Correspondence

Indicators of Life-Threatening Malaria

N Engl J Med 1995; 333:1011-1012October 12, 1995

Article

To the Editor:

Marsh et al. (May 25 issue)1 identified impaired consciousness, respiratory distress, hypoglycemia, and jaundice as ominous prognostic indicators in African children with a primary diagnosis of malaria. Our findings, made in a different setting (the urban environment of Ouagadougou, capital of Burkina Faso in West Africa), support these data but raise some additional points.

During the rainy seasons of 1993 and 1994, 739 children with severe malaria (age, 6 months to 15 years; mean [±SD], 4.55±3.03 years) who were admitted to the pediatric ward of Ouagadougou University Hospital were included in a study aimed at relating immune responses to the severity of malaria. Inclusion criteria followed the World Health Organization (WHO) definition of severe and complicated malaria.2 Symptoms and signs on admission as well as the outcome were carefully noted. The children were treated according to WHO guidelines2 with a complete regimen of drugs that were provided free as part of the study.

Of 612 children whose outcomes were known, 85 died while in the hospital. The mean (±SD) age of the children who died was 3.38±2.07 years. The prevalence of certain conditions at admission and the related case fatality rates are given in Table 1Table 1Prevalence of Certain Conditions at Admission among Children Hospitalized for Malaria and the Related Case Fatality Rates..

If a prevalence analysis in the study by Marsh et al. is restricted to patients with severe malaria (according to WHO criteria), major differences become obvious between the two study groups. In Kenyan patients, the prevalences of the most common signs were as follows: convulsions, 48.4 percent; severe anemia, 45.8 percent; respiratory distress, 36.0 percent; prostration, 27.5 percent; and coma, 26.5 percent. The overall mortality rate for patients with severe malaria was 7.3 percent, as compared with 13.9 percent in our study.

Despite these differences, univariate analysis of data on the Ouagadougou patients showed the same signs and symptoms identified by Marsh et al. to be significantly associated with higher case fatality rates than that for the group as a whole. The single important exception was severe anemia. The case fatality rate associated with this sign was 24 percent in our group as compared with 4.7 percent in Kenyan patients. The key role of severe anemia in identifying different clinical patterns of severe malaria in African children living in areas with different rates of transmission of malaria has recently been discussed by Snow et al.3

Our findings confirm the need to concentrate on a few conditions as key prognostic indicators for life-threatening malaria, as proposed by Marsh et al. The identification of these conditions, however, has to be tailored to the local epidemiologic situation and to the resulting clinical pattern. This approach will be useful in order both to adopt rational, cost-effective clinical management protocols and to deepen the pathophysiologic knowledge of still poorly understood important clinical symptoms, such as severe anemia and respiratory distress.

David Modiano, M.D., Ph.D.
Università di Roma “La Sapienza”, 00100 Rome, Italy

Alphonse Sawadogo, M.D.
Centre Hospitalier National Yalgado Ouedraogo, Ouagadougou 03, Burkina Faso

Franco Pagnoni, M.D.
Centre National de Lutte contre le Paludisme, Ouagadougou 01, Burkina Faso

3 References

1. 1

   Marsh K, Forster D, Waruiru C, et al. Indicators of life-threatening malaria in African children N Engl J Med 1995;332:1399-404.
   

2. 2

   Warrell DA, Molyneux ME, Beales PF. Severe and complicated malaria. Trans R Soc Trop Med Hyg 1990;84:Suppl 2:1-65
   

3. 3

   Snow RW, Bastos de Azevedo I, Lowe BS, et al. Severe childhood malaria in two areas of markedly different falciparum transmission in east Africa. Acta Trop (Basel) 1994;57:289-300
   CrossRef | Web of Science | Medline

To the Editor:

Marsh et al. determined the relative risk of death using certain prognostic clinical criteria in a large group of hospitalized children with a primary diagnosis of malaria in Kenya. Though the relative risk was statistically significant for four prognostic indicators, the results are inconclusive. The analysis was flawed because the authors did not control for certain other factors (sickle cell disease and trait,1 thalassemia, glucose-6-phosphate dehydrogenase deficiency, and HLA antigens2), which have been reported to be independently associated with the severity and outcome of Plasmodium falciparum malaria and are confounding factors.

Soumitra Sarkar, M.D., M.P.H.
University of Southern California School of Medicine, Los Angeles, CA 90033-9987

2 References

1. 1

   Luzzatto L. Genetics of red cells and susceptibility to malaria. Blood 1979;54:961-976
   Web of Science | Medline

2. 2

   Hill AV, Allsopp CE, Kwiatkowski D, et al. Common west African HLA antigens are associated with protection from severe malaria. Nature 1991;352:595-600
   CrossRef | Web of Science | Medline

Author/Editor Response

Dr. Marsh replies:

To the Editor: We thank Modiano and colleagues for making their data available and strongly support their suggestion that this prognostic approach needs to be tailored to the local epidemiologic situation. The clinical spectrum of severe malaria appears to change markedly with the rate of transmission, particularly in terms of the age at presentation and the relative importance of severe anemia and coma.1,2 Although we expect that basic principles of management can be usefully extended from one situation to another, there may also be important differences. For example, we find little evidence that children with anemia and respiratory distress are in congestive cardiac failure, and our approach to rapid resuscitation is tailored accordingly. In a situation in which children are exposed to higher rates of transmission (and possibly higher background levels of anemia) it is possible that the same apparent clinical end point has a different pathogenesis, and this needs to be established. We encourage colleagues to perform such analyses to guide the development of local management strategies.

We do not share Sarkar's worry about our results being inconclusive (though they are, in some sense, inconclusive). Our study was designed to examine the use of practical prognostic indexes in children admitted to the hospital with severe malaria. It was, in fact, the clinical arm of a case–control study that examined a wide range of risk factors involved in the development of severe malaria and whose results will be published separately. We know of no evidence that any putative risk factors affect the risk of death once clinically severe malaria has developed, and if any were to be found it seems unlikely that they would have a major role in the immediate assessment of sick children.

Kevin Marsh, M.B., Ch.B.
Kenya Medical Research Institute, Kilifi, Kenya

2 References

1. 1

   Snow RW, Bastos de Azevedo I, Lowe BS, et al. Severe childhood malaria in two areas of markedly different falciparum transmission in East Africa. Acta Trop (Basel) 1994;57:289-300
   CrossRef | Web of Science | Medline

2. 2

   Snow RW, Marsh K. Will reducing Plasmodium falciparum transmission alter malaria mortality among African children? Parasitol Today 1995;11:188-190
   CrossRef

Citing Articles (2)

Citing Articles

1. 1

   Marcel Hommel, Herbert M. Gilles. 2010. Malaria. .
   CrossRef

2. 2

   World Health Organization. (2000) Severe falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 94, 1-90
   CrossRef