four autopsy reports. Annals of Tropical Medicine and Parasitology 48, 30G306. Edington, G. M. (1967). Pathology of malaria in West Africa. British Medical Journal i, 7 15-7 18. Ehrich, J. H. H., Beck, E. J., Haberkorn, A. and Meister, G. (1984). Causes of death in lethal rat malaria. Tropenmedizin und Parasitologie 35, 127-1 30. Essien, E. (1989). The circulating platelet in acute malaria infection. British Journal of Haematology 72, 589-590. Evans, R. D., Argiles, J. M. and Williamson, D. H. (1989). Metabolic effects of tumour necrosis factor (cachectin) and interleukin 1. Clinical Science 77,357-364. Facer, C. A. (1980). Direct Coombs’ antiglobulin reactions in Gambian children with Plasmodiumfalciparum malaria. 11. Specificity of erythrocyte bound IgG. Clinical and Experimental Immunology 39, 279-288. Facer, C. A., Bray, R. S. and Brown, J. (1979). Direct Coombs’ antiglobulin reactions in Gambian children with Plasmodium falciparum malaria. I. Incidence and class specificity. Clinical and Experimental Immunology 35, 1 19-127. Fairley, N. H. (1947). Sidelights on malaria in man obtained by subinoculation experiments. Transactions of the Royal Society of Tropical Medicine and Hygiene 40,521-676. Fajardo, L. F. and Rao, S. (1971). Platelet enlargement in malaria. Military Medicine 136,463464. Falconer, A. W. (1919). The pulmonary manifestations in malaria. Quarterly Journal of Medicine 13, 25-34. Fein, L. A., Rackow, E. C. and Shapiro, L. (1978). Acute pulmonary edema in Plasmodium falciparum malaria. American Review of Respiratory Diseases 118, 425-429. Feldman, H. A. and Murphy, F. D. (1945). The effects of alterations in blood volume on the anaemia and hypoproteinaemia of human malaria. Journal of Clinical Investigation 24, 780-792. Feldman, R. M. and Singer, C. (1987). Non-cardiogenic pulmonary edema and pulmonary fibrosis in falciparum malaria. Review of Infectious Diseases 9, 134139. Field, J. W. (1949). Blood examination and prognosis in acute falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 43, 3 3 4 8 . Field, J. W. and Niven, J. C. (1937). A note on prognosis in relation to parasite counts in acute subtertian malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 30, 569-574. Fisher, C. W. S. (1983). Clinical curio: acidosis and hypoglycaemia in malaria. British Medical Journal 286, 126 1 .



Fitz-Hugh, T., Pepper, D. S. and Hopkins, H. U. (1944). The cerebral form of malaria. Bulletin of the U S Army Medical Department 83, 3948. Fleming, A. F. (1989). Tropical obstetrics and gynaecology. 1. Anaemia in pregnancy in tropical Africa. Transactions of the Royal Society of Tropical Medicine and Hygiene 83, 44 1 4 8 . Fleming, A. F. and Allan, N. C. (1969). Severe haemolytic anaemia in pregnancy in Nigerians treated with prednisolone. British Medical Journal 4, 461466. Fleming, A. F., Allan, N. C. and Stenhouse, N. S. (1969). Haemolytic anaemia in pregnancy in Nigeria: recognition by simple laboratory procedures. West African Medical Journal 18, 82-88. Futrakul, P., Boonpucknavig, V., Boonpucknavig, S., Mitrakul, C. and Bhamarapravati, N. (1974). Acute glomerulonephritis complicating Plasmodium falciparum infection. Clinical Pediatrics 13, 281-283. Gage, A. (1926). Algid malaria. Therapeutics Gazette 50, 77-81. Galbraith, R. M., Fox, H., Hsi, B., Galbraith, G. M. P., Bray, R. S. and Faulk, W. P. (1980). The human materno-foetal relationship in malaria. 11. Histological, ultrastructural and immunopathological studies of the placenta. Transactions of the Royal Society of Tropical Medicine and Hygiene 74, 61-12. Gardiner, M., Smith, M. L., Kagstrom, M. L., Shohami, E. and Siesjo, B. K.(1982). Influence of the blood glucose concentration on brain lactate accumulation during severe hypoxia and subsequent recovery of brain energy metabolism. Journal of Cerebral Blood Flow and Metabolism 2,429438. Garnham, P. C. C. (1966). “Malaria Parasites and other Haemosporidia.” Blackwell Scientific Publications, Oxford. Garnham, P. C. C. (1970). The role of the spleen in protozoal infections with special reference to splenectomy. Acta Tropica 27, 1-14. Garnham, P. C. C. (1988). Malaria parasites of man: life cycles and morphology. In “Malaria: Principles and Practice of Malariology” (W. H. Wernsdorfer and I. McGregor, eds), pp. 61-69. Churchill Livingstone, Edinburgh. Gaskell, J. F. and Miller, W. L. (1920). Studies on malignant malaria in Macedonia. Quarterly Journal of Medicine 13, 38 1 4 2 6 . Gilles, H. M. and Hendrickse, R. G. (1963). Nephrosis in Nigerian children: role of Plasmodium malariae and effect of antimalarial treatment. British Medical Journal ii, 27-31. Gilles, H. M. and Ikeme, A. C. (1960). Haemoglobinuria among adult Nigerians due to glucose-6-phosphate dehydrogenase deficiency with drug sensitivity. Lancet ii, 889-89 1. Gilles, H. M., Lawson, J. B., Sibelas, M., Voller, A. and Allan, N. (1969). Malaria anaemia and pregnancy. Annals of Tropical Medicine and Parasitology 63, 245263. Goodall, H. B. (1981). Three aspects of intravascular coagulation in subtertian malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 75,


Grau, G. E., Piguet, P. F., Engers, H. D., Louis, J. A., Vassalli, P. and Lambert, P.-H. (1986). L3T4+ T lymphocytes play a major role in the pathogenesis of murine cerebral malaria. Journal of Immunology 137, 2348-2354. Grau, G. E., Gretner, D. and Lambert, P. H. (1987a). Prevention of murine cerebral malaria by low dose cyclosporin A. Immunology 61, 521-525. Grau, G. E., Del Guidice, G. and Lambert, P.-H. (1987b). Host immune response and pathological expression in malaria: possible implications for malaria vaccines. Parasitology 94, supplement, S 123-S 137.



Grau, G. E., Fajardo, L. F., Piquet, P. F., Allet, B., Lambert, P.-H. and Vassalli, P. (1987~).Tumor necrosis factor (cachectin) as an essential mediator in murine cerebral malaria. Science 237, 121C1212. Grau, G. E., Piguet, P. F., Gretener, D., Vesin, C. and Lambert, P.-H. (1988a). Immunopathology of thrombocytopenia in experimental malaria. Immunology 65, 501-506. Grau, G. E., Kindler, V., Piguet, P. F., Lambert, P. H. and Vassalli, P. (1988b). Prevention of experimental cerebral malaria by anticytokine antibodies. Journal of Experimental Medicine 168, 1499-1 504. Grau, G. E., Piguet, P. F., Vassalli, P. and Lambert, P. H. (1989a). Involvement of tumor necrosis factor and other cytokines in immune-mediated vascular pathology. International Archives of Allergy and Applied Immunology 88, 34-39. Grau, G. E., Piguet, P. F., Vassalli, P. and Lambert, P. H. (1989b). Tumour necrosis factor and other cytokines in cerebral malaria: experimental and clinical data. Immunological Review 112,49-70. Grau, G . E., Taylor, T. E., Molyneux, M. E., Wirima. J. J., Vassalli, P., Hommel, M. and Lambert, P. H. (1989~).Tumor necrosis factor and disease severity in children with falciparum malaria. New England Journal of Medicine 320, 1 5 8 6 1591. Grau, G. E., Heremans, H., Piguet, P. F., Pointaire, P., Lambert, P. H., Billiau, A. and Vassalli, P. (1989d). Monoclonal antibody against interferon can prevent experimental cerebral malaria and its associated overproduction of tumor necrosis factor. Proceedings of the National Academy of Sciences of the USA 86, 55725574. Grau, G. E., Frei, K., Piguet, P. F., Fontana, A., Heremans, H., Billiau, A., Vassalli, P. and Lambert, P. H. (1990). Interleukin 6 production in experimental cerebral malaria: modulation by anticytokine antibodies and possible role in hypergammaglobulinemia. Journal of Experimental Medicine 172, 1505-1 508. Greenwood, B. M. (1968). Autoimmune disease and parasitic infections in Nigeria. Lancet ii, 38C382. Greenwood, B. M., Bradley, A. K., Greenwood, A. M., Byass, P., Jammeh, K., Marsh, K., Tulloch, S., Oldfield, F. S. J. and 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. Greenwood, B. M., Greenwood, A. M., Snow, R. W., Byass, P., Bennett, S. and Hatib-N’jie, A. B. (1989). The effects of malaria chemoprophylaxis given by traditional birth attendants on the course and outcome of pregnancy. Transactions of the Royal Society of Tropical Medicine and Hygiene 83, 589-594. Gurman, G., Schlaeffer, F., Alkan, M. and Heilig, I. (1988). Adult respiratory distress syndrome and pancreatitis as complications of falciparum malaria. Critical Care Medicine 16, 205-206. Hall, A P., Charoendhum, D. and Sonkom, P. (1975). Pulmonary oedema due to fluid overload in falciparum malaria. Annual Progress Report: S E A T 0 Medical Research Laboratory, Bangkok, Thailand, pp. 226-233. Hamilton-Fairley, N. and Bromfield, R. J. (1934). Laboratory studies in malaria and blackwater fever. Part 11. Blackwater fever and haemoglobinuria. Transactions of the Royal Society of Tropical Medicine and Hygiene 28, 141-156. Handunnetti, S. M., David, P. H., Perera, K. L. R. L. and Mendis, K. N. (1989). Uninfected erythrocytes form ‘rosettes’ around Plasmodium falciparum infected erythrocytes. American Journal of Tropical Medicine and Hygiene 40, 11 5-1 18.



Hartenbower, D. L., Kantor, G. L. and Rosen, V. J. (1972). Renal failure due to acute glomerulonephritis during falciparum malaria. Case report. Military Medicine 137, 74-76. Hasler, T., Handunnetti, S. M., Aguiar, J. C., van Schravendijk, M. R., Greenwood, B. M., Lailinger, G., Cegielski, P. and Howard, R. J. (1990). In-vitro rosetting, cytoadherence and microagglutination properties of Plasmodium falciparum infected erythrocytes from Gambian and Tanzanian patients. Blood 76, 18451852. Hendrickse, R. G., Hasan, A. H., Olumide, L. 0. and Akinkunmi, A. (1971). Malaria in early childhood. An investigation of five hundred seriously ill children in whom a “clinical” diagnosis of malaria was made on admission to the Children’s Emergency Room at University College Hospital, Ibadan. Annals of Tropical Medicine and Parasitology 65, 1-20. Hendrickse, R. G., Adeniyi, A., Edington, G. M., Glasgow, E. F., White, R. H. R. and Houba, V. (1977). Quartan malarial nephrotic syndrome. Collaborative clinicopathological study in Nigerian children. Lancet i, 1 143-1 149. Henquin, J. C. (1979). Opposite effects of intracellular calcium and glucose on the potassium permeability of pancreatic islet cells. Nature 280, 66-68. Henquin, J. C., Horeman, B., Henquin, M., Verniers, J. and Lambert, A. E. (1975). Quinine-induced modifications of insulin release and glucose metabolism by isolated pancreatic islets. FEBS Letters 57, 280-284. Herzog, C., Ellis, C. J., Innes, J. A. and Fletcher, K. A. (1982). Possible role of drug malabsorption in recrudescence of falciparum malaria. Lancet ii, 1157-1 158. Hill, A. V. S., Allsapp, C. E. M., Kwiatkowski, D., Anstey, N. M., Twumasi, P., Rowe, P. A., Bennett, S., Brewster, D., McMichael, A. J. and Greenwood, B. M. (1991). Common West African HLA antigens are associated with protection from severe malaria. Nature 352, 595-600. Hill, C. J., Knight, V. and Jeffrey, G. M. (1964). Thrombocytopenia in vivax malaria. Lancet i, 240-24 1. Hills, A. G. (1971). Malarial jaundice. American Journal of the Medical Sciences 212, 45-53. Ho, M. and Webster, H. K. (1990). T cell responses in acute falciparum malaria. Immunology Letters 25, 135-1 38. Ho, M., Webster, H. K., Looareesuwan, S., Supanaranond, W., Phillips, R. E., Chanthavanich, P. and Warrell, D. A. (1986). Antigen-specific immunosuppression in human malaria due to Plasmodium falciparum. Journal of Infectious Diseases 153, 763-77 1. Ho, M., Webster, H. K., Green, B., Looareesuwan, S., Kongchareon S. and White, N. J. (1988). Defective production of and response to interleukin 2 in acute falciparum malaria. Journal of Immunology 141, 2755-2759. Ho, M., Webster, H. K., Green, B., Looareesuwan, S., Kongchareon, S. and White, N. J. (1988). Defective production of and response to interleukin 2 in acute host defence and anaemia in acute falciparum malaria. Journal of Infectious Diseases 161, 555-561. Ho, M., Webster, H. K., Tongtawe, P., Pattanapanpysat, K. and Weidanz, W. P. (1990b). Increased gamma/delta T cells in acute falciparum malaria. Immunology Letters 25, 139-142. Ho, M., Singh, B., Looareesuwan, S., Davis, T. M. E., Bunnag, D. and White N. J. (1991a). Clinical correlates of in vitro Plasmodium falciparum cytoadherence. Infection and Immunity 59, 873-878.



Ho, M., Davis, T. M. E., Silamut, K., Bunnag, D. and White, N. J. (1991b). Rosette formation of Plasmodium falciparum infected erythrocytes from patients with acute malaria. Infection and Immunity 59, 2135-2139. Hoffman, S. L., Piessens, W. F., Ratiwayanto, S., Hussein, P. R., Kurniawan, L., Piessens, P., Campbell, J. R. and Marwoto, H. A. (1984). Reduction of suppressor T lymphocytes in the tropical splenomegaly syndrome. New England Journal of Medicine 310, 337-341. Hoffman, S. L., Rustama, D., Punjabi, N. H., Surampaet, B., Sanjaya, B., Dimpudus, A. J., Campbell, J., Marwoto, H., Paleologo, F. P., McKee, K. and Laughlin, L. W. (1988). High-dose dexamethasone in quinine-treated patients with cerebral malaria: a double-blind, placebo-controlled trial. Journal of Infectious Diseases 158, 325-331. Holloway, P. A. H., Krishna, S. and White, N. J. (1991). Plasmodium berghei: lactic acidosis and hypoglycaemia in a rodent model of severe malaria. Effects of glucose, quinine and dichloroacetate. Experimental Parasitology 72, 123-133. Holmquist, G., Udomsangpetch, R., Berzins, K., Wigzell, H. and Perlmann, P. (1988). Plasmodium chabaudi antigen Pf105, Plasmodium falciparum antigen Pfl55, and erythrocyte band 3 share cross-reactive epitopes. Infection and Immunity 56, 1545-1 550. Hommel, M. and Semoff, S. (1988). Expression and function of erythrocyteassociated surface antigens in malaria. Biology of the Cell 64, 183-203. Hommel, M., David, P. and Oligino, L. (1983). Surface alterations of erythrocytes in Plasmodium falciparum malaria. Antigenic variation, antigenic diversity, and the role of the spleen. Journal of Experimental Medicine 157, 1 137-1 148. Horstmann, R. D. and Dietrich, M. (1985). Haemostatic alterations in malaria correlate with parasitaemia. Blut 51, 329-333. Horstmann, R. D., Dietrich, M., Bienzle, U. and Rashe, H. (1981). Malaria induced thrombocytopenia. Blut 42, 157-1 64. Houba, V. (1975). Immunopathology of nephropathies associated with malaria. Bulletin of the World Health Organization 52, 199-207. Howard, R. J. (1987). Antigenic variation and antigenic diversity in malaria. Contributions to Microbiology and Immunology 8, 176-2 18. Howard, R. J. and Gilladoga, A. D. (1989). Molecular studies related to the pathogenesis of cerebral malaria. Blood 74, 2603-261 8. Igarashi, I., 00, M. M., Stanley, H., Reese, R. and Aikawa, M. (1987). Knob antigen deposition in cerebral malaria. American Journal of Tropical Medicine and Hygiene 37, 5 1 1-5 15. Janota, I. and Doshi, B. (1979). Cerebral malaria in the United Kingdom. Journal of Clinical Pathology 32, 769-772. James, M. F. M. (1985). Pulmonary damage associated with falciparum malaria: a report of ten cases. Annals of Tropical Medicine and Parasitology 79, 123-138. James, S. P., Nichol, W. D. and Shute, P. G. (1932). A study of induced malignant tertian malaria. Proceedings of the Royal Society of Medicine 25, 1153-1 186. Jaroonvesama, N. ( 1972). Intravascular coagulation in falciparum malaria. Lancet i, 22 1-223. Jelliffe, E. F. P. (1968). Low birth weight and malarial infection of the placenta. Bulletin of the World Health Organization 38, 69-78. Jensen, M. D., Conley, M. and Helstowski, L. D. (1983). Culture of Plasmodium falciparum: the role of pH, glucose and lactate. Journal of Parasitology 69, 1 0 6 s 1067.



Joshi, P., Alam, A., Chandra, R., Puri, S. K. and Gupta, C. M. (1986). Possible basis for membrane changes in non parasitised erythrocytes of malaria infected animals. Biochimica et Biophysica Acta 862, [email protected] Joshi, Y. K., Tandon, B. N., Acharya, S. K., Babu, S. and Tandon, M. (1986). Acute hepatic failure due to Plasmodium falciparum liver injury. Liver 6, 357-360. Karney, W. W. and Tong, M. J. (1972). Malabsorption in Plasmodium falciparum malaria. American Journal of Tropical Medicine and Hygiene 21, 1-5. Kaul, D. K., Roth, E. F., Nagel, R. L., Howard, R. J. and Handunetti, S. M. (1991). Rosetting of Plasmodium falciparum infected red blood cells with uninfected red blood cells enhances microvascular obstruction under flow conditions. Blood 78, 8 12-8 19. Kayembe, D., Maertens, K. and De Laey, J. J. (1980). Complications oculaires de la malaria cerebrale. Bulletin de la Sociiti Belge d’Ophtalmalogie 190, 53-60. Kawo, N. G., Msengi, A. E., Swai, A. B. M., Chuwa, L. M., Alberti, K. G. M. M. and McClarty, D. G. (1990). Specificity of hypoglycaemia for cerebral malaria in children. Lancet 336, 454-457. Kean, B. H., Smith J. A. (1944). Death due to aestivo-autumnal malaria. A resume of one hundred autopsy cases 1925-1942. American Journal of Tropical Medicine and Hygiene 24, 3 17-322. Kelton, J. G., Keystone, J., Moore, J., Denomme, G., Tozman, E., Glynn, M., Neame, P. B., Gouldie, J. and Jensen, J. (1983). Immune-mediated thrombocytopenia of malaria. Journal of Clinical Invesrigarion 71, 832-836. Kern, P., Hemmer, C. J., Van Damme, J., Grubs, H. J. and Dietrich, M. (1989). Elevated tumor necrosis factor alpha and interleukin 6 serum levels as markers for complicated Plasmodium falciparum malaria. American Journal of Medicine 87, 1 39- 143. Kety, S. S., and Schmidt, C. F. (1948). The effects of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. Journal of Clinical Investigation 27, 484-49 I . Keys, A,, Wells, S., Hofiauer, F. W., Longstreet-Taylor, H. and Henschel, A. (1950). Experimental malaria in man. 11. Liver function. Journal of Clinical Investigation 29, 6 M 7 . Khan, N. U. (1945). Cerebral malaria. Journal of the Royal Army Medical Corps 84, 263-267. Kharazmi, A., Hoby, N. and Theander, T. G. (1987). Pseudomonas aeruginosa septicaemia in a patient with severe Plasmodium falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 81, 49-50. Kidwai, T., Ahmad, S. H., Ahmed, S., Hussain, Z., Malik, A. and Khan, T. A. (1986). Serum complement levels in cerebral malaria. Indian Journal of Paediarrics 23, 185-188.

Kiire, C. F. (1986). Hypoglycaemia and cerebral malaria. Postgraduate Medical Journal 6 2 , 4 0 1 4 0 2 . Kitchen, S. F. (1949). Symptomatology: general considerations and falciparum malaria. In “Malariology” (M. F. Boyd, ed.), Vol. 2, pp. 996-1017. W. B. Saunders, Philadelphia. Knisely, M. H., Stratman-Thomas, W. K. and Elliot, T. S. (1941). Observations on circulating blood in the small vessels of internal organs in living Macaca rhesus infected with malarial parasites. Anatomical Records 79, 90. Knowles, D. M., Tolidjian, B., Marboe, C., D’Agati, V., Grimes, M. and Chess, L. (1984). Monoclonal anti-human monocyte antibodies O K M 1 and O K M 5 possess



distinctive tissue distributions including differential reactivity with vascular endothelium. Journal of Immunology 132,2170-2173. Knuttgen, H. J. (1987). The bone marrow of non-immune Europeans in acute malaria infection: a topical review. Annals of Tropical Medicine and Parasitology 81, 567-576. Kofi-Ekue, J. M., Phiri, D. E. D., Mukunyandela, M., Sheth, U. K. and Wernsdorfer, W. (1988). Severe orthostatic hypotension during treatment of malaria. British Medical Journal 2%, 396. Krishna, S. and Ng, L. L. (1989). Calcium metabolism in malaria infected red cells. Experimental Parasitology 69,402-406. Krishna, S., Shoubridge, E. A., White, N. J., Weatherall, D. J. and Radda, G. K. (1983). Plasmodium yoelii: blood oxygen and brain function in the infected mouse. Experimental Parasitology 56, 39 1-396. Kutner, S., Ginsburg, H. and Cabantchik, Z. I. (1983). Permselectivity changes in malaria (Plasmodium falciparum) infected human red blood cell membranes. Journal of Cellular Physiology 114, 245-251. Kwiatkowski, D. (1989). Febrile temperatures can synchronize the growth of Plasmodium falciparum in vitro. Journal of Experimental Medicine 169, 357-361, Kwiatkowski, D. and Nowak, M. (1991). Periodic and chaotic host parasite interaction in human malaria. Proceedings of the National Academy of Sciences of the USA 88, 5111-5113. Kwiatkowski, D., Cannon, J. G., Manogue, K. R., Cerami, A., Dinarello, C. A. and Greenwood, B. M. (1989). Tumour necrosis factor production in falciparum malaria and its association with schizont rupture. Clinical and Experimental Immunology 77, 361-366. Kwiatkowski, D., Hill, A. V. S., Sambou, I., Twumasi, P., Castracane, J., Manogue, K. R., Cerami, A., Brewster, D. R. and Greenwood, B. M. (1990). T N F concentrations in fatal cerebral, non-fatal cerebral, and uncomplicated Plasmodium falciparum malaria. Lancet 336, 1201-1204. Lam, K. M. C., Syed, N., Whittle, H. and Crawford, D. H. (1991). Circulating Epstein-Barr virus-carrying B cells in acute malaria. Lancet 337, 876879. Le Camus, J., Bernard-Catinat, G., Bernard, J. and Sarrony, J. (1988). Insuffisance cardiaque aigue au cors d’un acces pernicieux palustre. La Presse Midicale 17, 165. Lee, M. V., Ambrus, J. L., De Souza, J. M. and Lee, R. V. (1982). Diminished red blood cell deformability in uncomplicated human malaria. A preliminary report. Journal of Medicine 13, 479485. Lee, S. H., Looareesuwan, S., Wattanagoon, Y., Ho, M., Wuthiekanun, V., Vilaiwanna, N., Weatherall, D. J. and White, N. J. (1989). Antibody dependent red cell removal during P.falciparum malaria: the clearance of red cells sensitised with IgG anti-D. British Journal of Haematology 73, 39W02. Leech, J. H., Barnwell, J. W., Miller, L. H. and Howard, R. J. (1984). Identification of a strain-specific malarial antigen exposed on the surface of Plasmodium falciparum infected erythrocytes. Journal of Experimental Medicine 159, 15671575. Lewis, D. J. M. (1987). Hypophosphataemia: a feature of malaria? British Medical Journal 295,4 1&4 1 7. Guo, X., Jian, H., Fan, T. and Huang, W. (1983). Development state of Li, Q. Q., Plasmodium falciparum in the intradermal, peripheral and medullary blood of patients with cerebral malaria. National Medical Journal of China 63, 692-693.



Looareesuwan, S., Warrell, D. A., White, N. J., Chanthavanich, P., Warrell, M. J., Chantaratherakitti, S., Changswek, S., Chongmankongcheep, L. and Kanchanaranya, C. (1983a). Retinal haemorrhage, a common physical sign of prognostic significance in cerebral malaria. American Journal of Tropical Medicine and Hygiene 32, 9 1 1-9 15. Looareesuwan, S., Warrell, D. A., White, N. J., Suntharasamai, P., Chanthavanich, P., Sundaravej, K., Juel-Jensen, B., Bunnag, D. and Harinasuta, T. (1983b). Do patients with cerebral malaria have cerebral oedema? A computed tomography study. Lancet i, 434-437. Looareesuwan, S., Phillips, R. E., White, N. J., Kietinun, S., Karbwang, J., Rackow, C., Turner, R. C. and Warrell, D. A. (1985). Quinine and severe malaria in late pregnancy. Lancet ii, 4-8. Looareesuwan, S., Ho, M., Wattanagoon, Y., White, N. J., Warrell, D. A., Bunnag, D., Harinasuta, T. and Wyler, D. (1987a). Dynamic alterations in splenic function in falciparum malaria. New England Journal of Medicine 317, 675479. Looareesuwan, S.,Merry, A. H., Phillips, R. E., Pleehacinda, R., Wattanagoon, Y., Ho, M., Charoenlap, P., Warrell, D. A. and Weatherall, D. J. (1987b). Reduced erythrocyte survival following clearance of malarial parasitaemia in Thai patients. British Journal of Haematology 67,473478. Looareesuwan, S., Davis, T. M. E., Pukrittayakamee, S., Supanaranond, W., Desakorn, V., Silamut, K., Krishna, S., Boonamrung, S. and White, N. J. (1991). Erythrocyte survival in severe falciparum malaria. Acta Tropica 48, 263-270. Lumlertgul, D., Keoplung, M., Sitprija, V., Moollaor, P. and Suwangool, P. (1989). Furosemide and dopamine in malarial acute renal failure. Nephron 52, 4 W . Luse, S. A. and Miller, L. H. (1971). Plasmodiumfalciparum malaria. Ultrastructure of parasitized erythrocytes in cardiac vessels. American Journal of Tropical Medicine and Hygiene 20, 65-60. Luzzi, G., Merry, A. N., Newbold, C. I., Marsh, K., Pasvol, G. and Weatherall, D. J. (1991). Surface antigen expression on Plasmodium fakiparum infected erythrocytes is modified in y and thalassemia. Journal of Experimental Medicine 73, 785-792. Mabey, D. C. W., Brown, A. and Greenwood, B. M. (1987). Plasmodium falciparum malaria and Salmonella infections in Gambian children. Journal of Infectious Diseases 155, 1319-1321. MacPherson, G. G., Warrell, M. J., White, N. J., Looareesuwan, S. and Warrell, D. A. (1985). Human cerebral malaria: a quantitative ultrastructural analysis of parasitized erythrocyte sequestration. American Journal of Pathology 119, 385401. Maegraith, B. G. (1948). “Pathological Processes in Malaria and Blackwater Fever”, pp. 348-349. Blackwell Scientific Publications, Oxford. Maegraith, B. G. (1952). Recent advances in tropical medicine: blackwater fever. West African Medical Journal 1, 4-10 and 203. Maegraith, B. G. and Findlay, G. M. (1944). Oliguria in blackwater fever. Lancet ii, 403-404. Maegraith, B. G. and Fletcher, A. (1972). The pathogenesis of mammalian malaria. Advances in Parasitology 10, 49-75. Magowan, C., Wollish, W., Anderson, L. and Leech, J. (1988). Cytoadherence by Plasmodiumfalciparum-infected erythrocytes is correlated with the expression of a family of variable proteins on infected erythrocytes. Journal of Experimental Medicine 168, 1307-1320.



Maguire, P. A., Prudhomme, J. and Sherman, I. (1991). Alterations in erythrocyte membrane phospholipid organisation due to intracellular growth of the human malaria parasite, Plasmodium falciparum. Parasitology 102, 179-1 86. Malhotra, K., Salmon, D., Le Bras, J. and Vilde, J. L. (1988). Susceptibility of Plasmodium falciparum to a peroxidase-mediated oxygen-dependent microbicidal system. Infection and Immunology 56, 3305-3309. Malloy, J. P., Brooks, M. H., Barry, K. G., Wilt, S. and McNeil, J. S. (1967). Pathophysiology of acute falciparum malaria. 11. Fluid compartmentalization. American Journal of Medicine 43, 745-750. Marchiafava, E. and Bignami, A. (1894). “On Summer-Autumnal Fever”. The New Sydenham Society, London. Marsh, K., Marsh, V. M., Brown, J., Whittle, H. C. and Greenwood, B. M. (1988). Plasmodium falciparum: the behavior of clinical isolates in an in vitro model of infected red blood cell sequestration. Experimental Parasitology 65, 202-208. Martell, R. W., Kallenbach, J. and Zwi, S. (1979). Pulmonary oedema in falciparum malaria. British Medical Journal i, 1763-1 764. Mbelepe, R., Beaucaire, G. and Dobscker, Y. (1986). Hypoglydmies skveres iteratives au cours de l’acces pernicieux palustre a Plasmodium falciparum. Semaine des H6pitaux de Paris 62, 1021-1023. McGregor, I. and Wilson, R. J. M. (1988). Specific immunity: acquired in man. In “Malaria” (W. H. Wernsdorfer and I. McGregor, eds), pp. 559-620. Churchill Livingstone, Edinburgh. McGregor, I. A., Gilles, H. M., Walters, J. H., Davies, A. H. and Pearson, F. A. (1956). Effects of heavy and repeated malaria infections on Gambian infants and children. British Medical Journal ii, 686-692. McGregor, I. A., Wilson, M. E. and Billewicz, W. Z. (1983). Malaria infection of the placenta in The Gambia, West Africa: its incidence and relationship to stillbirth, birthweight and placental weight. Transactions of the Royal Society of Tropical Medicine and Hygiene 77, 232-244. Menon, R. (1972). Pregnancy and malaria. Medical Journal of Malaysia 27, 11-1 19. Merry, A. H., Looareesuwan, S., Phillips, R. E., Chanthavanich, P., Supanaranond, W., Warrell, D. A. and Weatherall, D. J. (1986). Evidence against immune haemolysis in falciparum malaria in Thailand. British Journal of Haematology 64, 187- 194. Metzger, B. E., Ravnikar, V., Vileisis, R. A. and Freinkel, N. (1982). “Accelerated starvation” and the skipped breakfast in late normal pregnancy. Lancet i, 588592. Migasena, P. and Areekul, S. (1987). Capillary permeability function in malaria. Annals of Tropical Medicine and Parasitology 81, 549-560. Migasena, P. and Maegraith, B. G. (1967). Pharmacological action of antimalarial drugs: action of chloroquine and hydrocortisone on blood-brain barrier in Plasmodium knowlesi malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 61, 6. Migasena, S. (1983). Hypoglycaemia in falciparum malaria. Annals of Tropical Medicine and Parasitology 77, 323-324. Miller, K. D., White, N. J., Lott, J. A., Roberts, J. M. and Greenwood, B. M. (1989). Biochemical evidence of muscle injury in African children with severe malaria. Journal of Infectious Diseases 159, 139-142. Miller, K. L., Silverman, P. H., Kullgren, B. and Mahlmann, L. J. (1989). Tumor necrosis factor alpha and the anaemia associated with murine malaria. Infection and Immunity 57, 1542-1 546.



Miller, L. H. (1988). Genetically determined human resistance factors. In “Malaria” (W. H. Wernsdorfer and I. McGregor, eds), pp. 487-500. Churchill Livingstone, Edinburgh. Miller, L. H., Makaranond. P., Sitprija, V., Suebsanguan, C. and Canfield, C. J. (1967). Hyponatraemia in malaria. Annals of Tropical Medicine and Parasitology 61, 265-279. Miller, L. H., Chongsuphajaisiddhi, T. and Kanakakorn, K. (1968). Comparative studies on the pathology and host physiology of malarias. V. Hypovolaemia in Plasmodium coatneyi malaria. Annals of Tropical Medicine and Parasitology 62, 218-232. Miller, L. H., Usami, S. and Chien, J. (1971). Alteration in the rheologic properties of Plasmodium knowlesi-infected red cells. A possible mechanism for capillary obstruction. Journal of Clinical Investigation 50, 1451-1455. Miller, L. H., Chien, S. and Usami, S. (1972). Decreased deformability of Plasmodium coatneyi-infected red cells and its possible relationship to cerebral malaria. American Journal of Tropical Medicine and Hygiene 21, 133-1 37. Mohanty, D., Marwaha, N., Ghosh, K., Sharma, S., Garewal, G., Shah, S., Devi, S. and Das, K. C. (1988). Functional and ultrastructural changes of platelets in malaria infection. Transactions of the Royal Society of Tropical Medicine and Hygiene 82, 369-375. Molyneux, M. E., Looareesuwan, S., Menzies, I. S., Grainger, S. L., Phillips, R. E., Wattanagoon, Y., Thompson, R. P. H. and Warrell, D. A. (1989a). Reduced hepatic blood flow and intestinal malabsorption in severe falciparum malaria. American Journal of Tropical Medicine and Hygiene 40,[email protected] Molyneux, M. E., Taylor, T. E., Wirima, J. J. and Borgstein, J. (1989b). Clinical features and prognostic indicators in paediatric cerebral malaria: a study of 131 comatose Malawian children. Quarterly Journal of Medicine 71, 441459. Murray, M. J., Murray, N. J., Murray, A. 9. and Murray, M. 9. (1975). Refeeding malaria and hyperferraemia. Lancet i, 653-654. Nash, G. B., OBrien, E., Gordon-Smith, E. C. and Dormandy, J. A. (1989). Abnormalities in the mechanical properties of red blood cells caused by Plasmodium falciparum. Blood 74, 855-861. Neva, F. A., Howard, W. A., Glew, R. H., Krotoski, W. A., Gam, A. A., Collins, W. E., Atkinson, J. P. and Frank, M. M. (1974). Relationship of serum complement levels to events of the malarial paroxysm. Journal of Clinical Investigation 53,451-460. Newsome, F. (1984). Increased phagocytosis of non-parasitised red cells in Plasmodium berghei malaria. Annals of Tropical Medicine and Parasitology 78, 323-325. Newton, C. R. J. C., Kirkham, F. J., Winstanley, P. A., Pasvol, G., Peshu, N., Warrell, D. A. and Marsh, K. (1991). Intracranial pressure in African children with cerebral malaria. Lancet 337, 573-576. Nielsen, H., Kharazmi, A. and Theander, T. G. (1986). Suppression of blood monocyte and neutrophil chemotaxis in acute human malaria. Parasite Immunology 8, 541-550. Nnalue, N. A. and Friedman, M. J. (1988). Evidence for a neutrophil-mediated protective response in malaria. Parasite Immunology 10, 47-58. Nosten, F., ter Kuile, F., Malankiri, L., Decludt, 9. and White, N. J. (1991). Malaria in pregnancy in an area of unstable endemicity. Transactions ofthe Royal Society of Tropical Medicine and Hygiene 85,424-429.



Ockenhouse, C. F. and Chulay, J. D. (1988). Plasmodium fakiparum sequestration: OKM5 antigen (CD36) mediates cytoadherence of parasitized erythrocytes to a myelomonocytic cell line. Journal of Infectious Diseases 157, 584-588. Ockenhouse, C. F., Schulman, S. and Shear, H. L. (1984). Induction of crisis forms in the human malaria parasite Plasmodium falciparum by y-interferon-activated, monocyte-derived macrophages. Journal of Immunology 133, 1601-1608. Ockenhouse, C. F., Magowan, C. and Chulay, J. D. (1989a). Activation of monocytes and platelets by monoclonal antibodies or malaria-infected erythrocytes binding to the CD36 surface receptor in-vitro. Journal of Clinical Investigation 84, 468475. Ockenhouse, C. F., Tandon, N. N., Magowan, C., Jamieson, G. A. and Chulay, J. D. (1989b). Identification of a platelet membrane glycoprotein as a falciparum malaria sequestration receptor. Science 243, 1469-1471. Ockenhouse, C. F., Klotz, F. W., Tandon, N. N. and Jamieson, G. A. (1991a). Sequestrin, a CD36 recognition protein on Plasmodium fakiparum malariainfected erythrocytes identified by anti-idiotype antibodies. Proceedings of the National Academy of Sciences of the USA 88, 3175-3179. Ockenhouse, C. F., Ho, M., Tandon, N. N., van-Seventer, G. A., Shaw, S.,White, N. J., Jamieson, G. A., Chulay, J. D. and Webster, H. K. (1991b). Molecular basis of sequestration in severe and uncomplicated Plasmodium falciparum malaria. Differential adhesion of infected erythrocytes to CD36 and ICAM-I. Journal of Infectious Diseases 164, 163-169. Ogunye, 0. and Gbadebo, A. 0. (1981). Syndrome of inappropriate antidiuretic hormone (SIADH) in measles and malaria infections. Tropical and Geographical Medicine 33, 165-168. Okitolonda, W., Pottier, A. M. and Henquin, J. C. (1986). Glucose homeostasis in rats treated acutely and chronically with quinine. European Journal of Pharmacology 132, 179- 185. Okitolonda, W., Delacollette, C., Malengreau, M. and Henquin, J. C. (1987). High incidence of hypoglycaemia in African patients treated with intravenous quinine for severe malaria. British Medical Journal 295, 7 16-7 18. Olsson, R. A. (1967). The comparative pathophysiology of malarias. Annals of Tropical Medicine and Parasitology 61, 515-5 18. Olsson, R. A. and Johnston, E. H. (1969). Histopathologic changes and small bowel absorption in falciparum malaria. American Journal of Tropical Medicine and Hygiene 18, 355-359. Omanga, U., Ntihinyurwa, M., Shako, D. and Mashako, M. (1983). Les hemiplegies au cours de I'acces pernicieux a Plasmodium falciparum de I'enfant. Annales de PPdiatrie 30, 294-296. Onabanjo, A. 0. and Maegraith, B. G. (1970a). Kallikrein as a pathogenic agent in Plasmodium knowlesi infection in Macaca mulatta. British Journal of Experimental Pathology 51, 523-533. Onabanjo, A. 0. and Maegraith, B. G. (1970b). Pathological lesions produced in the brain by kallikrein (kininogenase) in Macaca mulatta infected with Plasmodium knowlesi. Annals of Tropical Medicine and Parasitology 64, 237-247. Onongbu, I. C. and Onyeneke, E. C. (1983). Plasma lipid changes in human malaria. Tropenmedizin und Parasitologie 34, 193-196. 00,M. M. and Than, T. (1989). Pathogenesis of ring haemorrhage in cerebral malaria. Annals of Tropical Medicine and Parasitology 83, 555-556.



00, M. M., Aikawa, M. and Than, T. (1987). Human cerebral malaria: a pathological study. Journal of Neuropathology and Experimental Neurology 46, 223-23 1. Oppenheimer, S. J., Gibson, F. D., Macfarland, S. B., Moody, J. B., Harrison, C., Spencer, A. and Bunari, 0. (1986). Iron supplementation increases prevalence and effects of malaria: report on clinical studies in Papua New Guinea. Transactions of the Royal Society of Tropical Medicine and Hygiene 80, 603-612. Oquendo, P., Hundt, E., Lawler, J. and Seed, B. (1989). CD36 directly mediates cytoadherence of Plasmodium falciparum parasitized erythrocytes. Cell 58, 95-

101. Pan, X.J., Mikkelsen, R. B., Wallach, D. F. H. and Asher, C. R. (1987). Synthesis of a somatostatin-like peptide by Plasmodiumfalciparum. Molecular and Biochemical Parasitology 25, 107-1 11. Panton, L. J., Leech, J. H., Miller, L. H. and Howard, R. J. (1987). Cytoadherence of PIasmodiumfalciparum-infected erythrocytes to human melanoma cell lines correlates with surface OKM5 antigen. Infection and Immunity 55, 2754-2758. Parillo, J. E., Parker, M. M., Natanson, C., Suffredini, A. F., Danner, R. L., Cunnion, R. E. and Ognibene, F. P. (1990). Septic shock in humans. Advances in the understanding of pathogenesis, cardiovascular dysfunction and therapy. Annals of Internal Medicine 113, 227-242. Pasvol, G., Weatherall, D. J., Wilson, R. J. M., Smith, D. H. and Gilles, H. M. (1976). Fetal haemoglobin and malaria. Lancet i, 1269-1272. Pasvol, G., Weatherall, D. J. and Wilson, R. J. M. (1977). Effects of fetal hemoglobin on susceptibility bf red cells to Plasmodium falciparum. Nature 270, 171-173. Pasvol, G., Weatherall, D. J. and Wilson, R. J. M. (1980). The increased susceptibility of young red cells to invasion by the malaria parasite Plasmodium falciparum. British Journal of Haematology 45, 285-296. Perrin, L. H., Mackey, L. J. and Miecher, P. A. (1982). The hematology of malaria in man. Seminars in Hematology 19, 70-82. Petchclai, B., Chutanondh, R., Hiranras, S. and Benjapongs, W. (I 977). Activation of classical and alternate complement pathways in acute falciparum malaria. Journal of the Medical Association of Thailand 60, 174176. Petersen, C. M., Nykjaer, A., Christiansen, B. S., Heickendorff, L., Mogensen, S. C. and Moller, B. (1989). Bioactive human recombinant tumor necrosis factor: an unstable dimer? European Journal of Immunology 19, 1887-1894. Petithory, J. -C., Lebeau, G., Galeazzi, G. and Chauty, A. (1983). L'hypocaldmie palustre. Etudes des correlations avec d'autres paramktres. Bulletin de la Socikt; de Pathologie Exotique 76, 455462. Pfaller, M. A., Parquette, A. R., Krogstad, D. J. and Nguyen-Dinh, P. (1982). Plasmodiumfakiparum: stage-specific lactate production in synchronized cultures. Experimental Parasitology 54, 39 1-396. Phanuphak, P., Hanvanich, M., Sakultamrung, R., Moollaor, P., Sitprija, V. and Phanthumkosol, D. (1985). Complement changes in falciparum malaria infection. Clinical and Experimental Immunology 59, 571-576. Philippe, E. and Walter, P. (1985). Les lisions placentaires du paludisme. Archives Francaise de Pkdiatrie 42, suppliment, 921-923. Phillips, R.E., Looareesuwan, S., Warrell, D. A., Lee, S. H., Karbwang, J., Warrell, M. J., White, N. J., Swadichai, C. and Weatherall, D. J. (1986a). The importance of anaemia in cerebral and uncomplicated falciparum malaria: role of complications, dyserythropoiesis and iron sequestration. QuarterIy Journal of Medicine 58, 305-323.



Phillips, R. E., Looareesuwan, S., White, N. J., Chanthavanich, P., Karbwang, J., Supanaranond, W., Turner, R. C. and Warrell, D. A. (1986b). Hypoglycaemia and antimalarial drugs: quinidine and release of insulin. British Medical Journal 292, 1319-1321.

Phillips, R. E., Bennish, M., Rahman, O., Turner, R. C., Williamson, D. H., Bloom, S. L. and Warrell, D. A. (1988). Pathophysiology of life-threatening hypoglycaemia in children with enteric infection in Dhaka, Bangladesh. Quarterly Journal of Medicine 69, 844-845. Piessens, W. F., Hoffman, S. L., Wadee, A. A., Piessens, P. W., Ratiwayanto, S., Kurhiawan, L., Campbell, J. R., Marwoto, H. A. and Laughlin, L. L. (1985). Antibody mediated killing of T suppressor lymphocytes as a possible cause of macroglobulinemia in the tropical splenomegaly syndrome. Journal of Clinical Investigation 75, 1821-1827. Playfair, J. H. L., Taverne, J., Bate, C. A. W. and de Souza, J. B. (1990). The malaria vaccine: anti-parasite or anti-disease. Immunology Today 11, 25-27. Polder, T., Jerusalem, C. and Eling, W. (1983). Topographical distribution of the cerebral lesions in mice infected with Plasmodium berghei. Tropenmedizin und Parasitologie 34, 235-243. Pongponratn, E., Riganti, M., Harinasuta, T. and Bunnag, D. (1989). Electron microscopic study of phagocytosis in human spleen in falciparum malaria. SouthEast Asian Journal of Tropical Medicine and Public Health 20, 31-39. Pongponratn, E., Riganti, M., Punpoowong, B. and Aikawa, M. (1991). Microvascular sequestration of parasitized erythrocytes in human falciparum malaria -a pathological study. American Journal of Tropical Medicine and Hygiene 44, 168-175.

Ponnudurai, T., Lensen, A. H. W., van-Gemert, G. J. A., Bolmer, M. G. and Meuwissen, J. H. E. (1991). Feeding behaviour and sporozoite ejection by infected Anopheles stephensi. Transactions of the Royal Society of Tropical Medicine and Hygiene 85, 175-1 80. Pukrittayakamee, S., White, N. J., Clemens, R., Chittamas, S., Karges, H. E., Desakorn, V., Looareesuwan, S. and Bunnag, D. (1989). Activation of the coagulation cascade in falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 83, 762-766. Pukrittayakamee, S., Davis, T. M. E., Levy, J., Looareesuwan, S., Roche, R. J., Jeerawathanapun, P. and White, N. J. (1991). The metabolic response to rapid intravenous glucose injection in acute falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 85, 189-193. Pukrittayakamee, S., White, N. J., Davis, T. M. E., Looareesuwan, S., Supanaranond, W., Desakorn, V., Chaivisuth, B. and Williamson, D. H. (in press). Hepatic blood flow and metabolism in severe malaria: the clearance of intravenously administered galactose. Clinical Science. Punyagupta, S., Srichaikul, T., Nitiyanant, P. and Petchclai, B. (1974). Acute pulmonary insufficiency in falciparum malaria: summary of 12 cases with evidence of disseminated intravascular coagulation. American Journal of Tropical Medicine and Hygiene 23, 551-559. Raventos-Suarez, C., Kaul, D. K.,Macaluso, F. and Nagel, R. L. (1985). Membrane knobs are required for the microcirculatory obstruction induced by Plasmodium falciparum-infected erythrocytes. Proceedings of the National Academy of Sciences of the USA 82, 3829-3833. Reid, H. A. and Nkrumah, F. K. (1972). Fibrin-degradation products in cerebral malaria. Lancet i, 218-221.



Riganti, M., Pongponratn, E., Tegoshi, T., Looareesuwan, S., Punpoowong, B. and Aikawa, M. (1990). Human cerebral malaria in Thailand: a clinical-pathological correlation. Immunology Letters 25, 199-206. Rigdon, R. H. (1944). Pathological lesions in the brain in malaria. Southern Medical Journal 37, 687494. Rigdon, R. H. and Fletcher, D. E. (1945). Lesions in the brain associated with malaria. Pathological studies on man and on experimental animals. Archives of Neurology and Psychiatry 53, 19 1-1 98. Riley, E. M., Anderson, L., Otoo, N., Jepsen, S. and Greenwood, B. M. (1988). Cellular immune responses to Plasmodium falciparum antigens in Gambian children during and after an acute attack of falciparum malaria. Clinical and Experimental Immunology 73, 17-22. Riley, E. M., MacLennan, C., Kiatkowski, D. K. and Greenwood, B. M. (1989a). Suppression of in-vitro lymphoproliferative responses in acute malaria patients can be partially reversed by indomethacin. Parasite Immunology 11, 509-5 17. Riley, E. M., Schneider, G., Sambou, I. and Greenwood, B. M. (1989b). Suppression of cell-mediated immune responses to malaria antigens in pregnant Gambian women. American Journal of Tropical Medicine and Hygiene 40, 131-144. Riley, E. M., Jakobsen, P. H., Allen, S. J., Wheeler, J. G., Bennett, S., Jepsen, S. and Greenwood, B. M. (1991). Immune responses to soluble exoantigens of Plasmodium falciparum may contribute to both pathogenesis and protection in clinical malaria: evidence from a longitudinal, prospective study of semi-immune African children. European Journal of Immunology 21, 1019-1025. Roberts, D. D., Sherwood, J. A., Spitalnik, S. L., Panton, L. J., Howard, R. J., Dixit, V. M., Frazier, W. A,, Miller, L. H. and Ginsburg, V. (1985). Thrombospondin binds falciparum malaria parasitised erythrocytes and may mediate cytoadherence. Nature 318, 64-66. Rock, E. P., Roth, E. F., Rojas-Corona, R. R., Sherwood, J. A., Nagel, R. L., Howard, R. J. and Kaul, D. K. (1988). Thrombospondin mediates the cytoadherence of Plasmodium falciparum-infected red cells to vascular endothelium in shear flow conditions. Blood 71, 71-75. Rockett, K. A., Targett, G. A. T. and Playfair, J. H. L. (1988). Killing of blood-stage Plasmodium falciparum by lipid peroxides from tumor necrosis serum. Infection and Immunity 56, 3 18&3 183. Rosenberg, E. B., Strickland, G. T., Yang, S.-L. and Whalen, G. E. (1973). IgM antibodies to red cells and autoimmune anemia in patients with malaria. American Journal of Tropical Medicine and Hygiene 22, 1 4 6 152. Rosenberg, R. and Wirtz, R. A. (1990). An estimation of the number of malaria sporozoites ejected by a feeding mosquito. Transactions of the Royal Society of Tropical Medicine and Hygiene 84, 209-2 12. Sadun, E. H., Williams, J. S., Meroney, F. C. and Hutt, G. (1965). Pathophysiology of Plasmodium berghei infections in mice. Experimental Parasitology 17,271-304. Sander, B. J., Lowery, M. S. and Kruckenberg, W. C. (1982). Plasmodium berghei acid insensitive phosphofructokinase in infected mouse erythrocytes. Experimental Parasitology 53, 11-16. Sanohko, A., Dareys, J. P. and Charreau, M. (1968). Etat endphalitique prolonge et acces pernicieux palustres. Bulletin de la SociPtP MPdicale d’Afrique Noire Langue Franpise 13, 662469. Schmid, A. H. (1974). Cerebral malaria. On the nature and significance of vascular changes. European Neurology 12, 197-208.



Schmidt, J. A., Udeinya, I. J., Leech, J. H., Hay, R. J., Aikawa, M., Barnwell, J. W., Green, I. and Miller, H. (1982). Plasmodium falciparum malaria. An amelanotic melanoma cell line bears receptors for the knob ligand on infected cells. Journal of Clinical Investigation 70, 379-386. Schmidt, L. H. (1978). Plasmodium falciparum and Plasmodium vivax infections in the owl monkey (Aotus trivirgatus). I. The course of untreated infections. American Journal of Tropical Medicine and Hygiene 27, 67 1-702. Schwartz, R. W., Olsen, J. A., Raventos-Suarez, F., Ye,M., Heath, R. H., Lubin, B. and Nagel, R. L. (1987). Altered plasma membrane phospholipid organisation in Plasmodium falciparum infected erythrocytes. Blood 69, 401407. Segal, H. E., Hall, A. P., Jewell, J. S., Pearlman, E. J., Na-Nakorn, A. and Mettaprakong (1974). Gastrointestinal function, quinine absorption and parasite response in falciparum malaria. South-East Asian Journal of Tropical Medicine and Public Health 5, 499-503. Shaffer, N., Grau, G. E., Hedberg, K., Davachi, F., Lyamba, B., Hightower, A. W., Breman, J. G. and Nguyen-Dinh, P. (1991). Tumor necrosis factor and severe malaria. Journal of Infectious Diseases 163, 96101. Sheagren, J. N., Shulman, N. P. and Canfield, C. J. (1970). Malaria: host-defence mechanisms and complications. Annals of Internal Medicine 73, 295-306. Sherman, I. W. (1979). Biochemistry of Plasmodium (malaria parasites). Microbiological Reviews 43, 453495. Sherwood, J. A., Roberts, D. D., Marsh, K., Harvey, E. B., Spitalnik, S. L., Miller, L. H. and Howard, R. J. (1987). Thrombospondin binding by parasitized erythrocyte isolates in falciparum malaria. American Journal of Tropical Medicine and Hygiene 26,228-233. Shute, P. G. and Maryon, M. (1954). A contribution to the problem of strains of human plasmodium. Rivista di Malariologia 33, 1-21. Singh, B., Ho, M., Mathai, E., Looareesuwan, S., Warrell, D. A. and Hommel, M. ( I 988). Plasmodium falciparum: inhibition/reversal of cytoadherence of Thai isolates to melanoma cells by local immune sera. Clinical and Experimental Immunology 72, 145-150. Sitprija, V. (1988). Nephropathy in falciparum malaria. (Clinical conference.) Kidney International 34, 867-877. Sitprija, V., Indraprasit, S., Pochanugool, C., Benyajati, C. and Piyaratn, P. (1967). Renal failure in malaria. Lancet i, 185-188. Sitprija, V., Vongsthongsri, M., Poshyachinda, V. and Arthachinta, S. (1977). Renal failure in malaria: a pathophysiologic study. Nephron 18, 277-287. Skirrow, M. R., Chongsuphajaisiddhi, T. and Maegraith, B. G. (1964). The circulation in malaria. 11. Portal angiography in monkeys (Macaca mulatta) with Plasmodium knowlesi and in shock following manipulation of the gut. Annals of Tropical Medicine and Parasitology 58, 502-5 10. Skudowitz, R. B., Katz, J., Lurie, A., Levin, J. and Metz, J. (1973). Mechanisms of thrombocytopenia in malignant tertian malaria. British Medical Journal ii, 515518.

Spellacy, W. M. and Goetz, F. C. (1963). Plasma insulin in late normal pregnancy. New England Journal of Medicine 268,988-991. Spitz, S. (1946). Pathology of acute falciparum malaria. Military Medicine 99, 555572. Sprague, H. B. (1946). The effects of malaria on the heart. American Heart Journal 31,426-430.



Srichaikul, T. (1959). A study of pigmentation and other changes in the liver in malaria. American Journal of Tropical Medicine and Hygiene 8, 11&118. Srichaikul, T., Panikbutr, N. and Jeumtrokul, P. (1967). Bone marrow changes in human malaria. Annals of Tropical Medicine and Parasitology 61, @51. Srichaikul, T., Pulket, C., Sirisatepisarn, T. and Prayoonwiwat, W. (1988). Platelet dysfunction in malaria. South-East Asian Journal of Tropical Medicine and Public Health 19, 225-233. Stein, C. M. (1987). Leucocytosis in acute Plasmodium falciparum malaria. South African Medical Journal 12, 363. Stone, W. J., Hanchett, J. E. and Knepshield, J. H. (1972). Acute renal insufficiency due to falciparum malaria. Archives of Internal Medicine 129, 62M28. Sucharit, P., Chongsuphajaisiddhi, T., Harinasuta, T., Tongprasroeth, N. and Kasemsuth, R. (1975). Studies on coagulation and fibrinolysis in cases of falciparum malaria. South-East Asian Journal of Tropical Medicine and Public Health 6, 33-39. Sullivan, J. (1876). Pernicious fever: febris algida and febris comatosa. Medical Times and Gazette 1, 277-279. Supanaranond, W., Davies, T. M. E., Dawes, J., Silamut, K., Vilaiwanna N. and White, N. J. (in press). In-vivo platelet function in severe falciparum malaria. Platelets. Taliaferro, W. H. and Mulligan, M. W. (1937). The histopathology of malaria with special reference to the function and origin of the macrophages in defence. Indian Medical Research Memoirs 29, 1-1 38. Tandon, N. N., Kralsz, U. and Jamieson, G. A. (1989). Identification of glycoprotein IV (CD36) as a primary receptor for platelet-collagen adhesion. Journal of Biological Chemistry 264, 75767583. Taverne, J., Tavernier, J., Fiers, W. and Playfair, J. H. L. (1987). Recombinant tumour necrosis factor inhibits malaria parasites in vivo but not in vitro. Clinical and Experimental Immunology 61, 1 4 . Taverne, J., Bate, C. A. W., Kwiatkowski, D., Jakobsen, P. H. and Playfair, J. H. L. (1990a). Two soluble antigens of Plasmodium falciparum induce tumor necrosis factor release from macrophages. Infection and Immunity 58, 2923-2928. Taverne, J., Bate, C. A. W., Sarkar, D. A., Meager, A., Rook, G. A. W. and Playfair, J. H. L. (l990b). Human and murine macrophages produce TNF in response to soluble antigens of Plasmodium falciparum. Parasite Immunology 12, 3343. Taylor, T. E., Molyneux, M. E., Wirima, J. J., Fletcher, K. A. and Morris, K. (1988). Blood glucose levels in Malawian children before and during the administration of intravenous quinine in severe falciparum malaria. New England Journal of Medicine 319, 104O-lO47. Tharavanij, S., Warrell, M. J., Tantivanich, S., Tapchaisri, P., Chongsa-Nguan, M., Prasertsiriroj, V. and Patarapotikul, J. (1984). Factors contributing to the development of cerebral malaria. I . Humoral immune responses. American Journal of Tropical Medicine and Hygiene 33, 1-1 1. Theander, T. G., Bygbjerg, I. C., Andersen, B. J., Jepsen, S., Kharazmi, A. and Odum, N. (1986). Suppression of parasite specific response in Pfasmodium falciparum malaria. A longitudinal study of blood mononuclear cell proliferation and subset composition. Scandinavian Journal of Immunology 24, 73-8 I . Thomas, J. D. (1971). Clinical and histopathological correlation of cerebral malaria. Tropical and Geographical Medicine 23, 232-238.



Thumwood, C. M., Hunt, N. H., Clark, I. A. and Cowden, W. B. (1988). Breakdown of the blood-brain barrier in murine cerebral malaria. Parasitology 96, 579-589. Thurnham, D. I., Oppenheimer, S. J. and Bull, R. (1983). Riboflavin status and malaria in infants in Papua New Guinea. Transactions of the Royal Society of Tropical Medicine and Hygiene 77, 423424. Toro, G. and Roman, G. (1978). Cerebral malaria. A disseminated vasculomyelinopathy. Archives of Neurology 35, 271-275. Tubbs, H. (1980). Endotoxin in human and murine malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 74, 121-123. Udeinya, I. J. and Miller, L. H. (1987). Plasmodium falciparum: effect of infected erythrocytes on clotting time of plasma. American Journal of Tropical Medicine and Hygiene 31, 246-249. Udeinya, I. J., Schmidt, J. A., Aikawa, M., Miller, L. H. and Green, I. (1981). Falciparum malaria-infected erythrocytes bind to cultured human endothelial cells. Science 213, 555-557. Udeinya, I. J., Miller, L. H., McGregor, I. A. and Jensen, J. B. (1983). Plasmodium falciparum strain-specific antibody blocks binding of infected erythrocytes to amelanotic melanoma cells. Nature 303, 429-43 1. Udomsangpetch, R., Aikawa, M., Berzins, K., Wahlgren, M. and Perlmann, P. (1 989a). Cytoadherence of knobless Plasmodium falciparum-infected erythrocytes and its inhibition by a human monoclonal antibody. Nature 338, 763-765. Udomsangpetch, R., Wahlin, B., Carlson, J., Berzins, K., Torii, M., Aikawa, M., Perlmann, P. and Wahlgren, M. (1989b). Plasmodium falciparum-infected erythrocytes form spontaneous erythrocyte rosettes. Journal of Experimental Medicine 169, 1835-1840. Usawattanakul, W., Tharavanij, S., Warrell, D. A., Looareesuwan, S., White, N. J., Supavej, S.and Soikratoke, S. (1985). Factors contributing to the development of cerebral malaria. 11. Endotoxin. Clinical and Experimental Immunology 61, 562568. Van der Jagt, D. L., Hunsaker, L. A., Campos, N. M. and Baack, B. R. (1990). D-lactate production in erythrocytes infected with Plasmodium falciparum. Molecular and Biochemical Parasitology 42, 277-284. Van Zon, A. A., Eling, W. M., Hermsen, C. C. and Koekkoek, A. A. (1982). Corticosterone regulation of the effector function of malarial immunity during pregnancy. Infection and Immunity 36, 484-491. Van Zon, A. A., Eling, W. M. and Hermsen, C. C. (1985). Pregnancy-induced recrudescences strengthen malarial immunity in mice infected with Plasmodium berghei. Parasitology 91, 9-1 7. Vleugels, M. P. H., Eling, W. M. C., Rolland, R. and De Graaf, R. (1987). Cortisol and loss of malaria immunity in human pregnancy. British Journal of Obstetrics and Gynaecology 94,758-764. Wahlgren, M., Carlson, J., Ruangjirachuporn, W., Conway, D., Helmby, H., Martinez, A., Patarroya, M. E. and Riley, E. (1990). Geographical distribution of Plasmodium falciparum erythrocyte rosetting and frequency of rosetting antibodies in human sera. American Journal of Tropical Medicine and Hygiene 43, 3 33-3 38. Waller, D., Crawley, J., Nosten, F., Chapman, D., Krishna, S., Craddock, C., Brewster, D. and White, N. J. (1991). Intracranial pressure in childhood cerebral malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 85, 362-364.



Walter, P., Gavin, J. F. and Blot, P. (1982). Placental pathologic changes in malaria. American Journal of Pathology 109, 33S342. Ward, K. N., Warrell, M. J., Rhodes, J., Looareesuwan, S. and White, N. J. (1984). Altered expression of human monocyte Fc receptor in Plasmodium falciparum malaria. Infection and Immunity 44, 623-626. Warrell, D. A., Looareesuwan, S., Kasemsarn, B., Intaraprasert R., Warrell, M. J., Bunnag, D. and Harinasuta, T. (1982). Dexamethasone proves deleterious in cerebral malaria. A double blind placebo controlled study in 100 patients. New England Journal of Medicine 306, 3 I 3-3 19. Warrell, D. A., Looareesuwan, S., Phillips, R. E., White, N. J., Warrell, M. J., Chapel, H. M., Areekul, S. and Tharavanij, S. ‘(1986). Function of the bloodcerebrospinal fluid barrier in human cerebral malaria: rejection of the permeability hypothesis. American Journal of Tropical Medicine and Hygiene 35, 882889. Warrell, D. A., White, N. J., Veall, N., Looareesuwan, S., Chanthavanich, P., Phillips, R. E., Karbwang, J., Krishna, S. and Pongpaew, P. (1988). Cerebral anaerobic glycolysis and reduced cerebral oxygen transport in human cerebral malaria. Lancet ii, 534-538. Waters, A. P., Higgins, D. G. and McCutchan, T. F. (1991). Plasmodium falciparum appears to have arisen as a result of lateral transfer between avian and human hosts. Proceedings of the National Academy of Sciences of the USA 88,3140-3144. Waters, L. S., Taverne, J., Tai, P., Spry, C. J., Targett, G. A. and Playfair, J. H. L. (1987). Killing of Plasmodium falciparum by eosinophil secretory products. Infection and Immunity 55, 877-88 1. Weatherall, D. J. and Abdallah, S. (1982). The anaemia of Plasmodium falciparum malaria. British Medical Bulletin 38, 147-1 5 1. Webster, H. K., Boudreau, E. F., Pang, L. W., Permpanich, B., Sookto, P. and Wirtz, R. A. (1987). Development of immunity in natural Plasmodium falciparum malaria: antibodies to the falciparum sporozoite vaccine I antigen (R32tet32). Journal of Clinical Microbiology 35, 1002-1 008. Weiss, L. (1990). The spleen in malaria: the role of barrier cells. Immunology Letters 25, 165-172. Weiss, L., Geduldig, U. & Weidanz, W. (1986). Mechanisms of splenic control of murine malaria: reticular cell activation and the development of a blood-spleen barrier. American Journal of Anatomy 176, 251-285. Wernsdorfer, W. H. & McGregor, I. (1988). “Malaria. Principles and Practice of Malariology”. Churchill Livingstone, Edinburgh. White, N. J. (1985). Clinical pharmacokinetics of the antimalarial drugs. Clinical Pharmacokinetics 10, 187-21 5. White, N. J. (1986). Pathophysiology. In “Clinics in Tropical Medicine and Communicable Disease” (G. T. Strickland, ed.) Vol. 1, Malaria, pp. 55-90. Saunders, London. White, N. J. and Krishna, S. (1989). Treatment of malaria: some considerations and limitations of the current methods of assessment. Transactions of the Royal Society of Tropical Medicine and Hygiene 83, 767-717. White, N. J. and Looareesuwan, S. (1987). Cerebral malaria. In “Infections of the Nervous System” (P. G. E. Kennedy and R. T. Johnson, eds), pp. 118-143. Butterworths, London. White, N. J., Looareesuwan, S., Warrell, M. J., Bunnag, D. & Harinasuta, T. (1982). Quinine pharmacokinetics and toxicity in cerebral and uncomplicated falciparum malaria. American Journal of Medicine 73, 564-572.



White, N. J., Warrell, D. A,, Looareesuwan, S., Chanthavanich, P., Warrell, M. J., Krishna, S., Williamson, D. H. & Turner, R. C. (1983). Severe hypoglycaemia and hyperinsulinaemia in falciparum malaria. New England Journal of Medicine 309, 61-66. White, N. J., Warrell, D. A., Looareesuwan, S., Chanthavanich, S., Phillips, R. E. & Pongpaew, P. ( I 985). Pathophysiological and prognostic significance of cerebrospinal-fluid lactate in cerebral malaria. Lancet i, 776-778. White, N. J., Miller, K. D., Marsh, K., Berry, C. D., Turner, R. C., Williamson, D. H . and Brown, J. (1987a). Hypoglycaemia in African children with severe malaria. Lancet i, 708-71 1. White, N. J., Miller, K. D., Brown, J., Marsh, K. and Greenwood, B. M. (1987b). Prognostic value of CSF lactate in cerebral malaria. Lancet i, 1261. Whittle, H. C., Brown, J., Marsh, K., Greenwood, B. M., Seidelin, P., Tighe, H. and Wedderburn, L. (1984). T cell control of EB virus infected B cell is lost during P . falciparum malaria. Nature 312, 449450. Wickramasinghe, S. N., Phillips, R. E., Looareesuwan, S., Warrell, D. A. and Hughes, M. (1987). The bone marrow in human cerebral malaria: parasite sequestration within sinusoids. British Journal of Haematology 66, 295-306. Wickramasinghe, S. N., Looareesuwan, S., Nagachinta, B. and White, N. J. (1989). Dyserythropoiesis and ineffective erythropoiesis in Plasmodium vivax malaria. British Journal of Haematology 72, 91-99. Winograd, E. and Sherman, I. W. (1989). Characterization of a modified red cell membrane protein expressed on erythrocytes infected with the human malaria parasite Plasmodium falciparum: possible role as a cytoadherent mediating protein. Journal of Cell Biology 108, 23-30. World Health Organization (1986). Severe and complicated malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 80, supplement, 1-50. World Health Organization (1990). Severe and complicated malaria, 2nd edn. Transactions of the Royal Society of Tropical Medicine and Hygiene 84, supplement 2, 1 4 5 . Wyler, D. J., Quinn, T. C. and Chen, L. T. (1981). Relationship of alteration in splenic clearance function and microcirculation to host defence in acute rodent malaria. Journal of Clinical Investigation 67, 140&1404. Yayon, A., Vande Waa, J. A., Yayon, M., Geary, T. and Jensen, J. B. (1983). Stage dependent effects of chloroquine on Plasmodium falciparum in-vitro. American Journal of Tropical Medicine and Hygiene 30,642-647. Yoeli, M. (1976). Cerebral malaria-the quest for suitable experimental models in parasitic diseases of man. Transactions of the Royal Society of Tropical Medicine and Hygiene 70, 24-35. Zhang, Y., Asanke, K . S. and Jung, A. (1986). Stage dependent inhibition of chloroquine on Plasmodium falciparum in-vitro. Journal of Parasitology 72, 83CL 836. Zolg, J. W., Macleod, A. J., Scaife, J. G . and Beaudoin, R. L. (1984). The accumulation of lactic acid and its influence on the growth of Plasmodium falciparum in synchronized culture. In Vitro 20, 205-21 5 . Zuckerman, A. (1966). Recent studies on factors involved in malarial anemia. Military Medicine 131, supplement, 1201-1216.