FERTILITY AND STERILITY Copyright
©
Vol. 56, No.6, December 1991
Printed on acid-free paper in U.S.A.
1991 The American Fertility Society
Severe ovarian hyperstimulation syndrome: role of peripheral vasodilation
Juan Balasch, M.D.*t Vicente Arroyo, M.D.:\: Francisco Carmona, M.D. * Jose Llach, M.D.:\:
Wladimiro Jimenez, M.D.§ Juan C. Pare, M.D. II Juan A. Vanrell, M.D.*
Faculty of Medicine, Hospital Cl£nic i Provincial, Barcelona, Spain
Objective: To investigate the pathogenesis of the systemic hemodynamic disturbance and the renal production of vasodilator prostaglandins (PGs) in the ovarian hyperstimulation syndrome. Design: Prospective longitudinal study. Setting: Assisted Reproduction Unit of the Hospital Clinic i Provincial in Barcelona. Patients: Five in vitro fertilization patients with ascites because of severe ovarian hyperstimulation syndrome. Main Outcome Measures: Measurement during the syndrome and 4 weeks after recovery of the following: cardiac output, arterial pressure, estimated peripheral vascular resistances, hematocrit, standard renal function tests, plasma renin activity, plasma aldosterone, norepinephrine and antidiuretic hormone concentrations, and urinary excretion of PGE 2 and 6-keto-PGF la . Results: During the syndrome, all patients showed arterial hypotension (74.2 ± 3.8 versus 85.8 ± 1.0 mm Hg), tachycardia, increased cardiac output (6.4 ± 0.2 versus 4.4 ± 0.1 L/min), low peripheral vascular resistance (929 ± 52 versus 1,568 ± 51 dyn/sec per cm- 5 ), high plasma levels of renin (72 ± 25 versus 0.5 ± 0.1 ng/mL per h-l), norepinephrine (639 ± 141 versus 203 ± 21 pg/mL) and antidiuretic hormone (6.1 ± 1.6 versus 1.5 ± 0.1 pg/mL), and increased urinary excretion of PGE 2 (551 ± 152 versus 106 ± 44 pg/min) and 6-keto-PGF la (470 ± 76 versus 99 ± 11 pg/min). No evidence of hemoconcentration, as assessed by hematocrit, was observed in any patient. Conclusions: (1) Severe ovarian hyperstimulation syndrome is related to marked arteriolar vasodilation that leads to underfilling of the arterial vascular compartment and stimulation of endogenous vasoconstrictor systems and (2) the increased urinary excretion of PGs probably represents a homeostatic response to antagonize the renal effects of these systems. Fertil Steril 1991;56:1077-83
The ovarian hyperstimulation syndrome is a relatively common complication of ovulation induction with human menopausal gonadotropin (hMG) and human chorionic gonadotropin (heG).1 At present, the syndrome is being increasingly recognized beReceived December 11, 1990; revised and accepted August 9, 1991. • Department of Obstetrics and Gynecology. t Reprint requests: Juan Balasch, M.D., Department of Obstetrics and Gynecology, Hospital Clinic i Provincial, Cj Casanova 143, 08036-Barcelona, Spain. :j: Liver Unit. § Hormonal Laboratory. II Cardiology Unit. Vol. 56, No.6, December 1991
Balasch et al.
cause of the high number of women undergoing in vitro fertilization (IVF) techniques. 2 In its severe form, ovarian hyperstimulation syndrome is characterized by marked ovarian enlargement, ascites, pleural effusion, high plasma levels of renin and aldosterone, sodium retention, and oliguria. 1 Some patients may even develop marked arterial hypotension, renal failure, and death. 3 Severe ovarian hyperstimulation syndrome is related to an increase in vascular permeability, especially of the ovarian vessels, leading to leaking of intravascular fluid in'to the peritoneal cavity and acute reduction of intravascular volume. 1,3 Although several substances have been considered as possible mediators of this in-
Peripheral arteriolar vasodilation in ovarian hyperstimulation syndrome
1077
creased capillary permeability (histamine, prolactin, renin), prostaglandins (PGs) are the most widely accepted. 1,3,4 On this basis, the current management of the syndrome includes measures focused at maintaining intravascular volume with colloid volume expanders and decreasing vascular permeability with indomethacin. 1,4 However, up to the present there has been no study investigating circulatory function in this syndrome. Furthermore, it is not known whether, as in other high-renin states, the renal production of vasodilator PGs is homeostatically increased in ovarian hyperstimulation syndrome to maintain renal perfusion. In this case, indomethacin administration would be contraindicated. The current article reports the results of one study assessing systemic hemodynamics, the activity of several endogenous vasoactive systems (renin-angiotensin-aldosterone and sympathetic nervous systems and antidiuretic hormone), and the urinary excretion of the vasodilator PGE 2 and PGI 2 in five patients with severe ovarian hyperstimulation syndrome. The aim of this study was to investigate the pathogenesis of the systemic hemodynamic disturbance and the renal production of vasodilator PGs in this syndrome. MATERIALS AND METHODS
From January 1989 to June 1990, 268 IVF cycles were performed in 197 women in the Assisted Reproduction Unit of the Hospital Clinic i Provincial of Barcelona. Eight women developed severe ovarian hyperstimulation syndrome (grade 5) according to the clinical categories proposed by Rabau et al. 5 and modified by Schenker and Weinstein. 1 The present report includes five of these eight patients, who participated in the study after detailed explanation of the nature and purpose of the investigation. The protocol was approved by the Investigation and Ethics Committee of our hospital. Ovarian stimulation was carried out with pure follicle-stimulating hormone (FSH) and hMG under pituitary suppression with buserelin acetate (Suprefact; Behring SA, Barcelona, Spain), 300 J,Lg subcutaneously two times a day until pituitary desensitization was achieved. On days 1 and 2 of ovarian stimulation, 225 IU/d of hMG (Pergonal; Serono SA, Barcelona, Spain) and 225 IU/d of FSH (Fertinorm; Serono SA) were administered intramuscularly (1M). Thereafter, hMG (150 IU) was administered daily until the two leading follicles had reached an average diameter of> 17 mm and plasma estradiol (E 2 ) levels were> 1,000 pg/mL. Follicular 1078
Balasch et al.
aspiration was performed 36 hours after an 1M dose of 5,000 IU of hCG. Forty-eight hours later, three to four embryos per patient were transferred into the uterus. Additional doses of 5,000 and 2,500 IU hCG were given on the days of follicular aspiration and embryo transfer, respectively, to supplement the luteal phase. None ofthe five patients became pregnant. After diagnosis of severe ovarian hyperstimulation syndrome, patients were admitted to the hospital and given a 60 mEq sodium diet (before admission all patients had had a normal sodium diet). Urine volume was carefully collected to measure electrolytes and urinary concentration of PGE 2 and 6-ketoPGF 1a , a stable metabolite ofPG1 2 • The next morning after overnight fasting from food and after 2 hours of bed rest, an antecubital vein was catheterized. Thirty minutes later blood samples were taken to measure plasma renin activity, plasma aldosterone, norephinephrine and antidiuretic hormone concentrations, serum creatinine, serum electrolytes, and hematocrit. Subsequently, an ascitic fluid sample was obtained under echographic control for biochemical and cytologic examination. Patients were then transferred to the Echocardiographic Unit in which cardiac output and arterial pressure were measured and heart rate was recorded. Plasma renin activity, plasma aldosterone and antidiuretic hormone concentrations, and urinary concentration of PGE 2 and 6-keto-PGF 1a were measured by radioimmunoassay following methods previously described. 6 ,7 Norepinephrine was measured by the radioenzymatic assay of Peuler and Johnson. s Normal values for these measurements in our laboratory are: plasma renin activity, 1.17 ± 0.16 (mean ± SEM) ng/mL per h; plasma aldosterone, 10.5 ± 1.18 ng/dL; plasma norepinephrine, 224 ± 17.7 pg/mL; plasma antidiuretic hormone, 2.4. ± 0.2 pg/mL; urinary excretion of PGE 2 , 120 ± 20 pg/min; and urinary excretion of6-keto-PGF 1a , 145 ± 17 pg/min. Cardiac output was measured by twodimensional and Doppler echocardiographic examination (SSH-65A; Toshiba, Tokyo, Japan)9 and arterial pressure by sphygmomanometry. Mean arterial pressure was calculated as dyastolic blood pressure plus t of the difference between systolic and dyastolic blood pressures. Peripheral vascular resistance was estimated by the formula (mean arterial pressure - right atrial pressure/cardiac output) X 80. Because measurement of right atrial pressure was not performed in any patient, it was considered as zero in this calculation.
Peripheral arteriolar vasodilation in ovarian hyperstimulation syndrome
Fertility and Sterility
After completion of these measurements, patients were treated with diuretics, and one woman with renal failure (serum creatinine, 3.7 mgjdL; normal values < 1.2 mgjdL) also received low molecular weight dextran and human serum albumin. Ascites disappeared in all cases within a period ranging between 5 and 12 days after admission. In the patient with renal failure, renal function returned to normal in 7 days. Four weeks after being discharged, women were readmitted to repeat all the above mentioned measurements. Thus, patients were their own controls for hemodynamic, humoral, and renal excretory studies during ovarian hyperstimulation syndrome. Results are presented as means ± SEM. The paired Student's t-test and the nonparametric test of Wilcoxon were used for statistical analysis of results. RESULTS
Peak plasma E2 levels during ovarian stimulation in the five patients studied were 5,210 ± 586 pgjmL (range 3,220 to 6,533). The mean number offollicles observed by vaginal ultrasonography on the day of heG administration was 26.6 ± 3.6, range 19 to 40. The size of the ovaries at admission ranged between 12 X 12 and 20 X 17 cm. The patient admitted to the hospital with renal failure presented the highest peak plasma E2 level and number of follicles during ovarian stimulation, as well as the largest ovaries. At admission, all patients had marked abdominal distention because of the presence of enlarged ovaries and ascites, and two also had pleural effusion. No patient had peripheral edema. The amount of fluid retained, as estimated by the loss of body weight during hospitalization, was 4.3 ± 0.5 kg (range 4 to 5.2 kg) (Fig. 1). Serum creatinine was normal in four cases. As previously indicated, one patient was admitted to the hospital with severe renal failure. All patients showed normal standard liver function tests and an exudative ascites (total protein concentration in ascitic fluid, 4.2 ± 0.6 gjdL; range, 2.6 to 6.1 gjdL). The ascitic fluid concentration of leukocytes and red blood cells were 408 ± 121 per mL (range 120 to 850 per mL) and 83,000 ± 49,000 per mL (range 2,300 to 240,000 per mL), respectively. The percentage of polymorphonuclears over the total concentration of leukocytes in ascitic fluid was 52% ± 11% (range 15% to 80%). Table 1 shows the mean values of systemic hemodynamic, renal function and hormonal measurements at admission and 4 weeks after recovery from the ovarian hyperstimulation syndrome in the five Vol. 56, No.6, December 1991
Balasch et al.
HEMATOCRIT (%) 50 45
BODY WEIGHT (Kg)
80
/.
40
60
35
50
30
OHSS
R
~
70
40
~ ~
OHSS
R
Figure 1 Individual changes in hematocrit and body weight in the five patients studied. Values given in the figure represent those obtained during ovarian hyperstimulation syndrome and at the end of hospitalization (body weight) and 4 weeks after recovery (hematocrit). The open dots represent values of the patient with ovarian hyperstimulation syndrome and renal failure.
patients studied. Figure 2 illustrates the individual values of the hemodynamic measurements. As compared with recovery values, ovarian hyperstimulation syndrome was characterized by arterial hypotension, tachycardia, high cardiac output, and low estimated peripheral resistance. No evidence of hemoconcentration was observed (Fig. 1). During ovarian hyperstimulation syndrome, there was oliguria and marked sodium retention. All patients showed dilutional hyponatremia (serum sodium concentration ranged from 121 to 134 mEqfmL; normal values in our laboratory, 135 to 145 mEqfmL). Plasma renin activity, plasma aldosterone, norepinephrine and antidiuretic hormone concentrations, and urinary excretion of PGE2 and 6-ketoPGF la were markedly increased during ovarian hyperstimulation syndrome and normal 4 weeks after recovery (Table 1). Individual values are represented in Figure 3. The patient admitted to the hospital with ovarian hyperstimulation syndrome and renal failure showed the lowest values of mean arterial pressure and estimated peripheral vascular resistance, the highest plasma levels of renin, norepinephrine and antidiuretic hormone, and the lowest urinary excretion of PGE2 and 6-keto-PGF la (Figs. 2 and 3). DISCUSSION
According to the current hypothesis, the initial pathophysiological event in severe ovarian hyper-
Peripheral arteriolar vasodilation in ovarian hyperstimulation syndrome
1079
r
Table 1 Systemic Hemodynamic, Renal Function and Hormonal Measurements During Ovarian Hyperstimulation Syndrome and 4 Weeks After Recovery Ovarian hyperstimulation syndrome Body weight (kg)" Mean arterial pressure Heart rate (beats/min) Cardiac output (L/min) Estimated peripheral vascular resistance (dyn/sec per cm- 5 ) Hematocrit (%) Urine volume (mL/d) Sodium excretion (mEqJd) Serum sodium (mEq/L) Serum creatinine (mg/dL) Plasma renin activity (ng/mL per h) Plasma aldosterone (ng/dL) Plasma norepinephrine (pg/mL) Plasma antidiuretic hormone (pg/mL) Urinary PGE2 (pg/min) Urinary 6-keto-PGF 1• (pg/min)
62.4 74.2 94.8 6.4
± ± ± ±
929.6 38.2 0.8 25.0 128.6 1.4 72.7 228.6 639.8 6.1 551 470
± ± ± ± ± ± ± ± ± ± ± ±
" Recovery values of body weight represent those obtained at the end of hospitalization.
stimulation syndrome is an increased capillary permeability, especially in the enlarged ovaries, leading to the extravasation of fluid to the abdominal cavity.1,3 An increased local production of PGs3 and histamine,lO and the activation of the follicular renin-angiotensin systeml l observed in patients undergoing ovarian stimulation have been suggested as probable mechanisms of this abnormality. The extravasation of fluid into the peritoneal cavity would cause the formation of ascites, a contraction of circulating blood volume, arterial hypotension, and renal sodium and water retention. Renal failure would be the expression of an extreme diminution of intravascular volume. The observation of hemoconcentration in some patients with ovarian hyperstimulation syndrome12 and the demonstration of increased ovarian capillary permeability in experimental ovarian hyperstimulation syndrome13 are the main arguments given to support this hypothesis. The results of the present study are not in agreement with this theory. None ofthe patients studied showed hemoconcentration during the syndrome. Furthermore, if ovarian hyperstimulation syndrome was because of a shift of fluid from the intravascular compartment to the peritoneal cavity leading to a contraction of the circulating blood volume, a reduced cardiac output and increased peripheral vascular resistance would be found. However, cardiac output was increased, and estimated peripheral vascular resistance was markedly reduced in all the 1080
Balasch et a1.
4.4 b 3.8 6.7 0.2
52.7 2.1 0.1 2.0 2.3 0.6 24.9 61.8 141.3 1.6 152 76
Recovery
P values
58.0± 85.8± 73.2 ± 4.4 ±
4.5 1.0 1.0 0.1
©
Vol. 56, No.6, December 1991
Printed on acid-free paper in U.S.A.
1991 The American Fertility Society
Severe ovarian hyperstimulation syndrome: role of peripheral vasodilation
Juan Balasch, M.D.*t Vicente Arroyo, M.D.:\: Francisco Carmona, M.D. * Jose Llach, M.D.:\:
Wladimiro Jimenez, M.D.§ Juan C. Pare, M.D. II Juan A. Vanrell, M.D.*
Faculty of Medicine, Hospital Cl£nic i Provincial, Barcelona, Spain
Objective: To investigate the pathogenesis of the systemic hemodynamic disturbance and the renal production of vasodilator prostaglandins (PGs) in the ovarian hyperstimulation syndrome. Design: Prospective longitudinal study. Setting: Assisted Reproduction Unit of the Hospital Clinic i Provincial in Barcelona. Patients: Five in vitro fertilization patients with ascites because of severe ovarian hyperstimulation syndrome. Main Outcome Measures: Measurement during the syndrome and 4 weeks after recovery of the following: cardiac output, arterial pressure, estimated peripheral vascular resistances, hematocrit, standard renal function tests, plasma renin activity, plasma aldosterone, norepinephrine and antidiuretic hormone concentrations, and urinary excretion of PGE 2 and 6-keto-PGF la . Results: During the syndrome, all patients showed arterial hypotension (74.2 ± 3.8 versus 85.8 ± 1.0 mm Hg), tachycardia, increased cardiac output (6.4 ± 0.2 versus 4.4 ± 0.1 L/min), low peripheral vascular resistance (929 ± 52 versus 1,568 ± 51 dyn/sec per cm- 5 ), high plasma levels of renin (72 ± 25 versus 0.5 ± 0.1 ng/mL per h-l), norepinephrine (639 ± 141 versus 203 ± 21 pg/mL) and antidiuretic hormone (6.1 ± 1.6 versus 1.5 ± 0.1 pg/mL), and increased urinary excretion of PGE 2 (551 ± 152 versus 106 ± 44 pg/min) and 6-keto-PGF la (470 ± 76 versus 99 ± 11 pg/min). No evidence of hemoconcentration, as assessed by hematocrit, was observed in any patient. Conclusions: (1) Severe ovarian hyperstimulation syndrome is related to marked arteriolar vasodilation that leads to underfilling of the arterial vascular compartment and stimulation of endogenous vasoconstrictor systems and (2) the increased urinary excretion of PGs probably represents a homeostatic response to antagonize the renal effects of these systems. Fertil Steril 1991;56:1077-83
The ovarian hyperstimulation syndrome is a relatively common complication of ovulation induction with human menopausal gonadotropin (hMG) and human chorionic gonadotropin (heG).1 At present, the syndrome is being increasingly recognized beReceived December 11, 1990; revised and accepted August 9, 1991. • Department of Obstetrics and Gynecology. t Reprint requests: Juan Balasch, M.D., Department of Obstetrics and Gynecology, Hospital Clinic i Provincial, Cj Casanova 143, 08036-Barcelona, Spain. :j: Liver Unit. § Hormonal Laboratory. II Cardiology Unit. Vol. 56, No.6, December 1991
Balasch et al.
cause of the high number of women undergoing in vitro fertilization (IVF) techniques. 2 In its severe form, ovarian hyperstimulation syndrome is characterized by marked ovarian enlargement, ascites, pleural effusion, high plasma levels of renin and aldosterone, sodium retention, and oliguria. 1 Some patients may even develop marked arterial hypotension, renal failure, and death. 3 Severe ovarian hyperstimulation syndrome is related to an increase in vascular permeability, especially of the ovarian vessels, leading to leaking of intravascular fluid in'to the peritoneal cavity and acute reduction of intravascular volume. 1,3 Although several substances have been considered as possible mediators of this in-
Peripheral arteriolar vasodilation in ovarian hyperstimulation syndrome
1077
creased capillary permeability (histamine, prolactin, renin), prostaglandins (PGs) are the most widely accepted. 1,3,4 On this basis, the current management of the syndrome includes measures focused at maintaining intravascular volume with colloid volume expanders and decreasing vascular permeability with indomethacin. 1,4 However, up to the present there has been no study investigating circulatory function in this syndrome. Furthermore, it is not known whether, as in other high-renin states, the renal production of vasodilator PGs is homeostatically increased in ovarian hyperstimulation syndrome to maintain renal perfusion. In this case, indomethacin administration would be contraindicated. The current article reports the results of one study assessing systemic hemodynamics, the activity of several endogenous vasoactive systems (renin-angiotensin-aldosterone and sympathetic nervous systems and antidiuretic hormone), and the urinary excretion of the vasodilator PGE 2 and PGI 2 in five patients with severe ovarian hyperstimulation syndrome. The aim of this study was to investigate the pathogenesis of the systemic hemodynamic disturbance and the renal production of vasodilator PGs in this syndrome. MATERIALS AND METHODS
From January 1989 to June 1990, 268 IVF cycles were performed in 197 women in the Assisted Reproduction Unit of the Hospital Clinic i Provincial of Barcelona. Eight women developed severe ovarian hyperstimulation syndrome (grade 5) according to the clinical categories proposed by Rabau et al. 5 and modified by Schenker and Weinstein. 1 The present report includes five of these eight patients, who participated in the study after detailed explanation of the nature and purpose of the investigation. The protocol was approved by the Investigation and Ethics Committee of our hospital. Ovarian stimulation was carried out with pure follicle-stimulating hormone (FSH) and hMG under pituitary suppression with buserelin acetate (Suprefact; Behring SA, Barcelona, Spain), 300 J,Lg subcutaneously two times a day until pituitary desensitization was achieved. On days 1 and 2 of ovarian stimulation, 225 IU/d of hMG (Pergonal; Serono SA, Barcelona, Spain) and 225 IU/d of FSH (Fertinorm; Serono SA) were administered intramuscularly (1M). Thereafter, hMG (150 IU) was administered daily until the two leading follicles had reached an average diameter of> 17 mm and plasma estradiol (E 2 ) levels were> 1,000 pg/mL. Follicular 1078
Balasch et al.
aspiration was performed 36 hours after an 1M dose of 5,000 IU of hCG. Forty-eight hours later, three to four embryos per patient were transferred into the uterus. Additional doses of 5,000 and 2,500 IU hCG were given on the days of follicular aspiration and embryo transfer, respectively, to supplement the luteal phase. None ofthe five patients became pregnant. After diagnosis of severe ovarian hyperstimulation syndrome, patients were admitted to the hospital and given a 60 mEq sodium diet (before admission all patients had had a normal sodium diet). Urine volume was carefully collected to measure electrolytes and urinary concentration of PGE 2 and 6-ketoPGF 1a , a stable metabolite ofPG1 2 • The next morning after overnight fasting from food and after 2 hours of bed rest, an antecubital vein was catheterized. Thirty minutes later blood samples were taken to measure plasma renin activity, plasma aldosterone, norephinephrine and antidiuretic hormone concentrations, serum creatinine, serum electrolytes, and hematocrit. Subsequently, an ascitic fluid sample was obtained under echographic control for biochemical and cytologic examination. Patients were then transferred to the Echocardiographic Unit in which cardiac output and arterial pressure were measured and heart rate was recorded. Plasma renin activity, plasma aldosterone and antidiuretic hormone concentrations, and urinary concentration of PGE 2 and 6-keto-PGF 1a were measured by radioimmunoassay following methods previously described. 6 ,7 Norepinephrine was measured by the radioenzymatic assay of Peuler and Johnson. s Normal values for these measurements in our laboratory are: plasma renin activity, 1.17 ± 0.16 (mean ± SEM) ng/mL per h; plasma aldosterone, 10.5 ± 1.18 ng/dL; plasma norepinephrine, 224 ± 17.7 pg/mL; plasma antidiuretic hormone, 2.4. ± 0.2 pg/mL; urinary excretion of PGE 2 , 120 ± 20 pg/min; and urinary excretion of6-keto-PGF 1a , 145 ± 17 pg/min. Cardiac output was measured by twodimensional and Doppler echocardiographic examination (SSH-65A; Toshiba, Tokyo, Japan)9 and arterial pressure by sphygmomanometry. Mean arterial pressure was calculated as dyastolic blood pressure plus t of the difference between systolic and dyastolic blood pressures. Peripheral vascular resistance was estimated by the formula (mean arterial pressure - right atrial pressure/cardiac output) X 80. Because measurement of right atrial pressure was not performed in any patient, it was considered as zero in this calculation.
Peripheral arteriolar vasodilation in ovarian hyperstimulation syndrome
Fertility and Sterility
After completion of these measurements, patients were treated with diuretics, and one woman with renal failure (serum creatinine, 3.7 mgjdL; normal values < 1.2 mgjdL) also received low molecular weight dextran and human serum albumin. Ascites disappeared in all cases within a period ranging between 5 and 12 days after admission. In the patient with renal failure, renal function returned to normal in 7 days. Four weeks after being discharged, women were readmitted to repeat all the above mentioned measurements. Thus, patients were their own controls for hemodynamic, humoral, and renal excretory studies during ovarian hyperstimulation syndrome. Results are presented as means ± SEM. The paired Student's t-test and the nonparametric test of Wilcoxon were used for statistical analysis of results. RESULTS
Peak plasma E2 levels during ovarian stimulation in the five patients studied were 5,210 ± 586 pgjmL (range 3,220 to 6,533). The mean number offollicles observed by vaginal ultrasonography on the day of heG administration was 26.6 ± 3.6, range 19 to 40. The size of the ovaries at admission ranged between 12 X 12 and 20 X 17 cm. The patient admitted to the hospital with renal failure presented the highest peak plasma E2 level and number of follicles during ovarian stimulation, as well as the largest ovaries. At admission, all patients had marked abdominal distention because of the presence of enlarged ovaries and ascites, and two also had pleural effusion. No patient had peripheral edema. The amount of fluid retained, as estimated by the loss of body weight during hospitalization, was 4.3 ± 0.5 kg (range 4 to 5.2 kg) (Fig. 1). Serum creatinine was normal in four cases. As previously indicated, one patient was admitted to the hospital with severe renal failure. All patients showed normal standard liver function tests and an exudative ascites (total protein concentration in ascitic fluid, 4.2 ± 0.6 gjdL; range, 2.6 to 6.1 gjdL). The ascitic fluid concentration of leukocytes and red blood cells were 408 ± 121 per mL (range 120 to 850 per mL) and 83,000 ± 49,000 per mL (range 2,300 to 240,000 per mL), respectively. The percentage of polymorphonuclears over the total concentration of leukocytes in ascitic fluid was 52% ± 11% (range 15% to 80%). Table 1 shows the mean values of systemic hemodynamic, renal function and hormonal measurements at admission and 4 weeks after recovery from the ovarian hyperstimulation syndrome in the five Vol. 56, No.6, December 1991
Balasch et al.
HEMATOCRIT (%) 50 45
BODY WEIGHT (Kg)
80
/.
40
60
35
50
30
OHSS
R
~
70
40
~ ~
OHSS
R
Figure 1 Individual changes in hematocrit and body weight in the five patients studied. Values given in the figure represent those obtained during ovarian hyperstimulation syndrome and at the end of hospitalization (body weight) and 4 weeks after recovery (hematocrit). The open dots represent values of the patient with ovarian hyperstimulation syndrome and renal failure.
patients studied. Figure 2 illustrates the individual values of the hemodynamic measurements. As compared with recovery values, ovarian hyperstimulation syndrome was characterized by arterial hypotension, tachycardia, high cardiac output, and low estimated peripheral resistance. No evidence of hemoconcentration was observed (Fig. 1). During ovarian hyperstimulation syndrome, there was oliguria and marked sodium retention. All patients showed dilutional hyponatremia (serum sodium concentration ranged from 121 to 134 mEqfmL; normal values in our laboratory, 135 to 145 mEqfmL). Plasma renin activity, plasma aldosterone, norepinephrine and antidiuretic hormone concentrations, and urinary excretion of PGE2 and 6-ketoPGF la were markedly increased during ovarian hyperstimulation syndrome and normal 4 weeks after recovery (Table 1). Individual values are represented in Figure 3. The patient admitted to the hospital with ovarian hyperstimulation syndrome and renal failure showed the lowest values of mean arterial pressure and estimated peripheral vascular resistance, the highest plasma levels of renin, norepinephrine and antidiuretic hormone, and the lowest urinary excretion of PGE2 and 6-keto-PGF la (Figs. 2 and 3). DISCUSSION
According to the current hypothesis, the initial pathophysiological event in severe ovarian hyper-
Peripheral arteriolar vasodilation in ovarian hyperstimulation syndrome
1079
r
Table 1 Systemic Hemodynamic, Renal Function and Hormonal Measurements During Ovarian Hyperstimulation Syndrome and 4 Weeks After Recovery Ovarian hyperstimulation syndrome Body weight (kg)" Mean arterial pressure Heart rate (beats/min) Cardiac output (L/min) Estimated peripheral vascular resistance (dyn/sec per cm- 5 ) Hematocrit (%) Urine volume (mL/d) Sodium excretion (mEqJd) Serum sodium (mEq/L) Serum creatinine (mg/dL) Plasma renin activity (ng/mL per h) Plasma aldosterone (ng/dL) Plasma norepinephrine (pg/mL) Plasma antidiuretic hormone (pg/mL) Urinary PGE2 (pg/min) Urinary 6-keto-PGF 1• (pg/min)
62.4 74.2 94.8 6.4
± ± ± ±
929.6 38.2 0.8 25.0 128.6 1.4 72.7 228.6 639.8 6.1 551 470
± ± ± ± ± ± ± ± ± ± ± ±
" Recovery values of body weight represent those obtained at the end of hospitalization.
stimulation syndrome is an increased capillary permeability, especially in the enlarged ovaries, leading to the extravasation of fluid to the abdominal cavity.1,3 An increased local production of PGs3 and histamine,lO and the activation of the follicular renin-angiotensin systeml l observed in patients undergoing ovarian stimulation have been suggested as probable mechanisms of this abnormality. The extravasation of fluid into the peritoneal cavity would cause the formation of ascites, a contraction of circulating blood volume, arterial hypotension, and renal sodium and water retention. Renal failure would be the expression of an extreme diminution of intravascular volume. The observation of hemoconcentration in some patients with ovarian hyperstimulation syndrome12 and the demonstration of increased ovarian capillary permeability in experimental ovarian hyperstimulation syndrome13 are the main arguments given to support this hypothesis. The results of the present study are not in agreement with this theory. None ofthe patients studied showed hemoconcentration during the syndrome. Furthermore, if ovarian hyperstimulation syndrome was because of a shift of fluid from the intravascular compartment to the peritoneal cavity leading to a contraction of the circulating blood volume, a reduced cardiac output and increased peripheral vascular resistance would be found. However, cardiac output was increased, and estimated peripheral vascular resistance was markedly reduced in all the 1080
Balasch et a1.
4.4 b 3.8 6.7 0.2
52.7 2.1 0.1 2.0 2.3 0.6 24.9 61.8 141.3 1.6 152 76
Recovery
P values
58.0± 85.8± 73.2 ± 4.4 ±
4.5 1.0 1.0 0.1
