- Email: [email protected]
Endometrial wavelike activity, endometrial thickness, and ultrasound texture in controlled ovarian hyperstimulation cycles
FERTILITY AND STERILITYt VOL. 70, NO. 2, AUGUST 1998 Copyright ©1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
Endometrial wavelike activity, endometrial thickness, and ultrasound texture in controlled ovarian hyperstimulation cycles Marga M. IJland, M.D., Johannes L. H. Evers, M.D., Gerard A. J. Dunselman, M.D., and Henk J. Hoogland, M.D. Department of Obstetrics and Gynaecology, University Hospital Maastricht, Maastricht, the Netherlands
Objective: To describe endometrial wavelike activity, endometrial thickness, and texture in controlled ovarian hyperstimulation (COH) cycles. Design: Prospective observational ultrasound study. Setting: University hospital-based infertility clinic. Patient(s): Thirty-five COH cycles in 19 women with unexplained infertility. Intervention(s): Transvaginal ultrasound examination was performed throughout COH cycles. Intrauterine insemination was performed after hCG administration. Main Outcome Measure(s): Endometrial wavelike activity, wave frequency, wave velocity, endometrial thickness, and endometrial texture. Result(s): Endometrial wavelike activity increased from menstruation to ovulation and decreased in the luteal phase. On day hCG12, endometrial wave-like activity was observed in all cycles. Waves from cervix to fundus prevailed in the periovulatory phase. Endometrial wavelike activity was related significantly to endometrial thickness at the start of ovarian stimulation and in the luteal phase. Endometrial thickness increased throughout the cycle. Endometrial texture showed periovulatory a triple-line aspect. Conclusion(s): In COH cycles, endometrial wavelike activity is more pronounced than in spontaneous cycles. The number of follicles and endometrial wavelike activity were not correlated significantly. This is the first prospective study to provide longitudinal observational evidence that endometrial thickness increases throughout the COH cycle and that a triple line pattern develops. (Fertil Sterilt 1998;70:279 – 83. ©1998 by American Society for Reproductive Medicine.) Key Words: Endometrial activity, endometrial waves, wave velocity, endometrial thickness, endometrial texture, transvaginal ultrasound, controlled ovarian hyperstimulation
The endometrial response to ovarian stimulation can be visualized by ultrasound (US). Most US studies have involved attempts to relate endometrial thickness and endometrial texture to endometrial receptivity in IVF cycles (1– 4). Received July 24, 1997; revised and accepted April 1, 1998. Reprint requests: Henk J. Hoogland, M.D., Obstetrics and Gynaecology, Academisch Ziekenhuis Maastricht, P.O. Box 5800, 6202 AZ Maastricht, the Netherlands (FAX: 31-433874765). 0015-0282/98/$19.00 PII S0015-0282(98)00132-0
A possible additional index of endometrial receptivity, endometrial wavelike activity, has been studied by US in spontaneous cycles only (5–10). In related studies we have qualified and quantified endometrial wavelike activity longitudinally in spontaneous cycles (11, 12). Our studies in spontaneous cycles also have shown endometrial wavelike activity to be related to fecundability (13). To shed more light on the complex features of endometrium development and activity in
stimulated cycles, we decided to map endometrial wavelike activity patterns longitudinally in controlled ovarian hyperstimulation (COH) cycles. Endometrial wavelike activity was related to the more traditional subjects of US investigation, i.e., follicle diameter and number, and endometrium thickness and texture.
MATERIALS AND METHODS The US observations were performed in 35 COH cycles in which hMG (Humegon; Organon, Oss, the Netherlands) was administered. Nineteen patients with unexplained infertility undergoing COH and IUI were included in the study. Each cycle was seen as an independent series of events in this observational study. Ovarian stimulation was initiated on the 3rd 279
day of menstruation in the absence of ovarian cysts. Human menopausal gonadotropin was administered subcutaneously at a dosage of 75, 150, or 225 IU/d according to the ovarian response as determined by US. Human chorionic gonadotropin (5,000 IU, Pregnyl; Organon) was administered when at least one follicle had reached a diameter of $18 mm. Human chorionic gonadotropin was withheld if more than four dominant follicles were present. Intrauterine insemination was performed 39 hours after hCG administration. The progesterone level was determined on day hCG19 to document ovulation. Values are given as means 6 SD. Throughout the cycle, five US measurements were performed at five standardized moments, namely, before starting ovarian stimulation (start), on the day of hCG administration (hCG), on the day of ovulation (hCG12), and four (hCG16) and seven (hCG19) days after ovulation. Furthermore, between (start) and (hCG), the stimulation period was divided into two intervals, one in which the diameter of the dominant follicle was #14 mm and one in which the diameter of the dominant follicle was .14 mm and ,18 mm. During these two intervals, additional US examinations were performed, depending on the rate of follicle growth. If more US examinations were performed during a particular interval, the last examination in each period was analyzed. Ultrasound examinations were performed transvaginally (7.5-mHz transducer) with the use of a real-time scanner (Ultramark-9-HDI-ESP; Bothell, WA) (11). All measurements were performed by a single investigator (M.IJ.) and were recorded on videotape. At each investigation, a midsagittal scan of the uterus was performed to study the endometrium. The number and size of the follicles were recorded in both the left and right ovary. The recording was stopped when no endometrial wavelike activity was observed for 3 minutes. After recording, a time-code was added to each video frame, allowing a description of the aspects of endometrial activity with a potential accuracy of 0.04 seconds. Off-line analysis was performed at high-speed replay (4 times) (11). Analysis was focused on endometrial activity per se (activity/no activity), differentiated activity (wave type), wave velocity, and wave occurrence frequency. Endometrial activity was described with the use of the previously described wave direction classification (11), which distinguishes five types of endometrial movements, i.e., no activity during at least 3 minutes of recording, waves from cervix to fundus (CF), waves from fundus to cervix (FC), opposing waves starting simultaneously at cervix and fundus (OP), and random waves starting at various foci (R). The relative presence of the different wave types was calculated for the different phases of the cycle. Wave occurrence frequency (waves per minute) (11), wave velocity (the length of the displayed endometrium divided by the duration of a wave), and wave intervals were calculated for those 280
IJland et al.
Endometrial wavelike activity and thickness
waves that allowed such a calculation, i.e., waves from cervix to fundus and waves from fundus to cervix (12). The follicle diameter represents the mean of three perpendicular dimensions. The number of follicles is the number of follicles exceeding a mean follicular diameter of 10 mm. Endometrial thickness was measured as the maximum distance between the two myometrial and endometrial interfaces in the midsagittal plane of the uterus. Endometrial ultrasonographic morphology was classified into three types: type I was multilayered, triple-line endometrium consisting of a prominent outer and inner hyperechogenic line and inner hypoechogenic regions; type II was intermediate isoechogenic pattern, with the same reflectivity as the surrounding myometrium and a poorly defined central echogenic line; and type III was entirely homogeneous, hyperechogenic pattern (1). The relative distribution of endometrial texture patterns among patients was calculated for the different phases of the cycle. Statistical analysis was performed with use of the MannWhitney U test, with P values of ,0.05 indicating statistical significance.
RESULTS Controlled ovarian hyperstimulation using hMG was started in 19 patients in 35 cycles. In 6 cycles hCG was withheld because these patients showed more than four follicles with diameters of $14 mm. Ultrasound monitoring and ovarian stimulation in these cycles were discontinued. Progesterone as determined on day hCG19 was 76.6 6 39.2 nmol/L, indicating an ovulation in the cycles that were continued. The study population ranged from 26 to 36 years of age, with a median age of 32 years. The duration of infertility varied from 28 to 95 months, with a median of 47 months. Twelve patients had primary infertility, and seven had secondary infertility. None of the patients became pregnant during the cycle of investigation. A total of 202 recordings were analyzed. The presence and types of endometrial wavelike activity in the cycles studied are shown in Figure 1. Endometrial wavelike activity was present in 47% of the cycles studied before the start of ovarian stimulation, increasing to 100% at ovulation (hCG12) and decreasing to 52% of cycles on day hCG19. Waves from cervix to fundus increased from 11% to 39% between the start of ovarian stimulation and the day of hCG. On the day of ovulation (hCG12), waves from cervix to fundus were present in 45%, decreasing to 10% on day hCG19. Waves from fundus to cervix started at 11% and increased to 48% in the phase with follicles between 14 and 18 mm. Waves from fundus to cervix were seen in 35%, Vol. 70, No. 2, August 1998
FIGURE 1 Relative distribution of wave type patterns throughout controlled ovarian hyperstimulation cycles. ■ 5 opposing waves; z 5 waves from cervix to fundus; d 5 waves from fundus to cervix; p 5 random activity; h 5 no activity.
10%, and 3%, respectively, on the day of hCG administration and on hCG12 and hCG16. Opposing waves were observed from the period in which the follicles were #14 mm, up to and including hCG16 in 15%, 16%, 17%, 20%, and 6%, respectively. Before hMG was started, random waves were observed in 31%, on the day of hCG in 2%, and on hCG12 in 25%. On days hCG16 and hCG19 random waves were observed in 44% and 38%, respectively. Table 1 shows the number of recordings, the diameter of
the follicles, the number of follicles, and the endometrial thickness throughout all cycles and separately in the canceled cycles. Mean (6SD) endometrial thickness increased from 4.9 6 2.2 mm before starting COH to 10.6 6 1.9 mm on day hCG12, and continued to increase to 11.8 6 3.5 mm on day hCG19. Table 2 shows the relation between the presence of endometrial wavelike activity (no activity and activity) and US measurements, i.e., endometrial thickness, diameter of the
TABLE 1 Number of recordings, diameter and number of follicles, and endometrial thickness throughout ovarian hyperstimulation cycles. No. of recordings Phase of cycle Start Follicle diameter 14 mm ,follicle hCG hCG12 hCG16 hCG19 Cycles canceled Start Follicle diameter 14 mm ,follicle
of #14 mm ,18 mm
of #14 mm ,18 mm
35 32 19 29 29 29 29 6 6 6
Follicle diameter
No. of follicles
12.3 6 1.1 15.8 6 1.1 19.1 6 1.0
2.0 6 1.4 4.5 6 2.9 2.7 6 1.6
12.8 6 1.3 16.7 6 0.8
3.8 6 2.1 8.2 6 2.3
Endometrial thickness 4.9 6 2.2 8.1 6 2.1 9.5 6 1.9 9.9 6 1.8 10.6 6 1.9 11.4 6 2.5 11.8 6 3.5 5.9 6 1.6 10.0 6 1.7 11.1 6 1.9
Note: All values are means 6 SD unless otherwise indicated.
FERTILITY & STERILITYt
281
TABLE 2 Presence of endometrial wavelike activity, endometrial thickness, and characteristics of follicles in controlled ovarian hyperstimulation cycles. Endometrial thickness
Phase of cycle Start Follicle diameter #14 mm 14 mm ,follicle ,18 mm hCG hCG12 hCG16 hCG19
No activity (n)
Activity (n)
3.9 6 1.5 (17) 6.6 6 2.0 (5) 8.4 (1) 8.7 6 1.2 (3) (0) 9.4 6 2.6 (7) 9.8 6 1.8 (16)
5.9 6 2.4 (18) 8.3 6 2.0 (27) 9.5 6 2.0 (18) 9.9 6 1.8 (26) 10.5 6 1.9 (29) 11.9 6 2.2 (22) 14.2 6 3.7 (13)
Follicle diameter P value* 0.01 0.09 0.85 0.18
No. of follicles
No activity
Activity
P value*
No activity
Activity
P value*
12.0 6 1.1 14.3 19.8 6 1.1
12.5 6 1.1 15.9 6 2.0 19.0 6 0.9
0.3 0.1 0.3
2.0 6 0.7 3.0 1.7 6 1.1
2.2 6 1.6 4.4 6 3.1 2.9 6 1.6
0.6 0.8 0.2
0.02 0.0002
Note: All value are means 6 SD unless otherwise indicated; n 5 number of recordings. p Determined by the Mann-Whitney U test.
follicles, and number of follicles. At three moments in the cycle the presence of endometrial wavelike activity was significantly related to endometrial thickness. There was no relation between the presence of endometrial wavelike activity and dimensions or numbers of follicles. Table 3 shows the frequency, velocity, and intervals of the endometrial waves. The frequency of waves from cervix to fundus increased from the start of COH to day hCG16 and decreased thereafter. The frequency of waves from fundus to cervix increased from the start of COH to the day of hCG. In the second half of the cycle the number of observations of waves from fundus to cervix was too low to allow exact calculation of frequency, velocity, and intervals. The velocity of waves from cervix to fundus and from fundus to cervix remained unchanged throughout the cycle. Intervals between waves from cervix to fundus were
negative on days hCG12 and hCG16, whereas intervals between waves from fundus to cervix were negative in the midfollicular phase, indicating overlapping waves. Endometrial texture (Table 4) changed from homogeneous at the start (83%) to a triple-line pattern (73%) on the day of hCG administration and then back to a homogeneous pattern on day hCG19 (93%).
DISCUSSION The primary function of the endometrium is to prepare for nidation of the embryo. Ultrasound examination may provide a clinical basis for a functional index of this endometrial receptivity. Endometrial characteristics, such as thickness and morphology, have been studied in ovarian stimulation cycles on the day of hCG administration and in the luteal phase (14), but observational studies with repeated measure-
TABLE 3 Frequency, velocity, and intervals of endometrial waves (CF and FC) throughout controlled ovarian hyperstimulation cycles. COH cycle
Endometrial wave CF Frequency (waves/min) Velocity (mm/s) Intervals (s) FC Frequency (waves/min) Velocity (mm/s) Intervals (s)
Follicle diameter of #14 mm
14 mm ,follicle ,18 mm
hCG
hCG12
hCG16
hCG19
0.79 0.9 6 0.1
1.7 6 0.2 1.1 6 0.1 4.7 6 2.2
1.8 6 0.7 1.3 6 0.4 10.6 6 14.4
2.4 6 0.9 1.4 6 0.3 5.1 6 13.2
3.2 6 1.4 1.4 6 0.3 23.7 6 10.6
3.3 6 1.3 1.3 6 0.4 22.7 6 12.3
2.0 6 0.1
1.83 1.2 6 0.3 0.33
1.9 6 0.7 1.2 6 0.2 6.0 6 10.0
Start of cycle
2.6 6 0.5 1.3 6 0.3 23.3 6 6.4
0.6 6 9.9
2.7 6 1.6 1.3 6 0.4 6.7 6 15.2
Note: All values are means 6 SD. COH 5 controlled ovarian hyperstimulation; CF 5 waves from cervix to fundus; FC 5 waves from fundus to cervix.
282
IJland et al.
Endometrial wavelike activity and thickness
Vol. 70, No. 2, August 1998
TABLE 4 Endometrial texture throughout controlled ovarian hyperstimulation cycles. Endometrial texture Phase of cycle Start Follicle diameter #14 mm 14 mm ,follicle ,18 mm hCG hCG12 hCG16 hCG19
I
II
III
3 51 63 73 52 0 0
14 40 28 17 38 14 7
83 9 9 10 10 86 93
Note: All values are percentages. I 5 triple line; II 5 intermediate; III 5 homogeneous.
ments throughout stimulation cycles have not been published previously. We previously had studied endometrial wavelike activity in spontaneous cycles in a longitudinal design (11–13). To extend these observations and to gather optimal information about different aspects of the endometrium, we studied the endometrial characteristics and endometrial wavelike activity on fixed moments throughout COH cycles. The results confirm some of our findings in spontaneous cycles. Endometrial wavelike activity was recorded most frequently in the periovulatory phase. Similar wave patterns were observed in spontaneous and ovarian stimulation cycles, namely, waves from cervix to fundus, waves from fundus to cervix, random activity, and opposing waves (11, 12). However, in ovarian stimulation cycles endometrial wavelike activity was more pronounced in all corresponding phases of the cycles than in the phases of the spontaneous cycles. This difference perhaps may be explained by higher estrogen levels in ovarian stimulation cycles. Cross-sectional studies of the effect of ovarian stimulation on the endometrium suggest that the endometrium increases in thickness and that a triple-line pattern develops in the course of the cycle (14). The present longitudinal study in COH cycles confirmed this. A significant positive correlation between endometrial thickness and the presence of endometrial wavelike activity was found only in two phases of the cycle, i.e., before the start of ovarian stimulation and in the luteal phase. This may imply that at the start of the COH cycle the thickness of the endometrium facilitates the US detection of these waves. Theoretically, a higher number of follicles in the follicular phase of COH cycles with a subsequent rise of estrogens would result in a thicker endo-
FERTILITY & STERILITYt
metrium and a more pronounced endometrial wavelike activity. However, no relationship was detected between characteristics of the follicles (number and diameter) and endometrial wavelike activity, perhaps reflecting supranormal estrogen stimulation. The expected number of ongoing pregnancies occurring in 35 COH cycles is 5 (95% confidence interval [CI] 2–11); the observed number was 0. We have no explanation for this significant difference. One explanation could be that repeated US observations interfere with the embryo implantation process. However, observations of normal pregnancy rates in spontaneous cycles in women exposed to the same repeated US observations (13) appear to contradict this. The conception cycles in our previous study (13) showed less activity, and no waves from fundus to cervix were seen after ovulation. Controlled ovarian hyperstimulation cycles, on the other hand, showed an increase in endometrial wavelike activity in all phases of the cycle and in some cycles even waves from fundus to cervix after ovulation. These results support the contention that the embryo may need a quiet environment for nidation and a coherent, finetuned activity pattern of the endometrium, as has been suggested previously (13). References 1. Gonen Y, Asper RF. Prediction of implantation by the sonographic appearance of the endometrium during controlled ovarian stimulation for in vitro fertilization (IVF). J In Vitro Fert Embryo Transfer 1990; 7:146 –52. 2. Imoedemhe DAG, Shaw RW, Kirkland A, Chan R. Ultrasound measurement of endometrial thickness on different ovarian stimulation regimens during in-vitro fertilization. Hum Reprod 1987;2:545–7. 3. Lenz S, Lindenberg S. Ultrasonic evaluation of endometrial growth in women with normal cycles during spontaneous and stimulated cycles. Hum Reprod 1990;5:377– 81. 4. Randall JM, Fisk NM, McTavish, Templeton AA. Transvaginal ultrasonic assessment of endometrial growth in spontaneous and hyperstimulated menstrual cycles. Br J Obstet Gynaecol 1989;96:954 –9. 5. Abramowics JS, Archer DF. Uterine endometrial peristalsis—a transvaginal ultrasound study. Fertil Steril 1990;54:451– 4. 6. Birnholz JC. Ultrasonic visualization of endometrial movements. Fertil Steril 1984;41:157– 8. 7. Chalubinski K, Deutinger J, Bernaschek G. Vaginosonography for recording of cycle-related myometrial contractions. Fertil Steril 1993; 59:225– 8. 8. De Vries K, Lyons EA, Levi CS, Lindsay D. Contraction of the inner third of the myometrium. Am J Obstet Gynecol 1990;162:679 – 82. 9. Lyons EA, Taylor PJ, Zheng XH, Ballard G, Levi CS, Kredenster JV. Characterization of subendometrial myometrial contractions throughout the menstrual cycle in normal fertile women. Fertil Steril 1991;55: 771– 4. 10. Martinez Gauda M, Yoshiba T, Bentsson LP. Propagated and nonpropagated myometrial contractiona in normal menstrual cycles. Am J Obstet Gynecol 1973;115:107–11. 11. IJland MM, Evers JLH, Dunselman GAJ, van Katwijk C, Lo CR, Hoogland HJ. Endometrial wavelike movements during the menstrual cycle. Fertil Steril 1996;65:746 –9. 12. IJland MM, Evers JLH, Hoogland HJ. Velocity of endometrial wavelike activity in spontaneous cycles. Fertil Steril 1997;68:72–5. 13. IJland MM, Evers JLH, Dunselman GAJ, Hoogland HJ. Relation between endometrial wavelike activity and fecundability in spontaneous cycles. Fertil Steril 1997;67:492– 6. 14. Friedler S, Schenker JG, Herman A, Lewin A. The role of ultrasonography in the evaluation of endometrial receptivity following assisted reproductive treatments: a critical review. Hum Reprod 1996;2:323–5.
283
Endometrial wavelike activity, endometrial thickness, and ultrasound texture in controlled ovarian hyperstimulation cycles Marga M. IJland, M.D., Johannes L. H. Evers, M.D., Gerard A. J. Dunselman, M.D., and Henk J. Hoogland, M.D. Department of Obstetrics and Gynaecology, University Hospital Maastricht, Maastricht, the Netherlands
Objective: To describe endometrial wavelike activity, endometrial thickness, and texture in controlled ovarian hyperstimulation (COH) cycles. Design: Prospective observational ultrasound study. Setting: University hospital-based infertility clinic. Patient(s): Thirty-five COH cycles in 19 women with unexplained infertility. Intervention(s): Transvaginal ultrasound examination was performed throughout COH cycles. Intrauterine insemination was performed after hCG administration. Main Outcome Measure(s): Endometrial wavelike activity, wave frequency, wave velocity, endometrial thickness, and endometrial texture. Result(s): Endometrial wavelike activity increased from menstruation to ovulation and decreased in the luteal phase. On day hCG12, endometrial wave-like activity was observed in all cycles. Waves from cervix to fundus prevailed in the periovulatory phase. Endometrial wavelike activity was related significantly to endometrial thickness at the start of ovarian stimulation and in the luteal phase. Endometrial thickness increased throughout the cycle. Endometrial texture showed periovulatory a triple-line aspect. Conclusion(s): In COH cycles, endometrial wavelike activity is more pronounced than in spontaneous cycles. The number of follicles and endometrial wavelike activity were not correlated significantly. This is the first prospective study to provide longitudinal observational evidence that endometrial thickness increases throughout the COH cycle and that a triple line pattern develops. (Fertil Sterilt 1998;70:279 – 83. ©1998 by American Society for Reproductive Medicine.) Key Words: Endometrial activity, endometrial waves, wave velocity, endometrial thickness, endometrial texture, transvaginal ultrasound, controlled ovarian hyperstimulation
The endometrial response to ovarian stimulation can be visualized by ultrasound (US). Most US studies have involved attempts to relate endometrial thickness and endometrial texture to endometrial receptivity in IVF cycles (1– 4). Received July 24, 1997; revised and accepted April 1, 1998. Reprint requests: Henk J. Hoogland, M.D., Obstetrics and Gynaecology, Academisch Ziekenhuis Maastricht, P.O. Box 5800, 6202 AZ Maastricht, the Netherlands (FAX: 31-433874765). 0015-0282/98/$19.00 PII S0015-0282(98)00132-0
A possible additional index of endometrial receptivity, endometrial wavelike activity, has been studied by US in spontaneous cycles only (5–10). In related studies we have qualified and quantified endometrial wavelike activity longitudinally in spontaneous cycles (11, 12). Our studies in spontaneous cycles also have shown endometrial wavelike activity to be related to fecundability (13). To shed more light on the complex features of endometrium development and activity in
stimulated cycles, we decided to map endometrial wavelike activity patterns longitudinally in controlled ovarian hyperstimulation (COH) cycles. Endometrial wavelike activity was related to the more traditional subjects of US investigation, i.e., follicle diameter and number, and endometrium thickness and texture.
MATERIALS AND METHODS The US observations were performed in 35 COH cycles in which hMG (Humegon; Organon, Oss, the Netherlands) was administered. Nineteen patients with unexplained infertility undergoing COH and IUI were included in the study. Each cycle was seen as an independent series of events in this observational study. Ovarian stimulation was initiated on the 3rd 279
day of menstruation in the absence of ovarian cysts. Human menopausal gonadotropin was administered subcutaneously at a dosage of 75, 150, or 225 IU/d according to the ovarian response as determined by US. Human chorionic gonadotropin (5,000 IU, Pregnyl; Organon) was administered when at least one follicle had reached a diameter of $18 mm. Human chorionic gonadotropin was withheld if more than four dominant follicles were present. Intrauterine insemination was performed 39 hours after hCG administration. The progesterone level was determined on day hCG19 to document ovulation. Values are given as means 6 SD. Throughout the cycle, five US measurements were performed at five standardized moments, namely, before starting ovarian stimulation (start), on the day of hCG administration (hCG), on the day of ovulation (hCG12), and four (hCG16) and seven (hCG19) days after ovulation. Furthermore, between (start) and (hCG), the stimulation period was divided into two intervals, one in which the diameter of the dominant follicle was #14 mm and one in which the diameter of the dominant follicle was .14 mm and ,18 mm. During these two intervals, additional US examinations were performed, depending on the rate of follicle growth. If more US examinations were performed during a particular interval, the last examination in each period was analyzed. Ultrasound examinations were performed transvaginally (7.5-mHz transducer) with the use of a real-time scanner (Ultramark-9-HDI-ESP; Bothell, WA) (11). All measurements were performed by a single investigator (M.IJ.) and were recorded on videotape. At each investigation, a midsagittal scan of the uterus was performed to study the endometrium. The number and size of the follicles were recorded in both the left and right ovary. The recording was stopped when no endometrial wavelike activity was observed for 3 minutes. After recording, a time-code was added to each video frame, allowing a description of the aspects of endometrial activity with a potential accuracy of 0.04 seconds. Off-line analysis was performed at high-speed replay (4 times) (11). Analysis was focused on endometrial activity per se (activity/no activity), differentiated activity (wave type), wave velocity, and wave occurrence frequency. Endometrial activity was described with the use of the previously described wave direction classification (11), which distinguishes five types of endometrial movements, i.e., no activity during at least 3 minutes of recording, waves from cervix to fundus (CF), waves from fundus to cervix (FC), opposing waves starting simultaneously at cervix and fundus (OP), and random waves starting at various foci (R). The relative presence of the different wave types was calculated for the different phases of the cycle. Wave occurrence frequency (waves per minute) (11), wave velocity (the length of the displayed endometrium divided by the duration of a wave), and wave intervals were calculated for those 280
IJland et al.
Endometrial wavelike activity and thickness
waves that allowed such a calculation, i.e., waves from cervix to fundus and waves from fundus to cervix (12). The follicle diameter represents the mean of three perpendicular dimensions. The number of follicles is the number of follicles exceeding a mean follicular diameter of 10 mm. Endometrial thickness was measured as the maximum distance between the two myometrial and endometrial interfaces in the midsagittal plane of the uterus. Endometrial ultrasonographic morphology was classified into three types: type I was multilayered, triple-line endometrium consisting of a prominent outer and inner hyperechogenic line and inner hypoechogenic regions; type II was intermediate isoechogenic pattern, with the same reflectivity as the surrounding myometrium and a poorly defined central echogenic line; and type III was entirely homogeneous, hyperechogenic pattern (1). The relative distribution of endometrial texture patterns among patients was calculated for the different phases of the cycle. Statistical analysis was performed with use of the MannWhitney U test, with P values of ,0.05 indicating statistical significance.
RESULTS Controlled ovarian hyperstimulation using hMG was started in 19 patients in 35 cycles. In 6 cycles hCG was withheld because these patients showed more than four follicles with diameters of $14 mm. Ultrasound monitoring and ovarian stimulation in these cycles were discontinued. Progesterone as determined on day hCG19 was 76.6 6 39.2 nmol/L, indicating an ovulation in the cycles that were continued. The study population ranged from 26 to 36 years of age, with a median age of 32 years. The duration of infertility varied from 28 to 95 months, with a median of 47 months. Twelve patients had primary infertility, and seven had secondary infertility. None of the patients became pregnant during the cycle of investigation. A total of 202 recordings were analyzed. The presence and types of endometrial wavelike activity in the cycles studied are shown in Figure 1. Endometrial wavelike activity was present in 47% of the cycles studied before the start of ovarian stimulation, increasing to 100% at ovulation (hCG12) and decreasing to 52% of cycles on day hCG19. Waves from cervix to fundus increased from 11% to 39% between the start of ovarian stimulation and the day of hCG. On the day of ovulation (hCG12), waves from cervix to fundus were present in 45%, decreasing to 10% on day hCG19. Waves from fundus to cervix started at 11% and increased to 48% in the phase with follicles between 14 and 18 mm. Waves from fundus to cervix were seen in 35%, Vol. 70, No. 2, August 1998
FIGURE 1 Relative distribution of wave type patterns throughout controlled ovarian hyperstimulation cycles. ■ 5 opposing waves; z 5 waves from cervix to fundus; d 5 waves from fundus to cervix; p 5 random activity; h 5 no activity.
10%, and 3%, respectively, on the day of hCG administration and on hCG12 and hCG16. Opposing waves were observed from the period in which the follicles were #14 mm, up to and including hCG16 in 15%, 16%, 17%, 20%, and 6%, respectively. Before hMG was started, random waves were observed in 31%, on the day of hCG in 2%, and on hCG12 in 25%. On days hCG16 and hCG19 random waves were observed in 44% and 38%, respectively. Table 1 shows the number of recordings, the diameter of
the follicles, the number of follicles, and the endometrial thickness throughout all cycles and separately in the canceled cycles. Mean (6SD) endometrial thickness increased from 4.9 6 2.2 mm before starting COH to 10.6 6 1.9 mm on day hCG12, and continued to increase to 11.8 6 3.5 mm on day hCG19. Table 2 shows the relation between the presence of endometrial wavelike activity (no activity and activity) and US measurements, i.e., endometrial thickness, diameter of the
TABLE 1 Number of recordings, diameter and number of follicles, and endometrial thickness throughout ovarian hyperstimulation cycles. No. of recordings Phase of cycle Start Follicle diameter 14 mm ,follicle hCG hCG12 hCG16 hCG19 Cycles canceled Start Follicle diameter 14 mm ,follicle
of #14 mm ,18 mm
of #14 mm ,18 mm
35 32 19 29 29 29 29 6 6 6
Follicle diameter
No. of follicles
12.3 6 1.1 15.8 6 1.1 19.1 6 1.0
2.0 6 1.4 4.5 6 2.9 2.7 6 1.6
12.8 6 1.3 16.7 6 0.8
3.8 6 2.1 8.2 6 2.3
Endometrial thickness 4.9 6 2.2 8.1 6 2.1 9.5 6 1.9 9.9 6 1.8 10.6 6 1.9 11.4 6 2.5 11.8 6 3.5 5.9 6 1.6 10.0 6 1.7 11.1 6 1.9
Note: All values are means 6 SD unless otherwise indicated.
FERTILITY & STERILITYt
281
TABLE 2 Presence of endometrial wavelike activity, endometrial thickness, and characteristics of follicles in controlled ovarian hyperstimulation cycles. Endometrial thickness
Phase of cycle Start Follicle diameter #14 mm 14 mm ,follicle ,18 mm hCG hCG12 hCG16 hCG19
No activity (n)
Activity (n)
3.9 6 1.5 (17) 6.6 6 2.0 (5) 8.4 (1) 8.7 6 1.2 (3) (0) 9.4 6 2.6 (7) 9.8 6 1.8 (16)
5.9 6 2.4 (18) 8.3 6 2.0 (27) 9.5 6 2.0 (18) 9.9 6 1.8 (26) 10.5 6 1.9 (29) 11.9 6 2.2 (22) 14.2 6 3.7 (13)
Follicle diameter P value* 0.01 0.09 0.85 0.18
No. of follicles
No activity
Activity
P value*
No activity
Activity
P value*
12.0 6 1.1 14.3 19.8 6 1.1
12.5 6 1.1 15.9 6 2.0 19.0 6 0.9
0.3 0.1 0.3
2.0 6 0.7 3.0 1.7 6 1.1
2.2 6 1.6 4.4 6 3.1 2.9 6 1.6
0.6 0.8 0.2
0.02 0.0002
Note: All value are means 6 SD unless otherwise indicated; n 5 number of recordings. p Determined by the Mann-Whitney U test.
follicles, and number of follicles. At three moments in the cycle the presence of endometrial wavelike activity was significantly related to endometrial thickness. There was no relation between the presence of endometrial wavelike activity and dimensions or numbers of follicles. Table 3 shows the frequency, velocity, and intervals of the endometrial waves. The frequency of waves from cervix to fundus increased from the start of COH to day hCG16 and decreased thereafter. The frequency of waves from fundus to cervix increased from the start of COH to the day of hCG. In the second half of the cycle the number of observations of waves from fundus to cervix was too low to allow exact calculation of frequency, velocity, and intervals. The velocity of waves from cervix to fundus and from fundus to cervix remained unchanged throughout the cycle. Intervals between waves from cervix to fundus were
negative on days hCG12 and hCG16, whereas intervals between waves from fundus to cervix were negative in the midfollicular phase, indicating overlapping waves. Endometrial texture (Table 4) changed from homogeneous at the start (83%) to a triple-line pattern (73%) on the day of hCG administration and then back to a homogeneous pattern on day hCG19 (93%).
DISCUSSION The primary function of the endometrium is to prepare for nidation of the embryo. Ultrasound examination may provide a clinical basis for a functional index of this endometrial receptivity. Endometrial characteristics, such as thickness and morphology, have been studied in ovarian stimulation cycles on the day of hCG administration and in the luteal phase (14), but observational studies with repeated measure-
TABLE 3 Frequency, velocity, and intervals of endometrial waves (CF and FC) throughout controlled ovarian hyperstimulation cycles. COH cycle
Endometrial wave CF Frequency (waves/min) Velocity (mm/s) Intervals (s) FC Frequency (waves/min) Velocity (mm/s) Intervals (s)
Follicle diameter of #14 mm
14 mm ,follicle ,18 mm
hCG
hCG12
hCG16
hCG19
0.79 0.9 6 0.1
1.7 6 0.2 1.1 6 0.1 4.7 6 2.2
1.8 6 0.7 1.3 6 0.4 10.6 6 14.4
2.4 6 0.9 1.4 6 0.3 5.1 6 13.2
3.2 6 1.4 1.4 6 0.3 23.7 6 10.6
3.3 6 1.3 1.3 6 0.4 22.7 6 12.3
2.0 6 0.1
1.83 1.2 6 0.3 0.33
1.9 6 0.7 1.2 6 0.2 6.0 6 10.0
Start of cycle
2.6 6 0.5 1.3 6 0.3 23.3 6 6.4
0.6 6 9.9
2.7 6 1.6 1.3 6 0.4 6.7 6 15.2
Note: All values are means 6 SD. COH 5 controlled ovarian hyperstimulation; CF 5 waves from cervix to fundus; FC 5 waves from fundus to cervix.
282
IJland et al.
Endometrial wavelike activity and thickness
Vol. 70, No. 2, August 1998
TABLE 4 Endometrial texture throughout controlled ovarian hyperstimulation cycles. Endometrial texture Phase of cycle Start Follicle diameter #14 mm 14 mm ,follicle ,18 mm hCG hCG12 hCG16 hCG19
I
II
III
3 51 63 73 52 0 0
14 40 28 17 38 14 7
83 9 9 10 10 86 93
Note: All values are percentages. I 5 triple line; II 5 intermediate; III 5 homogeneous.
ments throughout stimulation cycles have not been published previously. We previously had studied endometrial wavelike activity in spontaneous cycles in a longitudinal design (11–13). To extend these observations and to gather optimal information about different aspects of the endometrium, we studied the endometrial characteristics and endometrial wavelike activity on fixed moments throughout COH cycles. The results confirm some of our findings in spontaneous cycles. Endometrial wavelike activity was recorded most frequently in the periovulatory phase. Similar wave patterns were observed in spontaneous and ovarian stimulation cycles, namely, waves from cervix to fundus, waves from fundus to cervix, random activity, and opposing waves (11, 12). However, in ovarian stimulation cycles endometrial wavelike activity was more pronounced in all corresponding phases of the cycles than in the phases of the spontaneous cycles. This difference perhaps may be explained by higher estrogen levels in ovarian stimulation cycles. Cross-sectional studies of the effect of ovarian stimulation on the endometrium suggest that the endometrium increases in thickness and that a triple-line pattern develops in the course of the cycle (14). The present longitudinal study in COH cycles confirmed this. A significant positive correlation between endometrial thickness and the presence of endometrial wavelike activity was found only in two phases of the cycle, i.e., before the start of ovarian stimulation and in the luteal phase. This may imply that at the start of the COH cycle the thickness of the endometrium facilitates the US detection of these waves. Theoretically, a higher number of follicles in the follicular phase of COH cycles with a subsequent rise of estrogens would result in a thicker endo-
FERTILITY & STERILITYt
metrium and a more pronounced endometrial wavelike activity. However, no relationship was detected between characteristics of the follicles (number and diameter) and endometrial wavelike activity, perhaps reflecting supranormal estrogen stimulation. The expected number of ongoing pregnancies occurring in 35 COH cycles is 5 (95% confidence interval [CI] 2–11); the observed number was 0. We have no explanation for this significant difference. One explanation could be that repeated US observations interfere with the embryo implantation process. However, observations of normal pregnancy rates in spontaneous cycles in women exposed to the same repeated US observations (13) appear to contradict this. The conception cycles in our previous study (13) showed less activity, and no waves from fundus to cervix were seen after ovulation. Controlled ovarian hyperstimulation cycles, on the other hand, showed an increase in endometrial wavelike activity in all phases of the cycle and in some cycles even waves from fundus to cervix after ovulation. These results support the contention that the embryo may need a quiet environment for nidation and a coherent, finetuned activity pattern of the endometrium, as has been suggested previously (13). References 1. Gonen Y, Asper RF. Prediction of implantation by the sonographic appearance of the endometrium during controlled ovarian stimulation for in vitro fertilization (IVF). J In Vitro Fert Embryo Transfer 1990; 7:146 –52. 2. Imoedemhe DAG, Shaw RW, Kirkland A, Chan R. Ultrasound measurement of endometrial thickness on different ovarian stimulation regimens during in-vitro fertilization. Hum Reprod 1987;2:545–7. 3. Lenz S, Lindenberg S. Ultrasonic evaluation of endometrial growth in women with normal cycles during spontaneous and stimulated cycles. Hum Reprod 1990;5:377– 81. 4. Randall JM, Fisk NM, McTavish, Templeton AA. Transvaginal ultrasonic assessment of endometrial growth in spontaneous and hyperstimulated menstrual cycles. Br J Obstet Gynaecol 1989;96:954 –9. 5. Abramowics JS, Archer DF. Uterine endometrial peristalsis—a transvaginal ultrasound study. Fertil Steril 1990;54:451– 4. 6. Birnholz JC. Ultrasonic visualization of endometrial movements. Fertil Steril 1984;41:157– 8. 7. Chalubinski K, Deutinger J, Bernaschek G. Vaginosonography for recording of cycle-related myometrial contractions. Fertil Steril 1993; 59:225– 8. 8. De Vries K, Lyons EA, Levi CS, Lindsay D. Contraction of the inner third of the myometrium. Am J Obstet Gynecol 1990;162:679 – 82. 9. Lyons EA, Taylor PJ, Zheng XH, Ballard G, Levi CS, Kredenster JV. Characterization of subendometrial myometrial contractions throughout the menstrual cycle in normal fertile women. Fertil Steril 1991;55: 771– 4. 10. Martinez Gauda M, Yoshiba T, Bentsson LP. Propagated and nonpropagated myometrial contractiona in normal menstrual cycles. Am J Obstet Gynecol 1973;115:107–11. 11. IJland MM, Evers JLH, Dunselman GAJ, van Katwijk C, Lo CR, Hoogland HJ. Endometrial wavelike movements during the menstrual cycle. Fertil Steril 1996;65:746 –9. 12. IJland MM, Evers JLH, Hoogland HJ. Velocity of endometrial wavelike activity in spontaneous cycles. Fertil Steril 1997;68:72–5. 13. IJland MM, Evers JLH, Dunselman GAJ, Hoogland HJ. Relation between endometrial wavelike activity and fecundability in spontaneous cycles. Fertil Steril 1997;67:492– 6. 14. Friedler S, Schenker JG, Herman A, Lewin A. The role of ultrasonography in the evaluation of endometrial receptivity following assisted reproductive treatments: a critical review. Hum Reprod 1996;2:323–5.
283