The Involvement of Neonatal Intensive Care Unit and Other Perinatal Factors in Postpartum PTSD After Cesarean Section

The experience of a neonate hospitalised in the Neonatal Intensive Care Unit (NICU) is an understandably traumatic experience for the parents, while the feelings of mental distress may persist even after the neonate leaves the NICU [1]. Parents face separation from their child in an unfamiliar and stressful environment with access difficulties and sparing of communication and information from the staff [2]. Such an experience can affect parental mental health, with an impact on the transition to parenthood.

Mental distress resulting from preterm birth and/or NICU hospitalisation can be understood as post-traumatic symptomatology [3]. Post-traumatic stress disorder (PTSD) symptoms include reexperiencing the event, increased reactivity and avoidance and negative mood, while at least one month must have passed since the traumatic event [4] according to the Diagnostic and Statistical Manual DSM-5 [5]. It is already known that one in three women experience their delivery as a birth trauma [6], while one in four of them can develop postpartum PTSD (P-PTSD) [7]. Phillips [8] underlines the importance of the ‘magic hour’, that is, the first minutes immediately after birth in the mother–infant bond. Maintaining this contact after birth facilitates this bond and prevents the development of P-PTSD. Nevertheless, P-PTSD can affect the mother–infant bond [9, 10] and increase the infants' risks of developing behavioural or emotional problems later [11].

Several studies have demonstrated that many stressful events during the perinatal period, such as pathology of gestation [12], emergency cesarean delivery [13, 14], preterm birth [15] and postpartum complications [14] contribute to the development of P-PTSD [16]. Especially, an emergency cesarean section (CS) has been linked to increased rates of maternal depression and PTSD in the postpartum period [17, 18, 19], due to the unexpected event and the lack of the mother's mental preparation [20].

To meet the diagnosis of PTSD, exposure to a potentially stressful factor (Criterion A) must apply. More specifically, there must be exposure to actual or threatened death, serious injury or sexual violence in any of the following ways: a) direct exposure, b) as a witness to the event, c) learning that the traumatic event happened to a significant other and d) daily exposure through work [21]. Based on the above, admission to the NICU is considered a traumatic event for mothers, who, in addition to a potentially traumatic birth experience, such as CS delivery, face the threat of death or physical unseemliness of the neonate.

On the other hand, the prevalence of PTSD in mothers of infants admitted to the NICU ranges between 24% and 44%, according to previous studies [22, 23]. So far, however, NICU admission has not been studied as an additional trauma variable in postpartum women. Furthermore, it is not known to what extent the NICU may contribute to the high rates of PTSD in women after CS. Given that women who undergo CS (emergency or elective) are more likely to develop post-traumatic symptoms [13, 14, 17, 24,], the admission of the neonate to the NICU of CS-exposed women makes them a particularly vulnerable group of postpartum mothers. So far, there are several studies that show a correlation between CS and PTSD [17, 24]. It has also found a positive relationship of the admission of the newborn to the NICU [25, 26]. However, there is no study that identifies the effect of the NICU on the special group of women after CS.

Therefore, the purpose of this study is to investigate the involvement of the admission of a neonate to the NICU, and other perinatal factors on the development of P-PTSD in a sample of women after CS.

The specific study was carried out between July 2019 and June 2020 at the obstetrics clinic of the University Hospital of Thessaly in Greece with a prospective design. Our initial sample consisted of postpartum women after CS, aged 18 or older, who gave written informed consent to participate in the study. Finally, 469 postpartum women responded to the follow-up and constituted the sample of our study. Ethics Commission, Approval: 18838/08-05-2019, University of Thessaly.

A total of 469 women were analysed in this study. Their mean age was 32.58 ± 6.15 (SD) years, with the majority of them being from the city (N = 373, 79.5%); 98.5% (N = 462) were married, engaged or in a relationship. Overall, 43.4% (N = 204) of the sample had completed undergraduate and/or postgraduate studies, with a similar percentage who had finished high school (N = 196, 41.8%); 10.7% (N = 69) participants had finished primary school or junior high. None of these demographic variables or the occupation of the participants relate to PTSD. However, it seems that the financial status contributes to PTSD, but without an adequate number in the last category (high), preventing us from assumptions about differences between categories (Table 1). For almost all independent variables presented in Table 1, and PTSD as dependent, one-way analyses of variance (ANOVA) were applied.

Taking into consideration perinatal and mental health variables from Table 1, except for a kind of conception variable, a statistically significant relationship with PTSD can be noted. Therefore, multiple analyses revealed that parity, type of previous labour, gynecologic history, gestational week, type of C-section, cause of C-section, C-section complications, birth expectations, admission to the NICU, breastfeeding, psychiatric history, partner support and traumatic C-section, as well as Criterion A1 and A2 variables, were independently associated with the presence of PTSD.

The next analysis conducted in the present study, with the independent variable admission to the NICU and the dependent PTSD, is univariate ANOVA, while for the independent variables being the four factors of PCL-5 (intrusion/re-experiencing, avoidance, negative alterations in cognitions and mood & alterations in arousal and reactivity) multivariate ANOVA was applied. Prior to this, the correlations analysis among PCL-5 factors showed high, positive correlations, ranging from .717 to .919. The results of the ANOVAs are presented in Table 2. The cutoff for the provisional PTSD diagnosis was 33.

At the univariate level, the F criterion showed a statistically significant relationship between admission to the NICU and all five independent variables. According to the Scheffé post-hoc criterion, it is observed that mothers of the present sample, whose child was admitted to the NICU, either because of prematurity or any other reason reported in Table 2, are more likely to develop PTSD (p < .001), as well as to have higher scores on the intrusion/re-experiencing (p < .001), avoidance (p < .001), negative alterations in cognitions and mood (p < .001), and in alterations in arousal and reactivity subscale (p < .001).

Finally, a multiple linear regression analysis was performed in order to investigate the relation between the reasons for the admission to the NICU, in correlation to pathology gestation, cause of C-section and Criterions A1 and A2, and PTSD as a dependent variable. Therefore, binary basis variables, also called dummy variables, are created to represent the independent variables with two or more distinct categories included in the present analysis; categorical independent variables cannot be directly entered into a multiple regression analysis. Results are presented in Table 3, as well as categories of the intercorrelated variables. Twenty-one independent variables were included in the analysis, as if they were all potential predictors of PTSD.

Through the linear regression, with PTSD being the dependent variable relating to the 21 perinatal health independent variables, the model introduced here has only nine predictors that were strongly related to PTSD: placenta previa type4/abruption/bleeding (β = .07, p = .049), premature contractions (β = .08, p = .039), heavy medical history/previous gynecological history/preeclampsia (β = .08, p = .034), abnormal heart rate/pathological NST/premature rupture of membrane/premature contractions/infection (β = .14, p = .004), life of child in danger (β = .12, p = .025), complications involving child (β = .15, p = .002), complications involving both (child and mother) (β = .12, p = .011), traumatic C-section (β = .041, p < .001) and prematurity (β = .12, p = .022). The regression equation reached significance, signifying that 46.64% of the variance of PTSD could be explained by this regression model. Consequently, the participants who experienced any of the situations appearing as a predictor in the present model were more likely to develop PTSD.

According to our results, 46.64% of postpartum women after CS experienced P-PTSD. The high rate of PTSD is consistent with an earlier study in which PTSD was present in 60% of mothers whose children were admitted to the NICU [33]. Also, the results of another study [23] showed that the acute PTSD symptoms of postpartum mothers were related to having a neonate in the NICU.

Several studies identified cesarean delivery as a major factor of P-PTSD [17, 24, 34], especially emergency CS [13, 24, 35]. In addition, it was found that P-PTSD was associated with placenta previa (type 4), abruption, bleeding, premature contractions, infections and preeclampsia. It is already known that bleeding significantly increases the risk of PTSD [36, 37], since bleeding is a traumatic event for the mother, as an obstetric emergency with a possible risk of losing her life or the life of the foetus. Premature contractions or premature rupture of membranes and preeclampsia increase the risk of preterm or complicated birth [38, 39] and increase the risk of development P-PTSD [40]. Also, conditions that may lead to preterm birth, such as infections, are also responsible, according to our findings, for the development of P-PTSD [41].

Our results show that previous heavy medical history or gynecological history, associated with P-PTSD. One explanation for this phenomenon is that women's gynecological history may affect the course of pregnancy and, therefore, create complications related to the health of the neonate. However, abnormal foetal heart rate during labour increases the chance of an emergency CS [42].

One of the important findings of this study related to P-PTSD was the fulfillment of the perinatal stressor Criterion A. More specifically, all life-threatening situations of mother, child, or both, before delivery, during CS, or after are development factors of P-PTSD. However, all of the above factors, including the inclusion of a neonate in NICU, for any reason are considered risk factors for P-PTSD [43].

P-PTSD of mothers who have children in the NICU is multifactorial. Two important factors that may play a decisive role in the emergence of post-traumatic symptomatology are the severity of the neonatal illness and the parents' perception of the severity of the illness [44]. Also, within the NICU environment, the maternal role is diminished, as there are strong restrictions governing the nursing and operating regulations of the unit. Therefore, mothers may perceive the infant's hospitalisation as a threat of death or disability, and furthermore, images within the incubator may further reinforce these fears [45]. As a result, prolonged mental distress in mothers often disrupts the mother–infant relationship and also appears to affect children's psychosocial development in the future [46, 47]. In addition, the removal of the neonate from the mother results in lack of breastfeeding, which increases the intensity of the traumatic experience and the emergence of post-traumatic symptoms [48].

According to our findings, the admission of a neonate to the NICU is an important factor in the development of P-PTSD in women after CS. Also, the perinatal factors that lead a neonate to NICU were identified as dangerous for the occurrence of P-PTSD, since they indicate the intense stress on the part of the mother for the risk of losing the life of the neonate. Therefore, it is very important that special groups of women who have undergone a CS, especially an emergency one, are treated as individuals who may develop PTSD in the postpartum period. Additional measures must be taken for mothers of children who have been admitted to the NICU with psychological support interventions and reassessment of their mental state. Special care is required to strengthen breastfeeding, in order to reduce the intensity of the birth trauma and develop the mother–child bond.