Their study concludes the method would reduce by 80 percent the need for a riskier but definitive test called amniocentesis, in which a needle withdraws fluid from the amniotic sac. That triggers miscarriage about 1 percent of the time.
Compared with standard U.S. testing, the new system would prevent the miscarriages of about 1,400 healthy fetuses each year and detect about 800 more fetuses with the abnormality, said Dr. Nicholas Wald, the lead researcher.
The screening also could reassure more expectant parents their baby doesn't have Down syndrome, a condition in which an extra chromosome causes mental retardation and physical abnormalities.
"You've got an improvement in detection, but the real gain is the reduction in the number of women who have to have an amnio," said Wald, chairman of the Wolfson Institute of Preventive Medicine in London.
The research is reported in Thursday's New England Journal of Medicine.
About one in 700 babies has Down syndrome, which is marked by a broad, flat face with slanting eyes and an early death. Older women have a higher risk of delivering a baby with Down syndrome -- for example, for a woman who is 40, the rate is one in 106.
Wald and colleagues developed a complicated computer program that integrates results of tests in both trimesters. First, a woman is given an ultrasound and a blood test for a protein linked to Down syndrome, both done 10 to 14 weeks into the pregnancy. A month later, she has blood tests for abnormal levels of four substances associated with a high risk of Down syndrome.
American women generally get only second-trimester screenings for three of those substances -- two hormones and a protein.
Wald's computer program can be set to allow only a 1 percent chance of a false positive result that would lead a doctor to suggest the invasive amniocentesis test. With that setting, the program correctly identified 85 percent of Down syndrome cases. Standard screening detects 46 percent.
"For women who don't want an invasive procedure, this offers them a very viable choice," said Dr. Anne Summers, director of the Maternal Serum Screening Program at North York General Hospital in Toronto, which will begin using the system soon.
Although Summers' hospital plans to use Wald's system, she said more research is needed to confirm the accuracy of his research. That's because his calculations were done by piecing together data from three other studies, each involving different women receiving only some of the tests.
Malone and Wald are now running studies of the tests on tens of thousands of expectant mothers.
In an editorial in the journal, Drs. Joshua Copel and Ray Bahado-Singh of Yale University argue that Wald's method wn't gain wide acceptance because it requires withholding results of first-trimester tests until they are combined with the later data. Women might prefer an earlier warning so they could terminate the pregnancy safely, they said.
Written By Linda A. Johnson