Abstract
Background
Postpartum depression (PPD) is widely recognized as the most prevalent mental health crisis following childbirth and has been linked to sleep disturbances. However, the potential causal relationships between various sleep traits and PPD remain unclear. This study employs a bidirectional two-sample Mendelian randomization (MR) approach to investigate these associations.
Methods
The inverse-variance-weighted method was used to evaluate the causally linked sleep traits on postpartum depression. The weighted median, weighted mode, and MR-Egger were used to estimate the robustness of the inverse-variance-weighted method. The leave-one-out method estimated the sensitivity of the result. Cochran’s Q method was used for the heterogeneous test. The MR-Egger intercept and MR-PRESSO methods detected the horizontal pleiotropy.
Results
We examined the genetic causal relationships between nine sleep traits and postpartum depression. Sleep apnea syndrome (OR: 1.122; 95%CI: 1.063–1.185; p = 0.000), sleeplessness/insomnia (OR: 1.465; 95%CI: 1.104–1.943; p = 0.008), and frequency of tiredness/lethargy in last 2 weeks (OR: 1.725; 95%CI: 1.345–2.213; p = 0.000) genetically predicted the increased risk of postpartum depression. The reverse Mendelian randomization analysis showed PPD caused sleeplessness/insomnia (β: 0.006; 95%CI: 0.001–0.010; p = 0.016) and frequency of tiredness/lethargy in last 2 weeks (β: 0.007; 95%CI: 0.002–0.011; p = 0.004). The remaining six sleep traits showed no significant association with PPD. There was no heterogeneity or horizontal pleiotropy.
Conclusions
Genetic evidence reveals causal relationships between specific sleep traits and PPD, including sleep apnea syndrome, sleeplessness/insomnia, and tiredness. Whether certain sleep health indicators suggest a risk of postpartum depression or sleep issues that are caused by PPD, both may offer insights into the prevention and treatment of PPD.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12884-024-06929-7.
Keywords: Postpartum depression, Mendelian randomization, Sleep traits
Introduction
Postpartum depression (PPD) is characterized by an episode of depression diagnosed after childbirth within 4 weeks [1]. People with PPD usually exhibit a series of symptoms: emotional suffering, loss of appetite, sleep disturbances, increased fatigue, impaired creativity, reduced self-esteem, and a lack of confidence in life [2]. The prevalence of PPD in healthy mothers without a history of depression is between 12 and 17% [3], and in fathers ranges from 2.3% to 8.4% [4], and in Asian countries is approximately 21.8% [5]. PPD can damage marital relationships, and reduce family functioning, resulting in a significant burden on the family [2, 6]. When severe PPD persists in mothers, their children confront a 4.84-fold elevated risk of behavioral problems at 3.5 years, a 2.65-fold increased likelihood of struggling with mathematics grades at 16 years, and a 7.44-fold greater susceptibility to depression at 18 years [7]. Furthermore, PPD has an influence on sleep, exacerbating psychomotor retardation, and anxiety [2]. In severe cases, severely depressed postpartum mothers may harm infants, and even kill their children [8].
Sleep, as a basic physiological need of human survival, its alterations are present in various mental disorders [9]. Among postpartum women, poor sleep is associated with worse depressive symptoms [10]. In addition, a decline in sleep quality and tiredness contributes to the escalation of depressive symptoms [11, 12]. Sleep disturbances during pregnancy increase the risk of PPD such as insomnia [13, 14]. Due to the demands of caring for their newborn baby, postpartum parents reported increased sleep disturbances related to functional impairments [15]. For those postpartum women with a personal or maternal history of depression, sleep disturbances predict higher depressive symptoms [16]. Especially among women with postpartum depression, prominent sleep issues can escalate the severity of postpartum depressive symptoms [17]. Therefore, PPD and sleep alteration may mutually influence each other. But we cannot determine definitively whether sleep alteration is a cause or a consequence of postpartum depression based on sociological and clinical research because of limitations of the research method and heterogeneity of results [18]. Therefore, the present study aims to use A bidirectional two-sample Mendelian randomization (MR) study to solve those shortcomings from the biological dimensions.
The MR study utilizes genetic variants, as instrumental variables, related to exposures to evaluate their potential causal link with outcomes, aiming to minimize confounding bias [19]. Three assumptions must be fulfilled [20]: (1) the genetic instrumental variable is highly correlated with the exposure; (2) exhibits no associations with confounding of exposure-outcome relationships, and (3) the instrumental variable affects the outcome exclusively through the exposure variable [21].
Methods
Study design
MR research design was adopted to establish a causal relationship between sleep traits and PPD. Firstly, we selected the single nucleotide polymorphisms (SNPs) as instrumental variables According to the standards in previous literature [22]. We extracted the SNPs for the exposures (sleep traits) in which the significance level of the P-value met 5 × 10–5 The linkage disequilibrium threshold should meet r2 > 0.001 and the length of the SNPs should be longer than 10,000 kb. The F-statistics (F = beta2/se2) of SNPs should be higher than 10, where beta signifies the estimated genetic effect of the SNP on the exposure, and se indicates the standard error of the genetic effect. Then, if the SNPs associated with outcomes (PPD) are correlated with 5 × 10–5, they should be excluded. Lastly, those remaining SNPs should be harmonized and screened by the Steiger filtering method. Then, we conducted a statistical MR analysis. In the reverse MR analysis, PPD was used as the exposure, and the outcome variable was changed to sleep traits.
Data sources
The data sources of sleep disturbances were obtained from the UK Biobank (UKB), including Sleeplessness/insomnia dataset and two Frequency of tiredness/lethargy in last 2 weeks dataset. The SNP for sleep apnea syndrome was selected from the IEU Open GWAS Project. To reduce sample overlap, we extracted SNPs for PPD from the FinnGen database. The characteristics of data sources are summarized in Table 1 in detail. In addition, the sleep traits including daytime dozing (n = 386,548), morningness (n = 345,552), ease of getting up(n = 385,949), sleep duration(n = 384,317), daytime napping (n = 386,577) and snoring (n = 359,916) from Center for Neurogenomics and Cognitive Research (CNCR) database [23].
Table 1.
The characteristics of data sources used in this study
| Traits | Datasets | Unit | Gender | Population | Sample size(ncase) | Number of SNPs | Consortium | Year |
|---|---|---|---|---|---|---|---|---|
| Sleep apnea syndrome | EBI-a-GCST90018916 | NA | NA | European | 476,853(13,818) | 24,183,940 | Sakaue S | 2021 |
| Sleeplessness/insomnia | UKB-a-13 | SD | Males and Females | European | 336,965(NA) | 10,894,596 | Neale Lab | 2017 |
| Frequency of tiredness / lethargy in last 2 weeks 1 | UKB-a-245 | SD | Males and Females | European | 327,528(NA) | 10,894,596 | Neale Lab | 2017 |
| Postpartum depression | FINN-b-O15_POSTPART_DEPR | NA | Males and Females | European | 66,665(7,604) | 16,376,275 | NA | 2021 |
SD Standard Deviation, NA Not Applicable
Statistical MR analysis
The “TwoSampleMR” package was utilized in R.4.3.0 to conduct MR analysis, using the inverse-variance-weighted (IVW) model, weighted-median estimator, weighted mode, and Mendelian randomization-Egger regression. IVW was recognized as the primary method of Mendelian randomization analysis [24], and the other three methods were used to estimate the Robustness of IVW [25]. To determine reverse causation, the study was a bidirectional MR study. The two results of Frequency of tiredness/lethargy in last 2 weeks were combined into a statistical synthesis by meta-analysis, and the fixed effect model or random effect model adopted depends on the value of I2 [26]. The leave-one-out sensitivity test, involving the stepwise removal of single nucleotide polymorphisms (SNPs), was conducted to evaluate the impact of individual SNPs on the MR results [25]. Furthermore, heterogeneous results were identified through Cochran’s Q statistic test [27]. The MR-Egger intercept and MR PRESSO methods were used to detect the presence of horizontal pleiotropy [25]. Funnel plots display the individual Wald ratios for each SNP plotted against their precision.
Results
As shown in Tables 1 and 2, we obtained 97 SNPs as instrumental variables for Dozing 151 SNPs for Ease of Getting up, 195 SNPs for Morningness, 130 for Napping, 153 SNPs for Sleep duration, 278 SNPs for Snoring, 141 SNPs for Sleep apnea syndrome, 223 SNPs for Sleeplessness/insomnia, 224 SNPs for Frequency of tiredness/lethargy in last 2 weeks, 73 SNPs SNPs for PPD according to the criteria of thresholds (p < 5 × 10–5, r2 < 0.001 and > 10,000 kb, F > 10). Detailed information on finalized single-nucleotide polymorphisms associated with exposure and outcomes are listed in Supplementary file 1. As illustrated in Table 2 and Fig. 1, among nine sleep traits, only three sleep traits showed a causal association with PPD, which were Sleep apnea syndrome (OR: 1.122; 95%CI: 1.063–1.185; p = 0.000), sleeplessness/insomnia (OR: 1.465; 95%CI: 1.104–1.943; p = 0.008), and frequency of tiredness/lethargy in last 2 weeks (OR: 1.725; 95%CI: 1.345–2.213; p = 0.000) with IVW method. Table 3 showed the reverse MR analysis showed the causal effect of PPD on sleeplessness/insomnia (β: 0.006; 95%CI: 0.001–0.010; p = 0.016) and frequency of tiredness/lethargy in last 2 weeks (β: 0.007; 95%CI: 0.002–0.011; p = 0.004) with IVW method. The results of the weighted median, weighted mode, MR-Egger, and IVW analyses are consistent in direction, thereby enhancing the robustness of the IVW findings. The remaining exposure-outcome association showed no significance.
Table 2.
The Mendelian randomization results of postpartum depression on risk of sleep traits
| Exposure | NO. SNPs | Method | OR (95%CI) | P |
|---|---|---|---|---|
| Dozing | 97 | Inverse variance weighted | 1.015 (0.955 to 1.078) | 0.634 |
| MR Egger | 1.047 (0.893 to 1.228) | 0.574 | ||
| Weighted median | 0.967 (0.883 to 1.058) | 0.464 | ||
| Weighted mode | 0.913 (0.744 to 1.120) | 0.383 | ||
| Ease of Getting up | 151 | Inverse variance weighted | 1.000 (1.000 to 1.001) | 0.457 |
| MR Egger | 0.993 (0.978 to 1.008) | 0.378 | ||
| Weighted median | 1.000 (0.999 to 1.001) | 0.872 | ||
| Weighted mode | 1.000 (0.998 to 1.003) | 0.765 | ||
| Morningness | 195 | Inverse variance weighted | 1.000 (0.999 to 1.001) | 0.548 |
| MR Egger | 0.987 (0.975 to 1.000) | 0.052 | ||
| Weighted median | 1.000 (0.999 to 1.001) | 0.959 | ||
| Weighted mode | 1.000 (0.997 to 1.003) | 0.853 | ||
| Napping | 130 | Inverse variance weighted | 0.985 (0.923 to 1.052) | 0.654 |
| MR Egger | 0.990 (0.840 to 1.166) | 0.905 | ||
| Weighted median | 0.972 (0.882 to 1.072) | 0.573 | ||
| Weighted mode | 0.911 (0.714 to 1.161) | 0.452 | ||
| Sleep duration | 153 | Inverse variance weighted | 1.000 (0.999 to 1.000) | 0.316 |
| MR Egger | 1.001 (0.985 to 1.016) | 0.928 | ||
| Weighted median | 1.000 (0.999 to 1.001) | 0.844 | ||
| Weighted mode | 1.001 (0.998 to 1.004) | 0.593 | ||
| Snoring | 278 | Inverse variance weighted | 0.991 (0.903 to 1.087) | 0.847 |
| MR Egger | 0.823 (0.605 to 1.119) | 0.216 | ||
| Weighted median | 0.949 (0.827 to 1.089) | 0.456 | ||
| Weighted mode | 1.045 (0.686 to 1.590) | 0.839 | ||
| Sleep apnea syndrome | 141 | Inverse variance weighted | 1.122 (1.063 to 1.185) | 0.000 |
| MR Egger | 1.054 (0.930 to 1.195) | 0.409 | ||
| Weighted median | 1.089 (1.007 to 1.177) | 0.032 | ||
| Weighted mode | 1.032 (0.849 to 1.255) | 0.752 | ||
| Sleeplessness/insomnia | 223 | Inverse variance weighted | 1.465 (1.104 to 1.943) | 0.008 |
| MR Egger | 1.603 (0.702 to 3.661) | 0.264 | ||
| Weighted median | 1.330 (0.867 to 2.040) | 0.191 | ||
| Weighted mode | 1.198 (0.463 to 3.095) | 0.71 | ||
| Frequency of tiredness/lethargy in last 2 weeks | 224 | Inverse variance weighted | 1.725 (1.345 to 2.213) | 0.000 |
| MR Egger | 1.894 (0.902 to 3.978) | 0.093 | ||
| Weighted median | 1.799 (1.252 to 2.585) | 0.001 | ||
| Weighted mode | 2.400 (0.841 to 6.852) | 0.103 |
OR Odds ratio, 95%CI 95% Confidence Interval, IVW inverse-variance-weighted, NO. SNP the number of single nucleotide polymorphisms
Fig. 1.
Scatter plots of Mendelian randomization analyses. The black dot denotes the genetic instrumental variable included in the Mendelian randomization analysis. The grey error bar denotes the 95% confidence interval of the coefficient for each genetic instrumental variable. SNP, single nucleotide polymorphism
Table 3.
The reverse Mendelian randomization results of postpartum depression on risk of sleep traits
| Exposure.id | NO. SNPs | Outcomes.id | OR (95%CI) | P |
|---|---|---|---|---|
| postpartum depression (id: finn-b-O15_POSTPART_DEPR) | 73 | Dozing | ||
| Inverse variance weighted | 1.015 (0.980 to 1.052) | 0.399 | ||
| MR Egger | 1.028 (0.968 to 1.091) | 0.372 | ||
| Weighted median | 1.038 (0.979 to 1.101) | 0.206 | ||
| Weighted mode | 1.036 (0.964 to 1.114) | 0.335 | ||
| Ease of Getting up | ||||
| Inverse variance weighted | 0.008 (0.000 to 29,801.299) | 0.534 | ||
| MR Egger | 0.040 (0.000 to 2,419,051,640.271) | 0.801 | ||
| Weighted median | 0.022 (0.003 to 0.172) | 0.000 | ||
| Weighted mode | 0.007 (0.002 to 0.028) | 0.000 | ||
| Morningness | ||||
| Inverse variance weighted | 0.993 (0.967 to 1.020) | 0.611 | ||
| MR Egger | 1.009 (0.965 to 1.055) | 0.690 | ||
| Weighted median | 1.013 (0.966 to 1.062) | 0.601 | ||
| Weighted mode | 1.014 (0.952 to 1.080) | 0.671 | ||
| Napping | ||||
| Inverse variance weighted | 0.990 (0.975 to 1.006) | 0.211 | ||
| MR Egger | 0.982 (0.957 to 1.008) | 0.178 | ||
| Weighted median | 0.987 (0.967 to 1.008) | 0.226 | ||
| Weighted mode | 0.984 (0.958 to 1.012) | 0.262 | ||
| Sleep duration | ||||
| Inverse variance weighted | 1.007 (0.973 to 1.042) | 0.697 | ||
| MR Egger | 1.012 (0.946 to 1.083) | 0.735 | ||
| Weighted median | 0.985 (0.939 to 1.033) | 0.536 | ||
| Weighted mode | 0.949 (0.862 to 1.044) | 0.281 | ||
| Snoring | ||||
| Inverse variance weighted | − 6.986 (− 23.369 to 9.397) | 0.403 | ||
| MR Egger | − 11.778 (− 42.874 to 19.318) | 0.462 | ||
| Weighted median | − 8.464 (− 12.107 to − 4.821) | 0.000 | ||
| Weighted mode | − 10.827 (− 12.597 to − 9.056) | 0.000 | ||
| Sleep apnea syndrome | ||||
| Inverse variance weighted | − 6.479 (− 20.808 to 7.850) | 0.375 | ||
| MR Egger | − 9.142 (− 33.797 to 15.514) | 0.471 | ||
| Weighted median | − 9.568 (− 12.018 to − 7.118) | 0.000 | ||
| Weighted mode | − 9.973 (− 11.420 to − 8.526) | 0.000 | ||
| Sleeplessness / insomnia | ||||
| Inverse variance weighted | 0.006 (0.001 to 0.010) | 0.016 | ||
| MR Egger | 0.001 (− 0.007 to 0.010) | 0.726 | ||
| Weighted median | 0.006 (− 0.002 to 0.014) | 0.133 | ||
| Weighted mode | 0.006 (− 0.004 to 0.016) | 0.235 | ||
| Frequency of tiredness / lethargy in last 2 weeks | ||||
| Inverse variance weighted | 0.007 (0.002 to 0.011) | 0.004 | ||
| MR Egger | 0.000 (− 0.008 to 0.008) | 0.951 | ||
| Weighted median | 0.003 (− 0.004 to 0.010) | 0.427 | ||
| Weighted mode | 0.003 (− 0.005 to 0.012) | 0.462 | ||
OR Odds ratio, 95%CI 95% Confidence Interval, NO. SNP the number of single nucleotide polymorphisms
Sensitivity analysis using the leave-one-out method indicated that the observed associations of the sleep disturbances with PPD were not driven by single SNPs as shown in Supplementary file 2. The causal association still existed even when a single SNP was eliminated one by one and their MR results were robust. As Table 4 indicated, Cochran’s IVW Q test and MR-Egger showed no significant heterogeneity in SNP effects (All p > 0.05). To investigate the direction of the horizontal pleiotropy, MR-Egger intercepts, and MR-PRESSO global test were utilized, the results showed no horizontal pleiotropy among those exposure-outcome with significant association (All p > 0.05).
Table 4.
The analysis of heterogeneity and horizontal pleiotropy in the risk of sleep traits on postpartum depression
| Exposures | NO. SNPs | Cochrane’s Q | Pleiotropy | ||||||
|---|---|---|---|---|---|---|---|---|---|
| MR-Egger | IVW | MR-Egger | MR-PRESSO | ||||||
| Q | P | Q | P | Intercept | P | Globle test | P | ||
| Dozing | 97 | 97.034 | 0.423 | 97.209 | 0.446 | -0.003 | 0.453 | 99.215 | 0.679 |
| Ease of Getting up | 151 | 164.437 | 0.183 | 165.381 | 0.185 | 0.033 | 0.356 | 167.681 | 0.198 |
| Morningness | 195 | 206.003 | 0.248 | 210.233 | 0.202 | 0.056 | 0.048 | 212.597 | 0.195 |
| Napping | 130 | 121.935 | 0.634 | 121.940 | 0.658 | 0.000 | 0.948 | 123.832 | 0.657 |
| Sleep duration | 153 | 158.652 | 0.319 | 158.674 | 0.339 | -0.005 | 0.886 | 160.959 | 0.313 |
| Snoring | 278 | 291.719 | 0.247 | 293.345 | 0.239 | 0.006 | 0.216 | 295.599 | 0.235 |
| Sleep apnea syndrome | 141 | 151.807 | 0.216 | 153.088 | 0.212 | 0.006 | 0.281 | 155.257 | 0.218 |
| Sleeplessness/insomnia | 223 | 191.397 | 0.926 | 191.448 | 0.932 | -0.001 | 0.820 | 193.147 | 0.942 |
| Frequency of tiredness/lethargy in last 2 weeks | 224 | 180.574 | 0.981 | 180.642 | 0.983 | -0.001 | 0.794 | 182.369 | 0.983 |
IVW inverse-variance-weighted, NO. SNP the number of single nucleotide polymorphisms
Discussion
Based on the Mendelian randomization (MR) framework, our study has proved sleep disturbances cause the risk of PPD. More specifically, sleep apnea syndrome increases the risk of postpartum depression (PPD) by 1.122 times, sleeplessness/insomnia increases the risk of PPD by 1.465 times, and the frequency of tiredness/lethargy in the past 2 weeks increases the risk of PPD by 1.725 times. Additionally, PPD causes sleeplessness/insomnia and frequency of tiredness/lethargy in the past two weeks. The sensitivity analysis and assessment of horizontal pleiotropy both yielded favorable results, enhancing the robustness of the findings.
Our study confirms that sleep disturbances genetically increase the risk of PPD, in line with observational research [11, 12]. Among 116 postpartum women, their poor sleep quality is associated with higher depressive symptoms [11]. Our results suggest that individuals with pre-existing conditions such as obstructive sleep apnea or insomnia prior to childbirth are at an increased risk of developing postpartum depression. Previous studies have already demonstrated that factors such as frequent urination and difficulty repositioning at night during pregnancy contribute to over half of pregnant women reporting poor sleep quality. The prevalence of certain primary sleep disorders significantly increases during pregnancy, and insomnia is notably common. These sleep disturbances heighten the risk of postpartum depression [28, 29]. After childbirth, the mother's biological state is in a fragile state. However, tending to the newborn such as nighttime breastfeeding can exacerbate the mother's stress, leading to physical and mental exhaustion, which is a significant factor contributing to postpartum depression. Therefore, social support is widely recognized as a crucial buffering effect of sleep disturbances on postpartum depression[30, 31]. Research indicates that postpartum mothers' marital satisfaction and their perception of care from their mother-in-law can influence PPD and postpartum sleep through the mediation of social support [32]. Social isolation caused by external circumstances can exacerbate PPD and insomnia [33].
The mechanism underlying sleep disturbances that increase the risk of PPD is related to the synaptic homeostasis hypothesis, which posits that sleep is the price the brain pays for plasticity. One of the primary functions of sleep is learning, memory consolidation, and integration. Neuronal assemblies activated during learning will re-activate during sleep, participating in processing memory traces, thereby leading to plastic changes in the brain [34]. Neuroplasticity is dependent on synaptic and cellular homeostasis [35]. Once normal sleep is disrupted, it results in neuroplastic changes in the hippocampus and prefrontal cortex, providing a neural basis for the occurrence of PPD [36].
Sleep is a circadian-related biological activity, whose pattern may be altered by the episode of depression, including the symptoms of insomnia or hypersomnia. Research shows that ketamine combined with sleep deprivation treatment can effectively restore the dysregulation of diurnal rhythmicity caused by depression via resetting and stabilizing clock genes [37]. Disruption of sleep activates the immune system by the induction of a hormonal constellation through several inflammatory mediators, such as cytokines, which is the underlying mechanism of depression [38, 39]. Sleep disorders and inflammatory factor levels coexist in patients with depression [40]. In addition, sleep disturbances lead to PPD by inducing abnormal diurnal secretion of hormone levels, such as Ovarian hormones glucocorticoids and oxytocin, resulting in impaired hypothalamic–pituitary–adrenal axis negative feedback [36, 41–43]. Brain imaging studies suggest that the posterior cingulate may play a role in increasing the risk of sleep disturbances in postpartum depression. PPD can lead to altered functional connectivity between the posterior cingulate and amygdala [44]. Additionally, after a night of sleep deprivation, patients with depression exhibit significantly reduced activation in the cingulate [45].
The reverse MR suggests that PPD can cause sleep alteration, suggesting that patients with postpartum depression may present with symptoms of insomnia and fatigue. According to the diagnostic criteria of PPD [44], PPD is characterized by the onset of depressive symptoms that meet the diagnostic criteria for major depression disorder within the first two weeks after childbirth. In outpatient clinics, patients with depression often experience difficulties in falling asleep or early morning awakenings (insomnia). Sleep disturbances represent one of the numerous physiological symptom clusters, which may be present in the diagnosis of PPD. Certain genotypes expressed by postpartum depression may increase the risk of sleep disturbances such as TNFRSF17 [46]. In the TNF receptor family, sTNF-RI is associated with arousal during sleep [47]. Furthermore, postpartum depression causes sleep disturbances by environmental factors such as midnight feedings and soothing children.
Limitation
This study has several notable limitations. Firstly, the assessment of sleep traits lacks clarity regarding whether they are self-reported or clinically diagnosed, which may lead to discrepancies in interpretation. Additionally, our data predominantly come from European populations, raising concerns about the generalizability of our results to other demographic groups. Moreover, in analyzing the association between exposures and instrumental variables, we employed a stringent P-value threshold (P < 5 × 10−5) for SNPs, which allowed for the inclusion of more SNPs associated with exposure variables in the statistical analysis but may have diminished the statistical power of our causal estimates. Additionally, the potential confounding effects of sex differences in the insomnia GWAS may not be adequately addressed, limiting the generalizability of the findings to the female population specifically.
Conclusion
Genetic evidence suggests potential causal relationships between specific sleep traits and postpartum depression (PPD), including conditions like sleep apnea syndrome, insomnia, and excessive tiredness. It is important to explore whether these sleep health indicators indicate a risk for developing PPD or if they result from the disorder itself. Understanding this relationship could inform future research and interventions aimed at improving sleep quality among new mothers. If sleep disturbances are found to precede PPD, it may be beneficial to consider early interventions. Conversely, recognizing that PPD can lead to sleep issues could help in developing treatment strategies that address both mental and physical aspects of recovery. While these insights may contribute to maternal mental health care, further investigation is needed to clarify these connections and their implications for prevention and treatment of postpartum depression.
Supplementary Information
Acknowledgements
We express our gratitude to the United Kingdom Biobank for generously sharing GWAS pooled data with one Sleeplessness/insomnia dataset (id: ukb-a-13) and two Frequency of tiredness/lethargy in last 2 weeks datasets (id: ukb-a-245 and id: ukb-b-929). Additionally, we extend our appreciation to the participants and investigators of the FinnGen database for providing valuable GWAS pooled data on postpartum depression for inclusion in this research.
Abbreviations
- PPD
Postpartum depression
- MR
Mendelian randomization
- SNP
Single nucleotide polymorphism
- UKB
UK Biobank
- IVW
Inverse-variance-weighted
- OR
Odds Ratio
- 95%CI
95% Confidence Interval
Authors’ contributions
Chaoyan Yue and Yifeng Xu: Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization and Writing - review & editing. Qianying Hu and Enzhao Cong: Conceptualization, Funding acquisition, Project administration, Validation, Visualization, Writing - original draft, and Writing - review & editing. Jianhua Chen, Jingjing Ma, Luting Li: Methodology, Project administration, Resources, Writing - review & editing.
Funding
The study was funded by the Shanghai Planning Office of Philosophy and Social Science, Grant Number: 2019BSH012. The funders had no role in study design, data collection, data analysis, data interpretation, or writing of the manuscript.
Data availability
Data and materials will be available from https://gwas.mrcieu.ac.uk/. And the GWAS IDs are ukb-a-13, ukb-a-245, id: ukb-b-929 and finn-b-O15_POSTPART_DEPR respectively.
Declarations
Ethics approval and consent to participate
The present Mendelian randomization analysis relies on pooled data, and ethical approval has been duly obtained. Our research process complies with the data usage regulations of the IEU database and the Finnish database, meeting ethical requirements.
Consent for publication
All authors confirm that the article is the author’s original work. The article has not received prior publication and is not under consideration for publication elsewhere. All authors have seen and approved the manuscript being submitted.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Qianying Hu and Enzhao Cong contributed equally to this work and shared the first authorship.
Contributor Information
Yifeng Xu, Email: xuyifeng@smhc.org.cn.
Chaoyan Yue, Email: yuechaoyan@sina.com.
References
- 1.Wisner KL, Moses-Kolko EL, Sit DK. Postpartum depression: a disorder in search of a definition. Arch Womens Ment Health. 2010;13(1):37–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lee DT, Chung TK. Postnatal depression: an update. Best Pract Res Clin Obstet Gynaecol. 2007;21(2):183–91. [DOI] [PubMed] [Google Scholar]
- 3.Shorey S, et al. Prevalence and incidence of postpartum depression among healthy mothers: A systematic review and meta-analysis. J Psychiatr Res. 2018;104:235–48. [DOI] [PubMed] [Google Scholar]
- 4.Glasser S, Lerner-Geva L. Focus on fathers: paternal depression in the perinatal period. Perspect Public Health. 2019;139(4):195–8. [DOI] [PubMed] [Google Scholar]
- 5.Roomruangwong C, Epperson CN. Perinatal depression in Asian women: prevalence, associated factors, and cultural aspects. Asian Biomedicine. 2017;5(2):179–93. [Google Scholar]
- 6.Hahn-Holbrook J, Cornwell-Hinrichs T, Anaya I. Economic and Health Predictors of National Postpartum Depression Prevalence: A Systematic Review, Meta-analysis, and Meta-Regression of 291 Studies from 56 Countries. Front Psychiatry. 2017;8:248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Netsi E, et al. Association of Persistent and Severe Postnatal Depression With Child Outcomes. JAMA Psychiat. 2018;75(3):247–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bergink V, Rasgon N, Wisner KL. Postpartum Psychosis: Madness, Mania, and Melancholia in Motherhood. Am J Psychiatry. 2016;173(12):1179–88. [DOI] [PubMed] [Google Scholar]
- 9.Baglioni C, et al. Sleep and mental disorders: A meta-analysis of polysomnographic research. Psychol Bull. 2016;142(9):969–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dørheim SK, et al. Sleep and depression in postpartum women: a population-based study. Sleep. 2009;32(7):847–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Okun ML, et al. Poor sleep quality increases symptoms of depression and anxiety in postpartum women. J Behav Med. 2018;41(5):703–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wilson N, Lee JJ, Bei B. Postpartum fatigue and depression: A systematic review and meta-analysis. J Affect Disord. 2019;246:224–33. [DOI] [PubMed] [Google Scholar]
- 13.Li H, et al. Association between sleep disorders during pregnancy and risk of postpartum depression: a systematic review and meta-analysis. Arch Womens Ment Health. 2023;26(2):259–67. [DOI] [PubMed] [Google Scholar]
- 14.Marques M, et al. Is insomnia in late pregnancy a risk factor for postpartum depression/depressive symptomatology? Psychiatry Res. 2011;186(2–3):272–80. [DOI] [PubMed] [Google Scholar]
- 15.Insana SP, Montgomery-Downs HE. Sleep and sleepiness among first-time postpartum parents: a field- and laboratory-based multimethod assessment. Dev Psychobiol. 2013;55(4):361–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lewis BA, et al. The effect of sleep pattern changes on postpartum depressive symptoms. BMC Womens Health. 2018;18(1):12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Posmontier B. Sleep quality in women with and without postpartum depression. J Obstet Gynecol Neonatal Nurs. 2008;37(6):722–35; quiz 735–7. [DOI] [PMC free article] [PubMed]
- 18.Maghami M, et al. Sleep disorders during pregnancy and postpartum depression: A systematic review and meta-analysis. Int J Dev Neurosci. 2021;81(6):469–78. [DOI] [PubMed] [Google Scholar]
- 19.Skrivankova VW, et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA. 2021;326(16):1614–21. [DOI] [PubMed] [Google Scholar]
- 20.Bowden J, et al. A framework for the investigation of pleiotropy in two-sample summary data Mendelian randomization. Stat Med. 2017;36(11):1783–802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Smith GD, Ebrahim S. Mendelian randomization: prospects, potentials, and limitations. Int J Epidemiol. 2004;33(1):30–42. [DOI] [PubMed] [Google Scholar]
- 22.Li C, et al. Causal association between gut microbiota and intrahepatic cholestasis of pregnancy: mendelian randomization study. BMC Pregnancy Childbirth. 2023;23(1):568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sun Z, et al. Causal relationship between nonalcoholic fatty liver disease and different sleep traits: a bidirectional Mendelian randomized study. Front Endocrinol (Lausanne). 2023;14:1159258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44(2):512–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32(5):377–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21(11):1539–58. [DOI] [PubMed] [Google Scholar]
- 27.Bowden J, et al. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol. 2016;40(4):304–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Silvestri R, Aricò I. Sleep disorders in pregnancy. Sleep Sci. 2019;12(3):232–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Pavlova M, Sheikh LS. Sleep in women. Semin Neurol. 2011;31(4):397–403. [DOI] [PubMed] [Google Scholar]
- 30.Qi W, et al. Psychosocial risk factors for postpartum depression in Chinese women: a meta-analysis. BMC Pregnancy Childbirth. 2021;21(1):174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Heh SS. Relationship between social support and postnatal depression. Kaohsiung J Med Sci. 2003;19(10):491–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Qi W, et al. Effects of family relationship and social support on the mental health of Chinese postpartum women. BMC Pregnancy Childbirth. 2022;22(1):65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Miranda AM, et al. Triggering of postpartum depression and insomnia with cognitive impairment in Argentinian women during the pandemic COVID-19 social isolation in relation to reproductive and health factors. Midwifery. 2021;102:103072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Maquet P. The role of sleep in learning and memory. Science. 2001;294(5544):1048–52. [DOI] [PubMed] [Google Scholar]
- 35.Tononi G, Cirelli C. Sleep and the price of plasticity: from synaptic and cellular homeostasis to memory consolidation and integration. Neuron. 2014;81(1):12–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Brummelte S, Galea LA. Postpartum depression: Etiology, treatment and consequences for maternal care. Horm Behav. 2016;77:153–66. [DOI] [PubMed] [Google Scholar]
- 37.Bunney BG, et al. Circadian dysregulation of clock genes: clues to rapid treatments in major depressive disorder. Mol Psychiatry. 2015;20(1):48–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Chen X, et al. The clock-controlled chemokine contributes to neuroinflammation-induced depression. Faseb j. 2020;34(6):8357–66. [DOI] [PubMed] [Google Scholar]
- 39.Besedovsky L, Lange T, Haack M. The Sleep-Immune Crosstalk in Health and Disease. Physiol Rev. 2019;99(3):1325–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Carhart-Harris RL. Serotonin, psychedelics and psychiatry. World Psychiatry. 2018;17(3):358–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Gervais NJ, Mong JA, Lacreuse A. Ovarian hormones, sleep and cognition across the adult female lifespan: An integrated perspective. Front Neuroendocrinol. 2017;47:134–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Raymond JS, et al. The influence of oxytocin-based interventions on sleep-wake and sleep-related behaviour and neurobiology: a systematic review of preclinical and clinical studies. Neurosci Biobehav Rev. 2021;131:1005–26. [DOI] [PubMed] [Google Scholar]
- 43.Szmyd B, et al. The impact of glucocorticoids and statins on sleep quality. Sleep Med Rev. 2021;55:101380. [DOI] [PubMed] [Google Scholar]
- 44.Chase HW, et al. Disrupted posterior cingulate-amygdala connectivity in postpartum depressed women as measured with resting BOLD fMRI. Soc Cogn Affect Neurosci. 2014;9(8):1069–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Clark CP, Frank LR, Brown GG. Sleep deprivation, EEG, and functional MRI in depression: preliminary results. Neuropsychopharmacology. 2001;25(5 Suppl):S79–84. [DOI] [PubMed] [Google Scholar]
- 46.Mehta D, et al. Genome-wide gene expression changes in postpartum depression point towards an altered immune landscape. Transl Psychiatry. 2021;11(1):155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Yue HJ, et al. The roles of TNF-alpha and the soluble TNF receptor I on sleep architecture in OSA. Sleep Breath. 2009;13(3):263–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data and materials will be available from https://gwas.mrcieu.ac.uk/. And the GWAS IDs are ukb-a-13, ukb-a-245, id: ukb-b-929 and finn-b-O15_POSTPART_DEPR respectively.