A Multicentre Study on Behalf of the European Society of Paediatric Gastroenterology and Nutrition Working Group on Acute Diarrhoea

Abstract

The main aim in managing gastroenteritis in children is the correction of dehydration and maintenance of hydration using an optimal oral rehydration solution (ORS). In 1992 the European Society of Paediatric Gastroenterology and Nutrition (ESPGAN) working group on oral rehydration therapy (ORT) published guidelines for the optimal composition of ORS for Europe (1). The second aim in the management of these children is nutritional repair. The positive effect of continuing breast-feeding during treatment of gastroenteritis with ORS is well established (2). However, debate has continued on the optimal nutritional therapy of non-breast-fed patients, particularly in relation to the reintroduction of oral feeds and the need for specific dietary restrictions (3-19). In 1926 Powers (20) described the major procedures in the treatment of “intestinal intoxication” as administration of fluids, transfusion of blood, with-holding of food for a time, and the administration of food at the end of the period of starvation in gradually increasing amounts. Therapeutic starvation became the established practice, in addition to intravenous fluid therapy. This practice has been changed by oral rehydration therapy, which has gradually brought about the principle of rapid realimentation, with reintroduction of oral feeds immediately after adequate rehydration. The principal controversy in the nutritional therapy of acute gastroenteritis centres on the relative risks of cow's milk feeds (Table 1)(21). First, decreased lactase activity is chiefly associated with rotavirus gastroenteritis, and temporary lactose intolerance is considered an important factor in chronic diarrhoea (22). Second, infants who have rotavirus diarrhoea show an enhanced absorption of cow's milk antigens (23). Foreign proteins transported across an impaired mucosal barrier could evoke local or systemic immune response and food sensitisation (24). Until recently, the results of clinical studies from the 1940s and 1950s showing that infants and children treated with an early or free feeding regimen had better weight gain and no prolongation of diarrhoea compared with patients on starvation and gradual refeeding were overlooked (25,26), and children continued to be starved for 24-48 h (27). Several studies over the last 15 years have shown that there is no need for “regrading” slowly, and after initial ORT, full feeding appropriate for age is well tolerated in children >6 months of age with no adverse effects (6-10,14,18,28). When these clinical investigations are taken together, two risk factors, however, suggest caution: age (children <1 year) and diarrhoeal severity (more dehydrated and acidotic children have a higher rate and amount of vomiting and a higher number of stools) (29). The outcome of realimentation appears to be closely linked to the success of the initial ORT. AIMS The aim of this study was to compare the effect of ORS (conforming to ESPGAN ORS working group recommendations (1) and “early” or “late” feeding on the duration and severity of diarrhoea, weight gain, and complications (carbohydrate intolerance and vomiting) in weaned European infants. PATIENTS AND METHODS This was a multicentre study based in twelve European hospitals: Royal Hospital for Sick Children, Bristol, U.K.; Ospedale Maria Nuova, Reggio Emilia, Italy; Department of Clinical Sciences, University of Tampere, Finland; Departimento Universita Di Napoli, Napoli, Italy; The Children's AMC, Amsterdam; Groot Ziekengasthius, `s-Hertogenbosch, The Netherlands; Children's Hospital Zagreb, Republic of Croatia; Maribor Teaching Hospital, Ljubljanska, Slovenia; Department of Paediatrics, Charles University, Czech Republic; Antwerp Children's Hospital, Belgium; Hospital de S. Joao, Porto, Portugal; Katedra Pediatrii Akademii Medycznej, Warszawa, Dzialdowska, Poland; and Soroka Medical Centre, Beer Sheva, Israel. Weaned children under the age of 3 years who presented consecutively with acute diarrhoea (defined as passage of four or more watery stools per day for >1 but <5 days and requiring hospitalisation) were entered into the study. Children with short gut syndrome, chronic inflammatory bowel disease, ileus, previous treatment with antidiarrhoeal drugs, and associated hepatic or renal disease were excluded, as were children who had already been on oral rehydration solutions or were on intravenous fluids. Ethical Committee approval was obtained in each participating institution. Informed written consent from the parent or guardian was obtained after the nature of the study had been explained. The patients were randomly allocated to either the A. “early” feeding or the B. “late” feeding group. Patient details, medical history, and epidemiological history were recorded. Stool samples were collected for identification of microorganisms and for light and electron microscopy and enzyme-linked immunosorbent assay, for rotavirus, and urine and blood for electrolyte analysis. The patient was weighed and examined, the degree of dehydration was assessed, and the fluid deficit was calculated. The patient was then rehydrated over the following 4 h. Then, if allocated to group A, the patient was given his or her usual diet, and if allocated to group B, was given maintenance ORS only for the following 20 h and then fed the usual diet. In addition, both groups were offered ORS 10ml/kg/stool for each watery stool. If a child was breastfed, breast-feeding was to continue throughout, and in addition the child was given ORS and diet as appropriate. Body weight was recorded at 4 h and then twice daily until diarrhoea stopped and on day 5 and 14. Stool frequency and the type of stool (watery, loose, or formed) was recorded, as was the total duration of diarrhoea. Diarrhoea was considered to have stopped at the time of passage of the last watery stool. Oral intake (volume and nature) was recorded. Diarrhoeal stools were checked for reducing substances daily to day 5 and then on day 14 (lactose intolerance was defined as the presence of >0.5% reducing substances). If stools were still watery on day 5, stool microbiology was repeated. Treatment failure was deemed to have occurred if diarrhoea became worse after the patient's entry into the study and the patient became >5% dehydrated at 4-6 h. The patient was then withdrawn from the study and treated as appropriate by the clinician. Treatment failure was also recorded if the patient required intravenous fluids. Any patient could be withdrawn from the study if consent was withdrawn, there was a treatment failure as above, or the investigator considered that it was in the patient's best interest to be withdrawn. Where possible, if a patient was withdrawn, final evaluation was completed and data collected up to the time of withdrawal for analysis. Statistical Analysis The Analysis of Variance (ANOVA) and the χ2 test were used in statistical comparisons. ANOVA for repeated measures was used when comparing weight gain, presence of diarrhoea, and vomiting. RESULTS A total of 230 patients were enrolled in the study: 134 in the early feeding group (group A) and 96 in the late-feeding group (group B). The difference in the numbers entered into the two groups is mainly due to chance, although eight patients in the late-feeding group had to be omitted, as they were given food early. The patients were allocated to the two groups according to random numbers by the 12 centres concerned. Clinical Epidemiological Data The two groups were similar in relation to male/female ratio, birth weight (3,307 vs. 3,437 g), gestation, duration of breast-feeding, age at introduction of cow's milk, history of previous diarrhoea, duration of present diarrhoeal episode, presence of vomiting, degree of dehydration, body temperature (37.5 vs. 37.8°C) and prior treatment with non-ORS clear fluids and antibiotics (Table 2). The age at introduction of solids was lower in the non-feeding (B) group, 16 weeks compared with 19 weeks in the feeding (A) group (p = 0.01), and weight at first assessment was higher. There was no significant difference in the degree of dehydration of the two groups. Most of the patients had mild dehydration (Table 2). None were severely dehydrated. Stool Characteristics The most common pathogen was rotavirus; 71% in group B and 55% in group A (p = 0.04) (Table 3). Parasites were isolated in 3% of group A and nil in group B. Stool cultures were positive for bacteria in 10% of group A and 1% of group B. Salmonella was the commonest bacterium cultured, followed by Campylobacter and E. coli. Not all the centres were able to carry out the specialised test for specific bacterial pathogens, e.g., Campylobacter and Aeromonas. At entry, more patients in group A had blood in the stool (15%) than in group B (5%); p = 0.02. The incidence of lactose intolerance (>0.5% reducing substances in stool) was 6% in group A and 1% in group B. Rehydration Phase The amount of ORS taken over the first 4 h was similar in the two groups: a mean of 46 ml/kg (41-50 ml/kg) for group A and 49 ml/kg (45-52 ml/kg) for group B. There was no significant difference between the two groups in the number of watery stools or the presence of vomiting. Weight gain during this phase was also similar: 85 g for Group A and 77 g for Group B (p = 0.76). After Rehydration: 4-24 h During this period, mean weight gain was significantly higher in group A (95 g) than in group B (2 g) (p = 0.01) (Fig. 1). More ORS was consumed by the nonfeeding group (B), but the total amount of fluid intake was similar in the two groups. There was no significant difference between the two groups in plasma sodium, potassium, or bicarbonate levels. Weight Gain During Hospitalisation Net weight gain during hospitalisation was significantly higher in group A than in group B (p = 0.001) (Fig. 1). Complications There was no significant difference between the two groups in the incidence of vomiting or watery stools on days 1-5 (Figs. 2 and 3). No patients had lactose intolerance on day 5 and diarrhoea or vomiting persisting by day 14. Milk intake was significantly higher in group A on day 2 (p = 0.008) and day 3 (p = 0.02) but became equal by day 4. Weight gain was similar in both groups by day 5 and day 14 (Fig. 1). Treatment Failures Four patients in each group required intravenous fluids by day 4, none of these patients had significant lactose intolerance. DISCUSSION The results of this multicentre study show that the complete resumption of a child's normal feeding, including lactose-containing formula, after 4 h of rehydration with glucose ORS (ESPGAN recommended composition) led to significantly higher weight gain after rehydration and during hospitalisation and did not result in worsening of diarrhoea, prolongation of duration of diarrhoea, increased vomiting, or lactose intolerance compared with the late feeding group who were given normal feeding after they had received ORS for 24 h. Conventional treatment of acute gastroenteritis in Europe has included a period of starvation, usually 24 h but often prolonged (27). Recent data suggest that there are no advantages to regrading infants with gastroenteritis (6-10,14,18). It has also become clear that withdrawal of food during the management of gastroenteritis is not beneficial. This study, involving a range of hospitals from around Europe, adds considerable weight to this evidence. In malnourished children the nutritional benefits of early feeding have been well illustrated (14). Most of the children in our study were not nutritionally compromised when they developed gastroenteritis; nevertheless, as some of them lost weight, they may gain nutritional advantages from nutritional supplements in addition to ORS. The incidence of lactose intolerance in this study on weaned children appears to be extremely low: 6% and 1% in groups A and B before entry into study and nil on day 5 and day 14. On the basis of these data, the restriction of lactose-containing foods for the vast majority of children with gastroenteritis does not appear to be justified. This study provides clear evidence that continuation of feeding during gastroenteritis has advantages for the children of Europe. Optimal management of mild to moderately dehydrated children should consist of oral rehydration with ORS over the first 4 h followed by resumption of normal feeding. Breast-feeding should be continued. Further dehydration can be prevented by supplementing the usual feeds with ORS (10ml/kg/liquid stool) as necessary.FIG. 1: . Weight gain.FIG. 2: . Frequency of vomiting.FIG. 3: . Frequency of watery stools.