The efficacy of conversion from IUI to IVF-ET in infertility patients with hyper-response to ovulation induction: A retrospective study

With the advance of assisted reproductive technology, a combination of ovulation induction with intrauterine insemination (IUI) has achieved great success in the treat- ment of infertility. However, the pregnancy rate remains still unstable and depends on the condition of each indi- vidual case. To select the appropriate time window for IUI, the common way is to utilize the natural cycle or ovulation induction and monitor the follicle phase. Cycle- stimulated or non-cycle-stimulated IUI is considered as the first-line treatment for infertility caused by common etiological factors, including asthenozoospermia, semen liquefaction, sexual dysfunction, abnormal secretion of the female cervical mucus, and immunological complica- tions 1,2. In addition, many retrospective and prospective studies have confirmed that using gonadotropin (Gn) to stimulate ovary and increase the number of follicles can achieve a better fertilization and implantation rate, thus ultimately improving the pregnancy rate 3. Although this approach generally increases the clinical pregnancy rate, a subset of hyper-responsive patients reports multiple follicle development and over-stimulation of the ovary, resulting in an increased risk of multiple pregnancy, preterm birth, miscarriage, and obstetric complications. For these patients, successful pregnancy depends on careful control of the cycles, adjustment of the Gn dose, cancella- tion of the cycle or use of in vitro fertilization and embryo transfer (IVF-ET). The level of estradiol (E2) and the num- ber of follicles determine whether the cycles need to be can- celled or IVF-ET should be performed. Fortunately, with the development and application of gonadotropin-releasing hormone analogues (GnRH-a), the conversion from IUI to IVF-ET becomes more feasible. Furthermore, cycle cancel- lation is frustrating for both the patients and physicians, and poses huge burdens, both psychologically and financially, on the patients and their families. The conversion from IUI to IVF-ET can largely avoid this problem. Interestingly, to date, few studies have investigated the efficacy of the conversion from IUI to IVF-ET. In this present study, we directly addressed this and reported the efficacy of IVF-ET in patients who suffered from hyper- response to ovulation induction for IUI.

Subjects All the subjects in this study were chosen from pa- tients admitted to the Center for Reproductive Medicine Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2012 Jun; 156(2):159–163. 160 Table 1. Basic clinical information of the subjects. Conversion from IUI to IVF-ET Age (yr) duration of infertility (yr) BMI (kg2/cm) Basic FSH (IU/L) LH (IU/L) PRL (ng/mL) E2 (pg/mL) T (ng/mL) 30.90±4.02 4.82±3.16 20.46±1.97 4.82±1.50 6.31±5.29 19.89±9.66 40.40±21.87 0.82±0.50 at the First Affiliated Hospital of Sun Yat-sen University between Jan 2007 and Oct 2010. The study was approved by the ethics committee of Sun Yat-sen University and written consent was obtained from all subjects. Patients with liver and kidney disorders, acute or chronic infec- tious diseases and reproductive tract inflammation were excluded. All subjects had polycystic ovary syndrome (PCOS). Oviduct liquid passing test, hysterosalpingog- raphy and/or laparoscopy examinations confirmed that at least one side of the oviduct was clear. All patients received IUI with informed consent. Among these pa- tients, 50 cases displayed hyper-response of the ovary to the ovulation induction drugs and were switched to IVF- ET treatment with further informed consent. Infertility was caused by ovulation failure, cervical factors, immu- nological factors, mild endometriosis, male factors (such as asthenozoospermia and semen liquefaction), and other unexplained infertility. Conversion from IUI to IVF-ET A total of 50 patients had 50 cycles. Starting 3-5 days after the menstrual cycle, the basic evaluation was conducted by ultrasound examination. If the size of the poly-cysts was smaller than 10 mm, medication was ini- tiated with commonly prescribed Gn, such as human menopausal gonadotropin (HMG) or follicle-stimulating hormone (FSH), at the dose of 75-150 IU. After constant medication for 4-6 days, transvaginal B-ultrasound exami- nation was performed to monitor the development of fol- licles and the dose of the drugs was adjusted accordingly. Based on our clinical experience and previous literature, hyper-response to Gn was operationally defined as the occurrence of more than 4 follicles with a diameter larger than 14 mm or the level of E2 higher than 1500 pg/mL (ref. 4). Once hyper-response was found, we consulted with the patients and decided whether the cycle needed to be cancelled or IVF-ET needed to be performed. Most of the patients consented to switch to IVF-ET. If more than 3 follicles had a diameter larger than 17 mm or more than 2 follicles had a diameter larger than 18 mm, or the urine luteinizing hormone (LH) test was positive, human chorionic conadotropin (HCG) was injected intramuscu- larly at a dose of 5,000-10,000 IU. Approximately 35-36 h after the HCG injection, vaginal ultrasound-guided oocyte retrieval was performed by the puncture of ovary. Three days after the oocyte retrieval, embryo transfer was con- ducted. Luteal support was accomplished by intramuscu- lar injection of progesterone at a dose of 40 mg per day, or intramuscular injection of 2000 IU HCG every three days for a total of 4 days, or utilization of progesterone vaginally. Fourteen days after the embryo transfer, a morn- ing urine pregnancy test was performed. If the pregnancy test turned out to be positive, the medication was con- tinued. The occurrence of the embryo sac by ultrasound examination 3 weeks later indicated clinical pregnancy. If the embryo sac examination was negative, the medication was terminated. Semen collection and processing According to World Health Organization standard procedures, semen samples from the patients’ couples were collected after 3-7 days of sex abstinence using the masturbation sperm retrieval method. Semen samples were delivered to the laboratory within 30 min. After gra- dient centrifugation and upstream processing, the sperm concentration was adjusted to 1×10 6/mL. The volume of the semen for fertilization was 0.2-0.5 mL. Regular testing was performed on the semen samples before and after washing. Statistical analysis The data were expressed as mean ± standard devia- tion (SD). The statistical analysis was performed using Microsoft Excel software.

Basic clinical information of the subjects The average age of the subjects was 30.90 ± 4.02 yr, ranging from 23 to 42 yr. Of the patients, 33 had primary infertility and 17 had secondary infertility. The average duration of infertility was 4.82 ± 3.16 yr, ranging from 1 to 15 yr. The average body mass index (BMI) was 20.46 ± 1.97, ranging from 16.80 to 27.34. The average level of basic FSH was 4.82 ± 1.50 IU/L, ranging from 0.5 to 7.28 IU/L. The average level of LH was 6.31 ± 5.29 IU/L, ranging from 0.7 to 26.43 IU/L. The average level of prolactin (PRL) was 19.89 ± 9.66 ng/mL, ranging from 4.3 to 45.74 ng/mL. The average level of E2 was 40.40 ± 21.87 pg/mL, ranging from 10 to 98 pg/mL. The average level of testosterone (T) was 0.82 ± 0.50 ng/mL, ranging from 0.07 to 2.75 ng/mL (Table 1). Cycle parameters of the patients The total dose of Gn used was 1339.2 5 ±826.74 IU, ranging from 450 to 2200 IU. The average number of fol- licles larger than 14 mm was 8.24 ± 2.98, ranging from 4 Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2012 Jun; 156(2):159–163. 161 Table 2. Cycle parameters of the patients. Conversion from IUI to IVF-ET Gn total dose Number of follicles>=14mm Thickness of the endome- trium (mm) Ovum number Number of fertilized ovum Number of normal embryo- genesis Number of available embryos 1339.25±826.74 8.24±2.98 10.85±2.43 14.37±7.76 7.65±5.03 7.44±4.92 5.29±4.20 Table 3. Clinical efficacy of conversion from IUI to IVF-ET. Group Cycles Embryos transferred Clinical pregnancy rate Biochemical pregnancy Spontaneous abortion Oviductal pregnancy Single fetus Two fetuses Three fetuses Fresh transfer 42 87 (12/42) 28.5% 0 21732 Frozen transfer 21 55 (8/21) 38.1% 1 10521 to 22. Ovum was collected from 49 patients with an aver- age number of 14.37 ± 7.76, ranging from 2 to 35. Ovum was not collected from 1 patient. Ovum was fertilized in 49 patients. The average number of fertilized ovum was 7.65 ± 5.03, ranging from 1 to 20. The average number of normal embryogenesis was 7.44 ± 4.92, ranging from 1 to 20. The average number of available embryos was 5.29 ± 4.20, ranging from 1 to 18. The average thickness of the endometrium was 10.85 ± 2.43 mm, ranging from 6 to 19mm when HCG was injected (Table 2). Clinical efficacy of conversion from IUI to IVF-ET Fresh embryo transfer was cancelled in 6 cases to prevent the occurrence of ovarian hyper-stimulation syn- drome (OHSS). The fresh transfer had 42 cycles and a total of 87 embryos were transferred. Urine pregnancy test showed positive in 15 patients and fetal heart beat was de- tected in 12 patients by transvaginal ultrasound, of which 3 patients had two fetuses, 2 patients had three fetuses, and 7 patients had single fetus. In the remaining 3 cases, extrauterine pregnancy occurred in 1 patient and no fetal heart beat was detected in 2 patients. The overall clinical pregnancy rate of IVF-ET was 28.5% (12/42). Among the 6 patients who did not receive fresh embryo transfer, 1 patient to date has not undergone embryo transfer yet. In another case, after thawing, the frozen embryo was completed lyzed and thus no transfer was performed. A total of 21 cycles of frozen embryo transfer with up to 55 embryos were conducted for the patients who were not pregnant at the end of fresh embryo transfer cycles or who did not receive fresh embryo transfer in the first place. Urine pregnancy test was positive in 10 patients and fetal heart beat was detected in 8 patients by transvaginal ultrasound, of which 2 patients had two fetuses, 1 patient had three fetuses, and 5 patients had single fetus. For the remaining 2 cases, no fetal heart beat was detected in 1 patient and biochemical pregnancy occurred in 1 patient. The clinical pregnancy rate was 38.1% (8/21) (Table 3).

In our hospital, the cycle pregnancy rate of patients receiving IUI treatment fluctuates between 8% and 15%. The fresh cycle pregnancy rate is 28.5% and the frozen cycle pregnancy rate is 38.1%, which are slightly lower than those reported (57% and 48%, respectively) (ref. 4,5). A randomized, case-control study reveals that in infertility patients caused by male factors or mild endometriosis, the live birth rate of artificial insemination for patients with mild ovarian hyper-stimulation is higher than that for patients without any condition 6. A recent comprehensive analysis of patients with unexplained infertility shows that the pregnancy rate of artificial insemination with moni- tored ovulation cycles is 4%. In comparison, the natural pregnancy rate guided by mild hyper-stimulation of cycles is 8%. Moreover, the pregnancy rate of artificial insemina- tion combined with mild hyper-stimulation of cycles can reach 18% (ref. 7). In this study, all 50 patients who converted from IUI to IVF were hyper-responders. Based on the Rotterdam criteria, all patients were diagnosed as PCOS, which was considered an indication of potential hyper-responder. For these patients, the traditional effective treatment ap- proach for infertility, the IUI combined with ovulation in- duction, has profound disadvantages including high rates of cycle cancellation and multiple fetal implantations 8. Previous studies have showed that the rate of multiple fetal implantations is 7.5-29% (ref. 9) in patients receiving IUI and ovarian stimulation10. Most previous studies which compared the pregnancy rate between the IVF group and the conversion from IUI to IVF group reported that the clinical pregnancy rate was similar between the two groups. However, the im- plantation rate in the conversion from IUI to IVF group was much higher than that in the regular IVF group 4. A related finding is that the E2 level in the conversion from IUI to IVF group was significantly lower than that in the Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2012 Jun; 156(2):159–163. 162 regular IVF group. High levels of E2 may increase uterine contraction and thus adversely affect the IVF implanta- tion rate. However, some other studies report opposite findings 11. The application of Gn for ovulation cycle stimula- tion and IUI treatment has the following characteristics. Firstly, Gn improves the quality of the eggs and sperms during the therapeutic cycle, which increases the probabil- ity of pregnancy 12. Secondly, administration of HMG can correct a variety of abnormalities caused by critical luteal dysfunction, including low LH peak, abnormal LH secre- tion, abnormal follicular development, low levels of E2 secretion cycle, luteinized unruptured follicle syndrome, and ovarian incompetence. Thirdly, recombinant Gn can ameliorate the low levels of FSH in patients with PCOS. Fourthly, direct injection of treated sperms into the uter- ine cavity provides a reasonable therapy for the treatment of infertility caused by oligospermia, asthenospermia and cervical factors. Finally, application of Gn can achieve a remarkable pregnancy rate of 23% for patients with unex- plained infertility. To sum up, it is a simple, effective, and economic method 13. If such complications happen in IUI medical centers with IVF capabilities, the patients should be offered IVF treatment as an alternative. For IUI medical centers with- out IVF faciltiies, the patients should be transferred to other hospitals where IVF is routinely performed. Conversion to IVF can reduce the occurrence of complications and potential risks that the patients may experience. More importantly, larger numbers of ova and embryos can be obtained after conversion to IVF and this increases the pregnancy rate. Compared to con- trolled ovarian hyperstimulation (COH) followed by IVF, the conversion from IUI to IVF uses lower doses of Gn. For this reason, patients can have better endometrial re- ceptivity, the implantation rate can be increased, and the cost can be lowed 14. When there are more than 4 follicles larger than 14 mm on the day HCG was injected; multiple follicles can develop 15. In the present study, the average number of follicles obtained from 49 patients was 14.37, which is consistent with previous findings. These results suggest that IVF treatment should be considered when there are more than 4 dominant follicles larger than 14 mm on the day HCG is injected. The selection of appropriate ovulation stimulation program and handling of multiple follicle development should be considered with extreme caution for the treat- ment of infertility patients with hyper-responses. PCOS patients have the potential for hyper response to ovulation stimulation. Therefore, cancellation of the cycle or conver- sion to IVF should be well-prepared ahead of ovulation stimulation. For example, electrocardiogram, chest X-ray and other IVF-related examinations should be performed in advance. However, there is relatively insufficient sup- port to conduct the tube baby test (body check). Cycle cancellation can prevent the occurrence of OHSS, but it also poses huge burdens on the patients, both psychologi- cally and financially. In medical centers with the ability to perform IVF, most patients with hyper-response to ovulation stimulation prefer conversion to IVF to cycle cancellation. One study has reported that a combination of follicle rupture and IUI can be used in patients with multiple fol- licle development during the process of ovulation stimula- tion 16. On the one hand, this can increase the possibility for pregnancy. On the other hand, rupture of the follicles and controlling the number of the implanted embryos can reduce the likelihood of multiple pregnancy and over- stimulation of the ovary. Multiple pregnancy rates can also be reduced by aspiration of supplementary follicles followed by IUI (ref. 15). In the present study, we also man- aged to aspirate some of the ova using ultrasound-guided follicle aspiration. The aspirated ovum was used for IFV and the formed embryo was frozen. The remaining 1 or 2 mature follicles were used for IUI treatment or directed natural pregnancy. This strategy worked well for the most of our patients. Application of an antagonist is also feasible during the process of ovarian stimulation and IUI treatment. However, this reduces the clinical pregnancy rate, possibly due to its impact on the endometrial receptivity. Rather than pure antagonist treatment, mild stimulation of the ovary can be achieved by personalized therapeutic pro- gram. For example, during the first ovulation stimulation, clomiphene or letrozole may be individually administered; whereas in the later stage, Gn can be used for ovulation stimulation. Finally, for those patients with high BMI, appropriate pretreatment should be conducted before in vitro reproductive techniques are used. For example, im- provements of living habits and increasing the physical activity can be beneficial for losing weight. In summary, our results demonstrate that conver- sion to IVF-ET is an efficient method for patients with hyper-responses to ovulation induction during regular IUI stimulation cycle. In addition, it has a relatively high preg- nancy rate. However, the choice of in vitro reproductive technique varies in each case. For example, single embryo transfer may be considered to avoid multiple pregnancies for those patients with relatively good conditions 5. The

of this study is small number of subjects and lack of a control group in which the hyper-responsive pa- tients remain on IUI treatment. We believe that prospec- tive, multi-site, randomized, and case-control studies are urgently need to determine the feasibility and effective- ness of conversion to IVF in hyper-responsive patients with unknown causes or male factors infertility.

The work was supported by science and technology projects of Guangdong Province (2009B030801155) and research projects for the members of the Guangdong population and family planning committee (2010243). Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2012 Jun; 156(2):159–163. 163 CONFLICT OF INTEREST STATEMENT Author’s conflict of interest diclosure: The authors stated that there are no conflicts of interest regarding the publication of this article.

1. Karande VC, Korn A, Morris R, Rao R, Balin M, Rinehart J, Dohn K, Gleicher N. Prospective randomized trial comparing the outcome and cost of in vitro fertilization with that of a traditional treatment algorithm as first-line therapy for couples with infertility. Fertil Steril 1999;71:468-75. 2. Pashayan N, Lyratzopoulos G, Mathur R. Cost-effectiveness of pri- mary offer of IVF vs. primary offer of IUI followed by IVF (for IUI fail- ures) in couples with unexplained or mild male factor subfertility. BMC Health Serv Res 2006;6:80. 3. Guzick DS, Sullivan MW, Adamson GD, Cedars MI, Falk RJ, Peterson EP , Steinkampf MP . Efficacy of treatment for unexplained infertility. Fertil Steril 1998;70:207-13. 4. Nisker J, Tummon I, Daniel S, Kaplan B, Yuzpe A. Conversion of cycles involving ovarian hyperstimulation with inteuterine insemination to in-vitro fertilisation. Human Reproduction 1994;9(3):406-8. 5. Antman AM, Politch JA, Ginsburg ES. Conversion of high-response gonadotropin intrauterine insemination cycles to in-vitro fertil- ization results in excellent ongoing pregnancy rates. Fertil Steril 2002;77:715-22. 6. Tummon IS, Asher JL, Martin JSB, Tulandi T. R a ndomized controlled trial of superovulation and insemination for infertility associated with minimal or mild endometriosis. F e rtil Steril 1997;68:8-7. Haydardedeoglu B, Bagis T, Simsek E, Cok T, Hacivelioglu SO, Erkanli S. The impact of rescue in vitro fertilization converted from high- response gonadotropin intrauterine insemination cycles in terms of implantation and pregnancy rates as compared with matched controls. Fertil Steril 2009;92:137-42. 8. Olufowobi O, Sharif K, Papaioannou S, Mohamed H, Neelakantan D, Afnan M. Role of rescue IVF-ET treatment in the management of high response in stimulated IUI cycles. J Obstet Gynaecol 2005;25:166-8. 9. Schenker JG, Yarkoni S, Granat M. Multiple pregnancies following induction of ovulation. Fertil Steril 1981;35:105-23. 10. Callahan TL, Hall JE, Ettner SL, Christiansen CL, Greene MF, Crowley WF Jr. The economic impact of multiple-gestation pregnancies and the contribution of assisted-reproduction techniques to their inci- dence. N Engl J Med 1994;331:244-9. 11. Valbuena D, Simón C, Romero JL, Remohí J, Pellicer A. Factors re- sponsible for multiple pregnancies after ovarian stimulation and intrauterine insemination with gonadotropins. J Assist Reprod Genet 1996;13:663-8. 12. Zarrilli S, Paesano L, Colao A, Mirone V, Lombardi G, De Rosa M. FSH treatment improves sperm function in patients after varicocelec- tomy. J Endocrinol Invest. 2000;23(2):68-73. 13. Guanglun Z. Modern assisted reproductive technology. Beijing: People's Medical Publishing House, 2005. p132-4. 14. Xianhong T. The natural cycle of IVF - ET in assisted reproductive technology in the application . Foreign family planning/reproduc- tive health branch 2006;25(1):18-9. 15. De Geyter C, De Geyter M, Nieschlag E. Low multiple pregnancy rates and reduced frequency of cancellation after ovulation induction with gonadotropins, if eventual supernumerary follicles are aspi- rated to prevent polyovulation. J Assist Reprod Genet 1998;15:111-6. 16. Kun-ming L, Xiao-ming T, Hui-fen C, Fan Y, Zhi-qin C, Li-ping Z, Jian- zhi Y, Yu W, Jia-ping P , Jing-yu S. Preliminary Discussion of Patients of Multi-follicle Development during Induced Ovulation Altered to IVF-ET or Mature Follicle Puncture +IUI. Reproduc contracep 2008;28(6):1-5.