The Rational Clinical Examination
November 10, 2010

Does This Patient Have Malaria?

Author Affiliations

Author Affiliations: Division of Infectious Diseases and International Health (Dr Taylor) and Department of Medicine (Dr Simel), Duke University Medical Center, Durham, North Carolina; Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of Medicine, Blantyre, Malawi (Dr Molyneux); School of Tropical Medicine, University of Liverpool, Liverpool, England (Dr Molyneux); Durham Department of Veterans Affairs Medical Center, Durham, North Carolina (Dr Simel); and Department of Epidemiology, Gillings School of Global Public Health (Drs Taylor and Meshnick), and Division of Infectious Diseases, School of Medicine (Dr Juliano), University of North Carolina, Chapel Hill.

JAMA. 2010;304(18):2048-2056. doi:10.1001/jama.2010.1578

Context Malaria commonly infects residents of and travelers to tropical regions. The clinical features of infection are notoriously nonspecific but have not been comprehensively evaluated.

Objective To systematically review and synthesize data related to the predictive value of clinical findings for the diagnosis of malaria in endemic areas and in travelers returning from endemic areas.

Data Sources, Study Selection, and Data Extraction The databases of MEDLINE and EMBASE (1950-July 2010) were searched to identify studies published in the English language of endemic and “imported” (acquired during travel) malaria. Additional studies were identified from reference lists. Studies were included that had patients suspected of having acute malaria (usually because of fever) and compared the presence or absence of clinical findings with blood smear confirmation. Two authors independently identified studies, appraised study quality, and extracted data on the patient population, outcome assessment, and clinical findings. Differences between reviewers were resolved by consensus.

Data Synthesis Fourteen studies for endemic malaria were identified that met review criteria. Individual symptoms are of limited diagnostic utility but presence of splenomegaly (summary likelihood ratio [LR], 3.3; 95% confidence interval [CI], 2.0-4.7) or hepatomegaly (summary LR, 2.4; 95% CI, 1.6-3.6) make malaria more likely. Combinations of findings can affect the likelihood of malaria, but their performance varies by setting. Seven studies of imported malaria were identified. The presence of fever (LR, 5.1; 95% CI, 4.9-5.3), splenomegaly (summary LR, 6.5; 95% CI, 3.9-11.0), hyperbilirubinemia (LR, 7.3; 95% CI, 5.5-9.6), or thrombocytopenia (summary LR, 5.6; 95% CI, 4.1-7.5) make malaria more likely.

Conclusions In endemic areas, the likelihood of malaria is increased by the presence of splenomegaly and hepatomegaly but individual findings are of limited utility and cannot reliably exclude malaria; combinations of findings may be useful to stratify risk in patients. In returning travelers, the clinical assessment can provide substantial diagnostic benefit, although all patients still require laboratory testing because malaria can be rapidly fatal.