Bad air and birth defects. (Air Pollution).

Women exposed to levels of ozone and carbon monoxide that occur in large urban areas may have a higher chance of giving birth to babies with serious heart defects, according to researchers at the University of California, Los Angeles (UCLA). The study, published in the American Journal of Epidemiology in January, provides the first compelling evidence that air pollution may play a role in causing some birth defects, according to lead author UCLA epidemiologist Beate Ritz.

Three observations make the work compelling, she says. The heart defects correlate with exposure in the second month, when the heart and other organs form. In addition, there is a "clear dose-response relationship to the data," and no chromosomal defects were associated with the air pollution.

The research was carried out at the UCLA School of Public Health and the California Birth Defects Monitoring Program, in Oakland, California. Previous research by the same team has linked air pollution to harmful effects on pregnancy, including premature birth and low birth-weight babies. Recent studies conducted in China, the Czech Republic, Brazil, Mexico, and the United States have related ambient air pollution to such adverse birth outcomes as low birth weight, preterm birth, and fetal mortality.

Ritz's team looked at air monitoring data for pollution that comes directly, or indirectly, from vehicles--carbon dioxide, nitrogen dioxide, ozone, and P[M.sub.10] (particulate matter less than 10 mm in diameter). The pollution monitoring data came from the South Coast Air Quality Monitoring District, the air pollution control agency for southern California. By matching ZIP codes, the researchers compared this with information from the California Birth Defects Monitoring Program, a large, population-based registry on birth defects.

Using information on more than 9,000 babies born between 1987 and 1993 in the Los Angeles area, the team found that at increased levels of ozone and carbon monoxide, pregnant women faced an elevated risk of having a child with serious heart defects. These defects include holes in the heart, arterial defects, or pervasively malformed hearts. Normally this group of heart defects occurs 1.76 times per 1,000 births, with about 935 cases in California each year.

Ritz split exposure levels into quartiles and then compared each quartile with the lowest exposed group. "The clear dose--response for these effects is striking," she said. For women living in areas with the highest pollution levels, the risk tripled in comparison with women living in clean air areas. At moderate pollution levels, the risk was doubled.

The lack of chromosomal defects also led Ritz to attribute the heart defects to air pollution. This is because chromosomal defects occur at conception and should not be influenced by environmental factors during pregnancy.

However, the link with ozone and carbon monoxide is not clear. "We're not sure carbon monoxide is the culprit because it could be just a marker for something else in tailpipe exhaust," said co-author Gary Shaw of the California Birth Defects Monitoring Program. "The fact that certain heart defects are turning up in the second month of pregnancy when hearts are being formed suggests something serious may be happening."

Although the study is the first rigorous effort to demonstrate a link between air pollution and birth defects, the findings do have limitations, said Ritz. Researchers were only able to estimate mothers' exposure to routinely measured air pollutants. They relied on air pollution concentrations collected at the air-quality monitoring station nearest a mother's home--in some cases up to ten miles away. Also they were unable to evaluate other potential risk factors for birth defects, including maternal smoking, occupational exposures, vitamin supplement use, diet, and obesity.

Ritz and her colleagues are currently addressing these limitations by improving exposure estimates in a new study investigating the effects of confounding factors on preterm birth and birth weight for a group of 2,000 infants.