Advertisement

Blastocyst collapse is not an independent predictor of reduced live birth: a time-lapse study

      Objective

      To ascertain the rate of blastocyst collapse observed by time-lapse monitoring in a retrospective cohort of unselected infertile patients undergoing single blastocyst transfer and to determine its association with live birth.

      Design

      Blastocyst collapse and morphokinetic variables were scored according to previously published criteria. The association between blastocyst collapse and live birth was evaluated by a multivariate logistic regression analysis including morphokinetic variables and other confounders.

      Setting

      Private infertility clinic.

      Patient(s)

      Patients who underwent 277 consecutive single blastocyst transfers (mean age, 38.4 ± 3.9 years; range, 28–47 years) after minimal ovarian stimulation.

      Intervention(s)

      Minimal ovarian stimulation, prolonged embryo culture in time-lapse monitoring incubator, elective vitrification with subsequent vitrified-warmed single blastocyst transfer.

      Main Outcome Measure(s)

      Live birth rate per single blastocyst transfer in different blastocyst collapse groups (no, single, multiple collapses).

      Result(s)

      No, single, or multiple blastocyst collapses occurred in 54% (150/277), 22% (61/277), and 24% (66/277) of the cohort, respectively. In the multiple collapse group on average 2.9 contractions were seen (range, 2–9 contractions). Live birth rate decreased progressively between blastocyst collapse groups (36%, 31%, 14%); significantly lower if multiple collapses occurred. In a multivariate analysis, however, blastocyst collapse was not found to be a significant predictor and was confounded by stronger predictors such as morphokinetic variables t2, texpB2, and female age.

      Conclusion(s)

      Blastocyst collapse pattern should not be evaluated alone without taking into account morphokinetic variables that are stronger predictors of reproductive outcome.

      Key Words

      Discuss: You can discuss this article with its authors and with other ASRM members at http://fertstertforum.com/bodrid-blastocyst-collapse-live-birth/
      In recent years time-lapse monitoring (TLM) technology has emerged as a promising new way to augment the classic morphological selection of embryos leading to a potential improvement in clinical outcome (
      • Meseguer M.
      • Herrero J.
      • Tejera A.
      • Hilligsoe K.M.
      • Ramsing N.B.
      • Remohi J.
      The use of morphokinetics as a predictor of embryo implantation.
      ,
      • Meseguer M.
      • Rubio I.
      • Cruz M.
      • Basile N.
      • Marcos J.
      • Requena A.
      Embryo incubation and selection in a time-lapse monitoring system improves pregnancy outcome compared with a standard incubator: a retrospective cohort study.
      ). A number of studies (
      • Cruz M.
      • Garrido N.
      • Herrero J.
      • Perez-Cano I.
      • Munoz M.
      • Meseguer M.
      Timing of cell division in human cleavage-stage embryos is linked with blastocyst formation and quality.
      ,
      • Dal Canto M.
      • Coticchio G.
      • Mignini Renzini M.
      • de Ponti E.
      • Novara P.V.
      • Brambillasca F.
      • et al.
      Cleavage kinetics analysis of human embryos predicts development to blastocyst and implantation.
      ,
      • Chamayou S.
      • Patrizio P.
      • Storaci G.
      • Tomaselli V.
      • Alecci C.
      • Ragolia C.
      • et al.
      The use of morphokinetic parameters to select all embryos with full capacity to implant.
      ,
      • Campbell A.
      • Fishel S.
      • Bowman N.
      • Duffy S.
      • Sedler M.
      • Hickman C.F.
      Modelling a risk classification of aneuploidy in human embryos using non-invasive morphokinetics.
      ,
      • Kirkegaard K.
      • Kesmodel U.S.
      • Hindkjaer J.J.
      • Ingerslev H.J.
      Time-lapse parameters as predictors of blastocyst development and pregnancy outcome in embryos from good prognosis patients: a prospective cohort study.
      ) have shown that optimal values of certain morphokinetic variables were closely correlated with blastocyst formation, implantation potential, and even aneuploidy status. Using these findings several groups have created sophisticated, hierarchical predictive models that were shown to improve clinical outcome in the setting of randomized clinical trials (
      • Rubio I.
      • Galan A.
      • Larreategui Z.
      • Ayerdi F.
      • Bellver J.
      • Herrero J.
      • et al.
      Clinical validation of embryo culture and selection by morphokinetic analysis: a randomized, controlled trial of the EmbryoScope.
      ,
      • VerMilyea M.D.
      • Tan L.
      • Anthony J.T.
      • Conaghan J.
      • Ivani K.
      • Gvakharia M.
      • et al.
      Computer-automated time-lapse analysis results correlate with embryo implantation and clinical pregnancy: a blinded, multi-centre study.
      ). These usually included the same static and interval morphokinetic variables from the early divisions of a cleavage-stage embryo (i.e., t3, t5, cc2a, s2). At present only one published model (that incorporated tSB and tB) has focused on TLM blastocysts and could successfully predict aneuploidy status and implantation potential (
      • Campbell A.
      • Fishel S.
      • Bowman N.
      • Duffy S.
      • Sedler M.
      • Hickman C.F.
      Modelling a risk classification of aneuploidy in human embryos using non-invasive morphokinetics.
      ,
      • Campbell A.
      • Fishel S.
      • Bowman N.
      • Duffy S.
      • Sedler M.
      • Thornton S.
      Retrospective analysis of outcomes after IVF using an aneuploidy risk model derived from time-lapse imaging without PGS.
      ).
      In addition to these morphokinetic variables other studies (
      • Rubio I.
      • Kuhlmann R.
      • Agerholm I.
      • Kirk J.
      • Herrero J.
      • Escriba M.J.
      • et al.
      Limited implantation success of direct-cleaved human zygotes: a time-lapse study.
      ,
      • Desai N.
      • Ploskonka S.
      • Goodman L.R.
      • Austin C.
      • Goldberg J.
      • Falcone T.
      Analysis of embryo morphokinetics, multinucleation and cleavage anomalies using continuous time-lapse monitoring in blastocyst transfer cycles.
      ,
      • Liu Y.
      • Chapple V.
      • Feenan K.
      • Roberts P.
      • Matson P.
      A time-lapse deselection model for human day 3 in vitro fertilization embryos: the combination of qualitative and quantitative measures of embryo growth.
      ,
      • Liu Y.
      • Chapple V.
      • Feenan K.
      • Roberts P.
      • Matson P.
      Clinical significance of intercellular contact at the four-cell stage of human embryos, and the use of abnormal cleavage patterns to identify embryos with low implantation potential: a time-lapse study.
      ,
      • Liu Y.
      • Chapple V.
      • Roberts P.
      • Matson P.
      Prevalence, consequence, and significance of reverse cleavage by human embryos viewed with the use of the Embryoscope time-lapse video system.
      ,
      • Yang S.T.
      • Shi J.X.
      • Gong F.
      • Zhang S.P.
      • Lu C.F.
      • Tan K.
      • et al.
      Cleavage pattern predicts developmental potential of day 3 human embryos produced by IVF.
      ) focused on qualitative markers of cleavage-stage embryos that were associated with a reduced implantation potential and could be used as deselection criteria such as direct cleavage, multinucleation, and abnormal cleavage patterns. In a recently published study (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ) another negative prognostic marker emerged, but in a blastocyst-stage embryo. In this retrospective review blastocyst collapse, precisely observed by TLM, was associated with a reduced rate of implantation (35% vs. 48.5%). Marcos et al. (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ) strongly recommended against transferring such blastocysts (if any alternative is available) and encouraged to include this marker as an additional criterion in existing embryo selection models. Although this previous study was the first to investigate the influence of blastocyst collapse on reproductive outcome, its predictive value of a negative test (PVN) still has to be confirmed on embryos that originated from different patient populations and clinical settings. Therefore the aim of the present retrospective analysis was to determine the rate of blastocyst collapse observed by TLM in a cohort of blastocysts from unselected infertile patients undergoing single blastocyst transfer after mild ovarian stimulation and to determine its exact influence on live birth outcome.

      Materials and methods

       Study Patients and Follow-up

      All consecutive infertile patients whose embryos were submitted to prolonged embryo culture and reached the expanded blastocyst stage (n = 501) in a TLM incubator between October 2012 (the acquisition of a single EmbryoScope incubator) and December 2014 at our center (Kobe Motomachi Yume Clinic, Kobe, Japan) and subsequently underwent vitrified-warmed ETs until August 2015 (n = 291) were included in this retrospective analysis. Fourteen cases involving zona pellucida (ZP)-free embryos were excluded because blastocyst collapse could not be precisely evaluated in these embryos (
      • Bodri D.
      • Kato R.
      • Kondo M.
      • Hosomi N.
      • Katsumata Y.
      • Kawachiya S.
      • et al.
      Time-lapse monitoring of zona pellucida-free embryos obtained through in vitro fertilization: a retrospective case series.
      ). The outcome of ETs was followed-up until a live birth occurred (including 4 still ongoing pregnancies beyond 20 gestational weeks at the time of writing) (Supplemental Fig. 1, available online). Institutional Review Board approval was not required for the present study because in our center all patients undergoing IVF treatment gave informed consent to the anonymous use of their data for retrospective reviews.

       Natural Cycle IVF and Minimal Ovarian Stimulation Protocols

      Clomiphene citrate (CC) or letrozole-based minimal stimulation was used in most cycles whereas unstimulated natural cycle IVF or other mild stimulation protocols represented only a smaller proportion of cases. Details of the CC-based minimal stimulation and natural cycle IVF protocols were described previously (
      • Bodri D.
      • Kawachiya S.
      • Kondo M.
      • Kato R.
      • Matsumoto T.
      Oocyte retrieval timing based on spontaneous luteinizing hormone surge during natural cycle in vitro fertilization treatment.
      ,
      • Kato K.
      • Takehara Y.
      • Segawa T.
      • Kawachiya S.
      • Okuno T.
      • Kobayashi T.
      • et al.
      Minimal ovarian stimulation combined with elective single embryo transfer policy: age-specific results of a large, single-centre, Japanese cohort.
      ). Patients were not selected according to their age and this treatment option was offered over a wide age range.

       Oocyte Retrieval and Fertilization

      Transvaginal ultrasound-guided oocyte retrieval was performed without anesthesia using a very thin 21–22 G needle (Kitazato Medical Co., Ltd.). Mature (metaphase II) oocytes were inseminated by conventional IVF or intracytoplasmic sperm injection (ICSI). Subsequently ICSI injected oocytes were placed in pre-equilibrated slides (EmbryoSlide, Vitrolife), whereas IVF inseminated oocytes were first cultured in a conventional, trigas, water jacket incubator (Astec), and transferred to a TLM incubator the next morning if fertilization was confirmed.

       Prolonged Embryo Culture in a TLM Incubator and Elective Vitrification

      In our center elective blastocyst culture, vitrification, and subsequent vitrified-warmed single blastocyst transfer are routinely practiced in most patients. Prolonged embryo culture was performed in a time-lapse incubator (EmbryoScope, Vitrolife) according to previously described methodology (
      • Meseguer M.
      • Herrero J.
      • Tejera A.
      • Hilligsoe K.M.
      • Ramsing N.B.
      • Remohi J.
      The use of morphokinetics as a predictor of embryo implantation.
      ). Briefly, normally fertilized 2-pronuclei (PN) zygotes were incubated at 37°C in a 5% O2 atmosphere, cultured individually until days 2–3 in a Quinn's Advantage Cleavage Medium (SAGE), and subsequently until days 5–7 in a Quinn's Advantage Blastocyst Medium (SAGE). Cleavage-stage selection was performed by classic morphology criteria according to published consensus criteria (
      Alpha Scientists in Reproductive Medicine, Embryology ESIGo
      The Istanbul consensus workshop on embryo assessment: proceedings of an expert meeting.
      ). According to the protocol of our group embryos that reached the blastocyst stage by day 5 or 6 were eligible for elective vitrification (Cryotop, Kitazato Medical Co., Ltd.) as soon as they expanded to a size ≥160 μm. Expanding blastocyst were observed two to four times a day (
      • Kuwayama M.
      Highly efficient vitrification for cryopreservation of human oocytes and embryos: the Cryotop method.
      ). Expanded blastocysts were graded into high-quality (inner cell mass [ICM] and trophoectoderm score: AA, AB, or BA) or low-quality (ICM and trophoectoderm score: BB, BC, CB, or CC) categories (
      • Kirkegaard K.
      • Kesmodel U.S.
      • Hindkjaer J.J.
      • Ingerslev H.J.
      Time-lapse parameters as predictors of blastocyst development and pregnancy outcome in embryos from good prognosis patients: a prospective cohort study.
      ,
      • Gardner D.K.
      • Surrey E.
      • Minjarez D.
      • Leitz A.
      • Stevens J.
      • Schoolcraft W.B.
      Single blastocyst transfer: a prospective randomized trial.
      ). Electively vitrified blastocysts were transferred in the next 1–3 months after the oocyte retrieval in a spontaneous natural or hormonal replacement (HR) cycle (
      • Zhang J.
      • Chang L.
      • Sone Y.
      • Silber S.
      Minimal ovarian stimulation (mini-IVF) for IVF utilizing vitrification and cryopreserved embryo transfer.
      ).

       Time-lapse Annotations

      Seven plane focal images were taken and recorded every 20 minutes shortly after insemination for approximately 160 hours. The duration of the TLM observation from the time point when the blastocyst reached full expansion and started pushing on the ZP (tfullB) until it reached a sufficiently expanded size for elective vitrification, and was removed from the TLM incubator, was calculated for each blastocyst. Early (PNf, t2–t9) and late (start of blastulation and full blastocyst) morphokinetic time points were scored in accordance with recently published consensus criteria (
      • Ciray H.N.
      • Campbell A.
      • Agerholm I.E.
      • Aguilar J.
      • Chamayou S.
      • Esbert M.
      • et al.
      Proposed guidelines on the nomenclature and annotation of dynamic human embryo monitoring by a time-lapse user group.
      ). To determine the degree of blastocyst expansion more objectively, two additional TLM variables (texpB1, texpB2) were also introduced (the time point when the horizontal diameter of the expanded blastocyst reached 130 and 160 μm) (
      • Bodri D.
      • Sugimoto T.
      • Serna J.Y.
      • Kondo M.
      • Kato R.
      • Kawachiya S.
      • et al.
      Influence of different oocyte insemination techniques on early and late morphokinetic parameters: retrospective analysis of 500 time-lapse monitored blastocysts.
      ). Annotations were performed retrospectively using EmbryoViewer by two experienced embryologists (J.Y.S. and T.S.) and rechecked by a third person (D.B.). To control for different insemination techniques used (34% and 66% of the cohort was fertilized with conventional IVF or ICSI, respectively) static morphokinetic variables were standardized to the common postfertilization time point of pronuclear fading (
      • Bodri D.
      • Sugimoto T.
      • Serna J.Y.
      • Kondo M.
      • Kato R.
      • Kawachiya S.
      • et al.
      Influence of different oocyte insemination techniques on early and late morphokinetic parameters: retrospective analysis of 500 time-lapse monitored blastocysts.
      ,
      • Cruz M.
      • Garrido N.
      • Gadea B.
      • Munoz M.
      • Perez-Cano I.
      • Meseguer M.
      Oocyte insemination techniques are related to alterations of embryo developmental timing in an oocyte donation model.
      ,
      • Liu Y.
      • Chapple V.
      • Feenan K.
      • Roberts P.
      • Matson P.
      Time-lapse videography of human embryos: using pronuclear fading rather than insemination in IVF and ICSI cycles removes inconsistencies in time to reach early cleavage milestones.
      ).

       Definition of Blastocyst Collapse(s)

      Blastocyst collapse was defined according to the methodology described in Marcos et al. (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ). Briefly, each video sequence produced by EmbryoViewer was carefully analyzed—from full blastocyst expansion (the frame before the ZP starts to thin) until approximately 160 hours after insemination—to detect instances when the surface of the trophoectoderm cells separated >50% from the inner side of the ZP (collapse or strong contraction). A separation of <50% (weak contraction) was not considered as collapse. When necessary drawing tools were used to ascertain the degree of separation (Fig. 1 and Videos 1 and 2, available online, are showing a blastocyst with a single and multiple collapses). The number of observed blastocyst collapses were counted and the entire cohort was divided into three groups (no collapse, a single collapse, and multiple collapses).
      Figure thumbnail gr1
      Figure 1Multiple blastocyst collapses observed with Embryo Viewer (in a 39-year-old patient with a negative pregnancy test). Upper row: first collapse at 107.7 hours and re-expansion by 109.4 hours; middle row: second collapse at 109.7 hours and re-expansion by 111.7 hours; lower row: third collapse at 116.8 hours and re-expansion by 118.1 hours.

       Outcome Measures and Statistical Analysis

      The main outcome measure was live birth rate per single ET in blastocyst collapse groups. Baseline cycle characteristics were also compared between groups. Metric variables were compared by the one-way analysis of variance (ANOVA) test. Nominal variables were analyzed by the χ2 test. P<.01 was considered statistically significant. The association between live birth and blastocyst collapse was examined by a multivariate logistic regression analysis. In addition to the blastocyst collapse variable (no, single, multiple collapses) the model included eight categorical morphokinetic variables (categorized into inside or outside of the optimal live birth ranges) that were shown to be associated with live birth outcome in our previous work. In addition eight patient-related and cycle-related confounding factors, such as female age (28–34 years [reference], 35–40 years, 41–47 years), body mass index (BMI; in kg/m2), parity (yes/no), history of previous IVF treatment at other center (yes/no), current cycle rank (cycles 1–2 [reference], 3–4, ≥5), stimulation type (unstimulated [reference]/CC/letrozole/other), blastocyst quality (low/high), and duration of TLM observations (hours), were included. These were chosen because in our previously published dataset they were significantly associated with live birth (
      • Bodri D.
      • Kawachiya S.
      • De Brucker M.
      • Tournaye H.
      • Kondo M.
      • Kato R.
      • et al.
      Cumulative success rates following mild IVF in unselected infertile patients: a 3-year, single-centre cohort study.
      ). Statistical calculations were performed with the “R” software package.

      Results

       Blastocyst Collapse Groups and Their Reproductive Outcome

      During the study a total of 277 single blastocyst transfers were performed from all consecutive infertile patients (female age, 38.4 ± 3.9 years; range, 28–47 years) whose embryos underwent prolonged embryo culture in a time-lapse incubator. The overall live birth rate per (single) ET in the examined cohort was 30% (82/277). One or more blastocyst collapse occurred in 46% (127/277) of the examined embryos. In 22% (61/277) of the cases only a single collapse was observed, whereas in 24% (66/277) multiple collapses were seen. In the multiple collapse group an average number of 2.9 collapses were registered (range, 2–9 collapses).
      Baseline characteristics of the ET cycles according to blastocyst collapse groups are summarized in Table 1. There were no significant differences among baseline variables such as female age, BMI, basal FSH, type of infertility, parity, the history of previous IVF treatment at other centers, current cycle rank, stimulation type, the number of retrieved eggs, and the male partner age. Female age did not influence the occurrence of blastocyst collapse(s) (Supplemental Table 1). However, the proportion of high-quality blastocysts decreased gradually in each blastocyst collapse group (38%, 21%, and 8%, respectively; P<.0001). The duration of the time lapse observations was increasingly longer among groups (no collapse, 13.1 ± 5.7 hours; single collapse, 16.7 ± 6.1 hours; multiple collapses, 24.6 ± 8.3 hours).
      Table 1Baseline cycle characteristics in groups with no, 1, or >1 blastocyst collapse(s).
      ET cyclesNo collapse (n = 150)1 collapse (n = 61)>1 collapse (n = 66)P Value
      Female age (y)38.4 ± 438.6 ± 3.939.1 ± 3.6.16
      One-way analysis of variance test.
      BMI, (kg/m2)20.4 ± 2.320.3 ± 2.520.8 ± 2.4.49
      One-way analysis of variance test.
      Basal FSH (IU/mL)9 ± 4.99.5 ± 5.49.3 ± 5.2.80
      One-way analysis of variance test.
      Primary infertility, n (%)102 (68)42 (69)39 (59).39
      χ2 test.
      Nulliparous, n (%)136 (91)56 (92)60 (91).97
      χ2 test.
      Duration of marriage (y)6.7 ± 3.96.8 ± 3.57.1 ± 3.9.77
      One-way analysis of variance test.
      No IVF treatment (at other centers), n (%)62 (41)17 (28)27 (41).17
      χ2 test.
      Current cycle rank.49
      χ2 test.
       1–2 cycles, n (%)47 (31)17 (28)16 (24)
       3–4 cycles, n (%)57 (38)20 (33)22 (33)
       5 or higher, n (%)46 (31)24 (39)28 (42)
      Stimulation type.91
      χ2 test.
       Natural cycle, n (%)15 (10)6 (4)4 (3)
       Clomiphene citrate, n (%)110 (73)44 (29)47 (31)
       Letrozole, n (%)22 (15)10 (7)14 (9)
       Other, n (%)3 (2)1 (0.7)1 (0.7)
      Retrieved mature (MII) eggs, n5.8 ± 3.45.7 ± 3.35.3 ± 3.2.68
      One-way analysis of variance test.
      High-quality blastocysts, n (%)57 (38)13 (21)5 (8)<.0001
      χ2 test.
      Male partner age (y)39.7 ± 5.140.9 ± 5.640.8 ± 5.7.23
      One-way analysis of variance test.
      Note: MII = metaphase II.
      a One-way analysis of variance test.
      b χ2 test.
      Live birth rates per single blastocyst transfer decreased progressively in blastocyst collapse groups (36%, 31%, and 14%, respectively; P=.004), but statistically significant differences were only observed between the multiple collapse and the other two groups (P=.0009 and P=.02).

       Comparison of Morphokinetic Variables among Blastocyst Collapse Groups

      There were no significant differences observed between blastocyst collapse groups in cleavage stage, static (t2–t9), and interval parameters (cc2a, cc2b, s2, s3). In contrast when comparing blastocyst-stage TLM parameters (from tSB to texpB2), there was a statistically significant, gradually increasing delay among blastocyst collapse groups (Table 2 and Supplemental Fig. 2, available online).
      Table 2TLM timings in groups with no, 1, or >1 blastocyst collapse(s).
      TLM variables (mean ± SD)No collapse (n = 150)1 collapse (n = 61)>1 collapses (n = 66)P value
      One-way analysis of variance test.
      t22.7 ± 0.72.7 ± 0.92.8 ± 0.9.96
      t314.0 ± 1.714.0 ± 1.814.1 ± 1.9.86
      t414.7 ± 2.214.8 ± 2.115.1 ± 2.4.5
      t528.2 ± 4.927.5 ± 5.528.2 ± 4.8.58
      t629.8 ± 4.229.2 ± 5.030.1 ± 4.1.56
      t731.4 ± 4.931.6 ± 5.331.9 ± 4.8.82
      t833.7 ± 6.535.1 ± 7.835.8 ± 7.7.15
      t949.9 ± 8.751.6 ± 7.849.8 ± 8.7.42
      cc2a (t3-t2)11.2 ± 1.711.2 ± 1.711.4 ± 2.0.84
      cc2b (t4-t2)12.1 ± 2.112.2 ± 1.712.4 ± 2.6.53
      s2 (t4-t3)0.9 ± 1.61.0 ± 1.51.1 ± 2.3.71
      s3 (t8-t5)5.6 ± 5.37.7 ± 8.67.8 ± 7.5.054
      tSB77.7 ± 8.578.6 ± 11.581.4 ± 8.7.027
      tfB88.2 ± 10.589.7 ± 13.092.5 ± 9.6.030
      texpB188.4 ± 9.790.5 ± 12.094.4 ± 9.2.0004
      texpB296.6 ± 10.899.0 ± 12.1108.9 ± 10.2<.0001
      Note: cc2a/cc2b/s2/s3 = duration of different embryo cell cycle events; PNf = pronuclear fading; t2–t9 = time to 2 to 9 discrete cells; tSB = time to initiation of blastulation; texpB1 = time to reach expanded blastocyst size of 130 μm; texpB2 = time to reach expanded blastocyst size of 160 μm; tfB = time to full blastocyst. All time-lapse monitoring (TLM) variables were standardized to PNf.
      a One-way analysis of variance test.

       Association between Live Birth and Blastocyst Collapse in a Univariate/Multivariate Analysis

      Unadjusted odds ratios (ORs) showed that only “multiple blastocyst collapse” was associated with a decreased probability of live birth (OR, 0.28; 95% confidence interval [CI] 0.12–0.59; P=.001). In contrast “single blastocyst collapse” did not decrease significantly the chance of a live birth (OR, 0.80; 95% CI 0.42–1.51; P=.502). Among patient-related confounders female age (≥41 years), higher blastocyst quality, and longer duration of TLM observations were inversely associated with live birth (Table 3).
      Table 3Results of univariate and multivariate analysis on the association between live birth and categorical TLM variables and other confounders.
      TLM variablesUnadjusted ORsP valueAdjusted ORsAdjusted P value
      Adjusted for eight TLM variables and nine confounders (blastocyst collapse, female age, body mass index (BMI), parity, history of previous IVF, cycle rank, stimulation type, blastocyst quality, duration of time-lapse observation).
      Blastocyst collapse
       No collapse (reference)
       1 collapse0.80 (0.42–1.51).5021.87 (0.80–4.40).148
       >1 collapse0.28 (0.12–0.59).0010.86 (0.26–2.74).803
       texpB24.09 (2.31–7.52)<.00014.76 (1.61–15.51).006
       texpB12.13 (1.26–3.67).0051.06 (0.32–3.43).917
       tfullB2.20 (1.27–3.9).0050.89 (0.22–3.42).867
       t41.84 (1.08–3.17).0261.24 (0.56–2.78).593
       t31.72 (1.02–2.94).0442.23 (1.05–4.86).038
       t22.29 (1.35–3.9).0022.50 (1.29–4.95).007
       cc2a1.94 (1.14–3.31).0141.54 (0.73–3.27).254
       cc2b2.06 (1.20–3.54).0091.79 (0.75–4.36).190
      Female age
       28–34 y (reference)
       35–40 y0.61 (0.31–1.21).1520.70 (0.29–1.66).413
       41–47 y0.21 (0.09–0.48).00020.21 (0.07–0.61).005
      BMI0.97 (0.86–1.08).5931.02 (0.88–1.19).770
      Previous live birth0.92 (0.34–2.20).8541.04 (0.30–3.34).950
      Previous IVF treatment0.84 (0.49–1.43).5080.86 (0.42–1.79).691
      Stimulation type
       Natural cycle (reference)
       Clomiphene citrate0.55 (0.24–1.32).1711.04 (0.32–3.61).946
       Letrozole0.35 (0.12–1.01).0531.19 (0.27–5.35).813
      Current cycle rank
       1–2 cycles (reference)
       3–4 cycles1.20 (0.64–2.27).5631.64 (0.73–3.81).237
       5 or higher0.64 (0.32–1.24).1860.79 (0.32–1.95).611
      Low-quality blastocyst0.48 (0.27–0.84).0090.73 (0.34–1.58).428
      Duration of TLM observations0.94 (0.90–0.97).0011.00 (0.93–1.07).966
      Note: cc2a/cc2b/s2/s3 = duration of different embryo cell cycle events; OR = odds ratio; PNf = pronuclear fading; t2–t9 = time to 2 to 9 discrete cells; tfB = time to full blastocyst; texpB1 = time to reach expanded blastocyst size of 130 μm; texpB2 = time to reach expanded blastocyst size of 160 μm; tSB = time to initiation of blastulation. All time-lapse monitoring (TLM) variables were standardized to PNf.
      a Adjusted for eight TLM variables and nine confounders (blastocyst collapse, female age, body mass index (BMI), parity, history of previous IVF, cycle rank, stimulation type, blastocyst quality, duration of time-lapse observation).
      After adjusting for eight TLM variables and eight patient-related and cycle-related confounders the previously shown association for the multiple blastocyst collapse group disappeared (OR, 0.86; 95% CI 0.26–2.74; P=.803). In the resulting model among TLM variables only texpB2 (adjusted OR, 4.76; 95% CI 1.61–15.51; P=.006), t3 (albeit at a less significant level, P=.038), and t2 (adjusted OR, 2.50; 95% CI 1.29–4.95; P=.007) remained significantly associated with live birth. Among patient-related confounders the only variable, that still had a significant impact was female age >41 years (adjusted OR, 0.21; 95% CI 0.07–0.61; P=.005), whereas the previously observed association for low-quality blastocysts and duration of TLM observations disappeared (Table 3).

      Discussion

      Our study involving a cohort of unselected advanced aged infertile patients undergoing minimal ovarian stimulation has found that blastocyst collapses (including multiple ones), observed by TLM, occurred frequently (in almost 50% of entire cohort). Although live birth rates decreased when blastocyst collapse(s) were observed (especially if they were multiple) blastocyst contraction was not an independent predictor of reduced live birth and was confounded by cleavage stage and blastocyst stage morphokinetic variables and female age.
      This study to our knowledge is the second one that has investigated the influence of time-lapse observed blastocyst collapse(s) on reproductive outcome after IVF treatment. A recently published article (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ) by a Spanish group that pioneered TLM research presented the results of a two-center, retrospective review involving 715 transferred blastocysts during a 1-year period. They found that 19.4% of the analyzed blastocysts presented collapse including 1.4% with multiple (2 or 3) collapses. Baseline characteristics were not significantly different between the two groups with the exception of a higher proportion of donor cycles in the group of collapsed blastocysts. Interestingly, embryo/blastocyst quality assessed by classic morphology criteria was not significantly different, but timing for several morphokinetic parameters (i.e., t2, t3, t4, tB) were shorter for collapsed embryos (implying faster development). Reproductive outcome was evaluated in a smaller subset of 502 embryos due to the necessary reliance on known-implantation or known-implantation data. The investigators found that implantation rates (albeit only in a univariate analysis) were significantly lower (35% vs. 48.5%) in collapsed blastocysts. They concluded that these blastocysts should not be replaced (except if no better alternatives are available) and that the collapse pattern should be considered as an additional criterion that could improve embryo selection.
      Our study partly confirms, but also partly contradicts, the findings observed in the previously mentioned study. First, the overall rate of blastocyst collapse(s) was much higher in our cohort (46% vs. 19.4%) and also more multiple (<9) collapses were seen (22% vs. 1.4%). Second, findings were different in relation to blastocyst quality (decreasing quality for collapsed blastocysts in ours vs. similar quality in the Spanish study) and timing of morphokinetic parameters. The collapsed group blastocyst-stage TLM parameters were on average longer in our study but shorter in the Marcos et al. study (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ). All of these discrepancies could be explained by a radically different patient profile (beside ethnicity) between the two studies. Our study involved advanced aged, nondonor, infertile patients exclusively, whereas more than half of the Spanish cohort (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ) were young oocyte donors. In addition there were some differences in culture conditions (e.g., different brands of sequential media, low vs. atmospheric O2 concentration).
      However, both studies were concordant in that, at least in a univariate analysis, they have found decreased implantation rates in the group of collapsed blastocysts. Regarding the group of multiple collapses this finding was statistically significant in our study but could not be determined exactly in the Spanish study (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ) due to the small number of observed multiple collapse cases (n = 10). Even more important, however, in our study decreased implantation rate in the collapse groups was not present anymore after adjusting for confounding variables in a robust multivariate analysis. This suggests that the influence of blastocyst collapse is strongly confounded by a handful of important morphokinetic parameters that influence reproductive outcome (in our study, t2, t3, texpB2). Although a less robust multivariate analysis was also part of the Spanish study (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ), the six chosen covariates were patient-related, or cycle-related only and did not include the morphokinetic parameters (t2, t3, t4, tB). Therefore it is impossible to determine whether the original Spanish findings would have been concordant with ours had a more extensive multivariate analysis been performed. This strongly implies that although the blastocyst collapse pattern might become a relatively important predictive variable it should not be evaluated without taking into account other (cleavage stage or blastocyst stage) morphokinetic variables that already have been shown to be strong predictors of implantation and have been included in various hierarchical predictive models (
      • VerMilyea M.D.
      • Tan L.
      • Anthony J.T.
      • Conaghan J.
      • Ivani K.
      • Gvakharia M.
      • et al.
      Computer-automated time-lapse analysis results correlate with embryo implantation and clinical pregnancy: a blinded, multi-centre study.
      ,
      • Campbell A.
      • Fishel S.
      • Bowman N.
      • Duffy S.
      • Sedler M.
      • Thornton S.
      Retrospective analysis of outcomes after IVF using an aneuploidy risk model derived from time-lapse imaging without PGS.
      ,
      • Basile N.
      • Vime P.
      • Florensa M.
      • Aparicio Ruiz B.
      • Garcia Velasco J.A.
      • Remohi J.
      • et al.
      The use of morphokinetics as a predictor of implantation: a multicentric study to define and validate an algorithm for embryo selection.
      ).
      At present many animal studies (
      • Seshagiri P.B.
      • Sen Roy S.
      • Sireesha G.
      • Rao R.P.
      Cellular and molecular regulation of mammalian blastocyst hatching.
      ,
      • Gonzales D.S.
      • Bavister B.D.
      Zona pellucida escape by hamster blastocysts in vitro is delayed and morphologically different compared with zona escape in vivo.
      ,
      • Erbach G.T.
      • Biggers J.D.
      • Manning P.C.
      • Nowak R.A.
      Localization of parathyroid hormone-related protein in the preimplantation mouse embryo is associated with events of blastocyst hatching.
      ) investigated the phenomenon of blastocyst contraction, its molecular mechanisms, and the relation to embryo hatching. Niimura (
      • Niimura S.
      Time-lapse videomicrographic analyses of contractions in mouse blastocysts.
      ) suggested that in mouse embryos weak contractions play a positive role in hatching, whereas strong ones are inhibitory. It is unclear, however, how the previously mentioned findings could be extrapolated to in-vitro cultured human blastocysts. Marcos et al. (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ) hypothesized that blastocyst collapse may be linked to the implantation process, either through embryo quality (caused by decreased oocyte quality or suboptimal culture conditions) or by directly affecting the blastocyst's implantation ability (through abnormal hatching or other mechanisms). Our findings suggest that collapsing blastocysts also display suboptimal timing of morphokinetic variables, implying that the underlying common cause is the defective embryo quality, which ultimately manifests itself in a reduced implantation potential. For example, a link between suboptimal morphokinetical timing and aneuploidy rates (and decreased implantation) was already established in previous studies (
      • Campbell A.
      • Fishel S.
      • Bowman N.
      • Duffy S.
      • Sedler M.
      • Hickman C.F.
      Modelling a risk classification of aneuploidy in human embryos using non-invasive morphokinetics.
      ,
      • Basile N.
      • Nogales Mdel C.
      • Bronet F.
      • Florensa M.
      • Riqueiros M.
      • Rodrigo L.
      • et al.
      Increasing the probability of selecting chromosomally normal embryos by time-lapse morphokinetics analysis.
      ).
      The main limitation of our study was the relatively low number of single blastocyst transfers (277). This sample size, however, which was achieved in a single center during a 2-year study period, is comparable with those published in other relevant TLM articles and is certainly sufficient to develop an initial, center-specific predictive model that later could be improved on. Second, the duration of the TLM observations was increasingly longer for the collapse groups that, in our opinion, is not an inherent bias, but that (especially repeated) collapse/re-expansion cycles increased the time needed to reach the necessary criteria for elective vitrification. Third, elective vitrification itself, which is a prominent feature of our clinic's treatment protocol, was not practiced in the Spanish study (
      • Marcos J.
      • Perez-Albala S.
      • Mifsud A.
      • Molla M.
      • Landeras J.
      • Meseguer M.
      Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
      ), where only fresh ET cycles were included. As this has equally affected all blastocyst transfer cycles in our cohort it is unlikely that it could have introduced any bias affecting live birth rates. As a considerable strength, by our center's routine policy, our dataset included only single ETs, thus we did not have to rely on known-implantation data embryos, only avoiding any potential bias that might arise from analyzing embryos with a “double or nothing” implantation potential.
      In conclusion in a group of unselected infertile patients undergoing minimal ovarian stimulation and vitrified-thawed single blastocyst transfer, single and multiple blastocyst collapses occurred frequently, but were not independently associated with a decreased chance of a live birth.

      Acknowledgments

      The authors thank Michael Want for the linguistic revision of the manuscript.

      Appendix

      Figure thumbnail fx1
      Supplemental Figure 1Flow chart of study events. ICSI = intracytoplasmic sperm injection; PN = pronuclei; PNf = pronuclear fading; SBT = single blastocyst transfer; ZP = zona pellucida.
      Figure thumbnail fx2
      Supplemental Figure 2Time-lapse monitoring variables according to blastocyst collapse groups (0/1/multiple collapse(s). (A) tSB P=.019 between “0” and “Mult”; (B) tfullB P=.023 between “0” and “Mult”; (C) texpB1 P=.0002 between “0” and “Mult”; (D) texpB2 P<.0001 between “0” and “Mult” and between “1” and “Mult.” PNf = pronuclear fading; texpB1 = time to reach expanded blastocyst size of 130 μm; texpB2 = time to reach expanded blastocyst size of 160 μm; tSB = time to initiation of blastulation; tfB = time to full blastocyst. All time-lapse monitoring variables standardized to PNf.
      Supplemental Table 1Relationship between blastocyst collapse patterns and female age groups.
      Female age versus blastocyst collapse patterns
      χ2 test, P=.59.
      28–34 y (n = 44)35–40 y (n = 146)41–47 y (n = 87)
      No collapse n, (%)28 (64)79 (54)43 (49)
      1 collapse n, (%)7 (16)34 (23)20 (23)
      >1 collapse n, (%)9 (20)33 (23)24 (28)
      a χ2 test, P=.59.

      Supplementary data

      References

        • Meseguer M.
        • Herrero J.
        • Tejera A.
        • Hilligsoe K.M.
        • Ramsing N.B.
        • Remohi J.
        The use of morphokinetics as a predictor of embryo implantation.
        Hum Reprod. 2011; 26: 2658-2671
        • Meseguer M.
        • Rubio I.
        • Cruz M.
        • Basile N.
        • Marcos J.
        • Requena A.
        Embryo incubation and selection in a time-lapse monitoring system improves pregnancy outcome compared with a standard incubator: a retrospective cohort study.
        Fertil Steril. 2012; 98: 1481-1489.e10
        • Cruz M.
        • Garrido N.
        • Herrero J.
        • Perez-Cano I.
        • Munoz M.
        • Meseguer M.
        Timing of cell division in human cleavage-stage embryos is linked with blastocyst formation and quality.
        Reprod Biomed Online. 2012; 25: 371-381
        • Dal Canto M.
        • Coticchio G.
        • Mignini Renzini M.
        • de Ponti E.
        • Novara P.V.
        • Brambillasca F.
        • et al.
        Cleavage kinetics analysis of human embryos predicts development to blastocyst and implantation.
        Reprod Biomed Online. 2012; 25: 474-480
        • Chamayou S.
        • Patrizio P.
        • Storaci G.
        • Tomaselli V.
        • Alecci C.
        • Ragolia C.
        • et al.
        The use of morphokinetic parameters to select all embryos with full capacity to implant.
        J Assist Reprod Genet. 2013; 30: 703-710
        • Campbell A.
        • Fishel S.
        • Bowman N.
        • Duffy S.
        • Sedler M.
        • Hickman C.F.
        Modelling a risk classification of aneuploidy in human embryos using non-invasive morphokinetics.
        Reprod Biomed Online. 2013; 26: 477-485
        • Kirkegaard K.
        • Kesmodel U.S.
        • Hindkjaer J.J.
        • Ingerslev H.J.
        Time-lapse parameters as predictors of blastocyst development and pregnancy outcome in embryos from good prognosis patients: a prospective cohort study.
        Hum Reprod. 2013; 28: 2643-2651
        • Rubio I.
        • Galan A.
        • Larreategui Z.
        • Ayerdi F.
        • Bellver J.
        • Herrero J.
        • et al.
        Clinical validation of embryo culture and selection by morphokinetic analysis: a randomized, controlled trial of the EmbryoScope.
        Fertil Steril. 2014; 102: 1287-1294.e5
        • VerMilyea M.D.
        • Tan L.
        • Anthony J.T.
        • Conaghan J.
        • Ivani K.
        • Gvakharia M.
        • et al.
        Computer-automated time-lapse analysis results correlate with embryo implantation and clinical pregnancy: a blinded, multi-centre study.
        Reprod Biomed Online. 2014; 29: 729-736
        • Campbell A.
        • Fishel S.
        • Bowman N.
        • Duffy S.
        • Sedler M.
        • Thornton S.
        Retrospective analysis of outcomes after IVF using an aneuploidy risk model derived from time-lapse imaging without PGS.
        Reprod Biomed Online. 2013; 27: 140-146
        • Rubio I.
        • Kuhlmann R.
        • Agerholm I.
        • Kirk J.
        • Herrero J.
        • Escriba M.J.
        • et al.
        Limited implantation success of direct-cleaved human zygotes: a time-lapse study.
        Fertil Steril. 2012; 98: 1458-1463
        • Desai N.
        • Ploskonka S.
        • Goodman L.R.
        • Austin C.
        • Goldberg J.
        • Falcone T.
        Analysis of embryo morphokinetics, multinucleation and cleavage anomalies using continuous time-lapse monitoring in blastocyst transfer cycles.
        Reprod Biol Endocrinol. 2014; 12: 54
        • Liu Y.
        • Chapple V.
        • Feenan K.
        • Roberts P.
        • Matson P.
        A time-lapse deselection model for human day 3 in vitro fertilization embryos: the combination of qualitative and quantitative measures of embryo growth.
        Fertil Steril. 2016; 105: 656-662
        • Liu Y.
        • Chapple V.
        • Feenan K.
        • Roberts P.
        • Matson P.
        Clinical significance of intercellular contact at the four-cell stage of human embryos, and the use of abnormal cleavage patterns to identify embryos with low implantation potential: a time-lapse study.
        Fertil Steril. 2015; 103: 1485-1491.e1
        • Liu Y.
        • Chapple V.
        • Roberts P.
        • Matson P.
        Prevalence, consequence, and significance of reverse cleavage by human embryos viewed with the use of the Embryoscope time-lapse video system.
        Fertil Steril. 2014; 102: 1295-1300.e2
        • Yang S.T.
        • Shi J.X.
        • Gong F.
        • Zhang S.P.
        • Lu C.F.
        • Tan K.
        • et al.
        Cleavage pattern predicts developmental potential of day 3 human embryos produced by IVF.
        Reprod Biomed Online. 2015; 30: 625-634
        • Marcos J.
        • Perez-Albala S.
        • Mifsud A.
        • Molla M.
        • Landeras J.
        • Meseguer M.
        Collapse of blastocysts is strongly related to lower implantation success: a time-lapse study.
        Hum Reprod. 2015; 30: 2501-2508
        • Bodri D.
        • Kato R.
        • Kondo M.
        • Hosomi N.
        • Katsumata Y.
        • Kawachiya S.
        • et al.
        Time-lapse monitoring of zona pellucida-free embryos obtained through in vitro fertilization: a retrospective case series.
        Fertil Steril. 2015; 103: e35
        • Bodri D.
        • Kawachiya S.
        • Kondo M.
        • Kato R.
        • Matsumoto T.
        Oocyte retrieval timing based on spontaneous luteinizing hormone surge during natural cycle in vitro fertilization treatment.
        Fertil Steril. 2014; 101: 1001-1007.e2
        • Kato K.
        • Takehara Y.
        • Segawa T.
        • Kawachiya S.
        • Okuno T.
        • Kobayashi T.
        • et al.
        Minimal ovarian stimulation combined with elective single embryo transfer policy: age-specific results of a large, single-centre, Japanese cohort.
        Reprod Biol Endocrinol. 2012; 10: 35
        • Alpha Scientists in Reproductive Medicine, Embryology ESIGo
        The Istanbul consensus workshop on embryo assessment: proceedings of an expert meeting.
        Hum Reprod. 2011; 26: 1270-1283
        • Kuwayama M.
        Highly efficient vitrification for cryopreservation of human oocytes and embryos: the Cryotop method.
        Theriogenology. 2007; 67: 73-80
        • Gardner D.K.
        • Surrey E.
        • Minjarez D.
        • Leitz A.
        • Stevens J.
        • Schoolcraft W.B.
        Single blastocyst transfer: a prospective randomized trial.
        Fertil Steril. 2004; 81: 551-555
        • Zhang J.
        • Chang L.
        • Sone Y.
        • Silber S.
        Minimal ovarian stimulation (mini-IVF) for IVF utilizing vitrification and cryopreserved embryo transfer.
        Reprod Biomed Online. 2010; 21: 485-495
        • Ciray H.N.
        • Campbell A.
        • Agerholm I.E.
        • Aguilar J.
        • Chamayou S.
        • Esbert M.
        • et al.
        Proposed guidelines on the nomenclature and annotation of dynamic human embryo monitoring by a time-lapse user group.
        Hum Reprod. 2014; 29: 2650-2660
        • Bodri D.
        • Sugimoto T.
        • Serna J.Y.
        • Kondo M.
        • Kato R.
        • Kawachiya S.
        • et al.
        Influence of different oocyte insemination techniques on early and late morphokinetic parameters: retrospective analysis of 500 time-lapse monitored blastocysts.
        Fertil Steril. 2015; 104: 1175-1181.e2
        • Cruz M.
        • Garrido N.
        • Gadea B.
        • Munoz M.
        • Perez-Cano I.
        • Meseguer M.
        Oocyte insemination techniques are related to alterations of embryo developmental timing in an oocyte donation model.
        Reprod Biomed Online. 2013; 27: 367-375
        • Liu Y.
        • Chapple V.
        • Feenan K.
        • Roberts P.
        • Matson P.
        Time-lapse videography of human embryos: using pronuclear fading rather than insemination in IVF and ICSI cycles removes inconsistencies in time to reach early cleavage milestones.
        Reprod Biol. 2015; 15: 122-125
        • Bodri D.
        • Kawachiya S.
        • De Brucker M.
        • Tournaye H.
        • Kondo M.
        • Kato R.
        • et al.
        Cumulative success rates following mild IVF in unselected infertile patients: a 3-year, single-centre cohort study.
        Reprod Biomed Online. 2014; 28: 572-581
        • Basile N.
        • Vime P.
        • Florensa M.
        • Aparicio Ruiz B.
        • Garcia Velasco J.A.
        • Remohi J.
        • et al.
        The use of morphokinetics as a predictor of implantation: a multicentric study to define and validate an algorithm for embryo selection.
        Hum Reprod. 2015; 30: 276-283
        • Seshagiri P.B.
        • Sen Roy S.
        • Sireesha G.
        • Rao R.P.
        Cellular and molecular regulation of mammalian blastocyst hatching.
        J Reprod Immunol. 2009; 83: 79-84
        • Gonzales D.S.
        • Bavister B.D.
        Zona pellucida escape by hamster blastocysts in vitro is delayed and morphologically different compared with zona escape in vivo.
        Biol Reprod. 1995; 52: 470-480
        • Erbach G.T.
        • Biggers J.D.
        • Manning P.C.
        • Nowak R.A.
        Localization of parathyroid hormone-related protein in the preimplantation mouse embryo is associated with events of blastocyst hatching.
        J Assist Reprod Genet. 2013; 30: 1009-1015
        • Niimura S.
        Time-lapse videomicrographic analyses of contractions in mouse blastocysts.
        J Reprod Dev. 2003; 49: 413-423
        • Basile N.
        • Nogales Mdel C.
        • Bronet F.
        • Florensa M.
        • Riqueiros M.
        • Rodrigo L.
        • et al.
        Increasing the probability of selecting chromosomally normal embryos by time-lapse morphokinetics analysis.
        Fertil Steril. 2014; 101: 699-704