◆ Malaria is a very common cause of severe infection from Africa, Asia, and Latin America. All severely-ill patients from these areas should have a malaria blood film.
◆ There is good evidence that artesunate reduces mortality compared to quinine, but do not delay treatment with quinine if artesunate is not available.
◆ Cerebral malaria, renal failure, and acute respiratory distress syndrome (ARDS) are the major syndromes in adults. There is little evidence for specific adjunctive treatments for these.
◆ Severe malaria is a significant risk factor for bacterial sepsis in children, but probably not adults.
◆ The prognosis for adult malaria patients if they survive is good.
Malaria is the commonest life-threatening tropical infection. Diagnosed and treated early it will seldom require critical care and the outlook is good. Once severe malaria is established, it is a serious multi-system disease, which can continue to evolve even once the malaria parasites have cleared from the blood. This chapter is written for critical care physicians caring for predominantly adult patients. There are important differences from the management of severe childhood malaria, and as much of the data comes from children these are highlighted.
Epidemiology and when to suspect malaria
Malaria occurs in most countries in Africa, Asia, and Latin America, but incidence varies widely. In many parts of sub-Saharan Africa people can have clinical attacks of malaria four or more times a year. Malaria is much less common from most of Asia and Latin America, with many areas malaria-free. This can mean the diagnosis is not suspected and therefore delayed. The great majority of malaria that leads to critical illness is falciparum malaria, which causes around 95% of malaria in Africa. Vivax malaria is the most common form in much of Asia and Latin America, and relatively rarely causes critical illness, although severe cases have been more frequently recognized recently and a benign course cannot be assumed. In returning travellers presenting in non-endemic countries over 90% of cases of falciparum malaria will present within 3 months of return and most within a month, although delayed presentation of up to a year after return does occasionally occur. Very occasionally unexpected malaria is caused by blood transfusion, or sharing needles with intravenous drug use.
Diagnosis of malaria
Malaria can present in multiple ways; no clinical symptom or sign either proves or excludes the diagnosis. In a critically unwell patient with an exposure history fever, unconsciousness, or fitting, jaundice or splenomegaly are suggestive. Whilst a history of fever is usual, around half of patients with malaria are afebrile at the time of presentation. A malaria blood test is therefore essential.
For management of the critically-ill patient blood film microscopy for parasites remains the best test. It identifies the species of malaria, and importantly, the percentage of red cells parasitized. Rapid diagnostic tests for malaria are mostly as sensitive as microscopy, and may speed the diagnosis, but they are not a substitute for microscopy in severely ill patients as they give an all-or-nothing response. In critically-ill patients prior to treatment a sufficient number of parasites for diagnosis will almost always be present so microscopy is sensitive and specific.
Initial assessment of severity
Severe malaria can present with a number of syndromes. These may occur in isolation, or in any combination.
The most well-known is cerebral malaria. Strictly unrousable coma in the presence of asexual malaria parasites, any reduction of consciousness, neurological signs or fitting should be treated as cerebral malaria. It generally presents before starting antimalarial treatment, but especially in patients who present with very high parasite counts cerebral malaria may occur in the first 48 hours after starting treatment as parasites mature and begin to stick down (sequester) in the blood vessels in the brain.
Renal failure may present as oliguria or anuria, or be diagnosed from raised creatinine or blood urea. Renal failure may be present at initial diagnosis or develop in the first few days after treatment has started, including when parasites have cleared from the blood. There are potentially two components to renal failure in malaria; pre-renal failure in febrile patient not drinking adequately, and the direct effects of malaria on the kidney (the exact pathological mechanism is currently unknown). If there is any doubt whether prerenal failure is a major component rapid fluid challenges will help identify it. Renal failure is common in critically-ill adults with malaria (but rare in children) .
Malaria-associated respiratory distress in adults is usually due to problems in the lungs, with two potential causes. Pulmonary oedema, often caused or exacerbated by enthusiastic rehydration in patients with renal failure, is reversible. Adult respiratory distress syndrome (ARDS)/acute lung injury (ALI) often presents late in the course of severe malaria, including days after malaria parasites have cleared from the blood. In children respiratory distress is also a sign of severe disease, but is generally caused by acidosis rather than poor oxygen transfer and lungs are normal. In both adults and children coexisting pneumonia should be excluded.
Of the uncommon manifestations of severe malaria the most important is disseminated intravascular coagulation (DIC), and occasionally cerebral or other serious bleeds where there is no DIC due to the very low platelets which are commonly found in severe malaria.
In contrast to most critically-ill patients with sepsis, shock is relatively rare except when there is coexisting bacterial sepsis. Bilirubin is often raised due to massive haemolysis, but liver function rarely significantly impaired.
Two warning signs in otherwise well patient with malaria make it likely the patient will deteriorate even after starting treatment. The first is finding a high parasite count in the peripheral blood. Patients with parasite counts of greater than 10%, even if appearing initially well, have a very high chance of deteriorating over the next 48 hours as parasites mature. In high resource settings where malaria is not endemic, it is prudent to treat any patient with parasite count greater than 2% with parenteral drugs even if they are otherwise well as their risk of severe disease is significantly raised . The second is jaundice which in the presence of malaria generally indicates massive haemolysis.
Three groups are at particular risk of poor outcomes; elderly patients as mortality increases steadily with age , pregnant patients where the risk of complications in the pregnancy are significant, and patients with AIDS.
All patients with severe malaria should be treated rapidly with high doses of effective parenteral antimalarials, and this is by far the most important step in managing critically-ill patients. In practice the choice is between quinine (or in the USA quinidine) and artesunate. Quinine has been the mainstay of managing severe malaria for over a century, and remains an effective drug. In the last few years however there has been clear evidence from large randomized trials in children and adults in Africa and Asia that artesunate has a significant mortality advantage [4,5]. When both drugs are available therefore artesunate is preferable. No head-to-head comparisons of quinidine with artesunate have been undertaken, but there is no reason to believe that it would perform better than quinine so artesunate remains the best drug available. At present however stocks of artesunate are limited in many countries and the priority should be to treat as fast as possible with a high dose of an intravenous effective drug; better to use quinine (or quinidine) immediately than artesunate after several hours delay. In most non-endemic countries specialist centres hold critical stocks of artesunate.
A number of factors make artesunate particularly appropriate. Where patients have hyperparasitaemia (>10%) artesunate brings down parasite counts much faster than quinine, kills all ages of parasites, and it is in this group that the biggest survival advantage was found in trials in adults . In patients who are at risk of arrhythmia quinine and quinidine are arrhythmogenic. Quinine has a number of unpleasant, but generally non-severe side-effects, but an important one is that it is hypoglycaemic, and in patients on quinine blood glucose must be monitored closely. The side effects of quinidine are very similar to those of quinine, but there is a roughly 2 times higher risk of arrhythmias with quinidine for the same malaria killing dose.
Quinine should have a loading dose of 20 mg/kg as a slow infusion, followed by 10 mg/kg 8–12-hourly. Artesunate is given at a dose of 2.4 mg/kg as a bolus, repeated at 12 and 24 hours, then daily. Both should be continued until either parasites have cleared from the blood, or the patient is able to take oral treatment and has no remaining signs of severe disease (again until parasites are clear). For oral follow-up a combination of two drugs should be used with either quinine or an artemisinin drug; most countries have local guidelines for oral treatment.
Treatment of complications
Other than rapidly treating malaria, the majority of the treatment of the critically-ill malaria patient is supportive, and does not differ significantly from the management of other critically-ill patients with multisystem disease. Many adjunctive therapies such as steroids have been tried, generally based on contemporary (often erroneous) beliefs about malaria pathophysiology, but almost all have fallen into disuse when proper clinical assessment demonstrated either they do not work or that they do harm. The evidence base for managing severe malaria in children is generally better than that in adults. Unfortunately, as the pathophysiology of the disease differs between adults and children, it is unwise to extrapolate between them.
The most important thing in cases of reduced consciousness and malaria is to ensure that they do not have hypoglycaemia, which can be caused by the parasite and in patients on quinine the treatment.
In children in particular there is also risk of subclinical fitting for which the treatment is of complex partial status with antiepileptic drugs. This can manifest by subtle twitchings or, where available, EEG monitoring. Based on the finding that they reduce the risk of fitting, for a time prophylactic anticonvulsants were recommended for adults and children, but a trial in Kenyan children which demonstrated reduced fitting, but increased mortality means they are now generally only recommended in established fitting .
Because of evidence of some degree of brain oedema in cerebral malaria, steroids and mannitol have both been tried and seem to increase risks if anything.
In adults who recover even from prolonged (i.e. many weeks) unconsciousness in malaria, neurological deficits are surprisingly rare compared to other neurological infections. The great majority of patients with cerebral malaria will recover with no noticeable deficits. The rate of neurological sequelae is higher in children.
As with all acute renal failure the key is to minimize electrolyte disturbance, particularly high potassium and this may well require haemodialysis or haemofiltration. There is very weak evidence that early haemofiltration speeds recovery. Although renal support may be needed for some weeks following recovery from malaria, kidneys almost invariably recover back to their normal function with only a few case reports of persisting renal failure.
A very rare complication, although it used to be common when quinine was given in repeated low doses as prophylaxis, is blackwater fever (almost black urine). Almost all cases now are associated with G6PD deficiency, and are probably caused by quinine related drugs causing haemolysis. It is not a reason to stop therapy provided the haemoglobin is monitored and maintained when necessary.
Adult respiratory distress syndrome, disseminated intravascular coagulation
These are treated as with any other patients with these syndromes with no malaria-specific therapy proven to help other than treating the underlying condition.
Hyperparasitaemia and exchange transfusion
There is a logic that in patients treated with quinine who have very high parasite counts (>20% red cells) physically removing parasites will prevent them from developing into mature forms that sequester in vital organs. Quinine exerts the majority of its effect on mature parasites so many of these immature forms will progress if not removed even when on quinine treatment. In some centres it is therefore the practice with these very high parasite counts either to undertake an exchange transfusion where six units of blood are removed and six replaced simultaneously, or mechanical red-cell exchange. There is relatively little evidence either that this does any good, or any harm, beyond the usual risks of transfusion and substantial fluid shifts . In patients treated with artesunate parasite counts drop rapidly, and the drug kills young parasite forms, so the logic of exchange transfusion largely disappears. Exchange transfusion is therefore now rarely recommended.
In adults, how best to manage fluids in malaria patients is hotly debated, largely because little convincing evidence exists. There is weak evidence that over hydration of patients leading to pulmonary oedema may increase the risk of ARDS. Malaria physicians are therefore usually cautious about minimizing the risk overhydration. In children a recent major trial showed convincingly that bolus fluids lead to increased mortality; it is not clear whether this is relevant to adults .
Shock, and coexisting bacterial sepsis
Shock is relatively rare in malaria except when patients are significantly dehydrated, there is coexisting bacterial sepsis (particularly Gram-negative sepsis) and in rare cases of occult gastrointestinal bleeds with DIC. Patients who become shocked with malaria therefore should be treated with broad spectrum antibiotics with Gram-negative cover in addition to supportive treatment. In children, severe malaria seems to increase substantially the risk of sepsis, but this association is less clear in adults . It is therefore prudent to treat children prophylactically with antibiotics if they have severe malaria even without shock.
Anaemia and thrombocytopenia
Life-threatening anaemia is a common manifestation of malaria in children in low resource settings, but in non-pregnant adults it is rare, although some degree of anaemia is likely in all malaria cases because malaria destroys red cells. Transfusion is only likely to be useful in cases where haemoglobin has fallen substantially, or there is cardiovascular compromise. There is no consensus over what level of haemoglobin is appropriate treat in the absence of cardiovascular compromise, although in high resource settings Hb below 7 g/dL is a reasonable threshold. Thrombocytopenia is invariable in severe malaria. Except when coexisting with DIC it rarely causes significant clinical problems, and there is little association between scale of thrombocytopenia and severity of malaria. Platelet transfusions are probably only useful in the context of active bleeding where platelets fall below 30/L as hypersplenism means transfused platelets will be rapidly consumed.
Depending on syndrome, mortality from severe malaria by the WHO criteria in adults is likely to be 5–15%. In a high resource settings childhood deaths from malaria are extremely rare  although in Africa childhood deaths are the great majority. In patients with isolated cerebral malaria and no other organ involvement, if the patient survives the first 48 hours the outlook is usually good. Deaths from isolated renal failure should not occur where haemodialysis and haemofiltration are available. The mortality from ARDS, the rarest of the major syndromes, is the highest with over 50% reported in some series.
The majority of those who die from malaria in well-resourced critical care settings do so from the complications associated with long intensive care unit stays, in particularly nosocomial infection once parasites have cleared. The great majority of patients with severe malaria will survive, even if unconscious for a long time, and their outlook for making a full recovery is good, with very rare cases of significant neurological impairment or other disability, although these do occur.
1. Marks ME, Armstrong M, Suvari M et al. (2013) Severe imported falciparum malaria among adults requiring intensive care: a retrospective study at the hospital for tropical diseases, London. BMC Infectious Diseases. 13, 118.Find this resource:
2. Phillips A, Bassett P, Zeki S, Newman S, and Pasvol G. (2009). Risk factors for severe disease in adults with falciparum malaria. Clinical Infectious Diseases, 48, 871–8.Find this resource:
3. Checkley AM, Smith A, Smith V, et al. (2012). Risk factors for mortality from imported falciparum malaria in the United Kingdom over 20 years: an observational study. British Medical Journal, 344, e2116.Find this resource:
4. Dondorp A, Nosten F, Stepniewska K, et al. (2005). Artesunate versus quinine for treatment of severe falciparum malaria: a randomised trial. Lancet, 366, 717–25.Find this resource:
5. Dondorp AM, Fanello CI, Hendriksen IC, et al. (2010). Artesunate versus quinine in the treatment of severe falciparum malaria in African children (AQUAMAT): an open-label, randomised trial. Lancet, 376, 1647–57.Find this resource:
6. Crawley J, Waruiru C, Mithwani S, et al. (2000). Effect of phenobarbital on seizure frequency and mortality in childhood cerebral malaria: a randomised, controlled intervention study. Lancet, 355,701–6.Find this resource:
7. Auer-Hackenberg L, Staudinger T, Bojic A et al. (2012) Automated red blood cell exchange as an adjunctive treatment for severe Plasmodium falciparum malaria at the Vienna General Hospital in Austria: a retrospective cohort study. Malaria Journal, 11, 158.Find this resource:
8. Maitland K, Kiguli S, Opoka RO, et al. (2011). Mortality after fluid bolus in African children with severe infection. New England Journal of Medicine, 364, 2483–95.Find this resource:
9. Nadjm B, Amos B, Mtove G, et al. (2010). WHO guidelines for antimicrobial treatment in children admitted to hospital in an area of intense Plasmodium falciparum transmission: prospective study. British Medical Journal, 340, c1350.Find this resource: