Pinar Kodaman, MD - Endometrioma: Timing of surgery vs. IVF
Scientific Symposium
Advancing the Science and Surgery of Endometriosis
Monday and Tuesday, April 19, 2016
The Union Club, New York
Thank you. Thank you for the opportunity to present this talk here today. This is a topic that is very relevant to reproductive endocrinologists on a daily basis. But it is also relevant to those of you who take care of women with endometriosis that are in their reproductive years and desire fertility. I am happy to present the latest on this challenging topic.
I will not spend too much time on background given that this is the second day of this conference. About one third to 45 percent of women with endometriosis have endometriomas and there are several theories just like with endometriosis as to the pathophysiology of endometrioma formation. Usually it involved invagination of the ovarian cortex either due to bleeding from superficial lesions or metaplasia of the coelomic epithelium. There is also the theory of endometriotic transformation of functional cysts. As you know you end up with endometrial glands and stroma that are juxtaposed to normal endometrium and this active tissue bleeds into the cyst with cyclic hormonal changes.
What do endometriomas do to the ovary? The contents of these cysts are toxic in many ways. The free iron from the hemorrhage can diffuse through the cyst wall and impair folliculogenesis. You may recall that iron serves as a catalytic agent for the fenton reaction which produces reactive oxygen species so oxidative stress is increased. As a result we have fibrosis, inflammation. There is also the mechanical disruption of the ovary itself due to the increased pressure from these growing lesions, vascular distortion and resulting ischemia. All this results in altered folliculogenesis. This has been shown in the form of increased granulosis apoptosis, which as a result following ovulation will result in decreased progesterone production.
Endometriomas also affect fertility. We know that women with endometriomas have higher FSH levels their AMH levels, anti-mullerian hormone and antral follicle count levels are lower suggesting impairment of ovarian reserve. In a large study of about 1250 menstrual cycles Maggiore showed the actual rate of ovulation from the ovary with endometrioma is similar to that of the contralateral non-affected ovary, which is interesting. However, we do know that women with endometriosis do have infertility problems at least 40 percent of the time.
The effect of endometriomas on ovarian reserve has become a very hot topic in recent years, certainly with the use of AMH as the most salient marker of ovarian reserve currently. We see in this recent study that the presence of endometriomas, certainly bilateral endometriomas has a significant effect on ovarian reserve as measured by AMH in each age group studied with the exception of those in their later reproductive years where ovarian reserve is impaired on everyone. You will see that as a woman ages there is a natural decline. The white/empty bars represent the controls, the grey the unilateral endometriomas and the black bars the bilateral endometriomas. There is a dose dependent effect in terms of endometrioma presence on AMH levels with advancing age.
What happens to IVF cycles in the setting of endometriomas? We know that women with endometriomas have decreased ovarian responsiveness to gonadotropins. As a result their cycles tend to be impaired and can often be cancelled. There is a similar rate of fertilization _________retrieved cleavage rates are similar and high quality embryo formation also appears to be similar. There appears to be no effective unilateral endometrioma on the number of eggs retrieved. In this study, women with endometriomas greater than 2.5 cm were compared to those with less than 2.5 cm, unfortunately there was no adjustment for age or basal FSH, which is an issue. But what we can see from this status that there is compensation from the unaffected ovary so total egg number may not be impaired in the setting of unilateral endometrioma.
A more elegant study was done more recently by Coccia in Italy where they used women with unilateral endometriomas and compared the response of the affected ovary to the unaffected ovary such that each woman served as her own control. As you can see here in this panel there are decreased numbers of follicles. Certainly follicles greater than 16 mm which are relevant clinically in IVF and also egg number is decreased in the endometriotic ovary compared to the unaffected ovary. It appears that the size of three cm is the threshold for having such a detrimental effect. In fact, in their model that they created which was adjusted for age and FSH and all that every single millimeter increase in endometrioma size caused a decrease in predicted number of oocytes retrieved by about 0.667 which in the world of fertility is relevant.
Of course as seen on previous studies women even with unilateral endometriomas have higher cancellation rates.
This is going to be hard from me to see. In this large meta-analysis of 33 studies that was done recently by Hamdan et al they looked at endometrioma versus no endometrioma and various outcomes of IVF. The writing is a little small but you can see perhaps that on this graph that live birth rate and clinical pregnancy rates are not affected, neither are miscarriage rates. There is an increase in cancellation rate as well as a decrease in the number of oocytes retrieved and then elevation in baseline FSH in women with endometrioma. A total of FSH dose and antral follicle count did not vary in this study. So those are the relevant findings there.
For those of you who do not do IVF on a daily basis what are the challenges of oocyte retrieval with endometriomas? Well, we stimulate ovaries to produce multiple follicles and then those ovaries tend to become heavy and plump and sag into the cul-de--sac where we can then access them transvaginally with our ultrasound probe of which there is a needle that is used then to enter each follicle, aspirate the follicular fluid and the eggs are hopefully in that mix. When you have women with endometriomas this is typically what you see. You see the endometrioma blocking your access to the follicles posterior to it and because these ovaries also tend to be affected by adhesions it is hard to move the ovary around to get to where you need to be to retrieve those eggs. The decreased number of eggs retrieved is not a surprising finding.
What are the risks of egg retrieval in women with endometriomas? As I already showed you in the last slide it is difficult sometimes and you may have to limit the number of eggs retrieved because there are dangers. We try to avoid the endometriomas we do not want them to leak or rupture. There can be an injury to adjacent organs such as the bowel or vessels, certainly bladder and ureter can also be affected in certain cases depending on where the ovary is sitting. There are reports of endometriomas becoming infected. We have all seen this in practice, it is underreported in the literature of course because it is a complication. But you know the bloody content serves as culture medium, it is an excellent culture medium for bacteria and even administered prophylactics antibiotics when we breach the endometrioma this may not prevent infection. When these women get infected they tend to form pelvic abscesses, which require frequent drainage because again antibiotics do not perfuse the cyst very well.
Then there is also the concern of follicular fluid contamination with the endometrioma content. When the follicular fluid from the healthy eggs is mixed with the endometrioma toxic fluid, at least in my studies there appears to be decreased blastocyst development in terms of hatching. And in two out of three human studies that looked at this issue it appears that there are lower pregnancy rates when those eggs are contaminated by that toxic endometrium of fluid. In this latter study there was a 40 percent relative reduction in live birth rate but given the complication does not occur that frequently, or at least is not reported that frequently, the number needed to treat, the number of surgeries that need to be done to remove that endometrioma results in one additional live birth to avoid this complication is actually quite high.
What are the risks of expectant management? Endometriomas can rupture spontaneously, this is reported in the literature. They can ___ if they are not adhered and these things can happen in pregnancy as well. There are several case series of pregnancy complications in the setting of existing endometriomas, which can compromise the pregnancy. It is not that common but when it happens it can be quite dangerous. As other speakers have alluded to there is a risk of ovarian cancer, of leaving that endometrioma in place, approximately a one percent risk. There is a slightly higher risk of atypical endometriosis which may be a precursor lesion to cancers.
Some people worry about the progression of endometriosis in the setting of leaving an endometrioma and proceeding with IVF and pregnancy. That really has not borne out in the studies to date but it is an area of ongoing investigation.
What happens when we remove endometriomas in terms of ovarian reserve? Well, there are two systematic reviews of almost a 1000 women each looking at the effective cystectomy on AMH. It appears that there is a significant decrease in AMH following ovarian cystectomy. Interestingly, and we will get back to this on the next slide, most studies measured, in these meta-analysis, measured AMH levels after only three months post-op, the great majority of them did and not further out. There is almost always inadvertent removal of ovarian cortex. With cyst wall you can see that here. You can see a secondary follicle here removed with the endometrioma cyst wall. In fact, primordial follicles are seen in up to 85 percent of surgical specimens. Then of course other than that iatrogenic removal of healthy ovarian tissue there is injury from inflammation, from cautery that is used and vascular disruption. So all of these things may potentially affect how the ovary functions subsequently.
Interestingly, a recent study by Vignalli showed, as well as a previous one by Sugita, suggested that AMH levels may recover. In fact if you wait long enough, by about a year or so, AMH levels seem to go up and this is similar to what we see in some cases of young women treated with chemotherapy that there is some follicular reorganization and cohort recovery that may result in improved AMH. In Vignali’s study they looked at bilateral and unilateral endometriomas and you can see here in the immediate post-operative state AMH levels go down from the pre-operative. Then with each passing interval you see an improvement in AMH levels suggesting a recovery and by a year there is no difference between the pre-operative and post-operative AMH levels. This is significant. This is an area that needs more research but it is a significant finding. In the world of infertility 12 months is a long time, certainly if a woman is getting older. There is still that consideration but the recovery is certainly reassuring.
I will not spend too much time on this because it has been addressed and will be further addressed in the next talk but surgical treatment of endometrioma the gold standard has been laparoscopic cystectomy. The lowest recurrence rate is seen with that. You see here the cyst wall being extracted from the ovary. Care must be taken. In this case you see the ovary is essentially filleted and there is some limited vascularization that is a result of the surgery. Certainly with careful dissection pregnancy rates are improved following endometrioma resection. Laparoscopic ablation has become favorable in recent years due to a lower effect on AMH levels in the setting of larger endometriomas. People are doing hybrid procedures where you remove most of the cyst and then try not to disrupt the ___ too much where the vessels come in by ablating the base with either cautery or preferably with like a CO2 laser.
Drainage is controversial. It is not recommended but there are some studies of ultrasound drainage of endometriomas which show the expected increase recurrence rate but that increase goes down to only 5.4 percent but after six aspirations, which is quite a few aspirations and carries that inherent risk of infection, so not recommended.
Importance of surgical technique is key. I think this has been repeated over and over again. Surgeon experience is critical. In a study by Canis there was no difference in the number of oocytes and embryos obtained following cystectomy when careful surgical technique was applied compared to women either with just endometriosis or with tubal factor. Identification of that cyst wall plane is critical and the judicious use of bipolar cautery using laser ablation. There is a randomized control trial of suturing which may affect AMH less similarly. Some people have described the use of hemostatic sealants. This was a randomized control study that showed a lesser effect on AMH levels but that is still to be used with caution given the risk of intravasation possible thrombosis with these thrombin products. There have also been reported some SBOs. An interesting technique involves the infiltration of vasopressin into the cyst wall which can help with control of the bleeding and also perhaps helped it develop a plane.
Bilateral endometriomas represent a very interesting clinical challenge. There is a greater decline in AMH levels with bilateral endometriomas reduced responsiveness to gonadotropins compared to unilateral endometriomas. And, very importantly for people who desire to maintain their fertility these women post-surgery tend to have an earlier menopause at 42 years compared to the average age of 51 and increased rate of primary ovarian sufficiency of 36.4 percent in the study compared to the one percent background rate. Bilateral endometriomas have to be approached very carefully.
What are the risks of surgery? The rate of spontaneous ovulation can be lower certainly in the months initially following the procedure. Decreased AMH levels we discussed. Decreased response to ovarian hyper-stimulation and also decreased clinical pregnancy rates and live birth rates with endometriomas.
This was a study of over 2000 women and there was no difference actually in IVF success rate in women with endometriosis compared to those with tubal factor; however, the subset of women with endometrioma had the decreased clinical pregnancy and live birth rates and reviewing the retrospective cohort 78 percent of those women had intervention.
Coming back to the meta-analysis Hamdan et al also looked at endometrioma versus surgically treated endometrioma and again there is no difference in live birth rate, clinical pregnancy rate following IVF. Miscarriage rate is similar. The cancellation rate actually has not changed but you can see in here that the oocytes retrieved are slightly less. Total gonadotropin dose is increased. The antral follicle count which is a measure of ovarian reserve is also decreased.
If we take all the data together from this large recent meta-analysis, patients with intact endometriomas have higher cycle cancellation rates when it comes to IVF but similar clinical pregnancy and live birth rates. Those with surgically treated endometriomas require more gonadotropins but also have similar pregnancy rates and live birth rates. This study is limited, mostly retrospective studies were included and they combined bilateral and unilateral endometriomas. They did no assess size, relationship due to the heterogeneity of the data.
So this picture just summarizes what we need to think of as surgeons who are treating women who desire to maintain fertility in terms of the risks of surgical treatment of endometrioma prior to assisted reproductive technologies versus the risks of leaving that endometrioma intact.
Considerations prior to surgery include the age of the patient, baseline AMH, previous response to ovarian stimulation, the size of the endometriomas, the number of endometriomas and the history of previous surgery. You should think carefully if you are operating on a woman who is 40 with a low MHA and already demonstrating poor ovarian response to stimulation because you may just push her over the edge into early menopause or ovarian failure. Factors supporting surgical intervention include the rapid growth of the endometriomas, suspicious appearance, size greater than four to five centimeters, certainly pelvic pain is a clear indication for surgery as is inability to access normal tissue for retrieval. Also if a patients plans to delay or avoid IVF altogether and she is young there is an increased spontaneous pregnancy rate following endometrioma resection. So that might be a good candidate to intervene on.
ASRM and ESHRE the governing bodies for reproductive endocrinology in the States and Europe are in agreement that surgery is the gold standard in the treatment of women with symptomatic endometriomas and asymptomatic endometriomas that are larger than or equal to four centimeters. However, endometriomas with a mean diameter below four centimeters should not be systematically removed prior to IVF.
Surgery versus IVF: Individualize the treatment approach. I think that is key, every patient is different and has different goals. Informed consent regarding the options and associated risks, whether you decide to pursue expected management or pursue surgery those have to be discussed. Baseline levels of AMH, the appearance, size and number of endometriomas, etc. should especially be taken into consideration. Remember the goal in these patients with endometriosis often is fertility preservation. If they are not having pain you may want to wait.
Sometimes less is more. There are a lot of unanswered questions why does decreased ovarian reserve, fewer oocytes retrieved still result in similar clinical pregnancy and live birth rates? There may be an age related effect. Remember these rates are calculated in a funny way. Cancelled cycles – those cycles that do not result in embryo transfers those are not included in the denominator here. These clinical pregnancies and live birth rates are somewhat inflated. What is optimal interval between surgery and IVF? That as far as I know has not been addressed.
Do AMH levels recover across all ages at 12 months? Or is it just the younger patients that have adequate reserve and the ability to reorganize those follicular cohorts? Obviously we need more prospective studies, specifically RCTs and hopefully that is in the works.
I will take any questions. I leave you with a picture of the fertility goddess in Mayan culture who we visited last week in Cozumel. She greets visitors there. Any questions I would be happy to take. Thank you.
Audience Member: Why the four centimeter cutoff?
Pinar Kodaman, MD: In asymptomatic women or symptomatic?
Audience Member: Asymptomatic.
Pinar Kodaman, MD: In asymptomatic women the four centimeter cutoff has to do with a lot of factors. The risk of atypical endometriosis, the risk of potential malignancies is higher as the cysts get bigger. The risk of complication in terms of retrieval, in terms of potential rupture, torsion, growth, etc. are also increased. I think traditionally that has been the cutoff in assisted reproductive technologies. Again, you have to individualize to the patient. I think the goal is just to not systematically remove those that are less than four centimeters if a patient is completely asymptomatic. Of course patients that are symptomatic may remove even two centimeters endometriomas.
Tamer Seckin, MD: How many failures of IVF you would move on to look inside?
Pinar Kodaman, MD: That is a good question. It depends on so many things. It depends on patient age, AMH, what their IVF coverage is. We are in a mandated state where they have two IVF cycles covered. The overall data does not support necessarily intervening in the asymptomatic patient. But in a young patient with adequate ovarian reserve who has time and coverage, and certainly if she is having pain obviously intervene. But in the patient who is not at risk for complications from intervention in terms of ovarian reserve or having to wait a year to recover her ovarian reserve that is the patient that we might intervene on in the right circumstance. Waiting 12 months for the AMH levels to recover in someone who is 41 or 42 may pretty much negate any chances of them successfully conceiving. Even when their AMH levels recover the quality of those eggs we know from preimplantation genetic screening that the quality of those embryos from those eggs is very, very impaired.
Audience Member: I read all this data, I hear all these lectures and I am surgeon who takes care of endometriosis, which is inflammatory, cytotoxic and damaging to our patients. I have a really hard time leaving that damage inside to continue to grow. The other thing that I think, because the patients come to me after they come to you guys and say, “What should I do?” and part of it is cost because you only have two cycles that are covered, or three cycles sometimes or no cycles are covered for some of our patients because they already did one or two or three. I have a really hard time saying, “You know I am going to do something to you that is going to take away the cytotoxicity that is going to take away your endometriosis that is going to be paid for by insurance that might get you pregnant on your own because I am going to put dye through the tubes and make sure they are open and clean, and the data shows either way you are going to have the same live birth rate or pregnancy rate. I just find as a surgeon that it is really hard for me to tell a patient I am going to leave this disease in you.
Pinar Kodaman, MD: I am completely with you on that. I do a lot of surgery for endometriosis patients and the patient that is not planning on doing IVF that wants to enhance her natural fertility or do something short of IVF it is very reasonable, certainly in a younger patient that has adequate reserve and time on her side. But the data to date just recommends that we take caution in terms of removing all the cytotoxicity, etc. and in the setting of IVF in particular where it may actually make no difference in terms of IVF success rate. If that patient’s ultimate goal is pregnancy and pregnancy tends to be a favorable state for endometriosis it may not be the time necessarily prior to the IVF cycle if that is the plan to pursue a radical procedure that may ultimately do more harm than good. It is very individualized to a specific patient, her age, her AMH, her coverage and what her desires are in terms of fertility and fertility treatments. I think IVF is a special category because a lot of the success of IVF can overcome the – the IVF itself can overcome a lot of the negative effects of endometriosis in certain settings.
Harry Reich, MD: Just an observation because I noticed you had the slide about specimens with – there were oocytes removed with the specimens and then you showed the slide of Canis, which did not seem to back that up. I just want to emphasize to everyone here that Canis’ group have been doing this – this is Brouha’s group – for over 30 years so it does show I think that surgical experience in excising the cyst wall is paramount. Many people are trying to do this work and do not have the experience to remove them, the endometriomas.
Pinar Kodaman, MD: So that comes back to our previous speaker’s point of observing cases and participating in cases, learning surgical experiences is critical here because you really can do more harm than good if there is not careful surgical technique.
Thank you.
