The Abortion Cocktail


Bernard N. Nathanson


Copyright (c) 1996 First Things 59 (January 1996): 23-26.

In April 1970, in the pages of the Obstetrical and Gynecological Review, I reviewed the history of attempts to control population growth by means of medical-i.e., nonsurgical-abortion. Starting in antiquity, I found that the ancient Greeks did not generally approve of abortion, while their Roman counterparts favored the use of thaumaturgy and incantations. Cultures in Africa and the Middle East tried concoctions of camel or goat dung. In the medical lore of Europe during the Middle Ages, pastes, emetics, purgatives, emmenagogues, sternutators, convulsants, clysters, physical maneuvers, and pessaries are mentioned. The Enlightenment had a vast pharmacopeia with dazzling names and no abortifacient effect whatever: aloes, colocynth, apioline, pennyroyal, tansy, pulegium oil, caulophyllia, cardomon, and cimicifuge.

Toward the end of the paper I discussed a newly emerging class of drugs known as folic acid antagonists and antimetabolites. These drugs were used primarily to treat cancer but they were also reasonably effective in producing medical abortion, at least up to eight weeks gestation. The enormous toxicity of these drugs to the pregnant woman, however, and the fact that they produced massive congenital deformities in the fetuses who somehow survived the pharmacological assault, virtually precluded their use as abortifacients.

In 1995, twenty-five years after my review, the New England Journal of Medicine carried a report that the combination of a powerful anticancer agent (methotrexate) and a relatively innocuous drug used primarily to treat gastric ulcer (misoprostol) will effect medical abortion in 96 percent of women pregnant less than nine completed weeks, with no currently identifiable damage to the pregnant woman.

Some thoughtful analysts, including some pro-life advocates, have suggested that the legal battle to end abortion has at last ended in defeat with this report of an apparently effective and safe medical form of abortion. An "abortion cocktail" that women take in the privacy of their homes would eliminate the clearly specifiable abortionists against whom the pro-life movement has been able to focus its efforts. In the event of a ban on legal abortion, it would force prosecutors into the difficult position of seeking convictions against individual women rather than against the abortionists. Any attempt to pursue the doctors prescribing the "cocktail" would likely result merely in the emergence of a black market or of a charade in which women seeking abortions would report the symptoms of the diseases for which the ingredients are currently and legally prescribed.

The legal battle to ban abortion, however, is not by any means defeated with this report of a medical abortifacient, and it may in fact be helped. The unprofessional and unethical means by which the procedure was tested, the unknown long-term medical effects on women of the drugs involved, and the extreme dangers to the human gene pool posed by partially completed medical abortions all suggest a strong societal interest in banning the "abortion cocktail." And the falsity of claims that medical abortions will "demedicalize" and "empower" women by granting "at-home" abortions will soon become apparent-allowing the pro- life movement to continue to focus its efforts on the offices of abortionists.

Methotrexate, the first drug used in the "abortion cocktail," is an unusually potent and dangerous drug. The Physician's Desk Reference (PDR), an encyclopedic compendium of prescription drugs marketed in the U.S., devotes six pages to a description of the powers and hazards of this drug, its uses and abuses. (The average entry for drugs runs about one-half page.) The entry opens with a black-bordered box reading, in part: "Methotrexate should be used only by physicians whose knowledge and experience includes the use of antimetabolite therapy. . . . Because of the possibility of serious toxic reactions the patient should be informed by the physician of the risks involved, and should be under a physician's constant supervision."

The PDR goes on to specify in excruciating detail the toxic potential of this drug: liver damage, kidney destruction, heart muscle compromise, pulmonary failure, gastrointestinal pathology, and bone marrow suppression. It has also been reported to cause loss of speech function, strokes, and convulsions. It is, in other words, one of the most toxic, dangerous, and potentially fatal drugs in the entire PDR, and it has been approved for use by the FDA only for the treatment of cancer-which is to say, the dangers of this drug are so massive that they are outweighed only by the benefit of abating or curing cancer.

There is a considerable literature concerning the use of methotrexate in ectopic, or tubal, pregnancies. In the past, ectopic pregnancy had been managed by surgical means, i.e., by removal of the tube containing the pregnancy. Moral theologians generally sanctioned this approach since, although it did effect abortion, the primary intention of the surgery was to preserve the life of the mother and the abortion was a regrettable side effect.

However, beginning in 1987, increasing numbers of gynecologists were reporting the successful management of ectopic pregnancy by the use of laparoscopy (the introduction of an optical system into the abdomen through a small incision in the umbilicus) and the delivery of a dose of methotrexate through the scope into the amniotic sac of the fetus developing in the tube. No longer was there a "double-effect" rationale to exonerate the surgeon; we now had a direct lethal assault on the developing child (albeit developing in the wrong place). The moral consequences of this "momentous advance" in gynecologic technology provided a bridge to the use of methotrexate to destroy the child developing in the correct place, the womb.

In normative bioethics the use of a drug such as methotrexate would require a sufficient benefit to outweigh the risks: the disease being treated must be of such gravity as to justify the use of the drug. Pregnancy, however, falls comfortably into none of the usual definitions of disease: it is not a state inimical to the way one is supposed to feel; it is not a condition in which body components and systems are acting inharmoniously (to the contrary, infertility would more likely fall within that definition); it is not a state of abnormality-for the pregnant woman functions satisfactorily within society, and the social and physical environment tolerates the pregnant woman perfectly. Pregnancy, then, is not a disease by any normative definition, and the administration of a drug as potent and as toxic as methotrexate cannot be justified within the traditional framework of medical risk/benefit analysis.

Richard Hausknecht, the author of the article in the New England Journal of Medicine, acknowledged that the "abortion cocktail" was indeed an experiment when he stated in his report that "the protocol [of the experiment] was approved by the investigative review board of the Mount Sinai Medical Center." The Code of Federal Regulations (49 CFR Part II) sets forth federal policy for the protection of human subjects in medical experimentation, mandating the establishment of an institutional review board (IRB) in each hospital in which human subjects are used for experimentation. The Code further stipulates how many members each IRB should have, from what disciplines the membership should be drawn, and precisely how each IRB should function. The language of the Code is explicit: each IRB is charged with the responsibility to assure that "risks to subjects are minimized by using procedures which are consistent with sound research design and which do not unnecessarily expose subjects to risk." The Code requires periodic reports from the experimenters and specifies the language and nature of the informed consent presented to each subject. (The explanation of the risks is so detailed that these consents often run to ten or fifteen pages.)

In the Hausknecht report there is no indication of the nature and content of the informed consent. There is no mention of periodic reports to the Mount Sinai Hospital's IRB. There were no precautions or measures designed to minimize the risk accompanying this experimentation. Where was the constant medical supervision the PDR recommends when using the drug methotrexate? Hausknecht, in the protocol section of the report, speaks of each subject receiving an intramuscular injection of methotrexate, then going home, to return five to seven days later for the administration of the second drug misoprostol; who was supervising the patient in those five to seven days? Where were the blood tests to assure that the subject was not reacting adversely to the methotrexate? Where were the liver and kidney function tests, the tests for pulmonary function, the tests to assure that the bone marrow was functioning normally, the neurological assessments?

Another section of the Code of Federal Regulations (Title 45, Part 46) directs that the Department of Health and Human Services through the National Institutes of Health establish and maintain an Office for Protection from Research Risks. Even if an experiment is not federally funded it falls within the purview of this Office in that human subjects are involved in the experimentation. This Office is responsible for overseeing the functioning of IRBs. Subpart B of this section of the Code deals with "research, development, and related activities involving fetuses, pregnant women, and in vitro fertilization" and requires that each institution conducting research on such human subjects establish, in addition to the IRB, an Ethical Advisory Board consisting of research scientists, physicians, psychologists, educators, lawyers, and ethicists-as well as representatives of the general public. Nowhere in the Hausknecht report is an ethical advisory board even mentioned.

It is customary and even ethically mandatory that experimentation that carries with it a substantial degree of risk to human subjects first be tested in an animal laboratory or, at the very least, on a computer model. There is no mention of any such preliminary testing in the Hausknecht report, probably because no such testing was done. Women were once again used as laboratory rats, much as they had been used in the early days of oral contraceptives, IUDs, silicone breast implants, and the Norplant contraceptive.

Two doctors, Creinin and Vitinghoff, carried out a similar and equally odious experiment on pregnant women using methotrexate and misoprostol at the San Francisco General Hospital and reported their results in the Journal of the American Medical Association in October 1994. They at least had the minimal decency to provide a group of controls (women to whom only misoprostol was given) but their experiment was otherwise every bit as ethically shabby as the Hausknecht fiasco.

Many troubling questions arise from these two experiments. How could the IRBs of two prestigious university teaching centers have sanctioned such an experiment, despite the multiple strictures, restraints, and codes regarding experimentation on pregnant women and human fetuses? Was there no concern for those women who, having been given the methotrexate, might wish to back out of the experiment with the fetus still alive but lethally afflicted with the well-recognized teratogenic effects of the drug? Where were the preliminary animal testing and computer models establishing some parameters of safety and efficacy for these experiments? Where were the ethical advisory boards mandated by the Office for Protection from Research Risks? How many equally invidious experiments of this sort are now ongoing-and how many are federally funded? Where was the elaborate bureaucratic multitiered machinery designed to scrutinize such experiments? Where were the usual whistle- blowers while these squalid experiments were going on? For that matter, where was the pro-life movement, and why has the public response been so muted?

Hausknecht and company make much of the increased privacy flowing from the "abortion cocktail." This is, to be charitable, a delusion fostered largely by dishonest reporting and slovenly procedure. In their book RU-486: Misconceptions, Myths, and Morals, the feminists Janice G. Raymond, Renate Klein, and Lynette J. Dumble point out that all medical abortions, far from being purely private affairs between women and their physicians, are even more public than the conventional surgical (suction) abortions. A woman seeking a medical abortion of any sort must submit to a general physical examination (to assure she is an apt candidate for the drugs in question) as well as to a pelvic examination. She must have a blood test to confirm the pregnancy and an ultrasound examination to assure that the pregnancy is within the established limits of gestation for such an abortion. A legally mandated waiting period may follow, after which the woman must return to be given an injection by a physician or nurse. She must return a third time for the ingestion of the second drug (misoprostol, used both in RU-486 and the new "abortion cocktail"), whereupon she is subjected to a second pelvic examination as well as to an elaborate mix of blood tests. Only then does the woman go home. "To call this an at-home abortion is deceptive," Raymond, Klein, and Dumble noted about medical abortions in general, "since most treatment transpires in the clinic, hospital, or office, and is extremely medicalized."

Moreover, with either RU-486 or with the "cocktail," the ordeal is still not complete. "What actually happens at home can be an excruciatingly long wait for the embryo to be expelled from the uterus, accompanied by pain, bleeding, vomiting, and nausea and other complications that are drawn out over a substantially lengthy period of time." Following the miscarriage the woman must return a fourth time for an examination to ascertain whether the abortion is complete; this usually entails another ultrasound examination as well as a quantitative blood pregnancy test and another pelvic examination. Despite the rhetoric of "privacy," "women in control," and "demedicalization," there have been four office/clinic visits (in the ideal uncomplicated case), two ultrasound examinations, multiple blood tests, and strict medical supervision.

A group of investigators at the University of Edinburgh questioned women who had undergone medical abortion. In their report What Do Women Want During Medical Abortion, they found that only 24 percent of the women interviewed preferred to have the actual miscarriage at home; fully three-quarters of the women desired the hospital ambience with physicians, nurses, other women undergoing the identical experience, and support partners present. The miscarriage of a fetus at home is a lonely, physically trying, frightening, and demoralizing experience-one that culminates in the woman viewing her own dead child in a terrifying mass of blood, clot, and tissue between her legs.

The public must be made aware of the deceptions, the ethically deeply flawed experiments, the conscienceless exploitation of women, and the potentially dire consequences to both mother and (surviving) fetus. The long-term transgenerational consequences to humanity itself must be examined: methotrexate is a drug known to be capable of damage to the human DNA, and as such may induce mutations so grotesque as to be virtually unimaginable.

If there is one ray of light in all this for pro-life forces, it is the prospect of the surgical abortionists crossing rhetorical swords with the medical abortionists-competing with each other for the obscene commercial traffic and trashing each other's methods with abandon. Regardless, by no means does this intolerably unethical and immoral technology spell the end of the legal battle to ban abortion. That battle has only begun.


Bernard N. Nathanson, M.D., the author of Aborting America, is a bioethicist at the Center for Research and Clinical Ethics at Vanderbilt University.