Severe malaria in west African children

Abstract

Sir Olaf Müller and co-workers (Jan 4, p 86)1 suggest that malnutrition may be a far more important cause of anaemia than malaria in west African children. This is not a trivial issue, since it may redirect malaria control efforts toward nutrition programmes. This supposition goes against our clinical experience. In places with seasonal malaria, anaemia is more common at the end of the rainy season, when malaria is flourishing and food is abundant, than at the end of the dry season, when malaria is infrequent and malnutrition common. In northern Ghana, K Koram and co-workers2 found severe anaemia in 22·1% and 1·4% of children aged 6–24 months at the end of the rainy and the dry season, respectively; there was no such difference in anthropometric indices. Few patients with anaemia are admitted to our hospital in southern Ghana during the dry season, whereas in the rainy season we see many anaemic patients. Between 20% and 25% of these patients with anaemia do not have visible parasites on microscopy (J Kurtzhals and colleagues, unpublished observations). To find out whether these cases of anaemia are caused by malaria, we retrospectively analysed plasma samples from 48 randomly selected patients with symptoms of malaria but no microscopically detectable parasitaemia. 24 of these patients had severe anaemia (haemoglobin <50 g/L). The other 24 patients, matched by age, clinical history, and temperature, had a haemoglobin count above 80 g/L. Of the 24 anaemic patients, 20 (83%) had soluble malaria antigens in their plasma, compared with only two (8%) controls (odds ratio 55, p<0·001). The anaemic patients had mean corpuscular haemoglobin concentrations and mean corpuscular volumes within the normal range (M Helleberg and colleagues, unpublished observations) which argues against a role for iron deficiency. By contrast, these patients with anaemia had high concentrations of neopterin—a sign of macrophage activation that has previously been associated with erythrophagocytosis in malaria3 —and moderate thrombocytopenia, which is common in patients with malaria. We did not test for HIV infection, but the anaemic patients had raised lymphocyte counts, arguing against a role for HIV infection. Thus, a high proportion of patients with severe anaemia may be wrongly classified as not having malaria. One possible explanation is that patients had been treated before admission to hospital, or parasites could have sequestered in the vascular bed.