Accuracy of day three criteria for selecting the best embryos.

Abstract

Objective: Recently there is a heightened awareness of the need to control the risk of multiple gestations with IVF. Ideally, programs should not transfer more than two embryos to avoid the potential complications seen with higher-order multiples. The main obstacle to such an approach is the difficulty in choosing which two embryos to replace. To aid in the selection process, we perform blastocyst transfer (BT) in patients with more than three 8-cell embryos. However, the effort needed for blastocyst culture, compounded by concerns about the occasional sub-optimal performance of extended media, have led many programs to maintain their day 3 ET practice. In order to evaluate whether culture to day 5 is warranted for better embryo selection, we started a prospective study to assess the accuracy of day 3 criteria in identifying the best embryos. Design: Prospective study. Materials/Methods: This is an ongoing study of BT since September 2000. On day 3, the embryologist chose the 2 embryos that would have been selected for transfer that day and cultured them separately. On day 5, all embryos were examined to determine the best and second best blastocysts. Typically, two blastocysts were chosen for transfer with additional ones frozen on day 5 or 6. All embryos were cultured under oil in P1 with 10% SSS and moved on day 3 to Blastocyst medium with 10% SSS. Results: 65 patients had BT in this study to date. The mean age was 34.6 (27-41). The mean number of ≥8-cell embryos was 6.3. Picks were ≥8-cells and typically grade I (87%). All cycles made blastocysts and 74% had cryopreservation. The mean number of blastocysts was 2.9 on day 5 and 1.7 on day 6. Neither pick was chosen in 27 cycles. In 24 cycles, one pick was transferred, while both picks were transferred in only 14 cycles (22%). Of 78 non-transferred picks, 39 were frozen and 39 arrested, with both picks arresting in 6 cycles. The single best blastocyst was chosen from the picks in 27/65 cycles (42%). Conclusions: The value of morphologic criteria for cleavage-stage embryo selection may fall short when the transfer is limited to only two embryos. Although a reasonable number of day 3 picks did end up forming blastocysts by day 6, surprisingly, in 42% of our cycles, the chosen blastocysts were not from the day 3 picks selected by experienced embryologists. In the remaining 58% of the cycles at least one of the transferred blastocysts was from the picks. However, the single best blastocyst did not come from either pick in 38 out of 65 cycles. This could suggest that the majority of patients who conceived a singleton would not have achieved a pregnancy if they had a day 3 transfer of only the 2 picks. Therefore, we believe that blastocyst culture is the method of choice in this population.