Risk of stillbirth in low-risk singleton term pregnancies following fertility treatment:
a national cohort study
生殖辅助治疗后低风险单胎妊娠的死胎风险
一项全国范围队列研究
ABSTRACT
Objective
To assess the risk of stillbirth in low-risk IVF pregnancies.
Design
Register-based national cohort study.
Setting
Denmark 2003-2013.
Population
Cohort of 425,732 singleton pregnancies including 10,235 conceived following IVF/ICSI, 4,521 conceived following IUI, and 410,976 spontaneous conceived.
Methods
Information on pregnancy, obstetrical risk factors, stillbirth, and fertility treatment was obtained from the Danish national health registers for all pregnancies after gestational week 21+6. We estimated the overall and gestational age specific risk of stillbirth in low-risk term pregnancies following IVF, ICSI, and IUI. Further, we estimated the association between stillbirth and IVF and ICSI respectively as well as fresh or frozen-thawed embryo transfer.
Main outcome measures
Risk of stillbirth.
Results
The number of stillbirth in spontaneously conceived and IVF/ICSI low-risk term pregnancies was 525 (0.1%) and 35 (0.3%), respectively. In multivariate analysis, the risk of stillbirth in pregnancies following IVF/ICSI was increased (odds ratio (95% CI): 2.1 (1.4; 3.1)). The risk of stillbirth was correspondingly increased in time to event analyses taking risk time for each fetus into account from gestational week 37 and onwards (Hazard Ratio (95% CI): 2.4 (1.6; 3.6). In subanalyses, the risk of stillbirth was increased for pregnancies following ICSI (OR 2.2 (1.2; 3.1)), but not IVF (OR 1.7 (0.9;3.1)).
Conclusion
We found a systematically increased risk of stillbirth in low-risk term pregnancies following IVF/ICSI. Whether the risk was related to the treatment or underlying subfertility is uncertain. The results may indicate a need for obstetrical surveillance for these pregnancies when reaching term.
Key words: Fertility treatment; IVF; ICSI; Stillbirth
https://doi.org/10.1111/1471-0528.15509
摘要
目的
评估低风险IVF妊娠死胎的风险。
设计
基于登记的全国范围的队列研究。
背景
丹麦2003-2013。
研究对象
425,732例单胎妊娠队列，其中10,235例在IVF / ICSI后受孕，4,521例在IUI后受孕，410,976例自发受孕。
方法
获得妊娠、产科危险因素、死胎和生育治疗方面的信息。在丹麦国家健康登记处记录所有妊娠21 + 6周后的怀孕人员。在经过IVF，ICSI和IUI之后，我们估计了低风险妊娠期间死胎的总体风险和胎龄特定风险。此外，我们分别估计死胎和IVF和ICSI以及新鲜或冻融胚胎移植之间的关系。
主要观察指标
死胎风险。
结果
自发受孕和IVF / ICSI低风险足月妊娠的死胎数分别为525（0.1％）和35（0.3％）。在多变量分析中，IVF / ICSI后妊娠死胎的风险增加（优势比（95％CI）：2.1（1.4; 3.1））。事件分析时，死胎的风险相应增加，从妊娠第37周开始考虑每个胎儿的风险时间（危险比（95％CI）：2.4（1.6; 3.6）。在亚分析中，ICSI后死胎的风险增加（OR 2.2（1.2; 3.1）），IVF后死胎的风险无明显变化（OR 1.7（0.9; 3.1））。
结论
我们发现IVF / ICSI后低风险足月妊娠的死胎风险会有系统地增加。风险是否与治疗或潜在的生育能力有关尚不能明确。结果可能表明在达到足月时需要对这些怀孕产妇进行产科监测。
关键词：生育治疗;IVF;ICSI;死胎
https://doi.org/10.1111/1471-0528.15509
Introduction
Fertility treatment offers unique possibilities for otherwise infertile couples to become parents. While the procedures are generally considered safe for the woman, there is substantial evidence of higher risk of pregnancy complications. Compared to spontaneous conception, in vitro fertilization (IVF) has been associated with increased risk of preeclampsia, intrauterine growth restriction, preterm birth, and low birth weight. A few studies have suggested an overall higher risk of stillbirth compared to spontaneously conceived pregnancies, but recently a multinational cohort study showed comparable risk of stillbirth after gestational week 28+0.
前言
生育辅助治疗为不孕夫妇成为父母提供了独特的可能性。虽然通常认为这些手术对女性是安全的，但有大量证据表明妊娠并发症的风险相较升高。与自发受孕相比，体外受精（IVF）与先兆子痫、胎儿宫内生长受限、早产和低出生体重的风险增加有关。一些研究表明，与自发怀孕相比，IVF后死胎的总体风险更高，但最近一项跨国队列研究显示妊娠28 + 0周后IVF与自发受孕后的死胎风险相当。
Due to the risk of pregnancy complications, doctors may be prone to induce labour at an earlier gestation than in spontaneously conceived pregnancies. Moreover, the presence of obstetrical risk factors triggering induction of labour is more frequent in pregnancies following IVF (i.e. maternal age above 40 years). Correspondingly, IVF has been associated with increased risk of induction of labour. When evaluating the risk of stillbirth at a given gestational age associated with IVF, failure to account for induction of labour, or reason hereof, may lead to underestimation of the true association, since it may give rise to comparison of pregnancies following spontaneous conception with the remaining on-going uncomplicated IVF pregnancies. Thus, for the uncomplicated majority of IVF pregnancies the risk of stillbirth at term is still controversial.
由于怀孕并发症的风险，医生可能更倾向于在妊娠早期引产，而不是自发怀孕。此外，在IVF（即母亲年龄超过40岁）的妊娠中，诱发分娩的产科风险因素更加常见。相应地可以看出，IVF与引产的风险增加有关。在评估与IVF相关的给定孕龄的死胎风险时，如果不说明引产情况或其原因，可能会导致低估真正的关联因素，因为这可能会影响自发受孕后怀孕的对比评估，让其他人仍然继续进行简单的IVF受孕。因此，对于无并发症的大多数IVF妊娠而言，足月死胎的风险仍存在争议。
With this large nationwide cohort study we aimed to investigate the risk of stillbirth in low-risk IVF pregnancies at term taking obstetrical risk factor and induction of labour into account. These results are important for the future obstetrical management of uncomplicated IVF pregnancies in terms of whether to induce labour or not?
通过这项全国范围的大型队列研究，我们旨在调查低风险IVF妊娠死胎的风险，同时考虑产科危险因素和分娩诱导。这些结果对于未来是否诱导分娩，对于IVF怀孕的产科管理具有重要意义。
Materials and methods
Design and population
We designed a historic cohort study including all births in Denmark from January 1 2003, to December 31 2013. We established the cohort based on data from the Danish Medical Birth Register and the Danish IVF register. By using the mother’s Danish person identification number (CPR-number) we were able to ensure accurate individual-level linkage between national health registers on all pregnancies in Denmark after gestational week 21+6. The study was conducted without patient involvement.
材料和方法
设计和人口
我们设计了一项回顾性的队列研究，涉及2003年1月1日至2013年12月31日在丹麦出生的所有婴儿。我们根据丹麦医疗出生数据库和丹麦IVF登记处的数据建立了该队列。通过使用母亲的丹麦身份证号码（CPR号码），我们能够确保能够联系到所有在丹麦国家健康登记处登记的妊娠21 + 6周的孕妇。该研究没有患者参与进行。
Information on fertility treatment
Exposure information on fertility treatment was obtained from the Danish IVF register. In Denmark, it is mandatory by law for clinicians in private and public clinics to report each initiated treatment cycle independently of pregnancy outcome to the register. Since the start of the register in 1994, information has been collected on specific type of treatment, i.e. IVF, intracytoplasmic sperm injection (ICSI), fresh or frozen cycle, and gamete donation. From 2006 information on intrauterine insemination (IUI) was included in the register. Further, the register collects information on the CPR-number of the woman along with the outcome of the treatment cycle and the CPR-number-number of any live-born child resulting from the treatment. For this study, fertility treatment was divided into three groups: pregnancies following IVF with or without intracytoplasmic sperm injection (ICSI), pregnancies after IUI (from 2006), and pregnancies following spontaneous conception (i.e. all births after gestational week 21+6 not registered in the IVF register).
生育辅助治疗资料
有关生育辅助治疗的资料来自丹麦IVF登记处。在丹麦，法律规定私人和公共诊所的临床医生必须将独立于妊娠结局的每个初始治疗周期报告给登记处。自1994年登记开始以来，已经收集了特定治疗类型的资料，即IVF，卵胞浆内单精子注射（ICSI），新鲜或冷冻周期以及配子捐赠。从2006年起，关于宫腔内人工授精（IUI）的资料被纳入登记处。此外，登记处还收集关于女性的CPR号码信息、治疗周期和由治疗产生的任何活产儿童的CPR-信息。在这项研究中，生育辅助治疗被分为三组：IVF后伴或不伴卵胞浆内单精子注射（ICSI）后妊娠，IUI后妊娠（2006年）和自发受孕后妊娠（即妊娠21 + 6周后未在登记处登记的分娩）。
Information on pregnancy and stillbirth
Information on pregnancy, obstetrical risk factors and stillbirth was obtained from the Danish medical birth register. In this register, all information on the pregnancy, delivery, and the newborn are registered by the attending physicians and midwifes to be used for the electronical report to the Danish health authorities. The health care system in Denmark ensures that all health care cost are free of charge, so beside possibilities for nationwide register-based research, the registration of health care activities are used for reimbursement of the hospital by the Danish government ensuring complete coverage. Using the CPR-number, we cross-linked information as registered in the medical birth register with information in the IVF register to ensure accurate individual-level exposure and outcome information.
怀孕和死胎的资料
关于怀孕，产科危险因素和死胎的资料来自丹麦医疗出生登记处。在该登记处中，所有关于怀孕、分娩和新生儿的信息都由主治医生和助产士登记，用于向丹麦卫生当局提交电子报告。丹麦的医疗保健系统确保所有医疗保健费用免费，因此除了全国范围的注册研究的可能性之外，丹麦政府还使用医疗保健活动的注册来确保医院的报销，确保完全覆盖。使用CPR号码，我们将在医疗出生登记处登记的信息与IVF登记册中的信息进行交叉链接，以确保准确的个人级别暴露和结果信息。
Statistical analyses
We aimed to in investigate the risk of stillbirth in low-risk term pregnancies following IVF in order to contribute evidence on the obstetrical handling of such pregnancies. Hence, we excluded all preterm births, multiple pregnancies, and all pregnancies of women 40 years or older, with a BMI of 35 or more, induced labour (except if the indication was stillbirth), and women having pregnancy complications known to trigger induction of labour or acute caesarean section such as pre-existing or gestational hypertension, preeclampsia, eclampsia, HELLP, pre-existing or gestational diabetes, intrahepatic cholestasis of pregnancy, or immunization.
统计分析
我们的目的是调查体外受精后低风险怀孕期间死胎的风险，以便为产科处理此类怀孕提供证据。因此，我们排除了所有早产、多胎妊娠和体重指数大于35年龄为40岁或以上女性的所有妊娠情况，此外，还有诱导分娩（除非指的是死胎），以及已知可引发妊娠并发症的女性分娩或急性剖腹产，如既往或妊娠期高血压，先兆子痫，子痫，HELLP，既往或妊娠期糖尿病，妊娠期肝内胆汁淤积或免疫接种。
First, using multiple logistic regressions we compared the estimated overall risk of stillbirth in low-risk term pregnancies following IVF/ICSI or IUI, respectively, with that of spontaneously conceived pregnancies. Second, we estimated the risk of stillbirth from a given gestational age (i.e. the risk from 37+0, the risk from 38+0, the risk from 39+0 etc.) while accounting for the time at risk for each fetus. Each pregnancy contributed with time at risk of stillbirth beginning at the day of fulfilling the Danish live-born criteria at gestational age 22+0 weeks and ending on the first coming date of a registered stillbirth, live birth or end of follow-up on 31 December 2014 (one year after the end of the study period). We estimated the risk of stillbirth associated with conception mode using standard Cox regression analyses and to account for correlations between siblings, all analyses were performed using robust standard errors. Further, we estimated the association between stillbirth and subtypes of procedures (IVF and ICSI respectively) as well as type of embryo transfer (fresh or frozen-thawed embryo transfer).
首先，我们使用多重逻辑回归分析，比较了IVF / ICSI或IUI后低风险足月妊娠和自发怀孕的妊娠后死胎的总体风险。其次，我们估计了给定胎龄的死胎风险（即37 + 0的风险，38 + 0的风险，39 + 0的风险等），同时考虑了每个胎儿的风险时间。每次怀孕都有从分娩当天开始的死胎的风险。丹麦活产标准为孕龄22 + 0周，结束于2014年12月31日（研究期结束后一年）登记死胎、活产或结束随访的第一个日期。我们使用标准Cox回归分析估计与受孕模式相关的死胎风险，并考虑兄弟姐妹之间的相关性，所有分析均使用稳健的标准误差进行。此外，我们估计了死胎和亚型程序（分别为IVF和ICSI）以及胚胎移植类型（新鲜或冻融胚胎移植）之间的关联。
In all analyses we adjusted for a priori determined potentially confounding variables including maternal age (continuous), parity (nulliparous /parous), smoking in pregnancy (yes, no), and child sex (male/female). To assess the extent to which unmeasured confounding may have affected the associations we calculated the E-value. The proportional hazards assumption was evaluated by graphical assessment of log-log plots and the assumption was met. In reporting the results we refer to the risks as odds ratio (OR) or hazard ratios (HR) with 95% confidence intervals (CI). P-values less than 0.05 were considered statistically significant. All statistical analyses were based on an a priori specified analysis plan and performed using complete case analyses with Stata/SE 12 .
在所有分析中，我们调整了先验确定的潜在混淆变量，包括母亲年龄（连续），胎次，怀孕期间吸烟（是、否）和儿童性别（男/女）。为了评估未测量的混杂可能影响关联的程度，我们计算了E值。通过对数 - 对数图的图形评价，对比例风险假设进行了评估，并且满足假设。在报告结果时，我们将风险称为比值比（OR）或风险比（HR），具有95％的置信区间（CI）。P＜0.05被认为具有统计学意义。所有统计分析均基于先验指定的分析计划，并使用Stata / SE 进行完整的案例分析。
Results
A total of 662,739 singletons were born in Denmark by 432,938 mothers during the study period from 1 January 2003 to 31 December 2013. We excluded all preterm births (n= 33,259), and all pregnancies of women 40 years or older (n= 19,942), with a BMI of 35 or more (n= 23,482), or women having pre-existing or gestational hypertension, preeclampsia, eclampsia, or HELLP (n= 30,072), pre-existing or gestational diabetes (n=18,908), intrahepatic cholestasis of pregnancy (n=5,388), or immunization (n=2,745). Except for immunization, all these obstetrical risk factors were more prevalent in pregnancies following IUI and IVF/ICSI compared to spontaneously conceived (Table S1). Baseline characteristics for the remaining 548,908 singleton pregnancies are shown in table 1 according to conception methods. To explore the risk of stillbirth in pregnancies with expectant spontaneous delivery, we excluded all induced deliveries unless the induction of labour was performed due to antenatal diagnosed stillbirth. Hence, for the analyses of an association between mode of conception and stillbirth, the study population comprised 425,732 pregnancies. Of these, a total of 10,235 (2.4%) were conceived following IVF/ICSI, a total of 4,521 (1.1%) were conceived following IUI, and the remaining 410,976 (96.5%) were conceived spontaneously.
结果
在2003年1月1日至2013年12月31日的研究期间，共有432,938名母亲生下662,739名独生子女。我们排除了所有早产情况（n = 33,259）和40岁或以上女性的所有妊娠女性（n = 19,942），此外还有 BMI为35或更高（n = 23,482）的女性，既往或妊娠期高血压，先兆子痫，子痫或HELLP（n = 30,072），既往或妊娠期糖尿病（n = 18,908），肝内胆汁淤积症的女性妊娠（n = 5,388），或免疫（n = 2,745）。除免疫外，所有这些产科风险因素在IUI和IVF / ICSI后与自发受孕者相比更为普遍（表S1）。其余548,908例单胎妊娠的基线特征按受孕方法见表1。为了探究预期自发分娩妊娠死胎的风险，我们排除了所有诱导分娩，除非由于产前诊断为死胎而进行分娩。因此，为了分析受孕模式与死胎之间的关联，研究人群包括425,732名孕妇。其中，IVF / ICSI后共有10,235例（2.4％）受孕，IUI后共有4,521例（1.1％）受孕，其余410,976例（96.5％）自发受孕。
In the study period a total of 2,180 stillbirths (0.3%) occurred. When only considering the included low-risk term pregnancies, the crude number of stillbirths was reduced to 572 (0.1%) of whom 525(0.1%) were born following spontaneous conception, 35 (0.3%) were born following IVF/ICSI, and 12 (0.3%) were born following IUI. The median gestational age at stillbirth was 277 days (25/75 percentile: 269/284 days) with no statistical difference between the conception groups.
在研究期间，共发生了2,180例死胎（0.3％）。当仅考虑所包括的低风险足月妊娠时，死胎的原始数量减少到572（0.1％），其中525（0.1％）在自发受孕后出生，35（0.3％）在IVF / ICSI后出生，12个（0.3％）在IUI之后出生。死胎的中位孕龄为277天（25/75百分位数：269/284天），受孕组之间无统计学差异。
Risk of stillbirth associated with conception mode
The overall risk of stillbirth is presented in table 2. Compared to spontaneously conceived low-risk term pregnancies, the risk of stillbirth in pregnancies conceived following IVF/ICSI was increased (OR 2.1 (1.4; 3.1)). Similarly, the risk of stillbirth in pregnancies following IUI was comparable increased although just short of statistical significance (OR 1.8 (1.0; 3.5)).
与受孕模式相关的死胎风险
表2列出了死胎的总体风险。与自发怀孕的低风险足月妊娠相比，IVF / ICSI后怀孕期间死胎的风险增加（或2.1（1.4; 3.1））。同样，IUI后妊娠死胎的风险也有增加，但差异不大（OR 1.8（1.0; 3.5））。
Using gestational age specific cox regression, we analysed the risk of stillbirth from a given gestational age and onwards (table 2). Compared to spontaneously conceived pregnancies, IVF/ICSI was associated with an increased hazard of stillbirth from gestational week 37+0 (HR 2.4 (1.6; 3.6)), from gestational week 38+0 (HR 2.3 (1.5; 3.6)), from gestational week 39+0 (HR 2.5 (1.5; 4.1)), from gestational week 40+0 (HR 3.0 (1.7; 5.2), from gestational week 41+0 (HR 2.4 (0.9; 5.9), and from gestational week 42+0 (HR 6.8 (1.3; 37). For pregnancies following IUI, the hazard ratios were also increased, although slightly less and without reaching statistical significance except for pregnancies beyond gestational week 41+0 (HR 3.7 (1.8; 11.6)).
使用孕龄特定的cox回归分析，我们分析了给定孕龄及以后死胎的风险（表2）。与自发怀孕相比，IVF / ICSI后死胎风险增加，其风险随孕周依次为妊娠37 + 0（HR 2.4（1.6; 3.6）），妊娠第38 + 0周（HR 2.3（1.5; 3.6）），妊娠周39 + 0（HR 2.5（1.5; 4.1）），妊娠周40 + 0（HR 3.0（1.7; 5.2），妊娠周41 + 0（HR 2.4（0.9; 5.9），妊娠周42 + 0（HR 6.8（1.3; 37）。对于IUI后妊娠，风险比率也有所增加，除了怀孕41 + 0周（HR 3.7（1.8; 11.6））外，风险略低，没有达到统计学意义。
From the hazard ratio of stillbirth from gestational week 37+0 (HR 2.4 (1.6; 3.6)), an E-value of 4.2 was obtained indicating that the observed risk ratio of 2.4 could potentially be explained by an unmeasured confounder that was associated with both the treatment and the outcome by a risk ratio of 4.2 each, above and beyond the measured confounders, but weaker confounding could not do so.
根据妊娠37 + 0（HR 2.4（1.6; 3.6））的死胎风险比，得到的E值为4.2，表明观察到的2.4的风险比的可能的解释为由与之相关的未测量混杂因素。治疗和结果的风险比分别为4.2，超出测量的混杂因素，排除较弱混淆因素的可能。
Hazards associated with IVF, ICSI, and with fresh or frozen-thawed embryo transfer
Based on the information in the IVF register, the group of IVF/ICSI consisted of 4,858 pregnancies conceived following standard IVF, and 4,661 pregnancies conceived after IVF with ICSI. The remaining 716 pregnancies were conceived following IVF/ICSI without specification (frozen embryo transfer or egg donation with no information on the preceding IVF or ICSI). Compared with spontaneously conceived pregnancies, the hazard of stillbirth was increased for pregnancies following ICSI, but not for pregnancies following IVF. The overall odds ratio of stillbirth was 2.2 (1.2; 3.1) and 1.7 (0.9; 3.1) for pregnancies following ICSI and IVF, respectively. Similarly, the gestational age-specific risk was increased for pregnancies following ICSI, but not for after IVF (Table 3).
与IVF，ICSI以及新鲜或冻融胚胎移植相关的危险
根据IVF登记处的信息，IVF / ICSI组包括按照标准IVF设想的4,858例妊娠案例，以及IVIF与ICSI后怀孕的4,661例妊娠案例。其余的716例妊娠是在没有规范的IVF / ICSI之后怀孕的（冷冻胚胎移植或卵子捐赠，没有关于前面的IVF或ICSI的信息）。与自发妊娠相比，ICSI后妊娠的死胎风险增加，但IVF后妊娠没有增加。ICSI和IVF后死胎的总体优势比分别为2.2（1.2; 3.1）和1.7（0.9; 3.1）。同样，ICSI后怀孕时的孕龄特异性风险增加，但IVF后则没有增加（表3）。
Among the 10,235 pregnancies conceived following IVF/ICSI, a total of 874 treatments were performed as frozen-thawed embryos, 6,027 were categorized as fresh IVF/ICSI treatment, and for the remaining 3,334 IVF/ICSI pregnancies, no information on fresh or frozen-thawed treatment was available, and thus these were excluded for this analysis. Compared with spontaneously conceived pregnancies, the hazard of stillbirth was increased for pregnancies following fresh IVF/ICSI treatment (OR 2.1 (1.2; 3.5)), but not for pregnancies following frozen-thawed embryo transfer (OR 1.0 (0.2; 6.2)). The number of stillbirths was too low to explore the gestational age-specific risk of stillbirths (Table 3).
在IVF / ICSI后怀孕的10,235例妊娠中，共有874例采用冻融胚胎，6,027例采用新鲜IVF / ICSI处理，其余3,334例IVF / ICSI妊娠，无新鲜或冷冻胚胎使用的信息，因此这些分析被排除在外。与自发怀孕的妊娠相比，新鲜IVF / ICSI治疗后妊娠死亡的危险性增加（OR 2.1（1.2; 3.5）），但冻融胚胎移植后的妊娠没有增加（OR 1.0（0.2; 6.2））。由于死胎数量少，无法探讨特定孕龄的死胎风险（表3）。
Discussion
Main Findings
In this large cohort study of low-risk term pregnancies conceived after IVF, ICSI, and IUI, we found an increased risk of stillbirth in pregnancies following IVF/ICSI compared to spontaneously conceived pregnancies. When considering the specific type of treatment, the increased risk of stillbirth was related to ICSI, but not IVF, and increased in pregnancies following fresh-embryo cycles, but not frozen-embryo transfer cycles.
讨论
主要发现
在这项针对IVF，ICSI和IUI后怀孕的低风险足月妊娠的大型队列研究中，我们发现IVF / ICSI后妊娠死胎的风险与自发怀孕的怀孕相比有所增加。当考虑特定类型的治疗时，死胎风险的增加与ICSI有关，但与IVF无关，并且在新鲜胚胎周期后妊娠的死胎风险增加，而与冷冻胚胎移植周期无关。
The increased risk of stillbirth following after ART could potentially be secondary to ovarian stimulation, the invasive procedure of ICSI where a single sperm is introduced into the oocyte by micro insemination or because men with low sperm quality are more likely to have chromosomal abnormalities that they may pass on to the offspring.
ICSI后死胎风险增加可能继发于卵巢刺激后ICSI的侵入过程，其中单个精子通过微量授精被引入卵母细胞，或者因为精子质量低的男性更容易出现染色体异常，他们会将这些特性传给后代。
Strengths and Limitations
The main strength of the present study includes a large sample size of an unselected cohort of all pregnancies that during the study period lead to the birth of a singleton after gestational week 21+6. Using the unique Danish CPR-number we were able to crosslink the information in different Danish national health registers, and to control for a wide range of potential confounding factors in the multivariable analyses. Additionally, this comprehensive information made it possible to perform subgroup analyses on the risk following fresh- and frozen-embryo transfer cycles. Yet, residual confounding cannot be excluded. However, based on the calculated E-value 4.2 it is unlikely that residual confounding could explain the findings of this study.
优点和局限性
本研究的主要优势是样本量大，便于选择出适用于研究的孕周21+6的单胎妊娠。使用独特的丹麦CPR编号，我们能够在不同的丹麦国家健康登记册中交换信息，并控制多变量分析中的各种潜在混杂因素。此外，这一全面的信息使得有可能对新鲜和冷冻胚胎移植周期后的风险进行亚组分析。当然这不能排除残余混杂。然而，根据计算出的E值4.2，残留混杂不太可能影响本研究的结果。
We aimed to assess the risk of stillbirth in low-risk pregnancies following IVF/ICSI. Therefore, the results cannot be generalized to all pregnancies following IVF/ICSI. If any pre-specified obstetrical risk factors or pregnancy complication was present, this pregnancy was excluded from the present study. Since the proportions of obstetrical risk factors were more prevalent among pregnancies following IUI and IVF/ICSI, the exclusion of at-risk pregnancies may have affected the results of this study, most likely with an attenuation of a higher risk of stillbirth among pregnancies following IVF/ICSI.
我们的目的是评估IVF / ICSI后低风险妊娠死胎的风险。因此，结果不能推广到IVF / ICSI后的所有妊娠。如果存在任何预先确定的产科危险因素或妊娠并发症，则将排除本研究之外。由于IUI和IVF / ICSI后产科危险因素的比例更为普遍，因此排除有风险的妊娠可能会影响本研究的结果，最有可能减少IVF / ICSI后妊娠死胎的风险。
The analyses were conditioned on gestational age since we only included term pregnancies. Preterm delivery is associated with the risk of stillbirth and occurs more frequently following IVF/ICSI. Hence, gestational age may be considered on the causal pathway between IVF/ICSI and stillbirth, which could have led to bias in this study, most likely towards the null.
因为我们只研究足月妊娠，因此分析以孕龄为条件。早产与死胎风险有关，并且在IVF / ICSI后更常发生。因此，IVF / ICSI和死胎之间的因果关系可以考虑胎龄，这可能导致本研究产生偏差，最有可能导致研究无效。
Most studies investigating the risk of stillbirth following ART were relatively small and often with stillbirth as a secondary outcome. Fortunately, stillbirth is a very rare event, which may cause insufficient statistical power when studying this outcome. Assuming a background risk of stillbirth of 0.1%, the sample sizes in the exposed and reference group should be at least 2,500 to test the hypothesis of a doubling of the risk with a power of 80% and a significance level of 0.05. Hence, most studies reporting stillbirth as an outcome following IVF have been underpowered.
调查ART后死胎风险的大多数研究相对较少，并且往往以死胎作为次要结果。幸运的是，死胎是一种非常罕见的事件，在研究这一结果时可能会导致统计能力不足。假设死胎的背景风险为0.1％，暴露组和参照组的样本量应至少为2,500，以检验风险加倍的假设，其功效为80％，显着性水平为0.05。因此，大多数报告死胎作为IVF后的结果的研究都缺乏动力。
Additionally, besides multiplicity, previous studies have only sparsely accounted for the increased prevalence of obstetrical risk factors among women undergoing ART, and no study has taken into account the management of these pregnancies with a considerable increased incidence of induced labour. Independent of risk factors, induction of labour in IVF pregnancies at no later than 41 weeks of gestation has been standard in Denmark until recently. Further, obstetrical risks such as maternal age above 40 years, pre-existing medical conditions, and preeclampsia often leading to induction of labour are more frequent in pregnancies following IVF. Therefore, any risk of bias could be reduced if obstetrical risk factors and induction of labour are taken into account, as in the case of the present study.
此外，除了多重性之外，之前的研究很少考虑到ART的妇女产科风险因素的增加，并且没有研究考虑到这些妊娠的管理，诱导分娩的发生率也有显着增加。在不受风险因素影响的情况下，不晚于妊娠41周的IVF妊娠引产一直是标准做法。此外，在IVF后怀孕期间，如果产妇年龄超过40岁，已存在的医疗条件和先兆子痫导致引产的产科风险将更为频繁。因此，如果将产科风险因素和分娩诱导因素考虑在内，如本研究的情况一样，任何偏倚风险都可以降低。
Interpretation
In line with our results, Wisborg and colleagues made comparable conclusions while evaluating the overall risk of stillbirth from gestational age 22 weeks among Danish singleton pregnancies in fertile and subfertile parents and parents conceiving following IVF/ICSI and non-ART treatments. Compared to pregnancies achieved spontaneously within 12 months, pregnancies following IVF/ICSI had a higher risk of stillbirth (OR 4.08 (2.11–7.93)), while pregnancies following non-ART had comparable risk of stillbirth (OR 0.53 (0.13–2.18)). Similarly, a large Australian study found an increased risk of stillbirth following any assisted conception treatment (OR 1.82 (1.34,2.48)), IVF with fresh-embryo cycles 2.35 (1.34,4.11), and ICSI with fresh-embryo cycles (OR 2.46 (1.29–4.68)), but not following less invasive fertility treatment (ovulation induction OR 0.52 (0.07,4.18)). Further supporting the results of present study, the authors showed that the risk of stillbirth was more pronounced following fresh-embryo cycles than following frozen-embryo cycles. Moreover, results from a Belgian matched case-control study on 6,096 singleton pregnancies demonstrated that IVF/ICSI was associated with a higher risk of stillbirth compared to spontaneously conceived pregnancies (OR 2.51 (1.24-5.20)). However, none of the above-mentioned studies conducted analyses stratified by gestational age.
解释
根据我们的结果，Wisborg及其同事在评估丹麦可育和不孕父母以及在IVF / ICSI和非ART的父母的单胎妊娠中孕22周死胎的总体风险时，得出了类似结论。与12个月内自发妊娠相比，IVF / ICSI后怀孕的死胎风险更高（OR 4.08（2.11-7.93）），而非ART后怀孕的死胎风险较低（OR 0.53（0.13-2.18））。同样，澳大利亚的一项大型研究发现，任何辅助受孕治疗后的死胎风险都会增加（OR 1.82（1.34,2.48）），IVF新生胚胎周期为2.35（1.34,4.11），ICSI新鲜胚胎周期为（OR 2.46（1.29-4.68）），但无创生育治疗（排卵诱导OR 0.52（0.07,4.18））并没有增加风险。为了进一步支持本研究的结果，作者表明，采用新鲜胚胎周期后死胎的风险比采用冷冻胚胎周期更明显。此外，比利时一项病例对照研究对6,096例单胎妊娠的结果表明，与自发妊娠相比，IVF / ICSI与死胎风险相关（OR 2.51（1.24-5.20））。然而，上述研究均未进行按胎龄分层的分析。
In contrast, the largest study to date on the risk of stillbirth showed comparable risk beyond gestational week 28 when comparing pregnancies following ART with spontaneously conceived pregnancies. In this well-conducted multinational Nordic cohort study, Henningsen and colleagues compared the risk of stillbirth between 62,485 singleton pregnancies following ART with that of 362,798 spontaneously conceived controls. Although the definition of stillbirth varied among the participating countries, they were able to appropriately evaluate the gestational-age-specific risk of stillbirth using “fetuses-at-risk” estimates. Unlike the present study, they did not find an increased risk of stillbirth among ART singletons in the third trimester. This discrepancy may partly be explained by differences in methodological approach. As opposed to our study, the Nordic collaborative study did not account for difference in maternal smoking or body mass index (BMI). As potentially modifiable risk factors, maternal obesity and smoking in pregnancy has together with fetal growth restriction been estimated to account for more than half of all stillbirths. Hence, this may have caused a considerable underestimation in the Nordic study. Additionally, the study did not account for variations in obstetrical management. Compared to spontaneously conceived pregnancies, labour following ART pregnancies is more frequently induced, which again may be explained by increased prevalence of obstetrical risk factors or complications, or as a consequence of anxiety or maternal request. However, if unaccounted for it may introduce survivorship bias and underestimation of the true risk of stillbirth, since it may lead to comparison of pregnancies following spontaneous conception with the remaining on-going uncomplicated IVF pregnancies.
相比之下，迄今为止关于死胎风险的最大规模研究表明，在妊娠28周后将ART后妊娠与自发妊娠相比，死胎风险增加。在这项多国队列研究中，Henningsen及其同事将ART后62,485例单胎妊娠与362,798例自发受孕的死胎风险进行了比较。尽管参与国的死胎定义各不相同，但他们能够使用“有风险的胎儿”估计值来适当评估特定孕龄的死胎风险。与目前的研究不同，他们没有发现妊娠晚期ART妊娠后死胎的风险增加。这种差异可能由各国家研究方法的差异来解释。与我们的研究相反，北欧合作研究没有考虑到母亲吸烟或体重指数（BMI）的差异。作为潜在可改变的风险因素，孕妇肥胖和妊娠期吸烟与胎儿生长受限一起估计占所有死胎的一半以上。因此，这可能导致北欧研究中对风险相当大的低估。此外，该研究未考虑产科管理的差异。与自发怀孕相比，ART妊娠后的分娩更常被诱导，这也可能是由于产科风险因素或并发症的流行率增加，或焦虑或母亲要求的结果。但是，如果不明原因，可能会引入生存偏差并低估了死胎的真实风险，因为这可能导致自发受孕后的妊娠与其他不复杂的IVF妊娠进行比较。
While the study by Henningsen and colleagues aimed to evaluate the risk in different categories of gestational age, the aim of this study was to estimate the risk from a given gestational age until birth using time-to-event analyses, which may mimic the everyday clinical setting in a slightly more applicable way. When advising parents to-be at a specific gestational age, the cumulative hazard of stillbirth may be easier to communicate since the exact time of birth is hard to predict no matter whether the physician chooses to advise on induction or await spontaneous onset of labour.
虽然Henningsen及其同事的研究旨在评估不同类别孕龄的风险，但本研究的目的是使用时间 - 事件分析估计从给定孕龄到出生的风险，这可能模仿日常临床更加适用。当父母处于特定的孕龄时，采用死胎的累积风险数值作为证据来沟通可能更容易，因为无论医生是建议诱导分娩还是等待自发性分娩，都难以预测出生的确切时间。
Similar to the Nordic collaborative study, a few other studies have concluded that the risk of stillbirth among ART pregnancies is comparable to that of spontaneously conceived pregnancies. A Norwegian study including 998 pregnancies reported no significant difference in the number of stillbirths with an incidence of 19.7‰ in a study group of pregnancies following various procedures of assisted reproduction and 12.4‰ in the control group matched for age and parity. Similarly, a Chinese study found no significant association between IVF and stillbirth among 870 IVF pregnancies and 3,433 spontaneously conceived matched for maternal age and parity (OR 1.96 (0.84–4.57). Nevertheless, the risk estimates in these studies were quite similar to those in the present study. Since these studies were rather small with less than 1000 exposed pregnancies, there is a considerable risk of a type II statistical error with failure to reach statistical significance due to low number of participants.
与北欧合作研究类似，其他一些研究也得出结论，ART妊娠死胎的风险与自发怀孕的风险相当。一项包括998例妊娠案例的挪威研究报告，在各种辅助生殖手术后，怀孕研究组的死胎率没有显着差异，发生率为19.7‰，而对照组中12.4‰与年龄和胎次相匹配。同样，一项中国的研究发现，在870例IVF妊娠中，IVF与死胎之间没有显着相关性，3,433例自发受孕与母亲年龄和胎次相匹配（OR 1.96（0.84-4.57））。然而，这些研究中的风险评估与本研究中的风险评估非常相似。由于这些研究规模相当小，暴露的妊娠少于1000例，因此存在相当大的II型统计误差的风险，无法达到统计学意义。
Although the majority of existing literature report of an increased incidence of stillbirth following ART. , it remains uncertain whether the increased risk is related to a causal association with the IVF procedures or with factors related to the underlying subfertility. In a Norwegian study, Romundstad and colleagues found an increased risk of perinatal mortality following assisted conceptions, but this association diminished when analyses were repeated in sibling-pair-analyses among siblings born to women who had conceived both spontaneously and after assisted fertilization. Hence, the differences could not be attributed to the reproductive technology but most likely to factors related to the underlying subfertility. Additionally, the authors noted that women who had had a perinatal death in a spontaneously conceived pregnancy were three times more likely to seek fertility treatment afterwards than those who had not.
虽然现有的大多数文献报道了ART后死胎的发生率增加，但仍然不确定风险增加是否与体外受精（IVF）程序的因果关系或与潜在的低生育力相关的因素有关。在挪威的一项研究中，Romundstad及其同事发现辅助受孕会增加围产期死亡的风险，但这种关联在兄弟姐妹间重复分析中会减弱。因此，差异不能归因于生殖技术，最有可能是与潜在的低生育力相关的因素。此外，作者还指出，在自发怀孕期间曾发生围产期死亡的女性，其后接受生育辅助治疗的可能性是未接受生育治疗的女性的三倍。
Only a few other studies have investigated the risk of stillbirth following fresh- and frozen-embryo transfer cycles. Similar t our results, these studies found any risk of stillbirth related to IVF to be lower and comparable to that of spontaneously conceived among the subgroup conceiving after frozen-embryo cycles. This is in line with several other studies finding significantly lower risk of other perinatal complication following frozen-embryo cycles than following fresh-embryo cycles1. Nevertheless, the results of the subgroup analyses in this study showing a reduced risk of stillbirth following frozen-embryo transfers must be interpreted with caution given the lower numbers of participants.
只有少数其他研究调查了采用新鲜和冷冻胚胎移植周期后死胎的风险。与我们的结果相似，这些研究发现，与冷冻胚胎周期后受孕的亚组5, 22相比，任何与IVF相关的死胎风险都更低，并且与自发受孕的风险相当。这与其他几项研究一致，这些研究发现冷冻胚胎周期后其他围产期并发症的风险显着低于新鲜胚胎周期。尽管如此，本研究中亚组分析的结果显示，冷冻胚胎移植后死胎风险降低，这个结果必须谨慎解释，因为研究参与者数量较少。
Conclusion
In this large national register-based cohort study we found anincreased risk of stillbirth following IVF/ICSI. While the study was designed to assess the risk of stillbirth in low-risk IVF pregnancies at term, the results must be interpreted with caution for the whole population of IVF pregnancies. Whether the risk may be related to the treatment itself or the underlying subfertility of women seeking IVF treatment is uncertain. Nevertheless, the results may indicate a need for closer obstetrical surveillance for these pregnancies when reaching term.
结论
在这项基于国家登记的大型队列研究中，我们发现体外受精/ICSI后死胎风险增加。虽然这项研究的目的是评估低风险体外受精妊娠足月死胎的风险，但对整个体外受精妊娠人群的解释结果必须谨慎。这种风险是否可能与治疗本身或寻求体外受精治疗的妇女潜在的低生育能力有关还不确定。然而，研究结果可能表明，在这些妊娠达到足月时需要更密切的产科监测。
附件：
表1.体外受精/卵胞浆内单精子注射（IVF / ICSI），宫内人工授精（IUI）和自发受孕（SC）后无并发症1 妊娠妇女的基线特征，丹麦2003-2013
表2.体外受精/卵胞浆内单精子注射（IVF / ICSI），宫腔内人工授精（IUI）和自发受孕后无并发症足月死胎的比值比（OR）或危险比（HR）和95％置信区间（CI） （SC）非诱导的阴道分娩，丹麦2003-2013
表3.体外受精（IVF），卵胞浆内单精子注射（ICSI）（包括新鲜或冻融循环）后无并发症足月妊娠和非诱导性阴道分娩的死胎率的危险比（HR）和95％置信区间（CI）和自发受孕（SC）非诱导的阴道分娩，丹麦2003-2013
联系客服
