Effect of baseline ovarian cysts on in vitro fertilization and gamete intrafallopian transfer cycles*

Transcription

1 FERTILITY AND STERILITY Copyright <> 1991 The American Fertility Society Printed on acid-free paper in U.S.A. Effect of baseline ovarian cysts on in vitro fertilization and gamete intrafallopian transfer cycles* Jeffrey M. Goldberg, M.D t Frederick A. Miller, M.S. Chad I. Friedman, M.D. William G. Dodds, M.D. Moon H. Kim, M.D. The Ohio State University, Columbus, Ohio The presence of ovarian cysts may compromise the success of in vitro fertilization (IVF) and gamete intrafallopian transfer (GIFT). We prospectively studied 212 consecutive ovulation induction cycles in 120 patients for IVF and/or GIFT. A baseline cyst was defined as any intraovarian cystic structure ~ 12 mm noted on ultrasonography before superovulation. Cycle outcomes were compared between patients with cysts (n = 62) versus those with no cysts (n = 150). There were no differences in follicular or luteal pbase lengths or amount of human menopausal gonadotropins used. Peak estradiol (E 2 ) levels were significantly lower and cancellation rates significantly higher in the cyst group. For noncanceled cycles, there were no significant differences in peak E2 levels, the mean number of follicles ~ 12 mm, mature oocytes retrieved, or ova transferred for GIFT or embryos for IVF. The pregnancy rates overall and for noncanceled cycles were not significantly different. Fertil Steril55:319, 1991 The presence of ovarian cysts on the baseline ultrasound (US) examination 'is frequently a cause for canceling controlled ovarian hyperstimulation for in vitro fertilization (IVF) and gamete intrafallopian transfer (GIFT). The concern is that the cysts may grow further, undergo torsion, rupture, or may compromise the potential for follicular development in response to gonadotropins. Postponing an IVF or GIFT cycle to await spontaneous resolution of the cyst is often a tremendous source of frustration and despair for the infertility patients who perceive these canceled cycles as lost opportunities for conception. Postponing a cycle is obviously preferable to canceling it after the patient has invested in it both financially and emotionally. Several investigators observed no differences be- Received April 23, 1990; revised and accepted October 19, * Presented at the 45th Annual Meeting of The American Fertility Society, San Francisco, California, November 11 to 16, t Reprint requests: Jeffrey M. Goldberg, M.D., Department of Obstetrics and Gynecology, The Ohio State University Hospitals, 1654 Upham Drive, Columbus, Ohio tween cycles with or without baseline cysts during superovulation alone 1 or for IVF. 2-4 Others have noted poorer outcomes of IVF cycles in the presence of baseline cysts. 5-7 It should be noted that the definition of what constituted a baseline cyst was not standardized between these studies. This prospective study was performed to determine if baseline ovarian cysts had any adverse effects on various parameters of IVF and GIFT cycles. MATERIALS AND METHODS Two hundred twelve consecutive controlled ovarian hyperstimulation cycles were evaluated prospectively in 120 infertility patients undergoing IVF and GIFT. A baseline ultrasonogram was performed before initiating controlled ovarian hyperstimulation using an ADR Ultramark 4 with a 5 MHz endovaginal transducer (Advanced Technology Laboratories, Inc., Bellevue, W A). A cyst was defined as any intraovarian cystic structure with a mean diameter ~ 12 mm. Controlled ovarian hyperstimulation was accomplished with human menopausal gonadotropins (hmg, Pergonal; Ser- Goldberg et ai. Ovarian cysts and IVF/GIFT 319

2 ono, Randolph, MA), two ampules per day beginning on cycle day 3. The dose of hmg was subsequently individualized according to the follicular response. In addition, some patients also received pure follicle-stimulating hormone (FSH, Metrodin; Serono), two ampules on each ofthe 1st 2 days of hmg, or clomiphene citrate (CC, Serophene; Serono) 100 mg on cycle days 3 through 7. Patients with previous canceled hmg cycles frequently underwent pituitary desensitization with the gonadotropin-releasing hormone agonist (GnRH-a) leuprolide acetate (LA, Lupron; TAP Pharmaceuticals, Chicago, IL) before administering hmg. The 1st day of hmg on LA was considered cycle day 3. The patients were monitored with serial ultrasonograms and serum estradiol (E2) levels measured by radioimmunoassay (Coat-a-Count Estradiol, Diagnostic Products Corp., Los Angeles, CA). In addition, serum luteinizing hormone (LH) was obtained to detect a premature LH surge. When;;::2 follicles were ;;::16 mm in mean diameter and E2 was ;;::500 pg/ml, 5,000 U of human chorionic gonadotropin (hcg, Profasi; Serono) was given intramuscularly. If these criteria were not met, a premature LH surge occurred, or the patient was felt to be at high risk for severe ovarian hyperstimulation, hcg was withheld, and the cycle canceled. Oocyte retrieval was performed by laparoscopy or transvaginal US-guided needle aspiration 34 hours later. The number and size of all follicles were recorded on the day ofhcg or the day of cancellation, as well as the peak E2 levels, and the amount of hmg used. The follicular phase length extended from day 1 of the cycle to the day of hcg administration, and the luteal phase length from the day after hcg to the onset of menses. Pregnancy was defined as a rising serum ~-hcg level and the visualization of an intrauterine gestational sac on transvaginal US. Statistical analysis was performed using the Student's t-test and x2, with Yates correction or Fisher's exact test where appropriate. RESULTS A baseline cyst was present in 62 of the 212 cycles (29.2%). Ten patients had 2 or 3 cyst cycles accounting for 24 (38.7%) of all cyst cycles. In 17 cycles there were 2 cysts, and in 4 cycles there were 3 cysts on baseline transvaginal US. s ranged in size from 12 to 45 mm (mean 21.3 ± 8.2). There was no significant difference in the mean age of the patients :n the cyst versus the noncyst group, 32.4 Table 1 Infertility History (n = 62) (n = 150) Primary infertility 45 (72.6)b 82 (54.7) Secondary infertility 17 (27.4) 68 (45.3) Factors' Tubal 22 (35.5) 63 (42.0) Endometriosis 30 (48.4) 51 (34.0) Adhesions 14 (22.6) 42 (28.0) Ovulatory 19 (30.6) 31 (20.7) Male 6 (9.7) 21 (14.0) Luteal defect 7 (11.3) 7 (4.7) Cervical 8 (12.9) 14 (9.3) Uterine 1 (1.6) 2 (1.3) Immune 8 (12.9) 10 (6.7) Unexplained 4 (6.5) 10 (6.7) P<0.05. b Values in parentheses are percents. c Exceeds 100% because of patients with multifactorial infertility. versus 32.7 years, respectively. The mean duration of infertility in the cyst group, 5.3 years, was significantly different from the noncyst group, 4.4 years (P < 0.05). The percentage of patients with primary infertility was also greater in the cyst group. Analysis of infertility factors demonstrated a significant difference for endometriosis only, with more affected in the cyst group (Table 1). There were no differences between the groups in the percentage of patients receiving FSH, CC, or LA in addition to hmg. As well, no controlled ovarian hyperstimulation regimen had a statistically significant influence on cycle cancellation rates either between the cyst and noncyst groups, or compared with the other controlled ovarian hyperstimulation regimens within each group. No differences were observed for the follicular or luteal phase lengths or for the number of hmg ampules utilized. The peak serum E2 levels overall were significantly lower in the cyst group; however, when only noncanceled cycles were considered, this difference was abolished (Table 2). The overall cycle cancellation rate was 43.5% in the cyst group and 28.0% for noncyst cycles (P < 0.05). Most ofthe cancellations were because of dominant follicles or low E2 levels. As for the individual reasons for cancellation, only premature LH surge reached significance, though the number of cases was very small (Table 3). In a similar percentage of cycles, oocyte retrieval was performed by transvaginal US. There were no significant differences in the mean number of follicles aspirated, oocyte maturity, fertilization rate, or num- 320 Goldberg et al. Ovarian cysts and IVF/GIFT Fertility and Sterility

3 Table 2 Ovulation Induction No." Canceled No." Canceled HMGalone HMG+FSH HMG+LA HMG+FSH+LA HMG+CC Ampules of hcgb Day ofhcg injectionb Luteal phase length b Peak E2 (pg/ml)b.e E2 in noncanceled cycles b Follicles> 12 mm b 12 (19.4) 22 (35.5) 1 (1.6) 18 (29.0) 9 (14.5) , o (16.7) 70 (46.7) 6 (4.0) 31 (20.7) 19 (12.7) , " Values in parentheses are percents. b Values are means. e P<0.05. ber of embryos transferred for IVF or oocytes transferred for GIFT (Table 4). Table 5 summarizes the pregnancy data. There were no differences in the pregnancy rates (PRs) overall or for noncanceled cycles. The spontaneous abortion rate was approximately l in both groups. Only 8 cysts (12.9%) were noted to have a slight increase in size, whereas in the remainder an equal proportion, 43.5%, were either unchanged or actually decreased in size during stimulation with hmg. In only six cases (2.8%) was there confusion about cyst versus follicle on US monitoring. Eleven cysts were found to be endometriomas on aspiration. The remainder were follicular or lutein cysts. There were no episodes of adnexal torsion or cyst rupture. The presence of a cyst did not impair oocyte retrieval. There was no significant difference in the mean size of the largest cyst in canceled versus noncanceled cycles, 21.8 ± 9.5 and 20.9 ± 7.1 Table 3 Canceled Cycles Overall" Dominant follicle Poor development LowE 2 Premature LH surgee Hyperstimulation risk Other d 27 (43.5)b 19 (30.6) 2 (3.2) 10 (16.1) 4 (6.5) 42 (28.0) 33 (22.0) 1 (0.7) 15 (10.0) 1 (0.1) 2 (0.1) " P<0.05. b Values in parentheses are percents. e P< d One cycle was canceled at the patient's request. One was canceled because of a very poor semen specimen. These cycles were not included in the x 2 statistical analysis because they were unrelated to the superovulation outcome. mm, respectively. Similarly, the number of baseline cysts, from 1 to 3, did not affect the incidence of cycle cancellation. In retrospect, it might have been useful to obtain an E2 level at the time of the baseline scan. These data were available on 22 cyst cycles (12 on LA). One patient had an E2 of 72 pg/ml, all the others were :::;;40 pg/ml. There was no correlation between the baseline E2 levels and the number or size of the cysts, or cycle outcome. A subgroup of 18 patients had cycles with and without baseline cysts. These 52 cycles (21 with cysts) were further analyzed separately exactly as for the entire study population. With each patient serving as her own control, differences in duration of infertility, previous conception, and infertility factors were eliminated. In this more homogeneous group, there were no significant differences in any variable. Table 4 Oocyte Retrieval Follicles aspirated Mature Immature Atretic Fertilization rate (%) No. of embryos/etb No. of oocytes/gift transfer No. of embryos cryopreserved/et " No. of ET /noncanceled IVF cycles 16 (59.3)e No. of oocyte retrievals by transvaginal US 14 (22.6) " Values are means. bet, embryo transfer. e Values in parentheses are percents. " (51.7) 36 (24.0) Goldberg et al. Ovarian cysts and IVF/GIFT 321

4 Table 5 Pregnancy Overall 3 (4.8)a 17 (11.3) N oncanceled 3 (8.6) 17 (16.0) PerET 2 (12.5) 5 (12.5) Per GIFT 1 (11.1) 12 (26.1) b Term 2 (66.7) 11 (64.7)C Spontaneous abortion 1 (33.3) 6 (35.3)d a Values in parentheses are percents. b Two patients had combination IVF-ET/GIFT. These were counted as GIFT pregnanciers. C Includes one set of twins. d One set of quadruplets reduced to twins subsequently terminated because of spontaneous ruptured membranes at 20 weeks' gestation. DISCUSSION A 56% incidence of ovarian cyst (:2:15 mm) formation after hmg cycles was reported by Tummon et al. 4 These cysts resolved spontaneously within 2 cycles. If the serum E 2 1evel was :::;;85 pg/ml, repeat ovulation induction with hmg was initiated. In 15 cycles that met these criteria, folliculogenesis was unimpaired, the cysts decreased in size, and six pregnancies were established. The authors did note that the hmg requirements for cyst cycles were increased over controls. Grazi et al. l evaluated 42 nonconception hmg cycles in 21 patients who had cycles with and without the presence of a baseline cyst. There were no differences in any of the follicular response parameters to hmg, and all the cysts decreased or resolved. Ovarian cysts also developed in response to GnRH-a therapy before hmg superovulation for IVF.8 During down regulation, 29% of patients on decapeptyl and 22.7% on buserelin acetate developed ovarian cysts (>25 mm). These cysts did not change in size and did not interfere with ovulation induction, the number of oocytes obtained, or fertilization and PRs. In a retrospective study of 97 IVF cycles stimulated with hmg, Hornstein et al. 2 reported no difference in 21 cyst cycles (:2:10 mm and :::;;45 mm) versus 76 noncyst cycles in peak E2 levels, day of hcg, number of follicles, oocytes, and embryos transferred, cycle cancellation or PRs. The amount of hmg utilized was not provided. Hung et al. 3 compared 14 patients with large ovarian cysts, between 3 and 5 cm (8 remnant ovarian cysts and 6 endometriomas), with 51 noncyst cycles for IVF or GIFT. There was no or minimal growth of the cysts during controlled ovarian hyperstimulation. As well, there were no episodes of torsion or rupture. These large cysts did not influence folliculogenesis, fertilization, or PRs. Several studies, on the other hand, do suggest that baseline ovarian cysts may compromise controlled ovarian hyperstimulation for IVF. Thatcher et al. 5 observed no differences in the number of oocytes recovered, fertilization and cleavage, or PRs between cyst versus noncyst groups for noncanceled cycles. They reported that the 40 cyst cycles had significantly lower peak E2 levels and a higher cancellation rate than the 86 controls. Further, they found that cycles with cysts 16 to 29 mm had a significantly higher baseline E 2, which indicates hormonal activity. These cycles were more likely to be canceled because of a precipitous E2 drop. Cycles with 30 to 60 mm cysts had normal baseline E2 levels, and cancellations in these cycles were because of poor response. Last, two prospective randomized studies assessed the effect of transvaginal US cyst aspiration versus no aspiration before controlled ovarian hyperstimulation for IVF. Rizk et al. 6 stated that cyst aspiration contributed to a marginal improvement in the outcome of IVF by providing more oocytes at aspiration. There were no differences between the groups in E2 levels, ampules of hmg used, day of hcg administration, number of follicles, or fertilization rates. In addition, 2 patients developed infection after aspiration of hemorrhagic cysts, which led to the loss of one ovary. Aboulghar et al. 7 conducted a similar study in 64 patients with inflammatory cysts and/or hydrosalpinges. Aspiration resulted in a lower cancellation rate, more 00- cytes retrieved, and more embryos transferred. Fertilization and PRs were not significantly different between the groups. They postulated that these masses may have a deleterious effect on the ovarian response to controlled ovarian hyperstimulation by mechanically compromising the ovarian blood supply or by reducing the space for follicular enlargement. In addition, they noted that the presence of a mass made oocyte retrieval technically more difficult. In the present study, cycles with baseline ovarian cysts had a higher incidence of cancellation. However, when only noncanceled cycles were evaluated, there were no differences in any of the cycle variabies. The PRs overall, and for noncanceled cycles only, were not significantly different. Although this study is prospective and larger than the studies cited, the reader is cautioned about interpreting these results. We evaluated 212 consecutive cycles. 322 Goldberg et al. Ovarian cysts and IVF/GIFT Fertility and Sterility

5 No attempt was made to match the cycles in the cyst and noncyst groups. As it turned out, the patients who had cyst cycles had a longer duration of infertility and a greater incidence of primary infertility. A subgroup of 18 patients, who underwent controlled ovarian hyperstimulation cycles in the presence and absence of a baseline cyst, was further analyzed separately. No differences in any of the parameters were noted in this subgroup though obviously the numbers, and hence statistical power, were diminished. Although there were no differences between the cyst and noncyst groups in the percentage of cycles supplemented with FSH, CC, or LA, the ovulation induction protocol was not standardized. Each cycle was individualized based on the response during previous controlled ovarian hyperstimulation cycles. In a randomized prospective study, we found no difference in IVF or GIFT cycles with hmg alone versus a "sweetened" regimen with two ampules of FSH added the 1st 2 days of hmg therapy.9 This confirms the findings of Benadiva et al. 10 All patients who received FSH in the present study were included in this protocol. In summary, controlled ovarian hyperstimulation cycles for IVF and GIFT had a significantly higher cancellation rate if an ovarian cyst was present on baseline ultrasonography. If the cycle progressed to follicle aspiration, there were no differences in peak E2 levels, oocytes retrieved, or fertilization and PRs. In addition, the PRs overall were not significantly different. Further studies are needed to determine if transvaginal aspiration of these cysts before controlled ovarian hyperstimulation improves the cycle outcome. Acknowledgments. We gratefully acknowledge Larry Sachs, Ph.D., The Ohio State University, for assistance with the statistical.analysis. REFERENCES 1. Grazi R, Taney FH, Gagliardi CL, Khoury A, Von Hagen S, Weiss G, Schmidt CL: The presence of ovarian cysts does not alter ovarian stimulation with gonadotropins. (Abstr. 085) Presented at the 44th Annual Meeting of The American Fertility Society, Atlanta, Georgia, October 10 to 13, Published in the Program Supplement, by The American Fertility Society, 1988, p S29 2. Hornstein MD, Barbieri RL, Ravnikar VA, McShane PM: The effects of base-line ovarian cysts on the clinical response to controlled ovarian hyperstimulation in an in vitro fertilization program. Fertil Steril52:437, Hung TT, Tsuiki A: Noninterference of ovarian cysts or endometriomas with the outcome of ovulation induction in assisted reproduction. Presented at the 36th Annual Meeting of the American College of Obstetricians and Gynecologists, Boston, Massachusetts, May 2 to 5, Published in the Program Supplement, by the American College of Obstetricians and Gynecologists, 1988, p Tummon IS, Henig I, Radwanska E, Binor Z, Rawlins R, Dmowski WP: Persistent ovarian cysts following administration of huma~ menopausal and chorionic gonadotropins: an attenuated form of ovarian hyperstimulation syndrome. Fertil Steril 49:244, Thatcher SS, Jones E, DeCherney AH: Ovarian cysts decrease the success of controlled ovarian stimulation and in vitro fertilization. Fertil Steril52:812, Rizk B, Kingsland C, Steer C, Mason B: Does cyst aspiration prior to ovarian stimulation improve the outcome in in-vitro fertilization? A prospective randomized trial. (Abstr. P204) Presented at the 45th Annual Meeting of The American Fertility Society, San Francisco, California, November 11 to 16, Published in the Program Supplement, by The American Fertility Society, 1989, p S Aboulghar MA, Mansour RT, Serour GI, Satter MA, Awad MM, Amin Y: Transvaginal ultrasonic needle guided aspiration of pelvic inflammatory cystic masses before ovulation induction for in vitro fertilization. Fertil Steril 53:311, Feldberg D, Ashkenazi J, Dicker D, Yeshaya A, Goldman GA, Goldman JA: Ovarian cyst formation: a complication of gonadotropin-releasing hormone agonist therapy. Fertil Steril51:42, Dodds WG, Goldberg JM, Friedman CI, Kim MH: Unpublished data 10. Benadiva CA, Ben-Rafael Z, Blasco L, Tureck R, Mastroianni L, Jr, Flickenger GL: An increased initial follicle-stimulating hormone/luteinizing hormone ratio does not affect ovarian responses and the outcomes of in vitro fertilization. Fertil Steril 50:777, 1988 Goldberg et al. Ovarian cysts and IVF/GIFT 323