Olestra (Olean, Procter & Gamble, Cincinnati, OH) is a mixture of hexa-, hepta- and octaesters of sucrose formed from fatty acids isolated from edible oils. Olestra has the organoleptic and thermal properties of regular fat (Bernhardt 1988, Kester 1993) but is not digested or absorbed (Mattson and Volpenhein 1972, Miller et al. 1995); therefore it contributes no calories or fat to the diet. Olestra is approved for replacing 100% of the fat used in preparing savory snacks such as potato chips and crackers (Federal Register 1996).

Because olestra is lipophilic and nonabsorbed, it has the potential to interfere with the absorption of lipophilic nutrients (Jandacek 1982). This interference occurs because a portion of the lipophilic nutrients present in the gastrointestinal (GI)⁵ tract with olestra partitions into the olestra and thus becomes unavailable to the intestinal micelles to be transported to absorptive sites. The absorption of water-soluble nutrients is not affected because such molecules do not partition into the lipophilic olestra.

Olestra has been shown to affect the absorption of the fat-soluble dietary components in humans (Jones et al. 1991a and 1991b, Koonsvitsky et al. 1997, Schlagheck et al. 1997) and in pigs (Cooper et al. 1997a-c, Daher et al. 1997a). However, these studies have shown no effect of olestra on water-soluble nutrients, vitamin K function or overall vitamin D status. Other studies have shown that olestra has no significant effect on the absorption of preformed vitamin A or triglyceride (Daher et al. 1997b and 1997c).

It has been shown in studies described elsewhere in this issue that it is possible to offset the effect of olestra on the availability of fat-soluble vitamins by adding extra amounts of the affected vitamins to foods prepared with olestra (Cooper et al. 1997a and 1997b, Koonsvitsky et al. 1997).

The primary purpose of this study was to determine the amounts of vitamins D and E required to offset the effect of olestra on the absorption of these vitamins. Because vitamin A stores can be measured directly in the pig, the amount of vitamin A required to offset the effect of olestra was determined from pig studies (Cooper et al. 1997a and 1997b). Other purposes of the study were to determine the olestra dose response on the serum concentration of 25-hydroyergocalciferol [25(OH)D₂] when subjects consume vitamin D at the recommended dietary allowance (RDA) and to confirm that olestra does not affect the functional status of vitamin K or the absorption of water-soluble micronutrients.

To maximize the opportunity for olestra to affect the absorption of the nutrients, the subjects were required to eat olestra at every meal every day and were requested not to eat other foods between meals. In addition, daily intakes of olestra substantially greater than expected intakes of olestra from savory snacks were used.

SUBJECTS AND METHODS

Table 1 shows the group designations, olestra doses and target amounts of vitamins A, D and E added to the test foods. Group designations indicate the olestra dose and the relative amounts of vitamins A and E added: L = low, M = medium and H = high. Only two levels of vitamin D were added; those were added to the diets containing low and high levels of the other two vitamins. The groups receiving 20 g/d olestra with graded levels of vitamins (20 g/d L, 20 g/d M, 20 g/d H) were used to determine the amounts of vitamins D and E that would be required to offset the olestra effect. The groups receiving 8, 20 and 32 g/d olestra with the same levels of added vitamins (8 g/d M, 20 g/d M, 32 g/d M) were used to determine the dose response of olestra on vitamin D, carotenoids, vitamin K, vitamin B₁₂ absorption and iron status. Because depletion of vitamin A tissue stores in normal healthy individuals with inadequate vitamin A intake can take months (Olson 1984), and because olestra has no significant effect on the absorption of retinyl palmitate (Daher et al. 1997b), changes in serum retinol concentration resulting either from olestra intake or from the added retinyl palmitate were not expected. However, serum retinol concentrations were measured to assure that the subject's vitamin A status was adequate.

Table 1. Test groups, olestra doses and amounts of vitamins A, D and E added to the diet1 [View Table]

Exclusion criteria included pregnancy or lactation, chronic use of drugs that can potentially interfere with vitamin absorption, use of tanning booths or high sun exposure within the previous 2 mo, physician-recommended diet restrictions or greater than average caloric need. Demographics and randomization parameters for subjects entering the study are shown in Table 2.

Table 2. Demographics and randomization parameters for groups consuming 0, 8, 20 or 32 g/d olestra with graded levels of vitamins A, D and E [View Table]

A core menu provided 9205 kJ/d (2200 kcal/d) of energy. This menu was adjusted if necessary to provide more or less energy to meet the needs of individual subjects. A subject was assigned to one of seven energy-intake levels ranging from 7531 kJ/d (1800 kcal/d) to 12,522 kJ/d (3000 kcal/d) at intervals of 837 kJ/d (200 kcal/d). The assignment was made on the basis of resting energy requirement as estimated by the Harris-Benedict equation, and an activity factor ranging from 25 to 50% of the subject's basal metabolic rate (Alpers et al. 1983). The subjects were fed to maintain their weight within 5% of their starting weight. Adjustments to the diet were made on the basis of body weights obtained weekly.

The diet was developed to supply 50% of energy from carbohydrate, 35% from fat and 15% from protein. Intakes of vitamin A, D, E and K were controlled at 80-120% of the RDA for 25- to 50-y-old males (NRC 1989). Unlike subjects in the dose-response study, the subjects did not receive a vitamin D supplement.

The intake of iron was targeted at 12.5 mg/d, 69% of the RDA for the female subjects and 125% for the male subjects. This value was chosen so that the male's intake would not be excessive and the female's intake would not be so low as to place them at risk of iron deficiency. The intake of vitamin B₁₂-containing foods was allowed to fluctuate as necessary to meet the target for iron. Digestible fat, carotenoid and calcium intakes were held constant across the groups. To keep digestible energy content constant across the groups, the amounts of triglyceride displaced by olestra were added back as butter, margarine or vegetable oil. Dietary concentrations of nutrients were determined and the data were analyzed as described previously (Schlagheck et al. 1997).

To remain on study, the subjects were required to eat at least 90% of the meals and to consume at least 90% of the olestra or placebo food items. If a subject missed more than one meal during the 2 d before a scheduled blood draw, or missed the dinner of the evening before the blood draw, data from that subject were excluded from the database for that time point.

Food consumption was determined by weighing the items served and the portions remaining after a meal. Olestra intake was determined from the amounts of olestra foods consumed and the analytically determined amounts of olestra in those foods.

Table 3. Specimen collection and measurement schedule for groups consuming 0, 8, 20 or 32 g/d olestra with graded levels of vitamins A, D and E [View Table]

Urine collections (24-h) were made biweekly and the total volume was determined. Aliquots were stored on dry ice until analyzed for -carboxyglutamic acid (Gla), zinc and creatinine. Vitamin B₁₂ absorption was measured at base line and at the end of the study. All subjects were weighed and the females were given a pregnancy test weekly.

Measurements used to determine the effect of olestra and added vitamins on the status of the fat-soluble nutrients were the same as those used in the dose-response study as was the method of measuring vitamin B₁₂ absorption (Schlagheck et al. 1997). In addition to serum iron concentration and hematology parameters, serum concentration of ferritin and total iron-binding capacity (TIBC) were used as measures of iron status.

Any undesirable symptom or change in health, reported either voluntarily or in response to a daily question about whether any changes in health had occurred, was recorded as an adverse event and was followed until the condition resolved. If the subject requested or the clinical staff judged it appropriate, the subject was seen by a physician. If the undesirable symptom or condition was GI-related, the subject was given a form describing eight common GI symptoms and requested to record the severity (1 = mild, 2 = moderate, 3 = severe), number of episodes and duration of the symptom(s).

To determine if vitamin B₁₂ absorption was affected by olestra, differences from base line were analyzed by paired t tests. Dose-response effects were determined by linear regression of the data from appropriate groups. Nutrient intakes, clinical chemistry, hematology and urinalysis data were analyzed as described in Schlagheck et al. (1997).

The amounts of vitamins D and E required to restore serum concentrations of these vitamins to control concentrations were calculated by linearly regressing the mean serum concentrations of 25(OH)D₂ and -tocopherol, respectively, on the dietary concentrations of the vitamins, for groups receiving 20 g/d olestra with different levels of the vitamins. The restoration level was taken as the inverse solution of the regression when the serum concentration of the particular vitamin was equal to the control concentration.

Before regressing the serum -tocopherol concentrations, a two-way ANOVA was conducted using base-line values as the covariate. This was done because the mean concentration for the control group at base line was lower than the means for the 20 g/d olestra groups, although the differences were not significant. Without this adjustment, it would have been impossible to determine the amount of vitamin E required to offset the olestra effect.

All analyses were conducted at the two-tailed 0.05 significance level using SAS Version 6.07 software (SAS Institute, Cary, NC).

RESULTS

All but two subjects consumed 95% of the meals; those two consumed 90 and 93%, respectively. Four subjects did not meet the consumption requirements for the 2 d preceding the wk-6 blood draw. Data from these subjects were excluded from analyses for that time point.

Mean daily intakes of olestra, total energy, macronutrients and micronutrients, averaged across the 8 wk, are shown in Table 4. Daily olestra intakes were within 5% of target; intakes for the 20 g/d groups were not significantly different. Energy intake did not differ significantly across the groups. The contributions from protein, fat and carbohydrate were about 16, 31 and 54%, respectively. Cholesterol intake was constant across the groups.

Table 4. Group mean daily intake of olestra and nutrients for groups consuming 0, 8, 20 or 32 g/d olestra with graded levels of vitamins A, D and E averaged across 8 wk1 [View Table]

No significant differences in average daily intake of micronutrients were observed among the groups. Intakes of micronutrients targeted to be within 20% of the RDA fell within that range. Vitamin A intakes for the 20 g/d H and 32 g/d M groups were exceptions; for those groups, the intakes were 121 and 124% of the RDA, respectively. For all groups, vitamin B₁₂ intake was ~175% of the RDA as a result of providing foods rich in vitamin B₁₂ in order to meet the iron intake target.

Total vitamin A intake increased slightly with increasing olestra dose because the amount of corn oil margarine added to the diet had to be increased with the olestra dose to keep triglyceride intake constant; corn-oil margarine contains higher amounts of vitamin A than fat. Vitamin D intake also increased slightly with increasing olestra dose because the corn oil margarine was fortified with vitamin D₃.

The average daily iron intake for the female subjects was about 10.9 mg, ~61% of the RDA for 18- to 44-y-old females; for the male subjects it was ~13.8 mg/d, or ~138% of the RDA for 18- to 44-y-old males (NRC 1989). Differences in iron intake among the groups were not significant for either males or females.

Subjects in all groups, including the placebo group, reported a variety of common GI symptoms. The symptoms included abdominal discomfort (e.g., gas, cramping, bloating or nausea), flatulence, changes in stool consistency (e.g., soft or loose stools) and fecal urgency. The symptoms were transient in nature, abating and recurring. The average severity (1 = mild, 2 = moderate, 3 = severe) and the percentage of possible symptom-days are shown in Table 5 for all symptoms reported, and specifically for diarrhea and cramping, the symptoms of most concern. Symptom-days, a day in which one or more symptoms were experienced with more than usual frequency, as reported by the subjects, were used to characterize the symptoms because of their intermittent nature. The maximum possible number of symptom-days is obtained for a given symptom by multiplying the number of subjects per group by the number of days in the study. The percentage of symptom-days for all symptoms was dose responsive with respect to olestra intake; the average severity was not, nor was the percentage of symptom-days for diarrhea or urgency.

Table 5. Average severity and percentage of symptom-days for all gastrointestinal (GI) symptoms, and diarrhea and cramping, reported by subjects consuming 0, 8, 20 or 32 g/d olestra [View Table]

Olestra did not affect serum concentrations of total lipids, cholesterol or triglycerides. Week 0 and 8 data are shown in Table A in the Appendix. As expected, neither olestra nor added vitamin A affected serum retinol concentrations (Table B in the Appendix).

Vitamin D. The serum concentrations of 25(OH)D₂ were low (3-6 nmol/L) for all subjects at wk 0, probably because dairy products in the area where the study was conducted (Chicago, IL) are fortified primarily with vitamin D₃ (Table 6). The mean concentration for the control group increased about twofold during the study because the milk provided with the diet was fortified with vitamin D₂.

Table 6. Serum 25-hydroxyergocalciferol [25(OH)D2] concentration for groups consuming 0, 8, 20 or 32 g/d olestra with graded levels of vitamins A, D and E1 [View Table]

Serum 25(OH)D₂ concentrations increased as the amount of vitamin D₂ in the diet was increased. The values for the 20 g/d L (0.14 µg vitamin D₂/g olestra) and 20 g/d H (0.58 µg vitamin D₂/g olestra) groups were significantly greater at all time points than the value for the 20 g/d M group, whose diet had no added vitamin D₂ (Table 6). Serum 25(OH)D₂ concentration measured at wk 6 for the groups receiving 20 g/d olestra with added vitamin D₂ are shown in Figure 1A; results for wk 8 are shown in Figure 1B. For each time point, the serum 25(OH)D₂ concentrations encompassed the mean concentrations for the control group, indicated by dashed horizontal lines in the figures. The following equations were obtained by regressing the mean serum 25(OH)D₂ concentrations on the amounts of added vitamin D: These equations are represented by the solid lines in Figures 1A and 1B.

Table 7. Serum 25-hydroxycholecalciferol [25(OH)D3], total 25-hydroxyvitamin D [25(OH)D] and 1,25-dihydroxyvitamin D [1,25(OH)2D] concentrations for groups consuming 0, 8, 20 or 32 g/d olestra with graded levels of vitamins A, D and E1 [View Table]

Inserting the mean serum 25(OH)D₂ concentrations for the control group (11 nmol/L at wk 6 and 10 nmol/L at wk 8) into the equations and solving for µg vitamin D/g olestra produced a value of 0.07 µg vitamin D/g olestra for wk 8 and 0.05 µg vitamin D/g olestra for wk 6. The average of the two, 0.06 µg vitamin D/g olestra, was taken as the amount of vitamin D required to restore serum 25(OH)D₂ concentration to control concentration (i.e., the restoration level).

Olestra affected serum 25(OH)D₂ concentrations weakly as illustrated by the data from the 8 g/d M, 20 g/d M and 32 g/d M groups (Table 6). The decreasing trend in serum 25(OH)D₂ concentration with increasing olestra dose in the 20 g/d L, 20 g/d M and 20 g/d H groups was significant only at wk 6. 

Table 8. Serum -tocopherol concentration adjusted for base line and normalized with respect to serum lipids for groups consuming 0, 8, 20 or 32 g/d olestra with graded levels of vitamins A, D and E1 [View Table]

No significant differences were found among the groups for serum concentrations of 25-hydroxycholecalciferol [25(OH)D₃] (Table 7 and Table C in the Appendix). The serum concentration of 25(OH)D₃ decreased for all groups during the study.

Differences in serum total 25-hydroxyvitamin D concentrations [25(OH)D] among the groups reflected differences in serum 25(OH)D₂ concentrations produced by the added vitamin D-2 (Table 7 and Table D in the Appendix). Because 25(OH)D₃ contributed >80% of serum total 25(OH)D concentration, and declined during the study, serum 25(OH)D concentration declined in all groups.

No significant differences were found among the groups for serum 1,25-dihydroxyvitamin D concentration [1,25(OH)₂D] (Table 7).

Vitamin E. At wk 0, the serum -tocopherol concentration for the control group was 20.9 µmol/L (Table E in the Appendix), lower than the values for the other groups (22.8-25.5 µmol/L). Normalizing with respect to serum total lipids did not change this situation (Table F in the Appendix). Statistical analysis showed that the base line was a significant covariate; therefore the data were adjusted. Serum -tocopherol concentrations adjusted for base-line differences and normalized with respect to serum lipids are shown in Table 8.

Addition of vitamin E to the diet increased serum -tocopherol concentration. Figure 2A shows serum -tocopherol concentrations measured at wk 6 for the groups given 20 g/d olestra with 1.5, 2.4 or 3.3 mg d--tocopheryl acetate (TA) added per gram of olestra, the 20 g/d L, 20 g/d M and 20 g/d H groups. Results obtained at wk 8 are shown in Figure 2B. The serum -tocopherol concentrations for the control group are shown by dashed horizontal lines in the two figures.

The following equations were obtained by regressing the mean serum concentrations on the added amounts of TA:

Table 9. Plasma des--carboxyprothrombin concentration, urinary -carboxyglutamic acid (Gla) excretion and plasma prothrombin concentration for groups consuming 0, 8, 20 or 32 g/d olestra with graded levels of vitamins A, D and E1 [View Table]

Solving these equations for the restoration levels, as done for vitamin D, produced values of 1.9 and 2.2 mg TA/g olestra. The average of these two values, 2.1 mg TA/g olestra, was taken as the amount of vitamin E required to restore serum -tocopherol concentration to control value.

Serum -tocopherol concentration not normalized with respect to serum lipids showed the same effects as the normalized values (Table G in the Appendix).

Vitamin K. No significant effects of olestra were found for plasma concentrations of des--carboxyprothrombin as measured by the PIVKA-II (protein induced by vitamin K absence-factor II) assay (Table 9 and Table H in the Appendix). During the study, Gla excretion declined in all groups, including the control group; however, no significant intergroup differences and no significant trends with olestra dose were found at any time point (Table 9 and Table I in the Appendix). This was also true when the data were expressed as percentage change from base line (data not shown).

No significant differences were found among the groups for plasma prothrombin concentrations (Table 9 and Table J in the Appendix), PT or PTT (Tables K and L in the Appendix).

Olestra reduced serum phylloquinone concentration in a dose-responsive manner (Table 10). At wk 8, the concentrations for the 8 g/d M, 20 g/d M and 32 g/d M groups were 64, 60 and 53% of control, respectively.

Carotenoids. Olestra reduced serum -carotene concentration (Table 11). At wk 8, the concentrations for 8 g/d M, 20 g/d M and 32 g/d M groups were 42, 4 and 39% of control, respectively. Olestra also reduced the serum concentrations of -carotene, lycopene, lutein (+ zeaxanthin) and total carotenoids (Tables M-P in the Appendix). The effect on lutein (+ zeaxanthin) was less than the effect on the other carotenoids; for example, serum lutein (+ zeaxanthin) concentration for the group given 8 g/d olestra was 73% of control. Serum carotenoid concentrations normalized with respect to serum lipids showed the same effects with respect to olestra dose (data not shown).

Iron. Measures of iron status were analyzed separately for male and for female subjects. The groups showed no significant differences for serum ferritin concentrations for either male or female subjects, with one exception: the concentration measured at wk 4 for female subjects in the 32 g/d M group was significantly less than control (Table 13 and Table Q in the Appendix). Serum ferritin concentrations for all olestra groups were less than the control value at wk 0 for both male and female subjects and remained so throughout the study; however, the differences were not significant. For males, serum ferritin concentrations at base line ranged from 87 to 153 µg/L, within the normal range of 19 to 370 µg/L established by the laboratory. For females, the values ranged from 23 to 116 µg/L, within but, for most groups, at the lower end of the normal range of 10 to 155 µg/L.

Ferritin concentration declined for both males and females in all groups during the study. Regression analysis of the data, expressed either in absolute units or as a percentage of base line, showed no significant differences in the rate of decline between olestra and the control groups for either males and females.

No significant differences in TIBC were found among the groups for either male or female subjects (Table 13 and Table R in the Appendix). A significant increasing trend in TIBC with increasing olestra dose was noted at wk 8 in both males and females.

No significant differences among the groups were found in serum iron concentration for either male or female subjects (Table 13 and Table S in the Appendix). A significant decreasing trend in serum iron concentration with increasing olestra dose was found at wk 8 for the female subjects.

The percentage of transferrin saturation was calculated from serum iron and TIBC data as follows: % saturation = (serum iron concentration)/(TIBC) × 100. No significant intergroup differences in percentage transferrin saturation were noted at any time point for either males or females (Table T in the Appendix). Significant decreasing trends in transferrin saturation with increasing olestra dose were noted for the female subjects at wk 6 and 8.

DISCUSSION

The relative contributions of the two sources of vitamin D, dietary and UV-induced, to overall vitamin D status were realistic in this study. The subjects attained ~20% of their overall vitamin D status from the diet, a contribution at least as great as that measured in individuals living in northern latitudes during the winter (Jones 1978, Jones et al. 1991b).

The amount of vitamin E required to offset the effect of olestra on vitamin E status, 2.1 mg -tocopherol equivalents (TE)/g olestra, determined in this study was the same as that determined from liver and serum vitamin E concentrations in pigs (Cooper et al. 1997a). This agreement supports the appropriateness of the pig as a model in which to conduct nutritional studies as well as providing evidence that serum -tocopherol concentration is a reliable measure of vitamin E status.

No safety concerns are posed by the amounts of vitamins D and E that must be added to the diet to offset the effects of olestra because tissue concentrations of the vitamins are not increased relative to the situation in which olestra is not eaten. The extra amounts of the vitamins simply maintain tissue concentrations at the level they would have been had olestra not been eaten.

One of the purposes of this study was to determine whether the amount of vitamin D eaten by the subjects in the dose-response study influenced the effect of olestra on serum 25(OH)D₂ concentration measured in that study (Schlagheck et al. 1997). Comparison of the olestra effect on serum 25(OH)D₂ concentration in the two studies shows that it did not. In the present study, in which the subjects ate ~5 µg/d of vitamin D, 8 g/d olestra reduced serum 25(OH)D₂ concentration by ~20%. In the dose-response study, in which the subjects ate ~25 µg/d of vitamin D₂, 8 g/d olestra reduced serum 25(OH)D₂ concentration by ~22%.

The decline with time in serum 25(OH)D₃ concentration observed for all groups in this study was most likely a seasonal effect. The study was conducted from October to December, when the cutaneous synthesis of vitamin D was declining from the summer high (Poskitt et al. 1979). Because vitamin D-3 contributed ~80% of total vitamin D status, the changes in serum 25(OH)D₃ concentration produced corresponding changes in serum 25(OH)D. However, olestra had no significant effect on overall vitamin D status as reflected by any significant olestra-related changes in serum total 25(OH)D concentration or serum 2,25(OH)₂ concentration.

The effects of olestra on serum carotenoids and phylloquinone concentrations were in agreement with those observed in the dose-response study (Schlagheck et al. 1997). Also, K function was unaffected in this study as observed in the dose-response study, in other human studies (Jones et al. 1991a, Koonsvitsky et al. 1997) and in the pig studies (Cooper et al. 1997a, 1997b and 1997c).

Data from this study support the notion that serum phylloquinone concentration primarily reflects phylloquinone intake during the 12-24 h preceding the measurement. In the dose-response study, the 6-d rotating menu provided substantially different amounts of phylloquinone for the day before the blood draws, and serum phylloquinone concentration for the control group differed as much as threefold at the various time points (Schlagheck et al. 1997). In this study, the 7-d rotating menu provided the subjects with equivalent amounts of phylloquinone on days preceding all blood draws, and serum phylloquinone concentration was essentially constant in the control group at the four time points, ranging from 1.1 to 1.3 nmol/L.

As in the dose-response study, the effects of olestra on the fat-soluble vitamins and carotenoids observed in this study are greater than those likely to occur when olestra snacks are eaten in free-living conditions because of the frequency with which the olestra foods and other foods were eaten together, as well as the daily amount of olestra eaten. The degree to which dietary pattern can influence the effect of olestra on the availability of fat-soluble nutrients was illustrated and discussed elsewhere in this supplement (Cooper et al. 1997d, Schlagheck et al. 1997) and does not require further discussion here.

This study provides additional evidence that olestra does not affect the absorption of water-soluble nutrients. In the dose-response study, clinical chemistry measurements showed a dose-responsive decrease in serum iron concentration for male subjects (Schlagheck et al. 1997). To confirm that this observation did not represent an effect of olestra on the availability of iron, serum ferritin was measured in the present study, in addition to serum iron and TIBC. The concentration of circulating ferritin is the first iron-related parameter to change when iron status is changed (Cook and Skikne 1989, Fairbanks and Beutler 1988, Walters et al. 1973).

Declines in serum ferritin concentration were observed in this study; however, they occurred in all groups, at the same rates, and therefore were not related to olestra ingestion. The most likely reason for the changes in serum ferritin was the repeated blood draws. Others have reported decreases in iron stores as a result of phlebotomy (Gallagher et al. 1989, Walters et al. 1973) or blood donations (Finch et al. 1977, Simon et al. 1981, Skikne et al. 1984). Over the course of the present study, ~525 mL of blood was taken from each subject for a loss of ~211 mg of iron for the males and ~195 mg for the females. Total iron stores are ~770 mg in adult males and ~210 mg in adult females (Walters et al. 1973); therefore the males lost ~28% of their stores and the females lost >90%. At the same time, iron intake by the females was ~0.73 RDA; by the males, ~1.4 RDA.

Changes noted in other measures of iron status were also consistent with the phlebotomy effect on iron stores. Serum iron concentration generally decreased, TIBC increased and percentage of transferrin saturation decreased during the study. The changes were greater in females than in males. These parameters showed no consistent changes across the various time points or between male and female subjects and therefore do not indicate an olestra effect on iron status.

As in the dose-response study (Schlagheck et al. 1997), GI symptoms were monitored in this study by means of questionnaires given to subjects who reported an undesirable symptom or condition that was related to the GI tract. The kinds of symptoms reported were similar to those reported in the dose-response study. In both studies, a significant number of symptoms were related to decreased stool consistency (soft, loose). The presence of large amounts of olestra in the bowel is expected to affect stool consistency. This happens because olestra interferes with the formation of firm stools, probably in the same way that substances such as liquid petrolatum and olive oil soften stools (Curry 1986). In extreme cases, this effect can produce stools that are perceived and reported by the subjects as diarrhea; however, these "diarrhea-like" stools are not accompanied by water or electrolyte loss commonly found with true diarrhea.

The percentage of days on which GI symptoms were experienced with more than usual frequency tended to be greater than observed in the dose-response study; this was true for all olestra groups as well as the placebo group and thus not olestra related. The average severity of the symptoms was essentially the same in the two studies. The symptoms did not affect protocol compliance; neither consumption of olestra nor meal attendance was affected. Only one subject withdrew from the study because of GI symptoms and that was an upper GI symptom. Further, the symptoms did not affect the accomplishment of study objectives. The tissue concentrations of vitamins D and E responded to addition of extra amounts of those vitamins to the diet. Other findings were in agreement with those from other olestra clinical studies (Koonsvitsky et al. 1997, Schlagheck et al. 1997) and from pig feeding studies (Cooper et al. 1997a, 1997b and 1997c).

GI symptoms experienced by individuals eating olestra snacks in real life are likely to be considerably fewer than those reported in this study because the pattern of eating snack products in real life is different than the pattern used in the study. In this study, olestra snacks were eaten at every meal, which means that olestra was always present in the GI tract in significant amounts. In a study in which potato chips prepared with either olestra or triglyceride were eaten over a 5-mo period, at the choice of the head of the household and in habitual amounts, <1% of the subjects who ate either kind of chips voluntarily reported GI symptoms (Lawson et al. 1997). The individuals who ate olestra chips (n = 2,327) reported no more symptoms than the individuals (n = 1,030) who ate chips prepared with triglyceride.

ACKNOWLEDGMENTS