New test predicts risk of twins from IVF

Couples undergoing in vitro fertilization for infertility often face an excruciating choice: Should they have two or possibly three embryos implanted and run the risk of having twins or triplets, or should they have a single embryo implanted with the greater likelihood that no pregnancy will be achieved at all?

Previously, doctors used age as a way to calculate the likelihood of multiple births, but a new formula that considers multiple factors beyond age — developed using data from the fertility center Boston IVF — is far more reliable and could ultimately reduce multiple births resulting from IVF procedures, according to a study published Monday in the journal Fertility and Sterility.


The model plugs in several variables that determine the viability of an embryo, including a woman’s age, her body mass index, and levels of reproductive hormones. It also considers a man’s sperm count and the quality of the embryos produced in a petri dish for transfer into a woman’s womb.

Developed using data from nearly 37,000 IVF treatment cycles conducted at Boston IVF from 2000 to 2009, the new test calculates the probability that a woman will wind up giving birth to twins if she has more than one embryo implanted.

If a woman’s risk of having multiples was as high as 30 or 40 percent, doctors might be reluctant to transfer more than one embryo, said study co-author Dr. Alan Penzias, a reproductive endocrinologist at Boston IVF. “But what is the level above which the risk is too high to transfer two embryos? We haven’t come up with that yet.’’

Regardless, the decision will largely rest in the couple’s hands and how comfortable they are with the possibility of raising twins — or how concerned about having a failed pregnancy. Boston IVF plans to start offering the predictor tool, called IVFsingle, to patients within the next few months, Penzias said. The clinic hasn’t yet calculated how much it will cost or how it will be factored into the state’s mandated insurance coverage for IVF procedures, he added.


Penzias will soon become a paid member of the advisory board of Univfy, the California company that developed the risk prediction tool.

Other fertility clinics in the region could, if they deem the test worthy, start offering IVFsingle in their own clinics, though that would likely require them to set aside about $2,500 in costs to run their own clinic’s data through the model in order to get more accurate predictions; the test weighs certain factors, like age or body mass index, based on the population that comes into the clinic to make its predictions for individual patients.

The ultimate goal, Penzias said, is to decrease the rate of multiple births that result from IVF procedures. More than 30 percent of IVF-assisted births nationwide result in twins and, more rarely, triplets, compared to just 1 to 2 percent of births that occur without the use of reproductive technologies.

Other IVF predictor tests have been developed by Univfy
, including a test, called PredictIVF, that predicts a couple’s chances of having a successful pregnancy after a failed IVF attempt. That’s being tested at a handful of fertility clinics outside of Massachusetts, according to Univfy president Dr. Mylene Yao, a study co-author who trained as a reproductive endocrinologist at Brigham and Women’s Hospital. That test costs $350 and could be covered by insurance if it’s bundled with other IVF services.

Another test, called PreIVF, can predict a couple’s chances of achieving a successful pregnancy using IVF for the first time and will be offered for sale directly to consumers this summer for a cost of $250.


In addition to the fee, couples will need to submit relevant medical data and two medical test results that predict IVF success: a sperm count for the man and a measurement of follicle stimulating hormone for the woman. Studies detailing the reliability of PreIVF, however, haven’t been published, though Yao said results from one study have been submitted to a medical journal and will be presented this fall at a scientific meeting.

Loading Comments...