Initially patients have fever, chills, sweating, headache, weakness, and other symptoms mimicking a “viral syndrome.” Later, severe disease may develop, with an abnormal level of consciousness, severe anemia, renal failure, and multisystem failure.
Plasmodia are protozoa. Only the species Plasmodium falciparum, P vivax, P malariae, and P ovale are usually infectious for humans. Of these, P falciparum is the most dangerous.
Structure and Life Cycle
In nature, uninucleate sporozoites in the salivary glands of infected mosquitoes are injected into a human host when the mosquito feeds. The sporozoites rapidly invade liver parenchymal cells, where they mature into liver-stage schizonts, which burst to release 2,000 to 40,000 uninucleate merozoites. In P vivax and P ovale infections, maturation of the schizont may be delayed for 1 to 2 years. Each merozoite can infect a red blood cell. Within the red cell, the merozoite matures either into a uninucleate gametocyte--the sexual stage, infectious for Anopheles mosquitoes--or, over 48 to 72 hours, into an erythrocyticstage schizont containing 10 to 36 merozoites. Rupture of the schizont releases these merozoites, which infect other red cells. If a vector mosquito ingests gametocytes, the gametocytes develop in the mosquito gut to gametes, which undergo fertilization and mature in 2 to 3 weeks to sporozoites.
The fever and chills of malaria are associated with the rupture of erythrocytic-stage schizonts. In severe falciparum malaria, parasitized red cells may obstruct capillaries and postcapillary venules, leading to local hypoxia and the release of toxic cellular products. Obstruction of the microcirculation in the brain (cerebral malaria) and in other vital organs is thought to be responsible for severe complications. Cytokines (e.g., tumor necrosis factor) are also felt to be involved, but at present their role is unclear.
Both innate and acquired immunity occur. Innate immunity consists of various traits of erythrocytes that discourage infection. The sickle-cell trait protects against the development of severe P falciparum malaria, and the absence of Duffy antigen prevents infection by P vivax. Recurrent infections lead to the development of humoral and cellular immune responses against all Plasmodium stages. Acquired immunity does not prevent reinfection but does reduce the severity of disease.
Malaria is distributed worldwide throughout the tropics and subtropics.
Diagnosis depends primarily on the identification of plasmodia in thick and thin blood smears.
Treatment: The widespread resistance of P falciparum to chloroquine complicates treatment of falciparum malaria. Alternative drugs such as mefloquine, pyrimethamine/ sulfadoxine (FansidarR), quinine, quinidine, halofantrine and artemisinin derivatives (qinghaosu) are used. Chloroquine remains highly effective against P malariae and P ovale malaria, and against P vivax everywhere except Papua New Guinea and parts of Indonesia, where significant resistance has developed. Disease caused by P vivax and P ovale requires primaquine to eradicate latent liver forms of the parasite.
Prevention: Malaria may be prevented by chemoprophylaxis and personal protective measures against the mosquito vector and by community-wide measures to control the vector. Exposure to night-feeding Anopheles mosquitoes is reduced by using protective clothing, insect repellents, insecticides, insecticide-impregnated bed nets, etc. Mosquitoes may be reduced by destroying breeding places and by application of insecticides. Vaccines are being developed.
Malaria has been a major disease of humankind for thousands of years. It is referred to in numerous biblical passages and in the writings of Hippocrates. Although drugs are available for treatment, malaria is still considered by many to be the most important infectious disease of humans: there are approximately 200 million to 500 million new cases each year in the world, and the disease is the direct cause of 1 million to 2.5 million deaths per year.
Malaria is caused by protozoa of the genus Plasmodium. Four species cause disease in humans: P falciparum, P vivax, P ovale and P malariae. Other species of plasmodia infect reptiles, birds and other mammals. Malaria is spread to humans by the bite of female mosquitoes of the genus Anopheles.
The most characteristic symptom of malaria is fever. Other common symptoms include chills, headache, myalgias, nausea, and vomiting. Diarrhea, abdominal pain, and cough are occasionally seen. As the disease progresses, some patients may develop the classic malaria paroxysm with bouts of illness alternating with symptom-free periods (Fig. 83-1). The malaria paroxysm comprises three successive stages. The first is a 15-to-60 minute cold stage characterized by shivering and a feeling of cold. Next comes the 2-to-6 hour hot stage, in which there is fever, sometimes reaching 41°C, flushed, dry skin, and often headache, nausea, and vomiting. Finally, there is the 2-to-4 hour sweating stage during which the fever drops rapidly and the patient sweats. In all types of malaria the periodic febrile response is caused by rupture of mature schizonts. In P vivax and P ovale malaria, a brood of schizonts matures every 48 hr, so the periodicity of fever is tertian (“tertian malaria”), whereas in P malariae disease, fever occurs every 72 hours (“quartan malaria”). The fever in falciparum malaria may occur every 48 hr, but is usually irregular, showing no distinct periodicity. These classic fever patterns are usually not seen early in the course of malaria, and therefore the absence of periodic, synchronized fevers does not rule out a diagnosis of malaria.
Typical temperature charts of malarial infections. (Adapted from Bruce-Chwatt LJ: Essential Malariology. 2nd Ed. John Wiley and Sons, New York, 1985, p. 52, with permission.)
Physical findings in malaria are nonspecific and offer little aid in diagnosis. In many cases there may be no positive findings other than fever. Splenomegaly is common but may not be apparent early in disease. Hepatomegaly, jaundice, hypotension and abdominal tenderness may also be seen. Malaria does not cause lymphadenopathy and is not associated with a rash.
A variety of laboratory abnormalities may be seen in a case of uncomplicated malaria. These include normochromic, normocytic anemia, thrombocytopenia, leukocytosis or leukopenia, hypoglycemia, hyponatremia, elevated liver and renal function tests, proteinuria, and laboratory evidence of disseminated intravascular coagulation (although clinically important bleeding is rare). Eosinophilia is not seen. Patients with complicated malaria occasionally show evidence of massive intravascular hemolysis with hemoglobinemia and hemoglobinuria.
If the diagnosis of malaria is missed or delayed, especially with P falciparum infection, potentially fatal complicated malaria may develop. The most frequent and serious complications of malaria are cerebral malaria and severe anemia. Cerebral malaria is defined as any abnormality of mental status in a person with malaria and has a case fatality rate of 15 to 50 percent. Other complications include: hyperparasitemia (more than 3 to 5 percent of the erythrocytes parasitized); severe hypoglycemia; lactic acidosis; prolonged hyperthermia; shock; pulmonary, cardiac, hepatic, or renal dysfunction; seizures; spontaneous bleeding; or high-output diarrhea or vomiting. These manifestations are associated with poor prognosis. Persons at increased risk of severe disease from malaria include older persons, children, pregnant women, nonimmune persons and those with underlying chronic illness. Other complications of malaria infection include gram-negative sepsis, aspiration pneumonia and splenic rupture.
Only four species of the protozoan genus Plasmodium usually infect humans: P falciparum, P vivax, P malariae, and P ovale (Fig. 83-2). P falciparum and P vivax account for the vast majority of cases. P falciparum causes the most severe disease.
Blood stages of Plasmodium. Column A, Plasmodium vivax; B, P ovale; C, P malariae; D, P falciparum. Row 1, young trophozoites (ring forms); 2, growing trophozoites; 3, mature trophozoites; 4, mature schizonts; 5, macrogametocytes; 6, microgametocytes. (more...)
Structure and Life Cycle
Like many protozoa, plasmodia pass through a number of stages in the course of their two-host life cycle. The stage infective for humans is the uninucleate, lancet-shaped sporozoite (approximately 1 × 7 μm). Sporozoites are produced by sexual reproduction in the midgut of vector anopheline mosquitoes and migrate to the salivary gland. When an infected Anopheles mosquito bites a human, she may inject sporozoites along with saliva into small blood vessels (Fig. 83-3). Sporozoites are thought to enter liver parenchymal cells within 30 minutes of inoculation. In the liver cell, the parasite develops into a spherical, multinucleate liver-stage schizont which contains 2,000 to 40,000 uninucleate merozoites. This process of enormous amplification is called exoerythrocytic schizogony. This exoerythrocytic or liver phase of the disease usually takes between 5 and 21 days, depending on the species of plasmodium. However, in P vivax and P ovale infections, maturation of liver-stage schizonts may be delayed for as long as 1 to 2 years. These quiescent liver-phase parasites are called hypnozoites.
Life cycle of malaria parasite. (Adapted from Miller LH, Howard RJ, Carter R et al: Research toward malaria vaccines. Science 234:1350, 1986, with permission.)
Regardless of the time required for development, the mature schizonts eventually rupture, releasing thousands of uninucleate merozoites into the bloodstream. Each merozoite can infect a red blood cell. Within the red cell, the merozoite develops to form either an erythrocytic-stage (blood-stage) schizont (by the process of erythrocytic schizogony) or a spherical or banana-shaped, uninucleate gametocyte. The mature erythrocytic-stage schizont contains 8 to 36 merozoites, each 5 to 10 μm long, which are released into the blood when the schizont ruptures. These merozoites proceed to infect another generation of erythrocytes. The time required for erythrocytic schizogony-which determines the interval between the release of successive generations of merozoites-varies with the species of plasmodium and is responsible for the classic periodicity of fever in malaria (Fig 83-1).
The gametocyte, which is the sexual stage of the plasmodium, is infectious for mosquitoes that ingest it while feeding. Within the mosquito, gametocytes develop into female and male gametes (macrogametes and microgametes, respectively), which undergo fertilization and then develop over 2 to 3 weeks into sporozoites that can infect humans. The delay between infection of a mosquito and maturation of sporozoites means that female mosquitoes mustlive a minimum of 2 to 3 weeks to be able to transmit malaria. This fact is important in malaria control efforts.
Clinical illness is caused by the erythrocytic stage of the parasite. No disease is associated with sporozoites, the developing liver stage of the parasite, the merozoites released from the liver, or gametocytes.
The first symptoms and signs of malaria are associated with the rupture of erythrocytes when erythrocytic-stage schizonts mature. This release of parasite material presumably triggers a host immune response. The cytokines, reactive oxygen intermediates, and other cellular products released during the immune response play a prominent role in pathogenesis, and are probably responsible for the fever, chills, sweats, weakness, and other systemic symptoms associated with malaria. In the case of falciparum malaria (the form that causes most deaths), infected erythrocytes adhere to the endothelium of capillaries and postcapillary venules, leading to obstruction of the microcirculation and local tissue anoxia. In the brain this causes cerebral malaria (Fig. 83-4); in the kidneys it may cause acute tubular necrosis and renal failure; and in the intestines it can cause ischemia and ulceration, leading to gastrointestinal bleeding and to bacteremia secondary to the entry of intestinal bacteria into the systemic circulation. The severity of malaria-associated anemia tends to be related to the degree of parasitemia. The pathogenesis of this anemia appears to be multifactorial. Hemolysis or phagocytosis of parasitized erythrocytes and ineffective erythropoiesis are the most important factors, and phagocytosis of uninfected erythrocytes and an autoimmune hemolytic anemia have also been implicated. Massive intravascular hemolysis leading to hemoglobinuria and renal failure is referred to as blackwater fever. It was described more frequently in the past than currently. Hemolysis may also occur after the use of certain antimalarials (especially primaquine) in patients with glucose 6-phosphate dehydrogenase deficiency.
Light micrograph of a cerebral capillary blocked with parasitized erythrocytes. This specimen is from a patient with cerebral malaria. (From Aikawa M: Morphological changes in erythrocytes induced by malarial parasites. Biol Cell 64:174, 1988, with permission.) (more...)
Susceptibility to malaria infection and disease is regulated by hereditary and acquired factors (Fig 83-5). It now seems clear that the sickle cell trait (which is the cause of sickle-cell anemia) developed as a balanced polymorphism to protect against serious P falciparum disease. Although individuals with sickle cell anemia or the sickle cell trait are as easily infected with malaria parasites as normal individuals, they rarely exhibit malaria disease because P falciparum develops poorly in their erythrocytes. The virtual absence of P vivax infections in many areas of Africa is explained by the fact that most blacks do not have Duffy blood-group antigens, which apparently function as erythrocyte surface receptors for P vivax merozoites; without the Duffy antigen, the parasites cannot invade. Malaria parasites do not develop well in ovalocytes, and it has been suggested that ovalocytosis, which is quite common in some malarious areas, such as New Guinea, may reduce the incidence of malaria. Some investigators have suggested that glucose 6-phosphate dehydrogenase deficiency, as well as a number of other hemoglobinopathies (including the thalassemias and hemoglobin E), also protect against malaria infection, but the evidence for these associations is less compelling.
Host defense against malaria. (Adapted from Miller LH, Howard RJ, Carter R et al: Research toward malaria vaccines. Science 234:1350, 1986, with permission.)
Acquired immunity can also protect against malaria infection and the development of malaria disease. In malarious areas, both the prevalence and severity of malaria infections decrease with age. However, in contrast to many viral infections, multiple infections with malaria do not confer longlasting, sterile protective immunity. Virtually all adults in malarious areas suffer repeated malaria infections. Individuals who are repeatedly exposed to malaria develop antibodies against many sporozoite, liver-stage, blood-stage, and sexual-stage malaria antigens. It is thought that antibodies acting against sporozoites, liver-stage and blood-stage organisms are responsible for the decreased susceptibility to malaria infection and disease seen in adults in malarious areas, and that antibodies against the sexual stages of plasmodia may reduce malaria transmission. Additional work also suggests that the naturally acquired immunity includes the release of cytokines that act against all stages of the parasite, and also a cytotoxic T cell response directed at liver stages of the parasite.
Acquired antibody-mediated immunity is apparently transferred from mother to fetus across the placenta. This passively transferred immunity is lost within 6 to 9 months, as is the immunity in adults if they leave a malarious area and are no longer exposed to plasmodia. Pregnant women, particularly primigravidas, are more susceptible to malaria infections and serious disease.
Malaria is transmitted primarily by the bite of infected anopheline mosquitoes. It can also be transmitted by inoculation of infected blood and congenitally. Anophelines feed at night and their breeding sites are primarily in rural areas. The greatest risk of malaria is therefore from dusk to dawn in rural areas. In many malaria-endemic areas, there is little or no risk in urban areas. However, urban transmission is common in some parts of the world, especially Africa.
Malaria was once transmitted in many parts of the world, for example, as far north as North Dakota in the United States. Dueboth to environmental changes and to eradication campaigns conducted in the years after World War II, endemic malaria transmission has been eliminated from many areas, including the United States and Europe. The disease is still widely transmitted in the tropics and subtropics (Fig. 83-6). In these areas malaria transmission may be endemic, occurring predictably every year, or it may be epidemic, occurring sporadically when conditions are correct. Endemic transmission of malaria may be year-round or seasonal. In some areas of Africa, 90 to 100 percent of children less than 5 years old have malaria parasites circulating in their blood all the time. Because naturally acquired immunity develops with increasing exposure, in endemic areas malaria disease is primarily found in children. In epidemic areas, on the other hand, naturally acquired immunity falls off between epidemics, and malaria therefore affects all age groups during epidemics.
Distribution of malaria and chloroquine-resistant Plasmodium falciparum, 1993. (Centers for Disease Control and Prevention).
Approximately 1,000 cases of malaria are reported each year in the United States in returning travelers. Of the 1016 imported cases reported in 1991, the majority were acquired in Africa (466 cases) and India (221 cases). P vivax accounted for 43% of the cases and P. falciparum for 39%. The risk to travelers of acquiring P falciparum is greatest in Africa. This is because it is the most prevalent species there, malaria transmission is much more intense there than in other parts of the world, and there is significant risk in urban areas.
Anopheles mosquitoes capable of transmitting malaria are found in a number of areas of the United States. Local transmission may therefore occur when these mosquitoes feed upon malariainfected individuals, generally immigrants from malaria-endemic areas. Local transmission has recently occurred in southern California, New Jersey, New York City, and Houston, Texas. Malaria may also occur when infected mosquitoes are transported into non-endemic areas, such as by airplanes or ships.
In the late 1950s and early 1960s, it was thought that malaria could be eradicated through the widespread use of insecticides such as DDT and by treatment of cases with chloroquine. Eradication is no longer thought possible, however, because of the development of drug resistance by both the mosquito and the parasite, and because of deteriorating social and economic conditions in many malaria-endemic countries. These changes have resulted in a dramatic increase in the incidence of malaria in many parts of the world, and an increase in malaria-related mortality in some of these areas.
In the United States, many of the deaths from malaria are the result of delayed diagnosis and treatment because the health care provider did not suspect malaria. The diagnosis of malaria requires a high index of suspicion; malaria should be considered in any individual who has a fever and has visited an endemic area for malaria, received a blood transfusion, or used intravenous drugs. Although 95 percent of individuals infected with malaria develop their primary illness within 6 weeks of exposure, somemay have primary attacks up to a year after exposure, and relapses of malaria can occur up to 2–3 years after exposure. Therefore, individuals having a febrile illness and a history of exposure in the last 2–3 years should be evaluated for malaria.
Definitive diagnosis of malaria generally requires direct observation of malaria parasites in Giemsa-stained thick and thin blood smears (Fig. 83-2). Thick blood smears are more difficult to interpret than thin blood smears but they are much more sensitive, as more blood is examined. Thin blood smears, in which parasites are seen within erythrocytes, are used to determine the species of the infecting parasite. The presence of diagnostic forms can vary markedly with the stage of the life cycle, especially early in disease. In falciparum malaria, most organisms are not present in the peripheral blood because they are sequestered in the microvascular tissue of internal organs. If malaria is suspected, blood smears should be examined every 6 to 12 hr for at least 2 days. New diagnostic methods include a rapid antigen-capture dipstick test and a technique for detecting parasites with a fluorescent stain. Both of these tests are fast, easy to perform and are highly sensitive and specific
Other diagnostic methods include assays to detect malaria antibodies and antigens, and polymerase chain reaction/DNA and RNA probe techniques. These techniques are used primarily in epidemiologic studies and immunization trials and rarely in the diagnosis of individual patients.
The principles of medical management of malaria reflect the fact that falciparum malaria can progress rapidly to a life-threatening state and that complications can occur even after the initiation of therapy. They are: (1) early recognition of infection due to P falciparum; (2) rapid institution of appropriate therapy; (3) recognition and therapy of complications; and (4) monitoring of clinical and parasitologic response to therapy.
Malaria therapy is complicated by the fact that parasites may be present in the blood and the liver and that different drugs are required to eradicate each. Drugs which kill malaria parasites in the blood are called blood-stage schizonticides and those that kill them in the liver are called tissue schizonticides. A clinical cure refers to the elimination of parasites from the blood, which will relieve the signs and symptoms of disease. A radical cure is the eradication of all parasites from the body, both blood and liver. In cases of P falciparum and P malariae, which do not have latent liver forms (hypnozoites), an effective dose of a blood schizonticide to which the parasite is sensitive should lead to radical cure. In cases of P vivax and Povale malaria, which do form hypnozoites, radical cure requires therapy with both a blood schizonticide and a tissue schizonticide.
Recurrence of malaria infections after treatment is due either to recrudescence or to relapse. Recrudescence occurs when the blood schizonticide does not eliminate all parasites from the blood stream, either because the dose was inadequate or because the parasite is resistant to the drug. Relapse occurs in P vivax and P ovale infections after the delayed development of liver- stage parasites that have not been treated adequately with a tissue schizonticide.
Resistance of malaria parasites to antimalarials may be complete or relative; relative resistance can be overcome by raising the dose of the antimalarial.
The choice of blood schizonticide depends upon the clinical condition of the patient, infecting species and possibility of drug resistance. Parenteral therapy is reserved for patients unable to take medications by mouth and for those with complicated malaria.
Chloroquine-resistant P falciparum is widespread and currently exists in all malarious areas of the world except Mexico, Central America, the Caribbean and parts of the Middle East. P falciparum resistant to multiple drugs is most prevalent in S.E. Asia but is also present in Africa and Brazil. Chloroquine-resistant P vivax is prevalent on the island of New Guinea. Primaquine-resistant P vivax is most prevalent in S.E. Asia and Oceania and is reported from other areas. Drug resistance has not been reported for P ovale or P malariae.
If ever in doubt as to infecting species or presence of resistance, clinicians should assume the infection to be chloroquine-resistant P falciparum. Such therapy will cover all malaria species, although side effects may be more common.
The response to antimalarial therapy is monitored both clinically and by examining repeated blood films. Blood smears should be continued daily in all malaria patients until parasites are no longer detected. In severe or complicated malaria, parasitemia should be evaluated twice daily. Parasitemia should decrease by 75% and clinical status improve within 48 hr after initiating therapy. If not, drug resistance, inadequate drug levels or the presence of clinical complications should be suspected.
Treatment of Specific Infections
Uncomplicated, chloroquine-sensitive infections
All patients with uncomplicated P malariae, P ovale,and P vivax and P falciparum from chloroquine sensitive areas (see above and Fig 83-6) should be treated with oral chloroquine. The drug is highly effective, well tolerated and inexpensive.
Uncomplicated, chloroquine-resistant P falciparum
Therapy of chloroquine-resistant P falciparum is complicated and depends primarily on area of disease acquisition. Patients with uncomplicated disease acquired in areas of chloroquine resistance can be treated with one of several regimens effective against chloroquine-resistant parasites. In the United States, two regimens are used primarily: (1) mefloquine alone, or (2) quinine, plus doxycycline or pyrimethamine/sulfadoxine (FansidarR). Other effective drugs include halofantrine, artemisinin (qinghaosu) derivatives, and clindamycin. Halofantrine and artemisinin are used widely overseas but are not currently available in the U.S.
Uncomplicated, chloroquine-resistant P vivax
Chloroquine-resistant P vivax is highly prevalent on the island of New Guinea (Papua New Guinea and Irian Jaya, Indonesia) and may be present elsewhere. Recent studies in Indonesia have shown halofantrine, and chloroquine plus primaquine to be highly effective against these resistant strains. Although not specifically tested, the above regimens for chloroquine-resistant P falciparum should also be effective.
Severe or complicated malaria is a medical emergency. It is caused almost exclusively by P. falciparum. Patients with complicated malaria (see Clinical Manifestations above) should be treated with intravenous antimalarials and in an intensive care unit whenever possible. The drugs of choice are intravenous quinidine or quinine (IV quinine no longer available in the U.S.). Patients on these regimens must be observed closely for signs of hypotension or myocardial conduction abnormalities. Therapeutic plasma levels are 5 to 15 μg/ml for quinine and 5 to 10 μg/ml for quinidine. Oral quinine, plus doxycycline or FansidarR, is substituted as soon as there is clinical improvement. If acquired in an area of chloroquine-sensitive parasites, parenteral chloroquine may also be given. Artemisinin compounds show promise for therapy of severe malaria because they decrease parasitemia faster than all other antimalarials.
Any complicated P malariae, P vivax, or P ovale infection should be treated in the same way as a complicated P falciparum infection, since mixed infections are common.
Radical cure of P vivax and P ovale infections
For infections due to P vivax or P ovale, primaquine should be given after therapy of the blood-stage infection to eradicate hypnozoites of these species and prevent relapses. P vivax with decreased sensitivity to primaquine is prevalent in SE Asia and Oceania, where up to 30% of cases relapse after the standard regimen of 15 mg/day primaquine base for 14 days, and is reported from other areas. Resistance is usually relative and most initial failures respond to 30 mg/day for 14 days. Primaquine should be used with caution in persons who are G6PD deficient due to its potential to cause severe hemolysis.
Ancillary Therapy and Treatment of Complications
Supportive care and therapy of malaria complications may be as critical as choosing the correct antimalarial. Clinicians should monitor patients for complications (see Clinical Manifestations) and treat them as they occur.
Hyperthermia should be treated with cooling blankets and antipyretics. Proper fluid management is essential to prevent renal failure or pulmonary edema. If oliguric renal failure persists after fluid status is corrected, the patient is treated like other patients in the oliguric stage of acute tubular necrosis. Pulmonary edema, which may present like the adult respiratory distress syndrome (ARDS), is an uncommon but frequently fatal complication of severe P falciparum infection. It is treated by careful fluid management and application of the principles used in treating ARDS. Transfusion of erythrocytes may be necessary for severe anemia. Seizures are frequent with cerebral malaria and should be treated with standard anticonvulsants. Corticosteroids are of no benefit in the therapy of cerebral malaria. Hyperparasitemia may be treated with exchange transfusion. Exchange transfusion is generally reserved for individuals with more than 15 percent parasitemia or more than 5 percent parasitemia with cerebral malaria or other severe manifestation. Plasma glucose levels should be monitored regularly and hypoglycemia treated if it occurs. Aspiration pneumonia may occur when unconscious cerebral malaria patients suffer seizures and vomiting. Aspiration is prevented by controlling seizures and by attention to general airway management in the unconscious patient. Gram-negative bacteremia is a frequent accompaniment of severe P falciparum infection. Gram-negative organisms probably enter the circulation in areas of the bowel wall that are ischemic as a result of microcirculatory obstruction by parasitized erythrocytes. Any patient who is not responding to antimalarial therapy as expected should be investigated for bacteremia. Hypotension and shock may complicate severe malaria. If these occur, treatable causes should be considered, including Gram-negative sepsis, gastrointestinal hemorrhage, hypovolemia, and splenic rupture. Splenic rupture is seen infrequently but is one of the few fatal complications of vivax malaria.
Recent studies have found a strong association between sustained lactic acidosis and poor outcome in severe malaria. Until further work defines the role of specific interventions (e.g., sodium dichloroacetate and sodium bicarbonate) in reversing lactic acidosis in severe malaria, treatment must be aimed at the correction of defects in oxygenation and tissue perfusion and metabolic abnormalities such as hypoglycemia.
Malaria during pregnancy presents a unique problem. Pregnant women are at higher risk of developing severe and fatal malaria. Hyperparasitemia, hypoglycemia and pulmonary edema are more common in pregnant women with P falciparum infections. Pregnant women should be treated promptly with appropriate doses of antimalarials. Quinine does not appear to induce labor as was once thought. Pregnant women with chloroquine-sensitive P vivax infections should be treated with chloroquine to eliminate the erythrocytic-stage infection and then placed on weekly chloroquine to prevent relapse, as the safety of primaquine in pregnancy is not known.
Prevention of Malaria
Individuals with little or no previous exposure who develop malaria may rapidly progress to severe, often fatal disease. Most cases of malaria in Americans can be prevented by chemoprophylaxis and by avoiding the mosquito vector.
The female Anopheles mosquito feeds from dusk until dawn. During these hours, individuals should avoid contact with the mosquito by wearing protective clothing, using an insect repellent containing N,N-diethyl-l-m-toluamide (DEET), staying in screened areas and spraying these areas with pyrethrumcontaining insecticides, and by sleeping under insecticide-impregnated bednets.
Travelers to endemic areas should be advised not only on avoiding the mosquito vector but also on chemoprophylaxis. It must be emphasized that chemoprophylaxis is not one hundred percent effective; regardless of prophylaxis, malaria must be considered in the differential diagnosis of any febrile illness in an individual who has been in an area endemic for malaria within the last 2–3 years.
Chemoprophylaxis is designed to kill the parasite after it has gained access to the body but before it leads to the rupture of host RBCs, which causes the symptoms of malaria. Drugs may accomplish this by attacking the parasite in either the liver or the blood. Causal chemoprophylaxis refers to killing the parasite in the liver before it gains access to the blood. Suppressive chemoprophylaxis is accomplished by drugs which attack asexual parasites in the blood. Most antimalarial drugs attack parasites in the blood and are therefore suppressive chemoprophylactics. Primaquine is the only antimalarial drug currently available which reliably kills liver stage organisms.
The choice of a chemoprophylactic regimen depends on several factors: the health of the individual (including factors such as pregnancy, age, and chronic illness); the risk and types of malaria in the areas to be visited; and the presence of drug-resistant P falciparum.
Chloroquine is the recommended chemoprophylactic for those travelling to areas where plasmodia are still chloroquine sensitive (Mexico, Central America, Haiti, the Dominican Republic, and the Middle East). There are very few contraindications to chloroquine. Most travellers, however, visit areas where there is chloroquine resistance and other drugs, generally with greater toxicity, must be used. For most of these travelers, mefloquine is the drug of choice and doxycycline is as acceptable alternative. Extensive mefloquine resistance makes doxycycline the drug of choice for those visiting the borders of Thailand. Chloroquine plus proguanil (proguanil is not available in the U.S.) is another possible regimen for chloroquine-resistant areas, but this regimen is much less effective than mefloquine or doxycycline. Recent work also suggests that primaquine, apparently acting against liver-stage organisms, is as effective as mefloquine and doxycycline for chemoprophylaxis in areas of chloroquine resistance.
Prophylaxis with chloroquine or mefloquine should begin 2 weeks before entering the malarious area (to ensure tolerance to the drug and to provide adequate blood levels) and should continue throughout the stay in the area and for 4 weeks after leaving. Doxycycline should be started 1 to 2 days before travel to a malarious area and should be taken daily during the stay in the area and for 4 weeks after leaving. Taking the drugs after leaving the malarious area is referred to as terminal prophylaxis and is necessary to kill organisms which emerge from the liver after the person returns home. When there has been a significant risk of exposure to P vivax or P ovale, primaquine should be taken for 14 days after returning home to eliminate remaining liver stage parasites. Primaquine may be taken any time during the 4 weeks in which the blood schizonticide is being taken. See Table 83-1 for drug dosages.
Drugs Used For Chemoprophylaxis of Malaria.
Control in Populations
Control of malaria is difficult and requires the sustained effort of many individuals from many disciplines (Fig. 83-8). It is much more easily accomplished in some areas of the world than others. Control can be extremely difficult in areas where the Anopheles vector is numerous, longlived, and feeds only on humans.
Strategies for prevention of malaria. (Adapted from Miller LH, Howard RJ, Carter R et al: Research toward malaria vaccines. Science 234:1350, 1986, with permission.)
Transmission of malaria requires the presence of three factors: (1) malaria-infected humans carrying gametocytes that are infective to mosquitoes, (2) Anopheles mosquitoes that live long enough for the malaria parasites to develop within them to the infective sporozoite stage, and (3) infected mosquitoes that bite noninfected humans. Malaria control can be applied at each of these points: by treating human malaria infections and thereby reducing or eliminating the number of infected humans that mosquitoes feed on, by eliminating or reducing the numbers of Anopheles mosquitoes, by shortening the life span of mosquitoes to less than that required for the parasite to develop, or by providing alternative hosts for the mosquitoes to feed on.
There have been numerous efforts to reduce transmission by treating infected humans with drugs that render them noninfectious to mosquitoes. The success of these efforts is unclear.
Major efforts are under way to develop vaccines against malaria. Vaccines may be directed against any of the multiple stages of the organism's life cycle. Some vaccines attempt to prevent or diminish disease in the individual by inducing immune responses against sporozoites, liver-stage parasites, or erythrocytic-stage parasites, or by preventing the release of cellular products thought to be involved in pathogenesis. Other vaccines attempt to block transmission to others by inducing antibodies or cytokines that attack gametocytes, or antibodies that prevent development of the extracellular stages within mosquitoes. Several vaccines are currently undergoing evaluation in clinical trials. Recently noteworthy is SPf66, a synthetic vaccine produced in Colombia which contains peptides from the organism's blood and sporozoite stages. Although initial results were promising, subsequent clinical trials have reported protective efficacies of only about 30% against first clinical episodes of P falciparum. Further evaluation is needed to define the role of this vaccine in reducing the morbidity and mortality from malaria in various settings.
A second approach in malaria control is to reduce transmission by eliminating mosquitoes--primarily by eliminating breeding places such as lagoons and swamps or by killing the larvae in these breeding places. This approach has been quite successful in some parts of the world, particularly in areas where malaria transmission is not intense. Transmission can also be reduced by the use of insecticide-impregnated bednets. Several studies have shown the effectiveness of these bednets in reducing the morbidity from malaria in areas of intense transmission.
Another approach is to treat dwellings with residual insecticides, such as DDT, that shorten the lifespan of mosquitoes, thereby reducing the chance that they will live long enough to transmit malaria. This approach has been quite successful in some parts of the world, but has had significant problems because of the development of mosquitoes resistant to insecticides.
In some areas of the world where malaria vectors prefer animals, such as cows, to humans, the introduction of these animals has reduced malaria transmission.
Treatment of acute malaria. (Adapted from Miller LH, Howard RJ, Carter R et al. Research toward malaria vaccines. Science 234: 1350, 1986, with permission.)
Beadle C, Long GW, Weiss WR. et al. SL Diagnosis of malaria by detection of Plasmodium falciparum HRP-2 antigen with a rapid dipstick antigen-capture assay. Lancet. 1994;343:564. [PubMed: 7906328]
Bruce-Chwatt LJ: Essential Malariology, 3rd ed. Edward Arnold, Boston, 1993.
Centers for Disease Control: Health Information for International Travel 1993. HHS Publication (CDC) 938280. Washington, DC, 1993 .
Good MF, Saul AJ,(eds.) Molecular Immunological Considerations in Malaria Vaccine Development. Boca Raton, CRC Press: 1993 .
Hoffman SL. Diagnosis, Treatment and Prevention of Malaria. Med Clin North Am. 1992;76:1327. [PubMed: 1405822]
Kozarsky PE, Lobel HO. Antimalarial agents: are we running out of options? Curr Opin Infect Dis. 1994;7:701.
Mendis KN, Carter R. Clinical Disease and Pathogenesis in Malaria. Parasitol Today. 1995;11:2–15. [PubMed: 15275352]
Tanner M, Teuscher T. Alonso PL. SPf66 - The first malaria vaccine. Parasitol Today. 1995;11:10.
Warrell DA, Molyneux, Beales PF, eds. Severe and Complicated Malaria. Trans R Soc Trop Med Hyg 84 (Suppl 2):1-65, 1990 .
Has malaria finally met its scientific match? Victoria Gill finds out whether a fresh round of research funding could put an end to the killer disease
When John F Kennedy told the world in 1961 that America would put a man on the Moon by the end of the decade, he triggered one of the biggest scientific and technical projects in history. ’That challenge is one that we are willing to accept, one we are unwilling to postpone, and one which we intend to win,’ he promised in a speech the following year.
Kennedy could easily have been talking about the fight against malaria. Shortly after the second world war ended, global eradication of malaria had seemed to be a realistic goal. Wartime use of DDT had been particularly effective, and policymakers believed that all it would take to stamp out malaria was enough money to supply insecticides and medicine.
It is no small irony that by 1969, when Kennedy’s vision was finally achieved, most public health officials had concluded that the fight to eradicate malaria had largely failed. Indeed, in many regions it had actually strengthened the disease’s grip.
But now there is a renewed sense of optimism and confidence among the scientists, medics and politicians who make it their life’s work to battle the disease. It is bolstered by an enormous fundraising effort from the Global Fund to fight Aids, tuberculosis and malaria; the World Bank; the President’s malaria initiative; and other partners, all of whom coordinate their efforts under the umbrella of the Roll Back Malaria Partnership.
This surge of funding (with particularly heavy-duty support from the Bill and Melinda Gates Foundation) has been a catalyst for action, and success is emerging from its public-private research partnerships - including the first clinically effective malaria vaccine that will enter Phase III trials later this year.
Confidence is so high that Margaret Chan, director-general of the World Health Organization (WHO), now dares to talk about eradicating malaria, rather than merely controlling it. On 14 February the UN appointed a special envoy for malaria, Ray Chambers, a businessman and philanthropist who founded US-based organisation Malaria No More.
He calls the previous lack of emphasis on eradicating malaria a ’genocide of apathy’ and has pledged to raise $8-10 billion (?4-5 billion) over the next four to five years from the ’massive public-private partnership’ that has already started to bear fruit.
Achieving the impossible
Malaria currently kills more than one million people every year - and causes 500 million more to become severely ill. More than 90 per cent of deaths are in sub-Saharan Africa, where malaria kills a child every 30 seconds.
The figures are daunting for anyone hoping to combat the disease. Yet malaria has been eradicated before. From 1945 to 1970, it was driven out of large swathes of East Asia, America, as well as Eastern and Mediterranean Europe.
But during that period, the most vulnerable people in countries worst affected by the disease often had poor access to the drugs and prevention measures they needed, while civil wars and patchy funding also helped to kill off the eradication campaign.
The biggest problem, though, was the fact that the Plasmodium falciparum parasite, responsible for the most common form of malaria, can mutate rapidly to build up resistance to medicines. What’s more, the parasite’s favoured mode of transport - via the Anopheles gambiae mosquito - has also developed its own resistance to insecticides.
The traditional malaria medicine in Africa had been chloroquine - a cheap, synthetic derivative of quinine, a natural product extracted from the bark of the South American cinchona tree.
Chloroquine blocks the parasite’s ability to digest the host’s red blood cells. As the parasite feeds, it breaks down the oxygen-carrying molecule haemoglobin, releasing the iron-based haem unit. This is toxic to the parasite, which normally polymerises it into harmless haemozoin chains. But chloroquine enters the digestive vacuole of the malaria parasite and caps the haemozoin chains, preventing further polymerisation and causing a build-up of toxic haem that eventually kills the parasite.
However, mutations in the parasite have gradually reduced chloroquine’s ability to create this blockage - a single alteration to a protein that transports chloroquine into the parasite can render the drug useless.
The new gold standard of treatment is now based on another natural product found in Artemisia annua, the Chinese wormwood plant. It has been used for over 2000 years in traditional Chinese medicine, and its active ingredient - artemisinin - was originally extracted and developed into an anti-malarial drug by the Chinese government to protect soldiers during the Vietnam war.
Artemisinin also operates in the parasite’s food vacuole, breaking open its characteristic endoperoxide bridge in a reaction with the iron present in haem and haemozoin.
This reaction releases toxic haem and a burst of free radicals into the parasite. This new mode of action, bypassing changes that have allowed chloroquine resistance to develop, is what makes artemisinin so powerful.
Serious lessons have been learned from past reliance on chloroquine’s single mode of action, and the WHO now endorses the use of artemisinin combination therapies (ACTs) that incorporate several antimalarial agents - hitting multiple targets in the parasite and offsetting the risk of one powerful genetic mutation overcoming the action of a drug.
’Initially none of the big pharmaceutical companies were interested in artemisinin - there was no financial incentive because of where in the world the disease was a threat,’ says Bob Laverty, vice president of communications for Novartis - the Swiss pharmaceutical company that makes Coartem, the most widely-used ACT in Africa. ’But in the 1990s one of the companies that merged to form Novartis approached the Chinese government and expressed an interest.’
The initiative that Novartis has embarked on is one of the many public-private partnerships that have formed the most important part of the global, combined effort against the disease. The financial risk of developing a drug with no significant market is removed by the funding, and companies are gladly offering their best scientists and resources to anti-malarial drug development schemes.
The next generation
The Medicines for Malaria Venture (which gets around 60 per cent of its funding from the Gates Foundation), also based in Switzerland, is one of the non-profit organisations that funds and manages research into new malaria treatments. Its key focus for the immediate future is on new ACTs.
’We work in direct partnerships with drug companies both to develop new drugs and to get them out to patients in the field,’ says Tim Wells, MMV’s chief scientific officer. In early-stage discovery projects, MMV has teamed up with companies including GSK and Novartis to identify compounds in their pipelines that could be developed into anti-malarials.
’ACTs are like anti-malarial grenades,’ says Wells. ’Artemisinin-based drugs are combined with others that have an entirely different mode of action to hit the parasite at multiple targets.’ In Coartem, for example, artemether is combined with lumefantrine, which blocks the haem polymerisation pathway. Lumefantrine is much longer acting than artemether, and provides a back-up - protecting against resurgence of the disease.
’We have four new ACTs coming onto the market in the next couple of years,’ says Wells. ’But we need to be aware of the danger of over-use of artemisinin too - new drugs are always needed.’
One joint project with GSK has discovered that an antibiotic, triclosan, which blocks fatty acid synthesis, a process essential for bacterial growth, can have the same effect in Plasmodium falciparum. In a separate project, GSK/MMV scientists have identified a group of small molecules that inhibit the activity of proteins called falcipains, which also have an essential role in the breakdown of haemoglobin. Wells says that the next generation of anti-malarials could be available by as early as 2015.
Malaria life cycle
Female Anophelesmosquitoes carry malaria parasites in their saliva, which they inject when they bite humans - the saliva also contains an enzyme that prevents blood from clotting, and allows the insect to have a decent meal ( 1 ).
Parasites travel to the human victim’s liver as tiny sporozoites, which invade liver cells ( 2 ). Here they grow and divide, until tens of thousands of daughter cells (merozoites) escape the liver and enter the bloodstream where they invade red blood cells ( 3 ).
While some of the parasites break down the red blood cells (and are continually released in a cycle of reinvasion), some transform into sexual forms of the parasite, called gametocytes, which are able to reproduce ( 4 ). When a mosquito bites and takes a blood meal from an infected human, it ingests these gametocytes, which mature, reproduce and divide in the mosquito’s gut ( 5 ).
The eventual products of this parasite reproduction are malaria sporozoites, which travel to the mosquito’s salivary glands to start the whole cycle of infection again ( 6 ).
As new drugs are rolled out, accurate diagnosis of malaria becomes essential in order to avoid prescribing them incorrectly.
Prudence Hamade leads the malaria working group for Switzerland-based M?decins Sans Fronti?res (MSF). In Sierra Leone, she has been involved in MSF’s ’total malaria response unit’ - a pilot project that trains local people to use diagnostic kits, supplies treatments free of charge and distributes insecticide-treated bed nets.
She points out that poorly-organised drug distribution has already caused more long-term problems than it has solved. ’Diagnosis is absolutely fundamental. Because the old anti-malarial drugs were so cheap, they were distributed very freely and administered whenever someone developed malaria symptoms. But the symptoms are so broadly indicative - a fever, headache, flu-like symptoms - that could be caused by a variety of other diseases, such as typhoid. The liberal use of chloroquine has now resulted in widespread drug resistance in the parasite.’
Hamade’s project uses simple diagnostic kits that provide a result within 15 minutes. The tests use a cellulose strip containing a band of antibodies that bind to a malaria biomarker called histidine-rich protein II.
The malaria parasite uses this protein to convert the toxic haem produced as it feeds into harmless haemozoin chains - thus making the protein an excellent indicator of the parasite’s presence, requiring nothing more than a drop of blood and a buffer solution to identify.
According to Hamade, the WHO plans to publish a list of its tried, tested and approved diagnostic tests later this year - enabling healthcare workers to pick the most effective ones from hundreds on the market.
Prevention is cure
A key part of the battle that will take a big step this year is the development of an effective malaria vaccine. Joe Coen, vice-president of R&D on emerging diseases and HIV at GSK Biologicals in Belgium, has spent over a decade formulating the company’s candidate vaccine RTS,S. He calls it his baby. RTS,S uses a section from a protein called circumsprozoite (CSP), which is produced by the malaria parasite and was identified as a vaccine candidate over two decades ago.
Mutation of the parasite is once again a key problem in vaccine design - it is tricky to get the body’s immune system to recognise something that is constantly changing. But the portion of CSP used in this vaccine appears to survive mutations, so the body can be trained to recognise it as the signal for an enemy invasion.
’The original idea was to link the CSP proteins to hepatitis B viral antigen and express the fused genes in yeast cells. They self-assemble into virus-like particles,’ explains Coen. These synthetic viral particles should fool the immune system into responding as if to an infection. ’But the first attempt to develop a vaccine in 1987 failed.’
Coen’s team finally managed to stimulate an immune response with a vital extra ingredient, known as an adjuvant. Immunologic adjuvants have no effect on their own, but are used to enhance the effect of a vaccine - in this case the addition of an aluminium salt made all the difference, although the exact mechanism is unclear. ’The development of a better adjuvant system was the second part of the project,’ says Coen. ’We were able to improve the vaccine and got our first positive clinical result in 1996 - it was a tremendous breakthrough.’
He praises the financial support, particularly of the Gates foundation, for reviving the project. And in Phase II, results have been promising. Importantly, the vaccine appears to be effective in very young children, who are most vulnerable to the disease. The GSK team is now poised for the make or break Phase III trials. ’I now see a light at the end of the tunnel - this is a potentially life-saving vaccine,’ says Coen.
The first spatial map of global malaria risk to be produced in four decades shows that many of the people exposed to malaria are at a lower risk than previously thought.
The Malaria Atlas Project (MAP) found1 that 2.37 billion people are at risk of contracting malaria from Plasmodium falciparum- but about 1 billion of them live under a much lower risk of infection than was assumed by previous historical maps.
’This gives some hope of pursuing malaria elimination because the prevalence isn’t as universally high as many people suppose,’ says David Smith, a University of Florida associate professor of zoology and a co-author of the paper.
Where malaria has been successfully eradicated, it has been insecticides rather than medicines that have made the biggest impact. The distribution of insecticide-treated bed nets and indoor spraying is a major preventative approach. In the last three years, washable treated bed nets have become available. Rather than forming a temporary coating on the outside of the net, the insecticide is dispersed throughout the polymer fibres so that it gradually diffuses to the surface and is continuously replaced as the net is washed.
But the mosquito, like the parasite it carries, is not a willing victim - it evolves rapidly to evade such chemical weapons. Hilary Ranson from the Liverpool School of Tropical Medicine, UK explains that there are two kinds of resistance. ’A mutation that changes the target site can disrupt, for example, the binding of the insecticide agent,’ she says. But the parasite can also develop metabolic resistance: ’an up-regulation of the production of enzymes which allows the insect to metabolise the chemical’.
Only four types of insecticide are approved by the WHO for indoor spraying: DDT, pyrethroids, organophosphates and carbamates. All of these target specific proteins essential in the function of the insect’s nervous system - either sodium channel proteins at the membrane between nerve and muscle, or the acetylcholinesterase enzyme that breaks down a key neurotransmitter. These proteins are sufficiently unique to the insect - distinguishable from mammalian proteins - to allow them to be used with a relatively low risk of human nerve damage.
Even so, human toxicity issues mean that of the four types, only pyrethroids can be used in the bed nets themselves. ’You have to assume that babies will be sucking the nets, whereas they’re unlikely to be sucking the walls,’ says Tom McLean, senior executive officer of the Innovative vector control consortium (IVCC), who is also based at the Liverpool School of Tropical Medicine.
The IVCC was set up to address the lack of development of new insecticides against the malaria-carrying mosquito, and received a ?50 million Gates Foundation grant in 2005. ’Half of our grant will be spent on information systems,’ explains McLean. ’We’re developing tools to tell researchers what the insects are resistant to - databases of insecticides, where they are being used and signs of resistance - so that insecticide exposure can be coordinated properly. We’ve also designed lab kits that are in field trials in Africa at the moment, and we will have early versions available this year.’ The kits consist of specific reagents which highlight the DNA markers of resistance.
IVCC is also working with agrochemical companies, including Bayer and Syngenta, to develop new insecticides. ’The public health market is not big enough to support the commercial development of a new insecticide - which from bench to market takes 12-15 years and costs up to $250 million,’ says McLean. ’But if we can absorb the financial risks of development, the companies are more than happy to supply the resources and do the work.’
As well as developing new insecticides from the bench, part of this project involves scouring company pipelines for insecticides that could be used against the mosquitoes, and reviving their development specifically for that purpose. Since many of these chemicals are already registered, and laborious toxicology work has already been carried out, in theory they could be brought onto the public health market within five years.
Supply and demand
One problem with artemisinin combination therapies (ACTs) is that their shelf life can be as little as three years - a major concern when delivering drugs to remote regions lacking fundamental infrastructure is so difficult.
One project based at the University of York in the UK has started to address the issue of artemisinin supply. The university’s Centre for Novel Agricultural Products (CNAP) received a $13.6 million (?7 million) grant from the Gates Foundation for a fast-track breeding research programme for the Artemisia annua plant. The goal is to optimise the plant for high-yielding agricultural production, explains Elspeth Bartlet, a spokesperson from the university. With a combination of genetic screening and selective breeding, the CNAP researchers are rapidly producing plants with bigger leaves and larger leaf glands that contain the vital ingredient.
There have been recent concerns about overproduction of artemisinin threatening the industry (see Chemistry World, January 2008, p6). But Bartlet points out that a balance needs to be struck between supply and demand to make it sustainable. ’This glut will turn into a deficit as more farmers turn away in favour of more profitable crops, such as biofuels. We need to make this crop [Artemisia annua] profitable.’
For the people
Back in Africa, field workers like Hamade and Wells are acutely aware of the gap between major innovative steps in scientific research, and the practicalities of life in a poverty-stricken malaria region.
’People used to reject the bed nets because they found them too hot,’ says Hamade. ’But many people are now finding that they can also help them to get a good night’s sleep, so they are more accepting of them.’ The offer of free bed nets is also used as an incentive to persuade families and pregnant women to attend treatment and infant vaccination clinics. Wells points out that, for children, something as simple as disguising the unpleasant taste of medicines can make a significant difference in how well they stick to their drug regimens - cherry flavour works particularly well.
It’s clear that every piece of this colossal jigsaw counts. Malaria was underestimated once, and the fight against it failed, allowing an even more deadly disease to bounce back. Importantly, the current approach incorporates research projects that will find out if prevention and treatment measures are actually compatible with people’s lives.
The Global Fund’s project aims to halve deaths from tuberculosis and malaria by 2015, and the proof of their strategy will be in those disease’s stark mortality figures. But with a long-term commitment of resources in place, scientists can at least work to arm those on the front line with a fresh set of weapons against malaria.1 C A Guerraet al. PLoS Med,2008, , e38. DOI:10.1371/journal.pmed.0050038
1 C A Guerraet al. PLoS Med,2008, 5, e38. DOI:10.1371/journal.pmed.0050038