Until recently, the precarious first moments of an embryo’s development have been a mystery to doctors and scientists. A team of Stanford researchers, led by obstetrics and gynecology Prof. Renee Reijo Pera, recently managed to film early embryonic development, thus discovering more accurate ways to predict the success of an embryo developing into a child.

In vitro fertilization (IVF) is an increasingly popular form of infertility treatment for women. But though the technology has existed since 1978 and the successful birth of Louise Brown, the treatment is still riddled with flaws and risks.

One major problem in IVF is the low pregnancy rate–that is, the percentage chance that a woman who undergoes IVF will actually become pregnant. Currently, IVF practitioners create several embryos, which are grown and observed for up to five days before implantation. The practitioners pick embryos that are dividing regularly, but up to now there has been no way for doctors to predict whether or not the fertilization will be successful.

Nationwide, the live birth rate with IVF is only 30 to 35 percent for women under 35 years old, according to the American Pregnancy Association. This number continues to drop as the woman’s age increases. Failures are costly–not only financially, but emotionally as well. Another potential consequence is the higher likelihood of twins and triplets, which can result when multiple embryos are transferred to increase the chance of success.

“It’s difficult to tell if an embryo is going to make it or not,” said Reijo Pera. “It’s been a problem ever since 1978, and this is the number-one problem in IVF.”

Motivated by the low success rates in IVF, Reijo Pera set out to study the human embryo. Working with a team at Stanford’s Institute for Stem Cell Biology and Regenerative Medicine, Reijo Pera employed time-lapse imaging techniques on more than a hundred fertilized eggs to study their development.

Quickly, her team realized that the first stages of embryonic development were in fact quite orderly and not as chaotic as scientists had always presumed. By the time the embryo had divided into eight cells–which is roughly “half the size of a point of a pin,” according to Reijo Pera–the cells were already acting autonomously.

As a result, Reijo Pera realized that the chances of survival were not a murky mystery, but rather a predictable process that could be modeled with mathematical formulas.

“We can predict success or failure with over 93 percent accuracy,” she said.

Not only did they realize they could predict the success rate, but they could do so by the second day of development.

“The embryo is still genetically silent on day two,” said Reijo Pera. “The embryo has not even turned on its genes. So the success or failure must be genetically inherited.”

Reijo Pera gives credit to developing technology as a reason for this latest discovery. She explained that embryo development could be affected by light, so the microscopic cameras had to work in dim conditions.

“There’s also a component that nobody has really thought to do this before,” she added. She hopes that the discovery will help increase pregnancy rates and reduce the complications for women.

The technology has been licensed by Auxogyn, Inc., a bio-tech company in Menlo Park co-founded by Reijo Pera. The company hopes to start clinical trials by 2011 to determine if success rates are equally accurate outside the lab.