Sucrose polyester in human volunteers.

The recent publication by Kelly et al. (1998) of their study on sucrose polyesters (SPE) raised concerns about the possible gastrointestinal (GI) and nutritional effects of olestra. These concerns were echoed in the popular press (‘Doubts raised over anti-obesity fat substitute’. The Times, 11 August 1998.). Olestra’s potential to cause GI and nutritional effects has been carefully reviewed by the US Food and Drug Administration (FDA), including two public meetings of the FDA Foods Advisory Committee in November 1995 and again in June 1998. The recent comprehensive committee review included results from US marketplace surveillance and from new clinical research. The committee reaffirmed, by an overwhelming majority, their decision that olestra snacks are safe. In light of this favourable review and the acceptance of olestra snacks by US consumers, we would like to make several specific points regarding the publication by Kelly et al. (1998). First, it should be made very clear that Kelly and colleagues tested SPE produced by Unilever which are substantially different from olestra manufactured by Procter & Gamble and approved by the FDA for use in savoury snacks. A significant error in this regard was stated by Kelly et al. (1998) ‘...SPE was approved by the United States Food and Drug Administration for use in savoury snack foods’. Although olestra falls into the broader class of SPE, the olestra which was approved by the FDA has very specific composition requirements which set it apart from other SPE. The FDA’s compositional requirements for olestra include a specification to prevent anal oil-leakage. The liquid SPE used by Kelly and colleagues in the margarine, biscuits, cake, cheese, chocolate spread, peanut butter, and salad cream consumed in the study would not pass this specification, and would therefore cause anal oil-leakage. Therefore, the findings by Kelly et al. (1998) of ‘anal leakage’ in 7⋅2 % of subjects should not be assumed to be relevant to olestra. Further, studies conducted by Procter & Gamble to elucidate the relationship between GI symptoms experienced when eating SPE which is liquid at room temperature and those noted after eating olestra as specified in the US FDA regulation have shown that liquid SPE associated with ‘anal leakage’ are also associated with increased incidence of GI symptoms (data on file, Procter & Gamble). Second, the study by Kelly et al. (1998) involved daily ingestion of large amounts of SPE in a variety of foods that would be eaten at essentially every meal. This regimen differs markedly from use of olestra in savoury snacks which are consumed much less frequently and with only a fraction of meals. Several randomized placebo-controlled studies have specifically addressed whether ingestion of olestra snacks will increase GI symptoms in consumers. Cheskin et al. (1998) studied 1092 subjects eating as much as they wanted (up to 13 ounces (about 368 g)) on a single eating occasion of olestra potato chips or placebo, full-fat chips. There were no significant differences in the occurrence or severity of GI symptoms between the subjects who consumed olestra chips and those who consumed full-fat chips. In another double-blind study, 3181 children, teenagers, and adults were allowed to eat essentially unlimited amounts of olestra snacks for 6 weeks (Sandler, 1998). There was no overall significant difference in the incidence of GI symptoms, with the exception of increased nausea in the full-fat group. The mean number of symptom days during the 6-week period was not different between groups except for the number of days on which ‘more frequent bowel movements’ were reported (3⋅7 v. 2⋅8 d, olestra v. full-fat). There was no difference in the impact of GI symptoms on the daily activity assessment ratings between the two groups. It is worth noting that there were no reports of ‘anal leakage’ in either of these two studies. Third, Kelly et al. (1998) raised the question of whether there are safety concerns in persons with underlying GI disorders. Zorich et al. (1997) conducted a specific study in persons with inflammatory bowel disease that assessed whether olestra may have adverse health effects on potentially sensitive subpopulations with bowel disease. Eighty-nine patients with mild to moderate ulcerative colitis (n 43) or Crohn’s disease (n 46) were randomly assigned to eat 20 g olestra in chips and cookies for 4 consecutive weeks, or equivalent full-fat products. There was no difference with respect to disease activity, and GI symptoms were comparable between the treatment groups. Marketplace surveillance has not revealed any patterns indicating groups of people who are intolerant to olestra. Controlled rechallenge testing of people reporting digestive effects in the marketplace has shown no differences in digestive effects when these people were eating olestra v. full-fat chips and that they were not uniquely intolerant to olestra (Zorich et al. 1998). Based on the results reported by Kelly et al. (1998), it is unclear how they came to the conclusion that there were ‘important deleterious’ GI effects from the SPE. For example, the authors state that no subjects discontinued the study because of adverse effects. The authors state that bowel movement frequency increased from seven bowel movements per week to ten. This is still well within what would be considered a normal range for a healthy adult in the UK. Importantly, the authors monitored ‘general well-being scores’ during the study which were not lower when the participants consumed SPE. In addition, a variety of routine and more sophisticated studies were conducted, i.e. rectal tissue biopsies, small-bowel biopsies, liver function studies, and a large variety of screening blood chemistry and haematologic tests. No health concerns were identified by any of these tests. The reduction in serum carotenoid levels measured in this study when a large variety of SPE-containing foods British Journal of Nutrition (1999), 81, 169–171 169