The Relationship of the Administration of Intrapartum Synthetic Oxytocin and Breastfeeding Initiation and Duration Rates

Abstract

Aim:

The consequences that intrapartum administration of hormones can have on breastfeeding are unclear. The aim of the study is to determine if synthetic intrapartum oxytocin, used routinely for induction/stimulation, has a relationship to initiation/duration of breastfeeding.

Patients and Methods:

We conducted a cohort study that was carried out in a tertiary university hospital distinguished by WHO-UNICEF as a BFHI (Baby-Friendly Hospital Initiative). A group of 53 mother and newborn dyads who had been exposed to intrapartum synthetic oxytocin were compared with 45 nonexposed dyads. A breastfeeding questionnaire was administered by a midwife blind to patient group through phone calls 3 and 6 months after delivery.

Results:

No statistically significant differences were observed between the two groups in the rates of mothers exclusively breastfeeding (EBF) or nonexclusively breastfeeding. The percentage of those who were EBF when discharged was 97.3% in the oxytocin-nonexposed group and 87.1% in the oxytocin-exposed group (p = 0.14). At 3 months, the group rates of exclusive breastfeeding were 72.5% in the nonoxytocin-exposed group versus 65.9% in the oxytocin-exposed group (p = 0.71). At 6 months, rates of breastfeeding were 31.4% versus 27.9% (p = 0.53) in the oxytocin-nonexposed and oxytocin-exposed groups, respectively.

Conclusions:

In this study, no statistically significant effect of intrapartum synthetic oxytocin administration was observed pertaining to the initiation or duration of breastfeeding.

Introduction

Evidence says that maternal breastfeeding is beneficial. Breastfeeding is associated with a decrease in the risk of diabetes and other autoimmune disorders, protection against infectious illnesses, improvement of intelligence quotient, and decrease in mortality among other benefits.^1–3

To obtain these beneficial effects, maintenance of breastfeeding is an important issue. Several factors can impact the number of months of breastfeeding, such as prenatal education, pacifier use, or sharing the room with the newborn.⁴ In addition, several interventions during birth can have an effect on breastfeeding. Among the most common interventions are cesarean surgeries⁵ and exposure to different drugs used to augment contractions or to decrease pain. The results of interventions such as epidural analgesia on breastfeeding duration are unclear.^6,7

During labor, it is common to administer synthetic oxytocin to obtain effective uterine activity, especially in nulliparous women.⁸

Some authors consider exogenous oxytocin to be the drug that is most associated with preventable or avoidable adverse effects in childbirth.⁹ Oxytocin administration is one of the single most frequently used induction method. The use of the hormone without any medical or obstetric indication is increasing in developed countries. For example, in a study in Italy in 1992, the drug was used in 17.5% of induced deliveries, while in a more recent study in Latin America, it was used in 65.9% of induced deliveries.^10,11 Synthetic oxytocin during delivery is so widespread that there is a tendency to assume that its effects are well known. At first, it was assumed that the use was safe for fetuses because the placental barrier and blood–brain barrier were thought to provide natural protection. We did not find studies supporting these assumptions. Malek et al.¹² studied the human placental transport of oxytocin and found that oxytocin administered to the mother could pass to the fetal circulatory system. Wahl showed that permeability of the blood–brain barrier can increase during stressful situations because of the liberation of proinflammatory cytokines.¹³ Labor is a high stress situation for both the mother and child, and could facilitate passing synthetic oxytocin across the placental barrier to the baby's brain.

Ounsted et al.¹⁴ were among the first researchers to describe how different strategies in labor induction in humans can affect breastfeeding. They observed in a group of 184 primiparous women that newborns delivered by natural childbirth received maternal breastfeeding more frequently, both during hospital intake and at discharge, compared with newborns whose labors were induced with oxytocin. In contrast with these findings, Out et al.¹⁵ studied 185 mothers and affirmed that the administration of oxytocin for induction of labor does not influence the establishment of breastfeeding.

In a secondary analysis study in 2011, Guerra et al.¹¹ analyzed data from 37,444 pregnancies. They found that delayed initiation of breastfeeding was more common among women who had an elective induction of labor. However, this retrospective study did not examine the possible effects that delivery induction could have on breastfeeding duration. Likewise, a retrospective study in Spain comparing 189 neonates exposed to oxytocin during labor with 127 unexposed neonates found that the drug was linked to greater difficulty in the initiation and maintenance of breastfeeding.¹⁶

In contrast to these findings, Bai et al.¹⁷ performed a multicenter prospective study in Hong Kong, observing that each labor induction, pain reliever opioid use, or cesarean section did not influence maternal breastfeeding. They did, however, indicate that multiple intrapartum interventions significantly increased the likelihood of breastfeeding cessation.

In a previous study, we found that intrapartum oxytocin administration might inhibit the expression of several primitive neonatal reflexes associated with breastfeeding.¹⁸ The aim of the study is to determine if synthetic intrapartum oxytocin has a relationship to initiation/duration of breastfeeding.

Patients and Methods

A prospective, observational cohort study was conducted at a tertiary hospital that holds the Baby-Friendly Hospital Initiative (BFHI) designation of the WHO-UNICEF. The study was reviewed and approved by the Review Board of the Hospital Universitario Puerta de Hierro, Majadahonda (Madrid). During intake, hospital midwives explained the study opportunity to all mothers who fulfilled inclusion criteria before delivery in the hospital. Mothers provided informed consent to participate. The sample comprised two groups: the exposed group, consisting of the mother–child dyads, in which oxytocin was administered during labor, and the comparison group, comprising the mother–child dyads, in which oxytocin was not administered. Inclusion criteria were healthy single-birth newborns who were delivered vaginally at full-term with an Apgar score of at least seven at 5 minutes. In addition, mothers must have expressed a desire to breastfeed during pregnancy. Exclusion criteria were preterm infants; fetal chromosomal abnormalities diagnosed in utero; maternal admission in Intensive Care Unit; newborn Neonatal Intensive Care Unit admission in the 48 hours following delivery; lack of maternal desire to breastfeed expressed before delivery; language difficulties; and caesarean sections. Families did not receive financial compensation for their enrollment. Information such as the newborn's gestational age, sex, weight, Apgar score, and the mother's educational level, previous pregnancies, and marital status was collected, as well as information regarding the use of instruments during delivery (such as forceps).

Oxytocin induction/stimulation in the exposed group was performed by obstetrical indication (not electively) according to Bishop test under the Cardiff method: preparation of 10 U of oxytocin (Syntocinon^®; Defiante Pharmaceutics) in 500 mL of saline solution at 0.9%. Administration began with 2 mUI and was doubled every 15 minutes until at least three contractions in 10 minutes appeared, with a maximum allowance of 40 mUI. The final oxytocin dose was recorded by midwives assisting with the labor process. The most common indications were overdue and premature rupture of membranes. All mothers received oxytocin during the third stage of labor as recommended to prevent postpartum hemorrhage.

For epidural anesthesia, mothers received levobupivacaine at 0.125% (Chirocane^®; Abbott) combined with fentanyl or ropivacaine at 0.2% (Naropin^® Polybag^®; AstraZeneca) combined with fentanyl.

Infants were considered exclusively breastfed (EBF) if they received no other liquids or breast milk substitutes (other than vitamins or medications). They were considered partially breastfed (PBF) if they were supplemented with infant formula and/or other breast milk substitutes and were considered formula fed if they had not received any amount of breastfeeding.

After discharge, a questionnaire was administered blindly at 3 and 6 months by phone. Data for the 1-month follow-up were collected during the 3-month call. If participants were unable to be reached after three calls on three different days, they were considered lost to follow-up.

Analyses were performed using SPSS version 14.¹⁹ Descriptive statistics were used to describe the sample characteristics and the duration of breastfeeding. Results are expressed in mean ± standard deviation or median and interquartile range. Qualitative variables are expressed as absolute frequencies and percentages. Shapiro–Wilk test was conducted to confirm the normal distribution hypothesis. The comparisons between groups have been performed with Student's t test for independent samples and the Mann–Whitney nonparametric test. For qualitative variables, we used χ² and Yates and Fisher corrections if necessary. All p-values were two-sided and values of 0.05 or less were considered to indicate statistical significance.

This study is a secondary objective of a project aimed to evaluate the effects of oxytocin administered during labor on primitive neonatal reflexes. It was approved by the local ethics committee and is registered in ClinicalTrials.gov under identification number NCT01891201. It was funded by the Health Research Fund, FIS No. PI10/00791.

Results

We recruited 57 dyads for the oxytocin-exposed group and 48 for the nonexposed group. After initial inclusion in the study, four mother–child dyads (7.5%) in the exposed group were excluded (two for revocation of informed consent and two for lack of data regarding the administered oxytocin dose). In the nonexposed group, three dyads (6.6%) were excluded: one newborn had a broken collarbone, another was admitted to receive phototherapy, and the third had a gastric lavage. The final sample was 53 dyads in the exposed group and 45 in the nonexposed group.

For the 3-month questionnaire, we were able to contact 40 subjects (75.5%) in the oxytocin-exposed group and 44 (97.7%) in the nonoxytocin-exposed group. For the 6-month evaluation, we collected data from 43 participants (81.1%) in the oxytocin-exposed group and 35 (77.7%) in the nonexposed group. We were not able to contact the 14% of the original sample at any time point.

As shown in Table 1, we did not find statistically significant differences between the two groups in several maternal and neonatal characteristics; nonetheless, the nonexposed group included a higher percentage of women with previous pregnancies (p = 0.01) and more women in this group had previous children (p < 0.01). All mothers with previous children had a previous breastfeeding experience. The nonoxytocin-exposed group introduced pacifiers earlier (16.2 ± 12.2 days) than the oxytocin-exposed group (34.8 ± 32.7 days) (p = 0.02).

Table 1.

Maternal and Neonatal Sample Descriptive Data

	Nonexposed group (n = 45)	Oxytocin-exposed group (n = 53)	p
Maternal education			0.98
Primary	2.4%	2.1%
Secondary	34.1%	33.3%
University	63.4%	64.6%
Previous gestation	78.6%	55.1%	0.01
Previous children	73.8%	40.8%	0.002
Gestational age (weeks)^a	39.2 ± 1.1	39.7 ± 1	0.06
Newborn weight (grams)^a	3,245.3 ± 472	3,323.8 ± 375.4	0.39
Sex of the newborn (% female)	52.4%	49%	0.16
Use of pacifier	54.1%	73.7%	0.07
Maternal return to work at 6 months	51.4%	51.2%	0.98
Age of children at maternal return to work^a	130.8 ± 34.8	144.8 ± 35.3	0.21
Babies day care at 6 months	20%	27.9%	0.45
Child age at day care onset^a	137.1 ± 18.2	130.8 ± 60.1	0.79

Indicates mean ± standard deviation.

Characteristics of labor are shown in Table 2. All deliveries in the nonoxytocin-exposed group were conducted without instruments, while in the exposed group, 22.4% involved instruments (p < 0.01). The oxytocin-exposed group received epidural anesthesia significantly more frequently than the nonexposed group (p < 0.01).

Table 2.

Labor and Delivery Characteristics

Characteristics	Nonexposed group (n = 45)	Oxytocin-exposed group (n = 53)	p
Instrumental delivery	0%	22.4%	0.005
Epidural analgesia	26.2%	95.9%	<0.001
Oxytocin dose during labor (mIU)^a	0	1,400 (390–4,000)	<0.001
Oxytocin dose after delivery (mIU)^a	15,000 (10,000–20,000)	10,300 (8,420–17,310)	0.09
Apgar score at 5 minutes^a	10 (10–10)	10 (10–10)	0.35
Neonatal resuscitation	14.3%	18.4%	0.7

Indicates median (interquartile range).

The rates of breastfeeding at the different time points are shown in Table 3. There were no statistically significant differences in the proportion of mothers using EBF or PBF at any of the evaluation time points (hospital release, 1, 3, and 6 months). Additionally, there were no differences between EBF and PBF durations between groups.

Table 3.

Breastfeeding Rates at Different Time Points (Mann–Whitney Nonparametric Test)

Time points	Nonexposed group	Oxytocin-exposed group	p
Breastfeeding at discharge (%)	n = 45	n = 53	0.27
EBF	97.2	86.8
PBF	0	10.5
FF	2.8	2.6
Breastfeeding at 1 month (%)			0.80
EBF	81.1	82.1
PBF	13.5	15.4
FF	5.4	2.6
Breastfeeding at 3 months (%)	n = 39	n = 37	0.78
EBF	70.3	64.1
PBF	18.9	25.6
FF	10.8	10.3
Breastfeeding at 6 months (%)	n = 43	n = 35	0.52
EBF	31.4	27.9
PBF	34.3	25.6
FF	34.3	46.5
Length of BF (EBF+PBF)^a	149.2 ± 55.1	148.8 ± 45.3	0.97
Length of EBF^a	114.7 ± 64.7	119.3 ± 65.1	0.75

Indicates mean ± standard deviation.

EBF, exclusively breastfed; FF, formula feeding; PBF, partially breastfed.

Discussion

Because of the increasingly frequent administration of synthetic oxytocin during labor and the lack of rigorous studies in this regard, we analyzed data collected in an observational study to evaluate the effects of intrapartum use of this hormone on breastfeeding. Based on previous studies, we expected a relationship between oxytocin administration during labor and the start and duration of breastfeeding. Contrary to our expectations, results did not show statistically significant differences between groups.

Given the variable findings relating synthetic oxytocin administration and breastfeeding found in the literature and in our own studies, many questions arise from these results. The present study results did not replicate those found in a previous study by our group at the same hospital.²⁰ Several factors may account for the lack of replication. Comparing our two samples reveals differences between the two datasets. For example, in our previous article, all women received epidural anesthesia. In our current data, the oxytocin-exposed group received significantly more epidural analgesia (p < 0.001) than the nonexposed group. Several reports observed a negative influence of epidural analgesia on breastfeeding.^21,22 Based on the literature, this difference should have reinforced the negative effect of oxytocin on breastfeeding instead of diminishing the effect. Analgesia is a main confounder in these types of studies given that induction of labor has been associated with greater need for anesthesia.¹¹

Another important descriptive data difference is that in our present study, women who received intrapartum oxytocin were more likely to be first-time mothers or to have an assisted delivery. All of those factors have traditionally been associated with shorter breastfeeding duration. However, in our data, we did not find statistically significant differences between groups in the start and duration of breastfeeding. Furthermore, in the first study, we did not have a nonsynthetic oxytocin-exposed group. The current study includes such a control group and adds follow-up measures, as well as it has recruited a larger number of participants (98 compared with 20 in the previous sample).

A possible explanation for our unexpected results could be hospital policies to encourage breastfeeding. It has been shown that in hospitals with policies encouraging breastfeeding, it may be more difficult to assess any adverse effects caused by the use of morphic derivatives during labor.²³ The hospital where the study was carried out is very supportive of breastfeeding, which may explain why oxytocin did not negatively affect breastfeeding rates. The BFHI policies may be a confounding factor and could explain the lack of statistically significant differences. In addition, our data came from a very breastfeeding-motivated sample: one of the inclusion criteria was maternal wish to breastfeed and there was no financial compensation to the study that could add additional motivation to breastfeed.

Another consideration is that the nonoxytocin-exposed group received a higher (not statistically significant) dose of synthetic oxytocin during the third stage of labor, used as recommended to prevent hemorrhage.²⁴ In a study in the United Kingdom, it was observed that women who received oxytocin during the third stage of labor were more likely to stop breastfeeding within 48 hours of delivery.⁷ The authors controlled for confounding factors, but were limited by the retrospective analysis and lack of long-term follow-up regarding breastfeeding outcomes. Recently, another study showed that mothers who had received uterotonics were significantly less likely to be breastfeeding at 2–6 weeks.²⁵ How the administration of synthetic oxytocin during the third stage of labor influences breastfeeding remains unknown.

It is surprising that hormonal administration during childbirth is becoming more frequent without empirical and systematic evidence of the possible effects of such administration.²⁶ There are good reasons to test the effects on breastfeeding given that exogenous oxytocin can potentially affect breastfeeding through various mechanisms, including desensitization and downregulation of myoepithelial receptors and local feedback mechanisms,^27,28 disruption of endogenous pulsatile secretion and fluctuating concentrations,²⁹ and infant or maternal behavior modification.^20,30

Our study has several strengths. Based on a review of the literature, there are no other reports about oxytocin dosage in relation to breastfeeding. We also include data from a nonexclusively breastfeeding group and provide follow-up of infant feeding outcomes beyond hospital discharge.

One of the main challenges in studies about perinatal interventions is the possible overlap of confounding factors in the results. For that reason, we only include healthy dyads without perinatal complications to limit risk factors that could have affected the delivery and maternal and neonatal outcomes. Given our restrictive inclusion criteria, the resulting population was a group of uncomplicated healthy neonates. By limiting the analysis to uncomplicated births, we may have precluded the detection of clinically important associations. In addition, this may impact the external validity of the results.

Our study also has several limitations. The main limitation is that the sample was derived from a study designed to assess the effect of synthetic oxytocin intrapartum administration on primitive neonatal reflexes; the size and composition of the sample was thus determined to best evaluate a separate research question. A larger sample size would be needed to lower the likelihood of type 2 errors that could affect our data. Due to our limited sample size and lack of association between synthetic oxytocin intrapartum administration and breastfeeding, we did not analyze the correlations with oxytocin dosage levels. With regard to outcome measures, data pertaining to the first month of life were collected through telephone interviews given at 3 months. This introduces the potential of memory bias, although given the consistency of data from discharge to 6 months, it seems improbable. The large number of missing data (over 20%) at 3 months in the oxytocin-exposed group also reduces the reliability of our results.

Further research is needed to establish the effects of intrapartum hormonal administration on breastfeeding. As Odent says, there are good reasons to test the hypothesis that earlier than desired cessation of breastfeeding is to a certain extent a phenomenon related to the widespread obstetrical use of synthetic oxytocin.³¹ Despite the challenges of longitudinal designs, it would be informative to collect data until babies stop breastfeeding. As mentioned previously, there was a difference between groups in the amount of anesthesia received. It would be useful to conduct this study with patients who did not receive epidural services to remove this confounding factor.

Conclusion

Our data do not support an association between breastfeeding and synthetic oxytocin administered during labor. Given the lack of understanding about the effects of oxytocin administration during delivery and other routines systematically used during labor in maternal breastfeeding, this relationship should be further studied.

Footnotes

Acknowledgments

The authors want to thank I. Santos Millán, V. Costarelli, and F. López Sanchez for their participation in the study.

Disclosure Statement

No competing financial interests exist.

References

Quigley

, Kelly

, Sacker

. Breastfeeding and hospitalization for diarrheal and respiratory infection in the United Kingdom Millennium Cohort Study. Pediatrics, 2007; 119:e837–e842.

Lucas

, Morley

, Cole

, et al. Breast milk and subsequent intelligence quotient in children born preterm. Lancet, 1992; 339:261–264.

Edmond

, Zandoh

, Quigley

, et al. Delayed breastfeeding initiation increases risk of neonatal mortality. Pediatrics, 2006; 117:e380–e386.

Jaafar

, Lee

, Ho

. Separate care for new mother and infant versus rooming-in for increasing the duration of breastfeeding. Cochrane Database Syst Rev, 2012; 9:CD006641.

Smith

. Impact of birthing practices on the breastfeeding dyad. J Midwifery Womens Health, 2007; 52:621–630.

Devroe

, De Coster

, Van de Velde

. Breastfeeding and epidural analgesia during labour. Curr Opin Anaesthesiol, 2009; 22:327–329.

Jordan

, Emery

, Watkins

, et al. Associations of drugs routinely given in labour with breastfeeding at 48 hours: Analysis of the Cardiff births survey. BJOG, 2009; 116:1622–1629.

Freeman

, Nageotte

. A protocol for use of oxytocin. Am J Obstet Gynecol, 2007; 197:445–446.

Clark

, Simpson

, Knox

, et al. Oxytocin: New perspectives on an old drug. Am J Obstet Gynecol, 2009; 200:35.e1–e6.

10.

Medication during pregnancy: An intercontinental cooperative study. Collaborative Group on Drug Use in Pregnancy (C.G.D.U.P.). Int J Gynaecol Obstet, 1992; 39:185–196.

11.

Guerra

, Cecatti

, Souza

, et al. Elective induction versus spontaneous labour in Latin America. Bull World Health Organ, 2011; 89:657–665.

12.

Malek

, Blann

, Mattison

. Human placental transport of oxytocin. J Matern Fetal Med, 1996; 5:245–255.

13.

Wahl

. Could oxytocin administration during labor contribute to autism and related behavioral disorders?—A look at the literature. Med Hypotheses, 2004; 63:456–460.

14.

Ounsted

, Hendrick

, Mutch

, et al. Induction of labour by different methods in primiparous women. I Some perinatal and postnatal problems. Early Hum Dev, 1978; 2:227–239.

15.

Out

, Vierhout

, Wallenburg

HCS

. Breast-feeding following spontaneous and induced labour. Eur J Obstet Gynecol Reprod Biol, 1988; 29:275–279.

16.

Garcia-Fortea

, Gonzalez-Mesa

, Blasco

, et al. Oxytocin administered during labor and breast-feeding: A retrospective cohort study. J Matern Fetal Neonatal Med, 2014; 27:1598–1603.

17.

Bai

, Wu

, Tarrant

. Association between intrapartum interventions and breastfeeding duration. J Midwifery Womens Health, 2013; 58:25–32.

18.

Marín Gabriel

, Olza Fernández

, Malalana Martínez

, et al. Intrapartum synthetic oxytocin reduce the expression of primitive reflexes associated with breastfeeding. Breastfeed Med, 2015; 10:209–213.

19.

IBM-Corporation.. IBM SPSS Statistics for Windows, Version 14.0. Armonk, NY: IBM Corporation, 2005.

20.

Olza Fernandez

, Marin Gabriel

, Malalana Martinez

, et al. Newborn feeding behaviour depressed by intrapartum oxytocin: A pilot study. Acta Paediatr, 2012; 101:749–754.

21.

Anim-Somuah

, Smyth

, Jones

. Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Rev, 2011:CD000331.

22.

Volmanen

, Valanne

, Alahuhta

. Breast-feeding problems after epidural analgesia for labour: A retrospective cohort study of pain, obstetrical procedures and breast-feeding practices. Int J Obstet Anesth, 2004; 13:25–29.

23.

Halpern

, Levine

, Wilson

, et al. Effect of labor analgesia on breastfeeding success. Birth, 1999; 26:83–88.

24.

Begley

, Gyte

, Murphy

, et al. Active versus expectant management for women in the third stage of labour. Cochrane Database Syst Rev, 2010:CD007412.

25.

Brown

, Jordan

. Active management of the third stage of labor may reduce breastfeeding duration due to pain and physical complications. Breastfeed Med, 2014; 9:494–502.

26.

Nabi

, Aflaifel

, Weeks

. A hundred years of induction of labour methods. Eur J Obstet Gynecol Reprod Biol, 2014; 179:236–239.

27.

Cassoni

, Marrocco

, Sapino

, et al. Oxytocin synthesis within the normal and neoplastic breast: First evidence of a local peptide source. Int J Oncol, 2006; 28:1263–1268.

28.

Leng

, Caquineau

, Sabatier

. Regulation of oxytocin secretion. Vitam Horm, 2005; 71:27–58.

29.

Leng

, Meddle

, Douglas

. Oxytocin and the maternal brain. Curr Opin Pharmacol, 2008; 8:731–734.

30.

Carter

. Developmental consequences of oxytocin. Physiol Behav, 2003; 79:383–397.

31.

Odent

. Synthetic oxytocin and breastfeeding: Reasons for testing an hypothesis. Med Hypotheses, 2013; 81:889–891.