INTRODUCTION — The dominant practice in the United States for many decades was to follow the dictum of Cragin, "once a cesarean, always a cesarean," which was first put forth in 1916 [1]. In 1980, however, a panel convened by the National Institutes of Health questioned the need for routine elective repeat cesarean delivery (ERCD) and opined that a trial of labor after a previous lower uterine segment transverse incision could be attempted safely in properly selected patients [2]. In 1990, the United States Public Health Service proposed that an overall cesarean delivery rate of 15 percent be achieved by the year 2000, with a vaginal birth after cesarean (VBAC) rate of 35 percent (ie, rate of vaginal births per 100 births to women with a previous cesarean delivery) [3]. Some insurers and managed care organizations even mandated that almost all women with a previous cesarean delivery attempt a trial of labor (TOL) in the subsequent pregnancy. As a result of this change in philosophy, the VBAC rate rose from about 3 percent in 1980 to 20 percent in 1990 and then to 28 percent in 1996 [4-6]. Since 1996, however, the rate has fallen, decreasing to 7.6 to 8.5 percent (depending on method of tabulation) in 2006 [6].

This issue is of particular importance, given the continuing rise in cesarean deliveries in the United States. Over 10 years (1996 to 2006), the cesarean delivery rate rose 46 percent, to 31.3 percent [6]. Over time, more women and their providers will be faced with the issues regarding mode of delivery after cesarean.

Fewer women are attempting a TOL after a cesarean (TOLAC). Possible reasons for this include medical and legal pressures, changes in patient and provider preferences, changes in obstetric practice, and publication of complications related to vaginal delivery and failed TOL. Another factor contributing to the decline may be publications by the American College of Obstetricians and Gynecologists recommending specific personnel requirements for hospitals offering TOLAC [7] and discouraging the use of prostaglandins and oxytocin for cervical ripening in these patients [8,9]. This is in spite of an overall 60 percent to 80 percent chance of successful vaginal delivery in patients electing TOLAC (see 'Success rate' below.

The selection, counseling, and management of women for TOLAC will be reviewed here. General issues pertaining to cesarean delivery are discussed separately. (See individual topic reviews on cesarean delivery).

CHOOSING THE ROUTE OF DELIVERY — Pregnant women with a prior cesarean birth must choose between ERCD and TOLAC. The ability to identify women who are likely to successfully TOLAC could potentially decrease morbidity since studies consistently show that women who fail an attempt to deliver vaginally are at highest risk of adverse outcome [10-14]. The most serious concerns related to TOLAC are increased risks of uterine rupture and perinatal death, which may occur as related or independent events (see 'Complications' below. In a seminal study of women delivering after a previous cesarean, those who failed TOLAC accounted for 49 of 77 (64 percent) patients with major complications (need for hysterectomy, uterine rupture, or operative injury) [12]. Chorioamnionitis and hemorrhage also occur significantly more often in women who fail TOLAC than in those who are successful (eg, chorioamnionitis 25.8 versus 5.5 percent, hemorrhage 35.8 versus 15.8 percent [14]).

A reliable system for predicting an individual woman's specific likelihood of successful vaginal delivery after a previous cesarean would be useful for reassuring her that she is not likely to labor for hours in vain. A system that reliably predicts her specific risk of uterine rupture would be even more useful, since women are probably more concerned about placing themselves and their babies at high risk of serious complications than whether or not they successfully TOLAC [15,16]. No randomized, controlled trials have been published. A systematic review of observational studies concluded that no system was sufficiently reliable to be clinically useful, and high quality data to guide decision making were sparse [17].

If additional pregnancies are planned, women also need to consider the risks associated with multiple repeat cesarean deliveries when making the decision for ERCD or TOLAC. (See "Repeat cesarean delivery", section on 'Is there an unsafe number of repeat cesarean deliveries?',Decision aids can help women with a previous cesarean delivery decide whether to attempt a TOLAC or choose ERCD [18-20]. Such aids appear to increase knowledge and reduce decisional uncertainty and anxiety, but their effect on the TOLAC rate remains unclear. A variety of paper based, video, and computer based decision aids have been created for this purpose. They provide structured information about the possible outcomes from different routes of delivery and the probability of these outcomes; some also factor in the woman's preferences and values.

One study that attempted to determine a preferred route of delivery used a decision analysis model to study this question. The model favored TOLAC if the chance of success was greater than 50 percent and if the desire for additional pregnancies was 10 to 20 percent [21].

CANDIDATES FOR TOLAC

Appropriate candidates — The American College of Obstetricians and Gynecologists' (ACOG) guidelines for identifying women who are potential candidates for TOLAC include the following criteria [7]:

• No traditional contraindications to labor or vaginal birth
• One previous low transverse uterine incision and no other uterine scars.
• A clinically adequate pelvis
• No history of uterine rupture
• A physician immediately available throughout active labor who is capable of making the decision for and performing an emergency cesarean delivery
• Availability of anesthesia and nursing personnel for emergency cesarean delivery

Possibly appropriate candidates — It may also be reasonable to offer TOLAC to other women, but there is less evidence confirming safety in these situations [7]:

• Two previous cesarean deliveries and a previous vaginal delivery. Studies have consistently shown that a vaginal delivery prior to cesarean birth reduced the risk of uterine rupture during TOLAC by about 50 percent [5,22-24].
• Suspected macrosomia and a previous vaginal delivery [25-28]. Most women with a macrosomic infant and a previous vaginal delivery have a successful TOLAC. In women without a previous vaginal birth, the success rate appears to be lower (<50 versus ≥60 percent) and the risk of uterine rupture appears to be higher (3.6 versus ≤1 percent) [28].
• Gestation beyond 40 weeks [29-31]. There are few data on the safety or efficacy of TOLAC in postterm pregnancy; no evidence-based recommendations for management of such pregnancies are available [32]. Although the rate of successful TOLAC appears to be lower after the estimated date of delivery (EDD) (69 versus 78 percent before the EDD) [31], the risk of uterine rupture does not appear to be related to pregnancy duration (ie, less than versus more than 40 weeks).
• Low vertical incision [33-36]. A low vertical hysterotomy is one that involves only the lower uterine segment; given the subjective nature of this assessment, this classification may be misleading. Although data are limited, women with a previous low vertical hysterotomy undergoing TOLAC appear to have similar outcomes to those with previous low transverse incisions.
• Unknown type of previous hysterotomy [23,37-39]. Most unknown hysterotomy incisions are low transverse incisions; thus, most women with an unknown scar type are at low risk of uterine rupture with TOLAC. However, women with an unknown type of hysterotomy in the setting of risk factors for a previous classical or T incision (preterm breech birth before 28 weeks of gestation, transverse lie) should be considered at higher risk for uterine rupture with TOLAC.
• Twin gestation. Women with twins have TOLAC outcomes similar to those of women with singleton pregnancies; however, the available data consist of too few twin pregnancies undergoing TOLAC to provide confidence in the safety of TOLAC in these gestations. (See "Delivery of twin gestations", section on 'Trial of labor after previous cesarean delivery'.)

Contraindications — TOLAC should not be attempted in women at high risk of uterine rupture or with contraindications to labor or vaginal birth. Some examples of women at higher risk of complications during a TOL include [7]:

• Prior classical or T-shaped uterine incision or extensive transfundal uterine surgery (eg, myomectomy) (see 'Uterine rupture' below.
• Previous uterine rupture (the frequency of repeat rupture varies from 6 to 33 percent, depending upon the site of the initial rupture)
• Medical or obstetrical complications that preclude vaginal birth (eg, placenta previa)
• Inability to perform emergency cesarean delivery due to factors related to the facility, surgeon, anesthesia, or nursing staff
• Two or more prior uterine scars and no prior vaginal deliveries. However, this is controversial, as there are conflicting data regarding the magnitude of risk of uterine rupture and other complications after multiple versus a single previous cesarean delivery [5,22-24] (see 'Uterine rupture' below.

SUCCESS RATE — There are no randomized trials comparing TOLAC with ERCD; several large observational studies have provided outcome data, which have been generally reassuring. Approximately 60 to 80 percent of TOLACs result in successful vaginal birth [5,11,40-42]; the pooled vaginal delivery rate from prospective studies is 76 percent and the largest population-based study reported a rate of 60 percent [43]. Some studies have reported higher success rates, but these are in selected populations, and likely reflect bias.

Several factors affect the probability of a successful vaginal delivery, as illustrated by pooled data (table 1) [7] and data from a large prospective study (table 2) [41]. The vaginal delivery rate probably exceeds 80 percent in the following subgroups:

• Previous successful TOLAC [41,44-49]

• Previous vaginal delivery [40,41,44-46,50]

• Previous cesarean delivery for nonvertex presentation [41,44,50,51]

• Women with spontaneous onset of labor [40,41,43,44,52,53]

Other factors that have been associated with successful TOLAC are young maternal age and favorable cervix [43,50,54].

The following characteristics decrease the probability of successful TOLAC to below 75 percent:

• Increasing number of cesarean deliveries [5,22-24,44,51]. In a multicenter study, the rate of successful TOLAC dropped significantly as the number of prior cesarean deliveries increased from one to four: 74, 67, 63, and 55 percent, respectively.
• Failure to progress. Women delivered abdominally because of failure to progress have the lowest rate of successful TOLAC, although approximately two-thirds are delivered vaginally [37,51,55]. The likelihood of successful TOLAC is higher if the cesarean was performed in the latent phase of labor, and it is lower if it was performed after full dilatation [56,57] or if the neonate's birthweight has increased compared with the prior pregnancy [58]. In one study, as an example, the rate of successful TOLAC was significantly higher after a prior cesarean delivery where maximum cervical dilatation was 6 to 9 cm than when full dilatation was achieved: 73 versus 13 percent [57].
• Use of oxytocin for induction or augmentation of labor [43].

The combination of previous cesarean for failure to progress or nonreassuring tracing, no previous vaginal delivery, and induced labor has a particularly poor prognosis, fewer than 50 percent of such women achieved a successful TOLAC (table 2) [44].

Other factors that have been associated with a lower success rate are gestational age greater than 40 weeks, birthweight greater than 4000 g, interdelivery interval less than 19 to 24 months, and maternal obesity [43,59-62].

X-ray pelvimetry findings are not predictive of outcome [63].

COMPLICATIONS

Composite complication rates — Although there are no data from randomized trials, a meta-analysis that included both prospective and retrospective studies found maternal morbidity did not differ significantly between women planning TOLAC and those planning ERCD (6.7 versus 4 percent) [42]. Large prospective studies have generally shown that TOLAC is associated with a higher relative risk of major complications than ERCD, but the absolute risk is small and mostly related to uterine rupture (see 'Uterine rupture' below.

Among women who are highly likely to have a successful TOLAC, such as those with a prior vaginal delivery, composite morbidity with TOLAC is as low or lower than with ERCD [64].

• The largest prospective study of complication rates in women giving birth after one or more previous cesarean deliveries included over 33,000 women with singleton pregnancies (TOLAC n = 17,898, ERCD n = 15,801) [13]. Women with obstetrical indications (eg, placenta previa, breech) for repeat cesarean delivery were excluded. TOLAC was associated with a significantly higher rate of maternal and perinatal adverse events than ERCD:

• - Maternal composite morbidity (uterine rupture, dehiscence, hysterectomy, thromboembolic disease, transfusion, endometritis, death): 5.5 versus 3.6 percent; OR 1.56, 95% CI 1.41-1.74.
• - Uterine rupture occurred in 124 women (0.7 percent) undergoing TOL, but no woman in the ERCD group. Of note, 100 of the 124 uterine ruptures occurred in women receiving a uterotonic drug (oxytocin, prostaglandin).
• - In the TOLAC group, there were 16 antepartum fetal demises ≥39 weeks, two intrapartum stillbirths, and 12 infants with hypoxic-ischemic encephalopathy. The corresponding numbers in the ERCD group were 5, 0, and 0. Overall, 588 ERCDs would need to be done to prevent one poor perinatal outcome [65].

• Another prospective longitudinal study including over 6000 women (TOLAC n = 3249, ERCD n = 2889) found that major maternal complications (combined incidence of uterine rupture, hysterectomy, operative injury) were twice as common in the TOLAC group than in the ERCD group (1.6 versus 0.8 percent, OR 1.8, 95% CI 1.1-3.0) [12]. However, minor complications occurred less often in the TOLAC group: puerperal fever (5.3 versus 6.4 percent with ERCD), abdominal wound infection (1.3 versus 2.2 percent with ERCD). The overall rate of maternal morbidity (8 percent) did not differ between groups.

Uterine rupture — Data from observational studies show that TOLAC is associated with a higher risk of uterine rupture than ERCD, and women who have a uterine rupture are at high risk of maternal morbidity and fetal morbidity/mortality.

Symptomatic uterine rupture generally refers to complete spontaneous disruption of the uterine wall (serosa, myometrium, decidua), which is usually accompanied by bleeding and sometimes accompanied by extrusion of part or all of the fetal-placental unit. Perinatal mortality/morbidity is higher in fetuses who experience complete extrusion into the maternal abdomen than in those who are not extruded (10/28 versus 5/58) [66]. Some authors do not distinguish between symptomatic defects and those that have no maternal/fetal consequences and where the full thickness of the uterine wall may or may not be disrupted (often termed asymptomatic uterine rupture, dehiscence, incomplete rupture). The absence of a standard definition sometimes impedes comparison of data from different studies.

Incidence — Most uterine ruptures occur in women with prior cesarean deliveries, but rupture can occur after other types of uterine incisions, such as after a myomectomy, or, rarely, in an unscarred uterus [67]. The actual magnitude of risk depends upon several factors, the most important of which is the location of the previous hysterotomy incision. The rate of rupture reported by incision location is: classical (4 to 9 percent), T-shaped (4 to 9 percent), low vertical (1 to 7 percent), and low transverse (0.2 to 1.5 percent) [7,13,68,69].

The reported rate also depends upon whether symptomatic or asymptomatic ruptures are being tabulated. Evaluation of data from multiple trials showed that:

• Symptomatic uterine rupture occurs in 0 to 7.8 per 1000 TOLACs (pooled weighted mean 3.16 per 1000 TOLACs) [43]. On average, the incremental risk of rupture with TOLAC compared with ERCD is 2.7 per 1000.
• Asymptomatic uterine rupture occurs in 0 to 19 per 1000 TOLACs (mean weighted average 12.6 per 1000 TOLACs) [43]. This rate is comparable to that in women undergoing ERCD [11,43,70].

Rupture during labor may also occur in an unscarred uterus, but this is rare in developed countries. In two large series the incidence was 0.6 and 0.7 per 10,000 deliveries [71,72]. In one of these series, intrapartum rupture of an unscarred uterus occurred in 10 of 168,491 deliveries (0.6/10,000 deliveries) and was associated with one or more of the following: oxytocin use (four cases), prostaglandin use (three cases), utilization of vacuum or forceps (three cases), grand multiparity (two cases), malpresentation (two cases) [71]. In the other series, there were 25 uterine ruptures in women with unscarred uteruses and these events accounted for 13 percent of ruptures in this study [72]. The incidence of rupture in unscarred and scarred uteruses was 0.7 and 5.1 per 10,000 deliveries, respectively.

In undeveloped countries, the prevalence of uterine rupture is much higher and many uterine ruptures occur in the unscarred uterus due to obstructed labor [73-75].

Risk factors — No single or combination of risk factors is sufficiently reliable to be clinically useful for prediction of uterine rupture [15,16]. Although numerous risk factors have been cited for uterine rupture during labor in women with a previous cesarean delivery [5,22-24,66,69,76-85], the risk factors have not been consistent across studies, and the studies were generally hampered by having only a few cases of uterine rupture [54]. Two studies attempted to use models to predict the risk of uterine rupture in women with one prior cesarean. One prospective cohort study involving 11,855 patients and one case-control study failed to show any predictive factors or models that would be helpful in predicting the risk of rupture [15,16].

Several common risk factors for uterine rupture are discussed below:

• No previous vaginal delivery — There is consistent evidence from observational studies that a previous successful vaginal delivery is predictive of scar integrity during TOLAC [5,49,69,86,87]. In two multihospital series, a prior vaginal delivery reduced the odds of uterine rupture by about 50 percent [5,86].
• Multiple prior cesarean deliveries — The risk of uterine rupture during TOLAC after multiple versus a single prior cesarean delivery has been addressed in four large observational studies, with discordant results [5,22-24]. Three studies reported a significantly increased risk of rupture in women with two or more cesarean deliveries compared with those with one prior cesarean delivery (1.7 to 3.7 versus 0.6 to 0.8 percent) [22-24]; the fourth study reported a similar risk for both groups (0.9 and 0.7 percent) [5]. Since these are observational studies, multiple factors might account for the discordant findings, including differences in the characteristics of the populations and in obstetric management. As an example, the TOLAC rate in these series ranged from 9 to 49 percent.
• Induced or augmented labor — There is consistent evidence that inducing labor increases the rate of uterine rupture compared with spontaneous labor. A systematic review of pooled data from controlled studies of women with prior cesarean delivery reported the odds ratio for uterine rupture was 6.15 (95% CI 0.74-51.4) for induction compared to spontaneous labor [88]. All but one study had a point estimate indicating increased risk of rupture, but confidence intervals were wide. A large, well designed prospective study, not included in the pooled estimate, reported an odds ratio of 2.86 (95% CI 1.75-4.67) (absolute risk of rupture with induction: 1 percent versus 0.4 percent with spontaneous labor) [13]. (See "Induction of labor in women with prior cesarean delivery".)

The effect of oxytocin augmentation on uterine rupture risk during TOLAC is unclear, as studies have reported discordant results [5,13,89-91]. A case-control study of women given oxytocin for induction or augmentation of TOLAC reported the risk of rupture was related to the maximum dose of oxytocin used [92].

• Term delivery — A prospective four-year observational study of women with a singleton gestation and a prior cesarean delivery at 19 academic centers found that, compared to TOLAC at term, preterm TOLAC had a significantly lower risk of uterine rupture (0.34 versus 0.74 percent in term TOLAC) and uterine dehiscence (0.26 versus 0.67 percent in term TOLAC) [93]. The likelihood of vaginal birth with TOLAC in preterm pregnancies was comparable to that in term gestations.
• Other — A thick uterine scar (greater than 3.5 mm) on ultrasound examination at 37 weeks of gestation is predictive of scar integrity at TOLAC [94]. The risk of rupture is increased with a thin uterine scar, but the quantitative risk associated with scar thickness has not been determined.

In addition to a thin uterine scar, purported risk factors for rupture include multiple gestation, macrosomia, post-cesarean delivery endometritis, single versus double layer uterine closure of the hysterotomy incision, and short interpregnancy interval. Further study is needed before conclusions about causality can be made.

Of note, uterine scarring may be due to events other than previous cesarean delivery; these include fetal surgery, uterine surgery (myomectomy, repair of congenital defects, excision of cornual ectopic pregnancy), uterine perforation, and prior uterine infection. Rupture of a scarred or unscarred uterus has also been associated with trauma (eg, traffic accident, domestic violence, gun shot wound), obstetric maneuvers (eg, internal version and breech extraction, instrumental delivery, manual removal of the placenta), obstructed or neglected labor, and induction (especially with use of prostaglandins or grand multiparity). There also may be an association with placenta accreta, congenital uterine anomalies, and Ehlers-Danlos syndrome [67].

Women with a previous low transverse cesarean delivery performed preterm have a similar risk of uterine rupture at TOLAC as those with a previous term cesarean delivery [95].

Clinical manifestations — Rupture of lower segment scars usually occurs during labor, but may occur antepartum, particularly with classical uterine scars [96].

Fetal bradycardia is the most common and characteristic clinical manifestation of uterine rupture, occurring in 33 to 70 percent of symptomatic cases [72,97-100]. Variable or late decelerations may precede the bradycardia, but there is no fetal heart rate pattern pathognomonic of rupture. Perinatal mortality/morbidity is higher in fetuses who experience complete extrusion into the maternal abdomen than in those who are not extruded (10/28 versus 5/58) [66].

Maternal manifestations are variable. In women with known uterine scarring or trauma, uterine rupture should always be strongly considered if constant abdominal pain and signs of intraabdominal hemorrhage are present. Vaginal bleeding is not a cardinal symptom, as it may be modest, despite major intraabdominal hemorrhage. Other clinical manifestations include maternal tachycardia, hypotension ranging from subtle to severe (hypovolemic shock), cessation of uterine contractions, loss of station of the fetal presenting part, uterine tenderness, and change in uterine shape. Pain may be partially or completely masked by regional analgesia.

Postpartum, uterine rupture is characterized by pain and persistent vaginal bleeding despite use of uterotonic agents. Hematuria may occur if the rupture extends into the bladder.

Diagnosis — The diagnosis is typically made at laparotomy by observation of complete disruption of all uterine layers with active bleeding. Uterine wall "thinning," asymptomatic dehiscence incidentally observed during cesarean delivery, or palpation of a defect in the lower uterine segment after vaginal delivery are generally not considered uterine ruptures.

Differential diagnosis includes placental abruption, subcapsular liver hematoma, splenic rupture, and uterine torsion [67].

Outcome — Uterine rupture is the major cause of excess maternal morbidity with TOLAC (table 3). The most comprehensive review of complications from TOLAC (literature review of 880 uterine ruptures in over 140,000 TOLACs) reported the following findings [101]:

• Blood transfusion (1.8 per 1000 TOLACs)
• Umbilical artery pH less than 7.00 (1.5 per 1000 TOLACs)
• Hysterectomy (0.9 per 1000 TOLACs)
• Genitourinary injury (0.8 per 1000 TOLACs)
• Perinatal death associated with uterine rupture (0.4 per 1000 TOLACs)
• Maternal death (0.02 per 1000 TOLACs)

Although cases of low pH were identified, this series did not determine the rate of hypoxic ischemic neonatal encephalopathy. Umbilical artery pH less than 7.00 is necessary, but not sufficient, for the diagnosis. A large prospective study identified 12 cases of neonatal encephalopathy among almost 18,000 infants exposed to TOLAC; there were no cases among the almost 16,000 infants delivered by ERCD [13]. Seven of the 12 cases occurred in the setting of symptomatic uterine rupture. The overall rate of neonatal encephalopathy related to uterine rupture in women undergoing TOLAC was 0.46 per 1000 women. This rate of neonatal encephalopathy does not exceed that described for the general obstetrical population [102]. (See "Clinical features, diagnosis, and treatment of neonatal encephalopathy".)

The risk of hysterectomy is probably the same for women who attempt TOLAC and those who opt for ERCD [12,13,43,103,104]. A report on Vaginal Birth after Cesarean prepared for the Agency for Healthcare Research and Quality concluded that the risk of hysterectomy was the same in both groups (0.2 percent) [43]. They based their opinion on one study [12] because it was the only study to date that excluded ERCDs performed for medical or obstetrical indications associated with a higher risk of hysterectomy (eg, placenta previa). A subsequent large prospective study with the same exclusion criteria also found the risk of hysterectomy was similar for the two groups [13]. Although a meta-analysis that compared the outcome of TOLAC to that of ERCD concluded women undergoing TOLAC had a significantly lower rate of hysterectomy (OR 0.39, 95% CI 0.27-0.57; 15 studies, almost 48,000 women) [103], this analysis did not exclude women at high baseline risk of hysterectomy.

Management — If the maternal and fetal conditions are stable and the diagnosis is uncertain, then close observation is warranted with continuous electronic fetal monitoring and frequent assessment of maternal vital signs, progression of labor, and abdominal/uterine examination. We suggest placement of an intrauterine pressure catheter, as well. If there are signs suggestive of fetal or maternal compromise, then an emergency laparotomy should be undertaken. Early suspicion, detection, and treatment of uterine rupture provide the best chance for mother and fetus.

After delivery of the fetus and placenta, the site and extent of rupture should be thoroughly evaluated and a decision made to repair the defect or perform hysterectomy (total or supracervical). There are insufficient data to recommend routine performance of one or the other. A hysterectomy is often performed to control massive hemorrhage, and often minimizes blood loss, but obviously results in sterility. For women who desire to preserve their childbearing potential, an attempt at repair of the rupture is reasonable if the patient is hemodynamically stable and the rupture is of a size and shape that would allow a successful repair. Tubal ligation at the time of repair is only indicated if sterilization had already been planned.

Subsequent pregnancy — Although there are small series describing uncomplicated pregnancy and delivery after repair of a uterine rupture [105-108], there are inadequate data to assess the safety of this approach. Furthermore, recurrent rupture has been reported [108-111].

Each patient must be counseled based on her individual circumstances (size and location of rupture). If pregnancy is undertaken, we believe delivery should be via a scheduled ERCD.

Perinatal death — Even in the absence of uterine rupture, the risk of perinatal death appears to be higher in women who attempt TOLAC than in women who undergo ERCD. This was illustrated in a well-designed, population-based cohort study of Scottish births between 1992 and 1997 [112]. The authors evaluated the overall risk of perinatal death at 37 to 43 weeks of gestation in four groups of women with singleton infants in cephalic presentation (table 4). The lowest rate of delivery-related perinatal death occurred among women undergoing ERCD: only one fetal/neonatal death occurred among 9014 such women. The absolute risk of perinatal death with TOLAC was higher, but still low at 1 in 775 births (95% CI, as high as 1 in 500) and not significantly different from that of nulliparous women with unscarred uteri attempting vaginal birth (perinatal death 1 in 1016 births).

Based upon this study and another [12], the Agency for Healthcare Research and Quality report on Vaginal Birth after Cesarean concluded that the risk of infant death was probably increased with TOLAC compared with ERCD (13 to 90 infant deaths per 10,000 TOLACs versus 1 to 50 infant deaths per 10,000 ERCDs) [43].

LABOR AND DELIVERY — Delivery in a hospital rather than a birthing center is recommended so that immediate access to an operating room is available, if needed [113]. The number of births in a hospital may be an important factor in outcome after rupture. A large study from Scotland found that although the rate of uterine rupture during TOLAC was similar between low volume and high volume hospitals, perinatal mortality after rupture was significantly higher in hospitals with fewer than 3000 births per year [86].

Most experts recommend continuous electronic fetal heart rate and uterine monitoring during labor; external monitoring appears to be sufficient [7]. The use of epidural anesthesia appears to be safe in women attempting TOLAC [7,66,97].

Factors that may contribute to uterine scar disruption include mode of labor onset (spontaneous or induced), the type of uterine incision previously performed (eg, low transverse or classical), the duration and dose of oxytocin administration, and the choice of cervical ripening technique.

ACOG recommends that misoprostol (prostaglandin E1) NOT be used for cervical ripening or labor induction in women with prior uterine incisions [7,114,115], and strongly discourages use of other prostaglandins, as well [7,116]. They do not make a specific recommendation regarding use of oxytocin. These issues are discussed in detail separately. (See "Induction of labor in women with prior cesarean delivery".)

It is not necessary to routinely examine the uterus for dehiscence postpartum [7].

INFORMATION FOR PATIENTS — Educational materials on this topic are available for patients. (See "Patient information: Cesarean delivery" and "Patient information: Vaginal birth after cesarean delivery (VBAC)".) We encourage you to print or e-mail these topic reviews, or to refer patients to our public web site, www.uptodate.com/patients, which includes these and other topics.

SUMMARY AND RECOMMENDATIONS

• Pregnant women with a prior cesarean birth must choose between elective repeat cesarean delivery (ERCD) and a trial of labor after cesarean (TOLAC). We feel the decision regarding mode of delivery must be made by each woman after informed discussion with her physician. The final assessment depends upon her individual clinical circumstances, tolerance for the various risks involved, and ability to give birth in a facility capable of emergency cesarean delivery if required. (See 'Choosing the route of delivery' above.)

• Based on available data, the risk of major maternal complications and perinatal death are higher among women undergoing TOLAC. (See 'Complications' above.)

Additional factors which must be considered are: (see 'Complications' above:

• - Uterine rupture (3.16 per 1000 TOLACs, incremental risk over ERCD 2.7 per 1000).
• - Hysterectomy (0.2 percent, same as with ERCD. Risk of hysterectomy due to uterine rupture 3.4 to 4.8 per 10,000 TOLACs).
• - Perinatal death (13 to 90 infant deaths per 10,000 TOLACs versus 1 to 50 infant deaths per 10,000 ERCDs).

• Approximately 75 percent of women who attempt TOLAC will be successful; this rate varies up or down depending upon the clinical situation that led to the first cesarean birth. It is highest in women with a previous successful TOLAC, previous vaginal delivery, previous cesarean delivery for nonvertex presentation, and women with spontaneous onset of labor. The combination of previous cesarean for failure to progress or nonreassuring tracing, no previous vaginal delivery, and induced labor has a particularly poor prognosis, fewer than 50 percent of such women achieved a successful TOLAC. (See 'Success rate' above.)