In the February 1997 issue of Scientific American there appears a debate on the use of animals in medical research. We argued the case that "Animal Research is Vital to Medicine" [1] and were required to produce references and photocopies from primary sources to justify our assertions. The case against, "Animal Research is Wasteful and Misleading," was put forward by Neal Barnard and Stephen Kaufman [2], representing the Physicians Committee for Responsible Medicine and the Medical Research Modernization Committee, well-known animal-rights organizations. In our view, Barnard and Kaufman based their arguments upon misrepresentations of scientific fact that should have no place in the pages of an authoritative scientific journal, as we pointed out to the editors prior to publication.

Our essay is based on the actual email letters we sent to Scientific American's editors after reviewing our opponents' draft. (Barnard and Kaufman reviewed ours as well.) Our hope was that the editors would come to their senses and realize that they would harm science by lending the magazine's prestigious name to an article that grossly distorted medical history. Our prediction has unfortunately come true, for reference to Barnard and Kaufman's article as a legitimate scientific source has appeared in the fundraising material of an animal-rights organization and in debate in the media. Because the organization of the "debate" did not permit rebuttal in Scientific American, we offer it here.

Barnard and Kaufman claimed that during the 1920s and 1930s, experiments on monkeys led to "gross misconceptions that delayed the fight against poliomyelitis." They stated that the erroneous conclusion that the polio virus affects only the brain contradicted "previous human studies demonstrating that the gastrointestinal system was the primary route of infection." In fact, the only way to study the then totally invisible virus was to show its presence by the paralysis it produced on administration to the spinal cord of monkeys. Indeed, the "previous human studies" mentioned showed that washings of the gut of patients ill with polio, or who had died of the disease, could produce paralysis in monkeys [3].

The subsequent claim that animal studies delayed the development of a vaccine, an advance that became possible only after Enders et al. in 1949 [4] managed to cultivate the virus on human tissues, is equally spurious. The primary inoculum used by Enders was obtained from mouse brain, and its identity was verified by the "character of the disease it produced in white mice following intracerebral injection." Enders did indeed prove that the virus was replicating in culture because, after many subcultures (and therefore dilution of the original inoculum), the culture fluid still "on inoculation into mice and monkeys, produced typical paralysis." The fact that the sources actually cited by Barnard and Kaufman refer to animal-based experiments as contributing to the polio workers' discoveries is sufficient condemnation of their contentions.

In further arguing that medical advances have been delayed by misleading results from experiments with animals, Barnard and Kaufman refer to an opinion paper by Wiebers et al. [5]. This secondary source gave its own interpretation of a study reported by Pulsinelli and Buchan [6]. Actually, Pulsinelli and Buchan point out that the events following brain ischemia shown by a variety of methods in several animal species accurately predicted those occurring in people. They then ask why various compounds did not prove effective in treatment. They answer by noting that their review of a variety of studies indicated that insufficient numbers of animals had been used before clinical trial or there had been weak experimental design and technique. In other words, the blame should be put on the research designs, not the animals. Although Wiebers et al. were critical of animal-based research on stroke for the reason that, according to them, acutely induced conditions in animals can not adequately replicate chronic human disease, they concluded their paper by saying that animals "will contribute substantially to research in stroke for the foreseeable future. . . ." They also acknowledge, "There is also no current alternative to animal studies of safety and effect in the screening of agents that may benefit patients with stroke." They then ask for the development of better research techniques to use directly in people. No one could quarrel with this, however far in the future it might be.

A further misrepresentation concerns the hackneyed accusation that medicines declared safe after animal tests have been subsequently shown to be toxic after widespread use with patients. Four examples, including the antiviral drug fialuridine, were given. This is a completely spurious argument against animal experiments. Before a drug is marketed - and after the standard animal safety studies (involving a maximum of 1,400 individual animals) - the potential medicine is subjected to the standard three-phase clinical investigation in normal volunteers and patients, usually involving up to 3500 persons. Obviously, side effects with an incidence of less than 1 in 2,000-3,000 are unlikely to be detected in either the animal or the more extensive human studies. It is fatuous to suggest that reliance on animal work is responsible for post-marketing toxicity.

The use of fialuridine as an example is particularly erroneous. This drug had shown a remarkable effectiveness for the treatment of hepatitis B infection in two short trials. During the planning of a more extensive study a patient in the second trial died of liver failure four months after its conclusion. In all, five patients died showing an unusual type of long-delayed toxicity possibly due to damage to mitochondrial DNA [7]. A review of the drug trial by the American Institute of Medicine concluded that animal experiments should be carried out to establish the nature of this type of long delayed toxicity [8]. Presumably experimental studies designed to detect such delayed toxicity will become part of the routine preclinical examination of a potential medicine, just as teratological studies were introduced after the thalidomide tragedy.

Barnard and Kaufman claim that "most animal species used in the laboratory do not develop the kind of limb defects seen in humans after thalidomide exposure; only rabbits and some monkeys do." This ignores the many other congenital defects induced by thalidomide: lesions of the eyes, ears, heart, kidneys, and digestive tract, for example [9]. Also, they do not mention that congenital deformity of some form can be produced by thalidomide in rats, mice and hamsters [10-12] as well as the rabbit and three species of monkey they do mention. One should also note that only a strong stand by an official in the Food and Drug Administration prevented marketing of the drug in the United States. She noted that thalidomide had been inadequately tested in Germany, thereby averting even more harm in the United States [9]. The dismissal of the case of thalidomide as an argument against animal experimentation is surely rendered complete by posing the question: would thalidomide pass the teratogenic tests subsequently implemented as a result of the tragedy, if it were produced as a new chemical entity today? The answer is an emphatic no!

To argue against the important role animals played in the discovery and development of insulin as Barnard and Kaufman did is silly as well as irresponsible. Even though it was known that pancreatic malfunction led to diabetes and death, physicians had no way of knowing without animal research what was essential in the pancreas. Furthermore, the isolation, purification, and manufacture of insulin required the use of animals. In the early 1920s, emaciated, comatose, diabetic children at the brink of death experienced "the closest approach to the resurrection of the body that our secular society can achieve" [13] thanks to the discovery and commercial availability of insulin.

We could continue rebutting practically line by line, but let us conclude by quoting from an editorial we wrote to alert biology teachers to Scientific American's unfortunate gaffe. "Our view was that the exercise would harm science because an animal-rightist article would finally have a respectable reference to cite for all time. Furthermore, although we were certain that many readers would be sophisticated enough to identify faulty logic and sweeping generalizations, we knew they would not know enough medical history (let alone the past history of the writers) to know what was false, or at best, trickily presented.

"In conclusion, the ethics of using animals in various ways can and should be debated. Furthermore, we respect and support the wish of most to use animals thoughtfully. But the history of medicine is so clear on their contributions to our understanding of the causes, preventions, and cures of diseases, as well as the development of a host of other treatments and surgical procedures, that only the untutored or those with a hidden agenda could argue that animal research has not been fundamental to medical progress and human understanding. We hasten to add, though, that scientists conduct such research in conjunction with or cognizant of a variety of other approaches. Nature's complexity demands this of us. Unfortunately, Scientific American fell prey to a political agenda and chose to discuss the merits of one methodology in unnatural isolation [14]."

The original debate articles, "Animal Research Is Vital to Medicine" and "Animal Research Is Wasteful and Misleading," are available online from Scientific American, along with an extensive editorial overview and reader feedback on the issue.

"Animal Research Articles Draw Fire" - this article from the March 31, 1997 issue of The Scientist discusses the reaction to the Scientific American articles.

Policies on the Use of Animals and Humans in Neuroscience Research - the Society for Neuroscience's guidelines for responsible and humane research on animals, plus suggested further reading.

Animal Health, Well-Being, and Rights - list of links including animal research-related sites (most under "miscellaneous"). From the WWW Virtual Library.

Alternatives to Animal Testing Bibliography - a list of references with abstracts that introduce alternatives to animal research. From the National Library of Medicine at the NIH.

Altweb - provides information on K-12 education about animal testing and news about regulations and alternatives. The site is still under construction; most information consists of links to other sites. From the Center for Alternatives to Animal Testing at Johns Hopkins School of Public Health.

American Association for Laboratory Animal Science - includes membership information, resources for conducting responsible animal research, and an extensive list of related sites.