Health: Born Too Soon

(2 of 3)

Assisted reproduction isn't the only problem. Doctors have long known that smoking, uterine infection, high blood pressure and a prior history of preterm delivery also place an expectant mother at greater risk of delivering early. They're looking into the possibility that other factors, such as stress, diet (both before and after conception) and inflammation may also play a role. But they have something of a medical mystery on their hands. "Nearly half of preterm births are from unclear causes," says Dr. Nancy Green, medical director of the March of Dimes, which is in the early stages of a five-year, $75 million campaign to address the issue. You can do everything right and still give birth to a premature baby.

Doctors admit that some of their best ideas for preventing early delivery haven't worked very well. A drug called Ritodrine, which was approved by the Food and Drug Administration in 1980, successfully stops preterm labor in many women, but subsequent studies have shown that it has no overall effect on a baby's health or survival. Treating all uterine infections, no matter how mild, also appears to make no difference on the timing of delivery--suggesting that infection is only one stage in a larger, much more complex process. "We've been taking the one-cause-at-a-time approach for 20 years," says Dr. Jay Iams at Ohio State University in Columbus. "But it doesn't work that way." Indeed, many researchers believe they won't really have a good grasp of how to prevent prematurity until they answer an even more fundamental question: How does a woman's body decide it's time to give birth in the first place?

One intervention has shown some promise in early trials of women with a history of premature delivery and no other known risk factors. Injecting these women once a week with the hormone progesterone reduces their chance of delivering early by a third. Researchers still don't quite understand why it works. The placenta produces a lot of the hormone, so it wouldn't seem that adding a little more would make a big difference. The treatment has not been studied in women who are carrying more than one child.

Caution may be in order. No one wants a repeat of the problems that happened with diethylstilbestrol (DES), a synthetic estrogen-like drug that was used in the 1950s and '60s to prevent premature delivery and turned out to cause, among other things, reproductive-tract abnormalities and a rare cancer. Unlike DES, however, progesterone has a long safety record. And it is not being used in the earliest days of pregnancy, when birth defects are more likely to occur. What progesterone doesn't have is a major manufacturer, because the drug is not protected by a patent. Instead it is usually produced in small batches at compounding pharmacies.