Pregnant women who have caffeinated drinks equal to two or more cups of coffee a day have twice the miscarriage risk of those who avoid caffeine altogether, a new study has found.

The Kaiser Permanente research, to be published today in the American Journal of Obstetrics and Gynecology, looked at 1,063 pregnant women in San Francisco over two years ending in October 1998.

The women who consumed 200 milligrams or more of caffeine per day were twice as likely to miscarry as those who consumed no caffeine.

That equals about two 8-ounce cups of coffee, five 12-ounce cans of cola or six cups of tea.

Previous studies have shown a link between caffeine and miscarriage. But critics questioned those findings, arguing that the results may have been skewed because many women who suffer morning sickness reduce their caffeine intake to ease nausea and vomiting. Women who suffer morning sickness are less likely to miscarry than those who do not, possibly because the hormonal changes that cause morning sickness also are part of a healthy pregnancy.

The Kaiser study addressed that issue by examining both women whose caffeine consumption changed during pregnancy and those whose caffeine use didn't change.

The study also adjusted for such factors as a mother's age, race and income level, and whether she smoked, consumed alcohol, used a hot tub or had a previous miscarriage.

The study appears online today in the Web site of the American Journal of Obstetrics and Gynecology, www.ajog.org.

Not only should pregnant women limit their caffeine consumption, women who are attempting to become pregnant should consider doing so as well because the first 20 to 40 days after an egg is fertilized is a key time in its development. As a result, harm could occur before a pregnancy is confirmed, said Dr. David Walton, a Kaiser perinatology specialist in California.

"Caffeine crosses the placenta barrier easily," said Dr. De-Kun Li, the study's lead investigator. Once in the fetus, it may stay there longer because fetuses have an underdeveloped metabolic system, Li said.

In large quantities, caffeine may also decrease placental blood flow and harm cell development, experts say.

"We recommend avoiding caffeine, but if people are compelled to have it, we tell them for sure to limit it," Walton said.

Interviewers asked the women about the frequency and amount of beverages they consumed and whether they changed their patterns after becoming pregnant.

The researchers then compared caffeine use with rates of miscarriage during the first 20 weeks.

"The increased risk of miscarriage appeared to be due to caffeine itself rather than other possible chemicals in coffee because caffeine intake from non-coffee sources showed the similarly increased risk of miscarriage," the study found.

While there was some indication that consuming less than 200 milligrams of caffeine daily might increase the risk, the numbers were not large enough to be statistically significant, and thus no conclusions could be drawn about smaller amounts of caffeine, Li said.

Walton said he is concerned that women who had a miscarriage several years ago will now blame themselves because they drank coffee during their pregnancy. He noted that many other factors can lead to miscarriages.

"What we're trying to tell people is that a lot of times we use caffeine because we have bad lifestyles," Walton said. "So if we can make our lifestyles better and exercise more and sleep better, then caffeine isn't such a compelling part of our life.

Experts suggest seeking a natural energy boost by taking a brisk walk, doing yoga stretches, or snacking on dried fruits and nuts.

McClatchy-Tribune and Bloomberg news services contributed to this report.