MATER IALS AND METHODScourse of diarrh ea. Maln utrition has been report ed to increase the absorption of potentially harmful antigens (2) and to impai r imm une responses (3). A poor nutrition al state may thu s increase the risk of protracted diarrh ea as well as enhance susceptibility to other infectio ns and to gastroi ntestinal allergy. The nutritio nal state of even well-no urished infants deteriorat es rapidly during diarrhea, either because of loss of appetite, deliberate withholding of oral feeds, or partial malabsorption caused by viral invasio n ofenterocytes (4).Rapid reintrodu ction of oral feedings after rehydration has been advocated to coun terac t the pote nti al hazards related to fasting during diarrhea (5). We have previously shown that rapid refeeding results in earlier cessatio n of diarrhea in well-nourished children; also, cow milk produ cts are tolerated (6). A furt her shortening of diarrhea resulted from administratio n of hu man Lactobacillus stra in (Lactobacillus GG, Gefilac, Valio Finni sh Co-operative Dairies' Associatio n, Helsinki, Finland ) together with the rapid refeeding schedule (7). The mechanisms behind such a favorable outc ome remain poorly understood. Th e effect of nutritional therapy may be immunologically med iated and may prove important in eradicating enteric infections in the imm unocompromised host.The objective of the present study was to evaluate the effect of L actobacillus GG on the intestinal immune response triggered by rotavirus infection in well-nourished children. For th is pu rpose, we used the ELISPOT assay, which measures ISC and sASC amo ng circulating lymphocytes. T hese cells are arrested du ring their maturation cycle in peripheral blood, giving indirect evidence of gut local im m une respo nse (8-10).Patients. Forty-four well-no urished children (33.4% female), between 7 and 37 mo of age, were enro lled in the present study. Th ey were admitted for acute gastroen teritis of less than 7 d du ration at the Departm ent of Pediatrics, Ta mpere U niversity Hospital, during a rotavirus epidemic .Informed consent was obtaine d from the patients' pare nts, and the protocol was approved by the hospital's Committee on Ethical Practice.Ma nagement and samples. U pon ad mission, the children were weighed and clinically examin ed. The severity of dehydration (%) was estimated. Th e children were treated according to sta ndard practice: oral rehyd ration was accom plished in 6 h with a , solution containing Na" 60 mmo l/ L and glucose 144 mm ol/L (Osmosal Novum, Leiras, Turku , Finland) given at twice the estimated fluid loss with a minimum of 30 mL/kg body weight. Th e patients were weighed daily. Th e attending nurs es followed the quality (characterized as watery, loose, or solid) and number of stools and vomitus. The du ration of diarrh ea was counted 14 1