ORIGINAL E n d o c r i n e
ARTICLE R e s e a r c h
Intrafollicular Endocrine Milieu After Addition of hCG to Recombinant FSH During Controlled Ovarian Stimulation for In Vitro Fertilization L. L. Thuesen, A. Nyboe Andersen, A. Loft, and J. Smitz The Fertility Clinic (L.L.T., A.N.A., A.L.), Section 4071, Copenhagen University Hospital, Rigshospitalet, 2100 Copenhagen, Denmark; and Follicle Biology Laboratory (J.S.), Vrije Universiteit Brussel, UZ Brussel, 1090 Brussels, Belgium
Context: The role of human chorionic gonadotropin (hCG) supplementation on the intrafollicular steroid milieu has been studied. Objective: The objective of the study was to assess the impact on steroid levels in follicular fluids (FFs) after different doses of hCG supplementation to recombinant FSH for controlled ovarian stimulation. Setting: This was a prospective randomized dose-response study conducted at Copenhagen University Hospital, Rigshospitalet, Denmark. Patients: From 62 in vitro fertilization patients, 334 FFs were selected for analyses. Interventions: Patients were treated using a GnRH agonist protocol with recombinant FSH 150 IU/d and randomized from stimulation day 1 to supplementation with hCG: D0, 0 IU/d; D50, 50 IU/d; D100, 100 IU/d; and D150, 150 IU/d. Main Outcome Measure: Intrafollicular hormone concentrations in relation to treatment groups, follicular sizes, and embryo quality were measured. Results: In large follicles, hCG supplementation induced a nearly 3-fold increase of estradiol (nanomoles per liter) [D0: 1496; D50: 3138; D100: 4338; D150: 4009 (P ⬍ .001)], a significant 3-fold increase of androstenedione, and a 5-fold increase of T (nanomoles per liter) [D0: 15; D50: 38; D100: 72; D150: 56 (P ⬍ .001)]. The estradiol to T ratio decreased significantly, with the lowest ratio in D100 and the highest in D0. Large follicles giving rise to good-quality embryos had significantly higher estradiol and progesterone levels and estradiol to T, estradiol to androstenedione, and progesterone to estradiol ratios, compared with small follicles, leading to poor-quality embryos. Conclusions: Increasing doses of hCG supplementation markedly stimulated the intrafollicular concentration of both estradiol and androgens, with a shift toward a more androgenic milieu. In large follicles with oocytes giving rise to good-quality embryos, the FFs were significantly more estrogenic than in small follicles with oocytes developing into poor quality embryos. (J Clin Endocrinol Metab 99: 517–526, 2014)
D
uring follicular growth and maturation, steroids are synthesized in the theca and granulosa cells, accumulate in the follicular fluid (FFs), and give rise to the changes in peripheral blood (1). Additionally, the oocyte
and the surrounding cumulus cells may be influenced by both the absolute concentrations and the ratios of intrafollicular steroids, leading to maturation or postfertilization alterations (2).
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received February 28, 2013. Accepted October 15, 2013. First Published Online December 2, 2013
Abbreviations: A, androstenedione; CI, confidence interval; D, day; E2, estradiol; FF, follicular fluid; GQE, good-quality embryo; hCG, human chorionic gonadotropin; HP-hMG, highly purified human menopausal gonadotropin; IVF, in vitro fertilization; P, progesterone; PQE, poor-quality embryo; rFSH, recombinant FSH; TQE, top-quality embryo.
doi: 10.1210/jc.2013-1528
J Clin Endocrinol Metab, February 2014, 99(2):517–526
jcem.endojournals.org
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
517
518
Thuesen et al
hCG Addition and Follicular Endocrine Milieu
J Clin Endocrinol Metab, February 2014, 99(2):517–526
In response to LH, the thecal tissue stimulates the production of androgens, which in turn can be converted into estrogens in the granulosa cells through FSH-induced aromatization (3– 6). Results from large clinical trials have indicated that stimulation with gonadotropin preparations containing LH activity in the form of human chorionic gonadotropin (hCG), eg, highly purified human menopausal gonadotropin (HP-hMG) from the start of stimulation induces significantly different hormone profiles compared with stimulation with recombinant FSH (rFSH). The hCG component in HP-hMG induced higher levels of androgens and estrogens in both serum and FFs (7). Additionally, the augmentation of androgen levels has been hypothesized to be the underlying mechanism explaining a more selective follicle growth (7, 8). Although inducing a significantly lower number of follicles than in rFSH-only-stimulated cycles, the proportion of top-quality embryos was significantly higher with HP-hMG treatment. This observation was proposed to be causally related to the altered endocrine environment initiated by hCG (9). The intrafollicular estrogenic environment has been associated with good follicular growth and antiatresia effects. A high absolute estradiol concentration in FFs has been related to oocytes, giving rise to successful pregnancies (10), whereas other studies have indicated that the follicles with the highest estrogen to androgen ratios were more likely to contain a healthy oocyte (11–13). Previous studies on intrafollicular milieu claimed that an androgenic milieu could antagonize the estrogen-induced granulosa proliferation and, if sustained, could promote degenerative changes, finally resulting in follicular atresia (14, 15). The potential to effectively convert androgens to estradiol can be considered a marker of health of the follicle, and an inability to convert androgens to estradiol may represent an early atretic change (1, 7). Not only the androgens and estradiol reflect the legacy of follicular metabolism, but the same circumstance also probably applies to progesterone, reflecting a combination of legacy and new biosynthesis. An optimal exposure to progesterone has been suggested to have positive effects on oocyte characteristics, whereas an excessive exposure might lead to a rapid decrease in the quality of the cell (2). Our previously published clinical data showed that daily supplementation of hCG doses from 0 to 150 IU to rFSH throughout controlled ovarian stimulation in a long GnRH agonist protocol significantly increased the number of top-quality embryos and influenced serum androstenedione, T, progesterone, 17-hydroxyprogesterone, and estradiol significantly (16, 17). The aim of this exploratory analysis was to investigate changes in intrafollicular hormone concentrations from the same in vitro
fertilization (IVF) patients in relation to hCG dose, follicular sizes, and embryo quality.
Materials and Methods Subjects and protocol This was a randomized, dose-response study conducted at the Fertility Clinic, Copenhagen University Hospital (Rigshospitalet, Copenhagen, Denmark). The study group comprised of 62 standard patients scheduled for IVF with an age of 25–37 years, a body mass index greater than 18 and less than 30 kg/m2, and with a regular menstrual cycle. Early follicular phase serum FSH levels were 1–12 IU/L and early follicular phase total antral follicle (2–10 mm) count of 6 or greater. A detailed description of the study population along with the clinical outcome of the study has previously been published (16). The patients underwent controlled ovarian stimulation in a GnRH agonist protocol. All patients were treated in a fixed-dose regimen with 150 IU/d of rFSH (Puregon, N.V. Organon) and randomized to supplementation from day 1 of stimulation with a daily hCG dose: 1) day (D) 0, 0 IU/d of hCG; 2) D50, 50 IU/d of hCG; 3) D100, 100 IU/d of hCG; and 4) D150, 150 IU/d of hCG. To induce a final follicular maturation, 10 000 IU of hCG (Pregnyl, N.V. Organon) was administered. The study was approved by the Danish Medicines Agency (2612-3928, EudraCT number 2008-008355-42), the Danish National Committee on Biomedical Research Ethics (HB-2008146), and registered in ClinicalTrial.gov with a registration number of NCT00844311.
Follicular fluid samples Oocyte retrieval was scheduled 36 hours after hCG triggering. The retrieval procedure was performed using a single lumen needle (17-gauge and 30 cm long; Cook Medical) with manual aspiration of each single follicle by using a 20-mL syringe. The vaginal wall was not punctured repetitively to retrieve individual FFs, but the needle was flushed in between each follicle aspiration. Only FFs from follicles of approximately 12 mm and greater were aspirated. FFs from all follicles were collected individually, numbered, and the volume assessed. After the fluid was collected, the flushing procedure could be applied if appropriate. The number of the follicle and its corresponding oocyte were recorded consecutively. Thus, each oocyte could be unambiguously linked to its individual follicle. The FF was centrifuged at 400 ⫻ g in 8 minutes; supernatants were aliquoted and stored at ⫺24°C for subsequent analyses. Subsequently, IVF was performed. Embryo quality evaluation was blinded according to treatment and included assessment of blastomere number, degree of fragmentation, blastomere uniformity, and multinucleation. All oocytes were followed up individually. The transfer of preferably one embryo (in 90% of patients) was done on day 3 after oocyte retrieval. The main outcome measures were intrafollicular hormone concentrations in relation to treatment groups, follicular sizes, and embryo quality.
Follicular selection for analysis and embryo quality assessment The subset analyses and the numbers of the FFs are presented in Supplemental Figure 1, published on The Endocrine Society’s
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2013-1528
Journals Online web site at http://jcem.endojournals.org. From 59 retrievals, FFs from 816 follicles were collected and 536 follicles contained oocytes. In 14 follicles, two oocytes were found. Because the link between these oocytes and the follicles of origin was uncertain, they were excluded from further analysis. A subset of 334 follicles (40.9%) was selected for the measurement of endocrine parameters. Regarding oocyte/embryo selection, the selection of FFs for analysis was done in a systematic predefined way using oocyte/ aspiration record sheet (starting with the first aspirated follicle and so forth) to be able to include follicles of both small and large size as well as to choose follicles giving rise to a certain embryo quality.
jcem.endojournals.org
519
rise to a PQE (possible in 41 patients). For the embryo selection, the ranking criteria previously described were used.
Comparison based on pregnancy The FFs from follicles with an oocyte developing into a transferred embryo resulting in a live birth (16 embryos, four in each group) were compared with FFs from follicles with an oocyte that did not result in a live birth (35 embryos). All hCG groups were analyzed together.
FF hormonal determination
A small follicle was defined as 2.0 mL or less (equal to an average follicle diameter ⱕ15 mm) and a large follicle greater than 2.0 mL (⬎15 mm). The limit of 15 mm in follicular size was chosen because in previous studies using a long GnRH agonist protocol, meiotic maturation was maximal in the follicles with sizes greater than 15 mm (18 –20). A good-quality embryo (GQE) was defined as six or more blastomeres on day 3, 20% or less fragmentation on day 3, and no multinucleation. A top-quality embryo (TQE) was defined as four to five blastomeres on day 2, seven or more blastomeres on day 3, equally sized blastomeres and 20% or less fragmentation on day 3, and no multinucleation. To identify the best GQE or TQE, the following ranking criteria were used: 1) blastomere count day 3 (seven or more blastomeres better than less), 2) synchronization of the embryo from day 2 to day 3 (preferably four blastomeres on day 2 and eight blastomeres on day 3), and 3) blastomere size (equally sized blastomeres better than unequal). The following definition and ranking of a poor-quality embryo (PQE) was used: 1) not fertilized (not reached two pronuclei stage), 2) arrested oocyte/embryo (ie, lowest blastomere count on day 3), or 3) discarded embryo (due to low quality because of multinucleation or excessive fragmentation).
All hormone measurements in FFs were performed at the Laboratory for Hormonology and Tumormarkers, UZ Brussel (Brussels, Belgium). All FFs were visually inspected for eventual blood contamination; the impression of a faint or pink color led to the exclusion of follicles for hormone analyses. FF samples were measured undiluted for LH, FSH, and hCG, whereas T samples were diluted in multiassay diluent as recommended by the assay manufacturer. Estradiol (E2) and progesterone required a 1:1000 dilution in a multiassay diluent and androstenedione 1:400 in fetal calf serum. All media for dilution had previously been checked for linearity. A competitive RIA (coated tube referent KIP0451; DiaSource) was used for analysis of androstenedione. For all other hormones an electrochemiluminescence immunoassay (COBAS 6000; Roche Diagnostics) was used. The sensitivity and total imprecision (between-run and within-run variation) were as follows: hCG, 0.1 IU/L and less than 6%; FSH, 0.1 IU/L and less than 5%; LH, 0.1 IU/L and less than 5%; E2, 0.02 nmol/L and less than 6%; progesterone, 0.1 nmol/L and less than 4%; T, 0.087 nmol/L and less than 7%; and androstenedione, 0.1 nmol/L and 8%, respectively. The molar ratios of E2 to progesterone, E2 to androstenedione, E2 to T, and androstenedione to T were calculated to evaluate the conversion capacity of the major steroidogenic enzymes in the follicle.
Comparison based on follicle size
Statistical analysis
From each patient the first two small and the first two large follicles that contained an oocyte were selected (if available) for hormone analysis. A total of 194 follicles were analyzed (86 small and 108 large follicles).
The sample size calculation and randomization procedure were previously described (16). Data were expressed as mean and 95% confidence interval (CI). An ANOVA or Kruskal-Wallis test, where appropriate, was carried out for between-group analysis comparisons of continuous variables. Comparisons of continuous variables between two groups (pregnancy/nonpregnancy) were made by the independent-samples t test. Comparisons of proportions between the groups were made by a 2 test. A two-sided value of P ⬍ .05 was considered statistically significant. The Bonferroni correction was used for multiple comparison corrections. FF analyses were logarithmically transformed before the between-group analyses were done. The logarithmic means were transformed to the original scale; thus, the presented means are the geometric means with the corresponding CI. For the comparisons within the same patient, a paired-samples t test was performed. The statistical analysis of the data was performed in collaboration with statisticians affiliated with the Juliane Marie Center Rigshospitalet from the Biostatistics Department at Copenhagen University using the Statistical Package for the Social Sciences 19.0 software for Windows (SPSS Inc).
Definitions
Comparison based on embryo quality A total of 262 follicles were selected for the embryo quality comparisons. All the follicles containing an oocyte that developed into a GQE or a TQE were selected for FF analyses (n ⫽ 170) (42 small and 128 large follicles). The first small (n ⫽ 44) and the first large (n ⫽ 48) follicles with oocytes giving rise to PQEs were selected and compared with the endocrine data from follicles with similar diameter and with oocytes that gave rise to GQEs (selection criteria described previously).
Within-patient comparison based on embryo quality Within the same patient, the FF from the follicle containing an oocyte that developed into a transferred embryo was compared with FF from a follicle with similar diameter and an oocyte giving
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
520
Thuesen et al
hCG Addition and Follicular Endocrine Milieu
J Clin Endocrinol Metab, February 2014, 99(2):517–526
Table 1. Follicular Characteristics From All Follicles Aspirated at Oocyte Retrieval
All follicles aspirated All follicles aspirated (n ⫽ 816) Volume, mL Follicles with volume ⱕ2.0 mL, n, % Follicles with volume ⬎2.0 mL, n, % Oocytes, n,% (n ⫽ 550) All follicles containing an oocyte Follicles with oocytes (n ⫽ 536) Volume of follicles with oocytes, mL Poor-quality embryos, n, % Volume of follicles with poor-quality embryos, mL Good-quality embryos, n, % Volume of follicles with good-quality embryos, mL
Dose 0
Dose 50
Dose 100
Dose 150
P Value
(n ⫽ 219) 2.9 (2.6 –3.1) 82 (37) 137 (63) 146 (67)
(n ⫽ 198) 2.8 (2.6 –3.0) 70 (35) 128 (65) 124 (63)
(n ⫽ 212) 3.0 (2.7–3.2) 80 (38) 132 (62) 134 (63)
(n ⫽ 187) 2.9 (2.7–3.2) 67 (36) 120 (64) 146 (78)
.76 .97 .97 .02
(n ⫽ 140) 2.9 (2.6 –3.1) 93 (66) 2.8 (2.5–3.0) 47 (34) 2.9 (2.5–3.4)
(n ⫽ 124) 3.0 (2.7–3.3) 89 (72) 2.8 (2.5–3.2) 35 (28) 3.4 (2.8 –3.9)
(n ⫽ 131) 3.2 (2.8 –3.5) 75 (57) 2.7 (2.4 –3.0) 56 (43) 3.7 (3.1– 4.2)
(n ⫽ 141) 3.2 (2.9 –3.5) 96 (68) 2.9 (2.5–3.2) 45 (32) 3.8 (3.2– 4.3)
.30 .09 .87 .09 .10
Values are means (confidence intervals) or number (column percentage). ANOVA or 2 test is used. Poor-quality embryos are not fertilized/atretic/ discarded embryos; good-quality embryos are six or more blastomeres on day 3, 20% fragmentation on day 3, and no multinucleation.
Results FF volume distribution As shown in Table 1, 816 follicles were aspirated (D0: n ⫽ 219; D50: n ⫽ 198; D100: n ⫽ 212; and D150: n ⫽ 187). The mean volume of the follicles was 2.9 mL (95% CI 2.8 –3.0). The distribution of follicular volume
at retrieval as well as the number of small (ⱕ2.0 mL) and large (⬎2.0 mL) follicles was similar for all groups (Figure 1). When analyzing all groups together (n ⫽ 536), follicles that gave rise to PQEs and GQE were significantly different with mean volumes of 2.8 mL (95% CI 2.6 –3.0) vs 3.5 mL (95% CI 3.2–3.7) (P ⬍ .001), respectively.
Figure 1. The distribution of the volume of all follicles aspirated in the different treatment groups.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2013-1528
jcem.endojournals.org
521
Table 2. Follicular Fluid Concentrations at Oocyte Retrieval From the First Two Large (⬎2.0 mL/⬎15 mm) Follicles Aspirated (n ⫽ 194) Follicle Volume >2.0 mL (n ⴝ 108)
Dose 0 (n ⴝ 30)
Dose 50 (n ⴝ 27)
Dose 100 (n ⴝ 30)
Dose 150 (n ⴝ 21)
P Value
hCG, IU/L FSH, IU/L Estradiol, nmol/L Progesterone, nmol/L Androstenedione, nmol/L Total T, nmol/L Estradiol to progesterone ratio Estradiol to androstenedione ratio Estradiol to T ratio Androstenedione to T ratio
126 (111–142) 3.4 (3.1–3.7) 1496 (1279 –1750)50,100,150 42 533 (35 718 –50 647) 63.2 (49.4 – 81.0)50,100,150 14.8 (12.8 –17.1)50,100,150 0.035 (0.028 – 0.044)50,100,150 23.7 (18.2–30.7) 101.1 (92.3–110.7)100,150 4.3 (3.3–5.5)
145 (127–165) 4.0 (3.6 – 4.4) 3138 (2796 –3522)0,100 45 344 (37 907–54 241) 103.0 (69.7–152.3)0,100 37.5 (30.2– 46.7)0,100 0.069 (0.059 – 0.082)0,100 30.5 (21.5– 43.2) 83.6 (72.4 –96.5)100 2.7 (2.1–3.6)
132 (114 –152) 3.8 (3.4 – 4.1) 4 338 (3943– 4772)0,50 35 055 (29 033– 42 326) 207.9 (153.5–281.7)0,50 72.0 (60.0 – 86.6)0,50 0.124 (0.103– 0.148)0,50 20.9 (15.4 –28.2) 60.3 (51.6 –70.1)0,50 2.9 (2.3– 4.0)
145 (121–174) 3.9 (3.5– 4.3) 4 009 (3352– 4794)0 46 741 (38 113–57 322) 163.4 (97.0 –275.2)0 55.8 (39.2–79.5)0 0.086 (0.065– 0.114)0 24.5 (15.5–38.9) 71.8 (55.6 –92.8)0 3.0 (2.2– 4.0)
.39 .14 ⬍.001a .11 ⬍.001 ⬍.001a ⬍.001 .40 ⬍.001a .04
Values are means with 95% CI. ANOVA or nonparametric test was used. 0,50,100,150 a
Significant differences (P ⬍ .05) in Bonferroni-corrected post hoc tests.
Nonparametric test was used.
FF hormone concentrations For all the selected follicles (n ⫽ 334), hCG, FSH, LH, E2, progesterone, androstenedione, and T were analyzed, and the ratios of E2 to progesterone (P), E2 to androstenedione (A), E2 to T, and A to T were calculated. Because 73% of the samples had undetectable concentrations of LH (⬍0.1 IU/L), the analyses of LH were excluded from the between group analyses. Intrafollicular hormone concentrations in relation to follicular size Large follicles As shown in Table 2 and Figure 2, the concentration of estradiol increased significantly by nearly 3-fold, from 1496 nmol/L (95% CI 1279 –1750 nmol/L) in D0 to 4338 nmol/L (95% CI 3943– 4772 nmol/L) in D100 (P ⬍ .001). For androstenedione, a 3-fold increase was seen, from 63 nmol/L (95% CI 49 – 81 nmol/L) in D0 to 208 nmol/L (95% CI 154 –282 nmol/L) in D100 (P ⬍ .001), whereas for T, an increase of almost 5 times occurred, from 15 nmol/L (13–17 nmol/L) in D0 to 72 nmol/L (60 – 87 nmol/L) in D100 (P ⬍ .001). The hCG-induced increases in E2, androstenedione, and T leveled off, from D100 to D150. The highest E2 to P ratio was found in D100 and D150 (P ⬍ .001). No difference was seen for the E2 to A ratio, but a significant decrease was found for the E2 to T ratio with the lowest ratio in D100 and the highest in D0 (P ⬍ .001). A significant difference between the control group and the hCG dose groups was observed for the A to T ratio (P ⫽ .04). The level of hCG, FSH, LH, and P did not differ between the groups. Small follicles As shown in Table 3, the absolute concentrations of the steroids were lower in small than in large follicles. Hormone values and ratios in relation to the hCG dose followed the same pattern as for the large follicles but with a
larger variability, except regarding the E2 to T and A to T ratios, in which no differences were observed between the hCG dose groups. To circumvent the statistical concern of the independence of observations required for using ANOVA testing, the first large and the first small follicle from every single patient was performed (data not presented). The total amount of analyzed follicles was reduced from 194 to 108. The results from the more strict analysis were comparable with those in which the two first large and the two first small follicles were compared. Overall, the CI was extended, and only for the A to T ratio for the large follicles, a change from significant to not significant was observed. Intrafollicular hormone concentrations in relation to embryo quality Figure 3 and Supplemental Table 1 show the endocrine findings in 262 follicles in relation to either poor or good quality embryos and follicular sizes. Irrespective of the hCG dosing, the large follicles with an oocyte developing into a GQE had a significantly higher concentration of P, E2 together with the E2 to A ratio, and the E2 to T ratio as well as a lower E2 to P ratio compared with the other embryo quality groups. No differences were observed for androstenedione and T. When evaluating the hormones in relation to the four different treatment groups, the patterns observed were similar, however, with the highest absolute hormone concentrations found in D100 and D150 (Supplemental Table 1). Within-patient analysis of intrafollicular hormone concentrations in relation to embryo quality Table 4 shows the within-patient comparison of FF hormone concentrations. P and the E2 to T ratio were significantly higher and the E2 to P ratio significantly lower from follicles with an oocyte, giving rise to the best embryo (transferred embryo) compared with a follicle of a similar size but yielding an oocyte developing into a PQE.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
522
Thuesen et al
hCG Addition and Follicular Endocrine Milieu
60000
J Clin Endocrinol Metab, February 2014, 99(2):517–526
5000
Progesterone
40000 30000
1000 Dose 50 Dose 100 Dose 150
Dose 0 100
Androstenedione**
Testosterone**
60 40 20 0
Dose 0
Dose 50
Dose 0
Dose 100 Dose 150 50
E2 / Progesterone**
Dose 50
Dose 100 Dose 150
E2 / Androstenedione
40
0.12
30
0.08
20
0.04
10
0
0 Dose 0
120
Dose 50 Dose 100 Dose 150
80 nmol/l
nmol/l
Dose 0
0.16
3000 2000
20000
300 250 200 150 100 50 0
Estradiol**
4000 nmol/l
nmol/l
50000
Dose 50
Dose 100
Dose 150 6 5 4 3 2 1 0
E2 / Testosterone**
100 80 60 40 Dose 0
Dose 50
Dose 100
Dose 0
Dose 150
Dose 50
Dose 100
Dose 150
Androstenedione/Testosterone*
Dose 0
Dose 50
Dose 100
Dose 150
Figure 2. Endocrine levels in FF from the first two large follicles aspirated at oocyte retrieval in the different hCG dose groups. Values are means with 95% CI. Statistical significance (*, P ⬍ .05 and **, P ⬍ .001) between the groups using ANOVA.
Intrafollicular hormone concentrations in relation to pregnancy Supplemental Table 2 shows the comparison of the FF from follicles with an oocyte developing into a transferred embryo resulting in a live birth compared with FF from follicles with an oocyte not resulting in a live birth. For the A to T ratio, a significant difference was found.
Discussion The present study is the first to investigate changes in intrafollicular hormone concentrations from IVF patients receiving increasing doses of hCG supplementation to rFSH throughout controlled ovarian stimulation in a long GnRH agonist protocol. Previously published clinical
Table 3. Follicular Fluid Concentrations at Oocyte Retrieval From the First Two Small (ⱕ2.0 mL/ⱕ15 mm) Follicles Aspirated (n ⫽ 194) Follicle Volume
ARTICLE R e s e a r c h
Intrafollicular Endocrine Milieu After Addition of hCG to Recombinant FSH During Controlled Ovarian Stimulation for In Vitro Fertilization L. L. Thuesen, A. Nyboe Andersen, A. Loft, and J. Smitz The Fertility Clinic (L.L.T., A.N.A., A.L.), Section 4071, Copenhagen University Hospital, Rigshospitalet, 2100 Copenhagen, Denmark; and Follicle Biology Laboratory (J.S.), Vrije Universiteit Brussel, UZ Brussel, 1090 Brussels, Belgium
Context: The role of human chorionic gonadotropin (hCG) supplementation on the intrafollicular steroid milieu has been studied. Objective: The objective of the study was to assess the impact on steroid levels in follicular fluids (FFs) after different doses of hCG supplementation to recombinant FSH for controlled ovarian stimulation. Setting: This was a prospective randomized dose-response study conducted at Copenhagen University Hospital, Rigshospitalet, Denmark. Patients: From 62 in vitro fertilization patients, 334 FFs were selected for analyses. Interventions: Patients were treated using a GnRH agonist protocol with recombinant FSH 150 IU/d and randomized from stimulation day 1 to supplementation with hCG: D0, 0 IU/d; D50, 50 IU/d; D100, 100 IU/d; and D150, 150 IU/d. Main Outcome Measure: Intrafollicular hormone concentrations in relation to treatment groups, follicular sizes, and embryo quality were measured. Results: In large follicles, hCG supplementation induced a nearly 3-fold increase of estradiol (nanomoles per liter) [D0: 1496; D50: 3138; D100: 4338; D150: 4009 (P ⬍ .001)], a significant 3-fold increase of androstenedione, and a 5-fold increase of T (nanomoles per liter) [D0: 15; D50: 38; D100: 72; D150: 56 (P ⬍ .001)]. The estradiol to T ratio decreased significantly, with the lowest ratio in D100 and the highest in D0. Large follicles giving rise to good-quality embryos had significantly higher estradiol and progesterone levels and estradiol to T, estradiol to androstenedione, and progesterone to estradiol ratios, compared with small follicles, leading to poor-quality embryos. Conclusions: Increasing doses of hCG supplementation markedly stimulated the intrafollicular concentration of both estradiol and androgens, with a shift toward a more androgenic milieu. In large follicles with oocytes giving rise to good-quality embryos, the FFs were significantly more estrogenic than in small follicles with oocytes developing into poor quality embryos. (J Clin Endocrinol Metab 99: 517–526, 2014)
D
uring follicular growth and maturation, steroids are synthesized in the theca and granulosa cells, accumulate in the follicular fluid (FFs), and give rise to the changes in peripheral blood (1). Additionally, the oocyte
and the surrounding cumulus cells may be influenced by both the absolute concentrations and the ratios of intrafollicular steroids, leading to maturation or postfertilization alterations (2).
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received February 28, 2013. Accepted October 15, 2013. First Published Online December 2, 2013
Abbreviations: A, androstenedione; CI, confidence interval; D, day; E2, estradiol; FF, follicular fluid; GQE, good-quality embryo; hCG, human chorionic gonadotropin; HP-hMG, highly purified human menopausal gonadotropin; IVF, in vitro fertilization; P, progesterone; PQE, poor-quality embryo; rFSH, recombinant FSH; TQE, top-quality embryo.
doi: 10.1210/jc.2013-1528
J Clin Endocrinol Metab, February 2014, 99(2):517–526
jcem.endojournals.org
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
517
518
Thuesen et al
hCG Addition and Follicular Endocrine Milieu
J Clin Endocrinol Metab, February 2014, 99(2):517–526
In response to LH, the thecal tissue stimulates the production of androgens, which in turn can be converted into estrogens in the granulosa cells through FSH-induced aromatization (3– 6). Results from large clinical trials have indicated that stimulation with gonadotropin preparations containing LH activity in the form of human chorionic gonadotropin (hCG), eg, highly purified human menopausal gonadotropin (HP-hMG) from the start of stimulation induces significantly different hormone profiles compared with stimulation with recombinant FSH (rFSH). The hCG component in HP-hMG induced higher levels of androgens and estrogens in both serum and FFs (7). Additionally, the augmentation of androgen levels has been hypothesized to be the underlying mechanism explaining a more selective follicle growth (7, 8). Although inducing a significantly lower number of follicles than in rFSH-only-stimulated cycles, the proportion of top-quality embryos was significantly higher with HP-hMG treatment. This observation was proposed to be causally related to the altered endocrine environment initiated by hCG (9). The intrafollicular estrogenic environment has been associated with good follicular growth and antiatresia effects. A high absolute estradiol concentration in FFs has been related to oocytes, giving rise to successful pregnancies (10), whereas other studies have indicated that the follicles with the highest estrogen to androgen ratios were more likely to contain a healthy oocyte (11–13). Previous studies on intrafollicular milieu claimed that an androgenic milieu could antagonize the estrogen-induced granulosa proliferation and, if sustained, could promote degenerative changes, finally resulting in follicular atresia (14, 15). The potential to effectively convert androgens to estradiol can be considered a marker of health of the follicle, and an inability to convert androgens to estradiol may represent an early atretic change (1, 7). Not only the androgens and estradiol reflect the legacy of follicular metabolism, but the same circumstance also probably applies to progesterone, reflecting a combination of legacy and new biosynthesis. An optimal exposure to progesterone has been suggested to have positive effects on oocyte characteristics, whereas an excessive exposure might lead to a rapid decrease in the quality of the cell (2). Our previously published clinical data showed that daily supplementation of hCG doses from 0 to 150 IU to rFSH throughout controlled ovarian stimulation in a long GnRH agonist protocol significantly increased the number of top-quality embryos and influenced serum androstenedione, T, progesterone, 17-hydroxyprogesterone, and estradiol significantly (16, 17). The aim of this exploratory analysis was to investigate changes in intrafollicular hormone concentrations from the same in vitro
fertilization (IVF) patients in relation to hCG dose, follicular sizes, and embryo quality.
Materials and Methods Subjects and protocol This was a randomized, dose-response study conducted at the Fertility Clinic, Copenhagen University Hospital (Rigshospitalet, Copenhagen, Denmark). The study group comprised of 62 standard patients scheduled for IVF with an age of 25–37 years, a body mass index greater than 18 and less than 30 kg/m2, and with a regular menstrual cycle. Early follicular phase serum FSH levels were 1–12 IU/L and early follicular phase total antral follicle (2–10 mm) count of 6 or greater. A detailed description of the study population along with the clinical outcome of the study has previously been published (16). The patients underwent controlled ovarian stimulation in a GnRH agonist protocol. All patients were treated in a fixed-dose regimen with 150 IU/d of rFSH (Puregon, N.V. Organon) and randomized to supplementation from day 1 of stimulation with a daily hCG dose: 1) day (D) 0, 0 IU/d of hCG; 2) D50, 50 IU/d of hCG; 3) D100, 100 IU/d of hCG; and 4) D150, 150 IU/d of hCG. To induce a final follicular maturation, 10 000 IU of hCG (Pregnyl, N.V. Organon) was administered. The study was approved by the Danish Medicines Agency (2612-3928, EudraCT number 2008-008355-42), the Danish National Committee on Biomedical Research Ethics (HB-2008146), and registered in ClinicalTrial.gov with a registration number of NCT00844311.
Follicular fluid samples Oocyte retrieval was scheduled 36 hours after hCG triggering. The retrieval procedure was performed using a single lumen needle (17-gauge and 30 cm long; Cook Medical) with manual aspiration of each single follicle by using a 20-mL syringe. The vaginal wall was not punctured repetitively to retrieve individual FFs, but the needle was flushed in between each follicle aspiration. Only FFs from follicles of approximately 12 mm and greater were aspirated. FFs from all follicles were collected individually, numbered, and the volume assessed. After the fluid was collected, the flushing procedure could be applied if appropriate. The number of the follicle and its corresponding oocyte were recorded consecutively. Thus, each oocyte could be unambiguously linked to its individual follicle. The FF was centrifuged at 400 ⫻ g in 8 minutes; supernatants were aliquoted and stored at ⫺24°C for subsequent analyses. Subsequently, IVF was performed. Embryo quality evaluation was blinded according to treatment and included assessment of blastomere number, degree of fragmentation, blastomere uniformity, and multinucleation. All oocytes were followed up individually. The transfer of preferably one embryo (in 90% of patients) was done on day 3 after oocyte retrieval. The main outcome measures were intrafollicular hormone concentrations in relation to treatment groups, follicular sizes, and embryo quality.
Follicular selection for analysis and embryo quality assessment The subset analyses and the numbers of the FFs are presented in Supplemental Figure 1, published on The Endocrine Society’s
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2013-1528
Journals Online web site at http://jcem.endojournals.org. From 59 retrievals, FFs from 816 follicles were collected and 536 follicles contained oocytes. In 14 follicles, two oocytes were found. Because the link between these oocytes and the follicles of origin was uncertain, they were excluded from further analysis. A subset of 334 follicles (40.9%) was selected for the measurement of endocrine parameters. Regarding oocyte/embryo selection, the selection of FFs for analysis was done in a systematic predefined way using oocyte/ aspiration record sheet (starting with the first aspirated follicle and so forth) to be able to include follicles of both small and large size as well as to choose follicles giving rise to a certain embryo quality.
jcem.endojournals.org
519
rise to a PQE (possible in 41 patients). For the embryo selection, the ranking criteria previously described were used.
Comparison based on pregnancy The FFs from follicles with an oocyte developing into a transferred embryo resulting in a live birth (16 embryos, four in each group) were compared with FFs from follicles with an oocyte that did not result in a live birth (35 embryos). All hCG groups were analyzed together.
FF hormonal determination
A small follicle was defined as 2.0 mL or less (equal to an average follicle diameter ⱕ15 mm) and a large follicle greater than 2.0 mL (⬎15 mm). The limit of 15 mm in follicular size was chosen because in previous studies using a long GnRH agonist protocol, meiotic maturation was maximal in the follicles with sizes greater than 15 mm (18 –20). A good-quality embryo (GQE) was defined as six or more blastomeres on day 3, 20% or less fragmentation on day 3, and no multinucleation. A top-quality embryo (TQE) was defined as four to five blastomeres on day 2, seven or more blastomeres on day 3, equally sized blastomeres and 20% or less fragmentation on day 3, and no multinucleation. To identify the best GQE or TQE, the following ranking criteria were used: 1) blastomere count day 3 (seven or more blastomeres better than less), 2) synchronization of the embryo from day 2 to day 3 (preferably four blastomeres on day 2 and eight blastomeres on day 3), and 3) blastomere size (equally sized blastomeres better than unequal). The following definition and ranking of a poor-quality embryo (PQE) was used: 1) not fertilized (not reached two pronuclei stage), 2) arrested oocyte/embryo (ie, lowest blastomere count on day 3), or 3) discarded embryo (due to low quality because of multinucleation or excessive fragmentation).
All hormone measurements in FFs were performed at the Laboratory for Hormonology and Tumormarkers, UZ Brussel (Brussels, Belgium). All FFs were visually inspected for eventual blood contamination; the impression of a faint or pink color led to the exclusion of follicles for hormone analyses. FF samples were measured undiluted for LH, FSH, and hCG, whereas T samples were diluted in multiassay diluent as recommended by the assay manufacturer. Estradiol (E2) and progesterone required a 1:1000 dilution in a multiassay diluent and androstenedione 1:400 in fetal calf serum. All media for dilution had previously been checked for linearity. A competitive RIA (coated tube referent KIP0451; DiaSource) was used for analysis of androstenedione. For all other hormones an electrochemiluminescence immunoassay (COBAS 6000; Roche Diagnostics) was used. The sensitivity and total imprecision (between-run and within-run variation) were as follows: hCG, 0.1 IU/L and less than 6%; FSH, 0.1 IU/L and less than 5%; LH, 0.1 IU/L and less than 5%; E2, 0.02 nmol/L and less than 6%; progesterone, 0.1 nmol/L and less than 4%; T, 0.087 nmol/L and less than 7%; and androstenedione, 0.1 nmol/L and 8%, respectively. The molar ratios of E2 to progesterone, E2 to androstenedione, E2 to T, and androstenedione to T were calculated to evaluate the conversion capacity of the major steroidogenic enzymes in the follicle.
Comparison based on follicle size
Statistical analysis
From each patient the first two small and the first two large follicles that contained an oocyte were selected (if available) for hormone analysis. A total of 194 follicles were analyzed (86 small and 108 large follicles).
The sample size calculation and randomization procedure were previously described (16). Data were expressed as mean and 95% confidence interval (CI). An ANOVA or Kruskal-Wallis test, where appropriate, was carried out for between-group analysis comparisons of continuous variables. Comparisons of continuous variables between two groups (pregnancy/nonpregnancy) were made by the independent-samples t test. Comparisons of proportions between the groups were made by a 2 test. A two-sided value of P ⬍ .05 was considered statistically significant. The Bonferroni correction was used for multiple comparison corrections. FF analyses were logarithmically transformed before the between-group analyses were done. The logarithmic means were transformed to the original scale; thus, the presented means are the geometric means with the corresponding CI. For the comparisons within the same patient, a paired-samples t test was performed. The statistical analysis of the data was performed in collaboration with statisticians affiliated with the Juliane Marie Center Rigshospitalet from the Biostatistics Department at Copenhagen University using the Statistical Package for the Social Sciences 19.0 software for Windows (SPSS Inc).
Definitions
Comparison based on embryo quality A total of 262 follicles were selected for the embryo quality comparisons. All the follicles containing an oocyte that developed into a GQE or a TQE were selected for FF analyses (n ⫽ 170) (42 small and 128 large follicles). The first small (n ⫽ 44) and the first large (n ⫽ 48) follicles with oocytes giving rise to PQEs were selected and compared with the endocrine data from follicles with similar diameter and with oocytes that gave rise to GQEs (selection criteria described previously).
Within-patient comparison based on embryo quality Within the same patient, the FF from the follicle containing an oocyte that developed into a transferred embryo was compared with FF from a follicle with similar diameter and an oocyte giving
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
520
Thuesen et al
hCG Addition and Follicular Endocrine Milieu
J Clin Endocrinol Metab, February 2014, 99(2):517–526
Table 1. Follicular Characteristics From All Follicles Aspirated at Oocyte Retrieval
All follicles aspirated All follicles aspirated (n ⫽ 816) Volume, mL Follicles with volume ⱕ2.0 mL, n, % Follicles with volume ⬎2.0 mL, n, % Oocytes, n,% (n ⫽ 550) All follicles containing an oocyte Follicles with oocytes (n ⫽ 536) Volume of follicles with oocytes, mL Poor-quality embryos, n, % Volume of follicles with poor-quality embryos, mL Good-quality embryos, n, % Volume of follicles with good-quality embryos, mL
Dose 0
Dose 50
Dose 100
Dose 150
P Value
(n ⫽ 219) 2.9 (2.6 –3.1) 82 (37) 137 (63) 146 (67)
(n ⫽ 198) 2.8 (2.6 –3.0) 70 (35) 128 (65) 124 (63)
(n ⫽ 212) 3.0 (2.7–3.2) 80 (38) 132 (62) 134 (63)
(n ⫽ 187) 2.9 (2.7–3.2) 67 (36) 120 (64) 146 (78)
.76 .97 .97 .02
(n ⫽ 140) 2.9 (2.6 –3.1) 93 (66) 2.8 (2.5–3.0) 47 (34) 2.9 (2.5–3.4)
(n ⫽ 124) 3.0 (2.7–3.3) 89 (72) 2.8 (2.5–3.2) 35 (28) 3.4 (2.8 –3.9)
(n ⫽ 131) 3.2 (2.8 –3.5) 75 (57) 2.7 (2.4 –3.0) 56 (43) 3.7 (3.1– 4.2)
(n ⫽ 141) 3.2 (2.9 –3.5) 96 (68) 2.9 (2.5–3.2) 45 (32) 3.8 (3.2– 4.3)
.30 .09 .87 .09 .10
Values are means (confidence intervals) or number (column percentage). ANOVA or 2 test is used. Poor-quality embryos are not fertilized/atretic/ discarded embryos; good-quality embryos are six or more blastomeres on day 3, 20% fragmentation on day 3, and no multinucleation.
Results FF volume distribution As shown in Table 1, 816 follicles were aspirated (D0: n ⫽ 219; D50: n ⫽ 198; D100: n ⫽ 212; and D150: n ⫽ 187). The mean volume of the follicles was 2.9 mL (95% CI 2.8 –3.0). The distribution of follicular volume
at retrieval as well as the number of small (ⱕ2.0 mL) and large (⬎2.0 mL) follicles was similar for all groups (Figure 1). When analyzing all groups together (n ⫽ 536), follicles that gave rise to PQEs and GQE were significantly different with mean volumes of 2.8 mL (95% CI 2.6 –3.0) vs 3.5 mL (95% CI 3.2–3.7) (P ⬍ .001), respectively.
Figure 1. The distribution of the volume of all follicles aspirated in the different treatment groups.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2013-1528
jcem.endojournals.org
521
Table 2. Follicular Fluid Concentrations at Oocyte Retrieval From the First Two Large (⬎2.0 mL/⬎15 mm) Follicles Aspirated (n ⫽ 194) Follicle Volume >2.0 mL (n ⴝ 108)
Dose 0 (n ⴝ 30)
Dose 50 (n ⴝ 27)
Dose 100 (n ⴝ 30)
Dose 150 (n ⴝ 21)
P Value
hCG, IU/L FSH, IU/L Estradiol, nmol/L Progesterone, nmol/L Androstenedione, nmol/L Total T, nmol/L Estradiol to progesterone ratio Estradiol to androstenedione ratio Estradiol to T ratio Androstenedione to T ratio
126 (111–142) 3.4 (3.1–3.7) 1496 (1279 –1750)50,100,150 42 533 (35 718 –50 647) 63.2 (49.4 – 81.0)50,100,150 14.8 (12.8 –17.1)50,100,150 0.035 (0.028 – 0.044)50,100,150 23.7 (18.2–30.7) 101.1 (92.3–110.7)100,150 4.3 (3.3–5.5)
145 (127–165) 4.0 (3.6 – 4.4) 3138 (2796 –3522)0,100 45 344 (37 907–54 241) 103.0 (69.7–152.3)0,100 37.5 (30.2– 46.7)0,100 0.069 (0.059 – 0.082)0,100 30.5 (21.5– 43.2) 83.6 (72.4 –96.5)100 2.7 (2.1–3.6)
132 (114 –152) 3.8 (3.4 – 4.1) 4 338 (3943– 4772)0,50 35 055 (29 033– 42 326) 207.9 (153.5–281.7)0,50 72.0 (60.0 – 86.6)0,50 0.124 (0.103– 0.148)0,50 20.9 (15.4 –28.2) 60.3 (51.6 –70.1)0,50 2.9 (2.3– 4.0)
145 (121–174) 3.9 (3.5– 4.3) 4 009 (3352– 4794)0 46 741 (38 113–57 322) 163.4 (97.0 –275.2)0 55.8 (39.2–79.5)0 0.086 (0.065– 0.114)0 24.5 (15.5–38.9) 71.8 (55.6 –92.8)0 3.0 (2.2– 4.0)
.39 .14 ⬍.001a .11 ⬍.001 ⬍.001a ⬍.001 .40 ⬍.001a .04
Values are means with 95% CI. ANOVA or nonparametric test was used. 0,50,100,150 a
Significant differences (P ⬍ .05) in Bonferroni-corrected post hoc tests.
Nonparametric test was used.
FF hormone concentrations For all the selected follicles (n ⫽ 334), hCG, FSH, LH, E2, progesterone, androstenedione, and T were analyzed, and the ratios of E2 to progesterone (P), E2 to androstenedione (A), E2 to T, and A to T were calculated. Because 73% of the samples had undetectable concentrations of LH (⬍0.1 IU/L), the analyses of LH were excluded from the between group analyses. Intrafollicular hormone concentrations in relation to follicular size Large follicles As shown in Table 2 and Figure 2, the concentration of estradiol increased significantly by nearly 3-fold, from 1496 nmol/L (95% CI 1279 –1750 nmol/L) in D0 to 4338 nmol/L (95% CI 3943– 4772 nmol/L) in D100 (P ⬍ .001). For androstenedione, a 3-fold increase was seen, from 63 nmol/L (95% CI 49 – 81 nmol/L) in D0 to 208 nmol/L (95% CI 154 –282 nmol/L) in D100 (P ⬍ .001), whereas for T, an increase of almost 5 times occurred, from 15 nmol/L (13–17 nmol/L) in D0 to 72 nmol/L (60 – 87 nmol/L) in D100 (P ⬍ .001). The hCG-induced increases in E2, androstenedione, and T leveled off, from D100 to D150. The highest E2 to P ratio was found in D100 and D150 (P ⬍ .001). No difference was seen for the E2 to A ratio, but a significant decrease was found for the E2 to T ratio with the lowest ratio in D100 and the highest in D0 (P ⬍ .001). A significant difference between the control group and the hCG dose groups was observed for the A to T ratio (P ⫽ .04). The level of hCG, FSH, LH, and P did not differ between the groups. Small follicles As shown in Table 3, the absolute concentrations of the steroids were lower in small than in large follicles. Hormone values and ratios in relation to the hCG dose followed the same pattern as for the large follicles but with a
larger variability, except regarding the E2 to T and A to T ratios, in which no differences were observed between the hCG dose groups. To circumvent the statistical concern of the independence of observations required for using ANOVA testing, the first large and the first small follicle from every single patient was performed (data not presented). The total amount of analyzed follicles was reduced from 194 to 108. The results from the more strict analysis were comparable with those in which the two first large and the two first small follicles were compared. Overall, the CI was extended, and only for the A to T ratio for the large follicles, a change from significant to not significant was observed. Intrafollicular hormone concentrations in relation to embryo quality Figure 3 and Supplemental Table 1 show the endocrine findings in 262 follicles in relation to either poor or good quality embryos and follicular sizes. Irrespective of the hCG dosing, the large follicles with an oocyte developing into a GQE had a significantly higher concentration of P, E2 together with the E2 to A ratio, and the E2 to T ratio as well as a lower E2 to P ratio compared with the other embryo quality groups. No differences were observed for androstenedione and T. When evaluating the hormones in relation to the four different treatment groups, the patterns observed were similar, however, with the highest absolute hormone concentrations found in D100 and D150 (Supplemental Table 1). Within-patient analysis of intrafollicular hormone concentrations in relation to embryo quality Table 4 shows the within-patient comparison of FF hormone concentrations. P and the E2 to T ratio were significantly higher and the E2 to P ratio significantly lower from follicles with an oocyte, giving rise to the best embryo (transferred embryo) compared with a follicle of a similar size but yielding an oocyte developing into a PQE.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 May 2014. at 13:57 For personal use only. No other uses without permission. . All rights reserved.
522
Thuesen et al
hCG Addition and Follicular Endocrine Milieu
60000
J Clin Endocrinol Metab, February 2014, 99(2):517–526
5000
Progesterone
40000 30000
1000 Dose 50 Dose 100 Dose 150
Dose 0 100
Androstenedione**
Testosterone**
60 40 20 0
Dose 0
Dose 50
Dose 0
Dose 100 Dose 150 50
E2 / Progesterone**
Dose 50
Dose 100 Dose 150
E2 / Androstenedione
40
0.12
30
0.08
20
0.04
10
0
0 Dose 0
120
Dose 50 Dose 100 Dose 150
80 nmol/l
nmol/l
Dose 0
0.16
3000 2000
20000
300 250 200 150 100 50 0
Estradiol**
4000 nmol/l
nmol/l
50000
Dose 50
Dose 100
Dose 150 6 5 4 3 2 1 0
E2 / Testosterone**
100 80 60 40 Dose 0
Dose 50
Dose 100
Dose 0
Dose 150
Dose 50
Dose 100
Dose 150
Androstenedione/Testosterone*
Dose 0
Dose 50
Dose 100
Dose 150
Figure 2. Endocrine levels in FF from the first two large follicles aspirated at oocyte retrieval in the different hCG dose groups. Values are means with 95% CI. Statistical significance (*, P ⬍ .05 and **, P ⬍ .001) between the groups using ANOVA.
Intrafollicular hormone concentrations in relation to pregnancy Supplemental Table 2 shows the comparison of the FF from follicles with an oocyte developing into a transferred embryo resulting in a live birth compared with FF from follicles with an oocyte not resulting in a live birth. For the A to T ratio, a significant difference was found.
Discussion The present study is the first to investigate changes in intrafollicular hormone concentrations from IVF patients receiving increasing doses of hCG supplementation to rFSH throughout controlled ovarian stimulation in a long GnRH agonist protocol. Previously published clinical
Table 3. Follicular Fluid Concentrations at Oocyte Retrieval From the First Two Small (ⱕ2.0 mL/ⱕ15 mm) Follicles Aspirated (n ⫽ 194) Follicle Volume
