ORIGINAL ARTICLE
Sonographic evaluation of the endometrium in in vitro fertilization IVF cycles A way to predict pregnancy? CHRISTINA BERGH, TORWORN HILLENSJO AND LARSNILSSON From the Department of Obstetrics and Gynecology, University of Gbteborg, Sahlgrenska Hospital, Gbteborg,
Sweden
Acta Obster Gynecol Scand 1992; 11: 624-628
The endometrial thickness and echogenic pattern were prospectively evaluated in 100 patients undergoing IVF-treatment. On the day before oocyte aspiration (day-1) the endometrium was significantly thicker in the group of patients achieving an ongoing pregnancy than in the group that failed to conceive. The minimal endometrial thickness on day-1 compatible with an ongoing pregnancy was 9 mm. Before oocyte aspiration two different endometrial patterns could be distinguished, a multilayered, ‘triple line’, hypoechogenic endometrium (A) and an isoechogenic pattern (B). Other studies have, in addition. described a hyperechogenic endometrium with a poor pregnancy rate but that pattern could in the present study only be demonstrated after follicle aspiration. 32 clinical pregnancies occurred. Among patients with endometrial pattern A, 36% conceived and among patients with pattern B 23% conceived which was not significantly different. It is concluded that the day before aspiration an endometrial thickness of at least 9 mm appears to be required for an ongoing pregnancy. In all cases a triple line (hypo-or isoechogenic) pattern was evident before aspiration. Key words: in vitro fertilization. transvaginal sonography. endometrium
Submitted February 24, 1992 Accepted May 20, 1992
The two most important variables giving success in in vitro fertilizatiodembryo transfer (IVFIET) are believed to be embryo quality and uterine receptivity. There have been several studies showing a correlation between embryo quality and implantation rate (1) but it has been more difficult to evaluate uterine receptivity. The use of endovaginal ultrasound to measure endometrial thickness and endometrial pattern has produced divergent results. Measuring endometrial thickness alone has not generally been successful in predicting pregnancy (2). A combination of assessing the type of endometrial reflectivity as well as its thickness appeared from a few studies to be a possible way to predict implantation (3, 4). The aim of the present study was to evaluate if a combination of the endometrial thickness, the endo@ Aria Ohster Gynecol Scand 71 (1992)
metrial growth and the pattern of endometrial echo, as studied by vaginal sonography, could be used to predict pregnancy. Special interest was focused on the appearance of the endometrium on the day before follicle aspiration to allow a comparison with earlier studies (3).
Patlents and methods The endometrial thickness and echogenic pattern were studied prospectively during gonadotropin stimulation, follicular aspiration and embryo transfer in 114 consecutive patients undergoing IVFtreatment in the IVF-unit, Sahlgren’s Hospital, GO-
--
20
3
-5
OnOdngpngnrM rpont-
16
Endometrial sonography in IVF
625
andor more than 15 oocytes retrieved).'llventythree patients received progesterone and 77 patients received hCG. The number of pregnancies in these groups were 8 and 25, respectively.
12 c
0
H
6
z9
4
0
Cycle monitoring
7
6
0
1 0 1 1 1 2 13 1 4
1s 16
Endometrld thickneu (mm) Fig. 1. Endometrial thickness on the day before oocyte retrieval (day-1)in the different IVF outcome groups.
teborg, Sweden. Inclusion criteria were tubal infertility of at least 3 years' duration, age below 38 years (range 25-38 years) and normal sperm parameters. Exclusions were made at embryo transfer if fewer than three embryos of presumed good quality (1) were present. One hundred and fourteen patients were initially recruited but fourteen of these were excluded at the time of embryo transfer when it was evident that there were less than 3 high-quality embryos available for transfer. These 14 patients excluded will only be used for comparative reasons in the 'Discussion'. Of the 100 patients still in the study 89 had been down regulated for at least 3 weeks with a GnRH-agonist (Buserelin; Suprefact@, Hoechst, Germany) 1,2 mg/d (nasal administration) before stimulation with human menopausal gonadotropin (hMG, Pergonal", Serono. Italy) 15Cb300 IU/day. Eleven women were stimulated with clorniphene citrate (Clomivid", Draco, Sweden) 100 mg/day, cycle day 4-8 and hMG, 150-225 IU/day, from cycle day 7. Human chorionic gonadotropin (hCG; Profasia, Serono, Italy), usually 10,OOO IU, was administered when 3 or more follicles with a mean diameter of at least 18 mm were identified by vaginal sonography and a continuous rise in serumestradiol had been measured for at least 7 days. Oocytes were retrieved 36 hours after hCG using transvaginal ultrasound guided puncture ( 5 ) . In vitro culture and insemination followed standard techniques. Embryo transfer took place two days after wcyte retrieval using a Frydman TDT catheter and a full urinary bladder. Abdominal ultrasound was used to verify the location of the tip of the catheter and the fluid echo in the cavity. Luteal support was given from the day of ET to all women, either with hCG (1250 IU every 3 days until the 12th day) or, in the case of impending hyperstimulation, progesterone (25 mg as intramuscular injection twice daily). Impending hyperstimulation was based on clinical judgements (e.g. estradiol levels above lhmoyl
The response to hormonal treatment was monitored by serum estradiol levels (assayed by standard RIA) and by transvaginal ultrasound measurements of follicles and endometrium (Ultrasound Scanner 1846 with a 7 MHz vaginal, mechanical sector transducer, Bruel & Kjaer, Denmark). Each measurement of the endometrium was made by two or three different persons. In case of disagreement a second measurement was made. Endometrial pattern, thickness and growth
The endometrium was scanned during the stimulation, starting from 3-4 days before oocyte aspiration and until the day of embryo transfer. The day of oocyte aspiration was designated day 0. The endometrial thickness was measured as the maximal endometrial width taken on a mid-line, longitudinal uterine section. The echogenic pattern of the endometrium was recorded, according to the following classification (slightly modified from Gonen and Casper, (3)) where the reflectivity of the endometrium is compared with that of the surrounding myometrium, into one of the following types: v p e A: A multilayered, 'triple line' endometrium consisting of prominent outer and central hyperechogenic lines and an inner hypoechogenic area. v p e B: An isoechogenic pattern characterized by the same reflectivity as the surrounding myometrium, with a non-prominent, central echogenic line. l j p e C: A homogenous, hyperechogenic endometrium characterized by increased reflectivity compared to the myometrium and no central echogenic line visible. Statistical evaluation
The data obtained were compared by Student's t-test for unpaired results and chi-square calculation. Correlation was checked by linear regression analysis. Data are presented throughout this paper as the mean f standard deviation. A p-value 3 weeks) used. It is conceivable
IVF
627
that a prolonged hypoestrogenic state may influence the echogenic pattern of the myometrium, thus giving an underestimate of the rate of hyperechogenic endometrium. Another possibility is different judgements between the observers. It is well-known that the inter- and intra-observer variation in sonographic measurements may be considerable (7). In our study two of the three authors were always present during the sonographic examination, and we made an effort to harmonize our judgements in order to diminish this bias. The endometrial pattern showed certain tendencies between pregnant and non pregnant cycles. In patients with a hypoechogenic endometrium (type A) on Day -1 36% conceived and in patients with an isoechogenic pattern (type B) 23% conceived. I t is possible that this difference would have been significant if a larger number of subjects had been included in the study. Nevertheless, it is clear that the variability involved in evaluation of endometrial thickness and echogenic pattern, depending both on biological and methodological variability, calls for additional, more precise methods to evaluate the endometrium. The practical implication of our study is that we recommend sonographic evaluation of the endometrium in addition to follicular diameter and estradiol levels in the monitoring to optimize the time of hCG injection. An endometrial thickness, the day before aspiration, of at least 9 mm appears to be required for an ongoing pregnancy. In all cases a triple line endometrium was evident before aspiration.
Acknowledgments This study was supported by grants from Swedish Medical Research Council (5978). Swedish Medical Society and The Faculty of Medicine, University of Goteborg.
References 1. Scott RT. Hofmann GE, Veeck LL. Jones HW, Muasher SJ. Embryo quality and pregnancy rates in patients attempting pregnancy through in vitro fertilization. Fertil Steril 1991; 55: 426-8. 2. Ueno J. Oehninger S, Brzyski RG, Acosta A A . Philput CB, Muasher SJ. Ultrasonographic appearance of the endometrium in natural and stimulated in-vitro fertilization cycles and its correlation with outcome. Hum Reprod 1991; 6: 9 0 1 4 . 3. Gonen Y, Casper RF. Prediction of implantation by the sonographic appearance of the endometrium during controlled ovarian stimulation for in vitro fertilization (IVF). J In Vitro Fertil Embryo Transfer 1990; 7: 146-52. 4. Wikland M, Attebo B, Granberg M , Holmgren E. @ Arta Obsrer Gynecol Srand 71 (1992)
628
C. Bergh, T. Hillensjo and L. Nilsson
Echocharacteristic of the endometrium a possible parameter for prediction of uterine receptivity. Abstract of the 7th World Congress on IVF and Assisted Procreations, Paris. 30 June-3 July 1991. 5. Wikland M, Nilsson L, Hansson R. Hamberger L. Janson PO. Collection of human oocytes by the use of sonography. Fertil Steril 1983; 39: 603-8. 6. Rogers AW, Polson D. Murphy CR, Hosie M, Susil B, Leoni M. Correlation of endometrial histology, morphometry, and ultrasound appearance after different stimulation protocols for in vitro fertilization. Fertil Steril 1991; 55: 583-7. 7. Robinson JN, Forman RG, Egan DH, Reynolds KM.
0 Acia Obsiei Gynecol Srand 71 (1992)
Bridgewood EA, Barlow DH. Measurements of follicular size by vaginal ultrasound: an assessment of accuracy. Abstract of the I1 Joint ESCO-ESHRE Meeting. Milan. 1990. Address for correspondence: Christina Bergh, M.D. Department of Obstetrics and Gynecology. University of Goteborg. Sahlgrenska Hospital, S-413 45 Goteborg. Sweden
Sonographic evaluation of the endometrium in in vitro fertilization IVF cycles A way to predict pregnancy? CHRISTINA BERGH, TORWORN HILLENSJO AND LARSNILSSON From the Department of Obstetrics and Gynecology, University of Gbteborg, Sahlgrenska Hospital, Gbteborg,
Sweden
Acta Obster Gynecol Scand 1992; 11: 624-628
The endometrial thickness and echogenic pattern were prospectively evaluated in 100 patients undergoing IVF-treatment. On the day before oocyte aspiration (day-1) the endometrium was significantly thicker in the group of patients achieving an ongoing pregnancy than in the group that failed to conceive. The minimal endometrial thickness on day-1 compatible with an ongoing pregnancy was 9 mm. Before oocyte aspiration two different endometrial patterns could be distinguished, a multilayered, ‘triple line’, hypoechogenic endometrium (A) and an isoechogenic pattern (B). Other studies have, in addition. described a hyperechogenic endometrium with a poor pregnancy rate but that pattern could in the present study only be demonstrated after follicle aspiration. 32 clinical pregnancies occurred. Among patients with endometrial pattern A, 36% conceived and among patients with pattern B 23% conceived which was not significantly different. It is concluded that the day before aspiration an endometrial thickness of at least 9 mm appears to be required for an ongoing pregnancy. In all cases a triple line (hypo-or isoechogenic) pattern was evident before aspiration. Key words: in vitro fertilization. transvaginal sonography. endometrium
Submitted February 24, 1992 Accepted May 20, 1992
The two most important variables giving success in in vitro fertilizatiodembryo transfer (IVFIET) are believed to be embryo quality and uterine receptivity. There have been several studies showing a correlation between embryo quality and implantation rate (1) but it has been more difficult to evaluate uterine receptivity. The use of endovaginal ultrasound to measure endometrial thickness and endometrial pattern has produced divergent results. Measuring endometrial thickness alone has not generally been successful in predicting pregnancy (2). A combination of assessing the type of endometrial reflectivity as well as its thickness appeared from a few studies to be a possible way to predict implantation (3, 4). The aim of the present study was to evaluate if a combination of the endometrial thickness, the endo@ Aria Ohster Gynecol Scand 71 (1992)
metrial growth and the pattern of endometrial echo, as studied by vaginal sonography, could be used to predict pregnancy. Special interest was focused on the appearance of the endometrium on the day before follicle aspiration to allow a comparison with earlier studies (3).
Patlents and methods The endometrial thickness and echogenic pattern were studied prospectively during gonadotropin stimulation, follicular aspiration and embryo transfer in 114 consecutive patients undergoing IVFtreatment in the IVF-unit, Sahlgren’s Hospital, GO-
--
20
3
-5
OnOdngpngnrM rpont-
16
Endometrial sonography in IVF
625
andor more than 15 oocytes retrieved).'llventythree patients received progesterone and 77 patients received hCG. The number of pregnancies in these groups were 8 and 25, respectively.
12 c
0
H
6
z9
4
0
Cycle monitoring
7
6
0
1 0 1 1 1 2 13 1 4
1s 16
Endometrld thickneu (mm) Fig. 1. Endometrial thickness on the day before oocyte retrieval (day-1)in the different IVF outcome groups.
teborg, Sweden. Inclusion criteria were tubal infertility of at least 3 years' duration, age below 38 years (range 25-38 years) and normal sperm parameters. Exclusions were made at embryo transfer if fewer than three embryos of presumed good quality (1) were present. One hundred and fourteen patients were initially recruited but fourteen of these were excluded at the time of embryo transfer when it was evident that there were less than 3 high-quality embryos available for transfer. These 14 patients excluded will only be used for comparative reasons in the 'Discussion'. Of the 100 patients still in the study 89 had been down regulated for at least 3 weeks with a GnRH-agonist (Buserelin; Suprefact@, Hoechst, Germany) 1,2 mg/d (nasal administration) before stimulation with human menopausal gonadotropin (hMG, Pergonal", Serono. Italy) 15Cb300 IU/day. Eleven women were stimulated with clorniphene citrate (Clomivid", Draco, Sweden) 100 mg/day, cycle day 4-8 and hMG, 150-225 IU/day, from cycle day 7. Human chorionic gonadotropin (hCG; Profasia, Serono, Italy), usually 10,OOO IU, was administered when 3 or more follicles with a mean diameter of at least 18 mm were identified by vaginal sonography and a continuous rise in serumestradiol had been measured for at least 7 days. Oocytes were retrieved 36 hours after hCG using transvaginal ultrasound guided puncture ( 5 ) . In vitro culture and insemination followed standard techniques. Embryo transfer took place two days after wcyte retrieval using a Frydman TDT catheter and a full urinary bladder. Abdominal ultrasound was used to verify the location of the tip of the catheter and the fluid echo in the cavity. Luteal support was given from the day of ET to all women, either with hCG (1250 IU every 3 days until the 12th day) or, in the case of impending hyperstimulation, progesterone (25 mg as intramuscular injection twice daily). Impending hyperstimulation was based on clinical judgements (e.g. estradiol levels above lhmoyl
The response to hormonal treatment was monitored by serum estradiol levels (assayed by standard RIA) and by transvaginal ultrasound measurements of follicles and endometrium (Ultrasound Scanner 1846 with a 7 MHz vaginal, mechanical sector transducer, Bruel & Kjaer, Denmark). Each measurement of the endometrium was made by two or three different persons. In case of disagreement a second measurement was made. Endometrial pattern, thickness and growth
The endometrium was scanned during the stimulation, starting from 3-4 days before oocyte aspiration and until the day of embryo transfer. The day of oocyte aspiration was designated day 0. The endometrial thickness was measured as the maximal endometrial width taken on a mid-line, longitudinal uterine section. The echogenic pattern of the endometrium was recorded, according to the following classification (slightly modified from Gonen and Casper, (3)) where the reflectivity of the endometrium is compared with that of the surrounding myometrium, into one of the following types: v p e A: A multilayered, 'triple line' endometrium consisting of prominent outer and central hyperechogenic lines and an inner hypoechogenic area. v p e B: An isoechogenic pattern characterized by the same reflectivity as the surrounding myometrium, with a non-prominent, central echogenic line. l j p e C: A homogenous, hyperechogenic endometrium characterized by increased reflectivity compared to the myometrium and no central echogenic line visible. Statistical evaluation
The data obtained were compared by Student's t-test for unpaired results and chi-square calculation. Correlation was checked by linear regression analysis. Data are presented throughout this paper as the mean f standard deviation. A p-value 3 weeks) used. It is conceivable
IVF
627
that a prolonged hypoestrogenic state may influence the echogenic pattern of the myometrium, thus giving an underestimate of the rate of hyperechogenic endometrium. Another possibility is different judgements between the observers. It is well-known that the inter- and intra-observer variation in sonographic measurements may be considerable (7). In our study two of the three authors were always present during the sonographic examination, and we made an effort to harmonize our judgements in order to diminish this bias. The endometrial pattern showed certain tendencies between pregnant and non pregnant cycles. In patients with a hypoechogenic endometrium (type A) on Day -1 36% conceived and in patients with an isoechogenic pattern (type B) 23% conceived. I t is possible that this difference would have been significant if a larger number of subjects had been included in the study. Nevertheless, it is clear that the variability involved in evaluation of endometrial thickness and echogenic pattern, depending both on biological and methodological variability, calls for additional, more precise methods to evaluate the endometrium. The practical implication of our study is that we recommend sonographic evaluation of the endometrium in addition to follicular diameter and estradiol levels in the monitoring to optimize the time of hCG injection. An endometrial thickness, the day before aspiration, of at least 9 mm appears to be required for an ongoing pregnancy. In all cases a triple line endometrium was evident before aspiration.
Acknowledgments This study was supported by grants from Swedish Medical Research Council (5978). Swedish Medical Society and The Faculty of Medicine, University of Goteborg.
References 1. Scott RT. Hofmann GE, Veeck LL. Jones HW, Muasher SJ. Embryo quality and pregnancy rates in patients attempting pregnancy through in vitro fertilization. Fertil Steril 1991; 55: 426-8. 2. Ueno J. Oehninger S, Brzyski RG, Acosta A A . Philput CB, Muasher SJ. Ultrasonographic appearance of the endometrium in natural and stimulated in-vitro fertilization cycles and its correlation with outcome. Hum Reprod 1991; 6: 9 0 1 4 . 3. Gonen Y, Casper RF. Prediction of implantation by the sonographic appearance of the endometrium during controlled ovarian stimulation for in vitro fertilization (IVF). J In Vitro Fertil Embryo Transfer 1990; 7: 146-52. 4. Wikland M, Attebo B, Granberg M , Holmgren E. @ Arta Obsrer Gynecol Srand 71 (1992)
628
C. Bergh, T. Hillensjo and L. Nilsson
Echocharacteristic of the endometrium a possible parameter for prediction of uterine receptivity. Abstract of the 7th World Congress on IVF and Assisted Procreations, Paris. 30 June-3 July 1991. 5. Wikland M, Nilsson L, Hansson R. Hamberger L. Janson PO. Collection of human oocytes by the use of sonography. Fertil Steril 1983; 39: 603-8. 6. Rogers AW, Polson D. Murphy CR, Hosie M, Susil B, Leoni M. Correlation of endometrial histology, morphometry, and ultrasound appearance after different stimulation protocols for in vitro fertilization. Fertil Steril 1991; 55: 583-7. 7. Robinson JN, Forman RG, Egan DH, Reynolds KM.
0 Acia Obsiei Gynecol Srand 71 (1992)
Bridgewood EA, Barlow DH. Measurements of follicular size by vaginal ultrasound: an assessment of accuracy. Abstract of the I1 Joint ESCO-ESHRE Meeting. Milan. 1990. Address for correspondence: Christina Bergh, M.D. Department of Obstetrics and Gynecology. University of Goteborg. Sahlgrenska Hospital, S-413 45 Goteborg. Sweden
