| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
Department of Obstetrics and Gynaecology (T.K.H.C., L.T.K., Y.S.K.), and the Department of Psychiatry (H.F.K.C., T.S.L.), The Chinese University of Hong Kong, Prince of Wales Hospital, Sha Tin, NT, Hong Kong, and the Department of Social Medicine (T.S.L.), Harvard Medical School, Boston, Massachusetts.
Address reprint requests to: Tony Chung, MD, FRANZCOG, Department of Obstetrics and Gynaecology, Prince of Wales Hospital, Sha Tin, NT, Hong Kong. Email: tonychung{at}cuhk.edu.hk
| ABSTRACT |
|---|
|
|
|---|
METHOD: In a prospective observational study, 959 women were followed up longitudinally from early pregnancy to postpartum. The level of depression was measured at baseline (first antepartum visit) and in late pregnancy using the Beck Depression Inventory (BDI). Adverse obstetric and neonatal outcomes were recorded at delivery.
RESULTS: Depression in late pregnancy was associated with increased risk of epidural analgesia (33% vs. 19%, p = .01, adjusted RR = 2.56, 95% CI 1.245.30), operative deliveries (caesarean sections and instrumental vaginal deliveries) (39% vs. 27%, p = .02, adjusted RR = 2.28, 95% CI 1.154.53), and admission to neonatal care unit (24% vs. 19%, p = .03, adjusted RR = 2.18, 95% CI 1.024.66). These effects remained significant even when controlled for potential confounders, such as antepartum complications.
CONCLUSION: Previous studies have shown that antepartum anxiety or stress was associated with growth retardation, premature delivery, and epidural analgesia. Our findings add to this body of evidence, which together suggest an adverse impact of antepartum psychological morbidity on maternal and neonatal well-being.
Key Words: depression pregnancy psychological complications obstetric outcomes epiduralanalgesia operative deliveries neonatal complications prevention
Abbreviations: BDI = Beck Depression Inventory; NCU = Neonatal Care Unit; CPD = cephalopelvic disproportion
| INTRODUCTION |
|---|
|
|
|---|
Some specific psychotic conditions, especially bipolar disorders, have an appreciable impact on perinatal outcomes (8). However, there is relatively little information on the effect of less severe antepartum psychological morbidity, such as depression, on obstetric outcomes. This is somewhat surprising given that the risks of unrecognized and untreated depression include poor nutrition, lack of volition to follow medical and antepartum care recommendations, and increased alcohol and substance abuse, all of which would be expected to impact adversely on perinatal outcomes (9, 10). Furthermore, psychological illnesses, especially relatively silent ones such as depression, may adversely affect the decision-making capacity of the pregnant woman, without this being recognized by the attending obstetrician or midwife (11).
A few studies have examined the adverse effects of anxiety in pregnancy. Teixeira et al. showed that maternal anxiety in pregnancy was associated with increased uterine artery resistance (12). Other studies have demonstrated that babies of stressed or anxious mothers have a significantly lower average birth weight for gestational age and tend to be born early (1317). Using ultrasonography, one study found that the fetus of anxious mothers spent significantly more time in quiet sleep and exhibited less gross body movement when in active sleep (18). There is a one study of the effects of anxiety on maternal well-being, which showed that anxious mothers were more likely to receive epidural analgesia (19). These results appear to suggest that anxiety adversely impacts both maternal and fetal well-being.
There is one report of antepartum depression being associated with growth retardation (20). However, in this and many other studies, potential confounding factors have not been adequately controlled for. In some studies, this was probably related to the small sample size. The present study sought to address these shortcomings by examining a broad array of potential confounders in a larger sample. In addition, it examined whether depression, as an indicator of maternal psychological distress, affects maternal obstetric and neonatal outcomes. As antepartum depressive illness is far more commonly diagnosed than anxiety disorders in pregnancy, any positive findings would have a broader clinical implication.
| METHODS |
|---|
|
|
|---|
Design and Instruments
At baseline (first antepartum visit), we collected socio-demographic and clinical data using a semi-structured interview. To assess the level of depression, we also handed the Beck Depression Inventory (BDI) to the participants. At 32 weeks of pregnancy, we repeated the depression assessment by mailing the BDI (with stamped return envelope) to the participants. Women who did not return the questionnaires after two weeks were reminded by phone. If the first reminder failed, the participants received a second reminder call at 36 weeks of pregnancy. The BDI is a 21-item self-report rating scale designed to measure the severity of depression (21). The reliability and validity of the BDI, including the Chinese version, have been established in previous studies (22, 23).
Following childbirth, two research nurses visited the participants in the postpartum ward to collect details of obstetric outcomes from the participants, the medical notes, and the computerized database. As some participants delivered at other hospitals, the research nurses contacted these participants by phone to collect data on their obstetric outcomes. A broad array of obstetric outcomes was assessed, including preterm labor, duration of labor, induction of labor, use of analgesia, mode of delivery, meconium stained liquor, fetal distress, birth weight, infant asphyxiation score, and admission to the neonatal care unit (NCU). Potential confounders of the obstetric outcomes, such as antepartum medical complications, were also recorded (see statistical analysis for the complete list).
We obtained approval from the Institutional Review Board to conduct this study.
Statistical Analysis
Participants with no labor (eg, elective caesarean section) were removed from the analysis because their obstetric outcomes were substantially confounded by the conditions that led to no labor. The characteristics of the remaining participants were summarized using descriptive statistics. The participants were stratified by a BDI cutoff score of 14.5, and the relative risks of specific obstetric outcomes occurring among women who scored high on the BDI (baseline or third trimester) were calculated. Using logistic regression with the BDI score status (high score vs. low score) as an independent variable, we adjusted the score for potential confounders, which included age, parity, gestation, past psychiatric history, medical complications (maternal diabetes, gestational hypertension, thyroid disorders, antepartum hemorrhage), male fetus, actual birth weight, low birth weight, induced labor, labor augmentation, fetal presentation, and previous caesarean section. Mode of delivery was also adjusted for when estimating the relative risk of NCU admission.
| RESULTS |
|---|
|
|
|---|
The mean third trimester BDI score was 6.6 (SD 6.3). Sixty-seven participants (10%) scored 15 or above on the BDI, 26 (4%) scored 20 or above, and 10 (1%) scored greater than 29.
Among the participants, 49 (8%) had gestational diabetes mellitus, 5 (0.8%) had hypertension, and 11 (2%) had thyroid dysfunction or goiter during the index pregnancy. Sixteen women (3%) experienced antepartum hemorrhage in the index pregnancy. Seventy-eight participants (13%) had labor induced and 150 participants (22%) required emergency caesarean section. The commonest reasons for emergency caesarean section fell under two categories: cephalopelvic disproportion/lack of progress in labor/failed induction of labor (29%), and suspected fetal distress (22%). The mean gestational age at birth was 39 weeks (SD 1.6, range 2842) and there were 34 (5%) women with preterm labor. The mean birth weight was 3206 g (SD 480), and 72 infants (10%) had birth weight lower than the gestation-specific 10th centile. One hundred and twenty-three neonates were admitted to the NCU for the following reasons: respiratory distress (17%), hypoglycemia (16%), prematurity and related problems (15%), maternal diseases (15%), infection risk (13%), others (26%).
The baseline depression scores were not associated with any of the obstetric outcomes. The third trimester BDI scores, however, were significantly associated with three obstetric outcomes (Table 1). Women who had elevated BDI in late pregnancy had increased risk of epidural analgesia (33% vs. 19%, p = .01, adjusted RR = 2.56, 95% CI 1.245.30), operative deliveries (caesarean sections and instrumental vaginal deliveries) (39% vs. 27%, p = .02, adjusted RR = 2.28, 95% CI 1.154.53), and their infants were more likely to be admitted to NCU after delivery (24% vs. 19%, p = .03, adjusted RR = 2.18, 95% CI 1.024.66).
|
Since both epidural analgesia and operative deliveries are associated with antepartum depression, and epidural analgesia is well known to be associated with operative deliveries, we attempted to examine the role of the latter effect in the association between antepartum depression and operative deliveries. When epidural analgesia was included as one of the independent variables in the logistic regression of operative deliveries, the effect of antepartum depression disappeared. Thus, there was no independent association between antepartum depression and operative deliveries.
| DISCUSSION |
|---|
|
|
|---|
Our findings echo the preliminary evidence that antepartum anxiety is associated with epidural analgesia (19). It appears that women who are more anxious or depressed are more likely to need and receive epidural analgesia. While the exact psychopathophysiology remains to be elucidated, we propose a few possibilities. First, it is possible that the negative effect, be it from anxiety or depression, may reduce pain threshold, magnify pain perception, and amplify physical and psychological distress. Second, as anxiety has been shown to impair uterine contractility (19), and there is a strong correlation between depressive and anxiety affect, it is possible that antepartum depression increases the risk of operative delivery through this effect. Among our participants who needed emergency caesarean section, the commonest recorded reasons were under the category of lack of progress, failed induction of labor and cephalopelvic pelvic disproportion (29%). Finally, health care professionals may, in conscious and unconscious manner, respond to the depression symptomology and or "amplified" distress by being more ready to opt for epidural analgesia and or operative deliveries.
We feel that epidural analgesia plays a central role in mediating between antepartum depression and operative delivery. When the statistical effects of epidural analgesia are controlled for in the statistical modeling, antepartum depression symptomology ceased to be a significant explanatory factor for operative delivery. This is not surprising, given that there is an established relationship between epidural analgesia and operative deliveries (24). To elucidate these intriguing relationships, a series of carefully designed studies would be needed.
Other confounding factors may also be responsible for the observed associations. For instance, women who suffer depression during pregnancy may be prescribed antidepressants or antianxiety drugs, which in turn increase the risk of neonatal admissions to intensive care unit. Depressed women are also more likely to smoke or to abuse drugs. Yet, we feel that these confounders are unlikely among the study participants. This is because there is substantial proscription in Chinese culture against medications, alcohol, or smoking during pregnancy, believing that the fetus is very vulnerable to environmental insults (25). Hence, our data, which were routinely collected in all pregnant women, indicated that none of the participants smoked and none were prescribed psychotropic medications during pregnancy.
The link between antepartum depression and admission to NCU is more difficult to explain. It remained significant after logistic regression. The neonatal staff, who had the admission discretion for the unit, were unaware of the BDI scores of the study participants as were the obstetricians. Usual indications such as prematurity, respiratory distress, and maternal disease such as diabetes and hypertension largely governed the admissions. How a high BDI score can interact with these indications is unclear. This observation, if repeated, warrants further investigation.
There are several limitations to the study. About 20% of the participants did not return the postal questionnaires and another 7% had to be excluded because of absence of labor. While the latter is unavoidable, these attritions nonetheless reduce the generalizability of the findings. This study did not have a large enough sample to test the hypothesis that maternal depression has an effect on infant birth weight. Given the current findings, we estimated that about 6000 participants would be needed in order to have good statistical power. We feel that replication in other populations is warranted as our participants were of only one ethnicity.
We hope our findings will stimulate further research that would yield empirical evidence to support better detection and screening of antepartum psychopathology. At present, despite the fact that empirical data indicate a tight correlation between antepartum depression and postpartum depression (1, 26), the latter attracts far more attention from the lay public and medical profession. This is probably because postnatal depression has been popularized to describe distress associated with childbearing. The research evidence on the detrimental effects of postnatal depression on child emotional and cognitive development also lends strong support to a proactive policy to detect depression in postpartum. In contrast, as described earlier, comparable studies in antepartum almost always conceptualize psychological distress as stress or anxiety. While such constructs are valid indicators of psychological distress, they cannot be readily translated into clinical practice: stress and anxiety are often regarded as normative in childbearing. This nontranslatability is reflected by the fact that while it is theoretically feasible to diagnose antepartum anxiety or postnatal anxiety disorders, such diagnosis is rarely made in clinical reality or research context. We hope, by switching to a more familiar indicator of peripartum distress, ie, depression, this study will help clinicians and policy makers to appreciate the health impact and the potential medical cost of unrecognized or untreated antepartum psychological morbidity. This hopefully will initiate a new discourse, which would eventually lead to a proactive treatment policy for antepartum psychopathology.
| ACKNOWLEDGMENTS |
|---|
|
|
|---|
Received for publication May 23, 2000.
Revision received December 15, 2000.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. M. Lee, S. K. Lam, S. M. Sze Mun Lau, C. S. Y. Chong, H. W. Chui, and D. Y. T. Fong Prevalence, Course, and Risk Factors for Antenatal Anxiety and Depression Obstet. Gynecol., November 1, 2007; 110(5): 1102 - 1112. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. A. Kendall-Tackett Violence Against Women and the Perinatal Period: The Impact of Lifetime Violence and Abuse on Pregnancy, Postpartum, and Breastfeeding Trauma Violence Abuse, July 1, 2007; 8(3): 344 - 353. [Abstract] [PDF] |
||||
![]() |
J. Evans, J. Heron, R. R. Patel, and N. Wiles Depressive symptoms during pregnancy and low birth weight at term: Longitudinal study The British Journal of Psychiatry, July 1, 2007; 191(1): 84 - 85. [Abstract] [Full Text] [PDF] |
||||
![]() |
Antidepressants During Pregnancy: Treating the Condition While Acknowledging the Risks Journal Watch Women's Health, April 26, 2007; 2007(426): 1 - 1. [Full Text] |
||||
![]() |
G. M Thormahlen Paroxetine Use During Pregnancy: Is it Safe? Ann. Pharmacother., October 1, 2006; 40(10): 1834 - 1837. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. R. Rubinow Antidepressant Treatment During Pregnancy: Between Scylla and Charybdis Am J Psychiatry, June 1, 2006; 163(6): 954 - 956. [Full Text] [PDF] |
||||
![]() |
F. A. Carter, C. M. A. Frampton, and R. T. Mulder Cesarean section and postpartum depression: a review of the evidence examining the link. Psychosom Med, March 1, 2006; 68(2): 321 - 330. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. Haddad, B. R. Pal, P. Clarke, A. Wieck, and S. Sridhiran Neonatal symptoms following maternal paroxetine treatment: Serotonin toxicity or paroxetine discontinuation syndrome? J Psychopharmacol, September 1, 2005; 19(5): 554 - 557. [Abstract] [PDF] |
||||
![]() |
C. Larsson, G. Sydsjo, and A. Josefsson Health, Sociodemographic Data, and Pregnancy Outcome in Women With Antepartum Depressive Symptoms Obstet. Gynecol., September 1, 2004; 104(3): 459 - 466. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Andersson, I. Sundstrom-Poromaa, M. Wulff, M. Astrom, and M. Bixo Implications of Antenatal Depression and Anxiety for Obstetric Outcome Obstet. Gynecol., September 1, 2004; 104(3): 467 - 476. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Andersson, I. Sundstrom-Poromaa, M. Wulff, M. Astrom, and M. Bixo Neonatal Outcome following Maternal Antenatal Depression and Anxiety: A Population-based Study Am. J. Epidemiol., May 1, 2004; 159(9): 872 - 881. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. A. Bennett, A. Einarson, A. Taddio, G. Koren, and T. R. Einarson Prevalence of Depression During Pregnancy: Systematic Review Obstet. Gynecol., April 1, 2004; 103(4): 698 - 709. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Green, K. Kafetsios, H. E. Statham, and C. M. Snowdon Factor Structure, Validity and Reliability of the Cambridge Worry Scale in a Pregnant Population J Health Psychol, November 1, 2003; 8(6): 753 - 764. [Abstract] [PDF] |
||||
![]() |
M. G. Spinelli and J. Endicott Controlled Clinical Trial of Interpersonal Psychotherapy Versus Parenting Education Program for Depressed Pregnant Women Am J Psychiatry, March 1, 2003; 160(3): 555 - 562. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. N. Schnyer, R. Manber, and A. J. Fitzcharles Acupuncture Treatment for Depression During Pregnancy: Conceptual Framework and Two Case Reports Complementary Health Practice Review, January 1, 2003; 8(1): 40 - 53. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |