Difference in Outcome from IVF and ICSI: Is it Gonadotrophin Dependent?

View the video (requires free Windows Media Player)

Cees Jansen, MD, PhD: “I am interviewing Dr. Johan Smitz, who has recently presented some interesting data on a study that has been performed with recombinant and with urinary FSH at the State of the A.R.T. Conference two days ago. Johan, could you tell me what you have investigated and what the results were?”

Professor Johan Smitz, MD, PhD: “This study was conducted in 22 centres from Europe and in Israel in 1999 and 2000. Endocrinologically normal women were down-regulated generally with the long GnRH-agonist protocol and then the patients got either highly purified hMG and the other arm got recombinant FSH. All these patients had to be treated by an ART so it was a combination of patients with IVF or ICSI. The results that got published earlier this year, came to the conclusion that both products had the same efficacy. 

When results were broken down according to the different ART treatments we found that, in fact, in the ICSI treated patients there was not such a difference in outcome but that, in the IVF group, the outcome with the HP hMG was significantly better than with the recombinant FSH. 

This was surprising and we therefore started to look more in detail into some parameters of the stimulation. This quite extensive statistical analysis brought up that there seems to be a strong relationship between the concentration of hCG and less for LH and the outcome of pregnancy in the group treated with HP hMG. So, apparently, the effect of having a higher concentration, of hCG was positively correlated with a higher chance of achieving clinical pregnancy.”

Cees Jansen, MD, PhD: “How do you explain the difference between ICSI and IVF?”

Professor Johan Smitz, MD, PhD: “Well, the difference could be attributed to two possibilities, in my mind. First of all, we all know that the women from ICSI differ from those in IVF treatment, female pathologies being higher in IVF women. So the first possibility could be that there exists an inherent, yet undefined, deficiency of LH in the IVF population. 

A second possibility could be related to the type of treatment of the female gamete. There is one consistent in-vitro manipulation difference between ICSI and IVF. In ICSI, you, in effect, remove the cumulus cells before injecting them with a male gamete, while with IVF, you keep these cumulus cells around the oocyte during and after fertilisation for quite a period. So, another possibility could be that exposure of the follicles during their maturation period to LH brings certain factors to the oocyte that are expressed very late in her development during the post – ovulatory period. So this is a very plausible second possibility that definitely needs confirmation of course in a prospective study and also in some experimental set ups that we could design in the ART laboratory.”

Cees Jansen, MD, PhD: “Because at the same conference, there was compelling evidence that there is a continuous inter-relationship between the granulosa cells and the oocyte, the zona pellucida, and the oolemma and vice versa, also an interaction from the oolemma to the granulosa cells. Basically, that means that there might be something in it where this interaction is, for instance, in case of ICSI, disturbed. That might lead, let’s say, to a more favourable effect in the case of hMG. Could you tell me what are, in your view, the clinical implications of your findings? Are there any and, if so, what are they?”

Professor Johan Smitz, MD, PhD: “Oh, I think there are very important implications. First of all, I started to re-analyse the patients that had been treated in our own centre and it was a relatively small group of 72 patients, where we had blood samples banked during the follicular stimulation. I was very interested in digging out some more on the endocrine profile of these patients when being treated in the two arms of the study. 

Now in our group of patients, all women were down-regulated with decapeptyl depot administered at day 21 and quite a number of women, around 30% of them, after this treatment had really extremely low LH levels three to four weeks after the start of the agonist. Approximately 30% in both groups treated by HP hMG or recombinant FSH had an LH lower than 1 IU (one international unit) per litre. 

Then, to my surprise, in fact as soon as you start to administer these women hMG or recombinant FSH, you see that proportionally more and more women get LH-suppressed during the growth phase of the follicles. This can be explained by the feedback effects by what is produced from the ovaries. When you make a mean concentration of LH throughout the follicle phase, you see that you end up in the recombinant FSH group with approximately 75% of the women with a very low LH, the same, but less drastically, that is true with hMG. So the big difference is that in the group with recombinant FSH, there is really insufficient LH throughout folliculo genesis while in the group with HP hMG, you have not only some endogenous LH that is present, but on top of that we know that small amounts hCG are added to this preparation to come up to the 75 units of LH activity per ampoule. 

So I think that it’s very clear from the endocrine data of the patients treated in Brussels that there is a compensation for the lack of LH in the HP hMG, while no such thing is true in the FSH group. This could be an explanation for the large difference in the outcome. When I calculated for our own patient group the ongoing pregnancy rate and also the pregnancies occurring after replacement of the cryopreserved embryos, we came to a very large significant difference between both types of treatment. So this, I think, should incite all of us to re-analyse our data because if you try to go into the literature and see how both these preparations were compared, it is very often in a non-homogenous groups, where ICSI was mixed with IVF patients.. I think this needs to be done also for all the preparations that are perhaps out.”

Cees Jansen, MD, PhD: “I think this discussion probably will go on for another couple of years, but I thank you very much for this interview.”

Professor Johan Smitz, MD, PhD: “Thank you.”