<< . .

( 30)

. . >>

et al., 1998, Mathur et al., 2000).


The development of in vitro fertilization (IVF) by Professor Robert Edwards
and Dr. Patrick Steptoe was the gateway to modern human reproduction
(Steptoe and Edwards, 1978). The impact of IVF on reproductive medicine has
been phenomenal. It opened new horizons in every discipline from cell biology
to genetics. Robert Edwards is a legend of the twentieth century, and it is always
fascinating to see that he had already thought of and debated issues in the 1960s
and 1970s that our profession and society are just discovering (Aboulghar et al.
1998a). In relation to ovarian stimulation, Louise Brown was conceived after
natural-cycle IVF without gonadotrophins. The use of gonadotrophins became
popular in the early 1980s. It is interesting to note that the incidence of OHSS
following IVF in the 1980s (Table II.1) was higher than that following ovulation
induction in the 1970s without the widespread use of estradiol monitoring or
ultrasonography (Table II.2).
Rizk and Smitz (1992) thought this high incidence possibly represents an
increase in aggressiveness in stimulation during the 1980s, and secondarily the
use of long GnRH agonist protocols. Professor Edwards was among the ¬rst
to question the wisdom of aggressive ovarian stimulation and advocated a gentle
approach (Edwards et al., 1996; Fauser et al., 1999). The best example of this
very serious epidemic has been clearly demonstrated by Abramov et al. (1999).
In a multicenter report of OHSS cases from 16 out of 19 tertiary medical
centers in Israel, the authors revealed some shocking ¬ndings. While the
number of severe cases of OHSS following ovulation induction treatments
remained unchanged, the number of cases following IVF increased dramatically
from 2 (0.06% of all IVF cases in 1987) to 41 (0.24% of all IVF cases in
1996)(Figures II.1 and II.2). The total number of IVF cycles performed during

Table II.1 Moderate and severe OHSS in relation to the nature of the gonadotrophin releasing hormone (GnRH) agonist used, the protocol
for human menopausal gonadotrophins (hMG) therapy, the type of luteal support and the occurrence of pregnancies
Reproduced with permission from Rizk and Smitz (1992). Hum Reprod. 7:320À7

Incidence of GnRh agonist % OHSS
Reference Study group OHSS (%) used Dose pregnant HMG regimen Luteal support

Golan et al. (1988) 143 cycles 8.4 D-Trp 6 3.2 mg long-acting 83 started with 3 ampules 2500 IU hCG every 72 h.
117 patients and adjusted to estradiol
Belaisch-Allart and 304 embryo 5.9 D-Trp 6 or NM NM NM 2500 IU hCG (151 patients),
De Mouzon (1989) transfers D-Ser(TBU 6) or placebo (153 patients)
Buvat et al. (1989) 171 embryo 1.8 D-Trp 6 Short-acting NM NM 3 ‚ 1500 IU hCG or 400 mg
transfers (moderate) progesterone oral daily
Herman et al. (1990) 36 embryo 14 D-Trp 6 3.2 mg long-acting 80 started with 2 ampules hCG 2500 IU every third day
transfers and adjusted to estradiol or placebo (18 patients)
Forman et al. (1990) 413 cycles 1.9 D-Trp 6 or 3.75 mg long-acting 88 started with 2 ampules hCG didrogesterone 30 mg/day
(severe) D-Ser (TBU) 6 injection 100 ng/day and adjusted to estradiol orally or 2500 IU hCG
or nasal spray 500 ng/day every 72 h
Smitz et al. (1990) 1673 cycles 0.6 D-Ser (TBU) 6 600 ng daily nasal spray 70 started with 2 ampules hCG 1500 IU every 72 h or
(severe) and adjusted to estradiol progesterone vaginal or
Rizk et al. (1992) 1562 1.3 D-Ser (TBU) 6 subcutaneous injection 57 started with 2 ampules hCG 2000 IU on days 2 and 5
(severe) 200 mg/day or nasal and adjusted to estradiol or progesterone 200 mg/day
spray 500 mg g/day vaginal suppository

NM, not mentioned; hCG, human chorionic gonadotropin


Table II.2 Incidence of OHSS in hMG/hCG cycles
Reproduced with permission from Rizk, B. Ovarian hyperstimulation syndrome.
In (Studd J, Ed.), Progress in Obstetrics and Gynecology, vol. 11.
Churchill-Livingstone, Edinburgh, 1993, Chapter 18, pp. 311À49

Author Year Mild (%) Moderate (%) Severe (%)

Rabau et al. 1967 3.5
Tyler 1968 23 1.5
Taymor 1968 20 2
Thompson and Hansen 1970 1.2
Goldfarb and Rakoff 1973 10 0.005 0.008
Jewelewicz et al. 1973 20 7 1.8
Hammond and Marshall 1973 21 10
Caspi et al. 1974 6 1.17
Lunenfeld and Insler 1974 8.4 0.8
Schwartz et al. 1980 6.3 1.4

Fig. II.1: OHSS following ovulation induction and IVF and the number of IVF cycles over
a decade between January 1987 and December 1996
Reproduced with permission from Abramov et al. (1999). Hum Reprod 14:2181À3

this time also increased from 2890 in 1987 to 17 283 in 1996. The authors
explained this epidemic by the over-utilization of high-dose gonadotrophin
protocols by assisted reproduction units. These units, in their opinion, seemed
to have become more competitive in the last decade, with oocyte and
embryo numbers being considered as the main criteria for success. With
re¬nements in embryo cryopreservation allowing repeat embryo transfers, these
numbers have become more relevant. Expansion of oocyte donation programs,
where high-dose gonadotrophin protocols play a key role in achieving the
maximum number of oocytes to be donated, may have also contributed to this
problem. Finally, GnRH agonist protocols have been blamed in part for the

Fig. II.2: The annual incidence of severe OHSS per 1000 IVF cycles between January 1987
and December 1996
Reproduced with permission from Abramov et al. (1999). Hum Reprod, 14:2181À3


Several studies have attempted to collect and analyze data in order to
characterize the patient population at risk for OHSS (Schenker and Weinstein,
1978; Navot et al., 1988; Golan et al., 1988) and to de¬ne risk factors for
developing this syndrome (Blankstein et al., 1987; Asch et al., 1991; Rizk et al.,
1991a, b; Rizk and Smitz, 1992). More recently, the epidemiology of OHSS has
been studied in large series with the same two objectives by three groups of
investigators from Belgium, Israel and Egypt. The Belgian study was a
retrospective analysis of 128 cases from 13 IVF centers. The series from Israel
was a retrospective analysis of 209 cases of severe OHSS after IVF in 16 out of
19 tertiary medical centers in Israel. The Egyptian series consisted of cases of
moderate and severe OHSS from a single IVF center study of 3500 consecutive
IVF cases (Delvigne et al., 1993a; Delvigne and Rozenberg, 2002; Serour et al.,
1998; Abramov et al., 1998, 1999).

Selection of Patients

It is commonly observed that women suffering from ovarian hyperstimulation
are signi¬cantly younger (Navot et al., 1988). This does not mean that older
women are not at risk for OHSS but it means that younger women are at higher
risk. Delvigne et al. (1993a) in a large Belgian study including 128 cases of
OHSS and 256 controls, observed that the mean age for OHSS patients was 30.2
+3.5 versus 32.0+4.5 years in controls. Enskog et al. (1999), in a prospective
cohort study of 428 patients undergoing controlled ovarian hyperstimulation,

observed that the patients in whom severe, moderate or mild ovarian
hyperstimulation syndrome developed were younger than the patients in
whom OHSS did not develop. The difference in the mean age between the
patients in whom severe OHSS developed and the control group was
approximately 2 years. The difference between all patients in whom OHSS
developed and the control group was somewhat less. It is interesting to note
that the difference in ages is very similar between the Belgian and the Swedish
studies; the Belgian study reporting a difference of 1.8 years versus 2 years in the
Swedish study, and the age factor seems to be a constant ¬nding in most reports
(Delvigne et al., 1993a; Enskog et al., 1999; Rizk and Abdalla, 2006).

Body-Mass Index
Most clinicians have the impression that OHSS is more common in patients
with a lower BMI. Navot et al. (1988) described a positive correlation between
a lean body mass and OHSS, whereas several other investigators could not
con¬rm such a correlation (Lewis et al., 1990; Delvigne et al., 1993a, b; Enskog
et al., 1999).

Etiology of Infertility
OHSS has been observed equally in primary and secondary infertility (Delvigne
et al., 2004). The duration of infertility does not in¬‚uence the occurrence of
OHSS (Navot et al., 1988). Most certainly, women who have previously
developed OHSS are at increased risk (Delvigne et al., 1993a, b). A signi¬cantly
higher incidence of OHSS is reported in group II patients (World Health
Organization classi¬cation; WHO, 1973). Lunenfeld and Insler (1974) found
the incidence of mild and severe OHSS in 621 cycles of patients belonging to
group I to be 5.5% and 0.6% respectively, compared with 10.8% and 1.2% in
784 cycles in group II. Thompson and Hansen (1970) analyzed 3002 hMG/hCG
cycles in 1280 patients and found no cases of hyperstimulation syndrome in
patients with primary amenorrhea (group I). Similar observations have been
made by world-leading investigators (Caspi et al., 1974; Schenker and
Weinstein, 1978; Tulandi et al., 1984).


Polycystic ovarian syndrome (PCOS) is the most common endocrinopathy
affecting 4À12% of women of reproductive age. PCOS is a syndrome of ovarian
dysfunction, along with the cardinal features of hyperandrogenism and
polycystic ovary (PCO) morphology (Rotterdam ESHRE/ASRM Sponsored
PCOS Consensus Workshop Group, 2003). The characteristic appearance of
sclerocystic ovaries at laparoscopy or laparotomy is well known to every
reproductive surgeon (Figure II.3). Clinically, PCOS is characterized by
hyperandrogenism and chronic anovulation. Approximately 40À60% of

Fig. II.3: The gross and microscopic characteristics of polycystic ovaries
Reproduced with permission from Chang (2004). Polycystic ovary syndrome and
hyperandrogenic states. In (Strauss, Barbieri, Eds), Yen and Jaffe™s Reproductive
Endocrinology: Physiology, Pathophysiology and Clinical Management, 5th edition.
Philadelphia: Elsevier, Saunders, Chapter 19, p. 600

women with PCOS are obese and 60% display insulin resistance. The
pathophysiology of PCOS has been extensively evaluated over the last two
decades (Figure II.4). The clinical picture of patients with PCOS exhibits
considerable heterogeneity (Balen et al., 1995). At one end of the spectrum,
the polycystic ovary detected by ultrasound is the only ¬nding. At the other
end of the spectrum, obesity, menstrual cycle disturbance, hyperandrogenism
and infertility may occur either singly or in combination (Balen et al., 1999)
(Table II.3).
Rizk and Smitz (1992) found polycystic ovarian syndrome (PCOS) to be
the major predisposing factor for OHSS. Schenker and Weinstein (1978) found
that 12 out of 25 patients who developed severe OHSS had PCOS as determined
by endoscopy. Charbonnel et al. (1987) encountered ovarian hyperstimulation
in all 33 cycles in PCOS patients. Bider et al. (1989) found a higher proportion
of severe OHSS, 38% in PCOS patients. Smitz et al. (1990) found a hormone
pro¬le suggestive of hyperandrogenism in eight out of ten patients who
developed severe OHSS. Aboulghar et al. (1992) reported that 15 of 18 patients
with severe OHSS had PCOS. Rizk et al. (1991a) found that 13 out of
21 patients with severe OHSS had PCOS con¬rmed by ultrasound and endo-
crine criteria. MacDougall et al. (1992, 1993) showed ultrasonically diagnosed
PCOS in 63% of severe OHSS cases from the Hallam Medical Center.

Fig. II.4: Pathophysiologic concept of polycystic ovary syndrome
Reproduced with permission from Chang RJ (2004). Polycystic ovary syndrome and
hyperandrogenic states. In (Strauss, Barbieri, Eds), Yen and Jaffe™s Reproductive
Endocrinology: Physiology, Pathophysiology and Clinical Management, 5th edition.
Philadelphia: Elsevier, Saunders, Chapter 19, p. 617

The outcome of IVF was compared in 76 patients with polycystic ovaries
diagnosed on ultrasound scan and 76 control patients with normal ovaries,
who were matched for age, cause of infertility and stimulation regimen. Of the
polycystic ovary patients, 10.5% developed moderate/severe OHSS compared
with none of the controls (p ¼ 0.006). Delvigne et al. (1993a, b) reported 37%
of 128 cases had PCOS, compared with 15% among the 256 controls.

Hyperinsulinism and OHSS
Hyperinsulinemia contributes to hyperandrogenism by increasing ovarian
androgen production and suppressing sex-hormone-binding globulin by the
liver, thereby increasing free testosterone levels. In PCOS patients, hyperinsu-
linemia is more profound in obese patients, although the presence of insulin
resistance is independent of body weight (Dunaif et al., 1989; Carmina and
Lobo, 1999; Rizk and Abdalla, 2006). PCOS patients who have hyperinsulinism
were reported to have a higher incidence of OHSS compared with PCOS

Table II.3 Heterogeneity of clinical manifestations of PCOS
Reproduced with permission from Balen et al. In (Brinsden P, Ed.), A Textbook of
In-Vitro Fertilization and Assisted Reproduction. Carnforth, UK: Parthenon
Publishing, 1999, Chapter 8, pp. 109À30

% patients Associated endocrine
Symptoms affected manifestations Possible late sequelae

Obesity 38 elevated androgens diabetes mellitus (11%)
and androstenedione)
Menstrual disturbance 66 elevated LH cardiovascular disease
Hyperandrogenism 48 increased LH:FSH ratio hyperinsulinemia
Infertility 73% anovulatory increased serum estrogens high LDL, low HDL
Asymptomatic 20 elevated fasting insulin endometrial carcinoma
elevated prolactin hypertension
decreased sex hormone
binding globulin

LH, luteinizing hormone; FSH, follicle stimulating hormone; LDL, low-density lipoprotein; HDL,
high-density lipoprotein

patients with normo-insulinism (Fulghesu et al., 1997). A complex interaction
between insulin and follicular maturation has been suggested. Granulosa cells
play a major role in OHSS development, and insulin increases the aromatase
activity of granulosa cells, resulting in a higher ratio of estradiol to
androstenedione (Fulghesu et al., 1997). Higher insulin levels alter the ovarian
response to FSH and enhance the production of antral follicles as observed in
OHSS. The authors hypothesized that hyperinsulinism might play an etiological
role in the development of OHSS in PCOS patients. Delvigne et al. (2002)
studied the metabolic characteristics of women who developed ovarian
hyperstimulation syndrome. The primary purpose of the study was to
investigate whether the higher incidence of hyperinsulinism is found in
women who developed OHSS, whether or not they were PCOS patients. There
were no differences in the distribution of patients with insulin resistance
between the OHSS group and the control group. Insulin resistance was found
in six women, three women in each group. The results are in agreement with
those of Fedorcsak et al. (2001), who found no relation between hyperinsu-
linemia and IVF outcomes or OHSS rates (Fedorcsak et al., 2001).


Interestingly, Enskog et al. (1999) observed an increased prevalence of allergy in
patients who developed OHSS in a study involving 420 patients undergoing
controlled ovarian hyperstimulation during a 6-month period. The authors

hypothesized that differences in the immunologic sensitivity of patients are
a predictor of OHSS. This was based on their observation that the
pathophysiologic changes that occur in the ovary in response to OHSS closely
resemble an overactive in¬‚ammatory response, with the participation of
immunomodulatory cytokines. Therefore, before starting their controlled
ovarian hyperstimulation, all 428 patients were questioned about allergy as a
sign of a hyperreactive immune system, and disposition to infection as a sign of
hyporeactivity. The interesting observation of a signi¬cantly higher incidence of
allergy in severe OHSS may indicate that general immunologic mechanisms
may play a role in the development of an in¬‚ammatory response. No previous
study has reported this association between allergy and OHSS.

The development of gonadotrophins (Figure II.5) is one of the most signi¬cant
advances in the treatment of infertility in the 20th century (Rizk, 1993a, b).
Gonadotrophins have been used worldwide since the 1930s. Animal extracts
from the urine of mares, and later of pigs, were used for 30 years. During the
1950s two extraction processes were pursued in parallel: to obtain human
gonadotrophins from the human cadaver pituitary glands or from the urine
of postmenopausal women. Postmenopausal urine had the advantage of a
relatively high concentration of gonadotrophins resulting from the hypergo-
nadotropic status of postmenopausal women. In the late 1950s, researchers

Fig. II.5: The evolution of gonadotrophins
Reproduced with permission from Edwards RG, Risquez F (Eds) (2003). Modern Assisted
Conception. Cambridge, UK: Reprod Biomed Online, Reproductive Health Care Ltd, p. 93

Fig. II.6: Expression of rFSH in Chinese hamster ovarian cells
Reproduced with permission from Howles (1996). Hum Reprod Update 2:172À91

working on pituitary gonadotrophins were the ¬rst to achieve success in terms
of ovulation and pregnancy (Gemzell et al., 1958). A few years later, human
menopausal gonadotrophin was developed from urine. Urine extraction was
scaled up by the pharmaceutical industry and superseded pituitary extraction,
which continued to be used by only a small number of state agencies in
Australia (Loumaye and Howles, 1999). As stated by Loumaye and Howles,
extensive use of urine-derived gonadotrophin was fortunate because it avoided
the sad story of CreutzfeldtÀJakob disease (CJD) transmitted by pituitary-
derived human growth hormone. In the 1980s, puri¬ed forms were developed
and in the 1990s highly puri¬ed urine derivatives of FSH became available, and,
¬nally, recombinant FSH was developed (Figure II.6). The characteristics of the
different human gonadotrophin preparations (Table II.4) have been summar-
ized by Loumaye and Howles (1999).

Pregnant Mare Serum Gonadotrophins
The evolution of gonadotrophins started in the 1930s when equine serum
gonadotrophin was extracted from pregnant mare serum (Cole and Hart,
1930). The results from using pregnant mare serum gonadotrophins (PMSG)
were inconsistent and disappointing because of the formation of antibodies to
heterologous gonadotrophin (Schenker and Weinstein, 1978). Until 1961,

Table II.4 Characteristics of different human gonadotrophin preparations
Reproduced with permission from Loumaye and Howles. In (Brinsden P, Ed.), A Textbook of
In-Vitro Fertilization and Assisted Reproduction, 1999. Carnforth, UK: Parthenon Publishing,
Chapter 7, p. 104

Year of Protein FSH LH Physico- mass
¬rst Source concentration bioassay bioassay chemical content Route of
Preparation registration of FSH (IU/mg) (IU) (IU) QC control administration

Menotropin 1/1 1964 urine 75 75 none no im
Menotropin 2/1 1989 urine 75 37.5 none no im

<< . .

( 30)

. . >>

Copyright Design by: Sunlight webdesign