P-174 Do culture conditions alter the efficacy of embryo selection algorithms using time-lapse technology? Development of novel embryo selection model with embryos cultured in different conditions

Abstract Study question Is the efficacy of embryo selection models altered by conditions employed for culture? Summary answer The predictive capacity morphology and morphokinetics-based might be dependent on incubator culture employed. What is known already Morphological assessment remains gold-standard selection. However, since introduction time-lapse technology to incubators used culture, many algorithms based morphokinetic annotations have been developed, adding objectivity process. reproducibility these questioned other authors, being potentially influenced characteristics patients differing conditions. Nowadays, there are several with in market, some them providing novel features such as high humidity conditions, which has related differences morphokinetics development. design, size, duration Retrospective external validation an algorithm annotations, a set 555 transferred blastocysts cultured system dry (DC, n = 281) or humid (HC, 274), comparison morphological criteria. A model was developed considering our dataset, including DC HC. Embryos belong autologous oocyte-donation ICSI cycles performed clinic over 3 years. Participants/materials, setting, methods were Geri (Genea Biomedx) automatically annotated (Connect&Assess2.0). published Motato et al, 2016, ASEBIR grading assessed Generalized Estimating Equations (GEE), possible confounders. Efficacy quantified Area Under ROC Curve (AUC), its 95% confidence interval (CI) statistical significance Mann-Whitney test. aided visual tool FertAI (Merck). Main results role chance Transferred implantation data classified A-D base according optimal range 2 parameters, tEB s3, empirically-obtained embryos Embryoscope incubator. had AUC 0.591, 95%CI(0.542−0.64), resulting significantly (P<0.001), but lower than reported authors (AUC 0.602, 95%CI(0.559−0.645)). different two conditions: AUC(DC) 0.608, 95%CI(0.54−0.676), P 0.002; AUC(HC) 0.588, 95%CI(0.518−0.657), 0.103. evaluation (3 categories: (best) C (worse)) resulted statistically implantation: 0.596, 95%CI(0.547−0.646), P<0.001. Again, 0.626, 95%CI(0.559−0.693), P<0.001) HC 0.589, 95%CI(0.52−0.657), 0.013). shown associated development Hence, scoring empirically-determinated ranges three considering, first time, HC: tEB<113.874; (t5-t3)/(t5-t2) [0.521, 0.554] cc2 [10.34, 11.58], yielding score from 0 3. 0.637, 95%CI(0.59−0.684), equally efficient 0.645, 95%CI(0.579−0.712); 95%CI(0.579−0.711); P<0.001). Limitations, reasons caution This primary approach using until blastocyst stage must validated dataset. Wider implications findings supports necessity adjusting tools specific each IVF laboratory. parameters recorded Trial registration number Not applicable