Accurate diagnosis of cerebral malaria: a role for parasite histidine-rich protein 2?

The World Health Organization (WHO) definition of cerebral malaria requires Plasmodium falciparum parasitemia and coma not attributable to convulsions, sedatives, hypoglycemia, or another detectable nonmalarial cause [1]. In a series of elegant autopsy-based studies, Taylor and colleagues demonstrated that as many as 23% of children who meet this WHO definition of cerebral malaria actually die from other causes and that the presence of malaria retinopathy has a >90% sensitivity and specificity for predicting “true” cerebral malaria, in which parasite sequestration is documented in cerebral capillaries [2]. Assessment of malaria retinopathy in cerebral malaria has led to significant research advances, but the assessment requires highly specialized training and equipment, so it has not been practical in clinical settings in most low-income countries. Parasite histidine-rich protein 2 (pHPR2) is produced by P. falciparum throughout its life cycle [3]. It is released from infected erythrocytes as a watersoluble protein [4]. In areas of low to moderate transmission, plasma concentrations of pHRP2 appear to correspond well to body parasite biomass [5]. Elevated pHRP2 concentrations may, therefore, reflect a high level of sequestered parasites. In the current issue of the journal, Seydel and colleagues provide evidence that high pHRP2 concentrations are an excellent marker for biopsyor retinopathy-confirmed cerebral malaria [6]. The study design has several strengths that boost confidence in the accuracy of its findings, including the use of 3 study groups, which allowed for initial testing, establishment of an optimized pHRP2 cutoff level, and prospective assessment of the sensitivity and specificity of that cutoff level. The first study group consisted of children with WHO-defined cerebral malaria who died. The pHRP2 concentrations in children with autopsy-confirmed parasite sequestration were compared with the pHRP2 concentrations in those without sequestration. The pHRP2 concentrations distinguished almost perfectly between the 2 groups (area under receiver operating characteristic curve, 0.98). The second study group consisted of children from an earlier study who had WHOdefined cerebral malaria and who did (cases) or did not (controls) have malaria retinopathy. The pHRP2 concentrations in this group were used to identify a cutoff pHRP2 concentration that yielded optimal sensitivity and specificity for malaria retinopathy. The third study group was a cohort of children with WHO-defined cerebral malaria who were examined prospectively for malaria retinopathy. The pHRP2 cutoff concentration established in the second study group was assessed in this third study group. The established pHRP2 cutoff concentration (1700 ng/ mL) had a positive predictive value of 94% and a negative predictive value of 79% for malaria retinopathy. With these positive and negative predictive values, a pHRP2 value above the cutoff in a child with a diagnosis of cerebral malaria would greatly strengthen confidence in the diagnosis, whereas a value below the cutoff might prompt more vigorous investigation of other potential causes of coma. Because almost a quarter of children classified as having cerebral malaria actually have other reasons for coma, a pHRP2 cutoff test may prompt the clinician to seek and address other diagnoses early in the disease process. This application is one for which a simple pHRP2 concentrationbased test holds real promise as a tool that could improve disease outcome. So, are we ready to move to field use of a pHRP2 concentration-based test to identify true cerebral malaria in children and adults with P. falciparum parasitemia and coma? Not quite yet. First, other studies are needed to replicate the findings reported here by Seydel et al. In particular, it will be important to assess whether findings are similar in older children and adults in Southeast Asia who develop cerebral malaria. Second, uniform protocols for measurement of Received 6 March 2012; accepted 17 May 2012; electronically published 25 May 2012. Correspondence: Chandy C. John, MD, Division of Global Pediatrics, University of Minnesota Medical School, 717 Delaware St SE, Rm 366, Minneapolis, MN 55414 (ccj@umn. edu). The Journal of Infectious Diseases 2012;206:307–8 © The Author 2012. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals. [email protected]. DOI: 10.1093/infdis/jis373