WO1998009669A1

WO1998009669A1 - Rice-based oral rehydration solution

Info

Publication number: WO1998009669A1
Application number: PCT/AU1997/000568
Authority: WO
Inventors: Geoffrey Cleghorn; Clare Wall; Ross Shepherd
Original assignee: The Royal Children's Hospital Foundation
Priority date: 1996-09-03
Filing date: 1997-09-03
Publication date: 1998-03-12
Also published as: ZA977884B; AUPO207896A0

Abstract

This invention relates to cereal-based oral rehydration solutions ('ORS') which are particularly useful in treating dehydrated humans. The particular ORS of interest are those which are based on a complex carbohydrate having a glucose chain length of greater than 9 and less than 36 units, the total solution having an osmolality of less than 200 mmol/l. These solutions, it has been found, are more effective than traditional glucose based solutions in treating patients suffering from dehydration. The invention also encompasses sports drinks including the ORS of the invention, and methods of veterinary rehydration treatment.

Description

RICE-BASED ORAL REHYDRATION SOLUTION

FIELD OF THE INVENTION

This invention relates to oral rehydration therapies. In particular it is directed to those therapies which utilise oral rehydration solutions based on cereals and particularly rice. It is also directed at cereal based oral rehydration solutions. BACKGROUND OF THE INVENTION

Dehydration in humans caused by fluid loss in acute cases of diarrhoea for example or from excessive perspiration during exercise can lead ultimately to death but more commonly can result in hospitalisation for rehydration therapy to replace both the fluids and electrolytes lost. In infants and children, dehydration through diarrhoeal episodes can be particularly severe and is of considerable concern in developing countries where chronic diarrhoea can lead to morbidity as a result of impaired intestinal function, catabolism and eventual growth failure.

Traditionally, fluid and electrolyte replacement was effected by intravenous therapies which required hospitalisation, expensive infusion equipment, sterile solutions and trained personnel. Such therapies are clearly not suited to Third World conditions and as such were challenged in the late 1940's and 1950s by oral treatments combining essential electrolytes and glucose. At this time the rationale for including glucose was for nutritional purposes although it is now appreciated that incorporation of glucose in oral rehydration solutions assists in physiological mechanisms including sodium transport and water absorption across the intestinal mucosa. Current oral rehydration therapies available in developed countries include sodium, potassium, chloride, bicarbonate, citrate and glucose and commonly have an osmolality (the measure of the rate of diffusion of a fluid through a semipermeable membrane) in the range of 230-240 mmol/l and above. In developed countries oral rehydration solutions also commonly contain flavouring and sweeteners to enhance acceptance amongst children.

Despite evidence that oral rehydration therapies are effective, they continue to be under utilised in both developing and developed countries in the treatment of diarrhoea although their acceptance in sport associated rehydration is increasing. Fluid therapy for dehydrated children often still consists of clear fluids such as fruit juice or carbonated beverages despite the fact that they have high osmolalities or inappropriate ratios of sodium to carbonate. One major difficulty with the use of existing oral rehydration therapies in developing countries is that their use does not necessarily result in a reduction of volume, frequency or duration of diarrhoea: factors which are of key concern to carers of dehydrated patients and which lead to overuse of antidiarrhoeal drugs and under administration of fluids. As an alternative to glucose based oral rehydration therapies, cereal based therapies were first trialled in the early 1980s. In cereals, both long and short chain glucose polymers in the form of starch are available leading, it was thought, to better assistance of electrolytes across the intestinal mucosa. Digestion of starches and proteins in cereals release glucose, amino acids and short chain peptides, all of which are organic solutes which enhance the absorption of sodium and water. It is thought that cereal based oral rehydration therapies thus provide more energy to the dehydrated patient as they can be used in larger amounts than glucose based therapies without incurring an osmotic penalty. A further advantage of cereal based therapies is that they are derived from a familiar food source of much of the Third World, are easily accessible to most of the Worlds population and are relatively inexpensive when compared to products based on refined products such as glucose.

One oral rehydration solution is described in US Patent 5,120,539 (BOATWRIGHT). This solution is derived from rice, cornstarch, wheat or potato which is boiled and then combined with α-amylase plus sodium or calcium and other electrolytes. The composition is characterised by the action of the α- amylase on the complex carbohydrate source to produce glucose polymers having a chain length of about 2-9 units. It has an osmolality in the order of 200- 300. An alternative solution is provided in Australian Patent 642483

(BRISTOL-MYERS SQUIBB). The feature of this solution is the use of rice dextrin having a glucose polymer profile of primarily 2-6 glucose units. US Patent 5,489,440 (ABBOTT LABORATORIES) also teaches the use of a rice based oral rehydration solution but in contrast to US5.120,539 states that the use of amylase enzymes in the product leads to a product having an unacceptably high osmolality. None of these therapies are however suited to the particular requirements of mass treatment in Third World environments since they are, at least, not readily reconstitutable in cold water and cannot be provided in a prepackaged form. Moreover, they all involve enzymatic hydrolysis of the complex carbohydrate present leading to compositions having limited osmotic advantage.

It is thus an object of this invention to provide an oral rehydration solution which may be commercially produced, can be reconstituted in cold water and is acceptable to dehydrated patients. In particular it is important to patient acceptability that the solution reduce stool output, shorten illness duration and time to resumption of solid food intake, be administrable to infants under 3 months of age and be inexpensive. It is also important that the solutions provide adequate rehydration in clinical settings in which viral gastroenteritis predominates. It is also an object of this invention to provide an oral rehydration solution which is also acceptable as a sport or exercise rehydration or hydration fluid, in post operative fluid replacement, for fluid replacement in the elderly, in amelioration of jet lag, and in other types of fluid replacement therapy, for example, heat stroke and hypothermia. It is a further object of this invention to provide an oral rehydration therapy for treatment of patients with chronic diarrhoea, particularly one suited to the treatment of children and infants. SUMMARY OF THE INVENTION

To this end there is provided an oral rehydration solution including a cereal source of complex carbohydrates and electrolytes characterised in that the average glucose polymer chain length of the complex carbohydrate is greater than 9 and less than 36. Desirably the solution has a glucose polymer profile of from about 90-99% short chain glucose polymers of greater than 9 units and less than 36 units. Preferably the osmolality of the solution is less than 200 mmol/l, and more preferably in the range 150 - 190 mmol/l. Most preferably the oral rehydration solution of the invention has an osmolality of about 170 mmol/l and a glucose polymer profile of about 97% short chain glucose polymers of greater than 9 units and less than 36 units.

This invention is predicated on the discovery that long chain glucose polymers derived form a source of complex carbohydrate are surprisingly well absorbed across the intestinal mucosa in young children and infants and particularly when present in a solution having a low osmolality result in rapid and effective rehydration of patients requiring rehydration therapy. The efficacy of the solutions according to the invention is particularly potentiated by the low osmolality which further enhances water movement as a function of the osmotic gradient.

In a preferred embodiment of the invention the source of the complex carbohydrate is rice, particularly in the form of rice flour. In a further preferred embodiment of the invention the rice flour is pregelatinised. Pregelatinising may take place by drum drying the rice. It has been found that drum drying the rice may result in optimal drying, optimal gelatinisation of the starch and maximal reduction of microbial contamination. The resulting rice cereal has good homogenicity, good gelatinisation and therefore good digestability as well as good chemical and bacterial stability. Preferably the oral rehydration solutions according to the invention additionally include a hydrolysing agent to assist in obtaining the correct proportion of glucose polymers of differing chain length. In preferred embodiments this hydrolysing agent is α-amylase.

In another aspect of the invention there is provided an oral rehydration solution including pre-gelatinised rice flour, sodium chloride, potassium, citrate and aspartame. In order to provide a formulation which is acceptable to patients, the solutions according to this aspect of the invention preferably also include flavouring. The flavouring may be chosen from, but is not restricted to lemon orange and blackcurrant flavours. In one embodiment of this aspect of the invention, the composition of the solutions per litre are as follows: rice flour 20-90 g

NaCl 2-4 g

K+ 1-3 g

Citrate 0.5-3 g flavouring 0.2-1.2 g aspartame 0 -1.0 g

In one preferred embodiment of this aspect of the invention the oral rehydration solutions include per litre: rice flour 30 g NaCl 2.35 g

KCI 1.5 g

Di-sodium citrate 2.65 g flavouring 0.60 g aspartame 0.38 g (0.19 g phenylalanine) In another preferred embodiment of this aspect of the invention the oral rehydration solutions include per litre: rice flour 30 g

NaCl 3.53 g

K Citrate 2.05 g

Citric acid 0.865 g flavouring 0.60 g aspartame 0.38 g (0.19 g phenylalanine)

Another embodiment of this invention excludes aspartame for patients with phenylketonuria (PKU) and replaces it by increased natural flavouring It has been found that 30 g of rice flour is optimum in the solution since, when reconstituted it can be delivered to a small child or infant through the teat of a standard infant bottle obviating the need for the messy use of cups or spoons.

The most preferred solutions according to the invention are preferably reconstituted in one litre of water, boiled and then allowed to cool to about 37°C - body temperature. In another aspect of the invention there is provided a method of treating dehydration in a human or animal patient comprising administering to said patient an oral rehydration solution including a cereal source of complex carbohydrates and electrolytes characterised in that the average glucose polymer chain length of the complex carbohydrate is greater than 9 and less than 36 units. Desirably the solution has a glucose polymer profile of from about 90-99% short chain glucose polymers of greater than 9 units and less than 36 units. Preferably the osmolality of the solution is less than 200 mmol/l, and more preferably in the range 150 - 190 mmol/l. Most preferably the oral rehydration solution of the invention has an osmolality of about 170 mmol/l and a glucose polymer profile of about 97% short chain glucose polymers of greater than 9 units and less than 36 units.

There is also provided a method of rehydrating or hydrating persons who have engaged in sport or exercise comprising administering to said persons an oral rehydration solution including a cereal source of complex carbohydrates and electrolytes characterised in that the average glucose polymer chain length of the complex carbohydrate is greater than 9 and less than 36. Desirably the solution has a glucose polymer profile of from about 90-99% short chain glucose polymers of greater than 9 units and less than 36 units. Preferably the osmolality of the solution is less than 200 mmol/l, and more preferably in the range 150 - 190 mmol/l. Most preferably the oral rehydration solution of the invention has an osmolality of about 170 mmol/l and a glucose polymer profile of about 97% short chain glucose polymers of greater than 9 units and less than 36 units. The invention also encompasses a sports drink based on the oral rehydration solution described hereinabove. Sports drinks are commonly used by both amateur and professional athletes as a means of rehydrating after significant physical exertion, particularly in hot weather. In a preferred embodiment of this aspect of the invention the sports drink includes a cereal source of complex carbohydrates and electrolytes characterised in that the average glucose polymer chain length of the complex carbohydrate is greater than 9 but is less than 36. In another preferred embodiment of the invention, the sports drink includes rice flour 20-90 g

NaCl 2-4 g K+ 1-3 g

Citrate 0.5-3 g flavouring 0.2-1.2 g aspartame 0 -1.0 g

It will also be appreciated that other aspects of the invention include methods of post operative fluid replacement, fluid replacement in the elderly, jet lag amelioration, rehydration of patients suffering from travellers diarrhoea, and fluid replacement in heat stroke and hypothermia in all cases comprising the administration to the person concern of oral rehydration solutions according to the invention. The oral rehydration solutions of the invention are also suited to use in veterinary environments such as in the rehydration of animals suffering from diarrhoea, for example, pigs suffering from TGE virus, and for general fluid replacement purposes pre and post operative, when the animal is suffering form hypothermia, after over exertion and in similar dehydrating conditions. The solution according to the invention may also be used in the equine racing industry to rehydrate and hydrate race horses.

The oral rehydration solutions of the invention are conveniently packaged according to need. In one embodiment they may be packaged as a dehydrated powder in semi permeable membranes allowing the product to be rehydrated or dissolved in an aseptic manner using water or fluid which may be contaminated, including for example, untreated water supplies or water from creeks or rivers, the product may also be packaged as a pre mixed solution in varying volumes for instant use.

BRIEF DESCRIPTION OF THE FIGURES Figure 1 shows the effect of different polymer lengths in a rice based ORS on water absorption in a segment of a secretory rat intestine induced by cholera toxin. Figure 2 shows the effect of different polymer lengths in a rice based ORS on water absorption in a segment of normal rat intestine.

Figure 3 shows the net water movement (μl/min/g) in a normal and secreting intestine perfused with R-ORS or G-ORS. Figure 4 shows the water influx and efflux (μl/min/g) from R-ORS and G-

ORS in normal and secreting intestines.

Figure 5 shows the net sodium (μmol/min/g) in a normal and secreting treated intestine perfused with R-ORS or G-ORS.

Figure 6 shows blood glucose measurements at 0, 75, 90 and 105 minutes in a normal intestine perfused with R-ORS or G-ORS.

Figure 7 shows blood glucose measurements at 0, 75, 90 and 105 minutes in cholera toxin treated intestine perfused with R-ORS and G-ORS.

Figure 8 shows estimates for time to recovery onto other fluids from admission after treatment with R-ORS and G-ORS. Figure 9 shows estimates for time to recovery onto solid foods from admission after treatment with R-ORS and G-ORS. PREFERRED EMBODIMENT OF THE INVENTION EXAMPLE 1

To confirm the optimal glucose chain length of the complex carbohydrates used in the solutions according to the invention, male Wistar rats (150-200 g) were fasted for 18 hours with free access to water. The rat was anaesthetised with intraperitoneal sodium pentobarbitone (60 mg/kg), top up doses were administered as required. A 3 cm midline abdominal incision was made and a 13-20 cm segment of the proximal jejunum 10 cm distal to the ligament of trietz and proximal to the ileo-caecal junction was cannulated with polyethylene tubing (0-105 inches in internal diameter), (Dow Corning, USA). The perfusion solution was perfused through the proximal tubing. Perfusion Solution

1.5 g of rice based oral rehydration solution (R-ORS), of differing polymer lengths type 3, 10, 18, 35. 0.1 g of potassium citrate. 0.04 g of citric acid.

0.18 g of sodium chloride.

1 (Ci of polyethylene glycol- C14 (PEG- C14), MW 4000, (Du Pont, USE, Lot No

2967-242), which acted as a non absorbable marker. 0.25 g of polyethylene glycol-C (PEG), MW 4000, (Ajax Chemicals, Australia).

50 ml of tap water.

The osmolality of the R-ORS perfusion solution was measured on a osmometer (osmomat 030, Gonotec).

The perfusion solution was pumped through the intestine with a Imed 800 syringe pump, (Warner-Lambert Health Technologies Ltd, Abingdon) for 105 minutes at a rate of 0.416 ml/min. The first 60 minutes provided time for the solution to equilibrate (Elliott et al 1988), during the next three 15 minute periods, 3 effusate collections were made from the distal tubing. After a number of experimental perfusion time periods it became apparent that 105 minutes was the optimum time that allowed a balance between the volume of effusate required for analysis (approximately 1.25 ml) and the maintenance of optimal physiological conditions. Blood glucose levels were measured after each effusate collection by a Ames glucometer 3, (Bayer Diagnostics, Australia).

During the perfusion period, body temperature was continuously monitored by rectal probe and maintained at 37°C using a servo-controlled heating mattress.

Interluminal pressure was monitored throughout the perfusion period with a vertically mounted piece of tubing connected by a three-way tap to the inflow cannula, a blockage in the lumen was indicated by fluid rising in the tube. After perfusion the intestine was removed, the perfusion length measured and dried in an oven at 90°C for 18 hours. The dry weight was recorded.

Induction of a secretory state with cholera toxin (Elliott et al 1991) A secretory state with cholera toxin was induced as described previously by Elliott et al 1991. After cannulation of the intestine with polyethylene tubing, the intestinal segment intestine was lavaged with isotonic chloride to remove any luminal contents. The distal cannula was clamped and 75 μg of pure cholera toxin, (Sigma Aldrich, Australia, product No 3012) in 1 ml of isotonic sodium chloride was introduced via the distal cannula to allow equal distribution throughout the intestinal segment. The proximal cannula was clamped and the abdomen closed for a period of 2 hours. At the end of this period the clamps were removed, the intestine drained of its contents and washed with 5 ml ORS. The intestine was then perfused with the perfusion solution as per the perfusion protocol.

To establish that the intestine was in a secretory state the above methodology was applied however instead of perfusion with R-ORS or gastrolyte the intestinal segment was perfused with saline. Saline was perfused instead of ORS as ORS are designed to correct the secretory state by promoting the absoφtion of water therefore is a ORS is effective there will be no change in net water movement compared to the control. It was expected based on Elliottis data that if the intestine was in a secretory state the perfusion of saline will cause secretion of water. Flame Photometry

The perfusion solution, the three effusate samples where analysed via flame photometry, (Corning-EEL flame photometer, England) for sodium and potassium concentrations. The effusate and perfusion solution samples were diluted 1/250 and the plasma samples 1/400 for sodium analysis and 1 in 100 for potassium.

The level of C14 in the perfusion solution, effusate was measured via dual label counting on the Beckman liquid scintillation counter. Calculations (Sladen and Dawson 1968) Net water movement (μl/min/g) = {flow rate (1-[PEG]/[PEG])}/dry weight Statistical Analysis

Results were presented as mean SD. Comparison between the groups studied was made by Anova two factor analysis of variance with replication and one factor analysis of variance with replication. All statistical analysis was done with SAS. A probability of less than 5% was considered significant. Results of these trials are shown in Figures 1 and 2 in which:

Figure 1 shows the effect of different polymer lengths in a rice based ORS on water absorption in a segment of a secretory rat intestine induced by cholera toxin. Despite the small size of the group studied (6), significant difference was found between type 3 and type 10 in promoting water absorption, and Figure 2 shows the effect of different polymer lengths in a rice based ORS on water absorption in a segment of normal rat intestine. No significant difference was found between the groups of 15. EXAMPLE 2

The following example illustrates the advantage of a Rice based oral rehydration solution (R-ORS) over a Glucose based oral rehydration solution (G- ORS) in both a normal and secretory intestine. To determine water absorption from the rice and non-rice solutions, an in vivo steady state animal perfusion was set up using rat small intestine in both a normal and secretory state as induced by cholera toxin. A model to establish bidirectional flow of water across the intestine was validated. Deuterium, a stable isotope of water was added to the test solution. Prior to this, Tritium a radioactive isotope of water, was given to the rat to equilibrate into the body water. The accumulation rate of the D20 in the circulation and loss from the effusate was used to determine water uptake. The loss of 3H from the circulation and accumulation in the effusate was used to determine water movement from the body into the intestinal lumen. Male Wistar rats (150-200 g) were fasted for 18 hours with free access to water. After the fasting period, 50mCi of triated water were administered orally via a gavage. A period of 2 hours was allowed for the tritium to equilibrate in body water before the surgical procedure was performed. The rats were anaesthetised with intraperitoneal sodium pentobarbitone (60 mg/kg), top up doses were administered as required. A 3 cm midline abdominal incision was made and a 13-20 cm segment of the proximal jejunum 10 cm distal to the ligament of trietz and proximal to the ileo-caecal junction was cannulated with polyethylene tubing (0-105 inches in internal diameter), (Dow Corning, USA). The perfusion solution was perfused through the proximal tubing. Perfusion Solution

30 g of pregelatinised rice flour 3.53 g NaCl 2.05 g Potassium Citrate

0.865 Citric acid.

1 uCi of polyethylene glycol- C14 (PEG- C14), MW 4000, (Du Pont, USE, Lot No

2967-242), which acted as a non absorbable marker. 5 g of polyethylene glycol-C (PEG), MW 4000, (Ajax Chemicals, Australia).

1000 ml of tap water.

The perfusion solution was pumped through the intestine with an Imed 800 syringe pump, (Warner-Lambert Health Technologies Ltd, Abingdon) for

105 minutes at a rate of 0.416 ml/min. The first 60 minutes provided time for the solution to equilibrate (Elliott et al 1988), during the next three 15 minute periods, 3 effusate collections were made from the distal tubing. After a number of experimental perfusion time periods it became apparent that 105 minutes was the optimum time that allowed a balance between the volume of effusate required for analysis (approximately 1.25 ml) and the maintenance of optimal physiological conditions. Blood glucose levels were measured after each effusate collection by a Ames glucometer 3, (Bayer Diagnostics, Australia).

To establish that the intestine was in a secretory state the above methodology was applied however instead of perfusion with R-ORS or Gastrolyte the intestinal segment was perfused with saline. Saline was perfused instead of ORS as ORS are designed to correct the secretory state by promoting the absorption of water therefore is a ORS is effective there will be no change in net water movement compared to the control. It was expected based Elliottis data that if the intestine was in a secretory state the perfusion of saline will cause secretion of water. The perfusion solution, the three effusate samples were analysed via flame photometry, (Corning-EEL flame photometer, England) for sodium and potassium concentrations.

The D20 concentrations in the perfusion solution effusate and final plasma sample were analysed using a FTIR spectrometer (Perkin Elmer system 2000). no and 3H in the perfusion solution effusate and blood samples were measured using dual label counting on a Beckman liquid scintillation counter (Beckman). The dual-label DPM instrument operating parameters were modified to minimise overlap between the HC and 3H spectra. Calculations (Sladen and Dawson 1968) Net water movement (ul/min/g) = {flow rate (1 -[PEG]/[PEG])}/dry weight Statistical Analysis

Results were presented as mean SD. Comparison between the groups studied was made by Anova two factor analysis of variance with replication and one factor analysis of variance with replication. Ail statistical analysis was done with SAS. A probability of less than 5% was considered significant.

The experimental design was a four cell design, n=11 in each group. The results were presented as a mean ± SEM. The comparison of differences between the groups was made using a two-factor analysis of variance with duplication and one factor analysis of variance with replication. The R-ORS promoted water absorption in both the normal and cholera-toxin treated intestine (fig. 3.). G-ORS also promoted net water absorption in both normal and cholera-toxin treated intestine (same fig). R-ORS was significantly more effective in promoting net water absorption than the G-ORS in the normal (p<.02) and cholera-toxin treated (p<. 05) intestine. The water influx and efflux rates are shown in fig 4. Water efflux was significantly reduced in the R-ORS group compared to the G-ORS group in both normal and (p<.01 ) and the secreting intestine (p<.01 ). In the normal intestine, R-ORS promoted net sodium absorption and G-ORS promoted net Na+ secretion (fig 5). In the cholera-toxin treated intestine both R-ORS and G-ORS caused net secretion of Na+. These data indicate that whilst not significant, R-ORS promoted 84% less secretion than G-ORS in the normal intestine and 64% less secretion in the cholera intestine. Blood glucose levels increased after perfusion with both ORS (fig 6). The G-ORS produced a greater blood glucose level at 75 min and 90 min than the R-ORS but this was not significant. Blood glucose levels increased after perfusion with both ORS in the cholera toxin treated intestine (fig 7). The G-ORS produced a significantly greater blood glucose level at 75 min compared to the R-ORS.

EXAMPLE 3

The following study illustrates the advantages of oral rehydration solutions according to the invention as well as the benefits of applying such rehydration therapies to children suffering from diarrhoeal induced dehydration. In the study an oral rehydration solution comprising 30 g of pregelatinised rice flour, 2.35 g Nad, 1.5 g KCI, 2.65 g Disodium Citrate, 0.60 g flavouring and 0.38 g aspartame (comprising 0.19 g phenylalanine) in which the rice flour had been pre-treated with α-amylase was used. The solution had a Dextrose Equivalent ("DE") number of 3 meaning that 97.5% of the glucose polymers had a chain length greater than 9. A basic analysis of 30 g of rice-oral rehydration solution in 1 litre showed the following profile: energy 110 kcal protein 1.8 g sodium 60 mmol potassium 20 mmol chloride 60 mmol citrate 10 mmol

The measured osmolality of the solution was 172 mmol/l.

Following study approval by the Ethics Committee of the Northern Region Health Authority and The Royal Children's Hospital, Brisbane, 100 infants and young children aged 4 weeks to 5 years admitted to hospital with acute watery diarrhoea (3 or more stools in the previous 24 hours) of less than 4 days duration and clinically assessed as having mild to moderate dehydration (using a standardised protocol) were eligible to enter the study. Exclusion criteria were severe dehydration, other medical complications, previous bowel surgery, significant abnormality evident on clinical or laboratory screening, treatment with antibiotics in the previous 7 days or concomitant medication (i.e. antispasmodics) likely to interfere with oral rehydration therapy. No patients were treated by any of the study investigators as the attending physician, thus no treatment outcomes or variables were influenced by the study investigators.

A randomised trial was designed in which the null hypotheses to be tested were that no difference existed between the treatment groups in terms of the following outcomes: stool output, duration of diarrhoea, and recovery time. The first three outcomes are consistent with those included in previous studies of oral rehydration, while recovery time (RT) was chosen as a measure of well- being and was defined as the time taken in hours from admission to when both normal diet (RTD) and other fluid (RTF) were first successfully consumed. It was determined that 50 subjects per treatment arm could be entered on study over a six month period which would permit a treatment difference of 8.5 hours to be detected with alpha equal to .05 and power equal to 80%. Table 1. Relative compositions in mmol per litre of rice ORS and glucose ORS.

* Rice ORS 30 g yields 120 mmol/l of glucose on complete hydrolysis ^b Measured osmolarity

Patients were randomised to either R-ORS (Rice-Oral Rehydration Solution) or G-ORS (Glucose-Oral Rehydration Solution) (Gastrolyte, Rhone- Poulenc Rorer, Australia) (see Table 1 ) using a pre-determined random number table assigned in the Accident and Emergency department on an open label basis. A blinded study was not possible as the visual appearance of the two solutions were vastly different. The ORS's were administered exclusively at 60- 100 ml/kg body weight for a minimum of 6 hours and the ORS was continued in combination with other fluids (as tolerated) as maintenance after the clinical resolution of dehydration. Where an inadequate volume of ORS was consumed a nasogastric tube was inserted if tolerated. The decision to introduce other fluids and food was made by the attending physicians for each patient and was not influenced by the study team and was based on clinical indicators of tolerance of the ORS without vomiting, improvement or cessation of diarrhoea and correction of dehydration. It is the general principle of the various attending staff to offer a normal diet as soon as dehydration has abated and increased amount as tolerated. If the patient vomited or refused food or other fluid this was recorded as unsuccessful introduction and the patient was reintroduced to diet as appropriate. It was considered that the recovery time reflected the overall well-being of the patient and represented the return of a normal appetite. On admission to hospital the degree of dehydration was assessed by the attending physician using a standardised protocol. History of diarrhoea, a retrospective 8 hour food and fluid intake and results of a physical examination, were recorded in a pre-coded patient proforma. Nude weights were recorded to the nearest 10 g using a SECA (727) balance. Stool samples were taken for laboratory analysis and identification of pathogens. Volume of ORS and introduction of other fluid intake (RTF), introduction of diet (RTD), number of episodes and volume of vomiting, and degree of dehydration were recorded after 8, 12 and 24h and then at 24h intervals until discharge from hospital. Stool and urine output were assessed using urine bags and pre-weighed disposable nappies. Stool frequency was also recorded as an adjunct to total faecal excretion. Frequency of stool, stool weight, and urine output were recorded as 8h, 12h, 24h and then at 24 hour intervals until cessation of diarrhoea. Duration of diarrhoea (defined as the first formed stool passed if followed by two consecutive non-watery stools or by 12h without defecation), duration of hospital stay, success of rehydration, occurrence of adverse events and treatment success or failure were recorded on completion of the study.

The variable "weight" used in the analyses is the weight for each child standardised by age quartile by sex (Z-Score). Univariate analyses (independent t-test, chi-squared analysis, Fisher's exact test, Wilcoxon rank sum test) were performed as appropriate to compare the two treatment groups with respect to patient characteristics at study entry and in terms of the outcome variable stool volume. Time to event analysis was used for duration of diarrhoea, length of hospital stay, and recovery time. Observations for subjects who did not receive other fluids or solid food during their time in hospital were right-censored as of the time the subject was discharged from hospital. Product- limit estimates were calculated for each treatment group and compared using the log-rank statistic. Multivariate regression analyses were performed with treatment as an explanatory variable with particular emphasis on detecting whether treatment had an effect on the course of illness after allowing for patient and pretreatment variables, i.e. age, sex, duration of diarrhoea, duration of vomiting, fluid intake and food intake. Models were fitted using a forward selection procedure and confirmed using a backward selection procedure to select out those explanatory variables which made a significant contribution to the regression. Stool volume was evaluated using multiple linear regression. The dichotomous variables, RTD and RTF, were analysed by logistic regression. Duration of diarrhoea, length of hospital stay and recovery time were analysed using Cox regression analysis. All analyses were performed using the SAS statistical package. There were no significant differences between the two groups in terms of age (glucose; 19.9+/-2.8 mths, rice; 21.4+/-4.7 mths), standardised weight (z-score 1.2+V0.1 vs 1.1 +V0.1 ), sex, race, duration of diarrhoea, duration of vomiting, previous food intake and presence of rotavirus (glucose 18/50 vs rice 22/50). Rice ORS subjects had somewhat higher mean pre-study fluid intake, 273.8 ml vs 232.0 ml on the glucose ORS subjects, but this was not statistically significant (P=12).

All subjects were successfully rehydrated and no adverse experiences were reported for any subject during the study. Amongst the outcome variables (Table 2a, b), significant differences were found in favour of rice ORS subjects compared with glucose ORS subjects in mean stool volume (160.4 ml vs 213.1 ml respectively, P=.02), mean duration of diarrhoea (17.3h vs 24.3hr respectively, P=.03) And mean length of hospital stay (38.8h vs 49.2h respectively, P=.02).

During the course of the study, more rice ORS subjects recommenced normal food and fluids (food: rice 26/50 {52%} vs glucose 11\50 {22%} p< 0.007, fluid: rice 35\50 {70%} vs glucose 25\50 {50%} n.s.) than did the glucose ORS subjects and at an earlier time (RTF rice 12.7 hrs +V0.6, glucose 18.1 hrs +V0.5 p< 0.001 ; RTD rice 17.7 hrs +V0.6, glucose 19.8 hrs +V0.4 p< 0.01 ).

Time to cessation of diarrhoea showed a near significant trend (P=.05) in favour of rice ORS compared with glucose ORS. Plots of the product-limit estimates of recovery time in terms of other fluid intake (RTF) for each treatment group (Figure 8) show significantly shorter times for rice ORS subjects compared with glucose ORS subjects (means 12h vs 20h respectively, P<.001 ). Similarly, it can be seen in Figure 9 that rice ORS subjects experienced shorter recovery time in terms of food intake (RTD) (mean 24h) than did glucose ORS subjects. Only 22% of glucose ORS subjects had resumed eating and therefore the mean RTD had not been reached (P=.001 ). Table 2a Results of patient variables - ORS intake, faecal, urine output and vomit during the first 24 h of treatment with G-ORS or R-ORS in infants and young children with acute diarrhoea.

0-8 h 8-24 h 0-24 h G-ORS R-O R S G-ORS R-OR S G-OR S R-OR S

mean (SEM) mean (SEM) mean (SEM)

Weight (kg) 1 1.02 (0.4) 1 1.20 (0.4) 1 1.78 (0.5) 11.90 (0.4)

Volume Intake (ml) 597 (24.4) 674 (27.8) 620 (37.1 ) 730 (25.1 ) 1217 (61.5) 1350 (52.9)

No. episodes diarrhoea 2.80 (0.3) 2.00 (0.3) 2.0 (0.1) 1.0 (0.2) 4.8 (0.5) 3.0 (0.4)

Volume stools (ml) 91 (6.4) 77 (6.4) 122 (8.4) 83 (9.3)^** 213 (14.8) 160 (15.4)^**

Volume urine (ml) 110 (9.6) 114 (11.6) 248 (19.7) 301 (22.4) 358 (29.3) 549 (34.0) O cr No. episodes vomit 1 (0.1 ) 1 (0.1 ) 1 (0.1 ) 0 (0.1 ) 2 (0.3) 1 (0.3)

^■Α Volume vomit (ml) 75 (11.1 ) 45 (7.4) 15 (5.0) 10 (2.9) 90 (16.1) 55 (10.3) p<.02 m

Table 2b. Univariate comparison of study outcomes by variable type by treatment group.

Glucose ORS Rice ORS c

Mean (SEM) Mean (SEM) ro σ

Stool volume (ml) 213.1 (14.8) 180.38 (15.4) .02 Duration of Diarrhoea (hrs) 24.3 (2.4) 17.31 (1 .9) .03 Hospital stay (hrs) 49.23 (3.1 ) 38.80 (3.1) .02

Multiple linear regression analysis was performed where stool volume was the dependent variable and patient and pre-treatment variables, i.e. age, sex, duration of diarrhoea, duration of vomiting, fluid intake, and food intake, and treatment were independent variables. Rice ORS was significantly predictive of outcome after adjusting for pre-treatment variables (P=.02).

Separate logistic regression analyses were also performed where recovery times (RTD, RTF) respectively were used as the dependent variables and the patient, pre-treatment and treatment variables comprised the independent variables. As before, rice ORS was found to be significantly predictive of other fluids being introduced (P=.025) and nearly significantly predictive of solid food being introduced (P=.05), after adjusting for pre- treatment variables.

As shown in Table 3, Cox regression analyses of recovery times (RTD, RTF) showed that rice ORS was highly significantly predictive of shorter time to recover (RTF P=.0009 and RTD P=.003) after adjusting for patient and pre- treatment variables. Similar results were found in favour of rice ORS with respect to duration of diarrhoea (P=.025) and length of hospital stay in these patients (P=.025).

Table 3. Results of Cox regression analyses on time-related variables: significant predictors.

OUTCOME COEFFICIENT RELATIVE 95% C.I. P VARIABLE RISK

DURATION OF DIARRHOEA -.430 .650 .428 - .988 .025

Rice ORS

RECOVERY TIME FOR -.926 .396 .229 - .684 .0009 OTHER FLUIDS Rice ORS

RECOVERY TIME FOR -1.320 .267 .112 - .638 .003 SOLID FOOD Rice ORS

LENGTH OF STAY -.465 .628 .418 - .943 .025

Rice ORS It is thus demonstrated that the oral rehydration therapies according to the invention are more effective than glucose based therapies in the treatment of acute diarrhoea in small children.

It is expected that similar results would be achieved in the other methods described hereinabove. It will be appreciated that other embodiments of the invention exist which are not specifically outlined hereinabove but nevertheless are within the scope of the invention claimed.

The rationale of all oral rehydration solutions is based upon the physiological balance of glucose and electrolytes in the subject being treated. with glucose and electrolyte transport there is also active transport of solute across the cell membrane with passive transport of water. The oral rehydration solutions of the invention utilize this phenomenon but its efficacy is potentiated by low osmolality which further enhances water movement as a function of the osmotic gradient. Further enhancement occurs because of the amino acids present in the cereal starch and rice starch in particular. In short, the invention provides an enhanced and unique solution to the difficulties of medical rehydration.

Claims

THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1. An oral rehydration solution including a cereal source of complex carbohydrates and electrolytes characterised in that the average glucose polymer chain length of the complex carbohydrate is greater than 9 and less than 36.

2. The oral rehydration solution as claimed in claim 1 wherein the glucose polymer profile of the solution is 90 - 99% short chain glucose polymers having a chain length of greater than 9 and less than 36.

3. The oral rehydration solution as claimed in claim 1 or 2 wherein the osmolality of the solution is less than 200 mmol/l.

4. The oral rehydration solution as claimed in anyone of claims 1 - 3 wherein the osmolality of the solution is in the range 150 - 190 mmol/l.

5. The oral rehydration solution as claimed in any one of claims 1 - 4 wherein the osmolality of the solution is about 170 mmol/l.

6. The oral rehydration solution as claimed in any one of claims 1 - 5 wherein the glucose polymer profile of the solution is about 97% short chain glucose polymers of greater than 9 and less than 36 units.

7. The oral rehydration solution as claimed in any one of claims 1 - 6 wherein the cereal source of complex carbohydrate is rice.

8. The oral rehydration solution as claimed in claim 7 wherein the rice is present as rice flour.

9. The oral rehydration solution as claimed in claim 8 wherein the rice flour is pregelatinised.

10. The oral rehydration solution as claimed in any one of claims 1 - 9 wherein the solution additionally includes a hydrolysing agent.

11. The oral rehydration solution as claimed in claim 10 in which the hydrolysing agent is α - amylase.

12. An oral rehydration solution including rice flour, sodium chloride, potassium, citrate and aspartame.

13. The oral rehydration solution as claimed in claim 12 which additionally includes flavouring.

14. The oral rehydration solution as claimed in claim 12 or 13 wherein the rice flour is pregelatinised.

15. The oral rehydration solution as claimed in any one of claims 12 - 14 including per litre of solution: rice flour 20-90 g

NaCl 2-4 g

K+ 1-3 g

Citrate 0.5-3 g flavouring 0.2-1.2 g aspartame 0 -1.0 g.

16. The oral rehydration solution as claimed in any one of claims 12 - 14 including per litre of solution: rice flour 30 g

NaCl 2.35 g

KCI 1.5 g

Di-sodium citrate 2.65 g flavouring 0.60 g aspartame 0.38 g (0.19 g phenylalanine)

17. The oral rehydration solution as claimed in any one of claims 12 - 14 including per litre of solution: rice flour 30 g

NaCl 3.53 g

K Citrate 2.05 g

Citric acid 0.865 g flavouring 0.60 g aspartame 0.38 g (0.19 g phenylalanine).

18. An oral rehydration solution as claimed in any one of claims 1 - 17 which is formulated as a sports drink.

19. A method of treating dehydration in a human or animal patient including administering to a said patient an oral rehydration solution as defined in any one of claims 1 -17.

20. A method of rehydrating or hydrating persons who have engaged in exercise or sport including administering to said persons an oral rehydration solution as defined in any one of claims 1 - 17.

21. An oral rehydration solution as claimed in any one of claims 1 - 17 which is packaged in a semi-permeable membrane.