Catherine Racowsky

The number of eight-cell embryos is a key determinant for selecting day 3 or day 5 transfer

OBJECTIVE To select patients for day 3 vs. day 5 embryo transfer. DESIGN Retrospective analysis of assisted reproduction technology (ART) cycles comparing outcomes of day 3 and day 5 transfers. SETTING ART program of Brigham and Womens Hospital. PATIENT(S) Patients with day 3 or day 5 embryo transfers (n = 221 and 141, respectively). INTERVENTION(S) Cycles with eight or more zygotes were stratified by the number of eight-cell embryos available on day 3 (none, one or two, or three or more). MAIN OUTCOME MEASURE(S) Number of blastocysts, implantation rates, ongoing pregnancy rates, and number of fetal heart beats. RESULT(S) With no eight-cell embryos on day 3, 0% and 33% pregnancies resulted from day 5 vs. day 3 transfers. With one or two eight-cell embryos on day 3, ongoing and high order multiple rates were not different between day 3 and day 5 transfers. With three or more eight-cell embryos, day 5 transfer resulted in a decrease in multiple gestations but no difference in ongoing pregnancy rates compared with day 3 transfer. CONCLUSION(S) With no eight-cell embryos on day 3, a day 3 transfer is warranted. With one or two eight-cell embryos, any benefit of day 5 transfer appears to be equivocal. With three or more eight-cell embryos, day 5 transfer is recommended.

Blastocyst-ET and Monozygotic Twinning

AbstractPurpose: To examine the rate of monozygotic twinning associatedwith blastocyst transfer using commercially available,cell-free culture systems with unmanipulated blastocysts. Methods: A retrospective analysis was conducted in multipleprivate and academic infertility centers throughout theUnited States, of 199 pregnant patients following in vitrofertilization (IVF) blastocyst embryo transfer (ET). Humanembryos obtained through standard IVF stimulation protocolswere cultured in commercially available, cell-free mediasystems and transferred as blastocysts. The main outcomemeasure was the rate of monozygotic twinning. Results: A total of 199 blastocyst-ET pregnancies wereachieved during the study period at the fertility centersexamined. Monozygotic twinning was noted in 10/199 (5;pc)of these pregnancies. All were monochorionic diamnionic. Conclusions: Monozygotic twinning previously has beenreported following IVF, especially in relation to assistedhatching. While blastocyst transfer has been available formany years using coculture, there have been no publishedmulticenter reports of monozygotic twinning associated withunmanipulated blastocysts. In a multicenter analysis, a definiteincrease in monozygotic twinning was seen followingblastocyst-ET. We believe this phenomenon is real and thatthis information should be considered when counselingpatients for treatment.

Day 3 and day 5 morphological predictors of embryo viability

Controlling multiple pregnancies in patients undergoing artificial reproductive procedures requires consideration of single embryo transfers. Therefore, refinements for embryo evaluation are needed that select for the most developmentally competent embryo. The present study was designed to identify day 3 and day 5 morphological predictors of viability following transfers in which the morphology and fate of each embryo was precisely determined. Assessments on day 3 included cell number, and the extent of fragmentation and asymmetry, and on day 5, the developmental stage. Embryos resulting in a viable fetus at 11 weeks gestation were considered developmentally competent. The relationships among individual and collective embryo morphological characteristics were evaluated. Analysis of the interactions among morphological characteristics of embryos transferred on day 3 enabled identification of a multivariable selection order. Assessment of day 5 embryos revealed that expanding and expanded blastocysts exhibited comparable developmental potential that was superior to that of either morulae or early blastocysts. However, expanding or expanded blastocysts derived from 7-cell or 8-cell embryos were developmentally superior to those derived from other cleavage stages, regardless of fragmentation or asymmetry. Collectively, these findings further understanding of morphological predictors of viability, thereby improving the ability to select the most viable embryo for transfer.

Effect of obesity on oocyte and embryo quality in women undergoing in vitro fertilization.

OBJECTIVE: To estimate the effect of body mass index (BMI) on oocyte and embryo parameters and cycle outcomes in women undergoing in vitro fertilization (IVF). METHODS: We evaluated a retrospective cohort of 1,721 women undergoing a first IVF cycle with fresh, autologous embryos between 2007 and 2010 in an academic infertility practice. Main outcome measures included number of mature and normally fertilized oocytes, embryo morphology, estradiol on the day of human chorionic gonadotropin administration, clinical pregnancy, spontaneous abortion, and live birth. We performed multivariable analyses, adjusting for potential confounders, including age at cycle start, infertility diagnosis, type of stimulation, total gonadotropin dose, use of intracytoplasmic sperm injection, and number of embryos transferred. RESULTS: Compared with women of normal BMI, women with class II (BMI 35–39.9) and III (BMI 40 or higher) obesity had fewer normally fertilized oocytes (9.3 compared with 7.6 and 7.7, P<.03) and lower estradiol levels (2,047 pg/mL compared with 1,498 and 1,361, P<.001) adjusting for age and despite similar numbers of mature oocytes. Odds of clinical pregnancy (odds ratio [OR] 0.50, 95% confidence interval [CI] 0.31–0.82) and live birth (OR 0.51, 95% CI 0.29–0.87) were 50% lower in women with class III obesity as compared with women of normal BMI. CONCLUSION: Obesity was associated with fewer normally fertilized oocytes, lower estradiol levels, and lower pregnancy and live birth rates. Infertile women requiring IVF should be encouraged to maintain a normal weight during treatment. LEVEL OF EVIDENCE: II

Human embryonic stem cell derivation from poor-quality embryos

During in vitro fertilization, embryos deemed clinically useless based on poor morphology are typically discarded. Here we demonstrate a statistical correlation between the developmental stage of such poor-quality embryos and the yield of human embryonic stem (hES) cell lines. Early-arrested or highly fragmented embryos only rarely yield cell lines, whereas those that have achieved blastocyst stage are a robust source of normal hES cells.

Clinical outcomes following selection of human preimplantation embryos with time-lapse monitoring: a systematic review

Sir, We have read the recent article by Kaserand Racowsky (2014) with great interest and we highly welcome this thorough systematic review. We entirely agree with the main conclusions as presented in the abstract. There are currently no high-quality data to firmly support the clinical use of this technology for selection of preimplantation embryos. Prospective studies are needed to clarify the role. We would, however, like to discuss the results from the review upon which Drs Kaser and Racowsky base this conclusion. The authors thoroughly review the time-lapse studies that have presented data on clinical outcome and present the results of their descriptive comparison, including their interpretation of the results from our prospective study (Kirkegaard et al., 2013). Amongst others the authors use the results from our study to argue that there are no differences in timing between the pregnant and the nonpregnant group of all the measured parameters. Our objection is that our study was not powered to test pregnancy as a clinical outcome for all the parameters that the authors review. Accordingly, we specifically desisted from drawing any conclusions on the ability of time-lapse parameters to predict pregnancy in general. This is clearly stated in our paper. Following standard scientific conduct, we did publish timings of all the parameters in the pregnant and non-pregnant group, yet acknowledged that the study was powered only to test the parameters from the targeted logistic regression analysis. We even clarified this in a response to a letter addressing the sample size (Kirkegaard et al., 2014). We believe that the underpowered sample size entails a high risk of falsely concluding that there is no difference. For example, we did not test appearance, abuttal, syngamy, and breakdown of the male and female pronucleus (PN) as predictors of implantation in our logistic regression analysis (as stated in the review), but only PN breakdown. Therefore it is hardly justified to state that we did not find any difference in the above parameters, without acknowledging the lack of power to detect such differences. This is true for several of the conclusions the authors draw from our publication, including the conclusion that we found no difference between implanting and non-implanting embryos in terms of cleavage and blastocyst kinetics in general. We consider it plausible that the majority of the other published studies are far too small to detect any presumed differences in timing with regard to pregnancy. No randomized controlled studies of single embryo transfers have been published so far. We therefore find it very poorly supported, that reliable prediction of blastocyst formation may be the main advantage of TLM, as stated in the review. It is correct that we conclude that TLM may decrease variability (Sundvall et al., 2013). But as the cited study involved manual, in contrast to computer-assisted annotation, it cannot be stated that the reduced variability is a result of the semi-quantification. The statement that TLM may decrease intraand inter-observer variability among embryologists, as a result (our underlining) of computer-assisted annotation of developmental milestones and semi-quantitative process for embryo evaluation, is therefore unsupported. In summary, we entirely agree that larger prospective studies with clinical outcomes are needed to clarify the role of time-lapse. That the existing literature suggests no association with implantation potential is in our opinion so far unjustified due to several factors, most importantly due to lack of power of the studies. This was acknowledged in the original publications, but unfortunately not in the review.

The development of fertilized human ova to the blastocyst stage in KSOMAA medium: is a two-step protocol necessary?

Current protocols for the culture of human zygotes to blastocysts use two-step sequential media systems. The efficacy of a one-step system involving potassium simplex optimized medium (KSOM(AA)) has been investigated. In study 1, development of zygotes from days 1 to 3 in KSOM(AA) was compared with that for medium P-1. In study 2, embryos were cultured from days 1 to 3 in P-1 followed by culture from days 3 to 5 either in KSOM(AA) or medium CCM. In study 3, the ability of KSOM(AA) to support development of embryos from days 1 to 5, without medium renewal, was compared with the sequential media system P-1-->CCM. The cell numbers and fragmentation scores of day 3 embryos were distributed similarly following culture in KSOM(AA) or P-1. Significantly more KSOM(AA) embryos exhibited cytoplasmic pitting. Blastocyst formation rates were not significantly different whether embryos were cultured in the P-1-->KSOM(AA) or the P-1-->CCM systems, or when cultured from days 1 to 5 in KSOM(AA) without medium renewal or in P-1-->CCM. Five babies have been born from nine blastocysts transferred after extended culture in KSOM(AA). A one-step protocol involving KSOM(AA) can be used successfully to culture human zygotes to the blastocyst stage.

Is there an advantage in scoring early embryos on more than one day

BACKGROUND This study was undertaken to determine what characteristics should be recorded on which days to build a predictive model for selection of Day 3 embryos. METHODS Embryos failing to form a clinical sac or that formed a viable fetus (to > or =12 weeks), and transferred singly (n = 269) or in pairs (n = 1326) were scored for early cleavage and pronuclear status on Day 1, and cell number, fragmentation, and symmetry on Days 2 and 3, with number of nuclei per blastomere also recorded on Day 2. Seven candidate models were identified using a priori clinical knowledge and univariate analyses. Each model was fit on a training-set and evaluated on a test-set with resampling, with discrimination assessed using the area under the ROC curve (AUC) and calibration assessed using the Hosmer-Lemeshow statistics. RESULTS Models built using Day 1, 2 or 3 scores independently on the 30 resampled data sets showed that Day 1 evaluations provided the poorest predictive value (median AUC = 0.683 versus 0.729 and 0.725, for Day 2 and 3). Combining information from Day 1, 2 and 3 marginally improved discrimination (median AUC = 0.737). Using the final Day 3 model fitted on the whole dataset, the median AUC was 0.732 (95% CI, 0.700-0.764), and 68.6% of embryos would be correctly classified with a cutoff probability equal to 0.3. CONCLUSIONS Day 2 or Day 3 evaluations alone are sufficient for morphological selection of cleavage stage embryos. The derived regression coefficients can be used prospectively in an algorithm to rank embryos for selection.

Risk factors associated with pregnancies containing a monochorionic pair following assisted reproductive technologies

BACKGROUND Although several factors have been identified to predispose to an increased incidence of monozygotic twinning in assisted reproductive technologies (ART), the relative risks associated with each have yet to be fully established. Moreover, the focus has been predominantly on monozygosity, which, in the absence of monochorionicity, does not increase perinatal risk. The present objective was to undertake an analysis of the relative risks of factors associated with monochorionic pairs resulting from ART. METHODS Study cycles included the last cycle, of each patient undergoing ART at Brigham and Womens Hospital from January 1998 to December 2004, that resulted either in a pregnancy with a monochorionic pair (n = 41) or a pregnancy without a monochorionic pair at 12 weeks (n = 2460). We used multivariable logistic regression to estimate odds ratios (OR) and 95% confidence intervals (CI) to identify factors significantly associated with a monochorionic pair. RESULTS Independent predictors of a monochorionic pair were assisted hatching (OR 2.23, 95% CI 1.06-4.67), ICSI (OR 2.42, 95% CI 1.22-4.83) and Day 5 embryo transfer (OR 2.48, 95% CI 1.62-3.80). The effects of ICSI and Day 5 transfer were amplified when cycles involved both interventions. CONCLUSIONS ICSI and Day 5 embryo transfer synergistically increase the risk of monochorionic placentation. Patients undergoing these procedures should be counselled regarding these increased risks.

A novel system for in vitro maturation of human oocytes

OBJECTIVE To compare in vitro maturation of cumulus-free oocytes in glucose-free medium (P1) and standard medium (TC199). DESIGN Prospective, cohort study. SETTING Assisted reproductive technology program. PATIENT(S) One hundred eight patients undergoing ICSI. INTERVENTION(S) Germinal vesicle-stage or metaphase I--stage oocytes were allocated to culture with P1 or TC199. Metaphase II oocytes were fixed for immunofluorescence analysis or fluorescence in situ hybridization at 24 or 48 hours (or both). Media were compared by performing conditional logistic regression analysis that controlled for egg-specific factors. MAIN OUTCOME MEASURE(S) Proportion of mature oocytes and appearance of normal spindle-chromosome cytoarchitecture. RESULT(S) At 24 hours, more P1 oocytes than TC199 oocytes reached metaphase II (59.7% vs. 44.9%). At 48 hours, 71.7% of P1 oocytes and 61.0% of TC199 oocytes reached metaphase II, but this difference was not significant. Metaphase II oocytes in P1 were 34.3% more likely than those in TC199 to have a bipolar spindle with aligned chromosomes. Compared with oocytes at the germinal vesicle stage at 0 hour, those at metaphase I at 0 hour were more likely to progress to metaphase II (72.6% vs. 46.1% at 24 hours; 84.1% vs. 60.6% at 48 hours). CONCLUSION(S) P1 is superior to TC199 for in vitro maturation of granulosa-free human oocytes.