Post-traumatic stress disorder in parturients delivering by caesarean section and the implication of anaesthesia: a prospective cohort study

Childbirth is a highly emotional and challenging event in a woman’s life. However, rather than experiencing the joys of welcoming a new child, some women may experience fear for the life or wellbeing of their newborn, as well as their own. This can lead to serious psychological complications such as Post-traumatic Stress Disorder (PTSD) [1].

First identified in relation to war experiences, PTSD is an anxiety disorder that can occur after many types of traumatic events. According to the 1994 Diagnostic and Statistical Manual of Mental Disorders fourth ed. (DSM-IV) manual [2] used for the study before publication of the DSM-V manual, the disorder follows a traumatic experience usually involving threat to ones’ own life or physical integrity. It can also be associated with the witnessing of an event that involves death, injury, or threat to the physical integrity of another person (Criterion A). Response to that stressful event must include a specific number of symptoms from each of the three following criteria: (B) at least one re-experiencing symptom (i.e. the traumatic event is re-experienced as repetitive intrusive memories, flashbacks and nightmares), (C) three avoidance symptoms (i.e. avoidance of hospital or medical encountering), (D) two hyperarousal symptoms (i.e. irritability and hypervigilance). Symptoms must be present for more than 1 month (Criterion E) and must cause clinically significant distress or impairment in social, occupational or other important areas of functioning (Criterion F).

On the other hand, “Partial” PTSD, also referred to as PTSD “Profile” or “Subthreshold” PTSD, describes clinically significant PTSD symptoms in trauma-exposed persons who do not meet full criteria for PTSD [3‐5]. This category deserves equal attention as full PTSD, as a number of victims suffer from severe disturbing symptoms while not fulfilling all the PTSD criteria [3, 5, 6]. For instance, significant associations have been demonstrated between PTSD profiles and increased rate of suicidal ideation [7], alcohol use [8], episodes of absence at work place or increased healthcare utilization [9]. Thus, incomplete PTSD (PTSD profile) can cause significant impairment in daily life activity [5].

In the context of childbirth, it is currently known that 1–7% of parturients will experience partial or fully characterized PTSD at any given time point following childbirth [1]. Parturients suffering from PTSD are impaired in many aspects of their daily life and have an increased risk of developing long term depression, associated with difficulties in childcare and bonding, apprehension of sexual intercourse and of future pregnancies [10‐14]. Their risk of suicide is significantly increased [15]. Since the first studies by Bydlowski in 1978 [16], researchers have isolated perinatal and maternal risk factors leading to the development of PTSD [1, 10, 17‐20]. These include psychiatric disorders, previous trauma, poor levels of education and social support, increased sense of loss of control and psychological distress during labour. Additional factors linked to the occurrence of neonatal complications (poor staff support as well as deliveries in emergency situations) have also been identified [1, 12, 19, 20]. Amongst these, childbirth following a scheduled or emergency C-section appears as one of the top predictors for PTSD occurrence [19, 21]. However, it is unknown whether a C-section per se or if concomitant factors linked to it, particularly anaesthesia, play a role in the occurrence of a PTSD.

The aim of this study was to measure in a prospective cohort of parturients having a delivery by C-section, known risk factors for postpartum PTSD and assess the specific impact of anaesthesia-linked factors on the disease’s development. As major suffering and psychological sequelae can emanate from partial PTSD [3, 5, 6], we measured the full acute PTSD disease, as well as the acute partial PTSD symptoms according to Olde et al. [1] as “PTSD profile”, which include the DSM criteria B, C and D only.

Parturients with PTSD profile were included in the study as well as those who had the full spectrum of DSM-IV criteria for PTSD in order to include the maximum of patients with severe psychological morbidity from the traumatic experience. Moreover, the group with PTSD profile was selected in order to emphasize the continuum of this disorder and the need to prevent and manage partial symptoms of PTSD as well.

This study was performed between 2008 and 2011 at the maternity of the University Hospital of Geneva, a tertiary referral hospital in Switzerland. It was approved by the Geneva University Hospital Ethics Committee (Switzerland). Approval N°-CER 07–240- Ref Matped 07–045 (7/07/2008)- Chairperson Prof Michel Boulvain.

We enrolled in our study cohort, 280 women having a C-section and willing to participate in the study. Of these, 217 (77.5%) answered to the PCLS questionnaire and 175 (62.5%) had an additional phone standardized clinical interview performed. We identified 6 parturients (2.7%) with a confirmed PTSD and 20 (9.2%) with a PTSD profile (Fig. 2). Study population characteristics are detailed in Table 1. Most parturients (62.8%) were younger than 35 years and primipara or secondipare (65.7%). Risk factors for acute PTSD identified following univariate analysis (Table 2) were a non-European status, having a past significant somatic disease and a high level of dissociative reactions during C-section. Being married compared to single decreased the risk of PTSD and there was a higher number of patients with a PTSD who wished receiving additional sedation during the procedure.

Risk factors for acute PTSD profile (Table 3) were the presence of a stressful factor such as prepartum, intrapartum or postpartum maternal complications, a high level of dissociative reactions during C-section and having a past significant somatic disease.

When a full multivariable model for risk factors for acute PTSD profile was built (Table 4), maternal prepartum, intrapartum or postpartum complications, anaesthetic complications and dissociative experiences during C-section were found to be the most significant predictors. These were independent of the quality of the information provided by the anaesthesia team or parturients’ family status.

We also found that 63 (22.5%) patients were totally lost to follow up. These parturients did not complete the PCLS questionnaire and had no phone interview (Fig. 2). 49 patients (17.5%) completed the PCLS questionnaire but had no phone interview. When assessing the characteristics of parturients without phone interview (109 patients), we found that they were more often single, had lower education level and had more often past traumatic experiences and abortion (Table 5).

In the current study, we found an incidence of 2.7% of full acute PTSD and 9.2% of acute PTSD profile, in accordance with previous studies [1]. Significant risk factors for full acute PTSD or acute PTSD profile included: demographic characteristics (such as non-European status and being single), stressful event before or during childbirth (such as obstetrical and anaesthetic complications), subjective appraisal of birth as being possibly traumatic (such as a high level of dissociative reactions during C-section) and past morbidity (such as having a past significant somatic disease). More specifically, stressful factors related to childbirth and subjective appraisal of birth as being possibly traumatic were the most significant predictors of acute PTSD profile. These results are in line with the model developed by L.B. Andersen et al. [19] following a large meta-analysis of studies on PTSD in obstetrics. They confirm that PTSD is the result of a combination of several risk factors that can be aggregated into three main categories: 1) traumatizing events before or during childbirth, 2) subjective perception of these events as being highly traumatic, 3) predisposing psychological factors such as previous psychological trauma or past psychiatric disorder. Among these elements, the first and the second ones are, as demonstrated in our study, the most important category of factors for PTSD development.

In our study, we identified obstetrical and anaesthetic complications as being predictive factors of PTSD profile. This can be explained by the fact that these complications are potentially dangerous for the mother and/or for the baby, and are therefore and unsurprisingly associated with a PTSD profile. However, two important risk factors identified by previous studies, emergency situation and neonatal complications [19, 20, 29], did not prove significant in predicting acute PTSD profile or acute PTSD disease in our study. This could be explained by a possible lack of study power due to the limited number of disorders identified in our cohort. Nevertheless our study is the first to our knowledge, to show that anaesthesia plays a non-negligible part in it. Interestingly, the type of anaesthesia, whether general or neuraxial, had no significant impact on further acute PTSD profile or acute PTSD disease. In addition, the management of anaesthesia by an anaesthetist in training and poor information during the procedure given by the anaesthesia staff didn’t prove to increase the risk for PTSD. However, if any complication due to anaesthesia occurred (i.e. pain during surgery, nerve injury) the risk for a PTSD profile occurrence was higher.

Previous literature [18, 20] has suggested that a C-section was one of the top predictors for postpartum PTSD and depression. Our study, by analysing exclusively patients having a C-section for childbirth, shows that this paradigm should be probably refined. A number of factors are associated to the decision and completion of a C-section. These include neonatal malposition or distress, past or current maternal complications and patient willingness to avoid the subjective trauma of childbirth. Furthermore, a C-section is always associated with anaesthesia. Therefore, rather than the C-section itself, it is quite likely that all of the factors leading to, or around the C-section, contribute to the occurrence of PTSD or PTSD profile.

It is well described in the anaesthesia literature that intra-operative awareness can lead to PTSD in a number of patients [30‐32]. Our study shows however that other complications related to anaesthesia can also lead to PTSD disease or profile. A failed spinal anaesthesia with intraoperative pain or a severe postpartum postdural puncture headache can also increase the risk of developing PTSD disease or profile. It is likely that these complications contribute to the emergence of negative emotions such as pain and distress, all known to be associated with PTSD [19, 33]. This shows that the presence of a potentially traumatic event including anaesthesia complication is a key factor for the development of a PTSD.

In the model developed by L.B. Andersen et al. [19], subjective appraisal of birth as being traumatic which include subjective distress during labour, such as loss of control, fear for oneself and/or the baby and perinatal dissociation, is another important factor predicting PTSD following labour and delivery. In the literature, it is increasingly recognized that these subjective factors play an important role in the development of a PTSD [34, 35]. The evidence for the importance of the individual appraisal of the event is highlighted in our study by the highly significant association found between the negative appraisals of the trauma, i.e. dissociative reactions by patients during the C-section, and the presence of a full and partial acute PTSD. More specifically, this risk factor proved to be a significant predictor of PTSD profile both in the univariate and multivariable analysis, as well as PTSD in the univariate model. This result is in accordance with the model developed by L.B. Andersen et al. [19] which shows that subjective distress during childbirth is one of the most important factor of the development of PTSD. It may suggest that these patients have insufficient coping mechanisms available [36]. The request for further sedation by these patients can be understood as an overwhelmed coping capacity resulting in a strong feeling of dissociation. However, when sedation was administered during the procedure, our study results show that this did not help preventing PTSD. Therefore, further sedation of the mother after delivery, whether asked by the patient or proposed by the staff, may help diminishing current anxiety symptoms but not preventing psychological consequences of poorly managed traumatic experiences.

A number of weaknesses of this study should be mentioned. First, due to strict inclusion and voluntary participation criteria, the number of patients followed up in our cohort was limited. Although, our incidence of 2.7% of PTSD is similar to the one described in previous studies [1], it is likely that a number of acknowledged risk factors predicting PTSD profile or PTSD such as emergency situation or neonatal complications did not prove significant because of a lack of study power. This could explain why some risk factors identified by previous studies [19, 20, 29] such as emergency situation and neonatal complications were not significant in our study. Secondly, as a number of patients from foreign origins or admitted during weekends were not included in the study, a selection bias cannot be excluded and may also explain why “emergency” did not appear as a significant risk factor for PTSD. Thirdly, a number of patients (22.5%) were totally lost to follow-up. Although a 77.5% retention rate is considered as acceptable for cohort studies [37], the fact that an additional proportion of the patients (17.5%) could not be interviewed for PTSD disease assessment, reduced study power for this outcome. When comparing patient characteristics between those who completed the study and those who could not be interviewed for PTSD disease assessment, it appeared that patients lost to follow up were more often single, had lower education level and had more often past traumatic experiences and abortion. The later are known risk factors for PTSD [19]. Thus it cannot be excluded these patients had subclinical depression or PTSD symptoms such as avoidance and did not want to be further interviewed. This is likely to have biased study results towards less severe cases of PTSD and artificially decreased its true incidence. Fourth, as telephone interviews were used to diagnose PTSD, a measurement bias cannot be excluded. To minimise this risk, we used a validated diagnostic method based on a combination of clinical interviews and standardised questionnaires [1], and all cases were discussed with a second psychologist to confirm diagnosis. Fifth, due to the small sample size of the full PTSD group, no comparison could be done with the partial PTSD group. Thus, the degree to which parturients with partial PTSD are comparable to those with a full PTSD remains unknown, as well as the specific mechanisms underlying these two categories of PTSD. These two groups were included in the present study as recommended in the literature [5]. Indeed, partial PTSD, as well as full PTSD, can be associated with significant impairment and can be chronic [5], which emphasize the importance of including this group in the present study. Further research should be done in order to better understand the characteristics of the whole spectrum of PTSD following childbirth, its repercussions and the need to treat it. Finally, in the present study, only acute PTSD and acute PTSD profile were measured. Thus, no conclusion can be drawn about the potential risk of parturients developing chronic PTSD. To address this issue, longitudinal studies are needed in order to characterize the course of full and partial PTSD and their associated impairment.

Despite these weaknesses, we identified a number of risk factors for acute PTSD and acute PTSD profile in line with existing literature and managed to show that anaesthesia complications are associated with postpartum PTSD. Future studies should look at the development of a predictive score for parturients at high risk of developing a PTSD in order to identify and treat this complication preferably in the immediate postpartum period.

This study identifies predictors for acute postpartum PTSD profile and acute PTSD after delivery by C-section that can be easily recognised before or during the surgical procedure, particularly as spinal anaesthesia is most often used. In patients with a high degree of dissociative experiences during the C-section, or when maternal or anaesthesia complications occur, the risk for a postpartum PTSD profile increases significantly. These patients are likely to benefit from a close follow up by members of the obstetrics and/or anaesthesia team. If needed, intensive psychological support should be put in place as soon as possible.