Background: The gold standard in diagnosis of malaria is microscopic detection of malaria parasites in thin blood smears. However, the sensitivity and specificity of blood smears depend mostly on the experience of the examiner. In the traveler returning from the tropics, diagnosis of malaria may be difficult when the parasitemia is low. In this circumstance any indicator that suggests the diagnosis of malaria is of great interest. The aim of this study is to determine the value of hypocholesterolemia to the diagnosis of imported malaria.
Method: A retrospective case-control study was performed among hospitalized patients with fever returning from a malaria-endemic area, to compare the results of routine biological parameters of 129 malaria cases with those of 92 control patients.
Results: Multivariate analysis, using a logistic regression model demonstrates that hypocholesterolemia was the strongest parameter associated with malaria (adjusted odds ratio [OR]: 75.22, 95% confidence interval [CI] 4.60-1227.78) and the most specific (98%, 95% CI 0.95-1.0). The most sensitive abnormality was thrombocytopenia (82%, 95% CI-0.77 0.87). With a malaria prevalence of 52% in our population study, hypocholesterolemia has the strongest positive predictive value among routine biological parameters for malaria diagnosis (96%). The combination of hypocholesterolemia and thrombocytopenia was always associated with diagnosis of malaria in this study.
Conclusion: These results show that hypocholesterolemia is significantly associated with malaria. Therefore, in the setting of negative thin and thick blood smears, the presence of hypocholesterolemia, particularly when it is combined with thrombocytopenia, in a febrile traveler returning from a malaria-endemic area, should prompt repetition and careful analysis of blood smears to avoid misdiagnosis.