Experimental Test May Aid Decisions After Failed IVF

July 20, 2010— -- Only one in four attempts at in-vitro fertilization, or IVF, results in a live birth, despite a slow but steady rise in the overall success rate of these procedures.

Now, a new model developed by a team of researchers from Stanford University could significantly improve the accuracy with which fertility specialists predict IVF success after an unsuccessful attempt.

Currently, patients who have attempted IVF and who have not been able to get pregnant from it have little more than general age-related prognostic criteria on which to base future tries. And with an average cost of $12,400 per cycle, according to the American Society for Reproductive Medicine, as well as the psychological distress associated with infertility, decisions surrounding IVF are complex to say the least.

The researchers hope this new test will help cut through at least some of the uncertainty when it comes to these procedures.

"We have established an externally validated, highly discriminatory, well-calibrated, and robust prediction model that can use available clinical data from a previous cycle to predict live birth rates in a subsequent cycle, without additional clinical or laboratory testing," Prajna Banerjee and colleagues reported online in the Proceedings of the National Academy of Sciences.

Using data from unsuccessful attempts at IVF, the model provided more accurate results in 60 percent of second-cycle IVF procedures. In more than half of the cases, the model improved the odds for live birth compared with age-based criteria.

In addition to age, factors such as hormone levels, endometrial thickness and body mass index influence the likelihood of IVF success. And after one attempt, the details of such factors are known.

"Any information you can give a patient that can help them make a decision about IVF is going to be very helpful," said Dr. Michael Thomas, professor and director of Reproductive Endocrinology and Fertility at the University of Cincinnati Academic Heath Center. "It looks like this model has potential, but it has to be validated by other centers."

Tool May Help Physicians Counsel Patients on IVF

Because IVF success rates vary between medical centers, whether the model will hold up across the board remains to be seen.

According to data from the Society for Assisted Reproductive Technology, the IVF success rate for a 35-year-old woman is 47 percent, dropping to two percent by age 45. But the new, more personalized model yielded more favorable odds in most cases.

"Every physician wants to be able to accurately and fairly counsel patients on success or futility of further treatment, and this model appears to be a valuable asset over generic age-based counseling, which does not take unique patient and embryo factors into account," said Dr. Jani Jansen, assistant professor of Obstetrics and Gynecology at the Mayo Clinic.

"Deciding whether or not to continue with further infertility treatment after a failed IVF cycle is a complex decision often made on emotional or financial grounds without good data to support the decision process," she said. "This model seems like it would provide evidence for patients to make the best decision for their unique circumstances."