BackgroundThrombocytopenia is a common finding in adults with severe falciparum malaria, but its clinical and prognostic utility is incompletely defined.MethodsClinical and laboratory data from 647 adults with severe falciparum malaria were analysed retrospectively to determine the relationship between a patient’s platelet count on admission to hospital and their subsequent clinical course.ResultsOn admission, 614 patients (94.9%) were thrombocytopenic (platelet count <150 × 109/L) and 328 (50.7%) had a platelet count <50 × 109/L. The admission platelet count was inversely correlated with parasite biomass (estimated from plasma PfHRP2 concentrations, rs = −0.28, P = 0.003), the degree of microvascular sequestration (measured with orthogonal polarizing spectral imaging, rs = −0.31, P = 0.001) and disease severity (the number of World Health Organization severity criteria satisfied by the patient, rs = −0.21, P <0.001). Platelet counts were lower on admission in the patients who died (median: 30 (interquartile range 22 to 52) × 109/L versus 50 (34 to 78) × 109/L in survivors; P <0.001), but did not predict outcome independently from other established laboratory and clinical prognostic indices. The 39 patients (6%) with profound thrombocytopenia (platelet count <20 × 109/L) were more likely to die (odds ratio: 5.00, 95% confidence interval: 2.56 to 9.75) than patients with higher platelet counts, but these high-risk patients could be identified more rapidly with simple bedside clinical assessment. The admission platelet count did not reliably identify the 50 patients (7.7%) with major bleeding during the study.ConclusionsThrombocytopenia is a marker of disease severity in adults with falciparum malaria, but has limited utility in prognostication, triage and management.