Induction of labor with oxytocin

INTRODUCTION — 

Clinicians recommend delivery when they believe the risk of allowing the pregnancy to continue carries similar or more risk than the maternal, fetal, or newborn risk of terminating the pregnancy by delivery. Induction (ie, the process of artificially stimulating uterine contractions to accomplish delivery) and cesarean birth are the only available options. Induction is generally preferred when there are no contraindications to labor and vaginal birth, given the increased maternal risks associated with cesarean birth. (See "Cesarean birth: Postoperative care, complications, and long-term sequelae".)

Use of oxytocin for induction in patients with an unscarred uterus will be discussed here. Methods of cervical ripening (eg, prostaglandins, balloon catheter) and induction in patients with a scarred uterus (eg, a previous cesarean birth or other extensive transmyometrial surgery) are reviewed separately.

●(See "Induction of labor: Techniques for preinduction cervical ripening".)

●(See "Cervical ripening and induction of labor after a prior cesarean birth".)

PREVALENCE — 

The rate of labor induction in the United States was 32.1 percent in 2021, more than tripling since 1990 when it was 9.5 percent [1].

The rate of labor induction is also rising globally, though it varies among obstetric facilities and countries [2-5]. The World Health Organization (WHO) noted that the rate is generally lower in low- and middle-income countries but in some such settings, it can be as high as those in high-income countries [6].

INDICATIONS AND CONTRAINDICATIONS

Medical and obstetric indications — Induction before the onset of spontaneous labor is indicated when the maternal/fetal risks associated with continuing the pregnancy are thought to be at least as great as the maternal/fetal/newborn risks associated with delivery [7]. The risk of continuing the pregnancy is related primarily to the severity of the maternal/fetal condition and the risk of delivery is related primarily to the gestational age (ie, neonatal morbidity). The relative risk can rarely be determined with precision.

Examples of some common conditions where induction is often medically/obstetrically indicated include, but are not limited to, the following [7,8]:

●Postterm pregnancy (see "Postterm pregnancy")

●Prelabor rupture of membranes (see "Preterm prelabor rupture of membranes: Clinical manifestations and diagnosis" and "Prelabor rupture of membranes at term: Management")

●Hypertensive disorders of pregnancy (see "Preeclampsia: Antepartum management and timing of delivery" and "Eclampsia" and "HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets)" and "Gestational hypertension" and "Chronic hypertension in pregnancy: Prenatal and postpartum care")

●Diabetes (see "Preexisting (pregestational) diabetes mellitus: Obstetric issues and pregnancy management" and "Gestational diabetes mellitus: Obstetric issues and management")

●Fetal growth restriction (see "Fetal growth restriction: Pregnancy management and outcome", section on 'Delivery')

●Twins (see "Twin pregnancy: Labor and delivery")

●Clinical chorioamnionitis (see "Clinical chorioamnionitis")

●Placental abruption (see "Acute placental abruption: Management and long-term prognosis")

●Oligohydramnios (see "Oligohydramnios: Etiology, diagnosis, and management in singleton gestations")

●Intrahepatic cholestasis of pregnancy (see "Intrahepatic cholestasis of pregnancy")

●Alloimmunization with fetal anemia (see "RhD alloimmunization in pregnancy: Management" and "Management of non-RhD red blood cell alloantibodies during pregnancy")

●Fetal demise (see "Stillbirth: Maternal care and prognosis")

Many other maternal disease states and fetal conditions also warrant labor induction. They are discussed in UpToDate topics that review management of specific maternal and fetal disorders.

Scheduled induction at 39 weeks — Scheduled induction of labor in low-risk patients at 39 weeks with well-dated pregnancies has been called "elective"; however, risk-reducing induction is a more accurate term, given evidence that outcomes are equivalent, if not in some cases better than, expectant management [9-11].

Compared with expectant management, potential advantages of induction at 39 weeks include [12-20]:

●Reduction in cesarean birth

●Reduction in other adverse neonatal and maternal outcomes (eg, hypertensive disorders of pregnancy)

●Reduction in macrosomia (and its consequences)

●Reduction in stillbirth

●Ability to control the time of birth when this could be important (eg, patients with a history of rapid labor or who live far from the hospital and thus risk an out-of-hospital birth)

The disadvantages of risk-reducing induction at 39 weeks versus expectant management are a potentially longer duration on the labor unit and loss of the opportunity to experience a natural onset of labor, which is important to some individuals. Ultimately, the balance between these advantages and disadvantages depends on an individual's values and preferences.

Risk-reducing induction should be avoided before 39+0 weeks of gestation [21]. The morbidity of birth before 37+0 weeks (ie, preterm birth) is well-established; early-term birth (37+0 to 38+6 weeks) is also associated with greater neonatal morbidity and health care utilization during the entire first year of life compared with birth at 39+0 to 40+6 weeks [22-29]. To emphasize the importance of avoiding risk-reducing induction before 39+0 weeks, the National Quality Forum, the Joint Commission, and the Leapfrog Group made risk-reducing delivery prior to 39+0 weeks a maternal performance measure ("the 39 week rule") and collect data from hospitals on risk-reducing deliveries performed at 37+0 to 38+6 weeks of gestation [30].

●Evidence

•ARRIVE trial – In the multicenter randomized controlled trial, ARRIVE, which evaluated the perinatal and maternal consequences of planned induction of labor at 39+0 to 39+4 weeks of gestation versus expectant management in over 6100 low-risk nulliparous patients across the United States, induction resulted in [16]:

-Reduced risk of cesarean birth (18.6 versus 22.2 percent, relative risk [RR] 0.84, 95% CI 0.76-0.93). In a planned secondary analysis, the cesarean birth rate at 39, 40, and 41 to 42 weeks was 17.3, 22, and 37.5 percent, respectively [18-20].

-Reduced risk of hypertensive disorders of pregnancy (9.1 versus 14.1 percent, RR 0.64, 95% CI 0.56-0.74).

-Reduced risk of neonatal respiratory support (3.0 versus 4.2 percent, RR 0.71, 95% CI 0.55-0.93).

-Probable reduced risk of the composite outcome of perinatal death or severe neonatal complications (4.3 versus 5.4 percent, RR 0.80, 95% CI 0.64-1.00). In a planned secondary analysis, the composite adverse perinatal outcome at 39, 40, and 41-42 weeks was 5.1, 5.9, and 8.2 percent, respectively [18-20]. The composite outcome included death, respiratory support, five-minute Apgar score 3 or less, hypoxic ischemic encephalopathy, seizure, sepsis, meconium aspiration syndrome, birth trauma, intracranial or subgaleal hemorrhage, or hypotension requiring vasopressor support.

Other outcomes of risk-reducing induction noted in planned secondary analyses included [18-20]:

-Fewer antepartum visits, tests, and treatments

-Shorter postpartum maternal and neonatal hospital durations, though a longer duration on the labor and delivery unit (by approximately six to seven hours)

-Similar total costs

•Other evidence – Data from meta-analyses of mostly observational studies support the concept that risk-reducing induction of labor at 39 weeks in typical clinical settings has an effectiveness similar to the efficacy demonstrated in the ARRIVE trial [17,31]. Although no large-scale trials have been performed in low-risk parous patients who have had a prior vaginal birth, observational studies in this population have reported similar findings to the ARRIVE trial [32-34].

Long-term outcome data regarding the outcomes of children born after risk-reducing induction versus expectant management are limited, but show similar neurodevelopmental outcomes for both groups [35-37].

●Recommendations of selected groups

•The American College of Obstetricians and Gynecologists (ACOG) has concluded that offering induction of labor to nulliparous patients without a medical indication for delivery at ≥39+0 weeks of gestation is a reasonable option that patients and their obstetric providers should discuss (eg, the person's desire for the timing and approach to delivery) [38]. Furthermore, hospital systems should evaluate their available resources to accommodate these inductions, with an active effort to achieve/maintain equitable delivery of care. These recommendations apply to well-dated pregnancies. ACOG recommends against risk-reducing induction of suboptimally dated pregnancies [39]. (See "Prenatal assessment of gestational age, date of delivery, and fetal weight", section on 'Suboptimally dated pregnancies'.)

ACOG has also stated that maternal anxiety or discomfort related to normal pregnancy, the distance between the patient’s residence and the hospital, or a previous pregnancy with shoulder dystocia are not appropriate indications for early-term induction (37+0 to 38+6 weeks) [40], although they may be appropriate reasons for induction at ≥39+0 weeks.

•The Society for Maternal-Fetal Medicine (SMFM) also concluded that it is reasonable to offer risk-reducing induction to low-risk nulliparous patients who are ≥39+0 weeks of gestation and recommended that providers who choose this approach ensure that patients meet the eligibility criteria of the ARRIVE trial [41].

Contraindications — In each of the following settings, there is general consensus that the maternal/fetal risks associated with labor and vaginal birth, and therefore induction, are greater than the risks associated with cesarean birth; therefore, induction of labor is contraindicated.

●Prior classical or other high-risk cesarean incision (see "Choosing the route of delivery after cesarean birth", section on 'Prior transfundal uterine incision')

●Prior uterine rupture (see "Choosing the route of delivery after cesarean birth", section on 'Prior uterine rupture or dehiscence')

●Prior extensive complete transmural uterine incision (see "Uterine fibroids (leiomyomas): Issues in pregnancy", section on 'Patients with prior myomectomy')

●Active genital herpes simplex infection (see "Genital herpes simplex virus infection and pregnancy", section on 'Route of delivery')

●Placenta previa or vasa previa (see "Placenta previa: Management" and "Velamentous umbilical cord insertion and vasa previa")

●Umbilical cord prolapse or persistent funic presentation (see "Umbilical cord prolapse")

●Transverse fetal lie (see "Transverse fetal lie")

●Invasive cervical cancer (see "Cervical cancer in pregnancy", section on 'Considerations about delivery')

●Category III fetal heart rate (FHR) tracing (see "Intrapartum category I, II, and III fetal heart rate tracings: Management", section on 'Category III pattern: Abnormal')

PREINDUCTION ASSESSMENT AND PATIENT PREPARATION

Checklist — A thorough evaluation of maternal and fetal status is important before undertaking labor induction to ensure the indication is appropriate and to confirm the absence of contraindications to labor or vaginal birth:

●Review the data used to estimate gestational age and date of delivery. The gestational age and the certainty of the estimate are factors that are considered in timing of induction. (See 'Medical and obstetric indications' above and 'Scheduled induction at 39 weeks' above.)

●Determine fetal presentation. Noncephalic presentation can be an absolute or relative contraindication to induction. (See "Transverse fetal lie" and "Compound fetal presentation" and "Overview of breech presentation".)

●Estimate fetal weight. Consider the risks associated with induction of small or large for gestational age fetuses. (See "Fetal growth restriction: Pregnancy management and outcome" and "Shoulder dystocia: Risk factors and planning birth of high-risk pregnancies".)

●Perform a cervical examination to decide whether a cervical ripening agent is indicated. (See 'Clinical assessment' below.)

●Review the fetal heart rate (FHR) pattern to confirm that the use of ripening agents or oxytocin is not contraindicated because of a nonreassuring pattern. (See "Intrapartum category I, II, and III fetal heart rate tracings: Management".)

Patient preparation and laboratory tests — Patient preparation and laboratory tests are generally the same as in patients in spontaneous labor (which includes a review for risk factors for problems that may develop during labor and birth [eg, past history of shoulder dystocia, postpartum hemorrhage]). (See "Labor and delivery: Management of the normal first stage", section on 'Management of the first stage of labor'.)

In addition to standard preparation and testing,

●Discuss the indications for and alternatives to induction.

●Determine whether the cervix is favorable for induction (see 'Assessing the chance of a successful induction' below). If the cervix is favorable, oxytocin is administered without cervical ripening. If the cervix is unfavorable, cervical ripening is recommended. Procedures to promote cervical ripening, especially use of prostaglandins, may initiate labor and obviate the need for oxytocin. (See "Induction of labor: Techniques for preinduction cervical ripening".)

●Describe planned medications and procedures, including side effects and complications, and the possibility of cesarean birth. (See "Informed consent in obstetrics".)

ASSESSING THE CHANCE OF A SUCCESSFUL INDUCTION

Clinical assessment

Basic principles — The likelihood that induction will result in vaginal birth depends on both cervical and noncervical factors.

●A favorable cervix (eg, high Bishop score) is associated with a shorter duration of induction and higher likelihood of vaginal birth whereas the converse is true when the cervix is unfavorable [42,43] (see 'Bishop score' below). However, an unfavorable cervix does not mean that avoiding labor induction and managing the patient expectantly will result in a higher chance of vaginal birth; patients with unfavorable cervixes are still at increased risk for cesarean birth with expectant management, and the ARRIVE trial demonstrated that the cesarean birth rate was lower with induction regardless of cervical status.

●Noncervical factors associated with a higher chance of successful induction (as well as with higher chance of vaginal birth after spontaneous labor) include [44-47]:

•Multiparity

•Ruptured membranes

•Lower body mass index

•Taller height

•Lower estimated fetal weight

•Absence of comorbidities associated with placental insufficiency (eg, preeclampsia)

In preterm gestations, a study of labor induction using data from the National Institute of Child Health and Human Development Consortium on Safe Labor found that multiparity was the best predictor of vaginal birth (multiparity versus nulliparity: odds ratio [OR] 6.8, 95% CI 6.4-7.2) [48]. Greater gestational age at induction was also predictive of vaginal birth. Although patients <34 weeks of gestation were less likely to give birth vaginally than those ≥34 weeks of gestation, vaginal birth occurred in 57 percent of patients induced at 24 to 28 weeks, 54 percent of those induced at 28 to 31 weeks, and 67 percent of those induced at 31 to 34 weeks.

Bishop score — The Bishop score (table 1) is the cervical assessment system most commonly used in clinical practice in the United States [49]. It is based on the station of the presenting part and four cervical characteristics: dilation, effacement, consistency, and position. Cervical dilation is considered to be the most important of the five scoring elements [44]. The scoring system was developed among individuals who were multiparous, ≥36 weeks of gestation, and had a normal past and current obstetric history [49]; however, it is now applied to all patients undergoing induction.

There is no universally accepted threshold for a favorable or unfavorable score. Higher Bishop scores are associated with a higher chance of vaginal birth [49-52], while lower Bishop scores have been associated with a higher chance of cesarean birth [45]. Using the Bishop scoring system, many obstetricians consider a score ≥6 as favorable and a score ≤3 as unfavorable; scores of 4 or 5 are in a gray zone.

Despite its limitations, the Bishop score appears to be as, or more, predictive of successful induction than sonographic measurement of cervical length [53] or fetal fibronectin level [44]. Cervical scoring systems other than the Bishop score exist (eg, Fields system; Burnett, Caldor, and Friedman modifications of the Bishop system [54]; simplified Bishop score [55]), but are rarely used for predicting labor outcome.

Nonprognostic factors

●The time of day when induction is started does not appear to be an important independent factor in success [56].

●Early administration of neuraxial analgesia does not appear to lower the chance of vaginal birth [57]. (See "Adverse effects of neuraxial analgesia and anesthesia for obstetrics", section on 'Effects on the progress and outcome of labor'.)

Calculators — The benefit of using a calculator to predict induction outcome is unproved, although several investigators have attempted to derive models to predict the chance that induction will result in cesarean birth [58-61]. One group developed an internet-based calculator to estimate this risk for patients at term with an unfavorable cervix [58]. A different group developed a seven-variable calculator, which is also available online [59]. A third group developed calculators for nulliparas with term and preterm pregnancies, which are available as a mobile app [60]. The area under the curve (AUC) in an external validation cohort of the first two models was 0.75 [59] and 0.77 [58] in a systematic review, and these were among the three models with the best discriminative capacity [62]. (Note: A perfect test has an AUC equal to 1.0, whereas a test that performs no better than chance would have an AUC of 0.5).

The usefulness of these calculators in clinical practice remains uncertain. The overall predictive value of the best models is modestly better than a coin toss. Also, they were developed to evaluate the chance that a cesarean might occur if an induction were undertaken and not to indicate which patients would have a higher chance of vaginal birth if they were expectantly managed rather than induced. Similarly, they may not provide information as to which patients would have lower morbidity by avoiding labor altogether.

INDUCTION PROCEDURE

Uterine and fetal heart rate monitoring — Whenever oxytocin is administered, uterine activity and fetal heart rate (FHR) should be continuously monitored so the dose can be adjusted based on labor progress, uterine activity, and the FHR pattern.

Oxytocin administration — Synthetic oxytocin administration is the most common and proven method of labor induction [63,64]. In a network meta-analysis of labor induction methods, use of intravenous (IV) oxytocin plus amniotomy and use of vaginal misoprostol ≥50 micrograms were the two approaches most likely to result in a vaginal birth within 24 hours [65].

Pharmacology/pharmakokinetics — Exogenous oxytocin administration produces periodic uterine contractions first demonstrable at approximately 20 weeks of gestation. Myometrial responsiveness increases with advancing gestational age until 34 weeks, at which time it levels off until spontaneous labor begins, when it increases rapidly [66]. Increases in myometrial sensitivity are due primarily to increases in myometrial oxytocin receptor binding sites [67]. Receptor activation triggers signaling events that stimulate contractions, primarily by elevating intracellular calcium levels [68]. It also triggers synthesis of prostaglandins in the decidua and chorioamniotic membranes via a local paracrine effect [69].

Oxytocin cannot be administered orally because the polypeptide is degraded into small, inactive forms by gastrointestinal enzymes. When given intravenously, the plasma half-life has been estimated to be three to six minutes [70]. Low-dose oxytocin protocols are based on studies showing approximately 40 minutes are required for any particular dose of oxytocin to reach a steady-state concentration and maximal uterine contractile response [71]. (See 'Side effects' below.)

Synthetic oxytocin infused within the usual dose range are not likely to pass through the placenta to the fetus or through the maternal blood-brain barrier [69].

Variations in genes related to the oxytocin receptor appear to be associated with the amount of oxytocin required during induction and the duration of labor [72]. By comparison, during spontaneous labor, progress is not related to increasing oxytocin concentration, uterine contractions are not associated with changes in plasma oxytocin concentration, and hypocontractile labor does not appear to be the result of a deficit of oxytocin [73].

Timing in patients who undergo cervical ripening — Timing of initiation of oxytocin administration is only an issue in patients who have received prostaglandins for cervical ripening. In such patients, oxytocin can be initiated, if indicated:

●Six to 12 hours after the final dose of dinoprostone gel

●30 minutes after removal of dinoprostone insert

●Four hours after the final misoprostol dose

The reason for the delay is that tachysystole occurs more often with concurrent administration of oxytocin and a prostaglandin since both drugs carry a risk of this complication [74]. Data from human and animal studies show that prostaglandin administration increases uterine sensitivity to oxytocin [75-79].

In patients who had cervical ripening with a balloon catheter, oxytocin can be initiated while the catheter is in place or after it has been removed. Oxytocin can also be initiated concurrently with amniotomy. (See 'Early amniotomy' below and "Induction of labor: Techniques for preinduction cervical ripening", section on 'Procedure'.)

Infusion — An infusion pump allows continuous, precise control of the dose administered. Using a standardized regimen minimizes errors in administration [80-83]. A common regimen is to make a solution of 60 units oxytocin in 1000 mL crystalloid (60 milliunits in one mL) to allow the infusion pump setting (mL/hour) to match the dose administered (milliunits/minute) (eg, a pump infusion rate of 1 mL/hour equals 1 milliunit/minute). (See "Reducing adverse obstetric outcomes through safety sciences", section on 'Oxytocin administration'.)

Dosing — Oxytocin dosing regimens vary among and within facilities [84], differing in initial dose, dose increments, time period between dose increments, maximum dose, and cumulative dose [69,80,85,86]. The regimens are categorized as either high or low dose. Examples are described in the table (table 2) and the advantages and disadvantages of each approach are discussed below. (See 'Choosing a low- versus high-dose regimen' below.)

For both high- and low-dose oxytocin regimens, the dose is typically increased until:

●Labor progress is normal or

●Contractions are at least moderately strong to palpation, occurring every two to three minutes (three to four contractions per 10 minutes), and lasting at least 60 seconds. This is approximately equivalent to at least 200 to 250 Montevideo units, the value expected to lead to a normal rate of cervical change and fetal descent. (See "Use of intrauterine pressure catheters", section on 'Interpretation of findings'.)  

The occurrence of adverse effects, such as sustained FHR abnormalities or tachysystole, should lead to a dose reduction until they resolve.

Maximum dose — Although many institutional regimens limit the infusion dose to no more than 40 milliunits/minute during labor with a live fetus in the third trimester, doses as high as 90 milliunits/minute have been used without adverse maternal or fetal effects in randomized trials [85,86]. Given these data and the variability in uterine response to the medication, we favor titrating the oxytocin dose according to the oxytocin responsiveness of the individual patient, primarily based on their contraction and FHR patterns, without regard to an arbitrary maximum dose.

Choosing a low- versus high-dose regimen — For most patients, either a low- or high-dose regimen is acceptable for induction (or augmentation) of labor. The decision should depend on local factors (eg, provider or institutional preferences, local resources for patient monitoring). For patients who have had a previous cesarean birth or other extensive transmyometrial surgery, some UpToDate contributors suggest using a low- rather than high-dose regimen, while others do not specifically avoid high-dose regimens in these patients. The maximum safe dose in this setting is also unclear. Although many labor units use 20 milliunits/min in these patients, this is somewhat arbitrary, absolute risk differences between this maximum and higher maximums appear to be relatively small, and confounding by indication is possible. A prudent approach is to require the physician to be notified when 20 milliunits/min is reached. Patients who have had a previous cesarean birth or other extensive transmyometrial surgery are discussed separately. (See "Cervical ripening and induction of labor after a prior cesarean birth", section on 'Issues related to use of oxytocin'.)

Moderate- to high-quality evidence from randomized trials is insufficient to allow a clear conclusion as to whether using a high- versus a low-dose regimen alters the cesarean birth rate or substantively alters other maternal or neonatal outcomes. High-dose regimens modestly reduce the time from induction to birth by one to two hours on average and reduce the risk for chorioamnionitis but result in a higher rate of tachysystole. Other maternal and perinatal rates of adverse outcome are similar for both regimens.

●In a 2025 meta-analysis (10 randomized trials, 5508 pregnancies) comparing high- versus low-dose oxytocin regimens for induction of labor, using a high-dose regimen:

•Reduced the time to delivery (mean difference [MD] -0.63 hours; 95% CI, -0.85 to -0.41).

•Modestly increased the rate of spontaneous vaginal birth (74.4 versus 69 percent; RR 1.10, 95% CI 1.03-1.18), with a statistically similar cesarean birth rate (12.4 versus 14.5 percent; RR 0.83, 95% CI 0.67-1.02).

•Reduced the rate of chorioamnionitis (8.9 versus 12.9 percent; RR 0.70, 95% CI 0.57-0.84).

•Increased the rate of tachysystole (39.6 versus 28.8 percent; RR 1.32; 95% CI 1.21-1.43). 

Rates of postpartum hemorrhage, postpartum endometritis, uterine rupture, shoulder dystocia, pH <7.10, base excess >12 mEq, Apgar score <3 at 5 minutes, NICU admission, and neonatal death were similar between the two groups.

Should oxytocin be discontinued in the active phase? — In patients who achieve a desirable labor pattern and progress, there is no consensus about whether the oxytocin dose should be discontinued or continued. Based on the evidence described below, either discontinuation or continuation is reasonable, as long as labor is progressing and the FHR pattern is reassuring. If discontinued, oxytocin can be restarted if labor progress slows. If continued, continuous FHR monitoring is essential to identify potential tachysystole with FHR changes requiring discontinuation or dose reduction.

●In a 2022 meta-analysis of randomized trials of continued versus discontinued oxytocin in the active phase of induced labor at ≥36 weeks of gestation, the risk of cesarean with discontinuation was statistically similar to that with continuation in the subgroup of patients who actually reached the active phase and thus received the intervention (RR 0.82, 95% CI 0.60-1.10) [87]. Furthermore, there were no between-subgroup differences in the indication for the cesarean births that were performed. Of note, the trials were limited by many sources of bias and differences in the cervical dilation (4, 5, or 6 cm) defining the onset of the active phase of labor.

●A subsequent open-label randomized trial that evaluated the impact of discontinuing versus continuing oxytocin during active labor (cervical dilation ≥6 cm) on neonatal morbidity found no significant difference between approaches [88].

Other approaches to dosing

●"Oxytocin rest" or break – We recommend not routinely stopping oxytocin and then restarting the infusion after a period of time for the purpose of improving labor progress. Oxytocin-induced desensitization of the oxytocin receptor has been demonstrated in vitro, but the clinical value of this finding in patients who receive prolonged oxytocin stimulation has not been demonstrated [89-91]. Although it has been hypothesized that stopping oxytocin if labor is not progressing and then restarting the drug several hours later will improve myometrial contractility, no randomized trials have shown clear evidence of benefit of this approach.

●Pulsatile dosing – Some investigators have described pulsatile administration of IV oxytocin at 6- to 10-minute intervals as it theoretically could better simulate normal labor than continuous oxytocin administration [92,93]. However, pulsatile administration does not improve outcomes such as reducing the frequency of cesarean birth. Furthermore, the time from the start of the infusion to delivery may be longer than with continuous oxytocin administration. Pulsatile oxytocin administration requires special equipment and is rarely used in contemporary obstetric practice.

Side effects

●Tachysystole – Tachysystole is the most common side effect. The frequency increases as higher doses of oxytocin are used, but reported absolute rates vary widely [94-97].

The American College of Obstetricians and Gynecologists (ACOG) uses the term tachysystole to describe >5 contractions in 10 minutes, averaged over a 30-minute window [98]. The presence or absence of associated FHR changes should be noted. When reviewing data on tachysystole, clinicians should be mindful of the different terminologies that have been used in the literature. For example, the term "uterine hyperstimulation without fetal heart rate changes" has been used to describe uterine tachysystole (>5 contractions in 10 minutes for at least 30 minutes) or uterine hypersystole/hypertonus (a contraction lasting at least 2 minutes) with a normal FHR.

•Consequences – Since uterine activity causes intermittent interruption of blood flow to the intervillous space, a prolonged period of excessive uterine activity may result in fetal hypoxemia and acidemia [99-103]. Rarely, tachysystole causes uterine rupture; this is more likely to occur in multiparous patients [95,104].

•Management – The oxytocin dose should be reduced or discontinued until the tachysystole resolves, even if unassociated with FHR changes. If FHR changes potentially associated with acidemia are present, standard interventions should be initiated (eg, left lateral position) [105,106]. (See "Intrapartum category I, II, and III fetal heart rate tracings: Management", section on 'Approach to selected category II patterns' and "Intrapartum category I, II, and III fetal heart rate tracings: Management", section on 'General approach'.)

There is no consensus regarding the optimal approach to managing oxytocin administration once tachysystole is noted. Clinicians should assess the full clinical scenario (fetal status, maternal status, access to emergency cesarean birth if required) for decision-making in individual cases. For example, discontinuing oxytocin may be the preferred approach for tachysystole with contractions occurring approximately every minute, whereas reducing the dose by 50 percent may be preferred for tachysystole that just meets criteria (five contractions over 9 minutes and 50 seconds). Similarly, discontinuing oxytocin may be the preferred approach for a patient receiving 2 milliunits/minute, whereas reducing the dose by 50 percent may be preferred for a patient receiving 40 milliunits/minute since the absolute reduction in this setting is relatively large. The presence FHR changes and, if present, their relationship with fetal acidemia is also a factor. FHR changes associated with acidemia favors oxytocin discontinuation rather than reduction and may also prompt administration of a tocolytic such as terbutaline (eg, 0.25 mg subcutaneously) [107]. Of note, in a small randomized trial of patients in active labor receiving intravenous oxytocin who developed tachysystole, the mean time to resolution after oxytocin discontinuation was 35±27 minutes [108].

Similarly, the optimal approach to resuming the drug after reduction or discontinuation for tachysystole is not known. One approach after discontinuation resumes oxytocin at one-half the last infusion dose if it has been discontinued for less than 30 minutes and at the initial infusion dose if it has been discontinued for more than 30 minutes [109-111].

●Hypotension – Because oxytocin relaxes vascular smooth muscle, hypotension and tachycardia can result from rapid IV injection. This has been observed at cesarean birth when large IV oxytocin boluses (>5 units) were administered [112-116]. In a randomized trial of 75 patients undergoing cesarean birth and assigned to one of five doses of oxytocin postpartum, the prevalence of hypotension one minute after bolus injection of 0 units of oxytocin was 7 percent (one patient), 20 to 30 percent after bolus injection of 0.5 to 3 units, and 47 percent (seven patients) after bolus injection of 5 units [112].

We found no reports of hypotension with use of contemporary IV oxytocin infusion doses for induction of labor. However, as discussed above, it is prudent to administer oxytocin for induction by infusion pump to optimally control the rate of infusion and avoid adverse cardiovascular effects (arrhythmia, myocardial infarction, hypotension), as well as tachysystole, nausea, vomiting, headache, and flushing [113].

●Hyponatremia – Oxytocin has a similar structure to vasopressin (antidiuretic hormone) and can cross-react with the renal vasopressin receptor. If higher doses (eg, 50 milliunits/minute) of oxytocin are administered in a large volume (eg, over 3 liters) of a hypotonic solution (eg, 5 percent dextrose in water [D5W]) over a prolonged period of time (≥7 hours [117-119]), excessive water retention can occur and result in severe, symptomatic hyponatremia, similar to the syndrome of inappropriate antidiuretic hormone secretion [120-122]. This risk may be as high as 5 percent when the conditions described above are met [121]. Excessive oral, rather than IV, intake of hypotonic liquids can have the same effect.

Symptoms of severe acute hyponatremia include headache, anorexia, nausea, vomiting, abdominal pain, lethargy, drowsiness, unconsciousness, generalized tonic-clonic seizures, and potentially irreversible neurologic injury. (See "Manifestations of hyponatremia and hypernatremia in adults".)

If water intoxication occurs, oxytocin and any hypotonic solutions should be stopped. Hyponatremia must be corrected carefully and consists of restricting water intake and cautious administration of hypertonic saline if the patient is symptomatic. (See "Overview of the treatment of hyponatremia in adults".)

●Prolonged QT interval – Oxytocin can prolong the QT interval, but the dose at which this becomes clinically relevant is uncertain [123-128]. There is no need to routinely screen individuals with an electrocardiogram before the use of oxytocin; however, clinicians should be aware of the potential for arrhythmia in patients with known long QT syndrome [125]. To our knowledge, torsade de pointes has not been reported in a pregnant patient receiving exogenous oxytocin. (See "Acquired long QT syndrome: Definitions, pathophysiology, and causes" and "Acquired long QT syndrome: Clinical manifestations, diagnosis, and management".)

●Pain – Whether induced labor is more painful than spontaneous labor is not well understood. The answer is not straightforward given the many factors involved in an individual's labor experience, including the physical circumstances surrounding labor (eg, frequency and strength of uterine contractions, fetal presentation), as well as individual emotional, motivational, cognitive, and social circumstances [129].

Data from randomized trials suggest that intrapartum pain is not worse with induction than expectant management. In a randomized trial comparing over 1700 individuals with postterm inductions with a similar number of expectantly managed/monitored postterm pregnancies, the rate of use of analgesia or anesthesia was similar for both groups (approximately 91 percent) [130]. In a more contemporary randomized trial of induction versus expectant management at 39 weeks among low-risk nulliparous individuals, the induction group had a small reduction in overall and worst pain scores during labor and more perceived control during childbirth; regional analgesia was used in the large majority of both groups and with similar frequency [16].

By comparison, some observational studies have suggested a higher frequency of epidural use among individuals undergoing medically-induced labor, although in many settings its use is common regardless of how labor is initiated [131,132]. Also, whether this finding reflects the presence of greater pain among those being induced is uncertain because of the strong likelihood of selection bias and confounding by indication (eg, those undergoing a medical induction are more likely to have medical circumstances that would predispose to recommendation of neuraxial analgesia). As with other outcomes, it is important to consider the difference between induction versus expectant management in terms of counseling regarding management choices.

Early amniotomy — Induction using a combination of oxytocin administration and amniotomy appears to shorten the time to vaginal birth compared with using oxytocin alone. Amniotomy is performed as soon as feasible if the head is well apposed to the cervix; if not well apposed, it is delayed to reduce the risk of cord prolapse. The best available evidence suggests that early amniotomy is safe and effective.

●A 1995 randomized trial suggested that early amniotomy (defined as "soon as deemed safe and feasible") during labor induction was associated with shorter labor (13.3 versus 17.8 hours) [133]. Although the risk of chorioamnionitis (22.6 versus 6.8 percent) was increased, two different oxytocin protocols were used (oxytocin increments every 30 or 60 minutes as required) and a statistical increase was only seen in patients who received oxytocin increments at ≥60 minutes, thus its relevance to contemporary practice where increments are faster is not clear. Furthermore, subsequent trials have not consistently shown an increased risk of infectious morbidity with early amniotomy.

●In a 2020 meta-analysis of randomized trials of patients undergoing induction (four trials, 1273 patients), early amniotomy after cervical ripening shortened the interval from induction to delivery by approximately five hours (95% CI -8.12 to -1.78) with no significant effect on the cesarean birth rate (31.1 versus 30.9 percent, RR 1.05, 95% CI 0.71-1.56) compared with usual care (late amniotomy or spontaneous rupture of membranes) [134]. There were no clinically important between-group differences for other obstetric or perinatal outcomes. Early amniotomy was defined as artificial rupture of the membranes soon after successful cervical ripening (eg, expulsion of Foley balloon, cervical dilation 3 cm, or favorable Bishop score); late amniotomy was defined as artificial rupture of membranes after the onset of the active phase of labor.

●A subsequent secondary analysis of data from >2800 nulliparous patients participating in a randomized trial of term labor induction found that amniotomy performed two or more hours after oxytocin initiation was associated with lower odds of labor duration >24 hours compared with leaving the membranes intact during the same time period [135]. Amniotomy after oxytocin initiation was also associated with either lower or similar odds of cesarean birth and other adverse outcomes (including chorioamnionitis) compared with no amniotomy.

●In a 2024 meta-analysis of randomized trials of early versus delayed amniotomy in individuals undergoing pre-induction cervical ripening with a balloon catheter (five trials, 849 participants), early amniotomy resulted in a higher proportion of births within 24 hours (79.9 versus 67.1 percent; RR 1.19, 95% CI 1.04−1.36) and shorter intervals from oxytocin to delivery (weighted mean difference [WMD] -1.5 hours; 95% CI -2.1 to -0.8), catheter expulsion to vaginal birth (WMD -2.5 hours; 95% CI -4.8 to -0.1), and start of oxytocin to vaginal birth (WMD -1.8 hours; 95% CI -3.2 to -0.4) [136]. Differences in other outcomes (cesarean birth, chorioamnionitis, intrapartum fever, endometritis, cord prolapse, neonatal sepsis) were not statistically significant.

●In a study of outcomes before and after implementation of an induction of labor protocol, which included both standardized cervical ripening and early amniotomy, the post-implementation period was associated with decreases in the times from the start of induction to rupture of membranes (13.3 versus 10.4 hours) and the start of induction to delivery (21.2 versus 19.7 hours) [137]. There was little to no difference between groups in complications such as chorioamnionitis and postpartum hemorrhage or in the cesarean birth rate.

Complications of amniotomy (whether "early" or "late") include rupture of a vasa previa (if present) and umbilical cord prolapse. These complications may occur at the time of spontaneous rupture of the membranes, as well. (See "Velamentous umbilical cord insertion and vasa previa" and "Umbilical cord prolapse".)

COMPLICATIONS

Uterine rupture — Oxytocin administration has been associated with an increased risk of uterine rupture, but the absolute risk is very low in patients with an unscarred uterus. In a series including over 226,000 births, 14 ruptures occurred in patients with unscarred uteruses, but 12 of these were in patients exposed to oxytocin for induction or augmentation of labor (two of the patients were nulliparous) [138]. Most uterine ruptures occur in laboring patients with a scarred uterus. (See "Uterine rupture of the unscarred uterus: Risk factors, clinical manifestations, management, and outcome" and "Uterine rupture after previous cesarean birth: Prediction, clinical manifestations, diagnosis, management, and outcome".)

Amniotic fluid embolism — Induction of labor with oxytocin or a prostaglandin appears to be associated with an increased risk of amniotic fluid embolism compared with amniotomy alone or spontaneous labor. In a population-based cohort study of amniotic fluid embolism, amniotic fluid embolism occurred more commonly in patients who had medical induction of labor than in those who did not (adjusted odds ratio [aOR] 1.8, 95% CI 1.2-2.7), but the absolute risk difference was small (10.3 versus 5.2 per 100,000 births) [139]. Given that patients were induced for medical indications, it is highly plausible that the amniotic fluid embolism was causally related to the indication and not the induction itself. (See "Amniotic fluid embolism".)

Postpartum hemorrhage — Oxytocin use, particularly when prolonged or at high dose, has been cited as a risk factor for postpartum hemorrhage [140]; however, this is likely related to cofounding by indication (eg, individuals receive oxytocin for augmentation in the setting of a labor curve abnormality) rather than causal (ie, the oxytocin affects physiology in a manner that increases the chance of hemorrhage). Longer labor and/or labor dystocia themselves are risk factors for postpartum hemorrhage [141-143]. In addition, in trials comparing induction (where nearly 100 percent of participants receive oxytocin) versus expectant management (small proportion of participants receive oxytocin), induction did not increase the risk of postpartum hemorrhage [16,144-146].

Other issues

●Induction of labor at term does not appear to be a risk factor for spontaneous preterm birth in the subsequent pregnancy [147,148].

●True allergic reactions to oxytocin are rare [149,150].

●There is no consistent evidence from well-designed studies that oxytocin administration is associated with autism spectrum disorder, attention-deficit/hyperactivity disorder, or any other long-term adverse outcomes in offspring [116,151]. Confounding by indication and ascertainment bias are among the factors that limit the findings of observational studies reporting an association.

●Induction of labor at term does not appear to be a risk factor for developing a severe perineal laceration [152].

LABOR PROGRESS — 

The average duration of the latent phase of labor is longer in induced labor than in spontaneous labor. (See "Labor: Overview of normal and abnormal progression", section on 'Normal progression in induced labors'.)

Once patients being induced enter active labor (cervical dilation ≥6 cm), progression appears to be comparable to that in patients with spontaneous-onset active labor [153]. The duration of the second stage is similar in induced and spontaneous labors as well [153-155]. Therefore, active phase and second stage protraction disorders and arrest are diagnosed and managed the same as in patients in spontaneous labor. (See "Labor: Overview of normal and abnormal progression".)

DEFINITION OF FAILED INDUCTION — 

There is no strong consensus as to the standard for defining a failed induction. We consider a reasonable approach to reserve the diagnosis specifically for cesareans performed in the latent phase because this phase has continued for an extended length of time and, in the clinician’s judgment, the patient is unlikely to enter the active phase by continuing oxytocin administration [156]. We would not use the term for inductions in which a cesarean is performed to treat active phase protraction or arrest, prolonged second stage, nonreassuring fetal status, or a maternal indication for prompt delivery; instead we use the terms that are typically used for these indications. We would also not use the term for inductions that are stopped and the patient sent home undelivered, which is a discontinued induction.

In 2024, the American College of Obstetricians and Gynecologists (ACOG)/Society for Maternal-Fetal Medicine (SMFM) advised that if the maternal and fetal status remain reassuring, oxytocin should be administered for at least 12 to 18 hours after membrane rupture before considering induction failure and individualize the decision to continue past 18 hours [157]. This is a reasonable approach since approximately 70 percent of nulliparous patients in a large multicenter trial exited the latent phase after 6 hours of oxytocin and membrane rupture and 20 percent between 6 and 12 hours [158]. Among the 5 percent who remained in the latent phase for more than 12 hours, 40 percent went on to give birth vaginally. In another multicenter study of nulliparous patients undergoing induction of labor, 96 percent reached the active phase within 15 hours [159]. Similar results have been reported in other studies [160-162]. Among multiparous patients, the ACOG/SMFM criteria essential eliminate the occurrence of failed induction [156].

LABOR INDUCTION WITHOUT INITIAL USE OF OXYTOCIN

Amniotomy alone — In patients with a favorable cervix, use of amniotomy alone is an option to initiate labor if the head is well apposed to the cervix; however, the combination of amniotomy and intravenous (IV) oxytocin administration is more effective. In a meta-analysis of randomized trials, this combination resulted in a substantial reduction in pregnancies undelivered at 24 hours compared with amniotomy alone (2.1 versus 16.3 percent, relative risk [RR] 0.13, 95% CI 0.04-0.41; two trials with a total of 296 participants) [163].

Prostaglandin E1 or E2 — In patients with unfavorable cervixes, prostaglandins are a proven method of cervical ripening and may initiate labor as a result of their uterotonic effects. (See "Induction of labor: Techniques for preinduction cervical ripening", section on 'Prostaglandins'.)

In patients with favorable cervixes, however, there are inadequate data to determine the safety and efficacy of using prostaglandins instead of oxytocin to initiate induction of labor. Until such data are available, we avoid using prostaglandins to induce labor in patients with favorable cervixes.

Membrane stripping — Membrane stripping to induce labor has not been associated with demonstrable improvements in many clinically important outcomes (eg, lower cesarean rate, more favorable neonatal outcome), and as such, the decision to perform it should be individualized based on patient choice.

Stripping or sweeping of the membranes involves inserting the examiner’s finger beyond the internal cervical os and then rotating the finger circumferentially along the lower uterine segment to detach the fetal membranes from the decidua. It is typically performed during an office visit when the cervix is partially dilated and the patient and clinician hope to hasten the onset of spontaneous labor. Thus, it should be performed only in those planning a trial of labor who are at least 39 weeks.

In a 2020 meta-analysis of 40 randomized trials (>6500 participants) comparing membrane sweeping with no or a sham intervention, sweeping modestly reduced the rate of formal labor induction (23 versus 31 percent, relative risk [RR] 0.73, 95% CI 0.56-0.94) and increased the rate of spontaneous onset of labor (72 versus 60 percent, RR 1.21, 95% CI 1.08-1.34) without affecting other maternal or neonatal outcomes [164]. Participants generally felt the benefits outweighed the harms (eg, discomfort) and would recommend it to others.

Group B Streptococcus colonization is not a contraindication to membrane sweeping as there is no direct evidence of harm. This issue is discussed in more detail separately. (See "Prevention of early-onset group B streptococcal disease in neonates", section on 'Antepartum procedures'.)

Nonstandard approaches — There is a paucity of data regarding the safety and/or efficacy of glucocorticoids, castor oil, hyaluronidase, isosorbide mononitrate, acupuncture, evening primrose oil, herbal preparations, breast stimulation, or sexual intercourse for labor induction, and none of these can be recommended as an evidence-based approach to labor induction [165-175].

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cervical ripening and labor induction".)

INFORMATION FOR PATIENTS — 

UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Labor and childbirth (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Indications for induction

•High-risk pregnancies – Induction is indicated when the maternal/fetal risks associated with continuing the pregnancy are thought to be at least as great as the maternal/fetal/newborn risks associated with delivery and there is no contraindication to labor and vaginal birth. The risks are influenced primarily by the gestational age and severity of the maternal/fetal conditions, and can rarely be determined with precision. Some of the most common indications are postdate pregnancy, hypertensive disorders, prelabor rupture of membranes, and diabetes. (See 'Medical and obstetric indications' above and 'Contraindications' above.)

•Low-risk pregnancies – Risk-reducing induction at 39 weeks of gestation is an option for patients without other obstetric/medical indications for delivery. The values and preferences of the patient and the resources available for procedure need to be considered when making this decision. (See 'Scheduled induction at 39 weeks' above.)

●Preinduction evaluation and management – Preinduction evaluation includes reviewing the estimate of gestational age, determining fetal presentation, estimating fetal weight, performing a cervical examination to decide whether to use a cervical ripening agent, evaluating the fetal heart rate (FHR) pattern, and reviewing the patient's pregnancy and medical history for risk factors for potential problems that may develop during labor and delivery. (See 'Preinduction assessment and patient preparation' above.)

●Cervical examination/Bishop score – Cervical status before oxytocin administration is a key factor associated with the duration of induction. The Bishop score (table 1) is the best available tool for assessing cervical status. Most obstetricians consider a score ≥6 as favorable and a score ≤3 as unfavorable; scores of 4 or 5 are in a gray zone. (See 'Clinical assessment' above.)

Patients with unfavorable cervixes who are scheduled for induction can benefit from preinduction cervical ripening, as discussed separately. (See "Induction of labor: Techniques for preinduction cervical ripening".)

For patients with favorable cervixes who are scheduled for induction, we suggest administration of oxytocin and amniotomy rather than amniotomy alone (Grade 2B). The head should be well apposed to the cervix before amniotomy. (See 'Amniotomy alone' above.)

●Oxytocin administration – Oxytocin is the standard medication for induction.

•Dosing – Oxytocin is administered intravenously by an infusion pump. Implementation of a standardized protocol can minimize errors in oxytocin administration. A common regimen is to make a solution of 60 units oxytocin in 1000 mL crystalloid (60 milliunits in 1 mL) to allow the infusion pump setting (mL/hour) to match the dose administered (milliunits/minute; ie, a pump infusion rate of 1 mL/hour equals 1 milliunit/minute). (See 'Infusion' above.)

Either a high- or low-dose oxytocin regimen is acceptable (table 2). (See 'Dosing' above.)

We increase the dose according to myometrial responsiveness in the individual patient, primarily based on their contraction and FHR patterns, without regard to an arbitrary maximum dose. The dose is typically increased until labor progress is normal; however, the dose should be reduced if sustained FHR abnormalities or tachysystole occurs. (See 'Dosing' above.)

•Timing of amniotomy – In patients receiving oxytocin, we suggest early rather than delayed or no amniotomy (Grade 2B). The combination of oxytocin administration and amniotomy appears to shorten the time to vaginal birth compared with using oxytocin alone. The head should be well apposed to the cervix before amniotomy. (See 'Early amniotomy' above.)

•Side effects – The most common side effect is tachysystole (>5 contractions in 10 minutes, averaged over a 30-minute window). Rare side effects include hyponatremia and, if administered rapidly by IV bolus, hypotension. (See 'Side effects' above.)

•Discontinuation in the active phase – In patients who achieve a desirable labor pattern and progress, there is no consensus about whether oxytocin should be discontinued or continued in the active phase or second stage, and either approach is reasonable as long as labor is progressing and the FHR pattern is reassuring. If discontinued, oxytocin can be restarted if labor progress slows. If continued, continuous FHR monitoring will identify tachysystole with FHR changes requiring oxytocin discontinuation or dose reduction. (See 'Should oxytocin be discontinued in the active phase?' above.)

●Labor progress – Once a patient undergoing induction has entered active labor (cervical dilation 6 cm), progression appears to be comparable to that with spontaneous active labor. The duration of the second stage is similar in induced and spontaneous labors. Therefore, active phase and second stage protraction and arrest disorders are diagnosed and managed similarly to patients in spontaneous labor. (See 'Labor progress' above.)

●Failed induction – There is no consensus for defining a failed induction. We use the term for cesareans performed in the latent phase because this phase has continued for an extended duration (at least 12 to 18 hours after membrane rupture) and, in the clinician's judgment, the patient is unlikely to enter the active phase by continuing oxytocin administration. The decision to administer oxytocin past 18 hours should be individualized. (See 'Definition of failed induction' above.)

ACKNOWLEDGMENT — 

The UpToDate editorial staff acknowledges Deborah A Wing, MD, MBA, who contributed to an earlier version of this topic review.