Evidence based treatments and IVF success rates
Like other areas of medical practice, fertility treatments should not be offered by clinics unless there is evidence to show that they work and are safe. Unfortunately there are often new treatments offered in clinics around the world and announced with excitement in the news media which do not meet this standard. Here at St Mary’s we offer only treatments that are based on evidence. We do however keep a careful eye on all new developments in the field, as part of our clinical practice and our research programme. We also do this through our regulator and professional bodies; two of our senior staff are members of the HFEA’s Scientific and Clinical Advances Advisory Committee and Chair of the British Fertility Society’s Practice and Policy Committee, both of which bodies continually evaluate new treatments in the UK.
A good example arose in the newspapers recently, when a “new test” of mitochondrial DNA in embryos was announced at the American Society for Reproductive Medicine annual conference in the US, which apparently promised to double or treble success rates in IVF! This is interesting research by a group in the UK which I know well, which is at a very early stage, but was totally oversold in the media. This frequently occurs at conferences, since journalists gather there to catch up on the latest science and meet the scientists. However announcements made at conferences are nearly always new by their nature, and very rarely “peer reviewed” (subjected to critical analysis by other scientists). A high proportion of these exciting new conference announcements are in fact never published as scientific articles because they later turn out not to be true. With this particular study, what the scientists actually said in their presentation was that if you apply this new test to a group of patients who already have a very high chance of success (65% pregnancy rate per cycle, because they are in the select group of patients with multiple embryos to choose from on day 5 or 6), then their success rate might increase to 75-80%. So this is an actual increase of about 20%. Nice to have, but hardly dramatic, and so far only in this small group of patients.
So how did journalists report this as an overall doubling or trebling of IVF success? Apparently they simply took the figure of 80% and divided it by the overall success rate of 30-40% reported by most clinics. But hopefully we would all understand that a success rate of 65% can hardly be increased by 2 or 3 times!
Unfortunately this test could never increase overall success rates in IVF even by a modest 20%, because it can only be applied to cycles which have multiple embryos available on day 5/6 for transfer (a minority of cycles in the UK). It also does nothing to solve the problem that no matter which embryo we select for transfer, some women unfortunately are not receptive at that particular time and so the embryo will not form a pregnancy. Finally, the test is invasive and involves removing cells from the embryo, a procedure which may cause stress to the embryo in itself and is currently only recommended for diagnosing inherited genetic disorders. It is equally likely that freezing embryos and replacing them at a later time will be just as successful. However when the evidence does become available for any new test in IVF, we at St Mary’s will look at it carefully on behalf of our patients.
Finally, the good news from St Mary’s is that even without invasive tests of the embryo, our recent success rates are now very high. This is partly to do with investment by our NHS Trust in state of the art technology such as Embryoscope timelapse incubators. The HFEA are also currently reviewing the way they present success rates from clinics, as it is widely understood that success rates cannot be directly compared between clinics in any way which is meaningful to patients. The likely outcome is that the way results are presented will change in the new year and senior staff at St Mary’s have again played an important part in this development.