From the Editor

Abstract

EDITOR,-In their paper on the basis of viscosimetric studies performed on human gastric mucus Markesich et al challenge the concept that Helicobacter pylon exerts unfavourable effects on the mucous layer coating the gastric mucosa (Gut 1995; 36: 327-9). Their results are clearly at variance with the bulk of the experiments carried out in this areal and Markesich and colleagues try to show the reasons for this discrepancy by pointing out the potential drawbacks of the previous studies. Even if we accept Markesich's concept, however, that the results of some experiments on H pylori have to be discarded because they are not performed on human mucus, other data are consistent with a mucus damaging effect of the micro-organism in humans. Assessing the quality of gastric soluble mucus in humans we found that, although on the whole mucus is normal both in chronic gastritis and in duodenal ulcer,2 the lowest values are detectable in subjects with H pylori infection.3 Sidebotham et al also observed a breakdown of gastric mucus in patients infected with Hpylori, although they ascribed this phenomenon to a carbonate-bicarbonate buffer at the mucosal surface due to the urease activity of the germ, rather than to a direct mucolytic activity.4 Measurements of the gastric mucus gel thickness in vivo showed that the mucous coating is significantly thinner in H pylori infected patients,5 possibly resulting from increased erosion. In keeping with the above results H pylori infection seems to reduce mucosal hydrophobicity of human gastric mucus, which retums to normal after successful eradication.6 The reason for the discrepancy of Markesich's results is probably the method used to examine mucus. Reference is made to a study7 to justify the use of viscosity analysis for assessing the characteristics of mucus, but it is well recognised that viscosimetry is an unreliable technique for evaluating mucus properties.8 The finding that gastric mucus viscosity is increased in duodenal ulcer patients7 is only a further demonstration of how misleading the method can be. As only luminal mucus was examined in Markesich's study, it must be also considered that proteolytic enzymes produced byH pylori could promote a greater peptic erosion of the adherent mucus gel with consequent increase in the mucoprotein content of gastric juice influencing viscosity measurements. Eradication ofH pylon would clearly reduce mucus shedding and thus apparently decrease the viscosity of intraluminal mucus. M GUSLANDI Gastroenterology Unit, S Raffaele Hospital, University ofMilan, Italy