Comparison of the 1-gram (14)C-D-xylose breath test and the 50-gram hydrogen glucose breath test for diagnosis of small intestinal bacterial overgrowth.

BACKGROUND/AIMS: Culture of small bowel aspirate is the most direct method and the gold standard for diagnosing small intestinal bacterial overgrowth. However, cultures are cumbersome and fluoroscopy is required for obtaining aspirate. Therefore, different breath tests such as the xylose breath test and the hydrogen breath test have been developed. There is no general agreement as to which test is to be preferred. In the only previous direct comparison between these two tests an advantage for the 1-gram-(14)C-D-xylose breath test was found. The aim of the study was to compare the 50-gram glucose hydrogen breath test and the 1-gram (14)C-D-xylose breath test in relation to results of cultures of small bowel aspirate.

METHODS: Forty-six consecutive patients, mean age 57 (range 27-87) years, 12 men and 34 women, were included because of suspicion of small intestinal bacterial overgrowth. After small bowel aspiration, all patients received a solution of 1 g xylose, labelled with 50 microg (14)C-D-xylose, and 50 g glucose dissolved in 250 ml water. The concentration of breath hydrogen was analyzed every 15 min for 2 h and (14)CO(2) was analyzed every 30 min for 4 h. A positive hydrogen breath test was defined as a rise in hydrogen concentration of 15 ppm. A positive xylose test was defined as an accumulated dose 4.5% after 4 h. Two definitions for a positive culture were used, either growth of 10(5 )colonic-type bacteria/ml or growth of 10(5) bacteria/ml of any type.

RESULTS: Twenty-four patients had growth of 10(5) bacteria, of whom 10 had growth of 10(5) colonic-type bacteria in small bowel aspirate. Twenty-two patients had no significant growth. The hydrogen breath test and the xylose breath test had a sensitivity for growth of 10(5) bacteria of 58 and 42%, respectively. For growth of 10(5 )colonic-type bacteria the sensitivity was 90% for the hydrogen breath test and 70% for the xylose breath test. The specificity was similar for the two tests.

CONCLUSION: Although no significant difference between the two tests was found, there was a tendency in favor of the 50-gram glucose hydrogen breath test. The simplicity in combination with high sensitivity makes the hydrogen breath test suitable as a screening method to select patients for further investigation.