About 5% of normal subjects fail to produce increased hydrogen breath concentration after ingestion of the non-digestible carbohydrate lactulose (low hydrogen producers). The existence of low hydrogen producers limits the diagnostic use of hydrogen (H2) breath tests. We studied the effects of lactulose and of magnesium sulphate (MgSO4) pretreatment on stool-pH and on hydrogen exhalation after oral loading with lactulose or lactose in 17 hydrogen producers and 12 low hydrogen producers. In seven hydrogen producers acidification of stool pH by lactulose pretreatment (20 g tid) decreased hydrogen exhalation and three of seven (43%) became low hydrogen producers. In contrast, after pretreatment of eight low hydrogen producers with magnesium sulphate (5 g twice daily) all eight produced hydrogen after a lactulose load. Similarly four lactose intolerant low hydrogen producers had abnormal lactose hydrogen breath tests after MgSO4 pretreatment. MgSO4 pretreatment neither resulted in false positive lactose hydrogen breath tests in five lactose tolerant hydrogen producers, nor increased the hydrogen exhalation in five additional hydrogen producing controls after ingestion of lactulose. The results of these studies confirm that hydrogen production from lactulose decreases when the colonic pH is lower (lactulose pretreatment), and increases when colonic pH is higher (MgSO4 pretreatment). In low hydrogen producers the lacking increase of H2 exhalation after ingestion of non-digestible carbohydrates can be overcome by MgSO4 pretreatment, thus increasing the sensitivity of the test by avoiding false negative hydrogen breath tests in low hydrogen producers with disaccharide malabsorption or maldigestion. The underlying mechanism of this remarkable effect of MgSO4 pretreatment warrants further investigation.