The study included more than 52,000 pregnant women in Canada. It found that miscarriage rates were more than twice as high among those who filled at least one prescription for a nonsteroidal anti-inflammatory drug (NSAID) compared to women who didn't get prescriptions for those kinds of medications.
Because they're so common, researchers say many women reach for them before they suspect they're pregnant.
"In our study, women used NSAIDs, on average, for four days during early pregnancy. That was enough to trigger an increased risk," says study researcher Anick Bérard, PhD. Bérard is an epidemiologist and professor of pharmacy at the University of Montreal.
"Many of the drugs that are available over the counter are believed to be safe during pregnancy, which is not always the case. NSAIDs are a good example of that," she says.
The study is published in the Canadian Medical Association Journal.
A similar, large study by Danish researchers, published in BMJ in 2001, found that NSAID use was associated with a nearly threefold higher risk of miscarriage. But such studies can only show associations. They do not prove that NSAIDs cause miscarriages.
And experts say they often have weaknesses that make the links they find less reliable.
Miscarriages and NSAIDs
For the current study, researchers identified more than 4,700 women in Canada who had lost a baby before the 20th week of pregnancy.
Each one of those women was compared to 10 other women in a national registry of the same age who had not miscarried by the same point in her pregnancy.
Overall, 352 women (7.5%) of those who had miscarried had filled one or more prescriptions for an NSAID during pregnancy.
That compares with NSAID use in 1,213 women (2.6%) of those who had not miscarried.
The risk varied slightly by medication.
Women with prescriptions for diclofenac, for example, had about three times the risk of miscarriage compared to those who didn't take that medication. Those taking naproxen had a 2.64 times higher risk.
Celebrex and ibuprofen were each associated with a slightly more than doubled risk. Those taking Vioxx, which is no longer sold in the U.S., had an 83% greater risk.
Those risks remained when researchers adjusted their results to remove the influence of other things known to increase miscarriage risk, including a history of miscarriage and underlying health conditions like diabetes, depression, rheumatoid arthritis, lupus, and untreated thyroid disease.
The dose of the medication didn't seem to be tied to the degree of risk.
"This leads me to believe that women using NSAIDs over the counter, which are known to be lower dosage, are at the same risk as those who are going and getting a prescription," Bérard says.
One limitation of the study was that it relied on NSAID prescriptions as a marker for their use.
In Canada, one popular NSAID, ibuprofen, is also available over the counter.
As a result, many women in the study who were counted as not taking NSAIDs because they didn't have a prescription might actually have been using them.
Also, getting a prescription doesn't mean that a person actually takes the drug, says Gideon Koren, MD. Koren is a pediatrician, toxicologist, and pharmacologist who directs the Motherisk Program at the Hospital for Sick Children, in Toronto. He was not involved in the study.
Researchers in Sweden, who were looking at the safety of using migraine medications during pregnancy, recently tested the reliability of prescription records by going one step further and interviewing women about their medication use.
They found that many women didn't actually take the drugs their doctors ordered for them. And Koren says that besides their prescription history, there were also important differences between the women who had miscarriages and the women who did not that likely muddied the study's results.
About 15% of the women who didn't have miscarriages got prescriptions for anti-nausea drugs, compared to about 3% of the women who miscarried, a fivefold difference.
"Morning sickness protects against miscarriages," Koren says.
Other differences between the groups were higher rates of depression and anxiety and poverty in the miscarriage group than in the comparison group, though researchers tried to remove the influence of those things by adjusting their data to account for them.
Advice to Women
Taken together, he says he thinks pregnant women should not worry if they have taken an NSAID medication.
"Scaring pregnant women is a national sport in Canada and the United States, and I don't think this is good," Koren says. "It may lead women to not treat themselves when they need it, or even to terminate a wanted pregnancy. I think we have to be extremely careful before we make an association that has major methodological issues."
Study researchers acknowledge that more research is needed to prove that NSAIDs cause miscarriage.
But Bérard says she thinks that in this case, when there are other, perhaps safer choices of pain relievers for pregnant women to use, it's good to get the word out in case the increased risk is real.
"At least during the first trimester, women should think about safer alternatives, and I'm thinking of acetaminophen here," she says. Acetaminophen is the generic name for Tylenol and multiple other medication brands.
Bérard says women can lower their risk of inadvertent exposure by planning their pregnancies.
Pregnant women should never self-prescribe medication, Koren says. "Talk to your health care provider," he says. "There may be important issues that need to be discussed."