Depression, Anxiety and Stress in Women with Breech Pregnancy - "Be not Afraid": A Prospective Study


 Background: Breech presentation at term is associated with higher perinatal complications through the increased incidence of cesarean section and complications by vaginal breech birth. The decision-making process between breech delivery, trial of external cephalic version or cesarean section may lead to maternal depression, stress and anxiety. Mental disorder in itself is associated with poorer maternal and fetal outcomes, such as preterm birth, small for gestational age infants, lower birth weight and increased risk for postnatal depression. This study aims to evaluate the level of psychological distress for women with breech presentation compared to cephalic presentation. We hypothesized, that women with breech presentation have higher levels of depression, stress and anxiety than other pregnant women. Secondary objectives were to analyze potential demographic risk factors and comorbidity of psychological distress in breech pregnancy. Methods: The breech study group was formed by 379 women with breech presentation. A sample of 128 women with cephalic presentation was recruited during routine clinical care. Depression, anxiety and stress symptoms were ascertained by means of the Depression-Anxiety-Stress-Score (DASS)-21 questionnaire. Categorial data was analyzed with Chi-square or exact test, continuous data with unpaired t-test or Mann-Whitney-U-test. Demographic risk factors were identified using a binary logistic regression model.Results: Prevalence of psychological distress among women with breech was not higher compared to those of other pregnant women. Symptomatic depression, anxiety and stress affected 5.8%, 14.5% and 11.9% of women with breech, respectively. Decreasing age was identified as a risk factor for anxiety (p = 0.006). Multiparity increased risk for depression (p = 0.001), for anxiety (p = 0.026) and for perinatal stress (p = 0.010). More than 80% of women with depressive symptoms had comorbidities of psychological distress.Conclusions: Breech presentation compared to cephalic presentation was not associated with higher levels of psychological distress. However, breech pregnancies are affected by symptoms of potential mental disorder. Multiparous women and younger women may need additional support and would benefit from a standardized screening tool for the assessment of perinatal psychological distress.Clinical Trial Registration: Ethical approval (EA2/241/18) was granted by the Ethics Commission of the Charité University Hospital on the 23.01.2019 (ClinicalTrials.gov Identifier: NCT03827226)

Confronted with a breech pregnancy, women may be overburdened with decision-making before delivery with an increasing level of depression, anxiety and stress expected. Psychological distress in breech pregnancy has however never been investigated.
Small for gestational age infants, lower birth weight and increased risk for preterm birth and postnatal depression have been observed in women with mental disorders during pregnancy. [16][17][18][19][20][21] Despite this, few units establish routine clinical screening for psychological distress. [22] The prevalence of antenatal depression decreases from rst to third trimester. [16, 23,24] A cohort study found 13.5% of the participants at 32 weeks of gestation at risk for depression. [25] Other authors indicated a point prevalence between 8.5% and 11.1% for minor and major depression in the third trimester. [16,24,26] Antenatal anxiety is more prevalent than depression. [27][28][29] Lee et al. revealed that 54% of the women experienced symptoms of anxiety in at least one trimester and 35.8% in the last trimester. [27] Pregnancy anxiety as fear linked to the pregnancy itself (e.g. fears about oneself and baby's well-being, concerns about labor and childbirth), is found to be one of the most potent risk factors for adverse outcomes. [17,18] High-risk-pregnancy, increasing gestational age, younger maternal age and history of drinking are shown to be predictors for anxiety. [27,30] Depression as well, is associated with a high-risk pregnancy, lower maternal education and social factors such as lack of social support and domestic violence. [30,31] The primary objective of the study was to evaluate the prevalence of symptoms of depression, anxiety and stress for women with breech compared to women with cephalic presentation. We hypothesized higher levels of psychological distress among women with breech.
Secondary objectives were to determine (1) potential risk factors for high levels of distress among breech pregnancies, (2) the in uence of gravidity on distress and (3) the level of comorbidity between depression, anxiety and stress.

Methods
We conducted a prospective observational study between February 2019 and September 2020 in the obstetric clinic of Charité University Hospital Berlin.
Ethical approval (EA2/241/18) was granted by the Ethics Commission of the Charité University Hospital (ClinicalTrials.gov Identi er: NCT03827226). Written informed consent was provided by all women who agreed to take part in the study.
We used a self-designed questionnaire to collect demographic data, including age, body mass index (BMI), gestational week, gravidity, parity, history of cesarean birth, spontaneous birth, miscarriage and pregnancy termination. Any pre-existing health-condition and gestational complication as well as smoking and consumption of alcohol or drugs were recorded. Clinical symptoms of depression, anxiety and stress were ascertained by means of the Depression Anxiety Stress Score (DASS)-21 questionnaire. [32] The breech study group was recruited from pregnant women with breech presentation attending the consultant-led breech clinic. Breech was con rmed on sonographic examination from the 36th week of pregnancy. The consultation involved discussing the options of vaginal breech birth, attempting external cephalic version or planning a cesarean birth.
Recruitment of the control group was conducted by direct approach in the general obstetric outpatient clinic. Those patients presented for normal delivery planning in the third trimester of pregnancy with cephalic presentation.
All women had a singleton gestation, were at least 18 years old, able to sign the informed consent and had basic German or English language skills. Exclusion criteria were history of mental disorder, use of antidepressant medication or anxiolytics and any fetal anomalies.
The DASS-21 is a self-report questionnaire consisting of 21 questions for measuring depression, anxiety and stress as negative emotional states. Developed by P. F. Lovibond and S. H. Lovibond as DASS-42, this short form version is also a well validated screening tool. [32,33] The questionnaire consists of three 7-item subscales that measure depression, anxiety and stress. Patients estimate the degree of symptoms they have experienced over the last seven days in a four-point-Likert-scale (0-3 points), with higher values indicating greater distress. The total score of each scale can range from 0 to 21 points and is built by summing all of the corresponding items. A total level of distress can be ascertained, ranging from 0 to 63.
The use of somatic items (e.g. fatigue, sleep disturbance, constipation and decreased appetite) in the assessment of depression is common. [35,36] This can cause an overestimation of depressive symptoms in pregnancy. [37] Due to the absence of somatic items in DASS-21, it is more appropriate for screening in pregnancy.

Results
A total of 564 women enrolled in the study, of which 57 women were excluded (Fig. 1). Of the remaining 507 pregnancies 379 had breech and 128 cephalic presentation.
Baseline characteristics group are shown in Table 1. There was no signi cant difference in maternal age and BMI between both groups (p = 0.070, p = 0.447). Table 2 shows median scores and prevalence of severity grades of the DASS-21 questionnaire. Clinical symptoms of moderate to severe depression symptoms were found in 5.8% of breech pregnancies.
Symptomatic antenatal anxiety occurred in 14.5%. Symptomatic stress was found in 12%.
Prevalence of symptomatic psychological distress (moderate to extremely severe) is presented in Fig. 2.
A Mann-Whitney-U-Test showed with no signi cant differences for scores of depression, stress and total distress between the groups. The median anxiety score of the control study group was signi cantly higher (p = 0.033).
No signi cant results were found for differences in severity grades, neither for the recommended ve grades (normal to extremely severe) nor for those de ned by our team (normal and symptomatic).
In order to determine demographic confounders and the in uence of the fetal presentation on the presence of symptoms of mental disorders, a multiple logistic regression model was performed (Table 3).
To calculate power we assumed the effect size of stress was 0.4, a sample of 379 in breech and 128 in control group, yielding a power of 97.4% with a signi cant level of 0.05 (two-sided). [38] Continuous variables were presented with mean and standard deviation (SD), or median and interquartile range (25th percentile, 75th percentile), depending on the distribution. Histogram and Shapiro-Wilk test were used to explore the normal distribution. Chi-square or exact test by Monte-Carlo-Method was calculated for categorical data, whereas continuous data was analyzed with unpaired t-test for normal distribution or Mann-Whitney-U test for non-normal distribution.
In order to determine in uencing factors for presence of psychological distress (moderate level or greater), we analyzed several potential demographic risk factors. Univariate association between distress and demographic factors was carried out using Chi-square. Variables tested were: maternal age, BMI before pregnancy, history of breech, family history breech, pregnancy risk, gestational diabetes, preexisting health condition, hypothyroidism, gravidity, parity, history of spontaneous birth, cesarean birth, miscarriage or pregnancy termination and fetal presentation. All variables that showed statistical signi cance and relevant factors (e.g. age, BMI) were included in a multiple logistic regression model. Presence or absence of depression, anxiety, stress and total distress were de ned as dependent variables.
A p-value < 0.05 was assumed as statistically signi cant. Data was analyzed with SPSS software version 25.0 (SPSS Inc., Chicago, Illinois, USA).
Gestational diabetes was found to increase the risk of stress by 2.42 times (p = 0.051). Preexisting health condition (hypothyroidism excluded) had a 1.78 times higher risk of anxiety (p = 0.042). Multiparity signi cantly increased risk for depression, stress and total distress.
After adjusting for age, BMI as well as gestational and medical history, fetal presentation was not a signi cant predictor of psychological distress in pregnancy. Breech pregnancies were 1.51 times more likely to be stressed during third trimester (p = 0.279), but had lower risk for depression and anxiety, however these results were not statistically signi cant.
A multiple logistic regression model was performed to determine risk factors for psychological distress with breech. Adjusted Odds Ratios (95% CI) are shown in Table 4.
Multiparity was found to be the most important risk factor for symptoms of mental disorder. Women who have given birth to at least two children had an eightfold higher risk for depression (p = 0.001), threefold higher risk for anxiety (p = 0.026) and fourfold higher risk for stress (p = 0.010) compared to nulliparous women. Being younger increased the risk for the presence of symptomatic anxiety (p = 0.006).
Gestational diabetes was 2.83 times more likely to cause high total distress scores (p = 0.050) and 2.61 times increased risk for stress, although this association was not signi cant (p = 0.088).
Breech pregnancies were divided into four groups: (1) primigravida, (2) multigravida with either history of term delivery, (3) history of miscarriage or (4) history of delivery and miscarriage.
Another relation was found for primigravida vs. multigravida with history of term delivery. Multigravida had higher depression scores (p = 0.007) as well as stress scores (p = 0.043).
All other combinations, especially the comparison between history of delivery and history of miscarriage did not show signi cant results. Comorbidity with depression and anxiety occurred in 3.6%, depression and stress in 4.2% and anxiety and stress in 6.3% of breech cases.
Multimorbidity (symptoms of depression, anxiety and stress simultaneously) was found in 11 women, representing 50.0%, 20.0% and 24.4% of the depressive, anxious and stressed population, respectively (Fig. 3). Of those with symptomatic depression, 9.1% were comorbid for anxiety and 22.7% for stress, the other 18.2% had no comorbid distress. Of those women with symptoms of anxiety, more than half had no comorbid distress, another 3.6% rated themselves as depressive and 23.6% as stressed. Combinations with either depression or anxiety was found in 11.1% and 28.8% of the participants suffering from stress, whereas a third of the stressed women had no other psychological comorbidities.

Main ndings
The prevalence of symptoms for moderate to extremely severe depression, anxiety and stress in the breech pregnancy was 5.8%, 14.5% and 11.9%, respectively. There was no statistical difference compared to cephalic pregnancy.

Interpretation
Studies show that women with high-risk pregnancy have a higher risk of psychological distress in pregnancy. [30,39] We used the well-established DASS-21 questionnaire to assess symptoms of depression, anxiety and stress in breech compared to cephalic pregnancy. [32,33] In the breech group symptoms of mild depression were present in 5% and moderate to extremely severe symptoms in 5.8%. One study by Barber et al. found 10.5% of pregnant women as mildly depressive and 21.5% as moderate or highly depressive.
[38] In this study however, women with known mental disorders were not excluded, pregnancies with breech were not de ned and an online survey was used. Other studies showed variable prevalence rates of 13.5% and 11.1% for depression in pregnancy with the Edinburgh Postnatal Depression Scale. [16,25] These variations are likely because women with known mental disorders were not excluded and gestational ages were also not de ned. A meta-analysis on prevalence and incidence of perinatal depression excluded studies based on self-report screens and found a point prevalence of 8.5% for minor and major depression. [26] Mild and symptomatic anxiety was found in 11.9% and 14.5% of the breech study group. Other studies such as Barber [27] However, these studies were limited as patients with known mental disorders were not excluded, gestational age and fetal presentation was not de ned and different screening tools were used. A meta-analysis by Dennis et al. could identify the difference between the prevalence of anxiety over all trimesters, when ascertained by self-report symptoms (18.0%) or clinical diagnosis as a measurement (15.2%).
[28] Prevalence in third trimester was determined as being 24.6% for self-report and 15.4% for clinical diagnosis of any anxiety disorder, similar to the incidence in our breech study group.
In our study, 9% of breech pregnancies suffered from mild stress and 11.9% from moderate to extremely severe stress. Compared to depression and anxiety, antenatal stress seems to be neglected in research as it does not offer a medical diagnosis of a mental disorder. Two studies on stress in pregnancy presented varying results. Woods et al. used a clinical screening protocol for psychosocial strain and found 78% had low to moderate and 6% high stress levels. [22] In comparison, Barber  We performed an analysis of prevalence rates of combined components of psychological disorders in the breech population. Comorbid depression and anxiety affected 3.6%, depression and stress 4.2% and anxiety and stress 6.3%. Of the women with symptoms of depression, we found more than 80% had high levels of other psychological comorbidities. Comorbidity of antenatal depression and anxiety has been investigated in numerous studies. [20,27,29,40,41] A meta-analysis showed in the third trimester that comorbid anxiety and mild to severe depression occurred in 9.5% of all cases and moderate to severe depression in 6.6%. [41] There remains a paucity in the literature however on the role of mental stress in pregnancy, therefore the association with other comorbidities has been rarely described.
We also analyzed predictors of psychological disorder in women with breech. We found, that women with multiparity, gestational diabetes and decreasing age were at higher risk. In the literature there are con icting results on the in uence of parity. Fairbrother et al. found no signi cant difference between nulliparas and multiparas, Dipietro et al. found a higher prevalence among multiparas whereas Gillespie et al. found a higher prevalence in primiparas. [42][43][44] We found that breech pregnancy beyond 36 weeks of pregnancy, even with the challenges of deciding on interventions did not have signi cantly higher levels of psychological distress symptoms, compared to cephalic pregnancies. In fact, women with cephalic presentation in the control group scored higher on the anxiety subscale. Presumably, this might be a result of the collocation of the control group with other obstetric complications (e.g. hypertension, gestational diabetes, oligohydramnios, previous cesarean birth, previous stillbirth) which are associated with higher rates of anxiety. [30] However, a logistic regression analysis of our study and control groups showed no in uence of these comorbidities on the presence of psychological distress. Anxiety could also be affected by fear of childbirth in general.
Laursen et al observed a signi cant association of fear of childbirth with depressive and anxious symptoms. [45] Rouhe et al. found that fear of childbirth affects primarily nulliparous women and women with a history of cesarean birth.
[46] A proportion of women presenting in the breech clinic may perhaps be more desirous for the chance to deliver vaginally and may have less fear of childbirth.

Strengths and Limitations
This is a large prospective study performed on over 370 participants with breech. This is the rst study looking at breech and the in uence of perinatal psychological distress symptoms. Importantly, all women with pre-known mental disorders were excluded to minimize selection bias and a well validated screening tool was used. [33] The DASS-21 self-report questionnaire is however time dependent, evaluating symptom over the past week. Negative emotional states at other time points may have in uenced self-assessment scores. Nevertheless, questionnaires are not used for diagnosis, but rather as a screening tool. Socio-economic confounders such as relationship status were not ascertained. There is already extensive research done on social predictors of psychological distress and we chose to look primarily at clinical parameters.
Despite no increase in psychological distress symptoms in women with breech in general, we found that those associated with multiparity, gestational diabetes and decreasing age are at higher risk of developing a mental illness and therefore require additional support. Screening for mental disorders should be established in clinical routine to detect women who may be at high risk of mental illness. Hare et al recently showed that anxiety in pregnancy was associated with a 15% increased risk of postnatal depression and this can negatively in uence mother-infant bonding. [47] Optimal antenatal care should therefore include assessment of the mental health status of expecting mothers and screening for stress in particular. Furthermore, perinatal and delivery complications such as can occur with breech pregnancy can increase a child's risk for anxiety independent from the parental psychopathology.
[48] Freed et al also found it useful to screen for anxiety in pregnancy especially in mothers with a known psychiatric disease such as bipolar disorder as this can in uence psychopathology in offspring. [49] Support should then be tailored to meet individual needs. We therefore recommend a standardized screening tool such as the DASS-21 questionnaire for the speci c categories of perinatal psychological distress in order to make comparisons between future studies compatible. Indeed, prenatal screening for anxiety can be implemented into prediction models used to earlier identify mothers and offspring at risk. [50] Conclusion In women with breech pregnancies signi cant symptomatic depression, anxiety and stress symptoms were found in 5.8%, 14.5% and 11.9% respectively. Compared to cephalic pregnancy this was not higher. However, multiparity, gestational diabetes and decreasing maternal age were identi ed as potential factors for developing mental distress in breech pregnancies and require additional support.

Declarations
Ethics approval and consent to participate Ethical approval (EA2/241/18) was granted by the Ethics Commission of the Charité University Hospital on the 23.01.2019 (ClinicalTrials.gov Identi er: NCT03827226). All study participants provided written consent.

Availability of data and materials
Data sharing is available on reasonable request from the corresponding author.

Disclosure of Competing interests
The authors report no con ict of interest.

Funding
There were no sources of nancial support.

Authors' Contributions
MS and LH wrote the paper. MS, LH, EL, MA-E and ER collected data. PG,MS and LH designed and analyzed the study. WH designed and reviewed the manuscript. All authors contributed, reviewed and Prevalence of symptomatic distress (moderate to extremely severe) measured by the DASS-21 for breech and control study group.

Figure 3
Psychological comorbidities of women with breech presentation. Subgroup analysis of women with symptoms of depression (A), anxiety (B) and stress (C). They had either no comorbidity (green: only one