Tropical Sprue in a Returning Traveler: How Long Does the Antibiotic Actually Need to Run
A single patient, home three months from an assignment that ended over a year ago, with a chronic diarrhea nobody could source until someone finally asked the right travel question. The disagreement is over how long a fifty-year-old treatment regimen, built on a different population, actually needs to run for her.
Anna K., a 39-year-old aid worker, spent fourteen months running a maternal health program in rural India before returning home just over a year ago, and had assumed the loose, greasy stools that started near the end of that posting were just her gut readjusting to a different diet. They never fully stopped. Three months ago they worsened — four to six watery stools daily, real weight loss she couldn't attribute to anything else, and a fatigue heavy enough that she finally saw a gastroenterologist rather than continuing to wait it out. Celiac serology was negative, stool studies found no ova, parasites, or Giardia antigen, and a duodenal biopsy showed partial villous blunting with chronic inflammation — a nonspecific picture that, paired with her travel history and the exclusion of celiac disease and infectious causes, fits tropical sprue.
Her labs added a second finding that shapes the treatment plan as much as the diagnosis itself: a macrocytic anemia with a low folate and a low B12, reflecting injury across both proximal small bowel, where folate absorption concentrates, and distal ileum, where B12 absorption depends on intact terminal-ileal function. The antibiotic-duration question in front of the team turns partly on which population's literature actually describes her. Much of the classic three-to-six-month treatment course comes from older, uncontrolled case series in long-term expatriate residents of endemic regions, a genuinely different exposure pattern than a fourteen-month posting followed by over a year back home before diagnosis. Current guidance has moved away from that framing without ever testing it directly: the AGA's clinical practice update on refractory celiac disease, which addresses tropical sprue as a differential, describes the disease as responding rapidly to folic acid and tetracycline rather than requiring a fixed months-long course. Nobody has run the trial that would settle the duration question in returning travelers specifically, so the shorter course is an inference from her exposure pattern, not a finding anyone can cite for her. Nothing about that duration question, though, settles her B12 repletion timeline on its own — ileal recovery and jejunal recovery are not guaranteed to track the same curve, and her B12 stores, already measurably depleted, may need monitoring and replacement well past whatever point her diarrhea and folate levels normalize.
GI clinic, new tropical sprue diagnosis
I want to start doxycycline and folic acid for a full three to six months. Tropical sprue is a slow-healing enteropathy, and the older descriptive literature is built around treating until mucosal healing is genuinely complete, not just until symptoms improve. Stopping early on the strength of how she feels risks a relapse we won’t see coming until the villous injury has already progressed again.
I’d treat for shorter, closer to four to six weeks, and I want to be specific about why: the long-course literature largely describes long-term expatriate residents of endemic regions, a genuinely different exposure pattern from a fourteen-month posting followed by over a year back home before we even made the diagnosis. And the current AGA practice update describes tropical sprue as responding rapidly to folate and tetracycline — not as demanding a fixed months-long course. I want to be honest that no trial has tested duration in returning travelers specifically, so what I'm proposing is an inference from her exposure pattern rather than a result I can cite for her. But that cuts both ways: the long course isn't a tested result for her either, it's an inference from a population she isn't in.
I’m not arguing mucosal healing doesn’t matter — I’m arguing the assumption that HER healing timeline matches the long-term-resident literature is exactly the population mismatch worth naming out loud before defaulting to the longer course.
I think the antibiotic-duration debate, however it resolves, is only answering half of what she actually needs. Her B12 deficiency reflects ileal involvement, and ileal recovery doesn’t necessarily run on the same clock as the jejunal recovery her folate level reflects. Whatever course length you settle on for the antibiotic, I don’t think her B12 repletion and monitoring plan should just inherit that same endpoint by default.
I’d set B12 as its own tracked variable — replete now, recheck levels independently of the antibiotic-duration decision, and keep monitoring past whatever point her diarrhea resolves and her doxycycline course ends, since a normalized folate and a resolved stool pattern wouldn’t actually tell us her ileum has caught up.
Agreed: start doxycycline and folic acid today; B12 repletion begins in parallel with its own independent recheck schedule, not tied to the antibiotic course length.
Not agreed: the actual antibiotic duration. The infectious disease physician wants the full three-to-six-month classic course; the gastroenterologist wants a four-to-six-week course reassessed at that point, specifically because he reads the long-course evidence as describing a different population than hers.