Edited by: Richard Eugene Frye, Children’s Hospital Boston and Harvard University, USA
Reviewed by: Norma Beatriz Ojeda, University of Mississippi Medical Center, USA; Barbara Fuhrman, University of Arkansas for Medical Sciences, USA
This article was submitted to Child Health and Human Development, a section of the journal Frontiers in Public Health.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
The purpose of this study was to examine variations in endogenous oxytocin levels in pregnancy and postpartum state. We also explored the associations between delivery variables and oxytocin levels. A final sample of 272 mothers in their first trimester of pregnancy was included for the study. Blood samples were drawn during the first trimester and third trimester of pregnancy and at 8 weeks postpartum. Socio-demographic data were collected at each time point and medical files were consulted for delivery details. In most women, levels of circulating oxytocin increased from the first to third trimester of pregnancy followed by a decrease in the postpartum period. Oxytocin levels varied considerably between individuals, ranging from 50 pg/mL to over 2000 pg/mL. Parity was the main predictor of oxytocin levels in the third trimester of pregnancy and of oxytocin level changes from the first to the third trimester of pregnancy. Oxytocin levels in the third trimester of pregnancy predicted a self-reported negative labor experience and increased the chances of having an epidural. Intrapartum exogenous oxytocin was positively associated with levels of oxytocin during the postpartum period. Our exploratory results suggest that circulating oxytocin levels during the third trimester of pregnancy may predict the type of labor a woman will experience. More importantly, the quantity of intrapartum exogenous oxytocin administered during labor predicted plasma oxytocin levels 2 months postpartum, suggesting a possible long-term effect of this routine intervention, the consequences of which are largely unknown.
Pregnancy is a unique state that is accompanied by numerous biological and psychological changes. While there are known changes in the levels of hormones such as progesterone and estrogen in pregnant women (
In non-human primates, no reliable pattern of variation of oxytocin during pregnancy or the postpartum period has been observed, but a fourfold increase on the day of delivery has been reported (
In humans, different methods for measuring oxytocin during pregnancy and labor have been used including radioimmunoassay and enzyme immunoassay, and a variety of results have been reported. Some studies have found that oxytocin levels increase at the onset of labor and during labor compared to 1 or 2 weeks before labor (
Given the lack of clarity concerning oxytocin in pregnancy and the postpartum period, the present longitudinal study aimed (a) to describe the range of oxytocin levels in a larger sample of pregnant women than had been studied hitherto; (b) to estimate plasma oxytocin variations during pregnancy and the postpartum period; and (c) to explore the relation between oxytocin levels and labor outcomes. Thus, our main objective was to describe how plasma oxytocin levels vary in a large cohort of pregnant women. Our secondary interest was to raise, or confirm, hypotheses about oxytocin correlates by exploring its association with a number of labor variables available in our cohort.
A total of 272 women were tested. We recruited 342 pregnant women during their first trimester of pregnancy at an obstetrics clinic in a general hospital (
The study was approved by the Jewish General Hospital Research Ethics Committee and the Maison de Naissance Côtes des Neiges. A recruiter approached women during their first prenatal visit or during a general information session. Women meeting all of the inclusion criteria and willing to take part in the study gave written informed consent and were tested the same day. A second meeting took place during their third trimester of pregnancy, and a third meeting took place at the mothers’ homes 8 weeks postpartum. At each meeting, women completed a series of questionnaires described below in English or French, and blood samples (10 mL) were then drawn by a nurse or a midwife. Women received $25 per visit. Blood samples were drawn between 9 a.m. and 5 p.m., as soon as the nurse was available or as soon as the meeting with the doctor was over. At the postpartum session, if the mother was breastfeeding or bottle-feeding when the research assistant and nurse arrived, blood samples were drawn at least 30 min after the completion of feeding. Data were collected between July 2009 and September 2011.
Data were extracted from participants’ medical charts. Variables included: length of pregnancy; use of an epidural anesthesia; quantity of intrapartum exogenous oxytocin administered (for induction or augmentation of labor), for mothers who gave birth at the hospital; route of delivery (vaginal or cesarean); and duration of labor. The type of cesarean delivery was classified, according to Lucas and colleagues’ definition (
Data about age, education level, marital status, parity, and previous miscarriages were collected from a self-report questionnaire completed by every participant at her first visit.
Here, somatization is defined as the experience of physical symptoms most likely due to mental factors. One hypothesis is that oxytocin might modulate the link between these mental factors and their physical symptoms through its influence on the immune system, observed in humans (
An interview, conducted through phone at about 3 weeks postpartum (mean = 26 days, SD = 15), evaluated whether the women had a very negative, somewhat negative, neither negative nor positive, somewhat positive, or very positive labor experience.
Since oxytocin is known to play a role in milk ejection during lactation (
Blood was collected from participants in a heparinized tube, stored on ice until it could be centrifuged at 1600 ×
Oxytocin levels were measured using a commercially available enzyme-linked immunosorbent assay (EIA: Enzo Life Sciences Inc.). Direct measurement was performed following the manufacturer’s protocols and samples were diluted 1:2 or 1:4 as described previously (
We first compiled descriptive statistics for all variables collected in the study, including means, standard deviations, medians, and inter-quartile ranges and proportions, as appropriate. We next calculated correlation coefficients to examine, which variables were associated with oxytocin, and with each other, to search for possible confounding variables. All variables associated with oxytocin levels were included in the models exploring the relationship between oxytocin levels and labor or postpartum variables.
Three oxytocin measurements were taken for each participant. It is important to note that previous research has suggested that oxytocin levels measured in three repeated blood samples collected from 65 participants at intervals of 5 min showed greater inter-individual variation than intra-individual variation, with an intraclass correlation coefficient suggesting modest reliability (personal communication from Elizabeth Hoge). The intraclass correlation coefficient indicates how much agreement there is between two or more repeated measures. Even though this agreement is moderate (
Linear and logistic regression models were used to estimate the best predictors of oxytocin levels at each time point, and which outcomes oxytocin levels predicted. We included the following variables: parity, location of delivery, transfer from the birth center to the hospital, epidural, duration of labor, categories of cesarean birth, somatoform symptoms, positive experience of labor, negative experience of labor, intrapartum exogenous oxytocin administered, breastfeeding at 2 months postpartum, change of oxytocin levels from the first to the third trimester, oxytocin levels during the first trimester, during the third trimester, and during the postpartum period. We ran several models, as displayed in Table
The longitudinal models were fit using WinBUGS software (version 1.4.3, MRC Biostatistics Unit, Cambridge, UK). Other models were run using R (version 2.12, R Development Core Team, 2012).
Descriptive statistics are shown in Table
Mean (standard deviation) | |
---|---|
Age (years) | 31.6 (4.6) |
Education (years) | 15.3 (2.6) |
Length of pregnancy (weeks, |
39.6 (1.2) |
Duration of labor (h, |
8.6 (6.7) |
Single | 4.4 |
Married | 68.0 |
Living with partner | 25.4 |
Divorced/separated | 2.2 |
0 | 47 |
1 | 38 |
2 | 10 |
3 or + | 5 |
Vaginal | 74 |
Vaginal after cesarean birth | 4 |
Elective cesarean birth |
9 |
Scheduled cesarean birth |
7 |
Urgent cesarean birth |
5 |
Emergency cesarean birth |
1 |
0 | 71 |
1 | 24 |
2 or + | 5 |
Figure
Our hierarchical model estimated that oxytocin levels increased from T1 to T2 [change of 0.60 pg/mL estimated per day, CI = (0.41; 0.79)] and decreased from T2 to T3 [change of −1.03 pg/mL estimated per day, CI = (−1.30; −0.76)]. The standard deviations of the slopes between T1 and T2 and between T2 and T3 were very large [T1–T2: mean SD = 1.57, CI = (1.45; 1.71); T2–T3: mean SD = 2.26, CI = (2.07; 2.46)]. This suggests that this pattern of positive slope from T1 to T2 and negative slope from T2 to T3 was not followed by all women. Indeed, as shown in Figure
The four groups defined above were compared for labor duration, somatoform symptoms, intrapartum exogenous oxytocin, and experience of labor, parity, and route of delivery (Table
Continuous variables | Rise–rise group |
Rise–fall group |
Fall–rise group |
Fall–fall group |
||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Mean | SD | Mean | SD | Mean | SD | Mean | SD | |||||
Labor duration (h) | 18 | 10.51 | 6.48 | 153 | 8.65 | 5.24 | 32 | 6.28 | 3.96 | 26 | 9.29 | 7.01 |
Somatoform symptoms T1 | 23 | 4.96 | 3.08 | 176 | 5.28 | 2.80 | 42 | 4.81 | 2.17 | 31 | 5.61 | 2.69 |
Somatoform symptoms T2 | 23 | 5.74 | 2.77 | 175 | 5.73 | 2.67 | 40 | 6.50 | 2.74 | 31 | 6.97 | 2.39 |
Exogenous oxytocin (IU) | 23 | 1.90 | 4.02 | 175 | 1.23 | 2.52 | 42 | 0.58 | 1.85 | 31 | 3.02 | 6.29 |
Cesarean birth | 23 | 13.0 | 175 | 5.1 | 42 | 4.8 | 31 | 6.5 | ||||
Epidural | 23 | 60.9 | 175 | 61.7 | 42 | 45.2 | 31 | 51.6 | ||||
Negative experience labor | 22 | 13.6 | 175 | 11.4 | 41 | 4.9 | 31 | 19.4 | ||||
Positive experience labor | 22 | 81.8 | 175 | 72.6 | 41 | 87.8 | 31 | 61.3 | ||||
First child | 23 | 43.5 | 176 | 51.1 | 42 | 66.7 | 31 | 48.4 |
To explore the associations between oxytocin levels and the other variables, we ran a series of multivariate models, summarized in Table
Parity | ||||
−33.853 (−59.345, −8.360) | ||||
Parity | ||||
−29.943 (−62.745, 2.860) | ||||
Exogenous oxytocin | ||||
15.592 (5.733, 25.452) | ||||
Epidural | Parity | |||
1.787 (1.038, 2.535) | −0.529 (−0.890, −0.168) | |||
Parity | Center | Exogenous oxytocin | Transfer to hospital | |
−1.456 (−2.021, −0.891) | −4.417 (−5.8525, −3.009) | 0.598 (0.430, 0.766) | 3.985 (1.990, 5.979) | |
Cesarean birth | Exogenous oxytocin | Oxytocin T2 | ||
17.813 (4.448, 71.336) | 1.296 (1.132, 1.484) | 1.002 (1.000, 1.003) | ||
Cesarean birth | Exogenous oxytocin | |||
0.790 (0.701, 0.889) | 0.081 (0.020, 0.323) | |||
Center | Exogenous oxytocin | Oxytocin T1 | Oxytocin T2 | |
0.116 (0.061, 0.223) | 1.478 (1.198, 1.824) | 0.997 (0.995, 0.999) | 1.003 (1.001, 1.004) | |
Center | Exogenous oxytocin | Oxytocin change T1 to T2 | ||
0.117 (0.061, 0.225) | 1.456 (1.181, 1.795) | 1.003 (1.001, 1.004) | ||
Center | ||||
7.653 (1.778, 32.940) |
Oxytocin levels at the third trimester of pregnancy and changes of oxytocin levels from the first to third trimester of pregnancy, were predicted by parity, with greater parity predicting lower oxytocin and a smaller variation in oxytocin between the first and third trimester. Postpartum oxytocin levels (T3) were predicted by the quantity of intrapartum exogenous oxytocin received, with more intrapartum oxytocin predicting higher plasma oxytocin levels at 2 months postpartum. There were no other significant predictors of plasma oxytocin levels at T3.
A “very negative” or “somewhat negative” labor was predicted by oxytocin at the third trimester, the type of cesarean delivery, and the quantity of intrapartum oxytocin received. In contrast, only intrapartum exogenous oxytocin and cesarean birth type were relevant variables explaining positive labor scores. Having an epidural was predicted by oxytocin levels at the first and third trimester (or oxytocin changes from first to third trimester), the birth location (hospital or midwife clinic), and the quantity of intrapartum exogenous oxytocin received.
Only 30 women had stopped breastfeeding at 2 months postpartum. They stopped breastfeeding at around 7 weeks postpartum (
We measured plasma oxytocin levels in healthy women at three time points: during the first and third trimesters of pregnancy and 2 months postpartum. There were large inter-individual differences in oxytocin levels, with some women having less than 50 pg/mL and others having more than 2000 pg/mL (Figure
Animal studies suggest that oxytocin levels rise as delivery approaches, in part because oxytocin has a contractile function on the uterine muscles (
It is unclear why some authors have found stable levels across pregnancy. As mentioned earlier, oxytocin has a very short half-life and is found in comparatively low concentrations, which makes it difficult to measure in blood (
Concerning the relationship between oxytocin levels and the other variables, it is important to keep in mind that all analyses were exploratory and are thus to be considered carefully, as they need to be replicated. In general, the women who were pregnant with their first child had higher oxytocin levels in the third trimester and showed a larger increase in oxytocin levels from the first to the third trimester compared to the mothers who already had one or more children. Parity also predicted the duration of labor as well as the quantity of intrapartum oxytocin administered to the mothers. The association of parity with the length of labor has long been known (
While there was no direct association between oxytocin levels and the duration of labor, the fact that first-time mothers underwent a greater rise in oxytocin from the first to the third trimester and exhibited higher levels of oxytocin in the third trimester may have a functional significance. Oxytocin may mediate some of the physiological changes that are required to give birth for the first time, and these changes may leave the body permanently sensitized to oxytocin. Subsequent pregnancies may thus require a smaller increase in oxytocin.
Four additional findings emerge from this study. The first is that both the increase in oxytocin levels from the first to the third trimester, as well as the levels of oxytocin at both of those times, predicted the probability of epidural use during delivery. The second finding is that high oxytocin levels at the third trimester of pregnancy predicted a negative experience of labor. Taken together, these findings highlight the possibility of using oxytocin levels and their variations to predict some of the features of labor.
The third finding is that women who had stopped breastfeeding when interviewed at 2 months postpartum had higher oxytocin levels than those who were still breastfeeding. This result is at odds with previous reports, where higher plasma oxytocin levels were observed for breastfeeding women compared to non-breastfeeding women (
The last finding is that the quantity of intrapartum exogenous oxytocin administered just before delivery predicted oxytocin levels at 2 months postpartum. This is contrary to Jonas and colleagues’ findings (
Togas Tulandi is a clinical investigator funded by HALT Medical and advisor for Watson Pharma. All other authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The study was supported by a CIHR Emerging Team Grant in Perinatal Mental Health. Nancy Feeley is supported by a Research Scholar Award from the Fonds de Recherche du Québec – Santé (FRQS). We thank Elizabeth A. Hoge for her help finalizing the present manuscript.