<< . .

( 30)

. . >>

antidiuretic hormone were observed in patients with hemoconcentration. The
authors suggested that, in addition to increased capillary permeability, severe
OHSS is associated with arterial vasodilatation. The argument is that if
circulatory dysfunction occurred solely as a result of extravascular ¬‚uid shift,
then we would expect that the contraction of the circulating blood volume
would result in a reduction in the cardiac output, as well as an increase in
peripheral vascular resistance as well as atrial natriuretic peptide. The actual
¬ndings were increased cardiac output and atrial natriuretic, and markedly
reduced peripheral vascular resistance. The authors concluded that these
¬ndings indicate a marked peripheral arteriolar vasodilataion.


The hemodynamic relationship between hematocrit and plasma volume was
nicely illustrated by Van Beaumont in 1872. It is accepted that, in the face of
a constant red cell volume, a rising hematocrit signi¬es a fall in plasma volume.
It appears that, when the red cell volume remains constant, the change in
hematocrit can never be numerically commensurate with the change in plasma
volume. The change in plasma volume must always be larger than the change
re¬‚ected by the hematocrit. Thus a change of two percentage points in the
hematocrit from 45% to 47% is four times smaller than the actual 8% drop in

plasma volume. It is extremely important to keep this in mind when treating
patients with OHSS. Any increase in the hematocrit as it approaches 45% does
not accurately re¬‚ect the magnitude of plasma volume depletion, and thus the
seriousness of the patient™s condition. Likewise, in the face of hemoconcentra-
tion, small drops in the hematocrit may represent signi¬cant improvements in
plasma volume (Rizk, 2001, 2002).
Evbuomwan et al. (2000) performed a prospective longitudinal study on
women undergoing ovulation induction to analyze the osmolality and
hematocrit prior to the onset of clinical symptoms of OHSS. In OHSS patients,
a 20% decrease of blood volume occurred between days two and four after
administration of hCG. This was followed by a sustained 30% increase above
baseline from day 8 to day 12 after administration of hCG. These blood volume
alterations were not seen in patients without OHSS (p < 0.006). In OHSS
patients an unexpected increase of 6 mOsm/kg in osmolality was observed
during the later stages of follicular growth two days before hCG administration.
Thereafter, the osmolality decreased from day 2 to day 12 after hCG
administration. Decreased osmolality in severe OHSS is maintained despite
signi¬cant increases and decreases in blood volume. In control patients, the
osmolality decreased gradually from the beginning of gonadotropin injections
until two days after injecting hCG and started to recover from the fourth day
after hCG. This prospective longitudinal study is the ¬rst to track changes in
osmolality and blood volume during spuperovulation, and in severe OHSS
from onset to resolution. Previous studies have investigated women who
already had the fully developed syndrome (Balasch et al., 1994) or who were
pregnant (Haning et al., 1985). Evbuomwan et al. (2000) demonstrated that
blood volume does not change signi¬cantly with superovulation alone, and that
alterations in osmolality are observed in women even if they do not develop
OHSS. The paradox of hypoosmolality with hypovolemia demonstrated during
severe OHSS is suggestive of osmoregulatory adjustments. Furthermore, the
unexpected but signi¬cant decrease in osmolality between two days before
and the day of hCG administration in patients who developed severe
OHSS demonstrates that signi¬cant changes are apparent even before hCG
Evbuomwan et al. (2000) therefore postulated that the signi¬cant decrease
in serum osmolality by day two after hCG may be achieved by a resetting of the
threshold for arginine vasopressin secretion to lower serum osmolality values,
as they have demonstrated in their second study (Evbuomwan et al., 2001).
There is a precedent for the resetting of the threshold for arginine vasopressin
secretion in pregnant women (Davison et al., 1981, 1984). The authors
investigated the hypothesis that the decrease in and the maintenance of a new
steady state in plasma osmolality and sodium level in OHSS are due to the
altered osmoregulation of arginine vasopressin secretion and thirst. They found
that the osmotic thresholds for arginine vasopressin secretion and thirst are
reset to lower plasma osmolality during superovulation for IVF-ET. This new
lower body tonicity is maintained until at least day 10 after hCG in OHSS. The
decrease in plasma osmolality and plasma sodium levels in OHSS are due to

altered osmoregulation rather than electrolyte losses; correction of apparent
˜˜electrolyte imbalance™™ in OHSS is therefore inappropriate.


Levin et al. (2004) conducted a study to evaluate the erythrocyte aggregation
in OHSS. The degree of erythrocyte aggregation is enhanced in the peripheral
venous blood of patients with both COH and OHSS. This ¬nding, known to
cause capillary leak, may contribute to the pathophysiology of the OHSS.


The major event in the pathophysiology appears to be an increase in capillary
permeability. Hypoalbuminemia occurs because of leakage of albumin into
the third space, which is a well-established feature of OHSS (Schenker and
Weinstein, 1978; Polishuk and Schenker, 1969). The status of other plasma
proteins and speci¬c immunoglobulins has been studied by Abramov et al.
(1999) in 10 patients with severe OHSS after induction of ovulation for IVF.
Signi¬cantly lower levels of gamma-globulins, speci¬cally IgG and IgA, were
detected in the plasma of patients with severe OHSS, whereas alpha-and beta-
globulin levels, as well as IgM levels, were not signi¬cantly different from those
in controls (Abramov et al., 1999). Both IgG and IgA levels increased as the
patients improved. Ascitic ¬‚uid contained high IgG, moderate IgA and
negligible IgM levels. Severe OHSS is therefore characterized by hypogamma-
globulinemia attributable to leakage of medium-molecular-weight immuno-
globulins, such as IgG and IgA, into the peritoneal cavity.


In the absence of ovarian stimulation, the volume of normally present
peritoneal ¬‚uid was found to be directly related to cyclical ovarian activity;
it was consistently low in postmenopausal women and women on oral
contraceptives. In normal ovulatory women, the volume of peritoneal ¬‚uid was
diminished in the early proliferative phase and increased until the time of
ovulation (Donnez et al., 1982). Following ovulation, there was a sudden
increase in the volume of peritoneal ¬‚uid, which lingered throughout the luteal
phase and diminished at the commencement of menses (Maathuis et al., 1978).
Since this peritoneal ¬‚uid production was not dependent on the patency or
presence of the fallopian tubes or uterus, its origin was felt to be ovarian or
Yarali et al. (1993) assessed the direct ovarian contribution to ascites
formation in OHSS in the rabbit by the microsurgical isolation of the ovaries
from the peritoneal cavity. Despite extraperitonealization of both ovaries,

ascites occurred in all animals. This was considered to be evidence against
a direct ovarian contribution to ascites formation. They postulated that the
development of ascites was caused by a substance that increased capillary
permeability of the peritoneum and the omentum, and possibly the pleura.


Extensive research has focused on the pathogenesis of increased capillary
permeability in OHSS (Rizk, 1992, 1993a, b; El-Chalal and Schenker, 1997;
Kaiser, 2003).

The association between severe OHSS and high estradiol levels is well
established (Rizk et al., 1991). Asch et al. (1991) reported an overall incidence
of 1% for severe OHSS, but when the estradiol concentration on the day of
hCG administration was more than 22 000 pmol/1, the incidence of severe
OHSS rose to 38%. Estrogens were shown to induce increased capillary
permeability of the uterine and ovarian circulation. Because levels of estrogen
are greatly increased in OHSS patients, it was postulated that the increase in its
production and levels causes the increase in capillary permeability.
On the other hand, large doses of estrogens could not reproduce OHSS
in the rabbit model. Furthermore, Pellicer et al. (1991) reported moderate to
severe OHSS during pregnancy in a woman with partial 17, 20-desmolase
de¬ciency and very low serum estradiol concentration. Levy et al. (1996)
presented a case report of a woman with hypogonadotropic hypogonadism
who developed severe OHSS during ovulation induction, with urinary FSH and
hCG in the presence of low circulating estradiol concentrations. Therefore, the
role of estrogen as a mediator of increased capillary permeability is seriously

It is well established that the use of progesterone rather than hCG in the luteal
phase decreases the occurrence of late-onset OHSS (Rizk and Nawar, 2004; Rizk
and Aboulghar, 2005). However, the role of progesterone in OHSS has been
unclear, because of evidence of its ability to induce Vascular Endothelial
Growth Factor (VEGF) and hence, vascular permeability (Novella-Maestre
et al., 2005). Ohba et al. (2003) demonstrated that, in the rat model, hCG elicits
VEGF production, whereas the potent progesterone synthetic antagonist
decreases VEGF production in a dose-dependent fashion. The VEGF gene
expression was stable. Their studies suggested that progesterone is implicated
in part in the development of OHSS to enhance ovarian VEGF production by
post-transcriptional and organ-speci¬c control.

Novella-Maestre et al. (2005) have recently studied the role of progesterone
on the VEGF expression in the ovary and vascular permeability in OHSS, using
the rat model with dopamine agonists with or without progesterone. To
induce clinical manifestations of OHSS, mature female Wistar rats (22 days
old), were injected with 10 IU Pregnant Mare Serum Gonadotrophin (PMSG)
on days 22 to 25 and with hCG 30 IU on day 26. This ovarian stimulation
protocol has been shown by the authors to result in signi¬cant ovarian
enlargement (ten-fold), ascites and increased vascular permeability, with
maximum levels 48 h after the hCG administration (day 28). In the ¬rst set of
investigations, Novella-Maestre et al. (2005) blocked progesterone synthesis
by administering dopamine agonist on the day of hCG injection. The rats were
treated with:

(1) Dopamine agonist, bromocriptine 100 or 600 mg
(2) Cabergoline, 3 or 10 mg
(3) Untreated (placebo group).

In the second set of investigations the previous study was repeated but the
dopamine agonist treated rats were supplemented with progesterone pellets at
a dose of 400 mg 24 h after hCG administration. In the ¬rst set of experiments,
signi¬cant inhibition of vascular permeability, prolactin and progesterone
synthesis was observed in the dopamine-agonist groups compared to the
untreated rats. However, VEGF levels were unchanged. In the second set of
experiments, progesterone administration recovered progesterone levels in the
dopamine-agonist treated rats but was unable to recover vascular permeability
or VEGF expression. Novella-Maestre et al. (2005) concluded that progesterone
neither affects ovarian VEGF expression nor vascular permeability. It is
postulated that the inhibitory effects of vascular permeability is due to
apoptosis, or the post-transcriptional mechanism, and, furthermore, exogenous
progesterone supplementation was unable to recover ovarian function once
prolactin was deprived in this animal model.

The Ovarian ReninÀAngiotensin System
The role of the ovarian reninÀangiotensin system (RAS) in OHSS has been
extensively investigated (Rizk, 1993a; Rizk and Abdalla, 2006). In humans, 90%
of circulating renin is prorenin. The major source of plasma renin and prorenin
is the kidney (Peach, 1977; Hsueh et al., 1983). However, the ovary is also
a source of circulating prorenin, as has been reported in a bilaterally
nephrectomized woman (Blankestijn et al., 1990). Prorenin is synthesized
without conversion to renin in the monkey and the human ovarian theca cells
and corpus luteum (Paulson et al., 1989). Luteinizing hormone (LH) and hCG
switch on the renin gene expression (Itskovits et al., 1992; Lightman et al.,
1987). In a spontaneous menstrual cycle, the peak of plasma renin expression
occurs in response to the LH surge. In early pregnancy, there is an increase in
plasma renin that correlates with the rise of hCG (Derkx et al., 1986).

Does the Ovarian ReninÀAngiotensin System Play a Role in OHSS?
The role of the ovarian reninÀangiotensin system in the pathogenesis of OHSS
has been investigated (Navot et al., 1987; Pepperell et al., 1993; Morris et al.,
1995; Aboulghar et al., 1996; Delbaere et al., 1997; Rizk et al., 1997). The
angiogenic properties of human follicular ¬‚uid demonstrated by Frederick et al.
(1984) in addition to the high levels of prorenin (Glorioso et al., 1986; Itskovitz
and Sealey, 1987) renin-like activity (Itskovitz and Sealey, 1987; Fernandez
et al., 1985a), angiotensin II (Culler et al., 1986) in the follicular ¬‚uid, as
compared with the plasma, is the cornerstone of the hypothesis that the ovarian
reninÀangiotensin system is central to the pathogenesis of OHSS.
Fernandez et al. (1985a) demonstrated preovulatory follicular ¬‚uid levels
of prorenin up to 12 times higher than those of plasma prorenin after
gonadotrophin stimulation. The magnitude of the mid-cycle rise of prorenin in
response to hCG is related to the number of ovarian follicles. Fernandez et al.
(1985b), in a study investigating the development of new vessel formation in
the New Zealand white rabbit cornea, concluded that angiotensin II not only
facilitates the activation of pre-existing collateral vascular pathways but also has
angiogenic properties, and could therefore play an active role not only in the
fast but also in the slow phase of collateral revascularization characterized by
formation of new vessels. Navot et al. (1987) studied plasma renin activity and
aldosterone in patients with ovarian hyperstimulation. A direct correlation
between plasma renin activity and the severity of OHSS was established.
Elevated plasma aldosterone levels were observed in OHSS cycles, especially in
conceptual cycles with OHSS. Pronounced elevations in plasma renin activity
and plasma aldosterone concentrations were reported in patients with OHSS by
Ong et al. (1991) despite signi¬cant therapeutic plasma volume expansion.
The investigations performed on the ascites collected from patients that
developed OHSS rekindled interest in the concept of a signi¬cant stimulation
of the ovarian reninÀangiotensin system in OHSS (Pride et al., 1990; Rizk,
1992, 1993a). Rosenberg et al. (1994) observed a very high renin concentration
in the ascites of severe OHSS compared to control ascites. Prorenin was the
measured form identi¬ed. Delbaere et al. (1994) demonstrated levels of
angiotensin II-immunoreactive 100 times higher in the ascites of severe OHSS
patients compared with control ascites, and 6.9 times higher than in the plasma
during OHSS. An ovarian origin of angiotensin II in the ascites was therefore
suggested by Delbaere et al. (1994). The absence of a parallel concentration
gradient between plasma and ascites for renin activity and angiotensin
II-immunoreactive during severe OHSS prompted these investigators to evalu-
ate more accurately the active as well as inactive levels of renin, together with
their respective plasma-to-ascites ratio in the syndrome. Delbaere et al. (1997)
measured total renin, active renin, prorenin, and aldosterone in the plasma and
ascites of nine patients who developed severe OHSS. Total renin and prorenin
concentrations were signi¬cantly higher in the ascites than in the plasma. The
concentration gradient between the plasma and the ascites supports the hypo-
thesis of an ovarian origin in the ascites, and to a large extent in the plasma also.

It is, however, likely that the high plasma renin and active renin activity re¬‚ect a
peripheral activation of the reninÀangiotensin system. Delbaere et al. (1997)
concluded that their ¬ndings are consistent with a marked stimulation of both
the ovarian and the renal reninÀangiotensin system during OHSS.

Studies of Angiotensin-converting Enzyme Inhibitors and
Angiotensin Receptor Blockers
Data on the effect of angiotensin-converting enzyme (ACE) inhibitors on
OHSS in a rabbit model are con¬‚icting (Rizk, 2001, 2002). Sahin et al. (1997)
investigated the possible effects of the ACE inhibitor, cilazapril and the
angiotensin II antagonist, saralasin on ovulation, ovarian steroidogenesis and
ascites formation in OHSS in a rabbit model. They concluded that the ACE
inhibitor cilazapril and the angiotensin II antagonist saralasin did not prevent
ascites formation in OHSS. The ovarian reninÀangiotensin system may not be
the only factor acting in ascites formation in OHSS. Morris et al. (1995)
conducted an experiment to determine whether the use of the ACE inhibitor
enalapril would prevent the occurrence of OHSS in a rabbit model. In contrast
to the work done by Sahin et al. (1997) ACE inhibition resulted in a 40%
decrease in the occurrence of OHSS in the rabbit model (Morris et al., 1995).
Teruel et al. (2002) studied the hemodynamic state in 16 hyperstimulated
New Zealand rabbits, and investigated the role of angiotensin II in the
pathophysiology of OHSS. Angiotensin-converting enzyme inhibition decreases
the incidence of OHSS in a rabbit model by 30%, suggesting that angiotensin II
may plan a role in the formation of ascites. These authors also studied the effect
of an angiotensin-converting enzyme inhibitor on renal function in OHSS in
rabbits. They found that angiotensin II may play a signi¬cant role in this
phenomenon, since angiotensin-converting enzyme inhibition normalized the
pressureÀnatriuresis relationship (Teruel et al., 2001).
Ando et al. (2003) studied the ef¬cacy of combined oral administration of
angiotensin-converting enzyme inhibitor and angiotensin II receptor blocker in
the prevention of early OHSS in IVF patients at very high risk of this syndrome.
Four women who had estradiol concentrations of ¸8000 pg/ml were treated
with a combination of the ACE inhibitor alacepril and the angiotensin II
receptor blocker candesartan cilexetil for eight days, starting the day after
oocyte retrieval. All embryos were cryopreserved and transfer postponed.
Despite the extremely large ovaries, no ascites accumulated, and hematocrit
and serum albumin remained normal. The authors concluded that dual
reninÀangiotensin blockage therapy may be useful in prevention of early
OHSS. Further prospective randomized studies should be encouraged. It is
tempting to speculate that ACE inhibitors may be useful in the treatment of
OHSS in humans. Since severe OHSS commonly occurs with pregnancy,
possible fetal effects are important (Rizk 1993a). ACE inhibitors may alter
steroid synthesis within the ovary and inhibit ovulation (Pellicer et al., 1988),
resulting in retention of oocytes and follicular ¬‚uid, leading to larger ovaries.

Prostaglandins have been investigated as possible mediators by Schenker and
Polishuk (1976) and prostaglandin synthetase inhibitors have been used to
prevent the ¬‚uid shift responsible for the manifestations of OHSS. However,
Pride et al. (1986) found that indomethacin, in pharmacological doses, did not
in¬‚uence the clinical features of OHSS (ovarian weight and ascites formation).
Katz et al. (1984) found indomethacin to be useful in the prevention of ascites
associated with OHSS. However, the same group later demonstrated that that
was not the case (Borenstein et al., 1989).
Therefore, the rationale for treatment of OHSS with nonsteroidal anti-
in¬‚ammatory drugs should be seriously questioned (Rizk, 1992, 1993a; Rizk
and Aboulghar, 1999; Rizk and Nawar, 2004). Furthermore, Balasch et al.
(1990) suggested that renal prostaglandin PGE2 and PGI2, by antagonizing the
renal vasoconstrictor effect of angiotensin II and norepinephrine (noradre-
naline), play a major role in the maintenance of renal function in severe OHSS.
They reported a case of prerenal failure in a patient treated with prostaglandin
synthetase inhibitors.

von Willebrand Factor
The von Willebrand factor (vWF) is a large adhesive plasma glycoprotein
produced mainly by vascular endothelial cells and released mainly as multimers
(Handin and Wagner, 1989). The largest vWF multimers are stored in the
endothelial cells in the organelles called weibel-palad bodies. The endothelium
deposits vWF into the basement membrane of blood vessels (Meyer et al.,
1991). von Willebrand factor is a marker of activation of endothelial cells. Its
levels are diminished in von Willebrand syndrome and increased in clinical
conditions characterized by endothelial cell dysfunction, such as pre-eclampsia
in pregnancy and thrombocytopenic purpura. Plasma levels of vWF are raised
by desmopressin due to a selective effect on endothelial permeability. This effect
is used in the treatment of patients with von Willebrand syndrome (Mannucci
et al., 1981).
Todorow et al. (1993) were the ¬rst to demonstrate elevated vWF in
patients with severe OHSS, which subsided when the clinical syndrome
improved. In a retrospective study, Ogawa et al. (2001) found that a rise of the
serum level of vWF occurs before clinical manifestation of the severe form of
OHSS, but not in patients with mild OHSS.

Endothelin-1 is a vasoconstrictor that increases vascular permeability that was
observed to be 100- to 300-fold higher in follicular ¬‚uid than in plasma.
In OHSS patients, serum endothelin-1 is elevated, but in parallel with other
neurohormonal vasoactive substances and without correlation with the severity

of OHSS, suggesting a homeostatic response rather than an initiating role in
the development of the syndrome (Balasch et al., 1994).

Discovery and Cloning
Rizk et al. (1997) extensively reviewed the role of vascular endothelial growth
factor (VEGF) in the pathogenesis of OHSS. VEGF is a member of a family
of heparin-binding proteins that act directly on endothelial cells to induce
proliferation and angiogenesis (Gospodarowicz et al., 1989; Ferrara and Henzel,
1989; Millauer et al., 1993; Rizk and Nawar, 2004). Vascular permeability
factor (VPF) was characterized as a protein that promotes extravasation
of proteins from tumor-associated blood vessels (Senger et al., 1983). It was
subsequently realized that the permeability inducing factor and the endothelial
cell growth factor are encoded by a single VEGF gene. Several VEGF

<< . .

( 30)

. . >>

Copyright Design by: Sunlight webdesign