Assessing the real advantage of day-5 embryo transfer requires taking into account cycle failure prior to transfer: an analysis of U.S. surveillance data, 2004–2006

Abstract

OBJECTIVE: Nonrandomized studies suggest that IVF cycles in which embryos are transferred at the blastocyst stage (day-5 transfers) have higher success rates than day-3 transfers, whereas randomized trials fail to show an advantage. Nonrandomized studies may overestimate the advantage of day-5 transfers by computing success rates only for procedures that progress to transfer, thus ignoring failures due to short embryo survival. This study expand previous analyses to include procedures that failed before a transfer.DESIGN: Retrospective cohort study using data from CDC's National Assisted Reproductive Technology (ART) surveillance system (NASS).MATERIALS AND METHODS: We analyzed data on IVF's of fresh non-donors from ART programs that perform most transfers either on day 5 or on day 3 in the U.S. Outcome was measured by whether or not the cycle was terminated at different stages of the procedure. Cumulative logit models with adjustment for patient and clinic factors were used to compare the outcomes of day5 and day3 programs.RESULTS: A total of 9,273 procedures from 35 clinics were identified in the study as the day-3 group; and 9,007 IVF cycles from 19 clinics as the day-5 group. Compared to the day3 group, procedures in the day5 group were less likely to be cancelled prior before retrieval(odds ratio (OR): 0.84, 95% confidence interval(95%CI): 0.77-0.91); but were more likely to be terminated before transfer, particularly for women over age 40 (OR: 1.85, 95% CI: 1.53-2.25). The odds of implantation failure were generally lower in day-5 programs than in day-3 programs (OR:0.92, 95%CI:0.87-0.98), as were the odds of pregnancy loss (OR:0.91, 95%CI:0.85-0.97), but not among women over 40 years of age (implantation failure, OR:1.48, 95%CI:1.11-1.97; pregnancy loss, OR:1.38, 95%CI:0.97-1.97). Live birth rates were slightly higher for the day-5 group (OR: 1.10, 95% CI: 1.03-1.18), but not among women over 40 years (OR:0.72, 95%CI:0.51-1.03). As expected, restricting the analysis to procedures that progressed to transfer exaggerated the advantage of day-5 transfer (data not shown).CONCLUSIONS: In this analysis, ART programs that offer mostly day-5 transfer yielded slightly higher odds of successful outcomes compared to programs that offered mostly day-3 transfer, after controlling for patient characteristics. The advantage seemed to be limited to younger women. This analysis confirms that comparisons limited to cycles that progress to transfer tend to overestimate the advantage of extended embryo culture. OBJECTIVE: Nonrandomized studies suggest that IVF cycles in which embryos are transferred at the blastocyst stage (day-5 transfers) have higher success rates than day-3 transfers, whereas randomized trials fail to show an advantage. Nonrandomized studies may overestimate the advantage of day-5 transfers by computing success rates only for procedures that progress to transfer, thus ignoring failures due to short embryo survival. This study expand previous analyses to include procedures that failed before a transfer. DESIGN: Retrospective cohort study using data from CDC's National Assisted Reproductive Technology (ART) surveillance system (NASS). MATERIALS AND METHODS: We analyzed data on IVF's of fresh non-donors from ART programs that perform most transfers either on day 5 or on day 3 in the U.S. Outcome was measured by whether or not the cycle was terminated at different stages of the procedure. Cumulative logit models with adjustment for patient and clinic factors were used to compare the outcomes of day5 and day3 programs. RESULTS: A total of 9,273 procedures from 35 clinics were identified in the study as the day-3 group; and 9,007 IVF cycles from 19 clinics as the day-5 group. Compared to the day3 group, procedures in the day5 group were less likely to be cancelled prior before retrieval(odds ratio (OR): 0.84, 95% confidence interval(95%CI): 0.77-0.91); but were more likely to be terminated before transfer, particularly for women over age 40 (OR: 1.85, 95% CI: 1.53-2.25). The odds of implantation failure were generally lower in day-5 programs than in day-3 programs (OR:0.92, 95%CI:0.87-0.98), as were the odds of pregnancy loss (OR:0.91, 95%CI:0.85-0.97), but not among women over 40 years of age (implantation failure, OR:1.48, 95%CI:1.11-1.97; pregnancy loss, OR:1.38, 95%CI:0.97-1.97). Live birth rates were slightly higher for the day-5 group (OR: 1.10, 95% CI: 1.03-1.18), but not among women over 40 years (OR:0.72, 95%CI:0.51-1.03). As expected, restricting the analysis to procedures that progressed to transfer exaggerated the advantage of day-5 transfer (data not shown). CONCLUSIONS: In this analysis, ART programs that offer mostly day-5 transfer yielded slightly higher odds of successful outcomes compared to programs that offered mostly day-3 transfer, after controlling for patient characteristics. The advantage seemed to be limited to younger women. This analysis confirms that comparisons limited to cycles that progress to transfer tend to overestimate the advantage of extended embryo culture.