The aim of the study was to determine whetherthe specificity of the [14C]d-xylose breathtest could be improved, by excluding false-positivetests due to premature colonic metabolism of the[14C]d-xylose caused by rapid colonictransit. Forty-seven patients with suspected small bowelbacterial overgrowth were investigated by (1) aspirationand culture of duodenal fluid and (2) a[14C]d-xylose breath test. Those with either a positiveduodenal culture or breath test had a repeat[14C]d-xylose breath test given with one ofthree transit markers (barium, Gastrografin or99mTc-labeled tin colloid) to determine if the site of metabolism was inthe small bowel or colon. Fourteen patients had positiveduodenal cultures, four of whom had a negative[14C]d-xylose breath test, 15 patients had apositive [14C]d-xylose breath test, three ofwhich were due to colonic metabolism of the xylose.Where transit markers were used, 14C wasdetectable in the breath and serum before barium hadentered the small bowel, thus the barium did not comigrate withthe xylose. Gastrografin accelerated small boweltransit, leading to malabsorption of the xylose in thesmall intestine and subsequent colonic metabolism of the xylose. 99mTc-labeled tincolloid had no obvious disadvantages and appeared to bethe marker of choice. The use of a transit markerincreased the specificity of the[14C]d-xylose breath test from 85% to 94%. The specificity of the[14C]d-xylose breath test for the detectionof small bowel bacterial overgrowth is improved togreater than 90% by the use of an appropriate transitmarker.