{"paper_id":"a2d8f923-d6f8-46c5-812f-81b0a2e2955f","body_text":"To improve clinical outcomes of medically assisted reproduction (MAR), the transfer of multiple embryos is frequently performed ( Gliozheni  et al.  2023 ). This leads to a significantly higher rate of multiple pregnancies after MAR, along with the associated increase of maternal, fetal and neonatal complications compared to singleton pregnancies ( Eapen  et al.  2020 ,  Gliozheni  et al.  2023 ,  ESHRE Guideline Group on the Number of Embryos to Transfer  et al.  2024 ). Consequently, several international and national professional bodies have recommended (elective) single embryo transfers ( Practice Committee of the American Society for Reproductive Medicine and the Practice Committee for the Society for Assisted Reproductive Technologies 2021 ,  ESHRE Guideline Group on the Number of Embryos to Transfer  et al.  2024 ). In Europe, this resulted in a constant increase in the rate of single embryo transfers (SET) and a decrease in the rate of multiple pregnancies ( Gliozheni  et al.  2023 ). Among European countries reporting to the European IVF Monitoring Consortium (EIM), the overall rate of SET performed in 2019 was 55.4%, with large differences among the reporting countries, ranging from 100% to as low as 1.5% ( Gliozheni  et al.  2023 ). In Germany, the percentage of SET was 34.4%, double embryo transfer (DET) was 62.6% and triple embryo transfer was 3.0% ( Gliozheni  et al.  2023 ). This resulted in a twin rate of 18.1% and triplet rate of 0.4% ( Gliozheni  et al.  2023 ). Due to a further increase in the proportion of SET until 2021, the twin and triplet rates were further reduced to 15.2 and 0.3%, respectively ( Barnitzky  et al.  2024 ).\nThe continuing transfer of multiple embryos suggests that this strategy is still considered beneficial to improve the effectiveness in many MAR treatments, while accepting the safety concerns. Female patients’ age is often used as a predictor for clinical outcomes. Recommendations on the number of embryos to transfer by the ASRM are based on female patient age and modifiable by an individual prognosis ( Practice Committee of the American Society for Reproductive Medicine and the Practice Committee for the Society for Assisted Reproductive Technologies 2021 ).\nIn addition to female patients’ age, embryo quality impacts clinical outcomes ( Gardner  et al.  2000 ,  Racowsky  et al.  2011 ,  Vernon  et al.  2011 ). Embryo morphology and kinetics can be assessed to predict the developmental potential before transfer ( Ciray  et al.  2014 ). Extended  in vitro  culture has been developed to identify the embryo with the highest developmental potential for elected single embryo transfers (eSETs) with the aim to improve clinical outcomes, while avoiding multiple pregnancies ( Gardner & Schoolcraft 1999 ). This specific benefit is often ignored, leading to significantly higher rates of multiple pregnancies due to the transfer of more than one blastocyst ( Cutting 2018 ).\nIn Germany, the Embryo Protection Act limits the number of embryos to be transferred up to three ( Act for the Protection of Embryos 1990 ). In addition, the number of fertilized oocytes per cycle that can be cultured  in vitro  to embryos is limited and depends on the number of fertilized oocytes that is deemed necessary to obtain the desired number of viable embryos on the day of transfer ( Taupitz & Hermes 2015 ,  Nationale Akademie der Wissenschaften Leopoldina and Union der deutschen Akademien der Wissenschaften 2019 ). A typical eSET, in which all fertilized oocytes are cultured  in vitro , irrespective of the total number, and a single embryo is selected from this group, is therefore not allowed in Germany ( Nationale Akademie der Wissenschaften Leopoldina and Union der deutschen Akademien der Wissenschaften 2019 ). Furthermore, the number of embryos to be transferred is determined before the  in vitro  culture of embryos and takes predictive factors such as female patients’ age, medical conditions of the patient, prior pregnancies, the experience from previous MAR attempts and the personal wish of the patient into account.\nThus, in clinical practice, the number of embryos to be transferred is still mainly a decision based on patients’ and embryo characteristics to balance between effectiveness and safety of the treatment ( Ma  et al.  2022 ).\nIn this retrospective cohort study, we explored in fresh transfer cycles if a cohort of patients can be defined that benefits from DET in terms of clinical pregnancy and live birth rates (LBRs), while mitigating the odds of multiple pregnancy, based on clinical factors including female patients’ age, embryo quality and further stratification based on embryo cohort quality on the day of transfer after extended embryo culture.\n\nA retrospective cohort study was conducted at the Department of Gynecologic Endocrinology and Fertility Disorders, Heidelberg University Women’s Hospital, Ruprecht-Karl University Heidelberg, Germany. Between October 2016 and June 2020, all fresh cycle IVF- and ICSI-patients with embryos cultured up to day 4 or 5 and an embryo transfer with a known pregnancy outcome were retrospectively analyzed. Data were obtained from the prospectively maintained hospital electronic medical record system. Following the German Embryo Protection Act, the number of embryos cultured  in vitro  was limited. Up to five embryos were cultured, both for SET (mean: 3.3 ± 0.04, min. 1, max. 5) and for DET (mean: 4.0 ± 0.04, min. 2, max. 5). During the study period,  n  = 1,863 ovum pickups were performed. These led to  n  = 1,203 transfers, of which  n  = 754 were performed on day 4 or 5. Exclusion criteria were no embryo transfer ( n  = 660) or embryo transfer before day 4 ( n  = 449), female age >42 years on the day of transfer ( n  = 32), preimplantation genetic testing ( n  = 12), assisted hatching ( n  = 43) and calcium-ionophore treatment ( n  = 21). To reduce bias, only the first fresh cycle of each female patient within this period was included for the analysis, excluding additional transfers during the study period ( n  = 112). Additional cycles were excluded due to relevant co-medication ( n  = 2), missing data ( n  = 3), unknown pregnancy outcome ( n  = 5) and transfer of an additional thawed blastocyst ( n  = 1), leaving  n  = 523 transfers for the final analysis.\nOutcomes were clinical pregnancy rate (CPR) (defined as the number of clinical pregnancies per number of embryo transfers), live birth rate (LBR) (defined as the number of live births per number of embryo transfers) and multiple pregnancy rate (MPR) (defined as the number of multiple pregnancies per number of clinical pregnancies).\nThe stimulation protocol was based on individual patients’ characteristics (e.g., ovarian reserve, polycystic ovary syndrome (PCOS) and endometriosis), time available and preference of the patient and physician (e.g. women with PCOS received mostly the gonadotropin-releasing hormone (GnRH) antagonist protocol, while women with endometriosis were more likely to receive the GnRH agonist protocol). A consistent rise in E2 levels and follicle growth was monitored until the presence of three or more follicles >17 mm in diameter. Ovum pick-up followed 36 h after ovulation induction by ultrasound-guided aspiration with a 17-gauge ovum aspiration needle (Cook, K-OSN-1730-B-90, Limerick, Ireland) and an aspiration pressure of 120 mmHg.\nEmbryo grading on day 4 and 5 was performed as previously described with modification ( Roesner  et al.  2017 ,  Dietrich  et al.  2020 ). Briefly, cleavage stage embryos were graded as A = stage-specific cell size and cytoplasmic fragmentation <10%; B = stage-specific cell size and cytoplasmic fragmentation 10–25%; C = cell size not stage-specific and/or cytoplasmic fragmentation 26–50%; and D = fragmentation >50%. Blastocysts were scored according to  Gardner & Schoolcraft (1999) . Good quality embryos (GQEs) were defined as 9–16-cell and grade A or B on day 4. Compacting or fully compacted morulae on day 4 were considered GQE if their quality was grade A or B on day 3 (independent of cell number on day 3). On day 5, GQE was defined as a blastocyst with an expansion 3–6, inner cell mass (ICM) quality A-B and trophectoderm (TE) quality A-B.\nThe embryo quality was translated into a score (embryo score) as follows. Embryo score 2 was attributed to GQEs, embryo score 1 was attributed to poor quality embryos (PQEs) that were considered for embryo transfer or cryopreservation and embryo score 0 was attributed to PQEs that were discarded. In the analyses, embryo score describes only the quality of the transferred embryos.\nThe embryo cohort quality is shown as a score calculated as the mean embryo score of all cultured embryos of the same treatment cycle. For analysis, the cohorts were grouped based on their embryo cohort score: 0.0–1.0, 1.1–1.5 and ≥1.6.\nIn addition to the assessment of morphology, time-lapse imaging was used to identify abnormally developing embryos not suitable for transfer. Conspicuous kinetics (such as reverse cleavage or direct cleavage) could lead to a deprioritization of embryos for transfer independent of morphological quality.\nEmbryos were transferred on day 4 or 5 to allow for more flexibility in scheduling the transfer. A previous study by our group showed that, in non-selected couples, a transfer on day 4 or 5 results in similar CPRs, ongoing pregnancy rates and MPRs ( Holschbach  et al.  2017 ).\nAfter IVF or ICSI embryos were cultured in Continuous Single Culture Complete (CSCM-C, 90165, FUJIFILM Irvine Scientific, FUJIFILM Europe B.V., The Netherlands) in an EmbryoScope (ES-D2, Vitrolife Sweden AB, Sweden) or EmbryoScope + (ES-P1, Vitrolife Sweden AB, Sweden) at 5% O 2 .\nContinuous variables were described by the mean values ± standard deviations (SD) and compared using independent samples  T -test (two-sided p, equal variances not assumed). Categorical variables are presented with their absolute and relative numbers and associations were tested using Pearson chi-Square or Fisher’s exact test (exact sig., two-sided). Generalized linear model (binary logistic) with Wald chi-square statistics and a robust estimator for covariance was used to analyze binary clinical outcomes (clinical pregnancy, live birth or multiple pregnancies of DET). Covariates were analyzed by binary logistic regression with forward selection (likelihood ratio). Based on the results of this analysis, female age, embryo quality and female smoking habit were included in the model to calculate Exp(B), shown as adjusted odds ratio (aOR), and the lower and upper 95% Wald confidence interval for Exp(B) (95 CI). Subgroups with less than ten events were analyzed with a univariate model and results are shown as odds ratio (OR).  P -values ≤0.05 were considered significant. The Institute of Medical Biometry (IMBI) of the Heidelberg University was consulted for advice on statistical analyses. Statistical analyses were performed using the SPSS Version 29.0 (SPSS, Inc, USA).\nThis study was approved by the Ethics Committee of the Medical Faculty Heidelberg (S-649/2016) and conducted according to the principles of the Declaration of Helsinki.\n\nIn our study population, the baseline characteristics female patients’ age, female smoking habit, endometrial thickness (<8 or ≥8 mm) and the number of cultured embryos were significantly different between SET and DET groups ( Table 1 ). Body mass index (BMI), fertilization method, number of oocytes retrieved and number of fertilized oocytes (pronuclear stage oocytes, PNs) did not differ significantly between SET and DET groups ( Table 1 ).\nBaseline characteristics of the study population. Continuous variables are shown as the mean ± SD (min – max) and were tested by independent samples  T -test. Categorical variables are shown as number (percentage) and were tested by Pearson chi-Square (asymptotic significance, two-sided).\nBMI, body mass index; D4, day 4; D5, day 5; DET, double embryo transfer; Gravida, pregnancies; OPU, ovum pick-up;  P ,  P -value; Para, live births; PNs, pronuclear stage oocytes; SET, single embryo transfer.\nBinary logistic regression with forward selection (likelihood ratio) showed that clinical pregnancy and live birth were significantly affected by female age (OR: 0.931, 95% CI: 0.892–0.972,  P  = 0.001 and OR: 0.923, 95% CI: 0.882–0.966,  P  < 0.001), embryo quality (OR: 2.763, 95% CI: 1.733–4.406,  P  < 0.001 and OR: 3.248, 95% CI: 1.926–5.478,  P  < 0.001) and female smoking habit (OR: 0.504, 95% CI: 0.277–0.914,  P  = 0.024 and OR: 0.444, 95% CI: 0.227–0.864,  P  = 0.017). Clinical pregnancy or live birth were not significantly affected by female BMI ( P  = 0.292 or  P  = 0.439), endometrial thickness ( P  = 0.691 or  P  = 0.510), number of prior ovum pick-ups ( P  = 0.623 or  P  = 0.996), number of prior embryo transfers ( P  = 0.098 or  P  = 0.545), number of embryos cultured ( P  = 0.597 or  P  = 0.080), number of prior gravida ( P  = 0.407 or  P  = 0.677), number of prior para ( P  = 0.573 or  P  = 0.479) or the day of embryo transfer ( P  = 0.295 or  P  = 0.274), respectively.\nAfter embryo transfer ( n  = 523), the CPR was 42.1% ( n  = 220/523) and the LBR was 32.1% ( n  = 168/523). The number of transferred embryos had a significant effect on clinical outcomes ( Table 2A ). The CPR was 33.5% ( n  = 82/245) after SET vs 49.6% ( n  = 138/278) after DET (aOR: 2.233, 95% CI: 1.529–3.261,  P  < 0.001). The LBR was 24.1% ( n  = 59/245) after SET vs 39.2% ( n  = 109/278) after DET (aOR: 2.416, 95% CI: 1.605–3.636,  P  < 0.001). The MPR was 0.0% ( n  = 0/82) after SET vs 25.4% ( n  = 35/138) after DET ( P  < 0.001).\nClinical outcomes after SET or DET. Data are shown as ratio (percentage). CPR and LBR tested with a generalized linear model (binary logistic) including number and quality of embryos transferred, female age and smoking habit. MPR was tested with a Fisher’s exact test.\nFisher’s exact test.\naOR (95 CI), adjusted odds ratio (95% confidence interval); CPR, clinical pregnancy rate; DET, double embryo transfer; LBR, live birth rate; MPR, multiple pregnancy rate; SET, single embryo transfer.\nWe then explored the dataset to test if a patients’ cohort can be identified that benefits significantly from DET while mitigating the odds of multiple pregnancy by performing subgroup analysis based on female age, embryo quality and cohort quality.\nConsidering female patients’ age, both younger (<35 years) and older (≥35 years) female patients had significantly higher CPR (36.2 vs 58.6%, aOR: 2.832, 95% CI: 1.589–5.047,  P  < 0.001 and 31.0 vs 43.7%, aOR: 1.716, 95% CI: 1.027–2.866,  P  = 0.039), LBR (28.4 vs 48.6%, aOR: 2.605, 95% CI: 1.448–4.686,  P  = 0.001 and 20.2 vs 32.9%, aOR: 2.024, 95% CI: 1.143–3.585,  P  = 0.016) and MPR (0.0 vs 30.8%,  P  < 0.001 and 0.0 vs 20.5%,  P  < 0.001) after DET compared to SET (SET vs DET) ( Table 2B and C ).\nFurther subgroup analysis considered embryo quality ( Table 3 ). Compared to the transfer of a single PQE, the transfer of two PQEs did not significantly increase the CPR (18.9 vs 32.4%, OR: 2.057, 95% CI: 0.835–5.066,  P  = 0.117), but significantly increased the LBR (10.8 vs 27.0%, OR: 3.056, 95% CI: 1.089–8.575,  P  = 0.034) and MPR (0.0 vs 33.3%,  P  = 0.033) ( Table 3A ). Compared to the transfer of a single GQE, the transfer of an additional PQE did not significantly improve the CPR (39.8 vs 43.9%, OR: 1.184, 95% CI: 0.716–1.958,  P  = 0.510) or LBR (29.8 vs 28.6%, OR: 0.941, 95% CI: 0.545–1.627,  P  = 0.828), but the MPR was significantly increased (0.0 vs 23.3%,  P  < 0.001). Compared to the transfer of a single GQE, the transfer of two GQEs significantly increased the CPR (39.8 vs 58.0%, OR: 2.095, 95% CI: 1.334–3.292,  P  = 0.001), LBR (29.8 vs 49.7%, OR: 2.320, 95% CI: 1.460–3.688,  P  < 0.001) and MPR (0.0 vs 25.3%,  P  < 0.001) ( Table 3A ).\nClinical outcomes after SET or DET based on embryo quality. Data are shown as ratio (percentage). CPR and LBR were tested with a generalized linear model (binary logistic). MPR was tested with a Fisher’s exact test.\nFisher’s exact test.\nEmbryo quality 1, poor quality embryo (PQE or PQE/PQE); 2, good quality embryo (GQE or GQE/GQE); 1.5, GQE/PQE; CPR, clinical pregnancy rate; DET, double embryo transfer; LBR, live birth rate; MPR, multiple pregnancy rate; OR (95 CI), odds ratio (95% confidence interval); SET, single embryo transfer.\nConsidering embryo quality and female age, there were similar trends for increased CPR, LBR and MPR in both younger (<35 years) and older (≥35 years) female patients after DET compared to SET ( Table 3B and C ).\nTo test if further stratification allows us to identify a cohort with significantly reduced MPR after DET, we considered the embryo cohort quality. Whereas embryo scores refer to the quality of transferred embryos only, the embryo cohort score is calculated as the mean quality score of all cultured embryos. Among cycles with a cohort score of 0–1.0 ( n  = 314), the scores of transferred embryos included SET (1 = PQE,  n  = 68), SET (2 = GQE,  n  = 89), DET (1 = PQE/PQE,  n  = 35), DET (1.5 = GQE/PQE,  n  = 88) and DET (2 = GQE/GQE,  n  = 34). Among cycles with a cohort score of 1.1–1.5 ( n  = 117), the embryo scores included SET (1,  n  = 4), SET (2,  n  = 40), DET (1,  n  = 2), DET (1.5,  n  = 7) and DET (2,  n  = 64). Among cycles with a cohort score of 1.6–2 ( n  = 92), the embryo scores included SET (1,  n  = 2), SET (2,  n  = 42), DET (1.5,  n  = 3) and DET (2,  n  = 45).\nThe CPR and LBR correlated to the cohort score, with higher scores resulting in higher CPR and LBR (Supplementary Table 1A, B, C (see section on  Supplementary materials  given at the end of the article)). After DET, the MPR was high in all cohort score subgroups; there was no subgroup with significantly reduced MPR ( Table 4A , Supplementary Table 1A, B, C).\nMultiple pregnancy rates after the transfer of embryos depending on cohort score in all patients and in female patients <35 or ≥35 years. Data are shown as ratio (percentage) and were tested with a generalized linear model (binary logistic).\nCS, embryo cohort score; MPR, multiple pregnancy rate; OR (95 CI), odds ratio (95% confidence interval).\nStratification based on embryo quality and additionally based on embryo cohort quality was only performed for DET of GQE because case numbers of other combinations were too low for analysis. After the transfer of two GQEs, the CPR and LBR showed a trend corresponding to the cohort score, with higher cohort scores resulting in higher CPR and LBR (Supplementary Table 2A, B, C). After DET, the MPR was high in all cohort score subgroups; there was no subgroup with significantly reduced MPR ( Table 4B , Supplementary Table 2A, B, C).\n\nData from international and national registries suggest that the transfer of multiple embryos is still performed in many treatment cycles in MAR ( Barnitzky  et al.  2024 ,  Gliozheni  et al.  2023 ). Our data confirm that overall, DET results in significantly higher CPR and LBR compared to SET. For the decision between SET and DET, the challenge is to find a balance between highest possible effectiveness and reducing the safety concern of multiple pregnancies with their associated maternal, fetal and neonatal risks.\nFemale patients’ age and embryo quality are known to affect the clinical outcome of MAR treatments ( Gardner  et al.  2000 ,  Racowsky  et al.  2011 ,  Vernon  et al.  2011 ,  Vitagliano  et al.  2023 ). Our data confirm that both female age and embryo quality significantly affect CPR and LBR.\nConsidering that in clinical practice embryos are usually transferred hierarchically with respect to embryo quality, we explored clinical outcomes for the transfer of a single PQE compared to DET of two PQEs or of a single GQE compared to DET of one GQE and one PQE or two GQEs. A trend for a better clinical outcome (CPR and LBR) was observed in most cases of DET, in line with results from a recent meta-analysis ( Ma  et al.  2022 ).\nOur data show that PQE also have the potential to implant and give rise to live births, although at a reduced rate. Compared to SET (PQE), a DET (PQE/PQE) significantly improved the LBR, but compared to SET (GQE), a DET with an additional PQE did not significantly increase the CPR or LBR. These data support findings showing that the additional transfer of a PQE does not negatively affect clinical outcomes ( Wintner  et al.  2017 ,  Dobson  et al.  2018 ,  Wang  et al.  2020 ,  Theodorou  et al.  2021 ). However, in all cases, the additional transfer of a PQE increased the MPR, consistent with previous reports ( Wintner  et al.  2017 ,  Dobson  et al.  2018 ,  Wang  et al.  2020 ,  Theodorou  et al.  2021 ). Although there was no statistically significant difference in the MPR after DET (PQE/PQE) compared to SET (PQE) in the subgroup analysis based on female patients’ age, the MPR was high after DET. The lost significance in the subgroup analysis may be due to small numbers.\nThus, our data support the recommendation that the decision to perform DET should not be based on female patients’ age and embryo quality ( ESHRE Guideline Group on the Number of Embryos to Transfer  et al.  2024 ).\nIn this study, we explored if additional stratification could identify a cohort of patients with significantly reduced odds of multiple pregnancy. A previous study found that cycles with cryopreservation of supernumerary embryos resulted in increased LBRs compared to those without cryopreservation, suggesting that the availability of supernumerary embryos suitable for cryopreservation may indicate good quality ( Stern  et al.  2012 ). The quality of supernumerary embryos was shown to have an impact on the clinical outcome, where the availability of additional GQE led to a better clinical outcome ( Salha  et al.  2000 ). Machtinger  et al.  showed that the presence of a non-cleaved embryo was an indicator of an overall reduced quality of embryos in the cohort ( Machtinger  et al.  2015 ). In this study, we used the cohort score of all cultured embryos as a proxy for overall cycle quality and explored its association with the clinical outcome of MAR cycles. Although the cohort score is affected by the score of the transferred embryo ( Romanski  et al.  2018 ), especially in relatively small cohorts of  in vitro  cultured embryos, the cohort score does not only reflect the score of the transferred embryos. Supernumerary embryos can impact on the cohort score such that cycles with only GQE will have a higher cohort score than cycles where only the transferred embryos were GQE. Thus, we anticipated an additional useful stratification by considering the cohort score.\nThe clinical outcome after the transfer of GQE correlated to the cohort score. Transfer of GQE from cohorts with increasing scores had a trend for a better clinical outcome. An explanation for the improved outcome with additional stratification based on cohort score may be that cohort scores better reflect the quality of the embryos available for transfer than the simplified binary classification GQE vs PQE. GQE and PQE each include a variety of different embryo qualities. Cycles with a high cohort score have more embryos of good quality and with a hierarchical transfer strategy the best ones among GQE will be chosen for transfer. Thus, in agreement with findings by Romanski  et al.  we found that the cohort score may be useful to describe the overall quality of the cycle ( Romanski  et al.  2018 ). However, after DET, all subgroups had high MPR and stratification based on cohort score did not enable us to identify a subgroup with reduced odds of MPR. Thus, in this exploratory study, we were unable to define a subgroup based on female age, embryo quality or further stratification based on the embryo cohort quality with acceptable reduction of the MPR.\nIt is important to highlight that for multiple pregnancies, there was a quasi-complete separation of the data in our study population based on the number of embryos transferred (SET vs DET). Whereas no multiple pregnancies had occurred after SET, all subgroups with a DET were affected by multiple pregnancies. Therefore, the number of embryos transferred was the only variable that could reliably control the odds of a multiple pregnancy. This is expected, since after SET, twin pregnancies may occur only in the rare case of monozygotic twinning ( Blickstein 2005 ). The spontaneous monozygotic twinning rate is estimated to being approximately 0.4% ( Blickstein  et al.  1999 ,  Chen  et al.  2023 ). In MAR, this rate is increased to 1.6% ( Chen  et al.  2023 ).\nConsidering effectiveness, an alternative strategy to the transfer of multiple embryos is sequential SET. With the improved outcome after cryopreservation, sequential treatment strategies with fresh and frozen-warmed cycles have become an option ( Nagy  et al.  2020 ). Considering cumulative pregnancy rates, SET and DET strategies result in a similar clinical outcome ( Kamath  et al.  2020 ). Thus, sequential SET appears to be as effective as DET, while mitigating the risks of a multiple pregnancy.\nWe acknowledge limitations in our study. A limiting factor is the restricted number of embryos that can be cultured due to legal regulations in Germany imposed by the German Embryo Protection Act ( Act for the Protection of Embryos 1990 ,  Taupitz & Hermes 2015 ,  Nationale Akademie der Wissenschaften Leopoldina and Union der deutschen Akademien der Wissenschaften 2019 ). Because of that, the analyzed embryo cohort quality may not fully reflect the cycle quality. In this study, extended  in vitro  culture included transfers on day 4 and day 5, which may be a limitation. Female age was tested for groups <35 years and ≥35 years. Choosing different age categories may result in other outcomes. A further limitation is the retrospective and exploratory design including subgroup analyses and resulting in small numbers analyzed. Results require confirmation with larger numbers. This is a single center study, and outcomes may differ in other settings.\n\nCompared to single embryo transfer, the transfer of two embryos is associated with significantly increased CPRs and LBRs. The additional transfer of a PQE in DET did not negatively affect the clinical outcomes (CPR and LBR). Subgroup analysis based on female patients’ age, embryo quality and further stratification based on embryo cohort quality revealed that after DET, the MPR remained high in all subgroups. Considering the risks associated with multiple pregnancies, these data support the preference of SET over DET.\n\nThe authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.\n\nFor the publication fee, we acknowledge financial support by  Heidelberg University . E Capp is a recipient of a scholarship from CNPq–Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil. No further funding was received for this study.\n\nJED helped in study conception and design, data acquisition, data analysis and interpretation, manuscript drafting and revision. ICV helped in study conception and design, data acquisition, data analysis and interpretation, manuscript drafting and revision. EC helped in data analysis and interpretation, manuscript drafting and revision. TS helped in study conception and design, data analysis and interpretation, manuscript revision. AG helped in study conception and design, data analysis and interpretation, manuscript drafting and revision. All authors gave final approval of the version to be published and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.","source_license":"CC-BY-4.0","license_restricted":false}