Transvaginal color Doppler determination of the ovarian and uterine blood flow characteristics in polycystic ovary disease

Transcription

1 FERTILITY AND STERILITY Copyright < American Society for Reproductive Medicine Printed on acid-free paper in U. S. A. Transvaginal color Doppler determination of the ovarian and uterine blood flow characteristics in polycystic ovary disease Fatma A. Aleem, M.D., Ph.D.* Mladen Predanic, M.D., M.sc. Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, The Brookdale Hospital Medical Center, Brooklyn, New York Objective: To determine blood flow characteristics of ovarian and uterine arteries in patients with endocrinologically and clinically confirmed polycystic ovary disease (PCOD) in comparison with blood flow parameters observed in patients with spontaneous ovulatory cycles. Design: Controlled clinical study. Setting: Patients from Infertility Service in a tertiary care institution. Patients: Forty patients with confirmed PCOD and 50 control patients in various phases of spontaneous menstrual cycles. Main Outcome Measure: Using transvaginal color Doppler sonography to determine ovarian morphology, ovarian blood vessel visualization rates, and ovarian and uterine arteries blood flow parameters (resistance index, pulsatility index, and maximal peak velocity). These parameters were correlated with serum hormone levels. Results: Polycystic ovaries showed typical vascular pattern: increased stromal vascularity, a positive correlation between increased blood velocities and serum LH levels, and a trend toward lower resistance index and pulsatility index values, whereas uterine arteries revealed significantly increased resistance index and pulsatility index values. Conclusions: The observed specific intraovarian and uterine vascular pattern in PCOD patients may provide additional data for conventional endocrinologic and ultrasonic diagnostic methods for PCOD. Fertil Steril 1996;65:510-6 Key Words: Polycystic ovary, color Doppler, ovarian and uterine blood flow Polycystic ovary disease (PCOD) is a topic likely to generate greater controversy about etiology or pathogenesis than any other disease in gynecological endocrinology. Polycystic ovary disease remains an enigma because of the superficial nature of our knowledge regarding its etiology and the pathophysiology. However, several subgroups ofpcod caused by hypothalamic-pituitary, ovarian, or adrenal steroidogenic defects can be distinguished. Polycystic ovary disease, ultrasonically characterized by commonly enlarged ovaries with increased numbers of small follicles interspersed by abundant stroma (1), is a consequence of broad spectrum in hormone imbalance. Received March 23, 1995; revised and accepted September 29, * Reprint requests: Fatma A. Aleem, M.D., Ph.D., Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Brookdale Hospital, 1335 Linden Boulevard, Suite 100, Brooklyn, New York (FAX: ). 510 Aleem and Predanic PCO and color flow sonography The hemodynamicity of ovarian and uterine microcirculation is a "marionette" of hormonal changes in general and at local levels. Therefore, transvaginal color Doppler sonography was used in the assessment of uterine and ovarian blood flow changes in normal and stimulated menstrual cycles in correlation with E2 and P serum levels (2-5). Changes in the vasculature of the cycling ovary, described as rapid and profound, suggest that vascularization of the follicles may playa role in their maturation from early follicular phase onward (6-10). However, an increased ovarian vascularity also may reflect a response to specific signals for angiogenesis and/or accelerated metabolism in growing follicles. If it is assumed in PCOD that follicular development and maturation are impaired due to hormone imbalance or vascular impairment, then ovarian vascularity assessment may provide additional information about PCOD pathophysiology and may guide in choosing appropriate hormonal regimen treatment. Therefore, the goal of the present work was to study basic

2 ovarian vasculature and uterine blood flow in PCOD patients, to see if a correlation exists between blood flow data and ovarian morphology or hormone serum levels and to compare PCOD blood flow characteristics with those obtained from spontaneous ovulatory menstrual cycles. MATERIALS AND METHODS A total of 90 women from our Reproductive and Infertility Service, 40 patients with confirmed PCOD and 50 women with spontaneous ovulatory cycles, were included in the present study. The routine workup of our infertility patients consists of a transvaginal sonogram of pelvic organs and laboratory determination of serum hormone concentrations. Regarding the PCOD patients, the mean age of this group was 27.8 years (range 18 to 38 years). A mean body mass index (EM!) was 32.5 :::'::: 1.2 (mean :::'::: SEM). Clinical evidence of hyper androgen ism was noted, while serum levels oflh, FSH, LH:FSH ratio, PRL, T, E2, and P levels, measured in a reliable laboratory using RIA, were done on the same day of transvaginal color Doppler sonography examination. Thirty-seven patients had oligomenorrhea, two had amenorrhoea, and one patient had dysfunctional bleeding. Although, PCOD patients have tonic or steady state of serum hormone levels, patients with oligomenorrhea were examined with transvaginal color Doppler sonography between the 3rd and 8th day of the menstrual cycle to avoid possible interference of corpus luteum presence. The control group of patients with normal menstrual cycles was selected according to ultrasonic findings of normal uteri and ovaries with documented ovulation and menstrual cycle duration of 26 to 32 days for at least the previous 3 months. These patients had normal ovulatory cycles but were diagnosed to be infertile due to tubal or male factors. The mean age was 31.2 years, ranging from 18 to 42 years and the mean BMI was 30.2 :::'::: 2.4. A menstrual cycle was divided into four phases: 3rd to 8th day, 9th to 14th day, 15th to 21st day, and 22nd day to the end of the menstrual cycle. The longest menstrual cycle in the control group of patients was 32 days. This classification was done according to hormonal changes in spontaneous-ovulatory cycle to represent as close as possible hormonal and vascular hemodynamic changes throughout spontaneous ovulatory cycle. It was based upon reports published previously, in which blood flow changes were correlated with the E2 and P serum concentrations (2-5), Along with our standard practice of examining infertility patients with transvaginal conventional so- nography to rule out uterine and ovarian abnormalities, color flow sonography was added to the present study, with the consent of each patient, for the assessment of ovarian and uterine vascularity. A color Doppler ultrasound machine Ultramark 9-HDI (ATL Laboratories, Seattle, WA) with a 5-MHz transvaginal transducer was used. Ovarian volume, number of follicles, and the greatest diameter of the largest follicles were recorded. Ovarian volume was calculated by a simplified formula for a prolate ellipsoid: D1 X D2 X D3 X 0.52, where D1, D2, and D3 represented largest ovarian length, width, and depth diameters. Both right and left ovaries were observed and analyzed in each patient, revealing no statistical significance in morphological parameter values between them for both groups of patients. Therefore, the mean value for all ovarian morphological parameters was calculated and used in the statistical analysis. After the assessment of ovarian morphology, color Doppler was used for demonstrating the main uterine artery and ovarian vascular network within the ovary. The ascending branch ofthe main uterine artery was located in the transversal section, lateral to the uterus at the level of the cervicouterine junction, and ovarian blood vessels were visualized within the ovarian stroma of PCOD ovaries (away from the ovarian surface) or around the preovulatory follicle and corpus luteum in the ovaries of the control group. The lowest color pulsed repetition frequency of 400 Hz with wall filter of 50 to 100 Hz was used to achieve an enhanced blood flow image of the ovarian vascularity. Ovaries were assessed as avascular if no blood flow was demonstrated by color Doppler modality, whereas any blood flow signal, even if only one vessel was visualized, was recognized as a positive finding. After the visualization of blood vessels inside the ovarian stroma, blood flow characteristics were analyzed with pulsed Doppler. The angle of ins on at ion was changed to obtain maximum blood flow velocity and quality of pulsed Doppler signals. For uterine arteries, maximal blood velocity values were corrected by determining the angle between the ultrasound beam and the vessels. However, the angle between ultrasound beam and intraovarian blood vessels was impossible to obtain, therefore, the probe was tilted until the clearest pulsed Doppler waveform was achieved. Obtained pulsed Doppler signals were assessed in terms of blood flow velocities (peak systolic velocity) and vascular impedance to blood flow (resistance index and pulsatility index). When more than one pulsed Doppler signal was detected inside the ovarian stroma, only those waveforms with the lowest vascular impedance to blood flow were selected for Doppler analysis. The means of the resistance index, pulsatility index, and peak systolic velocity values were ob- Aleem and Predanic pea and color flow sonography 511

3 tained for each patient from three consecutive pulsed Doppler waveform signals. Statistical analysis was done by Pearson correlation test for comparison between hormone levels, ovarian morphological parameters, and blood flow data for uterine and ovarian vessels, and by oneway analysis of variance for comparison of measured parameters between PCOD patients and control group. Results were reported as arithmetical mean ± SEM and range of values. RESULTS The hormone serum concentrations obtained from PCOD patients the day oftransvaginal color Doppler examination were as follows: LH, 12.8 ± 2.4 miui ml (conversion factor to SI unit, 1.00); FSH, 5.8 ± 0.6 miu/ml (conversion factor to SI unit, 1.00); T, 50.1 ± 4.5 ng/100 ml (conversion factor to SI unit, 3.46); PRL, 13.5 ± 2.5 ng/ml (conversion factor to SI unit, 2.714); E 2, 92.9 ± 18.6 JLg/L (conversion factor to SI unit, 3.671); and P, 0.6 ± 0.1 JLg/L (conversion factor to SI unit, 3.18). The calculated mean value for the LH:FSH ratio was 2.2 ± 0.5. The mean ovarian volume in PCOD patients was 8 ± 0.7 ml with a range from 3.1 to 19.3 ml, which was significantly larger than ovarian volumes in the follicular phase of the spontaneous cycling ovaries (4.1 ± 0.3; P < 0.01). A significantly larger number of follicles in PCOD ovaries (10 ± 0.7) when compared with early follicular phase of control group (3.3 ± 0.3) also was found (P < 0.01). However, there was no significant difference for the largest diameter ofthe largest follicle between PCOD and sponta neously cycling ovaries. Color Doppler demonstrated an increased vascularity in PCOD ovaries compared with the early follicular phase ofthe ovulatory menstrual cycle. Blood flow visualization rate of 88% for the PCOD group compared with only 50% in normal cycling ovaries in the patients from 3rd to 8th day was found (Fig. 1). A difference in the blood flow pattern also was noted. One prominent blood vessel within the ovarian stroma was observed in PCOD ovaries (Fig. 2A), whereas none or very weak color signal was demonstrated in normal cycling ovaries (Fig. 2B). However, an increased visualization rate of 83% for preovulatory follicles (one prominent blood vessel around developing follicle, 9th to 14th day) (Fig. 2C) and 100% visualization rate for corpora lutea blood vessels were observed. Depending on the size and the shape of the corpus luteum, color flow sonography showed an abundant vascular area or dense "halo" of blood vessels around the corpora lute a (Fig. 2D). No significant difference was found between the pulsed Doppler measurements of the left and right 512 Aleem and Predanic pea and color flow sonography pea Normal ovulatory cycle Days of the menstrual cycle Figure 1 Color Doppler visualization rates (%) in PCOD patients and 40 women with spontaneously cycling ovaries. uterine arteries, as well as for the lowest index values between left and right ovaries in PCOD patients, which allowed us to use the average value of these in statistical analysis. Blood flow characteristics (resistance index, pulsatility index, and peak systolic velocity) obtained from ovarian blood vessels in PCOD ovaries and control group of patients are summarized in Table 1. A significant difference was found for resistance index and pulsatility index values between PCOD ovaries and ovaries in the follicular phase (3rd to 8th day, P < 0.01). Although resistance index and pulsatility index values obtained from the corpus luteum blood vessels showed low values, especially in the late luteal phase (22nd day to the end of the cycle), a statistical significance was not achieved. However, a significant difference was noted between blood flow velocities (peak systolic velocity) in PCOD ovaries and corpora lutea (P < 0.01), indicating significantly increased blood velocities due to abundant angiogenesis within the corpus luteum. When blood flow parameters obtained from the ovarian stromal vessels in PCOD patients were correlated with ovarian morphological characteristics, no statistically significant difference was observed. The vascular impedance to blood flow was influenced neither by ovarian volume nor by the number offollicles. Furthermore, no correlation was found between serum hormone concentrations and resistance index or pulsatility index values, except for a positive correlation between LH levels and increased peak systolic velocity in ovarian vessels (P < 0.001) (Fig. 3). Additionally, a trend toward lower resistance index and pulsatility index values with an increase in LH serum levels, as well as a trend in higher T levels and, consequently, higher peak systolic velocities were observed, but in both cases a level of statistical significance was not achieved. Table 2 presents a summary of uterine artery blood flow data obtained from PCOD and control groups of patients. A higher resistance to blood flow

4 Figure 2 (A), A characteristic vascular pattern of a polycystic ovary, a prominent dilated blood vessel within the ovarian stroma (arrow). (B), Early follicular phase of the spontaneously cycling ovarian vascularity noted with color Doppler. (e), Preovulatory follicle, note a blood vessel close to the follicular wall (arrow). (D), A corpus luteum surrounded with typical vascular pattern-a dense "halo" of engorged blood vessels (arrows). Insufficient knowledge of PCOD etiology and pathophysiology makes proposed medical or surgical therapy for PCOD patients desiring fertility empirical and the subject of discussion. Regarding ultrasonic ovarian appearance, we found a large number of PCOD ovaries with smaller mean volumes and a smaller number of follicles than published in previous studies (1,11,12). However, according to a previous study, up to 30% ofpcod ovaries will have normal volume (13), whereas PCOD-like ovaries can be noted in a broad spectrum of hormonal imbalance and even in 16% of normal ovulatory women (14). We hope with the help of transvaginal color Doppler sonography, a new noninvasive modality in the assessment of organ vascularity, to shine more light on pathophysiological mechanisms of PCOD. Estradiol serum levels are considered to play an important role as moderator of uterine and ovarian vascularity (2-4). Several investigators demonstrated a correlation between serum LH rise and Aleem and Predanic was found in the control group of patients during the early follicular phase (3rd to 8th day), implicating very low E2 levels. With the rise of E2 levels in the preovulatory follicular phase (9th to 14th day) and in the luteal phase combined with an increase in P serum concentrations, a decrease in resistance index and pulsatility index values was noted. However, no significant changes in blood flow velocities during the normal spontaneous menstrual cycle was noted. No significant correlation was found between serum hormone levels and obtained uterine artery resistance index, pulsatility index, or peak systolic velocity values in PCOD patients. DISCUSSION pea and color flow sonography 513

5 Table 1 Blood Flow Parameters Found in Ovarian Vessels of PCOD and Control Groups' No. of Resistance Pulsatility Peak systolic Group patients index index velocity PCOD :t :t :t 3.2 Days 3 to 8 (0.36 to 0.71) (0.52 to 1.61) (5.5 to 26.2) Control group Days 3 to :t 0.06t 1.87 :t 0.38t 9.6 :t 2.1 (0.57 to 1.00) (0.94 to 3.92) (6.5 to 17.0) Days 9 to :t :t :t 2.9 (0.49 to 0.69) (0.73 to 1.34) (10.0 to 15.7) Days 15 to :t :t :t 3.1 t (0.50 to 0.63) (0.71 to 1.09) (8.0 to 51.0) Days 22 to :t :t :t 2.6t (0.40 to 0.54) (0.51 to 0.99) (5.4 to 48.3) * Values are means :t SEM with ranges in parentheses. t p < 0.01 compared with PCOD patients. Table 2 Blood Flow Parameters Found in Uterine Artery of PCOD and Control Groups' No. of Resistance Pulsatility Peak systolic Group patients index index velocity PCOD :t :t :t 1.5 Days 3 to 8 (0.79 to 1.00) (1.78 to 6.70) (21.1 to 58.3) Control group Days 3 to :t :t :t 2.6 (0.89 to 1.00) (2.39 to 6.25) (23.3 to 46.4) Days 9 to :t 0.04t 1.66 :t 0.53t 33.4 :t 3.8 (0.76 to 1.00) (1.66 to 5.10) (21.3 to 48.0) Days 15 to :t 0.02t 2.39 :t 0.15t 40.5 :t 2.8 (0.78 to 1.00) (1.64 to 3.53) (22.0 to 60.0) Days 22 to :t 0.02t 2.15 :t 0.16t 43.9 :t 2.9 (0.70 to 1.00) (1.34 to 4.42) (22.4 to 68.1) * Values are means :t SEM with ranges in parentheses. t p < 0.01 compared with PCOD patients. blood flow changes in intrafollicular vessels during the periovulatory period (6,7,9). Mter follicular rupture and corpus luteum development, intraovarian vascularity is transformed into a characteristic blood flow pattern called "luteal conversion." This increased turbulent flow can be explained by changes in the spiral intraovarian vessel network due to formation of arteriovenous shunts during the luteal phase (15). These specific vascular changes in spontaneously cycling ovarian changes presumably will not be observed in PCOD ovaries due to the lack of cyclic changes of hormones. However, our findings of higher visualization rates for stromal blood vessels in PCOD ovaries, when compared with ovaries in the follicular phase of the control group of patients, as well as resistance index and pulsatility index values similar to those found in corpus luteum blood vessels, indicate that stromal vessels probably are dilated and engorged in PCOD ovaries. Visualization of distinct stromal arteries in PCOD ovaries L H ze.8 10.S , zo.s Ovarian blood velocity (CIII/S) Figure 3 Scattergram of the correlation test between serum LH values and the blood velocity within the ovarian stromal vessels (Pearson correlation test). Slope = (r = ; t = ; df= 34). 514 Aleem and Predanic pea and color flow sonography as early as cycle days 3 to 8 is a first significant difference in vascular pattern from spontaneously cycling ovaries and is not correlated to cyclic serum E2 level. A significant decrease in vascular impedance to blood flow in the ovarian artery (16) and in vessels around the follicles, in correlation with an increase in the number of follicles and serum E2 concentrations (10), was observed after ovarian stimulation with gonadotropin preparations. Another study showed that PCOD patients stimulated with clomiphene citrate reveal low serum E2 levels and increased vascular impedance to blood flow in the main ovarian arteries (17). The authors suggested low serum E2 levels as responsible for impaired ovarian circulation. With respect to this data, iflow E2 levels are indeed the cause of poor ovarian vascularity in PCOD patients and consequently an inadequate response to hormonal therapy, then it would not be possible to observe higher blood vessel visualization rates in stromal ovarian vessels in our examined group ofpcod ovaries versus ovaries in early follicular phase of the control group. Furthermore, our observations of high vascular impedance of uterine arteries in PCOD patients showed similarity with blood flow parameters obtained in the early follicular phase of the normal cycle (3rd to 8th day), when low E2 levels usually are present. It appears that low E2 concentrations found in PCOD patients is the cause of high vascular impedance in uterine arteries but could not be the moderator of increased vascularity and vessel dilatation in their ovarian vessels. Therefore, it is possible that some factor other then E2 level could be the cause for increased stromal vascularity in PCOD ovaries. A linear increase in the capillary cross-sectional area of the theca interna has been observed after the LH surge in the spontaneously cycling ovaries (18). This increase in capillary area (blood flow) was attributed to vasodilatation, because there was no

6 corresponding increase in the number of blood vessels (18). Additionally, some authors reported that, at the time of the LH peak, an angiogenesis and penetration of blood vessels (with increased peak blood flow velocities) into the granulosa layer can be detected with color and pulsed Doppler (9). Our data showed a significant positive correlation between higher LH serum levels and an increase in blood flow velocities, as well as a trend toward lower resistance index and pulsatility index values with the increase of serum LH in PCOD patients. Elevated LH serum levels could be implicated as a reasonable cause of aforementioned hemodynamic changes in the stromal vessels of polycystic ovaries. However, it is likely that LH is not involved directly in vascular network hemodynamic changes but may be involved through the action of a mediator(s)-probably prostaglandins (PGs). One may speculate that PGs may playa secondary role as potent vascular vasodilatators in polycystic ovaries, which can be observed with transvaginal color Doppler sonography as an increased visualization rate of ovarian stromal blood vessels. Luteinizing hormone stimulates PG biosynthesis and steroidogenesis in preovulatory follicles before ovulation (19). A substantial increase in the production ofpge2a and PGF2a 4 to 5 hours after gonadotropin stimulation has been reported (20). It appears that PGF 2 is a stimulator of ovarian ovulation, whereas PGE 2 seems to suppress ovulation (21). Among prostaglandins, PGE2 and PGI2 are assumed to be potent vasodilatators, which markedly increase local blood flow (22), such as in corpora lutea (23), causing vascular "luteal conversion." However, a continuous infusion of LH throughout the cycle, in experimental animals, causes unbalance between PGE2 and PGF 2 levels, suppressing ovulation with an increase in ovarian blood flow (24). In another study, a significant reduction of peak systolic blood velocity within the preovulatory follicle vessels was observed (despite an adequate LH surge) after ingestion of paracetamol drug, pointing out a direct correlation between LH levels, prostaglandin activity, and blood flow changes (25). In conclusion, we found a vascular pattern for polycystic ovaries consisting of characteristic ovarian stromal vascularity with positive correlation between an increase in blood flow velocities and serum LH levels and a trend toward lower values for mean resistance index and pulsatility index values, whereas uterine arteries showed increased resistance index and pulsatility index values, possibly because oflow serum E2 levels. Therefore, we believe that assessment of ovarian and uterine vascularity may add greater understanding of the ovarian hemodynamic changes to traditional endocrinologic and ultrasonographic findings of PCOD. Such information hopefully may provide us with an answer as to why many PCOD patients are resistant to available treatment regimens. REFERENCES 1. Takahashi K, Eda Y, Okada S,Abu-MusaA, Yoshino K, Kitao M. Morphological assessment of polycystic ovary using transvaginal ultrasound. Hum Reprod 1993;8: Scholtes MCW, Wladimiroff JW, van Rijen HJM, Hop WCJ. Uterine and ovarian flow velocity waveforms in the normal menstrual cycle: a transvaginal Doppler study. Fertil Steril 1989;52: Steer CV, Campbell S, Pampiglione JS, Kingsland CR, Mason BA, Collins WP. Transvaginal colour flow imaging ofthe uterine arteries during the ovarian and menstrual cycles. Hum Reprod 1990;5: de Ziegler D, Bessis R, Frydman R. Vascular resistance of uterine arteries: physiological effects of estradiol and progesterone. Fertil Steril 1991;55: Kmjak A, Kupesic-Urek S, Schulman H, Zalud 1. Transvaginal color flow Doppler in the assessment of ovarian and uterine blood flow in infertile women. Fertil SterilI991;56: Collins W, Jurkovic D, Bourne T, Kurjak A, Cambell S. Ovarian morphology, endocrine function and intra-follicular blood flow during the peri-ovulatory period. Hum Reprod 1991; 6: Campbell S, Bourne TH, Waterstone J, Reynolds KM, Crayford TJB, Jurkovic D, et al. Transvaginal color blood flow imaging of the periovulatory follicle. Fertil Steril 1993; 60: Balakier H, Stronell RD. Color Doppler assessment offolliculogenesis in in vitro fertilization patients. Fertil Steril 1994;62: Bourne TH, Jurkovic D, Waterstone S, Campbell S, Collins WP. Intrafollicular blood flow during human ovulation. Ultrasound Obstet Gynecol 1991; 1: Weiner Z, Thaler I, Levron J, Lewit N, Itskovitz-Eldor J. Assessment of ovarian and uterine blood flow by transvaginal color Doppler in ovarian-stimulated women: correlation with the number of follicles and steroid hormone levels. Fertil Steril 1993;59: Pache TD, de Jong FH, Hop WC, Fauser BCJM. Association between ovarian changes assessed by transvaginal sonography and clinical and endocrine signs of the polycystic ovary syndrome. Fertil Steril 1993;59: Fos R, Hull M. Ultrasound diagnosis of polycystic ovaries. Ann NY Acad Sci 1993;687: Yeh HS, Futterweit W, Thornton JC. Polycystic ovarian disease: US features in 104 patients. Radiology 1987; 163: Gadir AA, Khatim MS, Mowafi RS, Alnaser HMI, Muharib NS, Shaw RW. Implications of ultrasonically diagnosed polycystic ovaries: 1. Correlations with basal hormonal levels. Hum Reprod 1992;7: Merce LT, Garces D, Barco MJ, de la Fuente F. Intraovarian Doppler velocimetry in ovulatory, dysovulatory and anovulatory cycles. Ultrasound Obstet GynecoI1992;2: Deutinger J, Reinthaller A, Bernaschek G. Transvaginal pulsed Doppler measurement of blood flow velocity in the ovarian arteries during cycle stimulation and after follicle puncture. Fertil Steril1989;51: Schurz B, Schon HJ, Wenzl R, Wolfgang E, Huber J, Reinold Aleem and Predanic pea and color flow sonography 515

7 E. Endovaginal Doppler flow measurement of the ovarian artery in patients with a normal menstrual cycle and with polycystic ovary syndrome during in vitro fertilization. J Clin Ultrasound 1993;21: Cavander JL, Murdoch WJ. Morphological studies ofthe microcirculatory system of peri ovulatory ovine follicles. BioI Reprod 1990;42: Larson L, Olafsson J, Hellberg 0, Brannstrom M, Selstam G, Hedin L. Regulation of prostaglandin biosynthesis by luteinizing hormone and bradykinin in rat preovulatory follicles in vitro. Prostaglandins 1991;41: Le Maire WJ, Yang NST, Behrman HH, Marsh JM. Preovulatory changes in the concentration of prostaglandins in rabbit graafian follicles. Prostaglandins 1973;3: Richman KA, Wright KH, Wallach EE. Local ovarian effects of prostaglandin E2 and F2 on human chorionic gonadotro- pin-induced ovulation in the rabbit. Obstet GynecoI1974;43: Raud J. Vasodilatation and inhibition of mediator release represent two distinct mechanisms of prostaglandin modulation of acute mast cell-dependent inflammation. Br J PharmacoI1990;99: Alila HW, Corradino RA, Hansel W. A comparison of the effects of cyclooxygenase prostanoids on progesterone production by small and large bovine luteal cells. Prostaglandins 1988; 36: Brannian JD, Terranova PF. Involvement of prostaglandins in LH-induced superovulation in the cyclic hamster. Prostaglandins 1991;41: Bourne TH, Reynolds K, Waterstone J, Okokon E, Jurkovic D, Campbell S, et al. Paracetamol-associated luteinized unruptured follicle syndrome:effect on intrafollicular blood flow. Ultrasound Obstet Gynecol 1991; 1: Aleem and Predanic pea and color flow sonography j I I j