Association between epidural analgesia and postpartum psychiatric disorders: A meta-analysis

Aim Women during the postpartum period are at risk for psychiatric disorders such as postpartum depression (PPD), post-traumatic stress disorder (PTSD), and anxiety. It is controversial whether labour epidural analgesia have a protective role on PPD, PTSD, and anxiety or not. This study is a meta-analysis of previously published observational studies to investigate the association between epidural analgesia and PPD, PTSD, and anxiety. Methods We searched Cochran Library, PubMed, ProQuest, EMBASE, Scopus, Web of Science, PsycINFO, and MEDLINE from inception until December 2022. The outcome measures were positive screen of PPD, PTSD, and anxiety among women who received labour epidural analgesia. The Newcastle-Ottawa Scale was used to assess the quality of the observational studies. Statistical analysis was performed using Stata 17.0 software. The mean differences or odds ratios were pooled based on random effect. Results We included 31 studies (12,064 women) in the review. The meta-analysis of PPD studies reported mean (standard deviation) showed no significant association between epidural analgesia and PPD (Mean Difference = 0.01; 95% Confidence Interval = −0.14 to 0.16; p-value = 0.88). The meta-analysis of PPD studies reported percentage showed no statistically significant difference in terms of the chance of depression between the exposed and non-exposed groups (Log Odds Ratio = −0.61; 95% CI = −1.48 to 0.26; p-value = 0.17). The meta-analysis of PPD studies reported OR showed indicate a lower chance of depression in the epidural group compared to the non-epidural group (Effect size = 1.01; 95% CI = 0.64 to 1.38; P-value = 0.00). In the case of PTSD, due to the heterogeneity of the tools and the methodology of the studies, it was impossible to reach a definitive conclusion and measure the studies in a meta-analysis. The meta-analysis of anxiety studies showed that no significant association between epidural analgesia and anxiety (MD = −1.36; 95% CI = −3.38 to 1.14; p-value = 0.29). Conclusion Based on the meta-analysis of observational studies, epidural analgesia for labour pain relief has no protective effect on postpartum psychiatric disorders.


Introduction
Birth is a women's most challenging physiological and psychological experience [1].Women during the postpartum period are at risk for psychiatric disorders such as postpartum depression (PPD), post-traumatic stress disorder (PTSD), and anxiety [2].
The most common postpartum psychiatric disorder is postpartum depression, which onset the first month following birth [3].Studies have shown that postpartum depression affects a significant portion of the global population, with reported rates ranging from 8% to 26% [4,5].PPD can cause physical and psychological complications for mother and infant [6].Post-traumatic stress disorder has been defined as "the complex somatic, cognitive, affective, and behavioral effects of psychological trauma" [7].The prevalence of PTSD following birth is 7.2% [7,8].Postpartum anxiety has been reported in 16.2% of women within six weeks following birth [9] and may coincide with PPD [10].
The etiology of postpartum psychiatric disorders is complex, and several factors have been identified as its predictors.The intensity of labour pain has been established as a potential predictor for the onset of these disorders [7,11].It is concluded that labour pain may cause severe emotional states, including anxiety, stress, and fear [12].Women who experience severe pain during labour release a high amount of catecholamines, which prevents the contraction of the uterus and the progress of labour.This increases women's anxiety and creates a vicious cycle of poor progress of labour, releases more catecholamins, causes more anxiety and other mental illnesses [13].Depression and chronic pain occur simultaneously in more than 80% of patients.A number of common neural substrates including neurotransmitters, neurotrophins, inflammatory mediators and neuroendocrine alterations are involved in these conditions.Any or all of them may also alter neural function in key brain regions responsible for regulating emotional and pain processing [14].Therefore, pain relief during labour may improve maternal well-being and enhance maternal and neonatal outcomes, particularly those pertaining to the mother's mental health in the postpartum period [15,16].A multitude of labour pain management methods have been devised, encompassing both non-invasive and invasive techniques.These include epidural anesthesia [17], inhalation of nitrogen oxide [18], and administration of opioids [19].According to previous studies, epidural administration is widely regarded as the most effective method for pain management [20].
Regarding the association between pain and psychiatric disorders, several studies assessed the effects of epidural administration on lowering the incidence of PPD, PTSD, and anxiety in women during postpartum [21][22][23].However, the findings of these studies are inconsistent [24][25][26][27].Therefore, the clinical issue of whether epidural analgesia can effectively prevent psychological outcomes by decreasing labour pain remains unsolved [28].This study is a meta-analysis of recently published observational studies investigating the association between epidural analgesia and psychiatric disorders such as PPD, PTSD, and anxiety.

Methods
Methods were aligned with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guideline.

Search strategy and study selection
Eight electronic databases (Cochran Library, PubMed, ProQuest, EMBASE, Scopus, Web of Science, PsycINFO, and MEDLINE) were systematically searched from inception until December 2022.Grey literature was searched via Google Scholar, ResearchGate, and various websites for dissertations.Reference lists of included studies and reviews were explored for further studies.A search strategy was developed using the keywords "Epidural analgesia", "Postpartum Depression", "PPD", "Post-Traumatic Syndrome Disorders", "PTSD", "Anxiety", and "Stress".SPJ and LFG performed the screening process and independently determined eligibility by reviewing the relevance of all study titles and abstracts.Regarding the lack of sufficient information in the title and abstract, the full text was read to decide whether to include or exclude the study.Disagreements between the two reviewers were resolved by a discussion involving a third reviewer (SGH).

Eligibility criteria
Inclusion criteria were as follows: observational studies (prospective or retrospective longitudinal, cross-sectional) in English investigating the association between epidural analgesia and postpartum psychiatric disorders, including PPD, PTSD, and anxiety.Standard scales were employed to measure the outcomes between one to 24 months postpartum.Standard scales included for PPD: EDPS, Stein's MB scale; For PTSD: PPQ, PC-PTSD, CBiTS, TES and IES; and for anxiety: STAI, PSS.Studies involving women who had an elective cesarean section were excluded.

Data extraction
Data were extracted into a data extraction form in Microsoft Excel by second (SPJ) and third authors (LFG).Data extraction for the included studies were: a) author, b) year of publication, c) country, d) study design, e) the number of participants, f) type of psychiatric disorders, j) final results of the studies (results reported in percentage or mean and standard deviation or odds ratio for case and control group).

Quality assessment
The Newcastle-Ottawa Scale was used to assess the quality of the observational studies included.This scale contains eight items within three domains and a total maximum score of nine.The final quality scores were categorized as follows: "low quality with very high risk of bias" (score< 3); "intermediate quality or intermediate risk of bias" (score 4-6); and "high quality" (score 7-9) [29].The quality of the included studies was independently assessed by the first author (LFG) and second author (SPJ).In the next stage, judgments were compared, disagreements among the reviewers were finalized by consultation with the first author (SGH), and the final result was obtained.

Data synthesis
All analyses were performed using STATA version 17.0 (Stata Corporation, College Station, TX, USA).The percentage, mean, and odds ratio of risk for psychiatric disorders such as PPD, PTSD, and anxiety were examined.The presence of heterogeneity was assessed using Cochran's Q-test, with a significance level of p < 0.05.The I-square test was also used to calculate the percentage of heterogeneity [30].The odds ratio (percentage) or mean (standard deviation) were extracted from cohort and case-control or cross-sectional studies.A random-effect model was used to estimate pooled effect sizes.Predefined subgroup analyses were conducted to explore the origin of heterogeneity, with a focus on the sample size, follow-up period, and baseline value for psychiatric disorders.Publication bias was analyzed using funnel plot analysis and Egger's regression asymmetry test [31].

Meta-analyses results related to PPD
The meta-analysis of the studies by Kountanis et al.Nahirney et al., and Ren et al. using the random effect model (I-squared = 77.6%,p-value<0.01)showed a statistically significant difference in terms of the chance of depression between the exposed and nonexposed groups (OR = 1.01; 95% CI = 0.64 to 1.38; P-value = 0.00) (Fig. 4).Also, the results of Egger's test indicate the absence of publication bias in the study results (p-value = 0.79).
A longitudinal cohort study by Smorti et al. [51] investigating the risk factors associated with postpartum depression has shown a positive and significant relationship between the duration of epidural analgesia administration and postpartum depression (MD = 1.87; 95% CI = 0.48 to 1.14; p-value = 0.00).A longitudinal study by Liu et al. [50] into the relationship between epidural analgesia and the reduction of the risk of depression in mothers two years after birth showed that the rate of postpartum depression was significantly lower in the group exposed to epidural analgesia than in the non-exposed group (7.3% vs. 13.6%;p-value = 0.029).A cohort study by Lim et al. [48] found a statistically significant relationship between epidural analgesia and the reduction of postpartum depression (OR = 6.6; 95% CI = 1.9 to 22.4; p-value = 0.003).A cross-sectional study by Banovcinova et al. [23] to determine the predictors of postpartum depression showed that epidural analgesia is not one of the predictors of postpartum depression (β = − 0.03; 95% CI = − 1.35 to 0.57; p-value = 0.425).A case-control study by Gailard et al. [46] evaluated the relationship between various psychological, midwifery, and individual-social factors with postpartum depression and found no significant difference between the case group (with postpartum depression) and the control group (without postpartum depression) regarding the use of epidural analgesia (OR = 1.1; 95% CI = 0.4 to 2.6; p-value> 0.05).The study by Johnstone et al. [45] investigated the relationship between obstetric risk factors and postpartum depression and found that epidural analgesia did not increase the risk of postpartum depression (OR = 1.36; p-value>0.05).The study by Suhitharan et al. [21] investigated the relationship between analgesic and psychological factors and risk of postpartum depression and found that the risk of postpartum depression was significantly lower in women who received epidural analgesia (33 of 329 [10.0%]) than those who did not (29 of 150 [19.3%]) (P = 0.0078).

Postpartum anxiety
The meta-analysis of studies by Floris et al.Shishido et al., and Lim et al. has shown that (I-squared = 46.5%,p-value = 0.13) the average anxiety score in the exposed group (women receiving epidural) was lower than the non-exposed group (not receiving epidural), however, it was not statistically significant (MD = − 1.36; 95% CI = − 3.38 to 1.14; p-value = 0.29) (Fig. 5).According to the results of Egger's test, no publication bias was observed between the studies (p-value = 0.53).

Discussion
This study is the first meta-analysis to investigate the association between receiving epidural analgesia and postpartum psychiatric disorders among women.According to the results of two different meta-analyses of depression studies, there was no statistically significant difference in the mean depression score between the exposed group (women receiving epidural) and the non-exposed group (women not receiving epidural).Another meta-analysis found a statistically significant difference between the exposed and nonexposed groups in terms of depression.There was no statistically significant association between epidural analgesia and PTSD in the cohort studies of Kountanis et al.Suchanecki  Because of the variety in the methods used to measure PTSD and the heterogeneity in the methodology of the investigations, it was impossible to include the studies in a meta-analysis.In the meta-analysis of anxiety studies, the difference in mean anxiety score between the exposed and the non-exposed groups was not statistically significant.
PPD: According to the results of two different meta-analyses of depression studies, there was no statistically significant difference in the mean depression score between the exposed group (women receiving epidural) and the non-exposed group (women not receiving epidural).These results are consistent with the meta-analysis results of Parise et al. [28], Kountanis et al. [55], Almeida et al. [56], Zinger et al. [57], and another meta-analysis found a statistically significant difference between the exposed and non-exposed groups in terms of depression.The results were controversial due to the fact that the outcome of the PPD will most likely vary depending on the diagnostic criteria and follow-up duration used.For example, depression is most common one year after birth, however, most studies have a shorter follow-up time [28].The follow-up duration in most studies ranged between 2 h and 8 weeks, with the exception of Hernández-Martnez et al.'s [7] and Kountanis et al.'s [44] studies, which had follow-up periods of 1 and 2 years, respectively.As a result, it can be stated that in most studies, the follow-up period was insufficient, and it is possible that this is why a significant statistical relationship between epidural analgesia and psychiatric disorders was not demonstrated.On the other hand, while the diagnosis of psychiatric disorders needs the approval of psychologist, in almost all studies, screening tools have been used as a diagnostic method [57].Different cut-offs have been used to diagnose depression in different studies, which can impact the ability of the tool to detect depression.In the study by Edipoghlu et al. [38] depression is defined as a score of 10 or higher; in the study by  Eckerdal et al. [35] a score of 12 or higher; and in the study by Hiltunen et al. [7] a score of 13 or higher.
Recently, epidural analgesia has been proposed as a possible modifiable factor for postpartum depression.This hypothesis is founded on the association between chronic pain and the onset of psychiatric disorders.Nevertheless, in this study the association between the use of epidural and postpartum depression was contradictory, the relationship between epidural analgesia and postpartum psychiatric disorders cannot be ruled out with absolute certainty [56].
PTSD: Due to the heterogeneity of the tools used to measure PTSD and the heterogeneity of the methodology of the studies, it was not possible to measure the studies in a meta-analysis.As a result, it is impossible to reach a definitive conclusion in this case.In addition, because the majority of the included studies were cross-sectional, it is impossible to accurately evaluate the cause-and-effect relationship in these studies.Longitudinal studies or randomized controlled clinical trials should be devised and conducted to more accurately assess the association between epidural analgesia and the incidence of PTSD [58].
Anxiety: In this study, there was no statistically significant difference between the anxiety score in the exposed group (women receiving an epidural) and the non-exposed group (women not receiving an epidural).These results are consistent with a longitudinl study by Shishido et al. [37].Anxiety is a complex emotion and women during labour experience both external (contractions and pain) and internal (anticipation and imagination of future birth) component of anxiety.Furthermore, some confounders (provider approach, parity, family or professional support, history of depression, preterm birth, negative birth experience, and poor maternal self efficacy) [59] can affect postpartum anxiety [60].
Although the association between pain and mood disorders following childbirth has been hypothesized in individual studies [61], and pain relief may be linked to a decrease in these disorders [56], there is no statistically significant association between pain relief and psychological disorders in the current meta-analysis.This may be because of the following factors: 1.Compared to the control group, women who request epidural analgesia typically exhibit more feelings of fear, anxiety, and depression.These confounding and predisposing factors for psychological disorders can prevent the positive and protective effects of epidural analgesia on postpartum psychiatric disorders.2. Research has shown that a number of various factors play a role in increasing risk of postpartum depression and anxiety [59], since postpartum psychiatric disorders are multifaceted.Pain was only one of the risk factors, and it is unlikely that only pain control with epidural analgesia is a protective factor for these disorders.3. Every woman experience pain differently because it is a complicated sensation impacted by psychological and genetic variables.It's possible that using epidural analgesia does not alleviate pain or lessen postpartum psychiatric disorders in those with limited pain tolerance.The length of time before utilizing epidural analgesia has not been assessed in studies, which may be connected to how intense the labour pain is felt.In order to determine the association between pain relief and the decline in psychiatric problems, it is recommended that future studies look into the association between mothers' reported pain intensity and the length of time their pain went unrelieved [58].4. In China, for example, the prevalence of depression and other psychiatric disorders is higher than in other countries.It is challenging to achieve definitive conclusions when comparing women from different cultures who receive diverse emotional and psychological support from people around them.Furthermore, the results may differ based on the study location [57].
It is recommended that future studies aim for greater consistency by implementing longer follow-up periods, utilizing confirmation of psychiatric disorder diagnoses by qualified professionals, and employing standardized screening tools with standard cut-off points.Additionally, conducting randomized clinical trials could provide more conclusive results [28].

Conclusion
Although epidural analgesia is the gold-standard technique for reducing labour pain, it is controversial whether using epidural analgesia is useful in preventing postpartum psychiatric disorders or not.Since they are complex and affected by various factors.It may be beneficial to consider conducting studies with longer follow-up periods, utilizing the expertise of psychologists or psychiatrists for the diagnosis of psychiatric disorders, standardized screening tools, and conducting randomized clinical trials in order to obtain more conclusive results.

Strengths and limitations
The individual studies included in this meta-analysis also have several strengths.These include appropriate eligibility criteria and using a valid screening tool to detect PPD, PTSD, and anxiety.Anyway, there were several limitations in this study: 1) Most of the studies were observational studies because of ethical concerns.Observational studies cannot account for unknown confounders that can only be adequately managed by randomized clinical trials.Due to the fact that most studies were not RCTs, the decision to perform neuraxial epidural analgesia may be affected by multiple factors, such as socioeconomic factors, which are important risk factors of PPD. 2) Because of the limited number of observational studies, we could not be able to perform subgroup meta-analysis by study design.3) Even if all studies used Edinburgh Postnatal Depression Scale (EPDS) as a screening tool, the cut-off values of EPDS differed between studies.In some studies, the lack of standardization of the screening cut-offs may over-or underestimate the effect of neuraxial analgesia on PPD. 4) EPDS is a highly validated screening tool, not a diagnostic instrument.The diagnosis of PPD may need confirmation by a psychiatrist.5) There is no universally accepted evaluation time point for PPD.Most studies screen PPD within 6 weeks after postpartum.However, approximately 7%-21% of depressive symptoms may last more than two years [62].It is recommended that future studies include long-term follow-ups.Sixth, other risk factors of PPD, including adverse life events, parity, and breastfeeding, were underlying confounders that may not be adjustable.

Fig. 2 .
Fig. 2. Meta-analysis of mean (SD) of postpartum depression in case and control studies.

Fig. 3 .
Fig.3.Meta-analysis of odds ratio of postpartum depression in case and control studies.

Fig. 4 .
Fig. 4. Meta-analysis of percentage of postpartum depression in case and control studies.
et al., and Lyons et al.In addition, in two cross-sectional investigations of Hernández-Martnez et al. and Nagle et al. epidural analgesia had a conflicting influence in PTSD symptoms.In the Hernández-Martnez et al. it was a protective factor and in Negle et al. study epidural analgesia was not a predictive factor for the development of PTSD.

Fig. 5 .
Fig. 5. Meta-analysis of mean (SD) of anxiety in case and control studies.

Table 1
Summary characteristic of included study.