Psychological Impact in Early Pregnancy Loss: The Effects of Disclosing the Causative Chromosomal Anomaly

Early pregnancy loss (EPL) is defined as a non-viable intrauterine pregnancy in the first trimester with either an empty gestational sac or a gestational sac containing an embryo or a fetus without fetal heart activity. This is the most common form of pregnancy loss, occurring in 15–20% of clinical pregnancies [1]. It is clearly associated with maternal age, increasing from 10% of pregnancies at 20 years to 75% at 40 years [2] due to an increase in chromosomally abnormal pregnancies with maternal age [3]. Our group has shown that 70% of first-trimester miscarriages are associated with chromosomal abnormalities, with autosomal trisomies being the most common type (65%) [4].

When an EPL occurs, women often blame themselves because they feel guilty and experience sadness, anger, and bewilderment [5, 6]. They have been seen to suffer from episodes of anxiety, depression, post-traumatic stress, and an increased frequency of ruminative responses [7]. The effects of psychological distress can be highly shocking and can last up to 9 months. For this reason, women who have suffered an early loss are at higher risk of psychological morbidity [8, 9]. Despite EPL being a common event, social support is often lacking, leaving the couple or woman to cope in isolation since their acquaintances are usually unaware of the pregnancy. Nowadays, it is well established that there is a relationship between low social support and an increased risk of mental health problems, such as depression and anxiety, during pregnancy [10]. Currently, the European Society of Human Reproduction and Embryology and the Royal College of Obstetricians and Gynaecologists endorse the genetic study of the embryo in recurrent loss [11, 12].

In the present project, we hypothesize that the implementation of genetic diagnostic techniques and the accompaniment of genetic counselors allow women to understand the causes of loss and the recurrence risk, therefore minimizing the psychological impact and that this effect is already present as of the first loss [13, 14]. The main objective of the present study was to assess the evolution of anxiety, depression, post-traumatic stress, and rumination following an EPL and determine whether identifying the genetic cause of the EPL could facilitate subsequent psychological adaptation.

This prospective cohort study was conducted at the Hospital Clinic Barcelona, a tertiary referral hospital in the city of Barcelona, although the vast majority of EPL cases occurred in the community of West Barcelona. During 21 months (October 2020–June 2022), women with a diagnosis of clinical EPL, defined as any pregnancy loss occurring between 5 weeks (pre-embryonic loss) and 13 weeks (the upper limit for transcervical chorionic villi sampling (CVS), established by ultrasound measurements), were asked to participate in the study at the time that they were offered CVS to reveal the cause of the loss employing a quantitative fluorescent polymerase chain reaction (QF-PCR) and a long-term karyotype. Women who consented to participate were asked to fill out two online surveys on the psychological impact of the EPL at two specific time periods. The baseline survey was filled out 1 week after the diagnosis of EPL, and the second, 1 month later (1 week after the genetic counseling session scheduled at about 4 weeks after EPL, in which the embryo genetic results were communicated) (Fig. 1).

Although QF-PCR and a long-term karyotype were intended in all cases, when villi culture failed, a chromosomal microarray analysis was performed on the remaining DNA after QF-PCR. In addition, parental karyotypes were requested in fetal trisomies 21 and 13 when villi culture failed or when a rearrangement was found in the fetal karyotype.

Surveys were confidential, and although participants consented to share their e-mail exclusively to assign the correct follow-up to each patient, the e-mails were deleted from the remaining database. The exclusion criterion was not having a proficient level of Spanish or Catalan. This study was approved by the Ethics Committee of the Hospital Clinic de Barcelona (HCB/2020/0128).

Respondents to the baseline survey were asked to provide demographic information, including their age, marital status, education level, religious beliefs, obstetric history, including if they already have children, previous EPL, the time taken to get pregnant at the index pregnancy, and use of assisted reproductive technology.

The psychological impact of EPL was assessed in the first and second surveys using four online questionnaires for anxiety, depression, and post-traumatic stress disorder (PTSD) to cover the commonest mental health symptoms already described in the literature on pregnancy loss [9], together with a scale on rumination. These psychometric scores were also used to reflect the impact of the genetic counseling intervention. No questionnaire item was removed or altered during translation. Furthermore, there was an open-ended question in the survey that inquired about the content of the respondents’ ruminative thoughts.

The Hospital Anxiety and Depression Scale (HADS) is an extensively used and well-validated questionnaire including seven items measuring depression and seven items measuring anxiety with a good internal consistency (Cronbach alpha for both scales >0.80) [15]. Scores for each subscale range from 0 to 21 and a cut-off of 11 for the anxiety scale was chosen as the clinical score, following previous studies [9].

The Beck Depression Inventory (BDI-II) is a 21-item revised questionnaire extensively used with good internal consistency and reliability ranging from 0.88 to 0.94 [16‒18]. Each item has four response options according to the intensity of the different symptoms. For the present study and based on the use of the BDI-II in the postpartum population [19], eight items related to the cognitive aspects of depression were selected. The eight-item scores ranged from 0 to 24 and 12 was selected as the clinical score cut-off.

The Spanish version of the Impact of the Event Scale-Revised (IES-R) [20] is a 22-item scale reflecting psychological distress after a traumatic event, with good internal consistency (Cronbach alpha = 0.96) [21, 22]. This scale comprises three distress dimensions according to DSM-IV PTSD diagnoses, such as intrusion, hyperarousal, and avoidance. A cut-off of 33 was considered to indicate a clinical score.

Developed as an extension of the original RRS by Nolen-Hoeksema (1991), the Ruminative Responses Scale (RRS-10) comprises 10 items that capture the frequency and intensity of rumination. Rumination is defined as passively and repetitively focusing on one’s symptoms of depression and the possible causes and consequences of these symptoms. The scale has two subscales (Brooding and Reflection) [23]. With an initial statement stating, “what you generally do, not what you think you should do when feeling down, sad, or depressed,” respondents rated each item on a four-point scale (1 = “never” to 4 = “always”). Thus, final scores range between 10 and 40, with higher scores reflecting greater tendencies toward ruminative response styles. The authors considered a score of 28 or above [24] as a threshold to indicate a high level of rumination.

The score changes on the four scales (anxiety, depression, post-traumatic stress, and rumination) were calculated as the absolute difference (delta-values) between the score obtained on the first survey 1 week after CVS was performed (when the EPL was diagnosed), and the second survey 1 month after CVS (when the genetic results were delivered to women). A clinical score for anxiety, depression, post-traumatic stress, and high-frequency ruminative responses was defined when score values exceeded the aforementioned cut-offs. Changes in the proportion of clinical scores in anxiety, depression, post-traumatic stress, and the frequency of ruminative responses between both surveys were calculated.

In calculating the sample size, we acknowledged the limited existing data on psychological interventions in EPL. However, our assumptions were informed by findings from Nikcevic et al. [25]. Their study demonstrated that women offered a post-miscarriage follow-up consultation, allowing them to express their emotions, experienced a significant reduction in psychological symptoms – nearly a 50% decrease in clinical anxiety and depression – compared with those who did not receive such comprehensive care. Drawing an analogy, we hypothesized that providing CVS results alongside counseling to explain the cause of miscarriage could yield a similar psychological benefit, alleviating feelings of guilt or uncertainty. Based on this, we assumed a 50% reduction in the proportion of women experiencing psychological distress between the first and second surveys. To calculate the required sample size, we considered a repeated-measures design involving paired data. Using a significance level (alpha) of 0.05 and a power of 0.80, we estimated that a minimum of 77 participants would be necessary to detect clinically meaningful differences in psychological outcomes.

The prospective cohort of participants were divided into three groups according to the chromosomal anomaly found in karyotyping (Fig. 1): group A1 included cases with an autosomal trisomy, group A2 included other chromosomal anomalies also explaining the EPL, and finally, group B included cases in which (a) genetic diagnosis of the EPL was not provided due to a normal karyotype (46,XX or 46,XY) being found, (b) a failed result, or (c) because women declined genetic testing. When fetal QF-PCR showed a normal female karyotype (46,XX), the QF-PCR result on maternal saliva was used to rule out maternal cell contamination. Changes between both surveys in the four studied scales were also assessed after stratification per study group, both for score changes (delta-values) and for proportional changes in the clinical scores obtained by the participants.

Continuous variables are expressed as means with standard deviation (SD) and t tests were performed for their comparison, except for non-normally distributed paired samples in which the Wilcoxon test was required. When more than two groups were compared, either ANOVA or the Kruskal-Wallis test was performed according to whether a normal distribution was confirmed. The distribution of the delta-values of the scales (HADS, BDI-II, IES-R, and RRS-10) was assessed for normality and treated statistically as non-normally distributed.

Qualitative variables were compared using the chi-square test and expressed as numbers (n) and proportions (%). The McNemar test was performed when analyzing dichotomous paired samples. We used SPSS version 20 (IBM SPSS Statistics 20) for all the statistical analysis. The level of statistical significance was established at α = 0.05.

A multivariate regression analysis of the median score decline (delta-values) was performed across the three karyotypic groups, adjusted for potential confounders, such as maternal age, previous EPL, and having previous children. Maternal age was divided into two groups (<35 years and ≥35 years), previous EPL into three groups (no previous EPL, 1–2 previous EPL, ≥3 previous EPL) with the latter serving as the reference group, and previous children were considered depending on their presence or absence.

We considered using the group with ≥3 EPL as the reference group as these women often face a greater emotional burden, including intensified feelings of grief, frustration, and hopelessness, compared with those with <3 EPL. Based on our clinical experience, experiencing three EPLs frequently represents a pivotal threshold for many women when deciding whether to risk another similar experience. Quantile regression directly estimates the distribution by fitting a function to any chosen quantile using linear programming, without distributional assumptions.

In the survey, 127 commentaries were collected from the open-ended question focused on the content of the respondents’ ruminations between both surveys. The answers stating “0” or “No” were counted as empty, thus giving no information. The commentaries were first organized for each participant (first and second surveys) and then analyzed to assess the evolution of their thoughts over time. One of the team investigators (V.A.-R.) employed a phenomenological approach to conduct a thematic content analysis. The process for the analysis involved counting each occurrence of new ideas, which were coded upon their first appearance and marked again if repeated. This systematic approach allowed for identifying and categorizing common themes within the respondents’ ruminative thoughts.

During the study period, 151 women with an EPL were referred to the Prenatal Diagnosis Unit for CVS before uterine evacuation and consented to participate in the study. Among them, 125 (83%) filled out the first online survey, while 87 (70% of those who filled out the first survey) also filled out the second. The mean maternal age was 36.5 years, 69% were 35 years old or older, 94% were married or in a partnership, 28% had a previous child, 46% had a previous EPL, and 41% used Assisted Reproductive Technologies (Table 1).

There were 62 cases in group A1 (an autosomal trisomy explaining the EPL), 17 in group A2 (other chromosomal anomalies explaining the EPL), and 46 in group B (no chromosomal anomaly explaining the EPL). Women and pregnancy characteristics were similar between karyotypic groups (A1, A2, and B), except for having a previous child (27% in group A1 vs. 53% in group A2 vs. 20% in group B), with a similar maternal age (36.6 years) in the three groups.

A significant score decline (delta-value) was observed between the first and second surveys in the four studied psychological scales, from 4.8–32 to 4.0–25 (Fig. 2). A significant decline was also demonstrated in the proportion of women with a clinical score in all measured psychological impacts except for anxiety, which only showed a non-significant trend toward a decline (Fig. 3). The highest decrease was observed in clinical scores for post-traumatic stress, which dropped from 51% in the first survey to 26% in the second, followed by the clinical score for depression, which decreased from 26% to 10% between the two surveys, and ultimately by the elevated levels of ruminative responses, which decreased from 19% to 4% (Fig. 3).

When this analysis was repeated after karyotypic group stratification, no significant differences between the two surveys were found between the three study groups (A1, A2, and B) neither in score delta-values (Fig. 4) nor in the proportion of women with clinical scores (Fig. 5). Interestingly enough, the A2 group showed the highest drop in anxiety, depression, post-traumatic stress, and rumination in absolute values and proportion of clinical scores.

Quantile regression analysis of the medians adjusted by potential confounders showed a significant negative correlation with depression delta-values in group A2, with a coefficient estimate of −5 (SE = 1.22, t = −4.1, df = 80, p < 0.001) (95% CI: −7.4–2.6), indicating a substantial decrease in depression scores in the group of other chromosomal anomalies. Nevertheless, no significant associations were found for anxiety, post-traumatic stress disorder, or rumination (Table 2).

Regarding the open question about ruminations, the reasons hypothesized by the participants for the EPL were diverse and fell under 24 distinct categories (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000543684). Six participants echoed sentiments such as, “I have been trying to find what could have been the cause of the miscarriage in order to be able to understand it” (P54). The most frequently mentioned reason for the EPL was stress (n = 26), with participants expressing feelings like, “If I hadn’t been so stressed at the beginning of August...” (P90). This was followed by concerns about diet (n = 23) and physical activity (n = 22). Additional common ruminative thoughts included work-related stress (n = 19), smoking or alcohol consumption (n = 15), and ruminations about one’s age (n = 15).

Forty-two participants, 26 with a chromosomal anomaly report (groups A1 and A2), and 16 without (group B), made comments in both surveys. Eight participants with a report expressed that knowing the cause of their miscarriage had alleviated their feelings of guilt. The change between the first and second survey is highlighted by the following words (P-25): “They were meaningless but inevitable thoughts. The first thing I thought was that maybe I did something wrong (eating habits, physical activity, etc.) to suffer the loss” and “I have spoken to the geneticist this week and he told me the cause of the loss, so I now know that it was out of my hands to prevent the loss.”

The main findings of this study may be summarized in four items. First, a significant score decline was observed in the four studied psychological scales assessing anxiety, depression, post-traumatic stress, and rumination, between the first survey (filled out 1 week after the diagnosis of EPL) and the second survey (filled out 1 week after the delivery of the karyotype on chorionic villi). Second, a significant decline was also demonstrated in the proportion of women with clinical scores at all psychological scales except for anxiety, from 19% to 51% in the first survey to 4%–30% in the second. Third, when these changes were assessed after stratifying the data by karyotypic group (A1, A2, and B), the group with other chromosomal anomalies (A2) showed the highest drop in anxiety, depression, post-traumatic stress, and rumination. Fourthly, at the multivariate 50th quantile regression, only this group (A2) demonstrated a significant decline in depression scores.

In 2004, a review by Brier concluded that a significant proportion of women experience elevated levels of anxiety after a miscarriage up until about 6 months post-miscarriage and that they are at increased risk of obsessive-compulsive disorder and PTSD, acknowledging the scarce data at that time on this subject [25]. In 2015, a survey administered to men and women aged 18–69 years across the USA showed that respondents erroneously believed that miscarriage was a rare complication of pregnancy, with the majority believing that it occurred in 5% or less of all pregnancies. There were also widespread misconceptions about the causes of miscarriage. Those who had experienced a miscarriage often felt guilty, isolated, and alone [26].

In 2018, Farren et al. [5] reviewed 27 studies on the psychological impact of EPL. Evidence of significant depression (8–20%) and anxiety (18–32%) was found, and also for post-traumatic stress symptoms (25%–39%) in three studies in the first month following miscarriage. Those rates are comparable to ours, 10%, 30%, and 26%, respectively, since all fall within the aforementioned ranges. The review by Farren et al. [5] pointed out that women who had a history of psychiatric illness, less support, multiple miscarriages, and those who had experienced subfertility, and/or were left childless by the loss were found to be at a higher risk of psychological impact.

Subsequently, the same group conducted a prospective study in 737 women with an EPL and 171 controls with a viable pregnancy [9]. The rates of clinical anxiety (24%), depression (11%), and post-traumatic stress (24%) 1 month after EPL were higher than in controls (13% anxiety and 2% depression). When compared with the rates of our series, Farren’s rates were lower for anxiety (30%) and similar for depression (10%) and post-traumatic stress (26%). In the same study, follow-up was carried out 9 months after miscarriage, showing that the proportion of women screening positive decreased over time for all three conditions to about half (17% for anxiety, 5% for depression, and 16% for post-traumatic stress) [9].

Another factor that has been argued to modify the psychological impact is the miscarriage treatment type. A Swedish study compared the psychological response to and recovery after EPL in women randomized to expectant delivery or misoprostol treatment and no differences were found between both groups [27].

Paradoxically, because miscarriage is such a common event, it has not been considered a major health issue, as compared, for instance, to stillbirth. The English organization Tommy claims that the grief and psychological impact of miscarriage are often not comprehended, both by society and healthcare professionals [28]. The loss can often be minimized by the view that a pregnancy before 12 weeks “isn’t a real baby” or that “it just wasn’t meant to be.” The organization has issued the following recommendations for policy and solutions to improve care in miscarriage: (a) miscarriages must be recorded so that the rate of miscarriage can be measured nationally; (b) access to care after sporadic miscarriage must be available with a clear pathway for follow-up mental health support; (c) tests and treatments must be standardized through a “graded approach” to recurrent miscarriage; (d) specific, personalized care pathways should be established for high-risk women; and (e) a clear pathway for preconception support and guidance must be established [29].

The significant score decline observed between the first and second survey in the four studied psychological dimensions showed an improvement in the psychological state measured through the questionnaires in most participants after the disclosure of the cause of the EPL, particularly when maternal age was not involved in the frequency of the specific chromosomal anomaly type. Our group supports the offering of a pregnancy karyotype as one of the strategies that helps the grief process of women since it may reveal the cause of the miscarriage. Without this information, women may blame themselves and feel guilty for the pregnancy loss. Disclosing the cause shifts the perception from self-blame (internal locus) to understanding that the EPL was beyond their control (external locus), thereby offering peace of mind to some individuals who have experienced a miscarriage.

It appears that a normal pregnancy karyotype is not associated with a better prognosis as it has a higher risk of recurrence. It has been pointed out that pregnancy loss at a young maternal age and recurrent pregnancy loss are both indicative of euploid losses, thus suggesting shared underlying pathologic processes between euploid pregnancy loss and cardiovascular disease. Understanding the pathophysiology of euploid pregnancy loss, therefore, presents an avenue to reduce the risk of further pregnancy losses, as well as later-life disease in women [30]. A chromosomal anomaly different from an autosomal trisomy, usually a monosomy X or a triploidy, is the most favorable result because it is not related to an advanced maternal age or increased risk of recurrence. On the contrary, autosomal trisomies in young pregnant women are associated with an excess risk of recurrence above the maternal age risk [3], while in older women, they may trigger self-blaming for having ‘waited so long to get pregnant. We recognize the importance of reinforcing the message that “age matters” within the community as many women today are unaware of the increasing risk of miscarriage associated with advancing maternal age.

Depression, anxiety, and post-traumatic stress are disorders with significant implications for individual care and public health. Prevention would be highly desirable, so the identification of specific risk factors becomes extremely important and is a first step toward accomplishing it. If we understand social support in its broad sense and view genetic counseling as a form of informational support, this addresses one of the existing gaps in supporting women after experiencing an EPL [27]. However, as our study reveals, and some participants have emphasized, improved protocols are particularly necessary for women who have had recurrent EPL.

Studies on rumination are scarce, although it appears that people who frequently ruminate when distressed are more likely to develop depression and are likely to remain depressed for longer [31]. As mentioned before, rumination has been defined as passively and repetitively focusing on one’s thoughts around depression; therefore, it confers a risk of prolonged episodes and future depressive episodes. The tendency to ruminate is trait-like rather than a proper clinical diagnosis and appears to remain fairly stable, even as depression improves. It has also been suggested that rumination was the form most strongly and consistently related to depressive symptoms [31, 32]. In this context, rumination appears to provide a necessary rationale for the EPL [33], particularly because it is often an unexpected outcome of pregnancy. Participants describe various topics they ruminate on, ranging from self-reflection on personal habits to contemplating spiritual justifications for the decision to terminate the pregnancy. Stress, notably, is one of the most frequently expressed concerns. Remarkably, some participants have found that understanding the reason for the EPL has brought them mental peace and alleviated their rumination and distress.

The primary limitation of our study is that only 58% of the women who consented to participate completed both surveys, which makes it difficult to assess the evolution of their mental health over time accurately. We hypothesized that the low response rate for the first survey (83% of the initial 151 participants) may be due to the emotional impact of their recent EPL, which likely made participation challenging for them. Furthermore, the attrition observed in the second survey, with a 70% response rate from the 125 participants who completed the first survey, could be attributed to nonresponse bias. Second, it is crucial to note that our study focused on assessing psychological clinical scores (depression, anxiety, and post-traumatic stress) and elevated levels of rumination rather than providing clinical diagnoses as a clinical psychologist did not make any formal diagnostic evaluations. Third, we have focused on the hypothesis that disclosing chromosomal anomalies may lessen the psychological impact of EPL; however, several other dimensions should also be explored to better understand how this information aids individuals, such as the quality and availability of social support, their level of health literacy, and social determinants related to emotional distress and resilience.

The study has the strength of being the first aiming to assess the effect of disclosing the chromosomal anomalies causative of an EPL. This marks an initial step toward providing more comprehensive care to pregnant individuals, who are particularly vulnerable and in need of specialized attention due to the inherent challenges and risks associated with pregnancy. Another notable strength of the study is its use of a mixed methods approach. This methodology allows for a thorough examination of both quantitative and qualitative aspects. The psychological impact on participants was assessed using reliable questionnaires, ensuring the robustness of the data collected. Additionally, this approach enabled the documentation and analysis of specific participant characteristics, providing a deeper understanding of the diverse experiences and responses within the study population.

This study protocol was reviewed and approved by the CEIM Hospital Clinic Barcelona, Approval No. HCB/2020/0128. Written informed consent was obtained from all participants in the study.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

A.B., M.P., and A.M. designed the study. V.B., I.M., and C.I. contacted the pregnant women to participate in the study. M.P. organized all the data of the surveys. M.B. performed the statistical analysis. V.A.-R. performed the qualitative analysis. A.B., M.B., and V.A.-R. drafted the manuscript.

Maria Borrell and Montse Pauta contributed equally to this work.

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.