Assisted reproductive technology

Transcription

1 Assisted reproductive technology FERTILITY AND STERILITY Vol. 60, No.2, August 1993 Copyright 'c; 199:~ The American Fertility Society Printed on acid-free paper in U. S. A. Natural cycle in vitro fertilization-embryo transfer at the University of Ottawa: an inefficient therapy for tubal infertility* Paul Claman, M.D. t Minerva Domingo, M.D. Peter Garner, M.D. Arthur Leader, M.D. John E. H. Spence, M.D. Goal Program, Department of Obstetrics and Gynecology, Ottawa Civic Hospital, Uaiversity of Ottawa, Ottawa, Ontario, Canada Objective: To describe our experience with natural cycle IVF making clinical and endocrine comparisons with our standard stimulated cycle IVF program. Design: We attempted 75 natural IVF-ET cycles with hcg given to preempt the LH surge and compared these with 450 attempts at standard superovulation IVF -ET done in our unit during the same time period. Patients: Natural cycle patients are normally ovulating women < age 38. Superovulation IVF ET patients are all <41 years old. Patients in both groups had partners with normal semen parameters and tubal factor infertility. Main Outcome Measures: Cancellation rates, pregnancy rates per egg retrieval, per ET procedure, and luteal phase E 2 :P ratios of the treatment cycles are compared. Results: There were 35 of 75 (47%) natural cycle and 112 of 450 (25%) superovulation cycle cancellations. An egg was retrieved in only 24 of 40 (60%) natural cycles and 336 of 338 (99%) superovulation egg retrieval procedures. Pregnancy rates per ovum pick-up procedure were significantly higher: 65 of 338 (19%) in the superovulation versus 2 of 40 (5%) in the natural cycle groups. Pregnancy rates per ET were not significantly different between natural IVF -ET, 2 of 18 (11 %) and superovulation IVF-ET, 65 of 298 (22%). The E 2 :P ratios 5 days after ET were similar in both groups at 18 ± 4 after natural IVF-ET and 21 ± 18 after superovulation IVF-ET. Conclusions: [1] Cancellation rates for natural cycle IVF are very high. [2] Midluteal E 2 :P ratios are the same in both groups. [3] Pregnancy rates per egg retrieval are significantly lower for natural versus superovulation IVF-ET. [4] In our experience, natural cycle IVF-ET is an inefficient therapy for tubal infertility compared with superovulation IVF-ET. Fertil Steril 1993;60: Key Words: In vitro fertilization, embryo transfer, natural cycle, unstimulated cycle The first successful pregnancy after human IVF - ET was done by retrieving an oocyte during an unstimulated natural menstrual cycle (1). With refinements in controlled ovarian stimulation, natural Received January 22, 1993; revised and accepted April 19, * Presented at the 48th Annual Meeting of The American Fertility Society, New Orleans Louisiana, November 2 to 5, t Reprint requests: Paul Claman, M.D., 737 Parkdale Avenue, Ottawa, Ontario, Canada K1 Y 4E9. cycle IVF was abandoned in the early 1980s in favor of superovulation IVF leading to higher pregnancy rates (PRs) per egg retrieval procedure than could be achieved in the natural cycle (2). In recent years, the technological advances of ultrasound (US) guided egg retrieval replacing laparoscopy and a better understanding of the LH surge to ovulation interval (3) have rekindled interest in the use of unstimulated cycles for IVF -ET in the treatment of infertility (4-7). We attempted 75 natural cycle IVF -ET cycles and compared these with 450 superovulation cycle attempts at IVF -ET done during 298 Claman et al. Natural cycle versus superovulation IVF-ET Fertility and Sterility

2 the same time frame under the same laboratory conditions. MATERIALS AND METHODS Seventy-five natural cycle IVF-ET cycles were studied and compared with 450 superovulation IVF -ET cycles. Both groups of patients had tubal factor infertility with normal semen parameters (8). Patients with tubal factor infertility were offered natural cycle IVF-ET as an option to a standard IVF-ET superovulation protocol, but allocation was not randomized. Inclusion criteria for the natural cycle protocol include the following: age < 38, normal ovulatory function (regular cycles, normal BBT chart, and in-phase late luteal timed endometrial biopsy), serum FSH <10 on cycle days 3 to 5, and both ovaries accessible to transvaginal US-guided egg retrieval. Inclusion criteria in the superovulation IVF-ET-treated patients were age < 41 years in addition to the criteria noted above. The natural cycle protocol involved drawing a baseline serum LH before cycle day 9 and monitoring daily serum LH, E2, and US starting on cycle day 10. Human chorionic gonadotropin (Profasi; Serono Canada, Mississauga, Ontario, Canada) 2,500 IU 1M (30 cases) or 5,000 IU 1M (10 cases) was given 34 hours (35 cases) or 36 hours (5 cases) before ovum pick-up when a follicle> 18 mm diameter was seen on US in association with an E2 > 180 pg/ml (650 pmol/l). Cycles were canceled if LH levels were found to be more than two times the baseline value or if E2 levels fell during monitoring. Egg retrieval was done by transvaginal guided US. Analgesia at ovum pick-up included 50 to 200 Ilg IV fentanyl and in some cases medazolam 1 mg. Lidocaine, 0.5%, was injected into the vaginal mucosa of the lateral vaginal fornices after preparing the vagina with a normal saline rinse. Patients received cefazolin, 1 g IV, before ovum pick-up. Standard superovulation protocol used leuprolide acetate (Lupron; Abbott Pharmaceuticals, Montreal, Quebec, Canada) in the midluteal phase before hmg therapy as previously described (9). Fertilization with Percoll-treated sperm (10) and ET of a maximum of four embryos was performed in the same way for both superovulation and natural cycle IVF-ET. Luteal phase support using hcg 1,500 IU 1M every 3 days for three doses starting 1 day after ovum pick-up was given in both natural cycle and superovulation IVF -ET cycle patients. Progesterone luteal support was used in superovulation cycles when E2 was >2,700 pg/ml (10,000 pmol/l) at the time of the preovum pick-up hcg 5,000 IU ovulation trigger. Progesterone in oil (Steris Labs, Phoenix, AZ) 12.5 mg/d 1M was given starting on the day after the hcg trigger. On the day of ET, the P dosage was increased to 25 mg/d. If pregnant, P was continued until fetal heart activity was seen on US. Serum E2 and P were drawn 5 days after ET in both natural and superovulation IVF -ET treatment cycles. Endocrine tests (E2, P, FSH, LH, and (3-hCG) were all done in serum using commercial RIA kits (FSH, LH, and (3-HCG by Amerlex-M, Amersham Canada Ltd, Ontario, Canada and E2 by Pantex, Santa Monica, CA). Estradiol was assayed on the day of hcg trigger, day of ET, and 9 days after the hcg trigger (i.e., midluteal phase). Progesterone was assayed on the day of ET and 9 days after the hcg trigger. Pregnancy testing with a serum (3- hcg assay was done 17 days after ET. Pregnancy is defined by the identification of a gestational sac on US done at least 4 weeks after ET. A two-tailed Student's t-test was used to compare endocrine data after testing for normality, and x2 or Fisher's exact test was used where appropriate. RESULTS There were 35 of 75 (47%) natural cycle and 112 of 450 (25%) superovulation cycle cancellations (P < ) (Table 1). An egg was retrieved at ovum pick -up in only 24 of 40 (60%) natural cycles and in 336 of 338 (99%) superovulation ovum pick-up procedures (P < ). Of those ovum pick-ups producing at least one egg, there was a trend to having a better chance of at least one embryo being available for transfer in the superovulation IVF cycles with 18 of 24 (60%) natural IVF -ET and 298 of 336 (89%) superovulation cases, resulting in at least one embryo for transfer (P = 0.098). Pregnancy rates per cycle initiated were much higher in the superovulation 65 of 450 (14.4%) versus the natural cycle cases 2 of 75 (2.7%) P = Pregnancy rates per ovum pick-up were also significantly higher in the superovulation 65 of 338 (19%) versus 2 of 40 (5%) in the natural cycle groups (P = 0.044). Pregnancy rate per ET procedure appear similar in both groups with 2 of 18 (11 %) pregnant after natural IVF -ET and 65 of 298 (22%) after superovulation IVF-ET (P = 0.434). Miscarriage or ectopic pregnancies appear similar in both groups, with pregnancies going to term in 48 of 65 (74%) superovulation and 2 of 2 natural IVF-ET Vol. 60, No.2, August 1993 Claman et al. Natural cycle versus superovulation IVF-ET 299

3 Table 1 Results of Natural Cycle IVF-ET Versus Superovulation IVF-ET Nautral cycle IVF-ET Superovulation IVF-ET Probability No. cycles started No. of cancellations ;:,:1 egg found per ovum pick-up procedure ;:,:1 cleaving embryo per ovum pick-up if egg retrieved Pregnancies per ET procedure Pregnancies per ovum pick-up procedure Term pregnancies per ovum pick-up procedure Pregnancies per cycle started E 2 :P ratio 5 days after ovum pick-up* (47)* 24/40 (60) 18/24 (60) 2/18 (11) 2/40 (5) 2/40 (5) 2/75 (2.7) 18 ± (25) 336/338 (99) 298/336 (89) 65/298 (22) 65/338 (19) 48/338 (14) 65/450 (14.4) 21 ± 18 < < NSt NSt * Values in parentheses are percents. t NS, not significant. :j: Values are means ± SD. pregnancies (P = 0_99). There were 13 of 65 (20%) multiple pregnancies (8 twins, 4 triplets, and 1 quadruplet) in the superovulation IVF-ET group. Both pregnancies from natural cycle IVF-ET were singleton. The E2:P ratio 5 days after ET was similar in both groups at 18 ± 4 after natural IVF-ET and 21 ± 18 (n = 67) after superovulation IVF -ET (P = 0.556). Furthermore, the E2:P ratio 5 days after ET was not statistically different between natural IVF -ET 18 ± 4 or superovulation IVF -ET whether hcg (n = 39) 17 ± 8 (P = 0.7) or P (n = 28) 28 ± 25 (P = 0.227) was used for luteal support. Baseline LH:FSH ratios in natural cycle IVF -ET did not help predict which patients would be canceled with LH:FSH = ± in canceled versus ± in noncanceled cases (P = 0.6). The baseline LH:FSH of ± in patients canceled because of a premature LH surge was not significantly different from ± in nonsurgers. The E2 on the day of hcg trigger was the same whether an egg was (234 ± 67 pg/ml [858 ± 245 pmol/ml» or was not (241 ± 74 pg/ml [886 ± 272 pmol/ml» retrieved at ovum pick-up in the natural IVF -ET group. DISCUSSION The first successful IVF -ET trials used laparoscopic oocyte retrieval timed to the LH surge in unstimulated cycles. Forty-four oocytes were retrieved in 68 laparoscopic procedures. Because only 65 % of laparoscopic procedures led to successful retrieval of an egg and 68% of these eggs fertilized and cleaved for transfer (1), efforts at improving the success of IVF -ET concentrated on developing protocols for stimulating multiple follicular development (2). The importance of ensuring the availability of mature oocytes after an egg retrieval procedure was especially compelling in an era when laparoscopy and general anesthesia were needed for follicle aspiration. However, availability of minimally invasive methods for follicle aspiration (11) and a better understanding of LH surge physiology (3) have led to renewed interest in performing IVF ET in natural cycles. The potential advantages of natural cycle IVF ET are as follows. First, the costs to patient and facility are lower because only hcg is given and the amount of hormone and US monitoring needed is halved. Second, without hmg stimulation, the risks of ovarian hyperstimulation syndrome and multiple pregnancy are obviated. Third, the emotional "roller coaster" of IVF -ET is minimized by removing the almost obsessive daily concerns surrounding the ovarian response to hmg. Fourth, with maintenance of physiological steroid hormone concentrations found in the natural cycle, there may be a higher PR per transferred embryo. Although some centers have attempted natural cycle IVF -ET timed to the endogenous LH surge (5, 7), we and others (4, 6) chose to preempt the endogenous LH surge with hcg to ensure adequate time between LH-like stimulation and ovum pick-up (3) and to prevent the need to perform ovum pick-up procedures in late afternoon or early evening hours, which are impractical in our clinic setting. Although monitoring for natural cycle IVF-ET is much less intensive than during the protocol for superovulation IVF -ET, the cancellation rate of 47% was believed to be unacceptable in the natural IVF -ET cycle protocol, despite careful attention to excluding women with ovulatory dysfunction. Most cancellations occurred because of the endogenous LH surge despite choosing a low morning value E2 of >180 pg/ml (650 pmol/l) to indicate a later hcg injection (given at 10:00 to 11:00 P.M.). The 47% cancellation rate was significantly higher than 300 Claman et ai. Natural cycle versus superovulation IVF-ET Fertility and Sterility

4 the 25% cancellation rate in the superovulation protocol despite the fact that recent hmg protocol adjustments have been made to reduce an unacceptably high cancellation rate for superovulation IVF-ET employing GnRH analogue down regulation with hmg administration (12). Although transvaginal US-guided follicle aspiration is less traumatic than laparoscopy, it is not without risk of infection or bleeding and does involve intensive preparation with the presence of physician, nursing, and laboratory personnel as well as culture media preparation. In this experience with natural cycles, successful aspiration of the dominant follicle with a 3- to 4-mL aspirate was almost always achieved, but egg retrieval was only successful in 60% of these ovum pick-up procedures, despite multiple attempts at flushing. This is compared with a 99% chance of retrieving at least one egg in a superovulation cycle. This low success rate for retrieving the egg in unstimulated cycles is similar to those originally reported by laparoscopy (1) but lower than the 98% (4) and> 100% (1.6 eggs per aspiration) (6) reported by others using a similar protocol with hcg preempting the LH surge in natural cycles. Of those groups reporting the use of the endogenous LH surge alone in unstimulated cycles, oocytes were obtained in 88% to 100% of cycles (5, 7). The only explanation for the lower 60% successful egg retrieval rate in this series, compared with that reported by others, is possibly the policy of canceling cases with an endogenous LH surge. Canceling LH surge cycles was also the reported protocol at the University of Southern California (6). Operator inexperience is an unlikely problem, noting that we have performed five to eight transvaginal US-guided ovum pick-up procedures in superovulated cycles weekly since early 1989 with almost all procedures yielding more than one egg. With an egg available after ovum pick-up, fertilization leading to a cleaving (2- to 8-cell stage) embryo available for transfer occurred 60% of the time in unstimulated cycles. With multiple eggs available most of the time after superovulation IVF-ET, at least one cleaved embryo was transferred in 89% of these cases. With only 18 ET procedures completed after natural cycle IVF -ET, there was no demonstratable significant difference in the chances for ET after ovum pick-up (P = 0.098) compared with superovulation IVF-ET cases. A 60% overall fertilization rate per egg is also observed in our IVF -ET cycles done for tubal disease when superovulation is employed. These rates seem lower than the 85% to 100% fertilization rates per egg after unstimulated IVF -ET reported by others (4-7) but similar to fertilization rates reported by most IVF ET clinics after superovulation IVF-ET (13). A higher fertilization rate per egg after unstimulated IVF -ET might be expected because the single egg retrieved has matured before the LH stimulus, as opposed to a somewhat heterogeneous cohort of eggs aspirated from both small and large follicles after superovulation IVF-ET. However, fertilization rates of 60% even after an egg is aspirated in a natural IVF -ET cycle may be a realistic expectation for many IVF -ET laboratories. Pregnancy rates per ovum pick-up are higher at 19% for stimulated versus 5% for natural cycle IVF-ET (P = 0.044). These higher PRs per ovum pick-up for stimulated cycles would be expected because of the higher chance for eggs and embryos to be transferred after an ovum pick-up in superovulation IVF -ET cases. The two pregnancies in the 40 natural cycle ovum pick-up cases went to term with 48 of 65 superovulation IVF-ET pregnancies (26% early pregnancy loss) going to term after 338 ovum pick-up procedures. With only two natural cycle IVF -ET pregnancies, both going to term, there was no demonstratable difference in pregnancy loss rates between natural and superovulation IVF -ET. Pregnancy rates per ET procedure are similar after both natural (11 %) and superovulation IVF ET (22%) (P = 0.434) despite the fact that between one and four embryos (mean = 3) were transferred after superovulation IVF-ET. High luteal E 2 :P ratios have been implicated as a reason for lower than expected rates of implantation per embryo transferred after super ovulated IVF-ET (9). There was no difference between the E 2 :P ratios in the midluteal phase of natural cycle IVF -ET versus that observed after superovulation IVF-ET. In conclusion, Cancellation rates after the start of daily monitoring for natural cycle IVF -ET are very high. Significantly more natural cycle IVF -ET egg retrieval procedures end without retrieving an egg at all compared with superovulation IVF -ET. This leads to large numbers of expensive and unproductive procedure room and gamete laboratory preparations. In this protocol, natural cycle IVF ET was not found to be an efficient therapy for tubal infertility when compared with superovulation IVF -ET. REFERENCES 1. Edwards RG, Steptoe PC, Purdy JM. Establishing full term human pregnancies using cleaving embryos grown in vitro. Br J Obstet GynaecoI1980;87: Vol. 60, No.2, August 1993 Claman et al. Natural cycle versus superovulation IVF-ET 301

5 2. Fishel SB, Edwards RG, Purdy JM, Steptoe PC, Webster J, Walters E, et al. Implantation, abortion and birth after in vitro fertilization using the natural menstrual cycle or follicular stimulation with clomiphene citrate and human menopausal gonadotropin. J Vitro Fert Embryo Transfer 1985;2: Taymor ML, Seibel MM, Smith D, Levesue L. Ovulation timing by luteinizing hormone assay and follicle puncture. Obstet GynecoI1983;62: Foulot H, Ranoux C, Dubuisson J-B, Rambaud D, Aubriot F -X, Poirot C. In vitro fertilization without ovarian stimulation: a simplified protocol applied in 80 cycles. Fertil Steril 1989;52: Ramsewak SS, Cooke ID, Li T-C, Kumar A, Monks NJ, Lenton EA. Are factors that influence oocyte fertilization also predictive? An assessment of 148 cycles of in vitro fertilization without gonadotropin stimulation. Fertil Steril 1990;54: Paulson FJ, Sauer MV, Francis MM, Macaso TM, Lobo RA. In vitro fertilization in unstimulated cycles: the University of Southern California experience. Fertil Steril 1992;57: Taymor ML, Ranoux CJ, Gross GL. Natural oocyte retrieval with intravaginal fertilization: a simplified approach to in vitro fertilization. Obstet Gynecol 1992;80: World Health Organization. WHO Laboratory manual for the examination of human semen and semen-cervical mucus interaction. 2nd ed. Cambridge: The Press Syndicate of the University of Cambridge, Claman P, Domingo M, Leader A. Luteal phase support in in-vitro fertilization using gonadotrophin releasing hormone analogue before ovarian stimulation: a prospective randomized study of human chorionic gonadotrophin versus intramuscular progesterone. Hum Reprod 1992;7: Tanphaichitr N, Agulnick A, Seibel MM, Taymor ML. Comparison of the in vitro fertilization rate by human sperm capacitated by multiple tube swim up and percoll gradient centrifugation. J Vitro Fert Embryo Transfer 1988;5: Ramsewak SS, Kumar A, Welsby R, Mowforth A, Lenton EA, Cooke ID. Is analgesia required for transvaginal single follicle aspiration in in vitro fertilization? A double blind study. J Vitro Fert Embryo Transfer 1990;7: Meldrum DR, Wisot A, Hamilton F, Gutlay AL, Kempton W, Huynh D. Routine pituitary suppression with leuprolide before ovarian stimulation for oocyte retrieval. Fertil Steril 1989;51: Medical Research International, Society for Assisted Reproductive Technology, The American Fertility Society. In vitro fertilization-embryo transfer (lvf-et) in the United States: 1989 results from the IVF-ET Registry. Fertil Steril 1991;55: Claman et al. Natural cycle versus superovulation IVF-ET Fertility and Sterility