1987). At present, there is no satisfactory animal model for human cerebral malaria.
The clinical management of patients with cerebral malaria requires prompt, specific antimalarial treatment, in addition to the general supportive care required for unconscious patients. Because of the extensive spread of chloroquine-resistant P. falciparum, the current drugs of choice for patients with cerebral malaria are intravenous quinine and, when quinine is unavailable, quinidine (Phillips et al., 1985; Miller et al., 1989; World Health Organization, 1990). In patients who have not undergone any previous treatment, a “loading dose” should be used with each drug to produce therapeutic blood levels as quickly as possible (White et al., 1983a). The size of the loading dose may vary, depending on local parasite sensitivities and on any preadmission drug treatment.
Serious side effects include quinine- or quinidine-induced hypoglycemia (White et al., 1983b; Phillips et al., 1986b) and quinidine-related changes in the electrical conduction of the heart. Both can be mitigated by slow, controlled infusions of the medications (Phillips et al., 1985; Miller et al., 1989; Molyneux et al., 1989b). Pregnant women are particularly likely to develop hypoglycemia during the course of treatment with intravenous quinine (Looareesuwan et al., 1985). When the patient is able to eat and drink, tablets or capsules can be substituted for the intravenous medication. There is no consensus regarding the optimal duration of treatment with these two drugs. In certain parts of the world (e.g., Thailand) where the malaria parasite is less sensitive to quinine, a second drug (tetracycline) is used to eliminate parasitemia following resolution of coma (World Health Organization, 1990).
The degree of supportive care provided to patients depends on the available facilities. Very high body temperatures can be reduced with tepid sponging and fanning. Convulsions can be treated with any of a number of standard medications. For example, a recent study showed that prophylactic administration of phenobarbitone decreased the number of subsequent convulsions in a group of adults with cerebral malaria (White et al., 1988). The number of patients studied was too small, however, to allow determination of whether decreasing the number of convulsions had any effect on overall survival or the incidence of sequelae.
Exchange transfusion (replacement of a patient's blood with donor blood over a short period of time) has been used effectively to treat patients with severe malaria, especially those with high blood levels of parasites (Kurathong et al., 1979; Roncoroni and Martino, 1979; Yarrish et al., 1982; Kramer et al., 1983; Files et al., 1984; Chiodini et al., 1985; Hall et al., 1985; Miller et al., 1989). The technique is not practical in malaria-endemic areas where the prevalence of human immunodeficiency virus (HIV) seropositivity is high or in settings where close clinical monitoring of patients is not fea-