Pii: S0031-9384(00)00273-0

A significant decrease in the bacterial count of small intestinal mucosa has been observed in children with recurrent diarrhea or abdominal pain in the time that has elapsed from the previous meal. Humans may be trained to recognize metabolic feelings of hunger that are associated with a steady and slightly lower glycemia than baseline, between 4.7 and 3.9mmol/L (intervention). An eating habit associated with a decrease in preprandial glycemia prevented diarrhea relapses, and was expected to impair intestinal microflora growth, including Helicobacter pylori in the stomach. The development of Helicobacter pylori infection might be prevented during childhood, and recovery from infection may be expected with intervention. The improvement in attention to metabolic feelings consisted of acquiring a predictive ability of glycemia by distinction between unsolicited hunger feelings (metabolic hunger) and those associated with external cues. Matching intake to the inbetween energy needs served to predict the subsequent emergence of the metabolic hunger. The matching was further compensated for the early or late emergence of metabolic hunger at the subsequent meals. Fruit and vegetables were increased to avoid abrupt glycemia lowering. This intervention was trained in 5-month periods. Subjects (209, 44, and 58) completed their training during 4-year periods between 1982 and 1994, and were enrolled in a prospective, controlled, randomized, interventional, preventive, and cohort study. The “prevention” hypothesis was tested in a subgroup of 86 healthy infants who were recalled in the years 1996 to 1998. A “recovery” study of approximately a 1-year intervention was investigated in 47 healthy subjects between ages 5 and 25, who were positive for antiH. pylori and had no need for an immediate antibiotic treatment at entry. The following behavioral factors were recorded in a 7-day home diary and calculated: the fraction of meals induced by metabolic hunger out of 21 main mealtimes; average preprandial glycemia (DAP glycemia); daily intakes, activity; and bedtime hours. The decrease in preprandial glycemia was the objective measure of compliance with the recognition of “metabolic” hunger. Anthropometric measures and blood tests were obtained for nutritional and functional verifications. Average preprandial glycemia was 8.5 and 8.6% lower in the intervention groups than the control groups in the “prevention” and “recovery” studies, respectively, at the end of follow-up ( p , 0.05 and , 0.001, respectively). A 4.7% seroprevalence of H. pylori infection was observed in the intervention group, with 30.2% in the control group at a mean age of 10 years after approximately an 8-year follow-up in the “prevention” study ( p , 0.0005). The seroprevalence decreased to 9 of 24 (37.5%) under intervention as opposed to 20 of 23 controls (87%) in the recovery study ( p , 0.002). A significant positive correlation was found between DAP glycemia and the antiH. pylori serum antibody concentration ( r 5 0.52; p 5 0.0002). A decrease in the level of immune stimulation by H. pylori infection was observed due to the intervention, which may have a preventive and therapeutic role on the infection. © 2000 Elsevier Science Inc. All rights reserved.