The last thing a pregnant woman wants to hear is that a medical test she needs to undergo may harm her child down the line.
New research alleviates one of these commonly held fears: that being exposed to radiation from procedures such as CT scans and radionuclide imaging while pregnant can lead to cancer in her child.
In a study of 1.8 million live births in Ontario from 1992 to 2008, researchers have found that the risk of childhood cancers is about the same with or without exposure to these forms of radiodiagnostic testing.
Lead author Joel Ray, who practises obstetrical medicine at Toronto's St. Michael's Hospital, says the findings may help women feel less conflicted about undergoing these increasingly common tests. They are often used to diagnose major conditions such as blood clots in the lung, stroke and appendicitis.
"We haven't eliminated the unknown, but we've filled up part of the unknown glass here," he says. "We hope we've made it easier on a woman and her health care provider to have a discussion that lowers her anxiety."
Some previous studies have shown a link between radiation exposure in utero and childhood cancer, but Dr. Ray says much of that work looked at the practice in the 1950s and 1960s of X-raying a woman's pelvis to see whether the size of the baby could present problems in childbirth.
While even that risk was found to be low, "We would never do that [now]" he says referring to X-raying a pregnant woman's pelvis. "That's why medicine needs to check itself - because it does some crazy things."
Despite the low incidence of cancer among children, it remains the second-leading cause of hospitalization and death among children aged 14 years and younger in industrialized countries. And since fewer than 10 per cent of childhood cancers are attributable to genetics, there remains a great deal to learn about what's causing them.
The newer diagnostic tools use more radiation than regular X-rays, and about one out of every 160 pregnant women undergo one, says Dr. Ray, also a scientist with the Institute for Clinical Evaluative Science (ICES), which supported this research. The study appears in the September issue of the Public Library of Science's journal, PLoS Medicine.
A computed tomography scan (CT scan) creates a three-dimensional image of a person's anatomy using a series of X-rays. Radionuclide imaging involves injecting a patient with a radioactive substance to enhance the images of internal organs.
Of the 1.8 million births, Dr. Ray and his colleagues found that 5,590 mothers were exposed to radiodiagnostic testing - 73 per cent of those CT scans. The researchers then followed all the children born for at least eight years - and in some cases up to 14 years. Four cases of childhood cancers arose in the exposed group and 2,539 cases in the unexposed group.
Based on the total figures, the incidence rate of cancers in the unexposed group was 1.56 cancers per 10,000 person-years observed, and the rate in the exposed group was measured at 1.13 - lower than the unexposed group. But the margin of error could put that rate at anywhere from .25 to 1.8.
Dr. Ray acknowledges that because the number of cancer cases was so low, his findings have a wide margin of error. For an ideal level of statistical accuracy, he would have needed about eight times the number of cancer cases to compare.
Even though the highest possible risk of 1.8 cancers per 10,000 person-years is still small, Dr. Ray and his colleagues advocate a cautious approach.
Lead aprons should continue to be used for any pregnant woman and woman of child-bearing age for radiodiagnostic testing. In some cases, pregnancy tests should be administered in the event that an ultrasound or MRI test could be used instead. And along with those measures, health care providers should offer counselling about the low-risk nature of the radiodiagnostic testing. In his practice, Dr. Ray has had to talk a seriously ill pregnant woman back into a CT scan after her fears caused her to abandon it.
"Today, somewhere in Canada, a pregnant woman is going to get a CT scan. And when she gets it, the best thing we can do is to minimize her anxiety at a time when she's not well and had worries on her mind," he says. "… we can't keep theory running higher than data."