Timing and success determinants of ECV in term breech presentation, evidence base for the decision on mode of delivery, and criteria and technique of vaginal breech delivery in selected cases.
⚕️ Target audience: This clinical review is intended for physicians and healthcare professionals; It is a summary of current literature and guidelines for educational purposes. Clinical decisions should be made based on patient-specific evaluation and applicable institutional protocols.
Breech presentation is seen in approximately 3-4% of term pregnancies and is the name given to the situation where the presenting part of the fetus is the hips and/or feet. The anatomy of most breech fetuses is normal; However, there is an increased risk of congenital malformation, mild deformity, torticollis and developmental hip dysplasia. Its clinical importance is concentrated in two axes: the risk of asphyxia and traumatic injury in vaginal birth is higher than in cephalic presentation, and accordingly, breech presentation has become one of the single most important determinants of cesarean section rates worldwide (UpToDate, Overview of breech presentation, November 2025).
The prevalence of breech presentation is inversely proportional to gestational age: in early pregnancy the relative volume of amniotic fluid allows for fetal mobility and presentation resolves spontaneously as the weeks progress. This natural course explains why the ECV timing was postponed to 36-37 weeks.
| Week of pregnancy | Frequency of breech presentation |
|---|---|
| <28 weeks | 20%-25 |
| 32 weeks | 7%-16 |
| 34 weeks | 5%-7 |
| 36 weeks – term | 3%-4 |
The three main types of breech presentation differ radically in terms of the diameter of the presenting part and thus the mechanics of birth:
Why the type matters: In frank and complete breech presentations, the thighs and trunk pass through the birth canal at the same time ; if this larger diameter passes easily, the shoulders and head that follow are also likely to pass. In a footling presentation, one or both hips are not flexed, so the trunk and thighs do not pass together — the smaller presenting part may pass through an incompletely dilated cervix or an inadequate pelvis, after which the wider shoulders and head become trapped . The same small diameter also leaves space for cord prolapse: cord prolapse has been reported in approximately 1% of frank breech presentations and ≥10% of footling presentations (UpToDate, breech presentation, November 2025).
Breech presentation is incidental in most pregnancies; Up to 15% of cases are due to a fetal, maternal or placental abnormality—it is suggested that in the presence of normal anatomy, fetal activity, amniotic fluid and placental location, the fetus chooses the cephalic position near term, and breech presentation becomes possible when one of these variables is disrupted. Reported risk factors: preterm pregnancy, uterine anomaly (bicornuate/septate uterus, myoma), placental location (previa, cornual), extremes of amniotic fluid volume, nulliparity, narrowed maternal pelvis, fetal anomaly, extended fetal legs, multiple pregnancy, short umbilical cord, fetal growth restriction, and advanced maternal age (UpToDate, breech presentation, November 2025).
Population-based registries show a significant tendency for recurrence: after one breech presentation the recurrence rate in the next pregnancy is approximately 9%, after two consecutive breech presentations it is 25%, after three consecutive breech presentations it is close to 40%. On the other hand, if the first pregnancy is not breech, the frequency of breech presentation in the next is approximately 2% (Ford 2010, BJOG). Recurrence is usually attributed to persistent maternal anatomical disorder; However, the fact that the first child of parents who were also born with term breech presentation is twice as likely to present breech suggests a hereditary component in fetal development (Nordtveit 2008, BMJ).
Assessment of fetal presentation in late pregnancy should be part of the routine antenatal examination; However, abdominal palpation can be misleading even in the hands of an experienced clinician. In a study in which 138 pregnant women between 30-41 weeks were examined by an experienced clinician just before ultrasound, the physician detected only three of eight breech presentations, whereas he misdiagnosed breech presentation in six cases (UpToDate, breech presentation, November 2025). Obesity, full bladder, leiomyoma, polyhydramnios, and anteriorly placed placenta obscure characteristic palpation findings; Ultrasound, on the other hand, detects all breech presentations at the time of examination (UpToDate, breech presentation, November 2025).
This difference is not just academic. In a prospective study, before ultrasound at 36 weeks, 54% of breech presentations were not noticed on clinical examination; In a cohort study, introducing routine ultrasound at 36 weeks reduced the rate of undiagnosed breech presentation before labor from 22.3% to 4.7% (RR 0.21; 95% CI 0.12-0.36) (Knights 2023, PLoS Med).
In ultrasound, not only the incoming part, but also the flexion/extension status of both hips and knees, the location of the feet, and critically the hyperextension of the fetal head (extension angle >90°) are evaluated; hyperextension is a contraindication for vaginal birth (UpToDate, vaginal breech birth, September 2025).
Breech presentation itself is likely not an independent risk factor for adverse neonatal outcome; Increased morbidity is partly due to accompanying underlying conditions (congenital anomaly, growth restriction, preterm birth) and partly due to birth trauma. Regardless of the mode of delivery, congenital anomalies are more common in newborns with intrauterine breech delivery (4.4% versus 2.4% with vertex delivery); mild deformities, torticollis, and developmental hip dysplasia have also increased (UpToDate, breech presentation, November 2025).
Hip screening: In babies who were in breech presentation at ≥34 weeks, hip ultrasound is recommended at 4–6 weeks after birth (using corrected age after preterm birth), even if the hip examination is normal. Late dysplasia has been reported despite a normal initial ultrasound; a normal first test therefore does not end follow-up (UpToDate, breech presentation, November 2025).
ECV is the conversion of the fetus from noncephalic to cephalic presentation by manipulation through the maternal abdomen; The aim is to prevent cesarean section due to breech presentation by increasing the chance of vaginal cephalic birth.
The timing of ECV is a balance of two opposing pressures: early administration increases the chance of success but also introduces the possibility of delivering a preterm baby due to an ECV complication. Performing ECV at 36–37 weeks supports both the possibility of the version being permanent and the fetus being close to maturity if an emergency caesarean section is required (ACOG PB 221, 2020).
Institutional recommendations fall within this range but do not overlap: ACOG recommends ECV be discussed at ≥36 0/7 weeks, administered at ≥37 0/7 weeks (ACOG Practice Bulletin 221, 2020; confirmed in 2023). RCOG recommends that ECV be offered from 37 0/7 weeks, but states that it can also be offered at 36 0/7 weeks in nulliparas (RCOG Green-top Guideline 20a, 2017). UpToDate authors offer it any time after 36 0/7 weeks, typically at 37 weeks.
Early ECV (34 0/7 – 35 6/7 weeks) is a separate question and the evidence is clear: in a meta-analysis of three randomized trials (1906 participants), early ECV reduced the rate of noncephalic presentation at birth to 42% versus 52% (RR 0.81; 95% CI 0.74–0.90), but the difference in the cesarean delivery rate was only borderline significant (RR 0.92; 95% CI 0.85–1.00) and preterm birth rates were similar (Hutton 2015, Cochrane; Hutton 2011, BJOG). Thus, early ECV turns the fetus more often without significantly reducing the cesarean rate—which is why it has not become routine practice.
ECV success rates vary widely between centers, ranging from 22% to 76%; In a meta-analysis of 84 studies covering approximately 13,000 attempts at term versioning, the pooled success rate was 58% (Grootscholten 2008, Obstet Gynecol). In a series of 2614 interventions, return to breech presentation after successful ECV (UpToDate, ECV, February 2026) was 2.2%, and spontaneous cephalic version after unsuccessful ECV was 4.3%.
| Decreasing success | Increasing success |
|---|---|
| Anterior, lateral or cornual placenta; decreased amniotic fluid | Amniotic fluid index >10 cm; posterior placenta |
| The breech has descended into the pelvis; the fetal head cannot be palpated | Unengaged incoming part; complete breech |
| Nulliparity (40% vs 64% in multiparity) | Nonlongitudinal posture (oblique/transverse) |
| Obesity (BMI ≥40: 59% vs 65%); frank breech | Dedicated ECV team (40% → 70%); standard protocol (47% → 61%) |
Factors considered controversial or invalid: As long as vaginal delivery is planned, estimated fetal weight above a certain threshold is not a reason to avoid ECV. Previous lower segment transverse cesarean section is also not a contraindication for ECV; In a meta-analysis of six cohort and two case-control studies (>14,500 pregnant women), the ECV success rate was similar in those with and without previous cesarean section (range 63-83%), but the rate of vaginal birth after successful ECV was lower (75% vs 92%; pooled OR 0.26; 95% CI 0.14-0.50), and uterine rupture was not reported in these series. ACOG states that if a trial of vaginal birth is clinically appropriate, the decision to ECV should be individualized (ACOG PB 221, 2020).
An absolute contraindication is any condition that constitutes an absolute contraindication to vaginal birth, regardless of fetal development—placenta previa and previous classic cesarean section are typical examples. Because ECV complications are rare, there is no consensus regarding other contraindications and recommendations are largely based on expert opinion. Conditions generally avoided on the grounds that the likelihood of success is low or the risk of fetal harm is high: fetal growth restriction, hyperextended fetal head, multiple pregnancy, nonreassuring fetal monitoring, placental abruption, previous vertical lower segment cesarean section, rupture of membranes, severe oligohydramnios, and significant fetal or uterine anomaly. Although it is biologically plausible that ECV increases the risk of detachment in severe preeclampsia/HELLP syndrome, there is no data to support this; UpToDate authors individualize this decision (UpToDate, ECV, February 2026).
The procedure is performed by an experienced clinician in a facility with rapid access to emergency cesarean delivery. Beforehand, fetal and placental positions are confirmed by ultrasound, and fetal well-being is documented with a reactive fetal heart rate tracing or an adequate biophysical profile score. Once the breech is disengaged from the pelvis, the UpToDate author prefers starting with a backward somersault, whereas ACOG describes a forward roll first. The fetal heart rate is auscultated every two minutes; if bradycardia develops, the procedure is paused, and no more than four attempts are made in a single session (UpToDate, ECV, February 2026).
Tocolysis is the best-studied and most effective adjuvant measure; is used in most patients and this approach is consistent with institutional guidelines. The preferred agent is beta-adrenergic agonists: 0.25 mg terbutaline subcutaneously 15-30 minutes before the procedure, or 0.1 mg salbutamol slowly intravenously (monitoring the maternal heart rate) in 20 mL physiological saline immediately before the procedure. In a systematic review of six randomized studies, ECV with a parenteral beta-agonist increased the frequency of cephalic presentation in labor (mean RR 1.68; 95% CI 1.14-2.48) and reduced the rate of cesarean section (mean RR 0.77; 95% CI 0.67-0.88) compared to ECV without tocolysis (Cluver 2015, Cochrane). Data on atosiban, nifedipine, and nitroglycerin are more limited, suggesting that these agents are less effective than beta-agonists (UpToDate, ECV, February 2026).
Neuraxial anesthesia—effective, but not routine: A meta-analysis of nine randomized trials involving 934 pregnant women receiving tocolysis compared ECV success with 58.4% versus 43.1% (RR 1.44; 95% CI 1.27-1.64) and vaginal delivery with 54.0% versus no neuraxial anesthesia, IV analgesia, or no analgesia group. increased to 44.6% (RR 1.21; 95% CI 1.04-1.41); Caesarean section was reduced by 46.0% versus 55.3% (RR 0.83; 95% CI 0.71–0.97) (Magro-Malosso 2016, AJOG ). In contrast, UpToDate authors recommend against routine use of neuraxial anesthesia (Grade 2C): it is invasive, costly, and most patients experience only mild discomfort. Reasonable indications are cases in which ECV with tocolysis alone has failed and in cases where neuraxial anesthesia will already be used during planned cesarean section.
Amnioinfusion, pre-procedure hydration and bladder fullness do not increase success (UpToDate, ECV, February 2026). There is no high-quality evidence that maternal postural maneuvers in non-ECV alternatives facilitate spontaneous version (review of six studies, 417 pregnancies). In a systematic review of seven studies (1152 pregnant women) for moxibustion, the rate of noncephalic presentation at birth was reduced (RR 0.87; 95% CI 0.78-0.99), but caesarean section rates were similar (Coyle 2023, Cochrane).
ECV complications are rare but not trivial. In a systematic review of 84 studies covering 12,955 pregnant women who underwent ECV after 36 weeks at term (Grootscholten 2008, Obstet Gynecol):
| Outcome | Pooled proportion (95% CI) | Absolute frequency |
|---|---|---|
| Total complications | 6.1% (4,7-7,8) | — |
| Stillbirth | 0.19% (0,12-0,27) | ~1/5,000 procedures |
| Placental abruption | 0.18% (0,12-0,26) | ~1/1,200 procedures |
| Emergency caesarean section | 0.35% (0,26-0,47) | ~1/286 procedures |
Only two of the 12 deaths within the stillbirth figure were attributed to ECV. In a review of randomized trials, perinatal death occurred in 2 of 644 infants in the ECV group versus 6 of 661 infants in non-ECV breech presentations (RR 0.39; 95% CI 0.09–1.64)—meaning there is no signal that ECV increases perinatal mortality. These risks cannot be interpreted without a comparison group: the inherent risks of persistent breech presentation and of planned cesarean delivery (surgical injury, hemorrhage, and increased placenta accreta spectrum and uterine rupture in subsequent pregnancies) must be weighed against ECV.
Anti-D immunoglobulin is mandatory: D-negative pregnant women receive anti-D immunoglobulin immediately after ECV; This is consistent with ACOG and RCOG recommendations. Since the risk of major fetomaternal bleeding is extremely low, measuring the amount of bleeding with a Kleihauer-Betke or similar test before application seems unnecessary: in more than 1200 pregnant women with a negative pre-ECV test, post-procedure positivity was seen in 2.4%, and bleeding almost always remained below 30 mL.
After the procedure, fetal well-being is evaluated by cardiotocography; Monitoring is not continued for a fixed period of time, but until a stable and reactive pattern is achieved. It is common for traces to remain nonreactive for 20-40 minutes after ECV. Termination of ECV due to abnormal fetal heart rate was required in approximately 5% of cases in a large series (UpToDate, ECV, February 2026).
Return to routine antenatal care after successful ECV; Avoid “stabilizing induction” to prevent reversion (ACOG)—unless another induction is indicated. After unsuccessful ECV, it may be retried once or twice at one or two day intervals; No re-attempts are made after two unsuccessful attempts on separate days (UpToDate, ECV, February 2026).
The effect of ECV on the cesarean section rate is significant: a meta-analysis covering nine studies (>180,000 participants) found a cesarean section rate of 43% versus 75% (RR 0.57; 95% CI 0.50-0.64) in pregnant women in whom ECV was attempted (Hofmeyr 2015, Cochrane). However, the cesarean rate after successful ECV remains approximately twice as high compared to cephalic delivery without ECV (21% vs 11%; RR 2.19; 95% CI 1.73-2.76)—the excess resulting from both dystocia and unreassuring fetal heart rate patterns.
The basis for contemporary practice is the Term Breech Trial (Hannah 2000, Lancet): At 121 centers in 26 countries, 2088 pregnant women with singleton fetuses presenting with frank or complete breech were randomized to planned cesarean section or planned vaginal birth. The primary composite outcome (perinatal death, neonatal death, or severe neonatal morbidity) was significantly lower in the planned cesarean group: 1.6% vs 5.0% (RR 0.33; 95% CI 0.19-0.56; p<0.0001). There was no difference between groups in maternal death or serious maternal morbidity (3.9% vs 3.2%; RR 1.24; 95% CI 0.79-1.95).
In a Cochrane review combining three randomized trials (2396 participants), planned caesarean section reduced perinatal/neonatal death (RR 0.29; 95% CI 0.10-0.86). It should be emphasized that this result comes largely from countries with low perinatal mortality:
| Country group | Planned caesarean section | Planned vaginal | Risk ratio (95% CI) |
|---|---|---|---|
| Countries with low perinatal mortality | 4/1000 | 57/1000 | 0,07 (0,02-0,29) |
| Countries with high perinatal mortality | 29/1000 | 44/1000 | 0,66 (0,35-1,24) |
The absolute effect size is decisive for clinical interpretation: it has been calculated that approximately 338 caesarean sections due to breech presentation would be required to prevent one perinatal death.
The study's long-term data softens the picture. The combined risk of death or neurodevelopmental delay at age two was similar in the two groups, and 17 of 18 neonates with severe early morbidity were neurologically normal at age two (Whyte 2004, AJOG). On the maternal side, planned cesarean section slightly increased bleeding, transfusion, and infection in the short term and reduced urinary and gas incontinence; There was no difference in the long term, and complications related to uterine scarring in future pregnancies were not evaluated. The study changed practice worldwide: in 2016, the cesarean section rate for breech presentation was ≥70% in Europe and approximately 95% in the USA (UpToDate, breech presentation, November 2025).
Methodological criticisms of the study (heterogeneity between centers, compliance with selection criteria, level of experience, outcome definitions) are widespread in the literature and led to the birth of the PREMODA study. In this prospective observational study involving 8105 term singleton breech fetuses at 174 centers in France and Belgium, the composite of fetal/neonatal death or severe neonatal morbidity did not differ significantly between the two groups when strict selection criteria and a strict intrapartum protocol were applied: 1.60% versus 1.45%; adjusted OR 1.10 (95% CI 0.75-1.61) (Goffinet 2006, AJOG ). Approximately 70% of the 2502 pregnant women in the planned vaginal birth group delivered vaginally, and negative perinatal outcomes were reported in 165 pregnancies (6.6%) in this group. Factors associated with unfavorable outcome were geographic origin, birth at <39 weeks, birth weight below the 10th percentile, and birth in a center with <1500 births per year (Goffinet 2006, AJOG).
No two studies have the same level of evidence. Term Breech Trial is randomized, PREMODA observational; PREMODA cannot be read as definitive proof of "equivalence". What it does provide is a reference to the magnitude of morbidity/mortality of planned vaginal breech birth under strict protocol—and in the same series, only about 30% of pregnant women met the criteria for a trial vaginal birth.
Criteria have been developed to define the lowest risk group; these are based largely on expert opinion and vary between practitioners.
| Criteria | UpToDate author | SOGC No. 226 (2009) | CNGOF (2020) |
|---|---|---|---|
| Fetal presentation | Frank or complete; footling is contraindicated | Frank or complete, head flexed/neutral; others are contraindicated | Complete breech alone is not a contraindication |
| Estimated fetal weight | ≥2000 – ≤4000 g (author's practice) | Available in 2500 – 4000 g | Does not specify a numerical threshold |
| Week of pregnancy | ≥36 weeks | Term | Term |
| Previous cesarean section / nulliparity | Caesarean section relative contraindication; not nulliparity | Caesarean section is not among the stated contraindications | Neither alone is a contraindication |
| Induction | Most experts recommend avoiding | Not Recommended (II-3B) | Not a contraindication if criteria are met |
| Pelvimetry | No convincing evidence that it improves selection | Radiologic pelvimetry is not required; the best indicator is progress in labor | — |
| Head hyperextension | Contraindication (≥90°) | Must be flexed or neutral | — |
| Growth restriction / macrosomia | Contraindication | Contraindication (I-A) | — |
The severity of head hyperextension should be emphasized: in a large series, extended fetal neck/head was reported in 33/445 (7.4%) of breech presentations; While no neurological abnormalities were observed in any of the seven fetuses born by cesarean section, neurological sequelae attributable to spinal, supraspinal and cerebellar injuries were detected in 5 of the 26 fetuses born vaginally (UpToDate, vaginal breech birth, September 2025). The fact that approximately half of the perinatal deaths in the Term Breech Trial occurred in growth-restricted fetuses also explains why growth restriction is a separate exclusion criterion: the tendency for fetal acidosis due to chronic placental insufficiency, combined with the inevitable cord compression in the second stage.
Consent framework: ACOG states that "planned vaginal delivery of a term singleton breech fetus may be reasonable under hospital-specific protocol guidelines for eligibility and labor management" and notes that if vaginal breech delivery is planned—including the risk that perinatal/neonatal mortality or short-term serious neonatal morbidity may be higher than with planned cesarean delivery—detailed informed consent must be documented (ACOG Committee Opinion 745, 2018; reaffirmed 2023). SOGC likewise explicitly states that a pregnant woman who chooses labor despite a contraindication to a trial of vaginal breech delivery "has the right to do so and should not be abandoned."
Outline of labor management: On admission, vaginal examination and/or ultrasound confirms whether the fetal hips are flexed or extended. Membranes are left intact—because the irregular contour of the breech increases the risk of cord prolapse if membranes rupture; when spontaneous rupture occurs, an immediate vaginal examination is performed to exclude cord prolapse. Most experts recommend continuous electronic fetal heart rate monitoring. Neuraxial analgesia relieves pain, eliminates the involuntary urge to push before full dilation, and provides anesthesia if maneuvers become necessary.
Progress criteria: In the first stage, descent is considered adequate when the breech reaches the level of the ischial spines at 6 cm cervical dilation and reaches the pelvic floor at full dilation. In the second stage, a passive phase delaying active pushing for up to 90 minutes is acceptable; however, failure to deliver the breech within 30–60 minutes of active pushing is managed by cesarean delivery, not breech extraction. Oxytocin in the active phase is controversial: the UpToDate author regards poor progress as an indicator of fetopelvic disproportion and proceeds to cesarean delivery, whereas SOGC considers augmentation acceptable in the presence of uterine dystocia (SOGC No. 226, 2009).
There are two basic principles in birth itself: spontaneous or assisted breech birth is acceptable, while breech extraction is avoided in singleton pregnancy; and birth is not intervened until the fetus reaches at least the umbilicus/scapula level with maternal pushing effort — at this point, neck/head flexion is supported by applying suprapubic compression. No traction is applied to the trunk; traction leads to extension of the neck and arms, the nuchal arms (UpToDate, vaginal breech birth, September 2025).
If the legs are extended, the foot is removed by applying pressure behind the knee and removing the thigh from the body (Pinard maneuver). If the arms are not born spontaneously, the fetus is grasped at the hip or bony pelvis — never the abdomen (kidney/adrenal injury is possible) — and rotated 180° in both directions, with the shoulders alternately brought in front of the pubic arch and the arms removed (Løvset maneuver); If unsuccessful, the index finger is slid along the scapula and inserted into the antecubital fossa, and the forearm is swept across the front of the face to the chest. Unlike cephalic birth, the head is born in flexion rather than extension: if the hairline is not visible, the torso is turned to face the mother's sacrum, the assistant applies suprapubic compression, and the head is removed with the Mauriceau-Smellie-Veit maneuver or Piper forceps—no method has been proven superior to the other (UpToDate, vaginal breech birth, September 2025). No traction is applied to the jaw (risk of temporomandibular joint injury).
45-degree rule: The fetal legs should be supported, but the torso should not be elevated more than 45 degrees above the horizontal plane of the birth canal. If the trunk is bent too far back (e.g., over the mother's abdomen), neck hyperextension can occlude the vertebral arteries and lead to cervical cord necrosis; Excessive load on the cervical spine with downward traction can also have the same result (UpToDate, vaginal breech birth, September 2025).
Episiotomy is performed only if necessary and only after the fetal anus becomes visible in the perineum. There is no strict threshold for time: in a review of 90 obstetrics textbooks, recommendations for a “safe” interval between umbilicus-head delivery ranged from 5 to 20 minutes (UpToDate, vaginal breech birth, September 2025).
Head entrapment is the most serious complication of breech birth. The preterm fetus is at high risk: the head circumference/abdominal circumference ratio is larger than that of the term fetus, so the head may get stuck in a partially dilated cervix and acute asphyxia develops due to cord compression. The same situation may occur with a larger fetus, especially if the mother begins to push before full dilatation.
The first step is uterine relaxation. The patient is placed in the McRoberts position and a uterine relaxant is administered: beta-adrenergic agonist (slow intravenous salbutamol boluses of 50 mcg until the maternal heart rate reaches 120 bpm, cumulative maximum 250 mcg; or 250 mcg terbutaline subcutaneous) or nitroglycerin (50 mcg; may be repeated at one-minute intervals as necessary, total maximum 250 mcg; initial bolus of 100-200 mcg increases the risk of hypotension). General anesthesia also provides uterine and cervical relaxation, but it is too late to be effective unless the patient is already in the operating room. Simultaneously, the mother's active pushing and suprapubic pressure are maintained.
Last resort options and UpToDate's stance: When maternal straining, suprapubic compression, and uterine relaxation fail, the remaining options all carry significant risk to the mother and fetus. Dührssen incisions (cutting the cervix at 6 o'clock, then 2 and 10 o'clock; avoiding 3 and 9 o'clock to avoid descending branches of the uterine arteries) can extend into the lower uterine segment and ligamentum latum, causing heavy bleeding; The safety of symphysiotomy with the Zavanelli maneuver is unclear. UpToDate authors recommend against their use, stating that few clinicians have the necessary experience for these procedures (UpToDate, vaginal breech birth, September 2025).
Good evidence to determine mode of delivery in preterm breech presentation is limited, and available data are almost entirely observational—significant residual confounding is expected because pregnancies that deliver vaginally and those who have a cesarean section differ in most centers. However, a rough curve emerges: the observational benefit of cesarean weakens as the gestational age decreases.
| Week of pregnancy | Finding |
|---|---|
| <37 weeks | In a review of seven observational studies (3557 participants), neonatal mortality was lower in the caesarean section group: 3.8% versus 11.5% (pooled RR 0.63; 95% CI 0.48-0.81) (Bergenhenegouwen 2015, Obstet Gynecol) |
| 26 0/7 – 36 6/7 weeks | No difference in perinatal mortality in the Dutch cohort (>8300 pregnant women) (1.3% vs 1.5%; adjusted OR 0.97; 0.60-1.57); When minor morbidities were added, the composite outcome decreased (OR 0.77; 0.63-0.93). Significant decrease between 28-32 weeks in the subgroup |
| <32 weeks | In a 2024 meta-analysis (19 observational studies, >16,000 infants) planned cesarean section did not improve survival (OR 0.87; 95% CI 0.53-1.44) (Unger 2024, AJOG) |
| 23 0/7 – 27 6/7 weeks | Caesarean section reduced the odds of death by 41% (OR 0.59; 0.36-0.95; NNT 8) and severe intraventricular hemorrhage by 49% (OR 0.51; 0.29-0.91; NNT 12) in actively resuscitated extremely preterms (Grabovac 2018, BJOG) |
| 22-23 weeks | Evidence is weak; Since a classic incision is usually required in these weeks, maternal future pregnancy risks are weighed against |
Head entrapment is the mechanistic explanation for this difference: in a retrospective review of pregnant women who delivered breech at 26–29 weeks, the rate of head entrapment was twice as high in the planned vaginal birth group as in the planned cesarean group (11/84 [13%] vs. 5/85 [6%]); Three out of four neonatal deaths attributed to head entrapment were in those born vaginally (UpToDate, breech presentation, November 2025). If vaginal birth is to be attempted, leaving the membranes as intact as possible is particularly valuable in this population; Delayed cord clamping (1-2 minutes) is also recommended in extremely preterm births (UpToDate, breech presentation, November 2025).
Multiple pregnancy appears on the list of relative contraindications for standard (singleton) ECV, and data on antepartum ECV of the presenting twin (twin A) are insufficient: in a systematic review of 23 cases in which ECV was attempted for a noncephalic presenting twin, the success rate was 57% and the vaginal delivery rate was 48%; in a subsequent prospective study, version of twin A was successful in 10 of 18 cases (56%) (UpToDate, ECV, February 2026).
Breech presentation of the second twin is a fundamentally different situation—because once the first twin is delivered vaginally, the pelvis has been "tested" and the cervix is fully dilated. For this reason, footling breech delivery/extraction, which by consensus is avoided in singleton pregnancies, may be performed for the second twin. Acceptable approaches are intrapartum ECV and breech extraction; breech extraction has been reported to be associated with a higher rate of successful vaginal delivery than ECV (UpToDate, vaginal breech delivery, September 2025).
If the second twin is in transverse or oblique posture, the classic method is the internal podalic version: the practitioner enters the uterus, grasps one or both feet, brings the fetus into longitudinal breech posture, and then performs breech extraction. In a retrospective cohort based on Danish national registry (≥34 weeks, 457 births), 39 cases who underwent internal podalic version and breech extraction showed a lower rate of asphyxia and less need for mechanical ventilation compared to 418 cases who underwent cesarean section for the second twin; Apgar and cord pH values did not show any significant difference. Thirty of the 39 interventions were performed by obstetricians with more than five years of experience—the authors emphasize that the procedure is a disappearing skill (Jonsdottir 2014, Acta Obstet Gynecol Scand ).
Last updated: . Guide numbers and publication years have been verified against current catalogues.