Difficulty in recruiting participants.

This is a randomised controlled trial on the effect of endometrial injury in the cycle preceding the frozen-thawed embryo transfer (FET) cycles. The study hypothesis is that endometrial injury will increase the ongoing pregnancy rate in FET cycles.

Consecutive women attending subfertility clinic at Queen Mary Hospital, University of Hong Kong who are scheduled for FET treatment will be recruited. Women who have normal uterine cavity will be randomized into study and control groups in 1 to 1 ratio according to a computer-generated randomization list.

For patients in the study group, in the cycle immediately preceding the scheduled FET treatment, an endometrial biopsy would be arranged on day 21-23 of the menstrual cycles and they will be instructed to use non-hormonal means of contraception during that cycle. For patient in the control group, a similar procedure would be performed on the same timing during the preceding cycles without entering the uterine cavity, ie endometrial biopsy catheter entering the endocervical canal without entering the endometrial cavity. All patients will then proceed to FET treatment in the next cycle as scheduled.

The FET would be carried out as per our protocol. In short, patients attend the clinic daily from 18 days before the next expected period for the determination of serum E2 and luteinising hormone (LH) concentrations until the LH surge, which was defined as the day on which the LH level was above 20 IU/L and doubled the average of the LH levels over the past three days. For patients with irregular menstrual cycles or no ovulation demonstrated during natural cycle monitoring, clomiphene citrate (CC, Clomid, Merrell, Staines, U.K.) 50-100 mg will be given daily for five days from Days 3-7. The cycle will be monitored by blood tests from day 10 of the cycle as above. FET was performed on the third day after the LH surge. Up to two frozen-thawed embryo(s) can be transferred. No luteal phase support is used. On-going pregnancy rates, defined as viable fetuses beyond 10-12 gestational weeks, between the two groups will be compared.

We should measure endometrial thickness on LH+1 in natural and clomid-induced cycles.

Inclusion Criteria:

• Undergoing FET treatment cycles
• Normal uterine cavity as shown on transvaginal scanning or saline sonogram done at baseline
• endometrial thickness >=8mm during the stimulated IVF cycle and FET cycle

Exclusion Criteria:

• Pregnancy in previous IVF or FET cycles
• Presence of hydrosalpinx not surgically corrected prior to FET
• Presence of endometrial polyp or fibroid distorting uterine cavity
• IVF cycles carried out for preimplantation genetic diagnosis
• Arrange for blastocyst transfer
• Use of donor oocytes