Determining the cause of eosinophilia in patients returning from the tropics continues to present a diagnostic challenge. The history, symptoms and degree of eosinophilia are often poor predictors of eventual diagnosis, but helminths are an important cause. The current British Infection Association recommendations use travel history to guide investigation of eosinophilia. However the global burden of helminth disease and travel patterns have changed over the last 3 decades and guidelines based on previous epidemiology need to be reviewed in the light of current data.

Consecutive patients presenting with, or referred for, investigation of eosinophilia were identified prospectively. Case notes, laboratory results and electronic records were reviewed for demographic and clinical data. Patients with an eosinophil count ≥0.50 × 10⁹/L were included, and grouped based on lifetime history of travel to: West Africa, elsewhere in Africa, and the rest of the world. Results were compared to published data from 1997 to 2002 collected at the same centre.

Of 410 patients who met the inclusion criteria, 407 had a documented travel history. Average yearly referrals for eosinophilia fell from 58 per year between 1997 and 2002, to 33 per year (2002–2015). The proportion of eosinophilia cases diagnosed with a parasitic cause fell from 64% to 50%, and yields for all parasitological investigations fell, the largest reduction in stool microscopy (20% yield to 9%) and day bloods for microfilariae (14% yield to 3%). Strongyloides stercoralis was the commonest diagnosis overall in our cohort, accounting for 50% of the total parasites diagnosed, and was present in 38% of patients from West Africa, 19% from rest of Africa, and 34% from rest of world; a relative increase compared to previous data. Schistosomiasis is slightly less common in those who had travelled to West Africa than the rest of Africa, and overall point prevalence has fallen from 33% (1997–2002) to 17% (2002–2015). Travellers were significantly less likely than patients who had immigrated to the UK to be diagnosed with any parasite (OR 0.54 95% CI 0.378–0.778 p = 0.0009).

A parasitic cause will still be found in half of people returning from the tropics with an eosinophilia, but we observed a fall in the overall prevalence of parasitic diagnoses when compared with the earlier data. This may, in part, be explained by the impact of control programmes on the prevalence of parasites globally, especially filarial disease. S. stercoralis now represents the majority of parasites diagnosed in our cohort from all continents. We identified significantly higher rates of strongyloidiasis in immigrants than returning travellers. Despite the falling yields of stool microscopy and filarial serology the current guidelines based on travel history remain relevant with adequate yield.