Eosinophilia After Three Countries and No Single Obvious Cause
A wildlife photographer returns from three months across Botswana, Zambia, and Uganda with no symptoms at all, just an eosinophil count flagged on routine post-travel screening and a trip full of genuinely distinct parasitic exposures. Nothing points clearly toward one diagnosis — the debate is whether to treat broadly now or test first and treat what's actually found.
A.K., a 38-year-old wildlife photographer, just wrapped up three months on assignment across Botswana, Zambia, and Uganda — a trip that had her wading barefoot along riverbanks in the Okavango Delta chasing a shot of a fish eagle, canoeing for four days down a tributary of the Zambezi she now admits she probably shouldn't have swum in afterward to cool off, and finishing with three weeks in a rural homestay outside Kampala with no reliable access to clean water or closed shoes. She feels completely fine. The only reason she's in this office at all is that her travel clinic recommends a post-travel screening panel for anyone returning from an extended stay like hers, and that panel came back with an absolute eosinophil count of 1,200 cells/µL — clearly elevated, and clearly not pointing toward any single explanation.
That's the actual problem the visit has to solve. Unlike a returning traveler with one dominant exposure and a matching symptom, A.K.'s itinerary hands the team at least three genuinely distinct, independently plausible causes: freshwater contact in a schistosomiasis-endemic stretch of the Zambezi system, barefoot soil contact consistent with strongyloidiasis or hookworm, and a broader rural homestay exposure that could in principle explain any number of other helminths. The British Infection Association's UK recommendations on eosinophilia in returning travelers, updated by Thakker and colleagues from the earlier Checkley guidance, favor a testing-driven approach for exactly this kind of undifferentiated finding: a travel-history-guided serology and stool panel built to identify what is present rather than treat blind. Her own dates are what complicate applying it. Schistosoma serology does not reliably turn positive until six to eight weeks after exposure, and her Zambezi swim sits close enough to that boundary that a negative result drawn this week would settle nothing, while the Okavango wading that could have given her Strongyloides is three months back and well inside the window. One test is being read too early and the other is not, from the same blood draw on the same afternoon.
Two drugs, two different clocks, one asymptomatic patient
I'd treat empirically now — sequential therapy covering the two highest-stakes possibilities — rather than lean entirely on a testing panel with real sensitivity gaps for exactly this exposure pattern. Strongyloides serology can be falsely negative in low-burden infection, and Schistosoma serology often isn't reliable for another several weeks. And by her own account, she's not someone who reliably comes back for a scheduled follow-up between assignments.
If she were the kind of patient I could count on seeing again in six weeks, I'd feel differently. I don't think I can count on that here.
Testing-driven management is the actual current standard for a finding like hers, and I don't think an uncertain follow-up history is enough reason to abandon it. Treating her empirically for two different parasites when she's genuinely asymptomatic and may have neither infection is real drug exposure without a confirmed target — and it also risks masking whatever the actual finding turns out to be if she does need to be re-evaluated later.
I take the follow-up concern seriously, but the answer to "she might not come back" is a better follow-up plan, not treating without knowing what we're treating.
I don't think this actually has to be one strategy applied to both possibilities equally. Ivermectin and praziquantel don't carry the same risk, and Strongyloides and Schistosoma don't carry the same urgency or the same testing timeline. Give her ivermectin today — low risk, addresses the one possibility with genuinely catastrophic downside if missed — and send the full panel, but hold praziquantel specifically until Schistosoma serology actually says something, given how timing- sensitive that drug already is against early-stage infection.
That's not splitting the difference for its own sake — it's matching each drug's real risk profile to each diagnosis's real urgency, instead of treating "empiric versus testing" as one decision that has to apply the same way to both.
Agreed: ivermectin given today, the full serology and stool panel sent in parallel, and praziquantel explicitly held pending the Schistosoma serology result rather than given empirically.
Not fully agreed: the infectious disease physician would still have preferred withholding even the ivermectin until Strongyloides serology results, and said so directly, viewing today's plan as a reasonable compromise rather than her own first choice. The travel medicine physician's own preference, unmoved, was to have treated both possibilities empirically today given the follow-up uncertainty, and accepted the asymmetric plan as an improvement over waiting on everything rather than a fully satisfying answer. Both agreed the plan actually adopted was better than either of their original positions taken alone.