Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Carolina Vettorazzi, Manolo Mazariegos, Susana Molina, Isabel de Ramirez, and Noel W. Solomons
In a companion paper we described the feasibility of liquefying high concentrations of rice flour with amylase to produce enhanced oral rehydration solutions (ORS) with suitably lower viscosities to remain drinkable. This offers a means of orally rehydrating patients, while eliminating much or all of the caloric deficit imposed by conventional, dilute, glucose-based ORS. To explore whether humans would tolerate these solutions, 63 dehydrated diarrhoeic children were randomly assigned to one of three four-hour oral treatments: ORS with 5% rice flour, 10% rice flour incubated with a -amylase, or 15% rice flour incubated with a -amylase. The sodium and potassium concentrations of the solution as well as the osmolality and viscosity were within the desired ranges. On average, the children consumed 70 ± 26%, 90 ± 28%, and 80 ± 45% of the respective ORS, equivalent to intakes of 8 ± 4, 23 ± 9, and 33 ± 18 kcal/kg/4 hours, respectively. The mean increases in weight in four hours were 1.1%, 2.7%, and 1.6%, respectively. The concept of enhanced energy content of ORS based on amylase-treated rice flour appears to be ready for exploration in clinical field trials.
Ingenuity and creativity in food technology and community resource development must be applied to continue to reduce the mortality from acute dehydrating diarrhoeal disease, which still accounts for millions of deaths annually in developing countries [1-4]. In a companion paper [5] we discussed the history and potential of rice-based oral rehydration solution (RB-ORS) to prevent and treat dehydrating diarrhoea. Conventional 2% glucose-based ORS (GB-ORS) is hypocaloric relative to the needs of a young child, especially with the superimposed stress of disease. Four times as much energy can be delivered in an equivalent volume of ORS at the 8% (w/v) upper limit of solubility for rice flour added to water and electrolytes. Even higher deliveries of energy would be possible if rice flour could be solubilized into drinkable solutions at concentrations beyond 8%, enhancing even more the energy nutrition during recovery from acute dehydration.
During preliminary studies [5], we examined in vitro the viscosity, osmolality, and stability of concentrated rice flour solutions that were treated with amylase to reduce the thickness to that of a flowing liquid that would be easy to drink. As the next step in the sequential development of this concept we tested the solution in human subjects. We present a cautious advance that looked at tolerance to the solutions and the apparent success of the initial period of rehydration in children with clinical dehydration due to acute diarrhoea.
Location
The study was carried out in the CeSSIAM Metabolic Ward in the Pediatric Department of the San Juan de Dios General Hospital in Guatemala City. The subjects were 63 children, age 3 to 26 months, who had mild to moderate dehydration due to diarrhoeal disease.
Study design
Previously in the same facility we explored the capacity of healthy, non-dehydrated children to finish an 8-oz (240 ml) portion of various enhanced concentrations of rice flour treated with a -amylase [6, 7]. We found equivalent tolerance when the enhanced RB-ORS and standard ORS were compared. The present study was a double-blind, controlled, clinical trial. The subjects were randomized into one of three treatment groups based on the concentration of RS-ORS: 5% alone, 10% + enzyme, and 15% + enzyme. Boys or girls age 5 to 26 months who were seen in the outpatient emergency facility with acute diarrhoea of 72 hours duration or less, and with mild or moderate dehydration, were eligible for enrollment. Patients were excluded if they had evidence of a systemic infection, had clinical dysentery, or had a weight-for-height deficit of more than 2.5 Z scores in relation to National Center for Health Statistics (NCHS) standards [8].
The degree of dehydration based on history and clinical findings was determined by a team of three physicians. On admission, appropriate amounts of replacement and maintenance fluids were calculated for each patient. The volume for replacement of accumulated losses was based on admission body weight and the estimated degree of dehydration. For maintenance, a fixed volume (25 ml/kg body weight) of one of the three RB-ORS was offered over the first four hours.
RB-ORS preparation
The solutions were prepared each day as necessary, using commercial rice flour (Pacific Rice, Inc., Woodland, California, USA; coded variety 7011). The preparation procedure was as described in the companion article [5]. The sodium (Na) and potassium (K) concentrations of the solutions were verified in a selective electrode photometer (Electrolyte Analyzer; AVL, Schaffhausen, Switzerland). The mean Na concentrations ± SD were 60.0 ± 6.7, 65.2 ± 4.2, and 68.5 ± 0.8 mmol/L for the 5%, 10%, and 15% RB-ORS concentrations, respectively, and the mean K concentrations were 21 ± 1.2, 22 ± 2.2, and 24 ± 0.4 mmol/L, respectively. The prepared solutions were kept under refrigeration until needed; a maximum of 72 hours was allowed between preparation and use before the solution was discarded.
The viscosity of the solutions was measured periodically with a microviscometer (Brookfield Laboratories, Inc., Stoughton, Massachusetts, USA) as a quality control measure. Similarly, osmolality, expressed as mOsm/L, was measured by freezing-point depression with an Osmette micro-osmometer (Precision Systems, Inc., Natick, Massachusetts, USA) at least once a week during the clinical trials.
Clinical procedures
The subject's weight was measured without clothes to the nearest 100 g on an electronic scale with a capacity of 15 kg and a precision of 5 g (Cardinal/ Detecto, Brooklyn, New York, USA) on admission and after the four-hour treatment period. Clinical signs such as tearing, fontanelle depression, skin turgor, state of alertness, irritability, moistness of the mucosa, fever, and urine output were recorded hourly. Each episode of vomiting and each faecal evacuation before admission and during the observation was also recorded, but not quantified.
The treatment was considered successful if the child showed clinical improvement in the signs and symptoms of dehydration at the four-hour evaluation compared with those on admission. If the child showed no change or showed deterioration, the effort was considered a therapeutic failure.
After participating for four hours, the rest of the child's rehydration and rehabilitation therapy became the responsibility of the medical staff of the Oral Rehydration Therapy Ward in the hospital's emergency room. This paediatric team assumed unimpeded control of patient care, generally returning the child to a standard regimen of 2% GB-ORS or instituting intravenous rehydration if initial oral efforts with rice had been unsuccessful.
The characteristics of the three concentrations of RB-ORS are shown in table 1. The viscosities, measured in centipoise (cp), were lower in the more concentrated states due to the action of amylase, making a more fluid beverage at the two higher levels of rice than in the 5% rice emulsion without the enzyme. This concurred with our earlier observations [5]. Osmolality increased in a stepwise fashion, reaching almost 300 mOsm/L in the 15% RB-ORS with added amylase. The metabolizable energy is calculated from the amounts of rice added, and corresponded to the 50, 100, or 150 g of rice flour that had been added to 1 L of solution.
TABLE 1. Analysed characteristics of the three types of RB-ORS
| Characteristics of RB-ORS | Rice flour concentration (w/v) of RB-ORS | ||
| 5% | 10%a | 15%a | |
| Viscosity (cp) | 273 | 105 | 115 |
| Osmolality (mOsm/L) | 174 | 220 | 295 |
| Metabolizable energy (kcal/L) | 176 | 351 | 526 |
a. Treated with a -amylase.
TABLE 2 Initial characteristics of the 63 dehydrated children according to type of RB-ORS treatment
| Characteristics of children | Rice flour concentration (w/v) of RB-ORS | ||
| 5% | 10%a | 15%a | |
| Age (mo) | 14.6 6.1 | 10.5±4.5 | 13.0 ±5.9 |
| Male/female | 15/8 | 13/7 | 7/13 |
| No. of diarrhoeal stoolsb | 4 ±5 | 3 | 2 5 ±5 |
| No. of vomiting episodesb | 4 ±5 | 4±5 | 6± 4 |
| Initial weight (kg) | 8.4 ± 1.2 | 7.6±1.3 | 8.0 ± 2.0 |
| Initial degree of dehydration (%) | 4.5 + 1.7 | 5.1±2.0 | 5.2± 2.0 |
Through the randomization process, 23 subjects received with 5% solution, 20 with 10% rice and amylase, and 20 with 15% rice and amylase. The groups' gender ratios, preadmission diarrhoeal evacuations, rates of preadmission vomiting, entry weights, and degrees of dehydration are shown in table 2. With the exception of the gender ratio, the variables were strictly comparable among the groups.
The percentages of liquid estimated to cover replacement of accumulated deficit plus the four-hour maintenance are shown graphically in figure 1 (see FIG. 1. Comparisons among the three types of RB-ORS offered during the first 4 hours of rehydration with respect to volume (left) and carbohydrate energy intake (right)). For 5% rice alone, this ranged from 12% to 100% (mean 70 ± 26%); for 10% rice plus amylase, it ranged from 34% to 148% (mean 90 ± 28%); and for 15% rice plus amylase, it ranged from 4% to 205% (mean 80 ± 45%). These were not significant differences. Also, a stepwise increment in metabolizable energy intake from rice starch progressed from 69 ± 30 kcal at 5%, to 174 ± 71 kcal at 10%, to 240 ± 134 kcal at 15%. Normalized for estimated hydrated body weight, the four-hour energy intakes for the respective RB-ORS were 8 ± 4, 23 ± 9, and 33 ± 18 kcal/kg.
Table 3 compares the clinical evolution of the children in terms of number of vomiting episodes, number of diarrhoea! evacuations, and rate of therapeutic failures of the three treatment groups. No statistically significant differences were seen across the regimens. The nonsignificant rate of vomits and the number of therapeutic failures were greater in the 15% solution group than in the other groups, even though the viscosity of the solution was very similar to that of the 10% solution, and the osmolality was much higher.
TABLE 3. Clinical evolution of the patients after 4 hours of treatment according to type of RB-ORS used
| Clinical events | Rice flour concentration (w/v) of RB-ORS | ||
| 5% | 10%a | 15%a | |
| No. of vomiting episodesb | 0 5 ± 1.3 | 05 ± 1.1 | 1.5 ± 2.3 |
| Frequency of diarrhoeab | 2.0 ± 2.2 | 3.4 ± 2.7 | 1.9 ± 2.4 |
| Therapeutic failures | 2/32 (9%) | 1/20 (5%) | 3/20 (15%) |
The children were weighed at the beginning and at the end of the rehydration course (see FIG. 2. Comparison of 4-hour mean weight gains during treatment with the three respective therapies (in grams): 95 ± 110, 197 ± 166, and 121 ± 160. The vertical axis represents the different concentrations of rice (w/v) in the respective RB-ORS concentrations used). For the 5% solution, the weight changes ranged from -1.4% to 4.3% of the estimated hydrated rates (mean 1.1 ± 1.2%), which translated into an average increase for the group of 95 ± 110 g. For 10% rice flour plus amylase, the changes in weight ranged from -0.3% to 8.2% (mean 2.7 ± 2.2%), equivalent to a 197 ± 166 g overall group increment. For the 15% rice flour and enzyme treatment, weight changes varied from -1.25% to 4.90% (mean 1.6 ± 1.8%), representing 121 ± 160 g.
Given the short interval and the dehydrated state of the patients, weight increases can be ascribed to rehydration. Moreover, the equivalency of the estimated weight deficits across groups would make any differences in weight gain attributable to the treatment regimen. Statistical comparison of the percentage weight changes by group did show a significant difference (p <.05).
Since ORS were established for the management of dehydration and fluid loss in acute diarrhoea, morbidity and mortality from this syndrome have been reduced [2, 3]. Dehydration due to diarrhoea, however, is still a major public health concern, principally in developing countries where most of the affected children also have some degree of undernutrition. Taking into account that the energy supply of a WHO ORS based on glucose cannot be increased without producing negative side effects due to elevated osmolality [4], other solutions based on cereals have been proposed both to improve the efficiency of hydration and to protect nutrition status.
Recently, several studies reported on different cereals in ORS [9 18]; especially prominent is rice, which is easily digested and very well tolerated. It was a logical extension to add an amylase to decrease the thickness of solutions at high concentrations to crease energy-rich cereal-based ORS.
In this study, three RB-ORS were used. The 5% RB-ORS (without enzyme), the effectiveness of which was demonstrated in other reports [13, 19, 20], served as a control and standard of comparison. It had a higher viscosity than the other two solutions treated with a -amylase, indicating the effectiveness of the enzyme to liquefy rice emulsions. Moreover, the maximum 290 mOsm/L is well within a tolerable range. The GB-ORS with 90 mmol Na/L recommended by WHO, for example, has an osmolality of 330 mOsm/L.
Cooking time and storage must be considered in relation to the preparation of the solution. They are critical with respect to complete inactivation of the a -amylase enzyme; otherwise continued hydrolysis occurs, and the osmolality rises to a level that would be dangerous to the dehydrated patient. Under strict refrigeration at 4°C, the solution can be used safely during a 72-hour interval. Because of its rich nutrient content, however, care must be exercised with the conditions of cold storage in warm tropical regions.
Recent studies in Peru emphasized the benefits of early introduction of nutrients during rehydration therapy [2123]. The higher the caloric content of the solution, the less the probability of deterioration of nutrition status during or shortly after diarrhoeal dehydration. Although this study was not designed to demonstrate the nutritional impact of the RB-ORS, a rehydration solution containing digestible carbohydrate is assumed to prevent further deterioration of nutrition status by reducing the catabolic state and keeping the energy equation close to balance.
In practice, the 10% RB-ORS provided almost three times more energy than the 5% solution, and the 15% solution provided four times more energy. If it is estimated that the daily requirement of a 10-kg child is about 108 kcal/kg body weight/day for boys and 97 kcal/kg/day for girls [24], only 2 L of a 15% rice solution or 3 L of a 10% solution would provide the usual requirements with some overage for the stress of infection.
At the frontier of an innovative idea, we have opted to proceed slowly and cautiously from in vitro observations [5] to the short-term initiation of therapy. Slight numerical differences were observed with one treatment or another, but no distinct trend toward either net positive or negative clinical effects in rehydration was seen. Thus, in theory, one could sustain rehydration with amylase-incubated rice solutions at 10% or 15% or beyond.
The next phase of evaluation is obvious. The differential ability of the RB-ORS over 8% to achieve successfully the full rehydration of diarrhoeic children must be determined in a quantitative metabolic setting. Thereafter, public health applications at the community level must be undertaken to determine if the complex exigencies of ingredient combination, enzyme incubation and inactivation, and safe handling and storage can all be managed successfully under the conditions of poverty and illiteracy that promote the high frequency of acute diarrhoea in developing countries.
This project was supported in part by a grant from the Applied Diarrheal Disease Research Project of Harvard University by means of a cooperative agreement with the US Agency for International Development. We are very grateful for the consultation provided under that grant by Dr. Kenneth Brown for the conceptual and technical bases of this research. We also thank Dr. Pablo Yurrita from the Guatemalan Institute of Social Security (IGSS), for his technical support in the determination of the osmolality of the solutions. We also acknowledge Lic. Fabiola Micheo and Lic. Carmencita de Arriola from the Central American Institute of Investigations and Industrial Technology (ICAITI) in Guatemala City for their laboratory support in the determination of ORS viscosity. We are grateful to Drs. Eric Rodas, Sarita Tobar, and Carlos Reyes, the medical staff of the metabolic ward from the Pediatric Department of the San Juan de Dios General Hospital, Guatemala City.
