Optimising IVF practice: Fresh or frozen embryos?

Is it better to transfer fresh embryos during IVF or frozen embryos that have been thawed? Traditionally, the thinking in fertility medicine had been that fresh embryos would be the natural choice – where possible – as it avoids the potential for damaging the embryos through freezing. But several studies over recent years have come out in favour of using frozen embryos, and some experts are now suggesting that all embryos should be frozen.

It may at first seem counterintuitive to prefer frozen over fresh embryos, but there are some good reasons why this might be better. If we transfer fresh embryos, we're placing them into the uterine environment of a woman who has just been stimulated to produce numerous eggs. This is much like placing a seed into poor soil because the hormones released by the developing ovaries are those that regulate uterine readiness, and during ovarian stimulation those hormone levels are much greater – often ten times greater – than natural levels. So it may be better to store the embryos and let the uterus recover from these extreme hormone levels.

To date, there have been four published randomised trials comparing fresh and frozen-thawed embryo transfers. All four of these studies reported greater success rates with frozen-thawed embryo transfers than with fresh transfers, and in three studies that difference was statistically significant. Infants resulting from frozen-thawed embryo transfer are also a little healthier, on average, than those resulting from fresh transfer.

In the most recent of those four studies, Professor Zi-Jiang Chen and colleagues at several universities in China performed a randomised trial comparing outcomes of fresh and frozen-thawed embryo transfers in 1500 patients with polycystic ovary syndrome (PCOS). They reported a live birth rate of 49.3 percent following frozen-thawed embryo transfer, and 42 percent following fresh transfer. This difference was statistically significant.

Our own center reported a generally similar, but much smaller, randomised trial in high responders, a population that includes PCOS patients. We reported clinical pregnancy rates of 80 percent with frozen-thawed transfers and 65 percent following fresh transfers. Our study had only 101 embryo transfers, so the observed difference was not statistically significant. We eventually halted that study because the study protocol (written in 2007) specified the transfer of two blastocysts in every patient whenever possible, and that proved to be one embryo too many. Most of the pregnancies were multiple pregnancies (mostly twins), and this was unacceptable for safety reasons.

The difference in success rates between our study and the recent study by Professor Chen and colleagues might be attributed to our uniform use of blastocyst-stage embryo transfer, while Professor Chen's study featured only cleavage-stage embryo transfer. Blastocyst-stage embryos are a few days older than cleavage-stage embryos and tend to implant more readily. Also, we compared clinical pregnancy rates (the proportion of transfers that resulted in at least one fetal heart tone observed on ultrasound), a proportion of which will not result in a live birth, while Professor Chen compared live birth rates.

That said, the two study results have substantial agreement in the ratio of successes. The ratio of successes for frozen compared to fresh embryos in our study was 1.23, and in Professor Chen's study it was 1.17.

We also recently published a comparison of birthweights in fresh and frozen-thawed transfers, in which babies born as a result of frozen-thawed transfers were on average around six ounces heavier at birth. Professor Chen and colleagues reported a very similar difference. Similar findings have been reported in many national registry studies. While six ounces is not a huge difference, we consider this slight increase in birthweight as a good omen for frozen-thawed embryo transfer as low birthweights are associated with more frequent, severe, and costly complications in childbirth.

Professor Chen and colleagues also reported other differences between fresh and frozen-thawed embryo transfers. The rates of ovarian hyper-stimulation syndrome (OHSS) and pregnancy loss in his study were reduced with frozen-thawed embryo transfer. OHSS is a rare but serious complication of ovarian stimulation that can require hospitalisation. However, pre-eclampsia, a potentially serious maternal complication of pregnancy, was more frequent in patients receiving frozen-thawed embryo transfer (4.4 percent vs 1.4 percent). Pre-eclampsia affects two to seven percent of spontaneous (non-IVF) pregnancies.

There are several additional studies currently underway comparing fresh and frozen embryos, and we await their outcomes with interest. In the meantime, several centers in the USA have adopted or are adopting protocols in which all embryos are cryopreserved. This trend is expected to continue, and we suspect the frozen embryo transfer is already more frequent than the fresh transfer in the USA.

The Progress Educational Trust is holding a free-to-attend event public entitled 'Frozen Assets? Preserving Sperm, Eggs and Embryos' in Edinburgh on the evening of Tuesday 25 October.

Click here for details, and email sstarr@progress.org.uk to book places.