Programming in vitro fertilization for a 5- or 3-day week

Transcription

1 Ass~st.d.reproductive tec"no'ogy FERTILITY AND STERILITY Copyright I[J 1991 The American Fertility Society Printed on acid-free paper in U.S.A. Programming in vitro fertilization for a 5- or 3-day week Essam S. Dimitry, M.D.* Susan A. Bates, M.D. Thordur Oskarsson, M.D. Raul Margara, M.D. Robert M. L. Winston, M.D. The Institute of Obstetrics and Gynaecology, Royal Postgraduate Medical School, Hammersmith Hospital, London, United Kingdom Objective: To schedule oocyte retrievals on either 5 or 3 days per week in in vitro fertilization (IVF) cycles. Design: Human chorionic gonadotropin (heg) administration was delayed by at least 24 hours in patients undergoing superovulation to avoid egg collections on weekends (group 1). Encouraged by the results, a further prospective study in which oocyte retrievals were programmed for only 3 days a week was undertaken (group 2). Setting: Hammersmith Hospital, a tertiary referral unit. Patients: All patients undergoing IVF treatment were included. Main Outcome Measures: To schedule oocyte retrievals on either 5 or 3 days per week. Results: Only 4.0% of egg collections (12/303) occurred on weekends compared with 12.6% (22/ 175) before delaying heg. In group 2 (n = 215), only four egg collections (2.1 %) had to be performed out of schedule. Delaying administration of heg had no detrimental effects. Conclusions: Delayed administration of heg allows scheduling oocyte retrievals on either 5 or 3 days per week, leading to a substantial decrease in out-of-hours oocyte retrievals, reducing cycle costs, and facilitating efficiency. The method is applicable to assisted reproduction specialized units as well as district general hospitals that use gonadotropin-releasing hormone analogs in their superovulation regimens before IVF, gamete intrafallopian transfer, or intrauterine insemination. Fertil Steril 55:934, 1991 Until recently, protocols for ovarian stimulation in assisted reproduction employed clomiphene citrate and human menopausal gonadotropin (hmg), starting at the beginning of a natural cycle. This made a 7 -day service essential, increasing expense and inconvenience. Templeton et all showed that use of oral contraceptive pills and norethisterone allowed a rigid regimen of ovulation induction and oocyte retrieval. Wardle et al. 2 were the first to describe the use of progestogens in the preceding menstrual cycle in an in vitro fertilization (lvf) program. They were able to perform oocyte retrievals on only 2 days a week. Progestogens have been shown to be effective by Received August 9, 1990; revised and accepted January 16, * Reprint requests: Essam S. Dimitry, M.R.C.O.G., Institute of Obstetrics and Gynaecology, Royal Postgraduate Medical School, Hammersmith Hospital, DuCane Road, London W12 OHS London, United Kingdom. other authors. 3-6 However, these studies used superovulation without gonadotropin-releasing hormone analogs (GnRH-a), necessitating intensive monitoring of urinary or serum luteinizing hormone to detect premature luteinization and spontaneous ovulation, which may occur in 30% and 15% of cycles, respectively, leading to cancellation of the egg collections. 7 The introduction of GnRH _a 8,9 in conjunction with hmg has eliminated the risk of premature luteinization and simplified cycle programming. Zorn et al. lo described a cycle programming protocol in which use of norethisterone/hmg combined with GnRH -a led to only three weekend oocyte retrievals in 119 cycles (2.5%). They reported a pregnancy rate (PR) of 26% per embryo transfer (ET). Vauthier and Lefebvre ll programmed oocyte retrievals for a 5-day working week using two formulations of injectable D-Trp6 -luteinizing hormone- 934 Dimitry et al. Programming IVF for a 5- or 3-day week

2 releasing hormone giving a PRjETof31 % and 27%, respectively. However, this program included work on Saturdays. At Hammersmith Hospital, before the use of the GnRH -a, buserelin acetate, 42.7% of oocyte retrievals were performed out of hours. In the first 98 patients treated with buserelin acetate, the number of weekend oocyte retrievals was reduced to 10.2%.12 Abdalla et a1. 13,14 and Conaghan et a1. 15 have shown that GnRH -a allow delayed administration of hcg after the follicles have reached the standard diameter criteria without cycle compromise. This study evaluated the impact of adjusting cycle programming to 3 or 5 workdays per week on the number of canceled treatment cycles and on achievement of clinical pregnancy. Standard Protocol MATERIALS AND METHODS Since April 1988, standard treatment regimen16 commenced on day 2 of the menstrual period with daily administration of 1.5 mg buserelin acetate nasal spray (Suprefact; Hoechst, London, United Kingdom) for a minimum of 14 days. Pituitary suppression was confirmed by measuring a serum estradiol (E2) level of <73 pmoljl. Daily hmg (Pergonal; Serono Ltd, Welwyn Garden City, United Kingdom) injections were then begun on a Wednesday. After 5 days, the dose of hmg was altered, if necessary, according to E2 levels. Follicular growth was then monitored by daily E2 levels and vaginal ultrasound. Human chorionic gonadotropin (hcg) was given when the largest three follicles reached a mean diameter of at least 17 mm with consistent E2 levels.17 Oocyte retrieval was scheduled 32 to 36 hours later. Our laboratory methods have been previously described.18 After IVF, a maximum of three embryos were transferred to the uterus. Patients thought to be at high risk of triplet pregnancy were advised to have only two embryos transferred.19 Progesterone was given for luteal phase support immediately after ET, as previously described.18 Serum ~-hcg levels were measured 14 days after oocyte retrieval and clinical pregnancy confirmed by sonographic visualization of a gestational sac 2 weeks later. Between April and July 1988, the standard protocol was followed in 212 consecutive patients. Of these, 175 reached oocyte retrievals, which were performed on 7 days of each week.12 Delayed HCG Protocols Group 1 From September 1988 to March 1989, 348 consecutive treatment cycles were commenced. The standard regimen, as described above, for superovulation and monitoring was adhered to, starting hmg injections on Wednesdays. Administration of hcg was, however, delayed by at least 24 hours after follicular and E2 criteria were met in those patients whose oocyte retrieval would normally fall on the weekend or on public holidays. Indications for out-of-hours oocyte retrievals were either excessive rise in serum E2 levels > 13,500 pmoljl with imminent risk of ovarian hyperstimulation syndrome or if E2 levels failed to rise on 2 consecutive days. Group 2 From July 1989 to December 1989, 215 consecutive treatment cycles were commenced. The standard regimen for superovulation and monitoring was used, but hmg injections were started on Thursday instead of Wednesday. Administration of hcg was delayed to facilitate scheduling of oocyte retrievals only on Monday, Wednesday, and Friday for unit organization reasons. Indications for out-of-hours oocyte retrievals were the same as in group 1. Statistical Analysis Statistical analysis data were carried out using X2 test with Yates (continuity) correction where appropriate. Results are presented as X2 statistics, df, and P values followed by 95% confidence intervals (CI) for the difference in percentages. RESULTS Table 1 shows patient profile in the delayed and standard subgroups in groups 1 and 2. With the exception of more gravid patients in delayed subgroups (51.55% versus 30.2%; X2 = 7.777; P value = 0.006), there was no statistically significant difference between delayed and standard subgroups. Table 2 shows the outcome of IVF treatment cycles in the standard protocol group and of those in group 1 and group 2. Cancellation of cycles (n = 37) in the standard protocol group was because of poor superovulation (21 patients with <3 follicles and serum E2 levels Dimitry et al. Programming IVF for a 5- or 3-day week 935

3 Table 1 Patient Profile in the Delayed and Standard Subgroups in Group 1 and Group 2 Delayed Standard (n = 76) (n = 227) X2 Age (y) <30 14 (18.4)a 35 (15.4)a 31 to (39.5) 81 (35.7) X2 = 1.10 >35 32 (42.1) 111 (48.9) Gravidity None 36 (47.4) 127 (55.9) ~1 40 (52.6) 100 (44.1) X2 = 1.36 Parity None 65 (85.5) 186 (81.9) ~1 11 (14.5) 41 (18.1) X2 = 0.29 Cause of infertility Tubal 54 (71.0) 159 (70) Unexplained 11 (14.5) 29 (12.8) Male factor 3 (4) 7 (3.1) X2 = 0.81 Others 8 (10.5) 32 (14.1) Delayed Standard Probability (n = 101) (n = 86) X2 Probability 19 (18.8)a 20 (23.2)a P = (33.7) 24 (27.9) X2 = 0.95 P = (47.5) 42 (48.9) P = 0.24 P = 0.59 P = (48.5) 60 (69.8) 52 (51.5) 26 (30.2) 92 (91.1) 76 (88.4) 9 (8.9) 10 (11.6) 60 (59.4) 46 (53.5) 26 (25.7) 22 (25.5) 11 (10.9) 12 (14) 4 (4) 6 (7) X2 = 7.78 P = X2 = 0.14 P = 0.71 X2 = 1.43 P = 0.70 a Values in parentheses are percents. that failed to rise above 3,000 pmoljl), poor semen quality (5 patients with sperm separation < 0.2 X 10 6 sperms/ml on 2 occasions), imminent risk of ovarian hyperstimulation syndrome (7 patients with serum E2 levels> 13,500 pmoljl before hcg administration), and personal circumstances (4 patients). In group 1, 45 cycles were canceled before oocyte retrieval. This was because of poor superovulation (31), poor semen quality (3), imminent risk of ovarian hyperstimulation syndrome (12), and personal circumstances (6). Similarly in group 2, there were 28 abandoned cycles because of poor superovulation (20), poor semen quality (3), and imminent risk of ovarian hyperstimulation syndrome (5). There was no statistically significant difference in the number of canceled cycles among the three groups (X 2 = 2.546; P = 0.280). Figure 1 shows the distribution of oocyte retrievals throughout the week in the three groups. Out-of-hours egg collections in groups 1 and 2 were because of either raised serum E2 with imminent risk of ovarian hyperstimulation syndrome (10/12 and 3/4, respectively) or if the E2 levels had reached a plateau (2/12 and 1/4, respectively). The proportion of out-of-hours oocyte retrievals in groups 1 (4%) and 2 (2.1%) was reduced when compared with the standard group (12.6%) (P = , CI = 3.4% to 12.9%, and P = , CI = 5% to 15%, respectively). The proportion of cycles in which the administration of hcg was delayed was 0% in the standard protocol group (7 d/wk), 25.1 % in group 1 (5 d/wk) and 54% in group 2 (3 days/wk). This increase among the groups was statistically significant with a X2 statistic of (P < ), almost all of which was explained by the proportion increasing linearly as the number of days decreased (x 2 = 135.4), with no significant departure from this linear trend X2 = Table 2 Outcome of IVF Treatment After Cycle Programming With Buserelin Acetate/HMG Standard Group 1 Group 2 IVF treatment cycles Canceled cycles Oocyte retrievals Oocyte retrievals out of hours No. of cycles with delayed hcg No.ofET Clinical pregnancies/ cycle Clinical pregnancies/oocyte retrievals Clinical pregnancies/et /212 (17.5)a /175 (12.6) 0/175 (0) 141/175 (80.6) 51/212 (24.1) 51/175 (29.1) 51/141 (36.2) /348 (12.9) /303 (4.0) 76/303 (25.1) 249/303 (82.2) 94/348 (27.0) 94/303 (31.0) 94/249 (37.8) /215 (13.0) 187 4/187 (2.1) 101/187 (54.0) 165/187 (88.2) 78/215 (36.3) 78/187 (41.7) 78/165 (47.3) a Values in parentheses are percents. 936 Dimitry et al. Programming IVF for a 5- or 3-day week

4 o Sal Standard III Group 1 Sun Mon Tue Wed Thu Fri Figure 1 The effects of delayed heg on timing of oocyte retrieval. In standard group, 12.6% of oocyte retrievals occurred on Saturdays and Sundays. In group I, 4% of oocyte retrievals occurred on Saturdays and Sundays. In group 2, 2.1 % of oocyte retrievals occurred on Saturdays and Tuesdays. There was a rising trend in the proportion of patients reaching ET among standard protocol group, group 1 and group 2 (X 2 = 3.9, P = 0.048). Significantly more clinical pregnancies/cycle occurred in group 2 than in the standard protocol group (P = 0.008, CI = to -3.0%), whereas the difference between group 1 and the standard protocol group did not reach statistical significance (P = 0.5, CI = -10.1% to 4.9%). However, the increase in clinical pregnancies among the groups was statistically significant with a X2 statistic of 8.74 (P < 0.012), almost all of which was explained by the proportion increasing linearly as the number of days decreased (x 2 = 7.81), with no significant departure from this linear trend X2 = Hospital admissions for ovarian hyperstimulation syndrome were six and two, respectively, in groups 1 and 2. In neither group was delayed administration of hcg responsible for cycle cancellation or ovarian hyperstimulation syndrome. The aim of this study was to determine if by delaying the administration of hcg we could reduce out-of-hours work without compromise to the cycle. In each group, 4 % or less oocyte retrievals were performed out of hours, and no more cycles were canceled than in the standard protocol group. Delayed administration of hcg improves implantation and PRS. 15 A trend in this direction was seen in group 1 in which 25% of patients had hcg administration delayed. In group 2, in which >50% of patients were delayed for >24 hours, the clinical PR was 41.7% per oocyte retrieval. This could be because of either better oocyte maturity leading to increased fertilization or improved uterine receptivity leading to higher implantation rates. The results demonstrate that a delay in administration of hcg in the monitored cycle of IVF-ET leads to a substantial decrease in the number of outof-hours oocyte retrievals beyond that already achieved using buserelin acetate/hmg, without compromise to PR. It is possible using buserelin acetate/hmg to reduce weekend work to about 12% of patients. This can be reduced further to 2% by delaying administration of hcg even when only 3 working days per week are used. This reduces cycle cost and improves efficiency without increasing cancelled treatment cycles. When oocyte retrievals are carried out on 3 rather than 5 days per week, more hcg injections are delayed beyond standard criteria. There are no detrimental effects, and PRs may be improved. Data gained from this study are applicable to assisted reproduction specialized units as well as district general hospitals that use GnRH in superovulation regimens before IVF, gamete intrafallopian transfer, or intrauterine insemination. Such procedures can be scheduled on as few as 3 working days per week, without reducing PRs or increasing the number of canceled cycles. DISCUSSION The addition of buserelin acetate to our superovulation protocol significantly reduced out-of-hours oocyte retrievals and greatly simplified the organization ofthe program. However, cycle programming must not be done at the expense of cycle quality. Close cycle monitoring, as described above, and adjustments in the dose ofhmg, are important to avoid large numbers of canceled cycles because of either poor response or risk of ovarian hyperstimulation. Acknowledgment. We are grateful to Mr. Neil Alexander, Dip.Math.Stat., Department of Medical Physics, Royal Postgraduate Medical School, Hammersmith Hospital, London, United Kingdom, for statistical analysis. REFERENCES 1. Templeton A, Look PV, Lumsden MA, Angell R, Aitken J, Duncan A W, Baird DT: The recovery of pre-ovulatory oocytes using a fixed schedule of ovulation induction and follicle aspiration. Br J Obstet Gynaecol 91:148, 1984 Dimitry et al. Programming IVF for a 5- or 3-day week 937

5 2. Wardle PG, Foster PA, Mitchell JD, McLaughlin EA, Williams JAC, Corrigan E, Ray BD, McDermott A, Hull MGR: Norethisterone treatment to control timing of the IVF cycle. Hum Reprod 1:455, Frydman R, Forman R, Rainhorn D, Belaisch-Allart J, Hazout A, Testart J: A new approach to follicular stimulation for in vitro fertilization programmed oocyte retrieval. Fertil Steril 46:657, Rainhorn JD, Forman RG, Belaisch-Allart J, Hazout A, Fries N, Testart J, Frydman R: One year's experience with programmed oocyte retrieval for IVF. Hum Reprod 2:491, (Dub) Howard WF, Chihal HJ, Strain CA, Smith M: Programmed follicular stimulation reduces cycle cost and stress with no compromise in cycle quality: use of a modified programmed protocol. J Vitro Fert Embryo Transfer 5:343, Bates RG, Fielding PM, Lindsay KS, White NJ, Edmonds DK: Programmed gamete intrafallopian transfer (GIFT). Br J Obstet Gynaecol 95:1220, Steptoe PC, Edwards RG, Walters DE: Observations on 767 pregnancies and 500 births after human in vitro fertilisation. Hum Reprod 1:89, Fleming R, Adam AH, Barlow DH, Black WP, Macnaughton MC, Coutts JRT: A new systematic treatment for infertile women with abnormal hormone profiles. Br J Obstet Gynaecol 89:80, Fleming R, Coutts JRT: Induction of multiple follicular growth in normally menstruating women with endogenous gonadotropin suppression. Fertil Steril 45:226, Zorn JR, Boyer P, Guichard A: Never on a Sunday: programming for IVF-ET and GIFT. Lancet 1:385, Vauthier D, Lefebvre G: The use of gonadotropin-releasing hormone analogs for in vitro fertilization: comparison between the standard form and long-acting formulation of D-Trp-6- luteinizing hormone-releasing hormone. Fertil Steril 51:100, Rutherford AJ, Subak-Sharpe RJ, Dawson KJ, Winston N, Packham D, Margara RA, Winston RML: Programmed IVF treatment using buserelin/hmg for superovulation. (Abstr.) Presented at the 4th Annual Meeting of the European Society of Human Reproduction and Embryology, Barcelona, Spain, July 3 to 6, Published by the European Society of Human Reproduction and Embryology, in the Program Supplement, 1988, p Abdalla HI, Leonard T, Baber R, Kirkland A, Stocker P, Owen E, Studd J: In-vitro fertilization. Br Med J 296:1470, Abdalla HI, Baber R, Leonard T, Kirlkand A, Mitchell A, Power M, Owen E, Studd JWW: Timed oocyte collection in an assisted conception programme using GnRH analogue. Hum Reprod 4:927, Conaghan J, Dimitry ES, Mills M, Margara RA, Winston RML: Delayed human chorionic gonadotropin administration for in-vitro fertilisation. Lancet 1:1323, Rutherford AJ, Subak -Sharpe RJ, Dawson KJ, Margara RA, Franks S, Winston RML: Improvement of in vitro fertilisation after treatment with buserelin, an agonist of luteinising hormone-releasing hormone. Br Med J 296:1765, Hillier SG, Parsons JH, Margara RA, Winston RML, Crofton ME: Serum estradiol and preovulatory follicular development before in-vitro fertilization. J Endocrinol 101:113, The Fourth Report of the Voluntary Licensing Authority for Human In Vitro Fertilisation and Embryology. London, Medical Research Council and Royal College of Obstetricians and Gynaecologists, 1989, p Rutherford AJ, Dawson KJ, Subak-Sharpe RJ, Winston N, Margara RA, Winston RML: Can the risk of multiple pregnancy (triplet or greater) after IVF be eliminated? (Abstr.) Presented at Sixth World Congress In Vitro Fertilization and Alternate Assisted Reproduction, Jerusalem, Israel, April 2 to 7, Published by Serono, in the Program Supplement, 1989, p Dimitry et al. Programming IVF for a 5- or 3-day week