Computer-automated time-lapse analysis results correlate with embryo implantation and clinical pregnancy: a blinded, multi-centre study

Abstract

Computer-automated time-lapse analysis has been shown to improve embryo selection by providing quantitative and objective information to supplement traditional morphology. In this multi-centre study, the relationship between such computer-derived outputs (High, Medium, Low scores), embryo implantation and clinical pregnancy were examined. Data were collected from six clinics, including 205 patients whose embryos were imaged by the Eeva(TM) System. The Eeva scores were blinded and not considered during embryo selection. Embryos with High and Medium scores had significantly higher implantation rates than those with Low scores (37% and 35% versus 15%; P < 0.0001; P = 0.0004). Similar trends in implantation rates were observed in different IVF centres each using their own protocols. Further analysis revealed that patients with at least one High embryo transferred had significantly higher clinical pregnancy rates than those with only Low embryos transferred (51% versus 34%; P = 0.02), although patients' clinical characteristics across groups were comparable. These data, together with previous research and clinical studies, confirm that computer-automated Eeva scores provide valuable information, which may improve the clinical outcome of IVF procedures and ultimately facilitate the trend of single embryo selection.

Keywords: Eeva Test; embryo selection; implantation; in-vitro fertilization; pregnancy; time-lapse.

Figure 1

Implantation rates for Eeva two-category High versus Low scored embryos. Error bars represent 95% upper confidence limit. The difference in implantation rates between Eeva High versus Eeva Low embryos is statistically significant: P = 0.003.

Figure 2

Implantation rates for Eeva three-category High, Medium and Low scored embryos. Error bars represent 95% upper confidence limit. Implantation rates between High versus Low and Medium versus Low were significantly different (P < 0.0001 and P = 0.0004, respectively).

Figure 3

Implantation rates for embryos with Eeva High, Medium and Low scores from three clinical sites with at least 50 embryos of known implantation data per site. Error bars represent 95% upper confidence limit. For all three sites, the difference in implantation rates between Eeva High versus Low embryos is statistically significant (chi-squared test): P = 0.02 (site A); P = 0.02 (site B); P = 0.002 (site C).