Women who drink alcohol while pregnant raise the risk that their child will develop a rare type of blood cancer called acute myeloid leukemia, or AML, a new study suggests.

In a pooled analysis of data from published studies, researchers found that drinking alcohol during pregnancy was associated with a 56 percent increased risk of AML in children.

But in an email to Reuters Health, Dr. Julie Ross of the University of Minnesota in Minneapolis, who was not involved in the study, said it’s important for women to know that childhood AML is rare (about 700 cases are diagnosed each year) and likely has many causes.

Moreover, she said: “The vast majority of women who consume alcohol during pregnancy will not have a child who develops leukemia. However, given other risks associated with alcohol drinking during pregnancy, these results can help to reiterate the message that it is probably best to abstain from alcohol if you are pregnant or planning to get pregnant.”

What causes leukemia in children is largely unknown, but researchers suspect it may be an interaction between genes and the environment, including drinking alcohol, Dr. Paule Latino-Martel of the Research Center for Human Nutrition in France and colleagues note in a report published online today.

When the researchers looked at 21 previous studies of women’s drinking habits during pregnancy and childhood leukemia, they found that women who drank during pregnancy had a 56 percent increased risk of childhood AML, they report in the journal Cancer Epidemiology, Biomarkers & Prevention.

Ross cautioned, however: “While a 56 percent increased risk sounds like a lot, in real terms it means that with a childhood AML incidence rate in the US of about 8 cases diagnosed per million children, the risk might increase to about 12 cases diagnosed per million children. Thus about 4 more cases per million children.”

Despite the recommendation that women abstain from alcohol while pregnant, it’s estimated that 60 percent of Russian women drink while pregnant, as do 59 percent of their Australian counterparts. Fifty-two percent of French women, 30 percent of Swedish women and 12 percent of American women drink while pregnant, according to estimates.

For American women, Ross further explained: “If we are to believe that the risk (of childhood AML) is increased by about 56 percent, and that about 12 percent of US women drink alcohol sometime during pregnancy, this means that perhaps up to about 6 percent of childhood AML in the US might be attributable to alcohol consumption.”

On the flip side, “this also means that 94 percent might be attributable to other causes,” Ross noted.

Only a few of the studies reviewed by Latino-Martel’s group reported results according to type of alcohol consumed - beer, wine, or spirits - and the existing evidence does not suggest that one type of alcohol could be more related to leukemia risk than another, they say.

The limited available data also make it impossible to tell whether it matters when women drink in the course of a pregnancy, although risk tended to be higher when alcohol was consumed later in pregnancy.

The researchers say it’s possible the results were skewed because some women who drank during pregnancy did not admit to it “due to the stigma.” There were other limitations: Only a few of the studies adjusted for smoking during pregnancy and factors such as exposure to pesticides, folate intake, birth weight and age of the women were largely not taken into account.

Despite these limitations and caveats, the current findings serve to strengthen the public health recommendation against drinking alcohol during pregnancy, the study team, and Ross, conclude.

SOURCE: Cancer Epidemiology, Biomarkers & Prevention, online May 6, 2010.