Olestra (Olean, Procter & Gamble, Cincinnati, OH) is a mixture of the octa-, hepta- and hexaesters of sucrose formed from long-chain fatty acids isolated from edible oils. Olestra is not digested or absorbed intact (Mattson and Volpenhein 1972, Miller et al. 1995) and therefore contributes no calories to the diet. Olestra is approved for replacing 100% of the fat used to prepare savory snacks, such as chips, crisps, extruded puffs, curls and crackers (Federal Register 1996).

Because of its properties, olestra has the potential to reduce the availability from the diet of lipophilic nutrients. This reduction can occur because, under certain circumstances, a portion of those nutrients partitions into the nonabsorbed olestra in the gastrointestinal (GI)³ tract and becomes unavailable to the intestinal micelles (Jandacek 1982). The absorption of water-soluble or weakly lipophilic nutrients will not be affected by olestra because they do not partition into the lipophilic olestra in any significant amounts (Cooper et al. 1997e). Because it is a physical interaction, olestra can affect the absorption of a nutrient only if the two are present at the same place in the gut at the same time. For this reason, eating patterns (i.e., whether foods containing olestra are eaten at the same time or nearly the same time as other foods) and the relative amounts of olestra and nutrients eaten (i.e., olestra-to-nutrient ratio) are important factors in determining the degree to which olestra affects lipophilic nutrient availability.

The effects of olestra on the absorption of essential nutrients have been determined in animal feeding studies and human clinical studies described elsewhere in this supplement (Cooper et al. 1997b and 1997c, Daher et al. 1997a and 1997b, Koonsvitsky et al. 1997, Schlagheck et al. 1997b). Other studies were conducted to show that these effects can be offset by addition of extra amounts of the affected nutrients to olestra foods (Cooper et al. 1997a, Schlagheck et al. 1997a).

The human studies were conducted in normal, healthy 18- to 44-y-old males and females because this age group is estimated to consume the greatest amounts of olestra from savory snacks (Webb et al. 1997). The animal studies were conducted with weanling domestic pigs. The pig was chosen as the animal model in which to conduct the studies for several reasons as discussed in Cooper et al. (1997d). Key among these are the similarities in the processes of digestion and absorption between pigs and humans. In addition, the GI tract of the pig is similar, both functionally and morphologically, to the human GI tract, especially to that of young children. The rapidly growing weanling pig also has extremely high nutrients demands. The domestic pig has been extensively used in nutritional research (Miller and Ullery 1987), including pediatric nutritional research (Moughan et al. 1992).

Hypothetically, in considering nutrient availability (and hence nutritional status), two factors exist that might cause the effects of olestra measured in 18- to 44-y old humans or the weanling pig to be different, or greater, than those that might occur among certain subgroups of the general population. These factors are differences in dietary pattern (i.e., ratio of olestra-to-nutrient intake) and metabolic need (i.e., different nutrient requirements) between the subgroups and those tested in the studies. Dietary pattern can alter the effects of olestra on nutrient availability because olestra affects the absorption of lipophilic nutrients through a physical interaction (i.e., a partitioning of the lipophilic nutrients into olestra in the GI tract). How much nutrient partitions into the olestra, and thus becomes unavailable, depends on how much olestra is present relative to how much nutrient. Therefore, it is important to know whether the relative amounts of olestra and nutrients fed in the studies are as great, or greater than those that might occur within subgroups of the population.

Metabolic need for nutrients might alter the effects of olestra on nutrient availability because the efficiency of nutrient absorption may vary depending on the bodys demand for the nutrient. Therefore, it is important to know whether less nutrient would be absorbed in the presence of olestra when the demand is high relative to the amount absorbed when demand is not as high.

The purpose of this study was to determine whether the dietary patterns and metabolic needs represented in the studies in 18- to 44-y old humans and weanling pigs encompassed those of the general population, and as a consequence, whether the results of the studies can be used to assess the potential for olestra to affect the nutritional status of subgroups of the population not included in the studies. The relevance of the results of the studies to subgroups not included in the studies was assessed by comparing dietary patterns and metabolic needs of subgroups to the dietary patterns used in the studies and the metabolic needs of the study population.

The approach used was to examine published nutrient requirements and nutrient intakes and estimated olestra intakes for subgroups of the population to identify those subgroups having the greatest nutrient requirements and the highest olestra-to-nutrient intakes for nutrients tested in the studies. Then, the requirements and olestra-to-nutrient intake ratios of those subgroups were compared with the requirements and intake ratios covered in the studies. If olestra had no (or minimal) effects on certain nutrients in the studies at olestra-to-nutrient intake ratios and metabolic requirements exceeding those calculated for the subgroups, it was reasoned that olestra would be unlikely to affect those nutrients in the subgroups. Further, for those nutrients found to be affected in the studies, it was concluded that the effects would not be different (i.e., greater) in the subgroups.

METHODS

The NFCS data were reorganized and calculated for age ranges consistent with those used in the estimation of olestra intake from savory snacks (Webb et al. 1997). To do this, the data were transferred from the USDA source, a magnetic tape, into a SAS (SAS Institute, Cary, NC) database and reorganized into the appropriate age groups. The data were then weighted by demographic characteristics and mean intakes of the nutrients of interest were calculated for subgroups.

Nutrient intakes among women and children of low income households were obtained from the Continuing Survey of Food Intake by Individuals (CSFII, U.S. Department of Commerce 1986). Nutrient intakes for pregnant and lactating women were taken from data published by the Institute of Medicine (Institute of Medicine 1990 and 1991). To be conservative, the lowest reported nutrient intakes were used for the subgroups.

Vitamins D and K intakes were estimated from data on the consumption of common foods rich in these vitamins. The amounts of foods consumed that were rich in vitamin D, such as milk, cheese, ready-to-eat cereals and shrimp, or vitamin K, such as broccoli, coleslaw, lettuce, spinach, peanut butter, mayonnaise, margarine, salad oil and coffee, were estimated from the 1977-78 NFCS database (Pao et al. 1982). The vitamin D contents of the foods were taken from Pennington and Church (1985); the vitamin K contents were taken from data compiled by Booth et al. (1993).

To estimate potential olestra intake for subgroups not specifically included in the menu census survey, the following assumptions were made. Lactating women were assumed to have the same intake as pregnant women. Vegetarians were assumed to consume as much olestra as the subgroup with the highest intake, 18- to 44-y-old males. Women and children in low income households were assumed to consume the same amount of olestra as the corresponding age groups of the general population. Low income families, (i.e., families on food assistance) do not spend any more on snack foods than the general population (Fraker 1990 and 1993, Nelson 1979). Therefore, it is unlikely that these families would have a greater exposure to olestra than the general population. Whenever the age groups in the CSFII (used for nutrient intake) and the MRCA survey (used for olestra intake) did not align exactly, the highest olestra intake within an age range was used.

RESULTS

Table 1. Estimated intakes of olestra (O), protein (P), carbohydrate (CHO) and fat (F), and calculated olestra-to-macronutrients intake ratios for subgroups of the population [View Table]

Table 2. Estimated intakes of olestra, water-soluble micronutrients and calculated olestra-to-micronutrient intake ratios for subgroups of the population [View Table]

Table 3. Estimated intakes of olestra, fat-soluble vitamins and -carotene, and calculated olestra-to-micronutrient intake ratios for subgroups of the population [View Table]

Children, teenagers and young adults, low income women and vegetarians potentially have the highest olestra-to-nutrient intake ratios. Specifically, vegetarians have the highest potential olestra-to-protein (0.19 g/g), olestra-to carbohydrate (0.053 g/g), and olestra-to-fat (0.18 g/g) intake ratios, followed by children and teenagers 17 y old of both genders and 18- to 44-y-old males (Table 1).

For water-soluble micronutrients (Table 2), teenagers, 13-17 y old, have the greatest potential olestra-to-folate intake ratio (46 g/mg), followed by 18- to 44-y-old males and vegetarians (43 g/mg). Individuals 17 y old of either gender have relatively high potential olestra-to-folate intake ratios, as do low income women of all ages. Vegetarians have the greatest potential olestra-to-vitamin B₁₂ intake ratio (3.6 g/µg), followed by 13- to 17-y-old females (2.2 g/µg). Vegetarians have the greatest potential olestra-to-calcium intake ratio (15 g/g), followed by low income women, 13- to 44-y-old women, and 18- to 64-y-old males, with ratios of 10-13 g/g. Vegetarians also have the greatest potential olestra-to-iron intake ratio (902 g/g), followed by 13- to 17-y-old females (814 g/g) and 18- to 44-y-old males (764 g/g). Children (age 2-17) have relatively large potential olestra-to-iron ratios as do low income children 1-3 y old and low income women, 35-50 y old. Vegetarians potentially have the greatest olestra-to-zinc intake ratio (1341 g/g), followed by 13- to 17-y-old females (979 g/g), 2- to 5-y-old children (904 g/g) and 18- to 44-y-old males (886 g/g).

For fat-soluble vitamins and -carotene (Table 3), 13- to 17-y-old females and 14- to 44-y-old males have the greatest potential olestra-to-vitamin A intake ratio (11 g/mg), followed by 35- to 50-y-old low income women (9.9 g/mg) and vegetarians, who have a ratio of 9.7 g/mg. Females, 13-17 y old, and low income children, 1-3 y old, potentially have the greatest olestra-to--carotene intake ratio, both 31 g/mg. These subgroups are followed by children 2-12 y of age (27-28 g/mg), low income children 3-5 y of age (28 g/mg), and males 13-44 y of age (27 g/mg). Children, 2-5 y old, females 13-17 y old and vegetarians have the greatest potential olestra-to-vitamin E intake ratio (1.4 g/mg). These subgroups are followed by children 2-12 y of age and males 13-44 y of age, both of whom have ratios of 1.2 g/mg.

The greatest potential olestra-to-vitamin D intake ratio (2.5 g/µg), was found for 18- to 44-y-old females. Females >65 y of age potentially have an olestra-to-vitamin D intake ratio of 1.7 g/µg. Reliable vitamin D intake data were not available for dieters, vegetarians or low income women and children.

Children <2 y of age potentially have the greatest olestra-to-vitamin K intake ratio (45 g/mg), followed by 13- to 17-y-old males (35 g/mg), and 2- to 5-y-old children (34 g/mg). Reliable vitamin K intake data were not available for pregnant or lactating women, dieters, vegetarians or low income women and children.

The greatest olestra-to-nutrient intake ratios fed in the nutritional studies are shown in Table 4. The largest olestra-to-protein (0.33 g/g) and olestra-to-carbohydrate (0.17-0.18 g/g) intake ratios were fed in the 12-wk pig studies (Cooper et al. 1997a and 1997c). The largest olestra-to-fat intake ratio (0.81 g/g) occurred in the fat absorption study in humans (Daher et al. 1997b). The largest olestra-to-micronutrient intake ratios fed to humans were in the two 8-wk studies (Schlagheck et al. 1997a and 1997b) with the exception of the olestra-to-vitamin A intake ratio (96 g/mg); that occurred in the retinyl palmitate absorption study (Daher et al. 1997a).

Table 4. Largest olestra-to-nutrient intake ratios fed in the human and pig studies [View Table]

The largest olestra-to-nutrient ratios consumed in the nutrition studies are compared with the highest potential ratios calculated for any of the subgroups of interest in Table 5. The highest olestra-to-protein, olestra-to-carbohydrate and olestra-to-calcium intake ratios were consumed by the pigs. Both human and pig data were available for all macro- and micronutrients with the exception of calcium, for which the only relevant data came from pigs, and -carotene, for which the only relevant data came from the human studies. Although the olestra-to-nutrient intake ratios for folate, vitamin B₁₂, vitamin A and vitamin E consumed by the pigs were greater than those consumed by human subjects, the ratios from the human studies are used for comparison with the maximum calculated ratios for the subgroups for those micronutrients. Intake ratios for protein, carbohydrate, calcium and vitamin A consumed by the pigs were used because there were more direct measures of the potential of olestra to affect these nutrients in the pig studies than in the human studies (e.g., growth and feed efficiency, bone concentration of calcium, liver stores of vitamin A). The olestra-to-fat intake ratio consumed by the human subjects in the fat-absorption study was used for the same reason.

Table 5. Largest olestra-to-nutrient intake ratios fed in the olestra nutritional studies, the largest ratio calculated for any subgroup, and the ratio of the two [View Table]

The largest olestra-to-nutrient intake ratio consumed in the nutrition studies exceeded the largest olestra-to nutrient intake ratio calculated for any subgroup for all nutrients except calcium. The olestra-to-calcium intake ratio is low in pigs because of the extremely high requirement of the rapidly growing pig for calcium. For calcium, the ratio fed the pigs was 0.6 times the ratio calculated for vegetarians, essentially equal to the ratios calculated for children and females 45-64 y old, and exceeded the ratios calculated for the elderly, pregnant and lactating women, and low income children.

Olestra-to-nutrient intake ratios were calculated for the identified subgroups by using estimated 90th-percentile olestra intakes and average nutrient intakes. A more conservative evaluation might be to use nutrient intakes for 90th-percentile snack eaters, individuals who have high intakes of snack foods, possibly at the expense of other foods. However, 90th-percentile snack eaters generally have higher total food consumption, which increases the intake of most nutrients, with concomitant reductions in olestra-to-nutrient intake ratios (data not shown).

Table 6. Body weights and protein and energy requirements for human subgroups and for pigs used in the olestra feeding studies [View Table]

Table 7. Micronutrient requirements for human subgroups and the pigs used in the olestra feeding studies [View Table]

The requirement of 7-wk-old pigs (the age of the pigs at the start of the studies) for protein exceeds that of children with the greatest requirement (1-6 y of age) by more than 11-fold (Table 6). For 19-wk-old pigs, the age of the pigs at the end of the 12-wk studies, the factor was almost fivefold. When expressed in MJ/d, the energy requirement of 7-wk-old pigs exceeds that of the subgroup with the greatest need (7- to 10-y-old children) by a factor of 1.6; when expressed in MJ/(kg·d), the requirement of the 7-wk-old pig exceeded that of the subgroup with the greatest need (children 1-3 y old) by a factor of 2.6.

The requirements of the 7-wk-old pig for vitamin B₁₂, calcium, iron and zinc exceed those of the subgroup with the greatest need, children 1-10 y of age, for all these micronutrients by severalfold. The requirement of the 7-wk-old pig for folate exceeds that of the subgroups with the greatest needs, pregnant and lactating women, by a factor of ~3-4.

For the fat-soluble vitamins A, E and K, the requirements of the 7-wk-old pig exceed those of the subgroup of children with the greatest requirements by factors of 1.3, 1.2 and 3.6, respectively. For vitamin D, the requirement of the young pig is greater than the need of 7- to 10-y-old children but slightly less than the need of 1- to 6-y-old children.

DISCUSSION

The comparison showed that the nutrient requirements of the test subjects exceeded those of the identified subgroups for all nutrients evaluated. Similarly, the olestra-to-nutrient intake ratios consumed by test subjects in one or more of the studies were greater than the ratios calculated for the identified subgroups for all nutrients except calcium. Therefore, the findings on the effects of olestra on nutrient availability and on the amounts of fat-soluble vitamins required to offset the olestra effects on these essential nutrients are relevant to subgroups of the population.

Although some of the identified subgroups had potential olestra-to-calcium intake ratios greater than any consumed in the human and pig studies, the calcium nutritional status of any subgroup of the population is unlikely to be affected by olestra. Studies in the pig (Cooper et al. 1997a-c) and in humans (Schlagheck et al. 1997a and 1997b) have provided ample evidence that olestra does not affect the absorption of water-soluble nutrients. Specifically, bone and serum calcium and phosphorus concentrations, serum parathyroid hormone concentration, and bone ash content were unaffected in pigs eating up to 155 g/d olestra as part of a diet that provided calcium at the NRC requirement for swine (NRC 1988). This daily intake of olestra is more than 14 times the estimated chronic 90th-percentile intake of olestra for any subgroup of the population (Webb et al. 1997).

Data collected in the domestic pig on the effects of olestra on nutrient availability and utilization can be used to make inferences about these same factors in humans because of the high degree of similarity between pigs and humans in GI structure and function, including processes of nutrient digestion and absorption, as discussed in Cooper et al. (1997d). In addition, the nutrient requirements of the weanling pig are similar to or greater than requirements of human adults and children for fat-soluble vitamins (Miller and Ullrey 1987, Pond and Houpt 1978). Numerous researchers have shown that the weanling pig is particularly appropriate for generating nutritional information relevant to children (Cooper 1975, Glauser 1966, Kidder and Manners 1978, Mellor and Cockburn 1986, Miller and Ullrey 1987, Moughan et al. 1992).

The findings on the effects of olestra on nutrient availability generated in 18- to 44-y-old subjects are relevant to the identified subgroups for two reasons. First, nutrient digestion and absorption processes in the identified subgroups do not differ markedly from those in the normal healthy adults used in the clinical trials. For example, the processes of nutrient digestion and absorption are virtually fully developed in children by 9-12 mo of age (Grand et al l976, Hamosh 1979). These processes remain essentially unchanged throughout childhood into maturity, in pregnancy, lactation and when vegetarian diets are eaten (Calkins 1993, Schneeman and Gallaher 1993, Tasman-Jones 1993). The second reason that the data from the human clinical studies can be applied to subgroups not included in the study is that the effect of olestra on nutrient absorption does not involve alterations in any of the processes of nutrient digestion or utilization. The GI tract simply serves as a vessel in which olestra and nutrients physically interact.

Although the process of identifying subgroups who might be at potentially greater nutritional risk than the test subjects did not include the elderly, the applicability of the findings to this group is nevertheless of interest. The overall conclusion that can be drawn is that the elderly will not be placed at increased nutritional risk by eating olestra for the following reasons. There is no evidence of maldigestion or malabsorption of fat, protein or carbohydrate in older adults on usual diets, in the absence of clinically evident disease (Holt and Balint 1993, Hosoda, 1992, Rosenberg et al 1989, Russell 1992). Because the mechanisms of fat absorption are intact in the healthy elderly, the absorption of fat-soluble vitamins should be unaffected by aging. Therefore, the effects of olestra on the availability of fat-soluble vitamins should not be different in the elderly than in the test subjects. Moreover, the restoration of tissue concentrations of the vitamins by adding extra amounts to olestra will be as effective in the elderly as in the test subjects. Because the availability of water-soluble micronutrients is not affected by olestra, there should be no adverse effect of olestra on these micronutrients in elderly persons even if they have a reduced absorptive capacity for them.

This assessment shows that the findings from the human and pig studies are relevant to subgroups not included in the tests. Further, this assessment indicates that under the same dietary conditions, the nutritional status of the identified subgroups should not be affected by olestra differently than that of the test subjects.