The prognostic and diagnostic value of intraleukocytic malaria pigment in patients with severe falciparum malaria

Abstract

Severe falciparum malaria is a major cause of death in tropical countries, particularly in African children. Rapid and accurate diagnosis and prognostic assessment are critical to clinical management. In 6027 prospectively studied patients diagnosed with severe malaria we assess the prognostic value of peripheral blood film counts of malaria pigment containing polymorphonuclear leukocytes (PMNs) and monocytes. We combine these results with previously published data and show, in an individual patient data meta-analysis (n = 32,035), that the proportion of pigment containing PMNs is predictive of in-hospital mortality. In African children the proportion of pigment containing PMNs helps distinguish severe malaria from other life-threatening febrile illnesses, and it adds to the prognostic assessment from simple bedside examination, and to the conventional malaria parasite count. Microscopy assessment of pigment containing PMNs is simple and rapid, and should be performed in all patients hospitalised with suspected severe malaria.

Fig. 1. New data from three randomised controlled trials in severe falciparum malaria.

The prognostic value of quantitative microscopy parasite counts, pigment containing neutrophil (PMN) counts and pigment containing monocyte counts. Panels a–c show data from Asian adults and children (AQ Vietnam and SEAQUAMAT trials); panels d–f show data from African children (AQUAMAT trial). Each panel shows the histogram of quantitative counts (top), and the relationship between the count variable and mortality (mean: black line; 95% confidence interval: grey shaded area) estimated under a generalised additive logistic regression model (bottom). Note the y-axis scales are different for panels a–c and panels d–f.

Fig. 2. The prognostic value of quantitative microscopy cell counts in the pooled data set of >30,000 children from sub-Saharan Africa diagnosed with severe falciparum malaria.

Each panel shows the histogram of quantitative counts (top), and the relationship between the count variable and mortality (mean: black line; 95% confidence interval: grey shaded area) estimated under a generalised additive logistic regression model (bottom). Panel a: parasite density; panel b: pigment containing neutrophils [PMNs]; panel c: pigment containing monocytes.

Fig. 3. Meta-analysis of the prognostic value of pigment-containing PMNs (polymorphonuclear leukocytes.

panel a) and monocytes (panel b) stratified into five levels with no pigment containing cells as the reference. Point estimates (filled circles) and 95% confidence intervals (thick lines) are shown, estimated under a logistic regression model with nested random effects for site, country and study.

Fig. 4. Meta-analysis of the prognostic value for mortality of >5% versus ≤5% pigment-containing polymorphonuclear leukocytes (pPMN) in 32,035 patients clinically diagnosed with severe falciparum malaria.

The individual point estimates (shaded squares with centered vertical ticks) and 95% confidence intervals (horizontal lines) are shown for each study broken down by country of enrolment except for SMAC, Gabon where the distribution of pigment-containing PMNs was substantially different between the two sites (Lambaréné and Libreville).

Fig. 5. The prognostic value of intraleukocytic pigment cell counts in severe falciparum malaria with and without adjustment for known prognostic clinical and laboratory variables (clinical: coma and acidosis; laboratory: hypoglycaemia and either venous lactate for AQ Vietnam and SMAC or blood urea nitrogen for SEAQUAMAT and AQUAMAT).

Panel a: PMNs (polymorphonuclear leukocytes); panel b: monocytes. The point estimates (95% confidence intervals) for the nonadjusted models are shown by the blue triangles (blue lines); for the models adjusted with clinical variables only are shown by the purple circles (purple lines); for the models adjusted with clinical and laboratory variables are shown by the red squares (red lines).

Fig. 6. The diagnostic value of the parasite count, pigment-containing PMNs (polymorphonuclear leukocytes) and monocytes in severe falciparum malaria.

Panels a–c show scatterplots of the parasite count and the proportions of pigment containing PMNs and monocytes, respectively, against the plasma PfHRP2 concentrations in the 2933 patients from the AQUAMAT trial who had all measurements recorded. Red: patients who died; blue: patients who survived. In panels d–f, the black lines (grey shaded areas) show the mean (95% confidence interval) probability of having a plasma PfHRP2 concentration ≥1000 ng/ml as a function of the parasite count and the pigment containing leukocyte counts.