Human Reproduction vol.5 no.l pp.36-39, 1990

Treatment of hyperstimulation during in-vitro fertilization

J.Salat-Baroux, S.Alvarez, J.M.Antoine, D.Cornet, Ch.Tibi, M.Piachot1 and J.Mandelbaum1 Department of Obstetrics and Gynaecology, Hopital Tenon, University Paris VI, 4 Rue de la Chine, 75020 Paris and 'Unite' 173, INSERM, Hdpital Necker, Paris, France

In 33 patients treated with a combination of an LHRH agonist (LHRH-A) and gonadotrophin in a long protocol, a biological hyperstimulation occurred (E^> > 2500 pg/ml on the day of HCG administration and 4722 ± 1190 pg/ml the day after, with > 10 follicles > 12 mm on each ovary). The replacement of fresh embryos were deferred and LHRH-A was continued, and an endometrial biopsy was performed on the theoretical day of replacement (2 days after oocyte recovery). With this technique, we obtained a mean number of 17.9 ± 7 oocytes, a fertilization rate of 49% and a replacement rate of 87% in a deferred cycles. The overall pregnancy rate of frozenthawed embryos was 27% in the seven spontaneous cycles, 12 induced cycles and 10 artificial cycles. Only one severe hyperstimulation occurred and this case emphasizes that caution remains necessary even with this technique. Key words: In-vitro fertilization/hyperstimulation/polycystic ovary disease/LHRH agonists/transfer after freezing embryos

Introduction The frequency of hyperstimulation during ovulation induction by human menopausal gonadotrophin (HMG)/human chorionic gonadotrophin (HCG) is 3%, with serious grades of hyperstimulation occurring in 0.84% (Lunenfeld et al., 1986). The ovarian hyperstimulation which is necessary for in-vitro fertilization has raised these levels respectively to 10% for the mild forms and 1.57—4% for the moderate and severe forms (Friedman et al., 1984; Fivnat-France, 1988). This is even more pronounced in women with polycystic ovary disease (PCOD) when stimulation is performed with normal doses (Salat-Baroux et al., 1987). The introduction of luteinizing hormone releasing hormone (LHRH) analogues has been associated with an improvement in the pregnancy rate per cycle (Charbonnel et al., 1986) but has barely modified the frequency of hyperstimulation, especially in cases of PCOD (Salat-Baroux et al., 1988). However, even though some degree of biological ovarian stimulation is necessary in order to obtain the greatest number of oocytes and therefore embryos, this can also lead to serious hyperstimulation after repeated injection of HCG (Salat-Baroux et al., 1987) or after the secretion of HCG originating from one 36

Downloaded from https://academic.oup.com/humrep/article-abstract/5/1/36/743544 by University of Sussex user on 18 July 2018

or several implanted embryos. This worry has led to the cancellation of cycles with an oestradiol level - 3000 pg/ml accompanied by a multifollicular reaction. To avoid losing the benefit of the stimulation, we nevertheless triggered ovulation by an HCG injection, but replaced the embryos after freezing in later cycles. This technique had several advantages, allowing us to aspirate all the follicles after the injection of HCG and to avoid any endogenous HCG secretion arising from an early single or multiple pregnancy. Ovarian secretion could be arrested by the continuation of LHRH analogues and the chances of implantation could be increased by deferring the transfer to a cycle in which the endometrium is not disturbed by excessive 17/3-oestradiol (E2) secretion (Friedman etal., 1984). The occurrence of clinical hyperstimulation could be avoided. Materials and methods A total of 33 patients was treated by a combination of LHRH agonist and gonadotrophin in a long protocol. This consisted either of an i.m. injection of Decapeptyl (Ipsen Beaufour Laboratories, 3.75 mg) at the start of the cycle (down-regulation being achieved in 15-20 days) or with buserelin (Hoeschst Laboratories), 300 ng s.c./day for 15 days. This was followed by 225 IU of human menopausal gonadotrophin (HMG, Neopergonal, Serono Laboratories, France) for 8 days in 29 cases and pure follicle stimulating hormone (FSH, Fertiline, Searle Laboratories, France) in four cases at a dose of 300 IU/day for 3 days then 225 IU for 5 days. One patient was treated with 300 IU for 2 days and 150 IU for 4 days. The patients' mean age was 31.4 ± 4.1 years. Tubal infertility was the indication for IVF in 26 patients and idiopathic infertility in seven. Only one women was known to have PCOD. There was no male subfertility factor in this series. The diagnosis of biological hyperstimulation was recorded firstly, when the level of E2 was >25O0 pg/ml on the day of HCG administration (10 000 IU, Organon Laboratories) and when this level was obtained after 8 days of stimulation. On the day following injection the mean plasma E2 level was 4722 ± 1190 pg/ml. Secondly, when > 10 follicles > 12 mm were visible in each ovary by ultrasound, on the day of HCG. LHRH analogue was continued s.c. in the same doses, associated in the 48 h following oocyte recovery with micronized progesterone (Utrogestan Besins-Iscovesco Laboratories, 300 mg daily by the vaginal route) with the aim of obtaining an endometrial biopsy on the theoretical day of replacement and daily measurements of plasma E2 and progesterone (P). Treatment with the LHRH analogue was continued into the luteal phase until an oestradiol © Oxford University Press

Treatment of hyperstimulation during IVF

level of 1 ng/ml. In the induced cycles, pure FSH (Searle Laboratories, Paris) was used at a dose of 150 IU/day from the third to the seventh day of the cycle to avoid the recurrence of hyperstimulation (Figure 1). Ovulation induction was performed with 5000IU of HCG when there was at least one 16-mm follicle present at ultrasound and the cervical score was 10-12 for at least 2 days. In artificial cycles, the dates for replacement were programmed in advance. DTRP6 3.75 mg (TPSEN Laboratories) was injected 44 days beforehand. Then, while blockade was continued with DTRP6, 0.1 mg s.c. until replacement, substitution therapy was initiated following the same regimen as that used for the replacement of embryos obtained following oocyte donation in women without functional ovaries (Figure 2). Firstly, E5 was given, one dose from the 1st to the 5th day of the cycle, two doses from the 6th to the 25th day then one dose from the 26th to the 28th day by the percutaneous route (oestrogel, Besins Iscovesco Laboratories). Secondly oestradiol valerate, 2 mg/day was given from the 11th to the 13th day of the cycle by the oral route (Progynova, Schering Laboratories). Thirdly, micronized P 100 mg was given from the 13th to the 15th day by the intravaginal route, 200 mg from the 16th to the 25th day and then 100 mg for 2 days. A maximum of three frozen-thawed embryos was replaced on the 14th day of the cycle, i.e. 2 days after the start of P administration.

Results The results are shown in Tables I and II. The mean number of ampoules of HMG was 24 ± 7 and 11 ± 2 of FSH. The mean duration of stimulation was short (9.6 ± 0.9 days) although the mean F^ levels were 3702 ± 885 pg/ml on the day of HCG and increased to very high levels of 4722 ± 1190 pg/ml on the day following HCG. Four women did not have any oocytes which fertilized, even though there was no male factor present. In the other patients, the mean number of oocytes obtained was 17.9 ± 7.0 (range 10-25). Of the 597 oocytes obtained, 246 fertilized (49.5%) and a mean of 7.4 ± 5.4 oocytes fertilized and 6.3 ± 4.7 embryos cleaved per patient. Twenty nine women underwent embryo replacement (87.8% of patients). In four cases, the embryos did not survive freezing and thawing. The pregnancy rates were similar in the three sub-groups. In spontaneous cycles, embryos were transferred in six of seven cycles. One twin pregnancy was obtained in a patient having three embryos replaced, one ongoing pregnancy and one spontaneous

Downloaded from https://academic.oup.com/humrep/article-abstract/5/1/36/743544 by University of Sussex user on 18 July 2018

rei

iso

ID/HI;

FSH or WC ub*o n a c M u r r according Co 0 . 5 and O r v l u l KDCU* SCOT*

-

KlcroolMd P r o f M t t r o o * 300 X(T/ Daj Orally

O n l A t l o n by bCC 9000 ID DO OvuUtion D»2

laplaca D«*

Fig. 1. Diagram of treatment in induced cycles.

unr* i.n •sr/m at vwrlanla data IB tna cycla 30 daya

mica 1 •»

1 MIUTI Dl1 Dl

• •

D5

7

D13

1 IMft•n't/

by

1 DM

4

mi

EJT1ADI0L V1LEUMATE 2 Bar/Day

100 aap-/day treat Qiy 11 to Day 11 by tha vaginal ronta

Cat a flaad day In advanta In tha cytla)

Fig. 2. Diagram of treatment in induced cycles.

Table I. Characteristics of hyperstimulation cycles Cycles with frozen embryos (" = 33) Number amps HMG (m ± SD) Number amps FSH (m ± SD) Mean duration of stimulation Ej D-2" (pg/ml) E, D-l (pg/ml) P D-2 (pg/ml) P D-l (pg/ml) LH D-2 (pg/ml) Number of oocytes collected/patient Number of fertilized oocytes Number of embryos/patient Mean length of luteal phase Pregnancies

24 ± 7 11 ± 2 9.6 ± 0.9 3702 ± 885 4722 ± 1190 0.70 ± 0.3 7.4 ± 3.3 5.5 ± 5.0 597 17.9 ± 7 (rate of fertilization 49%) 246 7.4 ± 5.4 215 6.3 ± 4.7 8 24%

*D-2: 2 days before ovulation (oocyte aspiration).

abortion also occurred in two patients having two embryos replaced. In the other three patients, the numbers of embryos replaced were two, three and four. In the 12 cycles stimulated with pure FSH, 11 women underwent embryo transfer and there were two twin pregnancies and one ongoing pregnancy following the transfer of three, two and three embryos respectively. In the other eight patients, between one and four embryos were replaced but pregnancies did not result. 37

J.Salat-Baroux et al.

TaWe D. Pregnancy rate after replacement of frozen and thawed embryos according to the type of treatment Cycles

Number of embryos (E) transferred

Pregnancies

Spontaneous n = 7 patients 3 pregnancies

1 pregnancy after replacing 1 pregnancy after replacing 1 pregnancy after replacing

3 E Twin pregnancy 2 E Miscarriage 2 E (Singleton)

Stimulated (FSH) n = 12 patients 3 pregnancies

1 pregnancy after replacing 1 pregnancy after replacing 1 pregnancy after replacing

3 E (Singleton) 2 E Twin pregnancy 3 E Twin pregnancy

Artificial n = 10 patients 2 pregnancies

1 pregnancy after replacing 1 pregnancy after replacing

3 E (Singleton) 2 E (Singleton)

Table ID. Endometnal dating according to Noye's criteria and levels of steroids in serum 2 days after oocytc retrieval Endometrium in phase

Pg/mJ P ng/ml

Advanced endometrium

n = 3

D + 4 n = 17

D + 6 n = 5

1626 ± 699.6

1422 ± 401 .0

1441 ± 922 0

NS

97 ± 44. 3

142.0 ± 67. 1

NS

85.3 ± 17.2

*D is the correct day according to Noye's criteria

In the 10 patients undergoing an artificial cycle, there were eight replacements of which two gave rise to ongoing pregnancies after the transfer of three and two embryos respectively. In the other six patients, between one and three embryos were replaced, although one patient had a two-pronuclear zygote transferred. One case of serious clinical hyperstimulation occurred despite follicular aspiration. In this patient, after treatment with buserelin (long protocol) associated with HMG (3 ampoules/day for 9 days) the oestradiol level at the time of HCG was 4800 pg/ml. Ultrasonographic oocyte retrieval enabled 27 oocytes to be obtained and 14 embryos were frozen. Administration of LHRH analogue was continued by itself and 10 days later, the patient was admitted to hospital with a clinical severe hyperstimulation. These clinical and biological parameters gradually improved with intensive medical treatment. Three embryos were replaced later in an artificial cycle and a pregnancy occurred. It is interesting to note the histological results of the endometrial biopsy performed 2 days after ovulation and the correlation to E2 and P concentrations in the luteal phase (Table ITI). Of the 25 samples obtained, the endometrium was in phase in three cases, 3—4 days in advance in 17 cases and 6 days in advance in five cases. Furthermore, there was a glandular-stromal asynchrony in 12 of the 25 cases which was not related to dating of the endometrium. Comparing the E2 and P levels on the day of the biopsy to the endometrial status, there was no significant difference in the endometria regardless of whether they were in phase or out of phase.

38

Downloaded from https://academic.oup.com/humrep/article-abstract/5/1/36/743544 by University of Sussex user on 18 July 2018

Discussion There is undoubtedly a correlation both between the E2 levels on the day of HCG and the number of embryos obtained and also the number of embryos replaced and the pregnancy rate (Seppala , 1985). However, it seems that elevated E2 levels in the preovulatory phase can have an adverse influence on the pregnancy rate after embryo replacement (Forman et al., 1988) or may even be detrimental to oocyte quality and ability to be fertilized. When the E2 level exceeds 4000 pg/ml after the HCG injection, the risks of hyperstimulation become particularly serious, especially when associated with a multi-follicular reaction (Hanning et al., 1983). The most cautious attitude and that which is usually adopted is to avoid the injection of HCG in these cases and cancel the cycles. Apart from the fact that this therapeutic attitude does not definitely prevent stimulation, as spontaneous ovulation can occur, it also loses the benefits of a great many days of stimulation, especially when LHRH analogues have been used. Aspirating several follicles after HCG does, according to certain authors, reduce the risk of hyperstimulation (Hazout and Belaisch-Allart, 1986) but this reduction is relatively small, especially if a multiple pregnancy occurs following the procedure. Deferred transfer in which the LHRH analogues are continued and no HCG injections are given in the luteal phase, seems to us the most pragmatic attitude. Nevertheless, we have experienced clinical hyperstimulation type III in 33 cycles. There were several features to note about this particular case; the patient did not have daily LHRH analogue injections regularly after the injection of HCG, there was a lag time of 10 days between induction of ovulation with HCG and the appearance of the hyperstimulation syndrome. It may be that other factors are implicated in this incident, e.g. the renin agiotensin system (Friedman etai, 1984; Navot etal., 1987). To be more scientifically rigorous it would have been necessary to compare this series with another randomly selected group who had the same criteria but in whom fresh embryo transfer would have occurred in the same cycles. However, it seems to us that this study would be unethical. The final point to discuss is that of the incidence of pregnancies. It is obviously relevant to question the state of the endometrium in cases of hyperstimulation and its capacity to support nidation. Forman et al. (1988) have shown that excessive E2 levels in the preovulatory phase have a deleterious effect either on embryo quality or on the state of the endometrium. Several authors have reported a deterioration of the endometrium after ovarian stimulation. Garcia et al. (1984) have even emphasized that there is an advance of 1 day or more in the histological appearance of the endometrium as recorded using Noyes' criteria (1950). In our series, we recorded an advance of > 2 days in >50% of cycles and this was irrespective of the E2 and P levels noted on the day of biopsy. This observation leads us to have certain reservations about endometrial quality in cases of hyperstimulation and for this reason we prefer deferred replacement. This attitude is even more justified by the pregnancy rate obtained after embryo freezing and thawing which reaches 27% per replacement. The hyperstimulation does not seem to affect either the condition of the oocytes, which fertilize at the normal

Treatment of hypersttaulatioii during IVF

rate, or the state of the embryos which are perfectly well able to resist the freezing-thawing procedures and lead to a very satisfactory pregnancy rate. Furthermore, as we have previously shown (Mandelbaum et al., 1988), the pregnancy rate does not significantly differ in transfers performed in spontaneous, stimulated or artificial cycles. In summary, hyperstimulation is a frequent complication using LHRH agonists and a possible solution to avoid severe clinical hyperstimulation syndromes while conserving a high pregnancy potential is to defer embryo transfer by freezing the embryos. This was associated with a 27% pregnancy rate for embryo replacement. Nevertheless, the occurrence of one serious hyperstimulation in 33 cases emphasizes that caution is necessary and preventative measures should also be adopted.

Seppala.M. (1985) The world collaborative report on in vitro fertilization and embryo replacement. Current state of the art in January 1984. NYAcad. Sci., pp. 442-558. Received on May 23, 1989; accepted on September 25, 1989

References Charbonnel.B., Bavriere,P., Paillard.B. and Lopes.P. (1986) L'association d'un analogue de la LRRH aux gonadotrophines ameliore le recrutement folliculaire et lc taux de grossesses (Fecondation In Vitro). Ann. EndocrinoL, 47, 262. Fivnat-France.P. (1988) Giacomini-Hyperstimulation, Etude multicentrique-Journees Periconceptologie. Avril 88, Montpellier. Forman,R., Fries.N., Testart.J., Bellaisch-Allart.J., Hazout.A. and Frydman.E. (1988) Evidence for an adverse effect of elevated serum estradiol concentrations on embryo implantation. Fertil. Steril., 49, 118-122. Friedman.C.I., Schmidt,G.E., Chang.F.E. and Kim.M.H. (1984) Severe hyperstimulation following follicular aspiration. Am. J. Obstet. Gynecoi, ISO, 436-437. Garcia,J.E., Acosta.A., HsinJ.G. and Jones,H.W. (1984) Advanced endometrial maturation after ovulation induction with human menopausal gonadotrophin/human chorionic gonadotropin for in vitro fertilization. Fertil. Steril., 41, 31-36. Harming,R.J., Austin,Ch., Carlson,I., Kuzime,D., Shapiro,S. and Zweibel.W. (1983) Plasma estradiol and urinary estriol glucuronide as prediction of ovarian stimulation during induction of ovulation with metrophins. Fertil. Steril, 40, 31-36. Hazout.A. and Belaisch-Allart,J. (1986) Presentation des complications de l'induction de l'ovulation par la reduction folliculaire et la reduction embryonnaire—induction et stimulation de Povulation. PARIS Doin Ed., pp. 155-162. Lunenfeld,B., Bloom.S. and BlanksteinJ. (1986) Resultats de l'induction de l'ovulation par les menotrophines chrorioniques (HMG—HCG). Induction et stimulation de l'ovulation. Paris Doin Ed., pp. 95-104. MandelbaumJ., Junca.A.M., Plachot.M., Alnot.M.O., Salat-BarouxJ., Alvarez.S., Tibi.Ch., CohenJ., Debache,C. and Tesquier,L. (1988) Cryopreservation of human embryons and oocytes. Hum. Reprod., 3, 117-119. Navot.D., Margalioth.E., Laufer,N., Birkenfeld.A., Relou.A., Rosler.A. and SchenkerJ. (1987) Direct correlation between plasma renin activity and severity of the ovarian hyperstimulation syndrome. Fertil. Steril., 48, 57-61. Noyes.R.W., Hertig.A. and Roch.S. (1950) Dating the pregnancy biopsy. Fertil. Steril., 1, 3 - 7 . Salat-BarouxJ., Comet,D., Antoine.J.M., Alvarez,S., Alfieri.D. and Bonnardot,J.P. (1987) Un cas de stimulation grave au cours d'une fecondation in vitro suivie de grossesse. Gynicologie, (38) 2, 113-116. Salat-Baroux,J., Alvarez.S., Antoine.J.M., Tibi.Ch., Cornet,D., MandelbaumJ., Plachot.M. and Junca.A.M. (1988) Comparison between long and short protocoles of LHRH agonist in the treatment of polycystic ovary disease by in-vitro fertilization. Hum. Reprod., 3, 535-539.

39 Downloaded from https://academic.oup.com/humrep/article-abstract/5/1/36/743544 by University of Sussex user on 18 July 2018