Oral Rehydration: An Evolving Solution

Fluid and electrolyte balance is vital for good health at all ages. Water comprises 75% of the body weight of a neonate. However, during the first 3 to 5 days of life, 5% to 10% of body weight is lost because of a loss of extracellular fluid. Total body fluid decreases gradually to about 60% of total body weight by 1 year of age. The limited ability of the neonatal kidney to excrete and reabsorb water and the impaired ability to excrete sodium compromise fluid and electrolyte homeostasis in the neonate. However, the neonate born at term can conserve sodium through tubular reabsorption. Fluid requirements change dramatically during the early neonatal period, through childhood, and into adult life. Fluid requirements when expressed as a proportion of body weight are greatest during infancy, when the child is most susceptible to fluid losses. Fluid losses from the gastrointestinal tract are the most profound and clinically devastating. Infants and young children are susceptible to intestinal infections, particularly those that cause watery diarrhea. The discovery that oral administration of oral rehydration therapy (ORT), a glucose-electrolyte solution, can rehydrate and maintain hydration in the infant with acute infective diarrhea is one of the greatest scientific advances during the past 50 years. THE BEGINNINGS OF ORAL REHYDRATION Children's caregivers have used oral fluid replacement remedies for diarrhea for many centuries (1,2). Ancient manuscripts describe using naturally occurring fluids such as coconut milk and liquids containing rock salt and molasses. Oral rehydration therapy began to evolve in the 1940s, largely because of the initiatives of Harold Harrison at the Baltimore City Hospital and Daniel Darrow at Yale (3–5). Darrow performed balance studies in children with acute diarrhea that identified the need for replacement of sodium, potassium, and alkali to correct acidosis. To a balanced electrolyte solution that he had used successfully for rehydration, Harrison added glucose, probably for its added value as a nutrient. In the early 1950s, the Medical Research Council of Great Britain also recommended that glucose-electrolyte solutions should be used in the United Kingdom. Again the rationale for adding glucose to these early defined oral rehydration solutions (ORS) was not for its beneficial effect on oral rehydration (for this was yet to be discovered), but for nutritional reasons, to prevent ketosis. In 1953, Chatterjee (6) first showed that ORS could rehydrate patients with cholera and avoid the need for intravenous fluid. Although the practice of ORT was established, the true scientific rationale remained to be discovered. SCIENTIFIC BASIS OF ORT THERAPY The science underpinning of ORT was soon established with the discovery that sodium and glucose transport was coupled in the small intestine (7–11). It also became evident that other solutes, such as amino acids, also were absorbed by active transport, again coupled with sodium ions. Subsequent work showed that sodium also enters the enterocyte by additional mechanisms including “electrogenic” nonnutrient-coupled transport using selective ion channels and neutral sodium absorption through a pair of linked ion exchangers, involving sodium and bicarbonate and sodium and hydrogen, respectively. Sodium-solute–coupled cotransport is an energy-dependent process, the sodium gradient within the cell being maintained by the Na-K adenosinetriphosphatase pump on the basolateral membrane of the enterocyte. ESTABLISHING ORT IN CLINICAL PRACTICE Clinical studies of ORT in human cholera in the Philippines and India confirmed that this approach could be used for oral replacement of water and electrolytes, even in this devastating diarrheal disease. Many further clinical studies followed, confirming the dramatic effect that ORT had on mortality in acute diarrheal disease, and within a year or two, the World Health Organization (WHO) was driving a major campaign to implement ORT worldwide (12–16). CONTROVERSIES IN ORS COMPOSITION Although the success of glucose-electrolyte based ORT is unquestioned, controversy continues about the ideal composition of these solutions. One major area for debate has centered on sodium and glucose concentrations and, thus, osmolality. Although glucose has traditionally been the main substrate for ORS, the possibility that efficacy may be increased by using complex substrates, such as cereals or defined glucose polymers, now has been explored extensively. Complex substrate ORS are hypotonic, and their early successes contributed to the development of our hypothesis that decreasing osmolality could be a major factor in improving ORS efficacy. Sodium The sodium concentration of ORS must be sufficient to replace sodium losses and correct hyponatremia but not so high as to cause or worsen hypernatremia, which can itself occasionally result in death. It has been suggested that adequate rehydration and sodium replacement can be achieved by merely using soft drinks, such as “flat” cola. However, all such drinks have extremely low sodium concentrations, and many have a high osmolality because of the glucose or sucrose content, which can itself worsen diarrhea because of monosaccharide intolerance. The sodium concentration of the WHO-ORS, 90 mmol/L, was probably derived from the fecal sodium concentration in adults with cholera (17). Concern about widespread use of the 90 mmol/L ORS relates to hypernatremia and periorbital edema, which occasionally occur with WHO-ORS in infants and children in the industrialized world who receive high-solute feeds. Occasionally, the effects of inaccurate preparation during use of high-sodium ORS can be devastating, as has been reported from Egypt (18). Sometimes overlooked, the WHO recommends that after an initial, relatively short period of rehydration, its ORS should be used with ad libitum water, at least in the ratio of 2 to 1. This would result in a sodium concentration of 60 mmol/L. Therefore, at least some of the reports of hypernatremia may relate to the continued use of WHO-ORS without additional water during the maintenance period. Concerns about hypernatremia in the developed world resulted in the proliferation of new ORS formulations with lower sodium concentration (30–60 mmol/L) (19). Part of the rationale for these developments was the observation that more than 50% of cases of acute diarrhea in infants and young children in industrialized countries were caused by rotavirus, in which the stool sodium concentration was much lower than that in cholera. All clinical trials in well-nourished and malnourished children of all ages in developed and developing communities have shown that ORS with sodium concentrations in the 50 mmol/L to 60 mmol/L range are safe and effective for rehydration and maintenance therapy of mild to severe dehydration from acute noncholera diarrhea (19). Hyponatremia occurred infrequently and was generally only of biochemical and not clinical significance. Therefore, decreased-sodium ORS probably are safe and effective for rehydration and maintenance, and avoid the inconvenience of having to recommend additional water during the maintenance phase. This may increase the precision of the composition of maintenance ORS. If this approach were adopted, there may be no need to recommend rehydration and maintenance ORS with varying sodium concentrations. Glucose Glucose concentrations in ORS have varied widely between those in the current WHO-ORS (111 mmol/L) to the much higher concentrations of 200 mmol/L to 300 mmol/L found in some of the early ORS that were produced commercially for use in industrialized countries. The rationale for including such high concentrations is not clear and is inconsistent with published data that suggest that maximal water absorption in the jejunum occurs with glucose concentrations between 80 mmol/L and 140 mmol/L. In addition, the use of high-glucose ORS is associated with the appearance of glucose in the stool (12,20,21). Recent studies in animal models of secretory diarrhea have shown that decreasing glucose concentration to less than 110 mmol/L does not significantly decrease glucose or sodium absorption and has the advantage of increasing water absorption (22–30). Recent clinical studies confirm that low-glucose (70–100 mmol/L), hypotonic ORS decrease stool volume in children with acute gastroenteritis (31,32). ORS OSMOLALITY Until several years ago, the WHO-ORS and the majority of commercially available ORS were isotonic or moderately hypertonic, because of the concentrations of sodium or glucose. However, providing there are sufficient substrate and sodium and chloride ions to promote optimal absorption, no evidence suggests that increasing sodium or glucose concentration will improve efficacy. On the contrary, evidence shows that high-glucose ORS may increase stool volume because of monosaccharide intolerance, and high-sodium ORS, particularly if administered without ad libitum water in the maintenance phase, can produce hypernatremia. The discovery that rice powder and other cereals could replace glucose and improve efficacy of ORT in the early 1980s had an important impact on our thinking, both regarding the physiologic principles of ORT and also its practical implementation (33–36). One important aspect of these solutions is that it is possible to deliver an increased amount of substrate without increasing the overall osmolality. Some suggest that this increased amount of glucose substrate and the small amount of protein and amino acids (approximately 10%) may account for the increased water absorption and decreased stool volume. However, studies in animal models have failed to show that this explains their increased efficacy. During the past decade, the hypothesis that reducing ORS osmolality is the major determinant for improving water absorption from both simple and complex carbohydrate-based ORS has been examined. Because of the difficulty in performing studies in infants and children with diarrheal disease, we developed a series of animal models in which to perform preliminary studies with ORS of varying composition and osmolality (26–37). We extended these models to humans by performing perfusion studies in normal human jejunum and in human jejunum in which intestinal secretion had been induced by cholera toxin (27,28). We showed that decreasing the osmolality of standard glucose-electrolyte ORS resulted in improved water absorption in all models (Fig. 1). Our work focused on a glucose-containing ORS with an osmolality of approximately 240 mOsm/kg, and this now has been evaluated in clinical trials and shown to decrease stool volume in the same way that the hypotonic, complex-carbohydrate ORS do (31,32). A working group of the European Society for Paediatric Gastroenterology and Nutrition (ESPGAN) considered these findings and how they related to ORT for infants and children in Europe; ESPGAN endorsed the use of hypotonic ORS containing lower concentrations of glucose and sodium than those of the WHO-ORS (38). The most widely used ORS in the United Kingdom is hypotonic, as are the majority of currently available commercial ORS in the United Kingdom.FIG. 1.: Net water absorption, showing that Hypotonic Oral Rehydration Solution (ORS) improves water absorption in normal and infected rat intestine (p < 0.05 vs WHO ORS & BNF ORS in all three conditions). BNF = British National Formulary (from Hunt et al 1991).Glucose polymers provide an attractive option for delivering glucose to the small intestine at low osmolality. In animal and human models of diarrhea, evidence is accumulating that water absorption from the small intestine can be substantially increased with rice powder and glucose polymer-based ORS that have osmolalities of 150 mOsm/kg to 170 mOsm/kg (37–41). A new concept has recently been introduced into the design of starch-based ORS. Resistant starch is digested only partially in the small intestine, and thus up to 20% reaches the colon to be fermented by colonic bacteria. This produces short-chain fatty acids, which are then absorbed with sodium and water by the colon, further enhancing the ORS efficacy. A large randomized, controlled trial in India has clearly shown that this approach to treating adults with cholera is superior to both standard WHO-ORS and a hypotonic, glucose-based ORS (42). These studies with complex carbohydrate ORS and the previous work with low-osmolality glucose ORS suggest that osmolality is an important determinant of ORS efficacy. The use of naturally occurring starches such as rice powder and other cereals may increase in the future but will depend on local customs and practices and the educational facilities available to ensure that home-produced ORS is appropriately prepared and administered.