Abstract
BACKGROUND AND AIM: Breath testing (BT) is used to identify carbohydrate malabsorption and small intestine bacterial overgrowth. Measuring methane alongside hydrogen is advocated to reduce false-negative studies, but the variability of methane production is unknown. The aim of this study is to examine the effect of high methane production on hydrogen excretion after ingesting lactulose, fructose, or lactose. METHODS: A retrospective audit was performed of patients with gastrointestinal symptoms who underwent BT. Following a low fermentable carbohydrate diet for 24-h, a fasting BT before consuming 35 ml lactulose, 35 g fructose, or lactose in 200 ml water, followed by BT every 10-15 min for up to 3-h, was performed. A positive test was defined as a ≥20 ppm rise of hydrogen or methane from baseline. A high methane producer had an initial reading of ≥5 ppm. Breath hydrogen and methane production were measured as area under the curve. Chi-squared tests were used to compare proportions of those meeting the cut-off criteria. RESULTS: Of patients, 26% (28/106) were high methane producers at their initial lactulose test. The test-retest repeatability of methane production was high, with the same methane production status before ingesting lactose in all (70/70) and before ingesting fructose in most (71/73). Methane production was highly variable during testing, with 38% (10/26) having ≥1 reading lower than baseline. Hydrogen produced by high or low methane producers did not differ (1528 [960-3645] ppm min vs 2375 [1810-3195] ppm min [P = 0.11]). Symptoms and breath test results were not positively related. CONCLUSION: The validity of including an increase of ≥20 ppm methane to identify carbohydrate malabsorption or small intestine bacterial overgrowth should be questioned due to the variability of readings during testing.