Breath tests: principles, problems, and promise.

Breath tests rely on the measurement of gases produced in the intestine, absorbed, and expired in the breath. Carbohydrates, such as lactose and sucrose, can be administered in physiologic doses; if malabsorbed, they will be metabolized to hydrogen by colonic bacteria. Since hydrogen is not produced by human metabolic reactions, a rise in breath hydrogen, as measured by gas chromatography, is evidence of carbohydrate malabsorption. Likewise, a rise in breath hydrogen marks the transit time of nonabsorbable carbohydrates such as lactulose through the small intestine into the colon. Simple end-expiratory interval collection into nonsiliconized vacutainer tubes has made these noninvasive tests quite convenient to perform, but various problems, including changes in stool pH, intestinal motility, or metabolic rate, may influence results. Another group of breath tests uses substrates labeled with radioactive or stable isotopes of carbon. Labeled fat substrates such as trioctanoin, tripalmitin, and triolein do not produce the expected rise in labeled breath CO2 if there is fat malabsorption. Bile acid malabsorption and small intestinal bacterial overgrowth can be measured with labeled cholylglycine or cholyltaurine. Labeled drugs such as aminopyrine, methacetin, and phenacetin can be used as an indication of drug metabolism and liver function. Radioactive substrates have been used to trace metabolic pathways and can be measured by scintillation counters. The availability of nonradioactive stable isotopes has made these ideal for use in children and pregnant women, but the cost of substrates and the mass spectrometers to measure them has so far limited their use to research centers. It is hoped that new techniques of processing and measurement will allow further realization of the exciting potential breath analysis has in a growing list of clinical applications.