Holly disputes her midwife's story, asserting that the midwife recommended induction against Holly's better judgment. "My body was made to have babies," she told me. With five vaginal births to her credit, Holly had confidence in her ability to labor.

Whatever the truth, both parties agree that over the next several hours the nurse-midwife gave Holly three 25-microgram doses of Cytotec. (Because of legal considerations, both parties requested anonymity.) What Holly didn't know and the midwife never told her was that it was an unapproved drug with potentially disastrous side effects.

One hour after the third dose, labor began, with contractions every two and a half minutes. According to Holly's 19-year-old daughter, Ann, who was present throughout labor, Holly handled herself very well.

Thirty hours later, her cervix not yet fully open, Holly stood up and walked around. Then her bag of water broke. A little later she heard a popping sound from her body. The midwife monitoring her labor noticed that the baby's heart rate had dived from a normal 130-140 beats to a frightening 40 beats per minute.

She exhorted Holly to push, and within five minutes, Holly's 8-pound, 13-ounce daughter was born, followed by a huge gush of blood. The baby was blue and didn't breathe on her own, so the resuscitation team intubated her.

Holly, meanwhile, continued to bleed. Frightened, she told her midwife that something was wrong. The midwife assured her that her blood loss was not enough to warrant a doctor's presence. Later, realizing that Holly was bleeding excessively, the midwife removed several huge clots from her vagina, gave her medication to stop the bleeding and left her in the care of nurses.

Ann and Darryl, Holly's husband, were far from reassured. By this time, Holly lay unconscious, white as a ghost. They helplessly watched her struggle for breath. Darryl begged the nurses to get a doctor and the midwife directed a nurse to call for a doctor on the intercom. The physician who entered the birth room two minutes later was shocked at Holly's condition. "This lady is dying," he shouted. "I'm taking her to the O.R.!"

Holly's heart stopped twice during the surgery. At one point, the doctor told Darryl that he did not expect her to survive. Her uterus had ruptured from the top down through the cervix. This kind of wound is characteristic of Cytotec-related ruptures, according to obstetricians I've since spoken to. (One doctor described them to me as "totally exploding.") Surgeons removed Holly's uterus along with one of her ovaries and a fallopian tube. Thirty-seven units of blood, plasma and platelets were required to replace the blood lost during her ordeal. Gone forever was her chance to have another baby.

Was Holly's labor a nightmare fated to happen, with or without intervention? Or did Cytotec cause her uterine rupture, thereby threatening her and her daughter's life? As with so many forms of obstetric intervention, even hindsight isn't 20/20. Every birth is unique, and the influences on labor are far more numerous than most studies can account for. And even with large, long-term, controlled studies, it is sometimes complicated to ferret out the facts about the efficacy or safety of a given medical procedure.

Cytotec, however, doesn't have the benefit of such scientific debate, because it is still essentially an experimental birth drug that is being tested ad hoc by trial and error. But most patients are never informed of this fact.

As a midwife of 30 years and one of the founders of the natural childbirth movement, I have overseen more than 2,000 births at my birthing center in Summertown, Tenn. Over the years I've listened to innumerable anecdotes about the dangers of medical intervention. But the stories I was hearing about Cytotec I found especially unsettling.

Sign up to receive free e-mail updates from Salon -- now in 17 different varieties!