Now, a new paper just out in the journal Fertility and Sterility reports that it's possible to predict with much-improved accuracy just how likely a given couple is to have twins. It finds striking differences in the chances that a given patient will have twins if two embryos are transferred, ranging from about 12% to about 55%.
You can just imagine how the statistical tool the researchers used — which belongs to a private company, Univfy — could play out in clinic conversations: "Ms. Jones, we could transfer two embryos, but you'd then have a 55% chance of having twins. Perhaps we should do a single-embryo transfer?"
What affects the chances of twins? The study's press release says: "Variables that most affected chances of multiple birth among the Boston IVF patients studied include age, number of four-cell embryos, total motile sperm, serum peak estradiol level, and Day 3 FSH levels." (Those last two are hormones in the woman.)
More from the press release:
Says Dr. Alan Penzias, surgical director at Boston IVF: “As happy as it seems when someone has twins, the risks in even a twin pregnancy are significantly higher than for a singleton. With the tool that Univfy has developed, we’re able to get an idea if someone has a relatively low risk or an exceptionally high risk of a multiple pregnancy. Depending on their chances, we can counsel them to put in one embryo to minimize the risk of multiples or two embryos to boost their odds of getting pregnant at all.”
In the journal article, entitled “Predicting personalized multiple birth risks after in vitro fertilization (IVF)-double embryo transfer (DET),” by Benjamin M. Lannon, M.D., et. al., the authors applied advanced statistical prediction analysis to 2,413 IVF double embryo transfer (DET) cycles that resulted in live births over a 10-year period (2000 to 2009) at Boston IVF, a large, private practice in Waltham, MA, whose providers are on the Harvard University faculty. The researchers found that individual IVF patients have inherently different risks of multiple birth probabilities, ranging from 11.8% to 54.8%--rates significantly different from probabilities based solely on age in over half the patients. The researchers’ model showed over 100% improvement in predictive power over age-based probabilities commonly used today.
This program aired on June 4, 2012. The audio for this program is not available.