Investigation of tropical eosinophilia; assessing a strategy based on geographical area.

Abstract

Patients with eosinophilia are an important clinical problem. This study aimed to assess the most efficient manner of investigating patients with peripheral eosinophilia (eosinophil count >0.5x10(9)ml(-1)) presenting from the tropics. Patients attending the Hospital for Tropical Diseases, London, from October 1997 to March 2002 for investigation of eosinophilia were identified prospectively. Laboratory, clinical and demographic data were recorded from laboratory and clinical records. An investigation set was proposed prospectively and assessed for all geographical areas (stool microscopy, strongyloides culture and serology), all of Africa (additional schistosomal serology, terminal urine microscopy and filarial serology) and West Africa (additional day-bloods for microfilaria). Data was analysed for 261 patients. At least one helminthic cause for eosinophilia was found in 64% of patients (median eosinophilia 1.2x10(9)ml(-1)). Seventeen per cent of patients had more than one helminth species found. Median eosinophilia increased with number of diagnoses per patient. The proposed investigation sets were validated, with high yield for all proposed tests apart from filarial serology outside West Africa, and good sensitivity. Initial investigation of eosinophilia in patients presenting from the tropics may be guided by a simple investigation set depending on broad area of travel which has high sensitivity and yield. Patients frequently have more than one helminthic cause of eosinophilia.