To date, there is no general agreement as to which test is to be preferred for the diagnosis of small-intestinal bacterial overgrowth. The 1-g carbon-14 D-xylose breath test has been proposed as a very sensitive and specific test for the diagnosis of bacterial overgrowth. However, in patients with severe gastrointestinal motor dysfunction, the lack of consistent delivery of 14C-D-xylose to the region of bacterial contamination may result in a "negative" result. The aim of this study was to determine whether the accuracy of 14C-D-xylose breath test for detecting bacterial overgrowth can be increased by correction with the gastric emptying rate of 14C-D-xylose. Ten culture-positive patients and ten culture-negative controls were included in the study. Small-intestinal aspirates for bacteriological culture were obtained endoscopically. A liquid-phase gastric emptying study was performed simultaneously to assess the amount of 14C-D-xylose that entered the small intestine. The results of the percentage of expired 14CO2 at 30 min were corrected with the amount of 14C-D-xylose that entered the small intestine. There were six patients in the culture-positive group with a 14CO2 concentration above the normal limit. Three out of four patients with initially negative results using the uncorrected method proved to be positive after correction. All these three patients had prolonged gastric emptying of 14C-D-xylose. When compared with cultures of small-intestine aspirates, the sensitivity and specificity of the uncorrected 14C-D-xylose breath test were 60% and 90%, respectively. In contrast, the sensitivity and specificity of the corrected 14C-D-xylose breath test improved to 90% and 100%, respectively. In conclusion, using the gastric emptying rate of 14C-D-xylose as a correcting factor, we found a higher sensitivity and specificity for the 14C-D-xylose breath test in the detection of small-intestinal bacterial overgrowth than were achieved with the conventional method.