- Email: [email protected]
Hormone excretion during early pregnancy following spontaneous and clomiphene citrate-induced ovulation*
FERTILITY AND STERILITY Copyright " 1986 The American Fertility Society
Vol. 46, No.2, August 1986 Printed in U.SA.
Hormone excretion during early pregnancy following spontaneous and clomiphene citrate-induced ovulation*
Francesca Strigini, M.D.H William P. Collins, Ph.D., D.Sc.§ Malcolm I. Whitehead, M.R.C.O.G.§ Gian Benedetto Melis, M.D. t Piero Fioretti, M.D. t Stuart Campbell, F.R.C.O.G.§ University of Pisa, Pisa, Italy, and King's College School of Medicine and Dentistry, London, United Kingdom
Endocrine changes during early pregnancy have been studied in 12 patients between days 35 and 91 from the last menstrual period. Ovulation had occurred spontaneously in five patients and was induced with clomiphene citrate (CC) in the remaining seven women. All the patients collected daily samples of early morning urine throughout the period of study; on each sample, human chorionic gonadotropin (hCG), pregnanediol-3-a-glucuronide (PGDG) and estrone-3-g1ucuronide (ElG) were measured by chemiluminescence immunoassay. No significant difference was observed between the two groups of patients with respect to the concentrations of hCG and PGDG. The excretion of ElG, however, was significantly higher in patients treated with CC. These data suggest that the induction of ovulation with CC may affect estrogen production in early pregnancy. Fertil Steril46:2.o9, 1986
Clomiphene citrate (CC) is currently used in the treatment ofluteal phase defects.1 Moreover, it has been used to induce ovulation before artificial insemination or in vitro fertilization. 2 The possible effects on gestational outcome have been repeatedly evaluated, and there is a general agreement on the higher in,cidence of twin preg-
Received December 27, 1985; revised and accepted April 7, 1986. *Supported in part by the Italian National Research Council (C.N.R., Rome, Italy) grant 84.00815.04. tDepartment of Obstetrics and Gynecology, University of Pisa. :f:Reprint requests: Francesca Strigini, M.D., Clinica Ostetrica e Ginecologica, Universita degli Studi di Pisa, Via Roma 67,56100 Pisa, Italy. §Department of Obstetrics and Gynaecology, King's College School of Medicine and Dentistry. Vol. 46, No.2, August 1986
nancies. The possible effects of the drug on the incidence of birth defects and abortions are more controversia1. 3 ,4 For example, it has been suggested that CC might induce chromosomal abnormalities or other developmental abnormalities of the ovum. 4 Alternatively, the drug might impair luteal function or directly affect the decidua during early pregnancy.5 Although a number of studies6 have been devised to evaluate the characteristics of the luteal phase in the ovarian cycle after the administration of CC, only sporadic reports 7 , 8 have been concerned with the endocrine events of early pregnancy. It is well known that the concentrations of pIasrna estrogen and progesterone (P) are characterized by a high day-to-day variability during pregnancy,9 so that repeated samples are often necessary for identification of the hormonal patterns. An alternative approach involves the assay of steStrigini et al.
Urinary steroids in induced pregnancies
209
roid metabolites in serial samples of urine. The production of reagents that enable steroid glucuronides to be measured in diluted urine without previous hydrolysis or extraction led to the development of specific immunoassays.l0 Both radioimmunoassay and chemiluminescence immunoassay have been used for this purpose. l l It has been shown that estrone-3-glucuronide (E 1G) represents a quantitatively important metabolite of estradiol (E 2), whereas the principal metabolite of P is pregnanediol-3-a-glucuronide (PGnG). nefined changes in their concentration or concentration ratio may be used in predicting and detecting ovulation. 12 The simultaneous assessment of estrogen metabolites and PGnG in urine has been reported to be useful for monitoring early pregnancy.13 In the present study, the concentrations of human chorionic gonadotropin (hCG), E 1G, and PGnG have been measured by chemiluminescence immunoassay in daily samples of early morning urine (EMU) obtained from pregnant patients between days 35 and 91 of amenorrhea. The aim of this study was to evaluate whether CC, administered during the follicular phase of the menstrual cycle, was able to affect the secretion of· ovarian steroids during the first trimester of pregnancy. MATERIALS AND METHODS PATIENTS
A total of 12 pregnant women (21 to 39 years of age) were studied. Five of the patients (control group) had not received any drug during the conception cycle and for the period of study, apart from iron supplements. The other seven patients (treated group) had received CC from days 1 to 5 of the conception cycle to induce follicular development; the drug had been administered at the daily doses of 50 mg (n = 2), 100 mg (n = 3), and 150 mg (n = 2). The occurrence of ovulation was monitored indirectly by means of ultrasound and/ or luteinizing hormone assay and/or changes in basal body temperature, and occurred in all of the patients between days 12 and 18 from the beginning of the menstrual period. Fetal viability and normal growth had been demonstrated during the period of study by means of weekly ultrasound scans. No major complication was observed during the follow-up of their pregnancies. All of the patients delivered at term normal infants, whose birthweight ranged between 2950 and 3880 gm. 210
8trigini et al.
Urinary steroids in induced pregnancies
URINE COLLECTION
An EMU sample was defined as the first urine passed after getting up from bed in the morning to perform daily duties. The patients collected daily specimens of EMU from days 35 to 91, as calculated from day 1 of the last menstrual period. All specimens were stored at - 15°C and thawed for analysis. ASSAY PROCEDURES
HCG, E 1G, and PGnG were measured by chemiluminescence immunoassay,11, 14, 15 slightly modified for detection of the higher values typical of pregnancy, compared with those obtained during the menstrual cycle. The calibration curve for hCG ranged from 2.6 to 166 ng/ml (1 ng = 0.0104 IU of the First International Reference Preparation for hCG, provided by the National Institute for Biological Standards and Controls, London, UK). The urine samples were mixed with different amounts of assay buffer, according to the gestational age, to give a dilution of 1:10 (vol/yol) before the sixth week, 1:100 between weeks 6 and 8, and 1:1000 after the eighth week. Before the evaluation of methods for the measurement of E 1G and PGnG, the urine was diluted 1:100. In both assays, dextran-coated charcoal was used to separate the antibody-bound and free fractions. The calibration curves ranged from 156 to 10,000 pg/ml for E 1G and from 62.5 to 500 ng/ml for PGnG. The precision of the methods was assessed by measuring the intraassay variation (which was the coefficient of variation of 20 analyses of the same sample within a single assay) and the interassay variation (which was the coefficient ofvariation of the same sample analyzed in 20 sequential assays). The intraassay variation was 11.2% for hCG (mean value, 27,000 ng/ml) , 8.5% for PGnG (mean value, 47.48 J-Lmolll), and 8.9% for E 1G (mean value, 250.4 nmol/l). The interassay variation was 21.0% for hCG (mean value, 21,900 ng/ml) , 14.7% for PGnG (mean value, 36.96 J-Lmolll) , and 9.9% for E 1G (mean value, 476.2 nmolll). STATISTICAL ANALYSIS
The analysis of results was not performed on a daily basis because of the high day-to-day variability of all three compounds. Consequently, the statistical an.alysis was performed on the comFertility and Sterility
--- spontaneous ovulation - induced ovulation
aoooo
: M!SO
I
II III I
Cl)
~
I
II
~
•
.
~
II o'4b
8~
!iI,
day 70, followed by a slight decrease. On the contrary, the excretion of PGDG did not show any significant modification throughout the period of study in either group of patients. No significant challge in the concentrations of El G was observed in pregnancies after spontaneous ovulation; after CC-induced ovulation, however, the excretion of El G showed a significant increase (P < 0.05) throughout the period of study. The comparison between the two groups of patients showed no significant difference with regard to either hCG or PGDG excretion for the whole period of study (Fig. 4). On the contrary, the excretion ofE 1G was significantly higher (P < 0 ..005) after CC-induced ovulation (38.637 ± 12.643 J.Lmolll for 56 days) than after spontaneous ovulation (14.543 ± 7.347 J.Lmolll for 56 days) (Fig. 4). DISCUSSION
7b
clays from last _rual period
Figure 1 Concentrations of hCG in serial samples of EMU obtained from pregnant patients after spontaneous (n = 5) and CCinduced (n = 7) ovulation.
The assay of urinary metabolites in EMU has been successfully used for monitoring ovulation,12 and the results have shown a good correlation with the data obtained from 24~hour speci-
puted area under the curve, which described the secretion of each metabolite for a defined period of observation. In order to evaluate the increase of each compound within the same group of patients during the first trimester of pregnancy, the area under the curve was calculated for each week. The distribution was best described by a log/ normal model. Therefore, the hormone increase was evaluated by application of the one-way analysis of variance to the computed areas under the curve transformed to log~o. A comparison between the two groups of patients was performed by application of Student's t-test to the area under the curve for the whole period of study that represented the total secretion of each metabolite between days 35 and 91 of the pregnancy.
-- spontan_.· ovulation Induced ovulation
RESULTS
Figures 1 to 3 show the trends of hCG, PGDG, .and E 1G in the two groups of patients, expressed as the mean and standard deviation for the corresponding days from the last menstrual period. In both groups, the concentrations of hCG showed a significant increase (P < 0.001) until Vol. 46, No.2, August 1986
o
~
•
•
ro
~
90
days from last menstrual period
Figure 2 Concentrations of PGDG in serial samples of EMU obtained from pregnant patients after spontaneous (n = 5) and CCinduced (n = 7) ovulation.
8trigini et al.
Urinary steroids in induced pregnancies
211
--- .pont......u•. owlation _ Induced ovulation
I
I MtSD
o
40
50
'10
80
days from la.t menstrual pertod
Figure 3 Concentrations of ElGin serial samples of EMU obtained from pregnant patients after spontaneous (n = 5) and CC· induced (n = 7) ovulation. A significant increase (P < 0.05 by one·way analysis of variance) of El G excretion was shown in CC·treated patients by comparing the area under the curve computed for each gestational week.
o lID
spontaneous
induced
II<
u
)(
W
c
w
....«
a:
..,..
•
T •
.., '". It>
--1
ovulation
40
z
Q
..
:fI
*
2
1 .
•
..,..
..,. '". It>
0
E
=>.
=>.
p
M!SD
•
.,.. ..,. So 20
.::::. 0>
*
ovulation
2
.... w
mens, corrected for the urine volume. l6 In the present study, all the evaluated metabolites showed a high day-to-day variability. This observation can be partially related to the variations of urine concentration, but it certainly also reflects the variations of hormone secretion, the high amplitude of which has been repeatedly shown during pregnancy.9 Therefore, the comparison between the results obtained on a daily basis may be affected by the day-to-day variability of the compounds. The evaluation of the area under the curve offers the advantage of minimizing the possible bias linked to the daily variability. The rapid rise in the concentrations ofhCG, the comparatively slower rise in the amount of ElG, and the lack of any significant increase of PGDG during the study period are in good agreement with the well-known changes in the values for parent hormones in plasmap·l9 As for PGDG, present data are also consistent with the normal range observed in women who conceived without treatment. l3 In pregnancies after ovulation induction, the trend of hCG was similar to that observed after spontaneous ovulation, suggesting that, as expected, the trophoblastic secretion of the hormone is not affected by CC administration during the follicular phase of the menstrual cycle. With regard to PGDG, no significant difference was observed between the two groups of patients. The excretion of El G was significantly higher in pregnancies obtained after CC-induced ovulation. In the same group of patients, the concentrations
1
~1 E E
(!)
w ....
~
0
212
0
0 hCG
8trigini et al.
E 1G
Urinary steroids in induced pregnancies
PGDG
Figure 4 Comparison among the quantitative secretion of hCG, E1G, and PGDG in ear· ly pregnancies after spontaneous (n = 5) and CCinduced (n = 7) ovulation . For each compound the area under the curve between days 35 and 91 of amenorrhea has been computed.
Fertility and Sterility
of El G showed a significant increase during the first trimester of pregnancy, whereas in pregnancies after spontaneous ovulation the rise was not statistically significant. These results are consistent with those reported by other authors,7,8 who observed an increase of blood E2 concentrations during the luteal phase of conception cycles in patients treated withCC. In the above studies the patients were also treated with hCG and/or underwent follicular aspiration for ovum pickup and in vitro fertilization. Our patients were treated· with CC alone. Therefore, the enhanced estrogen production cannot be related to other drugs or follicular disruption but seems to be a consequence of the treatment with CC. The mechanisms through which CC administration during the follicular phase can affect estrogen secretion during the luteal phase and early pregnancy are not clear. The increased estrogen secretion could be related to the maturation of multiple follicles, which may become luteinized. In this case, PGDG excretion also should be raised, because multiple corpora lutea are expected to release increased amounts of P as . well as estrogens. Conversely, in both this and· other studies,7, 8 the concentrations of P and its metabolite were not increased after the treatment with CC. Even if, as has been suggested, CC remains in the body for several weeks,20 it is unlikely that the amount present when hCG starts to rise would be sufficient to increase gonadotropin-stimulated aromatase. 21 It can be also hypothesized that CC induces morphologic and/or functional modifications within the follicles, which become transformed into corpora lutea with different secretory activity. Laufer et al. 7 observed that in the luteal phase of fertile cycles the P/E 2 ratio is reduced by almost 50% and suggested that this reduction may be related to an inadequate process of luteinization of granulosa cells, because of either the treatment schedules or an inherent metabolic failure in patients requiring therapy. The hypothesis of a luteal defect is not consistent with the observation that P and PGDG values were similar in patients and controls. It is more likely thatCC exerts some stimulatory effects on estrogen~producing cells. Quigley et al. 8 observed that even if the number of follicles was similar, during the follicular phase plasma levels of follicle-stimulating hormone were raised in patients treated wih higher doses of GC and hypothesized that follicle-stimulating hormone might induce additional granulosa cell Vol. 46, No.2, August 1986
proliferation. The resulting corpora lutea could have different functional.capabilities, causing estrogen hypersecreton. The same authors8 also suggested that CC might induce additional theca cell luteinization and/or producton of E2 from extraluteal sources. Because CC acts by binding estrogenreceptors in estrogen target tissues,21 this may in turn stimulate further ovarian estrogen production.. In conclusion, present data strongly suggest that the endocrine effects of the administration of CC for the induction of ovulation are not limited to the follicular phase of the cycle, but extend beyond the luteal phase into the first trimester of pregnancy. Further studies are necessary to clarify the pathophysiology and possible consequences of the increased estrogen secretion during early pregnancy .
REFERENCES
J
1. Huppert LC: Induction of ovulation with clomiphene citrate. Fertil Steril 31:1, 1979 2. Hoult IJ, De Crespigny LCh, O'Herlihy C, Speirs AL, Lopata A, Kellow G, Johnston I, Robinson HP: Ultrasound control of clomiphenelhuman chorionic 'gonadotropin stimulated cycles for oocyte recovery and in vitro fertilization. Fertil Steril 36:316, 1981 3. Adashi EY, Rock JA, Sapp KC, Martin EJ, Wentz AC, Jones GS: Gestational outcome of clomiphene-related conceptions. Fertil Steril 31:620, 1979 4. Jansen RPS: Spontaneous abortion incidence in the treatment of infertility. Am J Obstet Gynecol 143:451, 1982 5 .. Lamb EJ, Collifiower WW, WilliamsJW: Endometrial histology and conception rates after clomiphene citrate. Obstet Gynecol 39:389, 1972 6. Talbert LM: Clomiphene citrate induction of ovulation. Fertil Steril 39:742, 1983 7: Laufer N, Navot D, Schenker JG: The pattern of luteal phase plasma progesterone and estradiol in fertile cycles. Am J Obstet Gynecol 143:808, 1982 8. Quigley MM, Berkowitz AS, Gilbert SA, WolfDP: Clomiphene citrate in an· in vitro fertilization program: hormonal comparison between 50- and 150-daily dosages. Fertil Steril 41:809, 1984 9. Parrini D, Facchinetti F, Genazzani A: Physiological variability of endocrine indices normally employed in. clinical practice. In The Human Placenta: Proteins and Hormones, Edited by A Klopper, A Genazzini, PG Crosignani. London, Academic Press, 1980, p 245 10. Kellie AE: The radioimmunoassay of steroid conjugates. J Steroid Biochem 6:277,1975 11. Weerasekera DA, Kim JB, Barnard GJ, Collins WP, Kohen F, Lindner HR: Monitoring ovarian function by a solid phase chemiluminescence immunoassay. Acta Endocrinol (Copenh) 101:254,1982
Strigini et aI. Urinary steroids in induced pregnancies
213
12. World Health Organization: Temporal relationships between indices of the fertile period. Fertil Steril 39:647, 1983 13. Mendizabal AF, Quiroga S, Farinati Z, Lahox M, Nagle C: Hormonal monitoring of early pregnancy by a direct radioimmunoassay of steroid glucuronides in first morning urine. Fertil Steril 42:737, 1984 14. Eshhar Z, Kim JB, Barnard GJ, Collinis WP, Gilard S, Lindner HR, Cohen F: Use of monoclonal antibodies to pregnanediol-3a-glucuronide to the development of a solid phase chemiluminescence immunoassay. Steroids 38:89,1981 15. Brockelbank JL, Kim JB, Barnard GJ, Collins WP, Gaier B, Kohen F: The measurement of urinary LH by a solidphase chemiluminescence immunoassay. Ann Clin Biochern 21:284, 1984 16. Collins WP, Collins PO, Kilpatrick MJ, Manning PA, Pike JM, Tyler JPP: The concentrations of urinary oestrone-3-glucuronide, LH and pregnanediol-3a-glucuronide as indices of ovarian function. Acta Endocrinol (Copenh) 90:336, 1979
214
8trigini et al.
Urinary steroids in induced pregnancies
17. Braunstein GO, Karow WG, Gentry WC, Rasor J, Wade ME: First-trimester chorionic gonadotropin measurements as an aid in the diagnosis of early pregnancy disorders. Am J Obstet Gynecol 131:25, 1978 18. Harrison RF, Kitchin Y: Maternal plasma unconjugated oestrogens in early human pregnancy. Br J Obstet Gynaecol 87:686, 1980 19. Harrison RF, Youssefnejadian E, Brodovcky H, Johnson M, Dewhurst J: Secretion patterns of plasma-progesterone, 17-hydroxyprogesterone and 20a-hydroxy-preg-4en-3-one in early normal pregnancy. Br J Obstet Gynaecol 85:921, 1978 20. O'Herlihy C, Pepperell RJ, Brown JB, Smith MA, Sandri L, McBain JC: Incremental clomiphene therapy: a new method for treating persistent anovulation. Obstet Gynecol 58:535, 1981 21. Adashi EY: Clomiphene citrate: mechanism(s) and site(s) of action-a hypothesis revisited. Fertil Steril 42:331, 1984
Fertility and Sterility
Vol. 46, No.2, August 1986 Printed in U.SA.
Hormone excretion during early pregnancy following spontaneous and clomiphene citrate-induced ovulation*
Francesca Strigini, M.D.H William P. Collins, Ph.D., D.Sc.§ Malcolm I. Whitehead, M.R.C.O.G.§ Gian Benedetto Melis, M.D. t Piero Fioretti, M.D. t Stuart Campbell, F.R.C.O.G.§ University of Pisa, Pisa, Italy, and King's College School of Medicine and Dentistry, London, United Kingdom
Endocrine changes during early pregnancy have been studied in 12 patients between days 35 and 91 from the last menstrual period. Ovulation had occurred spontaneously in five patients and was induced with clomiphene citrate (CC) in the remaining seven women. All the patients collected daily samples of early morning urine throughout the period of study; on each sample, human chorionic gonadotropin (hCG), pregnanediol-3-a-glucuronide (PGDG) and estrone-3-g1ucuronide (ElG) were measured by chemiluminescence immunoassay. No significant difference was observed between the two groups of patients with respect to the concentrations of hCG and PGDG. The excretion of ElG, however, was significantly higher in patients treated with CC. These data suggest that the induction of ovulation with CC may affect estrogen production in early pregnancy. Fertil Steril46:2.o9, 1986
Clomiphene citrate (CC) is currently used in the treatment ofluteal phase defects.1 Moreover, it has been used to induce ovulation before artificial insemination or in vitro fertilization. 2 The possible effects on gestational outcome have been repeatedly evaluated, and there is a general agreement on the higher in,cidence of twin preg-
Received December 27, 1985; revised and accepted April 7, 1986. *Supported in part by the Italian National Research Council (C.N.R., Rome, Italy) grant 84.00815.04. tDepartment of Obstetrics and Gynecology, University of Pisa. :f:Reprint requests: Francesca Strigini, M.D., Clinica Ostetrica e Ginecologica, Universita degli Studi di Pisa, Via Roma 67,56100 Pisa, Italy. §Department of Obstetrics and Gynaecology, King's College School of Medicine and Dentistry. Vol. 46, No.2, August 1986
nancies. The possible effects of the drug on the incidence of birth defects and abortions are more controversia1. 3 ,4 For example, it has been suggested that CC might induce chromosomal abnormalities or other developmental abnormalities of the ovum. 4 Alternatively, the drug might impair luteal function or directly affect the decidua during early pregnancy.5 Although a number of studies6 have been devised to evaluate the characteristics of the luteal phase in the ovarian cycle after the administration of CC, only sporadic reports 7 , 8 have been concerned with the endocrine events of early pregnancy. It is well known that the concentrations of pIasrna estrogen and progesterone (P) are characterized by a high day-to-day variability during pregnancy,9 so that repeated samples are often necessary for identification of the hormonal patterns. An alternative approach involves the assay of steStrigini et al.
Urinary steroids in induced pregnancies
209
roid metabolites in serial samples of urine. The production of reagents that enable steroid glucuronides to be measured in diluted urine without previous hydrolysis or extraction led to the development of specific immunoassays.l0 Both radioimmunoassay and chemiluminescence immunoassay have been used for this purpose. l l It has been shown that estrone-3-glucuronide (E 1G) represents a quantitatively important metabolite of estradiol (E 2), whereas the principal metabolite of P is pregnanediol-3-a-glucuronide (PGnG). nefined changes in their concentration or concentration ratio may be used in predicting and detecting ovulation. 12 The simultaneous assessment of estrogen metabolites and PGnG in urine has been reported to be useful for monitoring early pregnancy.13 In the present study, the concentrations of human chorionic gonadotropin (hCG), E 1G, and PGnG have been measured by chemiluminescence immunoassay in daily samples of early morning urine (EMU) obtained from pregnant patients between days 35 and 91 of amenorrhea. The aim of this study was to evaluate whether CC, administered during the follicular phase of the menstrual cycle, was able to affect the secretion of· ovarian steroids during the first trimester of pregnancy. MATERIALS AND METHODS PATIENTS
A total of 12 pregnant women (21 to 39 years of age) were studied. Five of the patients (control group) had not received any drug during the conception cycle and for the period of study, apart from iron supplements. The other seven patients (treated group) had received CC from days 1 to 5 of the conception cycle to induce follicular development; the drug had been administered at the daily doses of 50 mg (n = 2), 100 mg (n = 3), and 150 mg (n = 2). The occurrence of ovulation was monitored indirectly by means of ultrasound and/ or luteinizing hormone assay and/or changes in basal body temperature, and occurred in all of the patients between days 12 and 18 from the beginning of the menstrual period. Fetal viability and normal growth had been demonstrated during the period of study by means of weekly ultrasound scans. No major complication was observed during the follow-up of their pregnancies. All of the patients delivered at term normal infants, whose birthweight ranged between 2950 and 3880 gm. 210
8trigini et al.
Urinary steroids in induced pregnancies
URINE COLLECTION
An EMU sample was defined as the first urine passed after getting up from bed in the morning to perform daily duties. The patients collected daily specimens of EMU from days 35 to 91, as calculated from day 1 of the last menstrual period. All specimens were stored at - 15°C and thawed for analysis. ASSAY PROCEDURES
HCG, E 1G, and PGnG were measured by chemiluminescence immunoassay,11, 14, 15 slightly modified for detection of the higher values typical of pregnancy, compared with those obtained during the menstrual cycle. The calibration curve for hCG ranged from 2.6 to 166 ng/ml (1 ng = 0.0104 IU of the First International Reference Preparation for hCG, provided by the National Institute for Biological Standards and Controls, London, UK). The urine samples were mixed with different amounts of assay buffer, according to the gestational age, to give a dilution of 1:10 (vol/yol) before the sixth week, 1:100 between weeks 6 and 8, and 1:1000 after the eighth week. Before the evaluation of methods for the measurement of E 1G and PGnG, the urine was diluted 1:100. In both assays, dextran-coated charcoal was used to separate the antibody-bound and free fractions. The calibration curves ranged from 156 to 10,000 pg/ml for E 1G and from 62.5 to 500 ng/ml for PGnG. The precision of the methods was assessed by measuring the intraassay variation (which was the coefficient of variation of 20 analyses of the same sample within a single assay) and the interassay variation (which was the coefficient ofvariation of the same sample analyzed in 20 sequential assays). The intraassay variation was 11.2% for hCG (mean value, 27,000 ng/ml) , 8.5% for PGnG (mean value, 47.48 J-Lmolll), and 8.9% for E 1G (mean value, 250.4 nmol/l). The interassay variation was 21.0% for hCG (mean value, 21,900 ng/ml) , 14.7% for PGnG (mean value, 36.96 J-Lmolll) , and 9.9% for E 1G (mean value, 476.2 nmolll). STATISTICAL ANALYSIS
The analysis of results was not performed on a daily basis because of the high day-to-day variability of all three compounds. Consequently, the statistical an.alysis was performed on the comFertility and Sterility
--- spontaneous ovulation - induced ovulation
aoooo
: M!SO
I
II III I
Cl)
~
I
II
~
•
.
~
II o'4b
8~
!iI,
day 70, followed by a slight decrease. On the contrary, the excretion of PGDG did not show any significant modification throughout the period of study in either group of patients. No significant challge in the concentrations of El G was observed in pregnancies after spontaneous ovulation; after CC-induced ovulation, however, the excretion of El G showed a significant increase (P < 0.05) throughout the period of study. The comparison between the two groups of patients showed no significant difference with regard to either hCG or PGDG excretion for the whole period of study (Fig. 4). On the contrary, the excretion ofE 1G was significantly higher (P < 0 ..005) after CC-induced ovulation (38.637 ± 12.643 J.Lmolll for 56 days) than after spontaneous ovulation (14.543 ± 7.347 J.Lmolll for 56 days) (Fig. 4). DISCUSSION
7b
clays from last _rual period
Figure 1 Concentrations of hCG in serial samples of EMU obtained from pregnant patients after spontaneous (n = 5) and CCinduced (n = 7) ovulation.
The assay of urinary metabolites in EMU has been successfully used for monitoring ovulation,12 and the results have shown a good correlation with the data obtained from 24~hour speci-
puted area under the curve, which described the secretion of each metabolite for a defined period of observation. In order to evaluate the increase of each compound within the same group of patients during the first trimester of pregnancy, the area under the curve was calculated for each week. The distribution was best described by a log/ normal model. Therefore, the hormone increase was evaluated by application of the one-way analysis of variance to the computed areas under the curve transformed to log~o. A comparison between the two groups of patients was performed by application of Student's t-test to the area under the curve for the whole period of study that represented the total secretion of each metabolite between days 35 and 91 of the pregnancy.
-- spontan_.· ovulation Induced ovulation
RESULTS
Figures 1 to 3 show the trends of hCG, PGDG, .and E 1G in the two groups of patients, expressed as the mean and standard deviation for the corresponding days from the last menstrual period. In both groups, the concentrations of hCG showed a significant increase (P < 0.001) until Vol. 46, No.2, August 1986
o
~
•
•
ro
~
90
days from last menstrual period
Figure 2 Concentrations of PGDG in serial samples of EMU obtained from pregnant patients after spontaneous (n = 5) and CCinduced (n = 7) ovulation.
8trigini et al.
Urinary steroids in induced pregnancies
211
--- .pont......u•. owlation _ Induced ovulation
I
I MtSD
o
40
50
'10
80
days from la.t menstrual pertod
Figure 3 Concentrations of ElGin serial samples of EMU obtained from pregnant patients after spontaneous (n = 5) and CC· induced (n = 7) ovulation. A significant increase (P < 0.05 by one·way analysis of variance) of El G excretion was shown in CC·treated patients by comparing the area under the curve computed for each gestational week.
o lID
spontaneous
induced
II<
u
)(
W
c
w
....«
a:
..,..
•
T •
.., '". It>
--1
ovulation
40
z
Q
..
:fI
*
2
1 .
•
..,..
..,. '". It>
0
E
=>.
=>.
p
M!SD
•
.,.. ..,. So 20
.::::. 0>
*
ovulation
2
.... w
mens, corrected for the urine volume. l6 In the present study, all the evaluated metabolites showed a high day-to-day variability. This observation can be partially related to the variations of urine concentration, but it certainly also reflects the variations of hormone secretion, the high amplitude of which has been repeatedly shown during pregnancy.9 Therefore, the comparison between the results obtained on a daily basis may be affected by the day-to-day variability of the compounds. The evaluation of the area under the curve offers the advantage of minimizing the possible bias linked to the daily variability. The rapid rise in the concentrations ofhCG, the comparatively slower rise in the amount of ElG, and the lack of any significant increase of PGDG during the study period are in good agreement with the well-known changes in the values for parent hormones in plasmap·l9 As for PGDG, present data are also consistent with the normal range observed in women who conceived without treatment. l3 In pregnancies after ovulation induction, the trend of hCG was similar to that observed after spontaneous ovulation, suggesting that, as expected, the trophoblastic secretion of the hormone is not affected by CC administration during the follicular phase of the menstrual cycle. With regard to PGDG, no significant difference was observed between the two groups of patients. The excretion of El G was significantly higher in pregnancies obtained after CC-induced ovulation. In the same group of patients, the concentrations
1
~1 E E
(!)
w ....
~
0
212
0
0 hCG
8trigini et al.
E 1G
Urinary steroids in induced pregnancies
PGDG
Figure 4 Comparison among the quantitative secretion of hCG, E1G, and PGDG in ear· ly pregnancies after spontaneous (n = 5) and CCinduced (n = 7) ovulation . For each compound the area under the curve between days 35 and 91 of amenorrhea has been computed.
Fertility and Sterility
of El G showed a significant increase during the first trimester of pregnancy, whereas in pregnancies after spontaneous ovulation the rise was not statistically significant. These results are consistent with those reported by other authors,7,8 who observed an increase of blood E2 concentrations during the luteal phase of conception cycles in patients treated withCC. In the above studies the patients were also treated with hCG and/or underwent follicular aspiration for ovum pickup and in vitro fertilization. Our patients were treated· with CC alone. Therefore, the enhanced estrogen production cannot be related to other drugs or follicular disruption but seems to be a consequence of the treatment with CC. The mechanisms through which CC administration during the follicular phase can affect estrogen secretion during the luteal phase and early pregnancy are not clear. The increased estrogen secretion could be related to the maturation of multiple follicles, which may become luteinized. In this case, PGDG excretion also should be raised, because multiple corpora lutea are expected to release increased amounts of P as . well as estrogens. Conversely, in both this and· other studies,7, 8 the concentrations of P and its metabolite were not increased after the treatment with CC. Even if, as has been suggested, CC remains in the body for several weeks,20 it is unlikely that the amount present when hCG starts to rise would be sufficient to increase gonadotropin-stimulated aromatase. 21 It can be also hypothesized that CC induces morphologic and/or functional modifications within the follicles, which become transformed into corpora lutea with different secretory activity. Laufer et al. 7 observed that in the luteal phase of fertile cycles the P/E 2 ratio is reduced by almost 50% and suggested that this reduction may be related to an inadequate process of luteinization of granulosa cells, because of either the treatment schedules or an inherent metabolic failure in patients requiring therapy. The hypothesis of a luteal defect is not consistent with the observation that P and PGDG values were similar in patients and controls. It is more likely thatCC exerts some stimulatory effects on estrogen~producing cells. Quigley et al. 8 observed that even if the number of follicles was similar, during the follicular phase plasma levels of follicle-stimulating hormone were raised in patients treated wih higher doses of GC and hypothesized that follicle-stimulating hormone might induce additional granulosa cell Vol. 46, No.2, August 1986
proliferation. The resulting corpora lutea could have different functional.capabilities, causing estrogen hypersecreton. The same authors8 also suggested that CC might induce additional theca cell luteinization and/or producton of E2 from extraluteal sources. Because CC acts by binding estrogenreceptors in estrogen target tissues,21 this may in turn stimulate further ovarian estrogen production.. In conclusion, present data strongly suggest that the endocrine effects of the administration of CC for the induction of ovulation are not limited to the follicular phase of the cycle, but extend beyond the luteal phase into the first trimester of pregnancy. Further studies are necessary to clarify the pathophysiology and possible consequences of the increased estrogen secretion during early pregnancy .
REFERENCES
J
1. Huppert LC: Induction of ovulation with clomiphene citrate. Fertil Steril 31:1, 1979 2. Hoult IJ, De Crespigny LCh, O'Herlihy C, Speirs AL, Lopata A, Kellow G, Johnston I, Robinson HP: Ultrasound control of clomiphenelhuman chorionic 'gonadotropin stimulated cycles for oocyte recovery and in vitro fertilization. Fertil Steril 36:316, 1981 3. Adashi EY, Rock JA, Sapp KC, Martin EJ, Wentz AC, Jones GS: Gestational outcome of clomiphene-related conceptions. Fertil Steril 31:620, 1979 4. Jansen RPS: Spontaneous abortion incidence in the treatment of infertility. Am J Obstet Gynecol 143:451, 1982 5 .. Lamb EJ, Collifiower WW, WilliamsJW: Endometrial histology and conception rates after clomiphene citrate. Obstet Gynecol 39:389, 1972 6. Talbert LM: Clomiphene citrate induction of ovulation. Fertil Steril 39:742, 1983 7: Laufer N, Navot D, Schenker JG: The pattern of luteal phase plasma progesterone and estradiol in fertile cycles. Am J Obstet Gynecol 143:808, 1982 8. Quigley MM, Berkowitz AS, Gilbert SA, WolfDP: Clomiphene citrate in an· in vitro fertilization program: hormonal comparison between 50- and 150-daily dosages. Fertil Steril 41:809, 1984 9. Parrini D, Facchinetti F, Genazzani A: Physiological variability of endocrine indices normally employed in. clinical practice. In The Human Placenta: Proteins and Hormones, Edited by A Klopper, A Genazzini, PG Crosignani. London, Academic Press, 1980, p 245 10. Kellie AE: The radioimmunoassay of steroid conjugates. J Steroid Biochem 6:277,1975 11. Weerasekera DA, Kim JB, Barnard GJ, Collins WP, Kohen F, Lindner HR: Monitoring ovarian function by a solid phase chemiluminescence immunoassay. Acta Endocrinol (Copenh) 101:254,1982
Strigini et aI. Urinary steroids in induced pregnancies
213
12. World Health Organization: Temporal relationships between indices of the fertile period. Fertil Steril 39:647, 1983 13. Mendizabal AF, Quiroga S, Farinati Z, Lahox M, Nagle C: Hormonal monitoring of early pregnancy by a direct radioimmunoassay of steroid glucuronides in first morning urine. Fertil Steril 42:737, 1984 14. Eshhar Z, Kim JB, Barnard GJ, Collinis WP, Gilard S, Lindner HR, Cohen F: Use of monoclonal antibodies to pregnanediol-3a-glucuronide to the development of a solid phase chemiluminescence immunoassay. Steroids 38:89,1981 15. Brockelbank JL, Kim JB, Barnard GJ, Collins WP, Gaier B, Kohen F: The measurement of urinary LH by a solidphase chemiluminescence immunoassay. Ann Clin Biochern 21:284, 1984 16. Collins WP, Collins PO, Kilpatrick MJ, Manning PA, Pike JM, Tyler JPP: The concentrations of urinary oestrone-3-glucuronide, LH and pregnanediol-3a-glucuronide as indices of ovarian function. Acta Endocrinol (Copenh) 90:336, 1979
214
8trigini et al.
Urinary steroids in induced pregnancies
17. Braunstein GO, Karow WG, Gentry WC, Rasor J, Wade ME: First-trimester chorionic gonadotropin measurements as an aid in the diagnosis of early pregnancy disorders. Am J Obstet Gynecol 131:25, 1978 18. Harrison RF, Kitchin Y: Maternal plasma unconjugated oestrogens in early human pregnancy. Br J Obstet Gynaecol 87:686, 1980 19. Harrison RF, Youssefnejadian E, Brodovcky H, Johnson M, Dewhurst J: Secretion patterns of plasma-progesterone, 17-hydroxyprogesterone and 20a-hydroxy-preg-4en-3-one in early normal pregnancy. Br J Obstet Gynaecol 85:921, 1978 20. O'Herlihy C, Pepperell RJ, Brown JB, Smith MA, Sandri L, McBain JC: Incremental clomiphene therapy: a new method for treating persistent anovulation. Obstet Gynecol 58:535, 1981 21. Adashi EY: Clomiphene citrate: mechanism(s) and site(s) of action-a hypothesis revisited. Fertil Steril 42:331, 1984
Fertility and Sterility