by Michael McGuffin
Editor's note: A previous version of this article was published in September 2004 in the AHPA Report,1 the monthly newsletter of the American Herbal Products Association, the herb industry trade association. Because we considered the subject so significant, we invited the author to submit the article for publication in HerbalGram. This version contains new information and additional references and has undergone peer review. In essence, the author is commenting on the practice of the Food and Drug Administration to release adverse event reports (AERs) to the media without qualifying each report and/or clarifying the overall significance of the data being released. The author is not attempting to provide a thorough review or evaluation of the AERs on bitter orange dietary supplements. The opinions expressed herein are those of the author and do not necessarily reflect the opinions or policy of HerbalGram or the American Botanical Council. HerbalGram provided a copy of this editorial to officials at the FDA for comment in this issue, but at press time FDA had not yet responded.
Many marketers of dietary supplements containing bitter orange (Citrus x aurantium L., Rutaceae) were surprised when an April 11, 2004 article in the New York Times2 stated that, according to an unidentified Food and Drug Administration (FDA, or the Agency) spokeswoman, there had been 85 adverse reactions—and 7 deaths—associated with this increasingly popular ingredient. As is often the case with such articles, other media picked up on the story, and it appeared in other papers around the country, including the Seattle Times, the York Sunday News (PA), the Kansas City Star, the Arizona Republic, the Akron Beacon Journal (OH), San Jose Mercury News (citing the New York Daily News), and others.
None of the primary suppliers of bitter orange extracts, as well as various other herb experts in the industry and academia, had heard of any such cases, and it was difficult to believe that reports of such a serious nature could have passed unnoticed.
This author, as president of AHPA, contacted the Agency and engaged in frustrating conversations in which it became clear that FDA personnel knew that the Times report was not accurate and that these officials had absolutely no intention of issuing a retraction or correction, or at the least, a clarification. It was later learned that FDA officials knew that many of these reported adverse reactions were associated with dietary supplement products that also contained the controversial and now banned herb ephedra (Ephedra sinica Stapf., Ephedraceae), which contains ephedrine alkaloids. The combination products containing both bitter orange, ephedra, and other ingredients had been sold, and adverse event reports related to them had been filed prior to the ephedra ban on April 12, 2004.3 Consequently, AHPA's legal counsel, Anthony L. Young, filed a request with the FDA under the Freedom of Information Act (FOIA) for copies of any adverse event reports (AERs) regarding dietary supplements containing bitter orange and/or its extracts.4
In early June, AHPA received from FDA two CDs containing files for 147 AERs to fulfill Young's request. As is typical of FDA's AER files, these reports varied greatly in content and completeness of documentation of the details of the cases. Some of them provided thorough descriptions of the subject patient, extensive medical records, and a copy of the label of the product associated with the report. Others consisted of only one- or two-page e-mails from one FDA employee to another and lacked basic information such as the name of the product. A quick review showed that the many of these reports were associated with products that also contained ephedrine alkaloids.
The Number Keeps Going Up
Even as AHPA's staff worked to organize all of these records and wondered how the number had suddenly jumped from the 85 reported in April to the 147 received two months later under the FOIA request, the number of alleged AERs jumped again. An article in the July 5, 2004 edition of Food Chemical News5 reported that Barbara Schneeman, PhD, the new director of the FDA's Office of Nutritional Products, Labeling, and Dietary Supplements (ONPLDS), had then tallied 169 bitter orange AERs.
To her credit, Dr. Schneeman was also quoted as saying, "One has to look at the quality of the data that's in those reports," and that just because FDA receives a report "doesn't automatically mean that the product itself has caused the problem."5 But when the Los Angeles Times ran a story on July 19,6 it cited a "spokeswoman for [FDA]"as its source for reporting that 169 "reactions" had now been "linked" to bitter orange. In preparing this article, AHPA was informed by the LA Times writer Shari Roan that this number had been provided by FDA's press office in response to her request that specifically asked that only "ephedra-free" products be counted. AHPA's concerns about FDA's failure to reveal that these cases were based on products that also contained ephedrine alkaloids escalated to concerns that FDA was simply misstating the facts, whether intentionally or through negligence.
AHPA's Analysis, Part 1: How Many Bitter Orange AERs Are There?
When AHPA completed its analysis of the 147 AERs received in June, some truly remarkable facts and figures were found (see Table 1).
In summary, AHPA was provided with 110 unique AERs and with information about 17 others for which no report numbers were assigned. The total of 127 cases represents 42 more than the amount revealed by the NY Times' anonymous source at FDA in April and 42 fewer than what Dr. Schneeman (FDA's Director of ONPLDS) believed to exist in early July. Any competently rational person who reviews this data must assume that something is seriously flawed at the Agency. Either FDA's management is receiving erroneous information from its AER staff, or the FDA (via the FOIA system) is not actually providing the public—in this case AHPA—with the kind of accurate information that assures transparency of the operation of government agencies.
|Table 1. Quantification of AERs in Freedom of Information Act Response|
|TOTAL quantity of AERs identified by FDA with unique case numbers.||147|
|Duplicate numbers (i.e., more than one case number for a single AER).||1|
|Quantity of case numbers included in FDA's Freedom of Information Act (FOIA) response that were assigned to events identified in tables sent to FDA by a consultant to Cytodyne LLC* related to internal records on 2 of the company's products.||79|
|Apparent duplicates included in the above identified "79 cases."||36|
|Quantity of unique AERs identified by FDA in its FOIA response.||110|
|Events that were included in the above-referenced Cytodyne tables but apparently not counted in the above identified "79 cases."||17|
|Actual quantity of unique AERs identified in FDA's FOIA response.||127|
* A marketer of dietary supplements under the Xenadrine¨ brand.
Calculations are: 147 ¯ 1 ¯ 36 = 110; 110 +17 = 127.
Why Were the Incomplete Cases Included?
It is difficult to understand how FDA decided to consider each of the cases identified by Cytodyne LLC (in the information submitted by Cytodyne's consultant and listed in Table 1) as a unique adverse event. (Cytodyne markets Xenadrine¨ EFX, a dietary supplement containing bitter orange extract.) The Cytodyne tables were provided as a reference to support a statement made by a consultant to the company that one described event (in which a seizure was reported) represented a unique report for the company's product. The information included in the Cytodyne tables consisted only of an internally assigned number, the date of the report, a reported outcome (each stated in not more than three words (e.g., "hair loss" or "loss of consciousness"), and the frequency of the reported outcome. All additional relevant and necessary information was lacking; for example, there was no age or gender identified for the subjects, nor were dosage and duration of use discussed. Also missing was any information on the medical history and the possibility of concomitant use of prescription or nonprescription drugs or other dietary supplements. It is impossible to view these cases as meaningful submissions to a federally managed adverse event reporting system.
AHPA Analysis, Part 2: Where's the Bitter Orange?
Further analysis of the 127 unique cases identified in the data provided to AHPA by FDA was undertaken to answer an important question—How many of these cases were related to products that were labeled or reported to contain bitter orange (or bitter orange extract) but no other source of a stimulant alkaloid? Of particular interest is another question—How many of these AERs were associated with products that were labeled or reported to contain both bitter orange and a source of the now-banned ephedrine alkaloids? (See Table 2.)
|Table 2. Quantification of AERs Uniquely Associated with Bitter Orange|
|ACTUAL quantity of unique AERs provided by FDA.||127|
|Quantity of AERs that did not identify a bitter orange ingredient in the associated product.||82*|
|Quantity of AERs that did identify a bitter orange ingredient but did not include any ephedrine-containing ingredients.||41|
|Quantity of AERs that did identify a bitter orange ingredient, did not include an ephedrine ingredient, but also included a source of caffeine.||40|
|Quantity of AERs that identified only a bitter orange ingredient (i.e., did not include either an ephedrine or a caffeine ingredient).||1|
|* Includes one case for a product with a label that identified Sida cordifolia L., Malvaceae as an ingredient with no mention of ephedrine, but that also stated the product was not for sale in numerous states, consisting of those states where ephedrine was not allowed to be sold (but where sida was presumably legal); one case for the same product for which no label was provided; and a third case for this same product with a label that listed an extract of S. cordifolia as an ingredient and that also identified and quantified ephedrine as a constituent of the product.|
What is most outrageous in Table 2 is that the agency has not in any way differentiated the 82 AERs purportedly associated with bitter orange that are also products with ephedrine-containing ingredients. In fulfilling AHPA's FOIA request, FDA simply lumped these together with other cases. Nor has the Agency separated these AERs in its statements to the media, even when the media has reportedly requested information only for products that did not contain ephedrine. In its recent ban on ephedra, the government branded any amount of ephedrine in a dietary supplement as presenting an unreasonable risk of illness or injury.3 Has not FDA already assigned ephedrine as the "associated" ingredient for each of these 82 cases and utilized that information as part of the basis on which it has banned ephedrine alkaloids from dietary supplements?
The One Case Still Standing
The above calculations leave a single case remaining where FDA received a report, in late 1998, of an adverse event associated with a product that was labeled to contain bitter orange but that did not contain either caffeine or ephedrine. It is, of course, useful to examine this case in detail in light of the pressing question. Could this incident have been a "signal"? Did the symptoms and other factors rise to the level of a "sentinel event"? (The term "sentinel event" was used in the 2003 report by the RAND Corporation on its safety evaluation of ephedra, under contract by the Agency for Healthcare Research Quality.7 This report defined such an event to be one that met certain criteria, including exclusion of alternative explanations for the adverse event.)
In fact, this report recorded an adverse experience in a 74-year-old woman who had been prescribed three separate herbal formulas, in the form of imported small black pills, from a Chinese herbalist. One of the products contained as an ingredient "immature bitter orange." The woman's prior medical history was noted to include osteoarthritis, hypertension, and diabetes. Concomitant prescription drugs were identified as DiaBeta¨ (a blood sugar-lowering drug for diabetics), Norvasc¨ (a blood pressure-lowering drug), Zocor¨ (a cholesterol-lowering drug), and a thyroid drug. And the recorded adverse experience: "Felt very unwell, weak," and later, "lower extremity swelling (2+ edema)."
This review has narrowed the focus of these AERs to this single incident to call attention to concerns about the way that FDA provides safety information about herbs to the media. But those reports that are associated with products that combine bitter orange and a caffeine-source ingredient should not be overlooked, as there are numerous such products available in the market.
What Next? Will FDA or the Media Set the Record Straight?
AHPA has verbally communicated the information contained in this article to FDA by forwarding the previous version of this article published in the AHPA Report1 to them as well as to the writers of the articles cited above in the New York Times and the Los Angeles Times.2,6 AHPA has also requested that FDA amend its flawed communication policy that currently provides inaccurate information to the public on the agency's own AER records for bitter orange, and, presumably, for all dietary supplement products. AHPA believes that FDA should establish a policy wherein it always provides clarifying information whenever it releases reports about AERs it has received. Such information should include, for example, a breakdown of adverse events by dosage; breakdown of adverse events by severity; identification of other contained ingredients (especially if such ingredients might also be considered to be associated with any reported events); information on age and gender; and any other data needed to ensure a clear understanding of any such reports.
It has now been reported in trade journals that, in response to AHPA's identification of errors and misrepresentations in FDA's record of adverse events reportedly associated with bitter orange, FDA's press office will no longer provide quantitative responses to inquiries about adverse events associated with dietary supplements submitted to FDA.8,9 However, FDA has not made any public statement to acknowledge that the next time an enterprising writer calls the Agency, he or she will receive any more accurate information than was provided to the reporters for the New York Times and the LA Times last spring.
Michael McGuffin is president of the American Herbal Products Association, the national trade association and voice of the herbal supplement industry. He is also the senior author/editor of several books that provide self-regulatory guidance for the herb industry in the areas of safety and nomenclature, The American Herbal Product Association's Botanical Safety Handbook (1997) and Herbs of Commerce, 2nd ed. (2000). Founded in 1983, AHPA represents the responsible core of the botanical products industry, and comprises growers, processors, manufacturers, and marketers of herbal products. Web site: www.ahpa.org.
1. McGuffin M. FDA spins numbers on bitter orange AERs: AHPA analysis finds only 1 actual report associated with bitter orange. AHPA Report. September 2004;19(9):2-4.
2. Hurley D. As ephedra ban nears, a race to sell the last supplies. New York Times. April 11, 2004.
3. US Food and Drug Administration. Final Rule Declaring Dietary Supplements Containing Ephedrine Alkaloids Adulterated Because They Present an Unreasonable Risk, 69 Fed. Reg. 6788 (February 11, 2004). Available at: http://www.fda.gov/ohrms/dockets/98fr/1995n-0304-nfr0001.pdf.
4. Young AL. Letter (via facsimile) to Food and Drug Administration, Freedom of Information Staff, April 14, 2004.
5. Wallace P. FDA taking cautious approach to AERs, bitter orange safety. ---Food Chemical News. July 5, 2004;46(21):1,20.
6. Roan S. Ephedra-free weight-loss supplements scrutinized. Los Angeles Times. July 19, 2004:F1.
7. Shekelle P, Morton S, Maglione M, et al. Ephedra and Ephedrine for Weight Loss and Athletic Performance Enhancement: Clinical Efficacy and Side Effects. Evidence Report/Technology Assessment No. 76 (Prepared by Southern California Evidence-based Practice Center, RAND, under Contract No. 290-97-0001, Task Order No. 9). AHRQ Publication No. 03-E022. Rockville, MD: Agency for Healthcare Research and Quality. February 2003. Available at: http://www.fda.gov/OHRMS/DOCKETS/98fr/95n-0304-bkg0003-ref-07-01-index.htm.
8. Wallace P. Herbal group challenges adverse event numbers for bitter orange. Food Chemical News. September 13, 2004;46(31):17-18.
9. CFSAN examines AER reporting following bitter orange miscalculation. The Tan Sheet. September 20, 2004;12(38):11.