M.von Ungern-Stenberg ¹*, M.Rautio ¹, K. Ariniemi ¹, U. Nieminen ², R. Korpela ³, T. Vesa ³, M. Saxelin ³, A. Siitonen ¹, M. Färkkilä ², and H. Jousimies-Somer ^1
1 National Public Health Institute, ² Helsinki University Central Hospital, ³ Valio Research Center, Helsinki, Finland

Despite low lactase activity, in some lactose maldigesters, the oral intake of even considerable amounts of lactose does not necessarily lead to symptoms. The aim of this study was to investigate whether the gas produced by fecal contents of lactose maldigesters is traceable to symptoms of intolerance.

Frozen stool samples from 21 patients with biopsy proven hypolactasia, 10 with severe and 11 with mild symptoms (cumulative score), were used to prepare fecal suspensions (0.2%) into three different carbohydrate PY broths (1% lactose, glucose and galactose). After incubation (24 h. 35°C), H₂, CO₂, air and methane contents were analysed in triplicate by gas chromatography.

Gas production was best demonstrated in galactose broth. Differences between patient groups with mild and severe symptom scores were found; the patients with severe symptoms in average produced 5.1% H₂ (share of the bottle's head space gas volume) and 10.1% CO₂, compared with 3.7% H₂ and 8.8% CO₂ produced by patients with mild symptoms. However, this difference was not statistically significant. Substantial intragroup variations in gas production among patients with severe (range of H₂ in galactose from 0.9% to 8.3% and CO₂ from 5.9% to 15.9%) as well as with mild symptoms (H₂ from 0% to 8.3% and CO₂ from 1.4% to 14.5%) were recorded. Three patients produced detectable amounts of methane (two with severe and one with mild symptoms). The individual symptoms i.e. flatulence or abdominal pain did not directly correlate with the gas production, but abundant gas production had a tendency to associate with greater difference in waistline measurements before and after lactose provocation. Preliminary comparisons between fecal H₂ and CO₂ production and bacterial groups isolated from the samples showed that high concentrations of E. coli tended to be more often associated with high gas production. Numbers of lactobacilli and H2 concentrations had a trend for negative correlation.

In this relatively small patient material we could not demonstrate a clear association between intestinal microflora-derived gas production and any single symptom. Individual physiologic differences in elimination of gases through circulation and expiration may partly explain the lack of correlation of gas production and symptoms. Whether the observed E. coli -associated gas production correctly depicts the situation in vivo or whether it is an artefact e.g. affected by freezing of the sample and diminishing the viability of anaerobic bacteria deserves further comparative studies using fresh and frozen samples in parallel.