POTENTIAL USEFULNESS OF HYDROGEN BREATH TEST WITH D-XYLOSE IN CLINICAL MANAGEMENT OF INTESTINAL MALABSORPTION

Hydrogen breath tests (H-2 BT) have been used extensively to investigate intestinal dissacharidase deficiencies. A potentially useful test for assessing intestinal absorptive function, the H-2 BT with D-xylose (H-2 BT-xylose), has received scant attention. We report here the results of our investigation of this test in 45 patients. Fifteen patients had proved malabsorption that was due to nontropical sprue in nine, and to lymphoma, Whipple's disease, or giardiasis in the remainder. Nine patients had small-bowel bacterial overgrowth secondary to either postsurgical sequelae or intestinal dysmotility. Twenty-one patients with irritable bowel syndrome and 21 healthy individuals served as control groups. All participants ingested 25 g of D-xylose, and alveolar breath samples were obtained thereafter at 30 min intervals for 5 hr. Breath H-2 was measured by chromatography. Basal H-2 production, peak change (DELTA) and area under the curve (AUC) were calculated. Simultaneously, 5-hr urinary excretion of D-xylose was measured by colorimetry and served as the reference test. In healthy individuals, D-Xylose ingestion increased H-2 production (DELTA = 5.8 +/- 1.4 ppm, P < 0.001). Changes were similar in patients with the irritable bowel syndrome. In contrast, the increase was of a much greater magnitude in the malabsorption group (DELTA = 49.9 +/- 7.2 ppm, P < 0.001 vs healthy controls). AUC analysis yielded comparable results. Test performance analysis showed that, in malabsorption the H-2 BT-D-xylose had a sensitivity index of 0.86, which was identical to that of the urinary D-xylose test. Specificity was 1 and 0.95, respectively; and predictability 1 and 0.93, respectively. All patients who responded to treatment normalized their H-2 production, whereas those who did not respond maintained their high H-2 production. In the bacterial-overgrowth group, the H-2 BT-D-xylose was only positive when the urinary excretion of D-xylose was positive (five patients), whereas that three of the remaining four patients with normal urinary excretion of D-xylose also had a normal breath test. We conclude that the hydrogen breath test with D-xylose is a useful, valid, and practical test for the diagnosis and follow-up of malabsorption.