Stopping oxytocin at active labor modestly reduces cesarean risk and uterine tachysystole, with about a 30-minute longer labor; certainty depends on trial trustworthiness.

Background

Oxytocin is commonly used for induction or augmentation of labor but can increase uterine tachysystole and nonreassuring fetal heart rate (FHR) patterns. Discontinuing oxytocin once active labor is established has been proposed to reduce adverse effects without increasing cesarean delivery (CD).

Patients

• Total: 5734 pregnant patients across 15 randomized controlled trials
• Population: Women receiving oxytocin for induction or augmentation; mean gestational age ~39.6 weeks; majority nulliparous; regional anesthesia common
• Setting: Multinational (Europe, Asia, Middle East, United States)
• Active labor definition: As defined by each trial (most 5–6 cm cervical dilation)

Intervention

Discontinuation of oxytocin at onset of active labor, with protocolized criteria allowing oxytocin restart for inadequate progress (oxytocin was restarted in about 29% of discontinuation-group patients when specified).

Control

Continuation of oxytocin through delivery per local protocol.

Outcome

• Primary: Cesarean delivery rate
• Secondary maternal: Uterine tachysystole, nonreassuring FHR tracing, postpartum hemorrhage (PPH), operative vaginal delivery (OVD), labor duration
• Secondary neonatal: Apgar score at 5 minutes <7, umbilical arterial pH <7.10, NICU admission, resuscitation at birth, therapeutic hypothermia, neonatal death

Study Design

• Type: Systematic review and meta-analysis of randomized controlled trials (PROSPERO CRD42024513295)
• Search: Multiple databases from inception to February 2024; PRISMA-compliant
• Analysis: Random-effects models; effect estimates as relative risk (RR) or mean difference with 95% CI; heterogeneity assessed with I²; GRADE applied
• Quality/Integrity: Risk of bias generally low to moderate; most trials unblinded; funnel plot suggested small-study/publication bias; OBGYN Editors’ Integrity Group criteria assessed; leave-one-out analyses performed for studies with trustworthiness concerns

Level of Evidence

Level I (systematic review and meta-analysis of randomized controlled trials).

Follow up period

Intrapartum through delivery and immediate postpartum/neonatal hospitalization (no long-term follow-up).

Results

Primary outcome

• Cesarean delivery: RR 0.80 (95% CI 0.66–0.97); pooled events 13.2% (discontinue) vs 14.7% (continue)
  • Absolute effect: ~15 fewer per 1000
  • NNT to benefit: ~67 to prevent 1 cesarean (approximate)

Secondary maternal outcomes

• Uterine tachysystole: RR 0.45 (95% CI 0.34–0.60)
  • Absolute effect: ~66 fewer per 1000
  • NNT to benefit: ~15
• Nonreassuring FHR tracing: RR 0.64 (95% CI 0.49–0.82)
  • Absolute effect: ~96 fewer per 1000
  • NNT to benefit: ~10
• Postpartum hemorrhage: RR 0.85 (95% CI 0.71–1.02) — no significant difference
• Operative vaginal delivery: RR 1.03 (95% CI 0.90–1.19) — no significant difference
• Labor duration: Active phase +~30 minutes (mean difference 30.4 min, 95% CI 14.8–46.0); second stage +~6 minutes (mean difference 6.3 min, 95% CI 3.9–8.8)

Secondary neonatal outcomes

• Apgar <7 at 5 minutes: RR 0.76 (95% CI 0.37–1.54) — no significant difference
• Umbilical arterial pH <7.10: RR 1.00 (95% CI 0.75–1.35) — no significant difference
• NICU admission: RR 0.82 (95% CI 0.62–1.07) — no significant difference
• Resuscitation at birth, therapeutic hypothermia, neonatal death: no significant differences

Key subgroup and sensitivity findings

• Benefit for CD persisted in trials limited to induction of labor (RR 0.70; 95% CI 0.52–0.93).
• No significant difference in CD when restricted to blinded trials, trials defining active labor as 6 cm, or trials from the U.S./Europe.
• Meta-regression suggested the definition of active labor modified the effect.
• Leave-one-out analyses indicated the pooled CD benefit was sensitive to inclusion of certain trials with trustworthiness concerns.

Limitations

• Heterogeneity in active labor definitions, oxytocin protocols, and restart criteria; most trials unblinded.
• Potential publication/small-study bias suggested by funnel plot.
• Effect on CD attenuated in sensitivity analyses excluding trials with trustworthiness concerns.
• Generalizability: Several trials excluded comorbidities; mean BMI <30 kg/m²; varied geographic practice patterns.
• Incomplete reporting of chorioamnionitis and oxytocin dosing/duration; nearly 29% required oxytocin restart in discontinuation arms.

Funding

Not reported. Authors declared no conflicts of interest.

Citation

Whitley J, Burd J, Doering M, Kelly J, Frolova A, Raghuraman N. Reduced risk of cesarean delivery with oxytocin discontinuation in active labor: a systematic review and meta-analysis. American Journal of Obstetrics & Gynecology. 2025; July. doi:10.1016/j.ajog.2025.03.015. PROSPERO: CRD42024513295.