This is a randomized controlled trial on the comparison of the ongoing pregnancy rate in frozen-thawed embryo transfer cycles with spontaneous ovulation and hCG-induced natural cycles.

Embryo cryopreservation is essential nowadays. It allows the usage of surplus good quality embryos in frozen-thawed embryo transfer (FET) cycles which avoids embryo wastage and facilitates the adoption of transferring a small number of embryos in the fresh stimulation cycle so as to reduce the risk of multiple pregnancy during in vitro fertilization (IVF) treatment. Elective cryopreservation of all fresh embryos is also required in special circumstances where fresh transfer is undesirable, for example in cases where a high risk of ovarian hyperstimulation syndrome is anticipated. Fresh transfer is not advisable when serum progesterone level is elevated or hydrosalpinx is detected during ovarian stimulation.

The availability of FETs would increase the likelihood of successful pregnancy from a single superovulation and oocyte retrieval cycle. It has been estimated that in modern IVF programmes which incorporates embryo cryopreservation, up to 42% of all conceptions could be derived from FET (Borini et al, 2008).

Although the pregnancy rate in FET is comparable with stimulated IVF cycles, the optimal regimen to prepare the endometrium for implantation is not yet well proven. By a recent Cochrane review, there is no evidence to prove the use of one regimen in preference to another (Ghobara T 2008).

Natural cycles (NC) are considered the best regimen as it is physiological, where natural conception occurs in. However, there are two most commonly employed methods to time the ovulation in natural cycles. One is to use ultrasound (USG) to monitor the follicular growth and endometrial thickness, with the use of human chorionic gonadotrophin (hCG) to trigger ovulation in the presence of a dominant follicle around 17-18mm in diameter. Another is to monitor the blood hormonal concentration to detect the luteinising hormone (LH) surge associated with natural ovulation. There is scarce information in this area as only one randomized controlled trial compared these two approaches. The authors aimed to recruit 240 subjects, but the study was prematurely terminated as significant results were resulted in the first interim analysis when the sample size reached 124. The ongoing pregnancy rate was 31.1% in the group with spontaneous LH surge and 14.3% in the hCG-induced group (Fatemi et al., 2010).

The use of natural LH surge and hCG trigger in intrauterine insemination was compared too. A meta-analysis revealed a significantly higher clinical pregnancy rate in IUI cycles with natural LH surge than that with hCG trigger. However, the data from randomized controlled trial showed no significant difference. Also the data were contradictory when the analysis confined to a specific indication of subfertility (Kosmas et al., 2007). A cochrane review in 2012 also suggested there was no evidence to advise that any regimen was better than another (Cantineau et al., 2012).

This randomized controlled trial aims to compare the ongoing pregnancy rate of FET in spontaneous NC and hCG-induced NC. The hypothesis of this trial is that the ongoing pregnancy rate of FET is similar for spontaneous NC and hCG-induced NC. The advantage of hCG-induced NC is likely a reduction in the duration of monitoring for timing FET when compared with spontaneous NC.

Inclusion Criteria:

• Age of women <43 years
• Regular menstrual cycles ranging from 21-35 days with not more than 4 days variation between cycles
• Undergoing FET in natural cycles
• Normal uterine cavity as shown on saline sonogram performed before the IVF cycle or normal uterine cavity shown on pelvic scanning during the stimulated IVF cycle
• Endometrial thickness >=8mm in both stimulated IVF and FET cycles

Exclusion Criteria:

• >3 previous IVF cycles
• >6 embryos replaced without pregnancy
• Irregular menstrual cycles
• Subjects requiring clomid-induced cycles and hormonal replacement (HRT) cycles
• History of previous FET cycles within the study period
• Blastocyst transfer
• Preimplantation genetic diagnosis treatment
• Use of donor oocytes
• Presence of hydrosalpinx not corrected surgically prior to FET
• Recurrent miscarriage
• Serum progesterone level on the day of LH surge or day of hCG >5 nmol/L or 1.5 pg/L (these patients will be retrospectively excluded)
• Refusal to join the study