Food Safety: General Information

Center for Science in the Public Interest

June 10, 1998

Dockets Management Branch (HFA-305)
U.S. Food and Drug Administration
Room 1-23
12420 Parklawn Drive
Rockville, MD 20857

Re: Docket No. 94G-0082

Dear Sir or Madam:

    In 1994, Nabisco filed a petition (GRASP 4G0404) with the FDA seeking affirmation that its salatrim modified fat is generally recognized as safe (GRAS).1   Although the deadline for filing comments on Nabisco’s petition has passed, the Center for Science in the Public Interest (CSPI) requests the FDA to accept and consider these comments, particularly since the FDA has not yet acted on the petition.

    CSPI’s comments are pertinent at this time because Nabisco has not waited for the FDA to affirm salatrim as GRAS, but has declared salatrim GRAS and begun marketing salatrim and using it in its own products.  In light of the potential for ill effects to consumers from salatrim, FDA should act quickly to deny Nabisco’s petition.

    We have reviewed the data that Nabisco submitted to the FDA concerning gastrointestinal effects of salatrim.2  We are shocked by the skimpiness of the research, by the degree to which salatrim causes gastrointestinal symptoms at levels that may be consumed by many Americans, by Nabisco’s pretending that those symptoms do not occur, and by the FDA’s failure to protect the public’s health.

Nabisco’s data shows gastrointestinal symptoms occur

    Nabisco conducted several preliminary clinical studies and then conducted its largest (but still very small) study, Study V.  Study V involved testing 30g/day, 45g/day or 60g/day of salatrim on groups of 24 free-living subjects (12 females, 12 males) between the ages of 19 and 63 years. Subjects were fed salatrim for four weeks.  The 30g/day level represents the projected 90th percentile of consumption "if salatrim were included at 100% substitution in dairy products, cookies, crackers, chocolates, and confections, snacks, margarine, and spread" (p. 000283).3  It is worth noting that the 90th percentile of potential consumption over three days in 12- to 19-year-old males is 52.57 g/day (and mean of daily users in that group is 24.64 g/day) (p. 000116).  We will focus on the 30g/day group in this comment (gastrointestinal symptoms were much more common and severe in the two higher-dosage groups).

    In Study V, four of the 24 subjects dropped out because of adverse effects or other reasons, leaving only 20 subjects.  Study V found statistically significant increases of % Subject Days in nausea, stomach cramps, and reduced quality of life (pp. 000284-7).  The intensity of stomach cramps was greater than that of controls, and the cramps sometimes were "uncomfortable or an embarrassment, impairs normal functioning" (p. 000288).  Medical assistance was not required for any subjects (but was for people consuming 45g/day or 60 g/day).  Likewise, the intensity of nausea was greater in subjects consuming 30g/day salatrim than in controls (and medical assistance was not required until subjects consumed 45g/day) (p. 000289).  Amazingly, despite those findings, Nabisco stated: "The incidences of gastrointestinal discomfort are essentially equivalent to control fats when SALATRIM is consumed at 30g/day.  We therefore conclude that when consumed at the anticipated use levels in foods, SALATRIM is unlikely to cause significant levels of gastrointestinal upset. (p. 000291)."

    Regardless of the symptoms that salatrim was found to cause, the 20 subjects completing the 30g/day arm of the study is hardly a sufficient number on which to base the safety of a substance that millions of people will likely consume for extended periods of time.  Only extremely common effects would be detectable above the background rates of problems seen in controls.  Furthermore, the general population includes many consumers older and younger than those tested and includes millions of people with gastrointestinal disorders, food sensitivities, and other characteristics that might put them at greater risk of gastrointestinal and other symptoms.  It is incredible that Nabisco did not test a single person under the age of 19, despite the fact that many children will consume chocolate chips, cookies, and other foods containing salatrim.

    By way of analogy, it is worth noting that another artificial fat substitute, olestra, was approved on the assumption that consumers would not experience severe symptoms (though a few severe symptoms were seen in small clinical trials).   Adverse-reaction reports, however, indicate that individual sensitivities vary widely and that many people are suffering symptoms far more severe, and in more inopportune situations, than the FDA or Procter & Gamble envisioned (or admitted publicly).  It is naive to assume that symptoms more severe than those seen in the perfunctory test on salatrim would not occur in the general population.

    In addition, Nabisco’s experimental conditions involved exposing subjects to only one ingredient that causes gastrointestinal symptoms.  In the real world, many consumers would also be consuming other food ingredients -- olestra, sorbitol, and mannitol -- that can cause such problems.  The FDA should require Nabisco to test combinations of such ingredients to ascertain the extent to which their effects are additive or synergistic.

    Gastrointestinal symptoms were seen at the 30g/day level, and Nabisco did not identify a highest no-observed-effect level (HNOEL) in Study V or other studies.  Nevertheless, Nabisco concluded, "The results, therefore, suggest that at normal levels of consumption (i.e. <30g/day) SALATRIM would not be expected to cause any adverse gastrointestinal effect."

GRAS review found dose-dependent gastrointestinal symptoms

    Nabisco hired the Life Sciences Research Office of the Federation of American Societies for Experimental Biology to convene an expert panel to review the use and safety of salatrim.  As Nabisco notes in its GRAS petition, the panel did support a GRAS designation.  Nabisco does not note, though, that the committee stated: "In Clinical Study V, the frequency of clinical symptoms appeared to be dose-dependent; that is, consumption of 60g/day SALATRIM was associated with complaints of nausea in all subjects .... At 30 and 45g/day the frequency of gastrointestinal symptoms was much lower and in most cases, absent."4  The committee also stated:

In addition, the matter of gastrointestinal symptoms noted after ingesting products containing SALATRIM deserves further attention.  While not a major concern in regard to safety, the symptoms may affect or limit acceptability of products containing SALATRIM.   Therefore, in due course, it would be prudent to investigate the gastrointestinal effects of SALATRIM, particularly motility, during relatively long-term ingestion at dosages representative of the levels of the intended uses.5

We are not aware that those additional studies have ever been conducted.

The GRAS review panel also stated:

While there is no evidence in the available scientific information to suggest adverse health effects, the Expert Panel recommends that, if SALATRIM is marketed for its intended uses, a formal mechanism be provided for post-marketing surveillance of consumer comments and possible complaints.  Consumer comments to the manufacturer, which should also be made available to the Food and Drug Administration, would provide an additional follow-up measure to the present assurance of safety based upon the current information and data evaluated by the Expert Panel.6

Packages of Nabisco products containing salatrim do not include any notice advising consumers of the possibility that salatrim can cause gastrointestinal symptoms, nor are consumers invited to call a toll-free number if they experience symptoms.  The only possibly relevant thing on the package is a toll-free number printed together with Nabisco’s address.  Obviously, few consumers are likely to associate salatrim with gastrointestinal symptoms or contact the company.

Gastrointestinal symptoms constitute harm and make salatrim not GRAS

    The fact that in a small study salatrim caused gastrointestinal symptoms, which were sometimes severe, demonstrates that the additive poses a health and safety threat to consumers.  Any additive that "impairs normal functioning" cannot be considered "generally recognized as safe."  Any additive that causes potentially severe cramps poses a safety risk, because those cramps could occur while someone was engaging in an activity requiring constant attention to avoid danger (driving a car, flying an airplane, overseeing children, etc.).  We believe that the average consumer would consider the gastrointestinal symptoms prima facie evidence that the substance is not safe, even though the LSRO panel that Nabisco hired was aware of the gastrointestinal symptoms yet still declared salatrim GRAS.7

    The symptoms that salatrim causes should also render it unapprovable as a food additive, although we recognize that the FDA’s new low standards for approving additives, reflected in the approval of another chemical (olestra) that causes severe cramps, diarrhea, and other gastrointestinal symptoms, might render salatrim "safe enough" for American stomachs.  During an advisory committee meeting in November, 1995, on olestra, then-Commissioner David Kessler said that, "If someone is going to the bathroom all day, and there is really an effect on someone’s life, that certainly can be -- I think one could argue that is harm."  We think his standard makes eminent sense for the regulation of additives that would be added to a multitude of common foods that would be consumed by millions of Americans.  To demand proof that salatrim, olestra, or any other additive causes permanent medical harm, as some have suggested, before considering it unacceptable as a food additive sets a standard that throws caution to the wind and results in large numbers of consumers experiencing harm.

    In light of the extraordinarily limited clinical testing, and because the one small 28-day study found increased rates of significant gastrointestinal symptoms in subjects consuming salatrim, we urge the FDA to deny Nabisco’s GRAS petition and halt the use of salatrim.


Michael F. Jacobson, Ph.D.
Executive Director

1.    59 Fed.Reg. 33774, June 30, 1994.

2.    We may comment on additional matters in the future.

3.    Table C.3.5 states that 90th percentile potential consumption for the U.S. population would be 37.43g/day (p. 000116).

4.     LSRO/FASEB. "The evaluation of the health aspects of using certain triacylglycerols as food ingredients." Prepared for Nabisco Foods Group. Bethesda, Md., August, 1993. At 38.

5.     LSRO/FASEB. "The evaluation of the health aspects of using certain triacylglycerols as food ingredients." Prepared for Nabisco Foods Group. Bethesda, Md., August, 1993. At 39.

6.     LSRO/FASEB. "The evaluation of the health aspects of using certain triacylglycerols as food ingredients." Prepared for Nabisco Foods Group. Bethesda, Md., August, 1993. At 39.

7.    That the LSRO panel believed olestra to be GRAS throws into question the reliability and objectivity of that organization for reviewing proposed GRAS substances.