Uterine Rupture in Pregnancy 

Uterine rupture in pregnancy is a rare and often catastrophic complication with a high incidence of fetal and maternal morbidity. Several factors are known to increase the risk of uterine rupture, but, even in high-risk subgroups, the overall incidence of uterine rupture is low. From 1976-2009, 20 peer-reviewed publications that described the incidence of uterine rupture reported 1,864 cases among 2,863,330 pregnant women, yielding an overall uterine rupture rate of 1 in 1,536 pregnancies (0.07%).

The initial signs and symptoms of uterine rupture are typically nonspecific, which makes diagnosis difficult and sometimes delays definitive therapy. From the time of diagnosis to delivery, only 10-37 minutes are available before clinically significant fetal morbidity becomes inevitable. Fetal morbidity occurs as a result of catastrophic hemorrhage, fetal anoxia, or both. The inconsistent premonitory signs and the short time for instituting definitive therapeutic action make uterine rupture a fearful event.

Definition

Uterine rupture during pregnancy is a rare occurrence that frequently results in life-threatening maternal and fetal compromise, whereas uterine scar dehiscence is a more common event that seldom results in major maternal or fetal complications. By definition, uterine scar dehiscence constitutes separation of a preexisting scar that does not disrupt the overlying visceral peritoneum (uterine serosa) and that does not significantly bleed from its edges. In addition, the fetus, placenta, and umbilical cord must be contained within the uterine cavity, without a need for cesarean delivery due to fetal distress.

By contrast, uterine rupture is defined as a full-thickness separation of the uterine wall and the overlying serosa. Uterine rupture is associated with (1) clinically significant uterine bleeding; (2) fetal distress; (3) expulsion or protrusion of the fetus, placenta, or both into the abdominal cavity; and (4) the need for prompt cesarean delivery and uterine repair or hysterectomy.

Although a scar from cesarean delivery is a well-known risk factor for uterine rupture, most events that involve disruption of the uterine scar result in uterine-scar dehiscence rather than frank uterine rupture. These 2 entities must be clearly distinguished because the options for clinical management and outcomes differ significantly.

Sources of information and study selection

The peer-reviewed literature was searched using the PubMed, Medline, and Cochrane databases for all relevant articles published in the English language. The search terms were uterine rupture, pregnancy and prior cesarean section, vaginal birth after cesarean, VBAC, trial of labor (TOL), trial of labor after cesarean (TOLAC) uterine scar dehiscence, and pregnancy and myomectomy. Standard reference tracing was also used.

Articles published from 1976-2010 that described the incidence of uterine rupture and that included sufficient information regarding the authors' definitions of uterine rupture and of uterine-scar dehiscence were incorporated for review. All studies were observational or reviews. A total of 109 published articles were included for data extraction and analysis.

Incidence and risk factors

Meta-analysis of pooled data from 20 studies in the peer-reviewed medical literature published from 1976-2009 indicated an overall incidence of pregnancy-related uterine rupture of 1 per 1,536 pregnancies (0.07%). When the studies were limited to a subset of 8 that provided data about the spontaneous rupture of unscarred uteri in developed countries, the rate was 1 per 8,434 pregnancies (0.012%).

Congenital uterine anomalies, multiparity, previous uterine myomectomy, the number and type of previous cesarean deliveries, fetal macrosomia, labor induction, uterine instrumentation, and uterine trauma all increase the risk of uterine rupture, whereas previous successful vaginal delivery and a prolonged interpregnancy interval after a previous cesarean delivery may confer relative protection. In contrast to the availability of models to predict the potential success of a TOL after a prior cesarean section, accurate models to predict the person-specific risk of uterine rupture for individuals are not available.

Major patient characteristics for determining risk of uterine rupture are noted below.

Uterine status is either native (unscarred) or scarred. Scarred status may include previous cesarean delivery, including the following:

• Single low transverse (further subcategorized by 1-layer or 2-layer hysterectomy closure)
• Single low vertical
• Classic vertical
• Multiple previous cesarean deliveries

Scarred status may also include previous myomectomy (transabdominal or laparoscopic).

Uterine configuration may be normal or may involve a congenital uterine anomaly.

Pregnancy considerations include the following:

• Grand multiparity
• Maternal age
• Placentation (accreta, percreta, increta, previa, abruption)
• Cornual (or angular) pregnancy
• Overdistension (multiple gestation, polyhydramnios)
• Dystocia (fetal macrosomia, contracted pelvis)
• Trophoblastic invasion of the myometrium (hydatidiform mole, choriocarcinoma)

Previous pregnancy and delivery history may include the following:

• Previous successful vaginal delivery
• No previous vaginal delivery
• Interdelivery interval

Labor status is determined as follows:

• Not in labor
• Spontaneous labor
• Induced labor - With oxytocin, with prostaglandins
• Augmentation of labor with oxytocin
• Duration of labor
• Obstructed labor

Obstetric management considerations include the following:

• Instrumentation (forceps use)
• Intrauterine manipulation (external cephalic version, internal podalic version, breech extraction, shoulder dystocia, manual extraction of placenta)
• Fundal pressure

Uterine trauma includes the following:

• Direct uterine trauma (motor vehicle accident, fall)
• Violence (gunshot wound, blunt blow to abdomen)

The normal, unscarred uterus is least susceptible to rupture. Grand multiparity, neglected labor, malpresentation, breech extraction, and uterine instrumentation are all predisposing factors for uterine rupture. A 10-year Irish study by Gardeil et al showed that the overall rate of unscarred uterine rupture during pregnancy was 1 per 30,764 deliveries (0.0033%). No cases of uterine rupture occurred among 21,998 primigravidas, and only 2 (0.0051%) occurred among 39,529 multigravidas with no uterine scar.^[1]

A meta-analysis of 8 large, modern (1975-2009) studies from industrialized countries revealed 174 uterine ruptures among 1,467,534 deliveries. This finding suggested that the modern rate of unscarred uterine rupture during pregnancy is 0.012% (1 of 8,434). This rate of spontaneous uterine rupture has not changed appreciably over the last 40 years, and most of these events occur at term and during labor. An 8-fold increased incidence of uterine rupture of 0.11% (1 in 920) has been noted in developing countries. This increased incidence of uterine rupture has been attributed to a higher-than-average incidence of neglected and obstructed labor due to inadequate access to medical care. When one assesses the risk of uterine rupture, this baseline rate of pregnancy-related uterine rupture is a benchmark that must be used as a point of reference.

Effect of maternal parity

Many authors have considered multiparity a risk factor for uterine rupture. Golan et al noted that, in 19 of 61 cases (31%), uterine rupture occurred in women with a parity of more than 5.^[2] Schrinsky and Benson found that 7 of 22 women (32%) who had unscarred uterine rupture had a parity of greater than 4.^[3] In a study by Mokgokong and Marivate, the mean parity for women who had pregnancy-related uterine rupture was 4.^[4] Despite the apparent increase in the risk of uterine rupture associated with high parity, Gardeil et al found only 2 women with uterine rupture among 39,529 multigravidas who had no previous uterine scar (0.005%).^[1]

Rupture of the unscarred uterus before labor versus during labor

Schrinsky and Benson reported 22 cases of uterine rupture in gravidas with unscarred uteri. Nineteen occurred during labor (86%), and 3 occurred before labor (14%). This percentage was markedly different from that of gravidas with a previous uterine scar, for whom the timing of uterine rupture between labor and the antepartum period was nearly evenly distributed.^[3]

Oxytocin augmentation and induction of labor in the unscarred uterus

Although distinguishing oxytocin use for labor induction versus labor augmentation is useful, many researchers who have investigated the use of oxytocin and the risk of uterine rupture in unscarred uteri unfortunately have not made this distinction.

In 1976, Mokgokong and Marivate reported 260 uterine ruptures among 182,807 deliveries that involved unscarred uteri, and 32 of the 260 (12%) were associated with oxytocin use.^[4] Rahman et al similarly found that oxytocin was administered in 9 of 65 cases (14%) that involved unscarred uterine rupture.^[5] Golan et al noted that, among 126,713 deliveries, oxytocin was used in 26 of 61 cases (43%) that involved unscarred uterine rupture.^[2] However, Plauche et al attributed only 1 of 23 unscarred uterine ruptures (4%) to the use of oxytocin.^[6]

Therefore, the increased risk of uterine rupture attributable to the use of oxytocin in gravidas with unscarred uteri is uncertain. However, women who have had a cesarean delivery appear to have an increased risk of uterine rupture associated with the use of oxytocin both when it is used for labor augmentation and labor induction (see Table 1).

Congenital uterine anomalies

In a review article, Nahum reported that congenital uterine anomalies affected approximately 1 in 200 women.^[7] In such cases, the walls of the abnormal uteri tend to become abnormally thin as pregnancies advance, and the thickness can be inconsistent over different aspects of the myometrium.^{[8, 9, 10, 11]}

Ravasia et al reported an 8% incidence of uterine rupture (2 of 25) in women with congenitally malformed uteri compared with 0.61% (11 of 1,788) in those with normal uteri (P =.013) who were attempting VBAC.^[12] Both cases of uterine rupture involved labor induction with prostaglandin E2. Pregnancies implanted in the rudimentary horn of the uterus pose special risk for those women undergoing induction of labor, with a uterine rupture rate of up to 81% (387 of 475).^[13] Importantly, 80% of ruptures occurred before the third trimester, with 67% occurring during the second trimester.

In contrast, a study of 165 patients with Müllerian duct anomalies who underwent spontaneous labor after 1 prior cesarean delivery reported no cases of uterine rupture.^[14] Of note, in this study 36% (60 of 165) had only a minor uterine anomaly (arcuate or septate uterus), and 64% (105 of 165) had a major uterine anomaly (unicornuate, didelphys, or bicornuate uterus). Moreover, only 6% (10 of 165) of patients with Müllerian duct anomalies underwent induction of labor. The decision of induction of labor in women with congenitally anomalous uteri and a previous cesarean section must therefore be considered carefully given the higher incidence of uterine rupture reported in this patient population.

Although the uterine rupture rate in anomalous, unscarred uteri during pregnancy appears to be increased relative to that for normal uteri, the precise risk for different uterine malformations remains uncertain.

Previous myomectomy by means of laparotomy

Nearly all uterine ruptures that involved uteri with myomectomy scars have occurred during the third trimester of pregnancy or during labor. Only 1 case of a spontaneous uterine rupture has been reported before 20 weeks of gestation.^[15] Brown et al reported that among 120 term infants delivered after previous transabdominal myomectomy, no uterine ruptures occurred, and 80% of the infants were delivered vaginally.^[16] In contrast, Garnet identified 3 uterine ruptures among 83 women (4%) who had scars from a previous myomectomy and who underwent elective cesarean delivery because of previous myomectomy.^[17]

Such reports do not often delineate the factors that were deemed important for assessing the risk of subsequent uterine rupture (eg, number, size, and locations of leiomyomata; number and locations of uterine incisions; entry of the uterine cavity; type of closure technique). Further studies to investigate these issues are needed.

Previous laparoscopic myomectomy

Dubuisson et al reported 100 patients who underwent laparoscopic myomectomy and found 3 uterine ruptures during subsequent pregnancies.^[18] Only 1 rupture occurred at the site of the previous myomectomy scar, resulting in the conclusion that the risk of pregnancy-related uterine rupture attributable to laparoscopic myomectomy is 1% (95% CI, 0-5.5%). However, the rarity of spontaneous uterine rupture raises the issue of whether the 2 uterine ruptures at sites that were not coincident with previous myomectomy scars were attributable to the previous myomectomies. If so, a markedly higher 3% uterine rupture rate is associated with previous laparoscopic myomectomy.

Different authors reported no pregnancy-related uterine ruptures in 4 studies of 320 pregnancies in women who previously underwent laparoscopic myomectomy.^{[19, 20, 21, 22]} However, in all 4 studies, the number of patients who were allowed to labor was low, and a high percentage of deliveries were by scheduled cesarean delivery (80%, 79%, 75%, and 65%, respectively).

In a prospective study from Japan, there were no uterine ruptures among 59 patients with a successful vaginal delivery after a prior laparoscopic myomectomy.^[23] In a multicenter study in Italy with 386 patients who achieved pregnancy after laparoscopic myomectomy, there was 1 recorded spontaneous uterine rupture at 33 weeks' gestation (rupture rate 0.26%).^[24]

Uterine rupture has been reported to occur as late as 8 years after laparoscopic myomectomy.^[25] This finding suggests that additional investigations with long-term follow-up are needed.

The effect of previous cesarean delivery on the risk of uterine rupture has been studied extensively. In a meta-analysis, Mozurkewich and Hutton used pooled data from 11 studies and showed that the uterine rupture rate for women undergoing a TOL after previous cesarean delivery was 0.39% compared with 0.16% for patients undergoing elective repeat cesarean delivery (odds ratio [OR], 2.10; 95% CI, 1.45-3.05). Restricting the meta-analysis to 5 prospective cohort trials generated similar results (OR, 2.06; 95% CI, 1.40-3.04).^[26]

Hibbard et al examined the risk of uterine rupture after previous cesarean delivery in 1324 women who underwent a subsequent TOL. They reported a significant difference in the risk of uterine rupture between women who achieved successful vaginal birth compared with women in whom attempted vaginal delivery failed (0.22% vs 1.9%; OR, 8.9; 95% CI, 1.9-42).^[27] The effect of previous cesarean delivery on the rate of subsequent pregnancy-related uterine rupture can be further examined according to additional subcategories, which are summarized in Table 1.

Previous classic cesarean delivery

Classic cesarean delivery via vertical midline uterine incision is infrequently performed in the modern era and currently account for 0.5% of all births in the United States.^[28] In a meta-analysis, Rosen et al reported an 11.5% absolute risk of uterine rupture (3 of 26 cases) in women with classic vertical cesarean scars who underwent an unplanned TOL.^[29] For women who underwent repeat cesarean section, Chauhan et al reported that the uterine rupture rate for 157 women with prior classical uterine cesarean scars was 0.64% (95% CI, 0.1-3.5%). All patients in that study underwent repeat cesarean delivery, but a high rate of preterm labor resulted in 49% of the patients being in labor at the time of their cesarean delivery.^[28]

Landon et al reported a 1.9% absolute uterine rupture rate (2 of 105 cases) in women with a previous classic, inverted T, or J incision who either presented in advanced labor or refused repeat cesarean delivery.^[30] These rates of frank uterine rupture in women with classic cesarean deliveries are in contrast to the higher rates of 4-9% that the American College of Obstetricians and Gynecologists (ACOG) had historically reported for women with these types of uterine scars. However, Chauhan et al observed a 9% rate of asymptomatic uterine scar dehiscence (95% CI, 5-15%).^[28] This result suggests that disruptions of uterine scars might have been misclassified as true ruptures instead of dehiscences in previous studies; this error may explain the bulk of the discrepancy.

Previous low vertical cesarean delivery

A meta-analysis of pooled data from 5 studies demonstrated a 1.1% absolute risk (12 of 1,112 cases) of symptomatic uterine rupture in women undergoing a TOL with a low vertical cesarean scar.^{[31, 32, 33, 34, 30]} Compared to women with low transverse cesarean scars, these data suggest no significantly increased risk of uterine rupture or adverse maternal and perinatal outcomes.

Interpretation of these studies is hampered by inconsistencies in how high the lower uterine segment could be cut before it was considered a classic incision. Even when the lower uterine segment is already well developed as a result of active labor, a low vertical incision of adequate length is often impossible to permit fetal delivery. Naef et al arbitrarily defined a 2-cm extension into the upper segment as a classic extension. For 322 pregnancies that occurred after a low vertical cesarean delivery, the overall rate of uterine rupture was 0.62%. This rate could be further divided as 1.15% for 174 women who underwent a TOL compared with no ruptures among 148 women who underwent elective repeat cesarean delivery.^[31]

Unknown uterine scar

In many instances, the type of incision used for a prior cesarean delivery cannot be confirmed due to unavailability of the operative report. Under these circumstances, the assessment of uterine rupture risk may sometimes be guided by the obstetric history to infer the most probable type of uterine scar. For example, a patient with a history of a preterm cesarean delivery at 28 weeks’ gestation has a much higher likelihood of having had a vertical uterine incision than a patient who underwent a cesarean section for an indication of arrest of fetal descent at term.

It has been argued that because most cesarean deliveries in the United States are accomplished via low-transverse uterine incisions, the risk of uterine rupture for patients with an unknown scar is similar to that for women who have previously undergone a low-transverse hysterotomy. This logic depends on the high ratio of low-transverse to vertical incisions performed for cesarean section, but it ignores the varying probability with which different types of uterine incisions are made under different obstetrical circumstances, as well as differences that occur due to varying medical resources and the prevailing local practitioner practices in countries other than the United States (eg, practices that occur in other countries, such as Mexico or Brazil). Nevertheless, the vast majority of cesarean deliveries performed in the United States are accomplished via low-transverse uterine incisions.

In a small case-control study of 70 patients by Leung et al, no association was found between an unknown uterine scar and the risk of uterine rupture; however, given the rarity of uterine rupture (see Table 1), this study was vastly underpowered to detect such a difference.^[35] Two additional, but similarly underpowered, case series have also reported comparable rates of uterine rupture and VBAC success in women with unknown uterine cesarean delivery scars versus those with documented previous low-transverse hysterotomies.^{[36, 37]} The Maternal-Fetal Medicine Units (MFMU) Network cesarean delivery registry reports a 0.5% risk (15 of 3,206) of uterine rupture for patients who underwent a TOL with an unknown uterine scar.^[30]

For cases in which there are 1 or 2 unknown prior uterine incisions, there is a single small, randomized, controlled trial by Grubb et al that compared labor augmentation with oxytocin (n=95) with no intervention (n=93) in women with prior cesarean deliveries involving either 1 or 2 unknown uterine incisions. Four uterine dehiscences and 1 uterine rupture occurred, all in the group that underwent labor augmentation. In the 1 case of uterine rupture, the unknown uterine scar was in a patient with 2 prior cesarean deliveries, one of which involved a vertical incision. Had the uterine scar status for this patient been known in advance, it would have represented a contraindication to a TOL.^[38]

Previous low transverse cesarean delivery

The risk of uterine rupture after a low transverse cesarean delivery varies depending on whether patients undergo a TOL or an elective repeat cesarean delivery and on whether labor is induced or spontaneous, as well as other factors. The vast majority of cesarean deliveries in the United States are of the low transverse type. For women who have had 1 previous cesarean delivery, examining the various risk factors for uterine rupture is instructive. These absolute risks for uterine rupture are discussed below, as well as in Table 1.

Previous cesarean delivery without a subsequent trial of labor

In a study of 20,095 women by Lydon-Rochelle et al, the spontaneous uterine rupture rate among 6,980 women with a single cesarean delivery scar who underwent scheduled repeat cesarean delivery without a TOL was 0.16%.^[39] This finding indicates that uteri with cesarean scars have an intrinsic propensity for rupture that exceeds that of the unscarred organ during pregnancy, which is 0.012% (OR increase of approximately 12-fold). Therefore, all other uterine rupture rates in women with a previous cesarean delivery should be referenced to this expected baseline rate.

Previous cesarean delivery with subsequent spontaneous labor

A study by Lydon-Rochelle et al showed that the uterine rupture rate among 10,789 women with a single previous cesarean delivery who labored spontaneously during a subsequent singleton pregnancy was 0.52%.^[39] This rate of uterine rupture implies an increased relative risk (RR) of 3.3 (95% CI, 1.8-6.0) for women who labor spontaneously compared with women who undergo elective repeat cesarean delivery.

In a study by Ravasia et al of 1,544 patients with a previous cesarean delivery who later labored spontaneously, the uterine rupture rate was 0.45%.^[40] Zelop et al found that, among 2,214 women with 1 previous cesarean delivery who labored spontaneously, the uterine rupture rate was 0.72%.^[41] The authors of this article performed a meta-analysis of 29,263 pregnancies from 9 studies from 1987-2004 and showed that the overall risk of uterine rupture was 0.44% for women who labor spontaneously after a previous cesarean delivery.

Previous cesarean delivery with subsequent augmentation of labor

Despite the clinical heterogeneity and different VBAC success rates for women undergoing spontaneous labor rather than either labor augmentation or induction, very few studies have stratified their data by labor augmentation versus labor induction and the data that do exist are conflicting. There is wide variance in the frequency of clinical use of oxytocin as well as in the dose and dosing schedules of oxytocin that are used. As a result, there is a paucity of specific evidence-based clinical guidelines for the use of oxytocin in VBAC trials.

Previous cesarean delivery with subsequent induction of labor

Emerging data indicate that induction of labor after a prior cesarean delivery appears to be associated with an increased risk of uterine rupture.

Zelop et al found that the rate of uterine rupture in 560 women who underwent labor induction after a single previous cesarean delivery was 2.3% compared with 0.72% for 2,214 women who had labored spontaneously (P =.001).^[41]

In a study by Ravasia et al of 575 patients who underwent labor induction, the uterine rupture rate was 1.4% compared with 0.45% for women who labored spontaneously (P =.004).^[40]

Blanchette et al found that the uterine rupture rate after previous cesarean delivery when labor was induced was 4.0% compared with 0.34% for women who labored spontaneously.^[42] This last finding suggests a 12-fold increased risk of uterine rupture for women who undergo labor induction after previous cesarean delivery.

Data on mechanical methods of labor induction for cervical ripening are limited but reassuring. In a small case series, Bujold et al found no statistically significant difference among the uterine rupture rates of 1.1% for spontaneous labor, 1.2% for induction by amniotomy with or without oxytocin, and 1.6% for induction by transcervical Foley catheter (P =0.81).^[43]

Conversely, Hoffman et al reported a 3.67-fold increased risk of uterine rupture (95% CI, 1.46-9.23) with Foley catheter use for preinduction cervical ripening. Importantly, however, many of these patients received concomitant oxytocin together with application of the transcervical Foley catheter.^[44]

Of particular note is that a recent randomized controlled trial by Pettker et al found that the addition of oxytocin to the use of a transcervical Foley catheter for labor induction does not shorten the time to delivery and has no effect on either the likelihood of delivery within 24 hours or the vaginal delivery rate.^[45] In light of these findings, induction of labor with a transcervical Foley catheter alone may be a reasonable option for women undergoing a TOLAC with an unfavorable cervix.

Use of prostaglandins for cervical ripening and induction of labor after previous cesarean delivery

Current ACOG guidelines discourage the use of prostaglandins to induce labor in most women with a previous cesarean delivery. This recommendation is based on considerable evidence for an increased risk of uterine rupture associated with prostaglandins. Lydon-Rochelle et al reported a 15.6-fold increased risk for uterine rupture (95% CI, 8.1-30) when prostaglandins were used in gravidas who underwent a TOL after previous cesarean delivery. In 366 women with scars from a previous cesarean delivery who underwent labor induction with prostaglandins, the uterine rupture rate was 2.45% compared with 0.77% without prostaglandin use.^[39]

Taylor et al identified 3 uterine ruptures among 58 patients with 1 previous cesarean delivery who received prostaglandin E₂ (PGE₂) alone for labor induction. The uterine rupture rate was 5.2% (3 of 58) compared with 1.1% (8 of 732) among patients not treated with prostaglandin.^[46] Ravasia et al found that 3 ruptures occurred among 172 patients who underwent labor induction with PGE₂ alone (1.7%), which was significantly higher than 0.45% (7 of 1,544) women who labored spontaneously.^[40]

In contrast, Flamm et al found a uterine rupture rate of 1.3% (6 of 453) in patients with a previous cesarean delivery who were treated with PGE₂ in combination with oxytocin. This result was not significantly different from the rate of 0.7% (33 of 4,569) in women who were not treated with PGE₂.^[47] In a small study, Delaney and Young also did not find a significant difference in uterine rupture rates between patients with scars from a previous cesarean delivery who underwent labor induction with PGE₂ and patients with previous cesarean scars who labored spontaneously (1.1 vs 0.3%; P =.15).^[48]

Landon et al reported no uterine ruptures among 227 patients who underwent induction with prostaglandins alone. Although the study was underpowered to detect small differences, the particular type of prostaglandin administered did not appear to significantly affect the uterine rupture rate (52 patients received misoprostol; 111, dinoprostone; 60, PGE₂ gel; and 4, combined prostaglandins).^[30]

Previous cesarean delivery with previous successful vaginal delivery

Several studies have shown a protective association of previous vaginal birth on uterine rupture risk in subsequent attempts at vaginal birth after previous cesarean delivery. Zelop et al compared 1,021 women who underwent a TOL after a single previous cesarean delivery with 1 previous vaginal delivery with 2,762 women who underwent a TOL with no previous vaginal delivery. The uterine rupture rate was 0.2% versus 1.1% (P =.01).^[49]

Among women with a single uterine scar, those with at least 1 previous vaginal delivery had one fifth the risk for uterine rupture compared with women without a previous vaginal delivery (OR, 0.2; 95% CI, 0.04-0.8). Caughey et al found that women with a previous vaginal delivery were about one fourth as likely as patients without a previous vaginal delivery to have a uterine rupture (OR, 0.26; 95% CI, 0.08-0.88).^[50] In a study of 205 patients who underwent a TOL after 1 previous cesarean delivery, Kayani and Alfirevic noted that all 4 of their cases of uterine ruptures occurred in women with no previous vaginal delivery.^[51]

A study of 11,778 women by members of the Maternal-Fetal Medicine Units (MFMU) Network found that in women with no prior vaginal delivery who underwent a TOL, there was an increased risk of uterine rupture with induction versus spontaneous labor (1.5% vs 0.8%, P =0.02). In contrast, no statistically significant difference was shown for women with a prior vaginal delivery who underwent spontaneous TOL compared with labor induction (0.6% vs 0.4%, P =0.42).^[52]

Previous cesarean delivery with subsequent successful VBACs

Multiple studies suggest a protective advantage with regard to the uterine rupture rate if a woman has had a prior successful VBAC attempt. Multiple potential explanations exist, but the 2 most obvious are that a successful prior VBAC attempt assures that (1) the maternal pelvis is tested and that the bony pelvis is adequate to permit passage of the fetus and (2) the integrity of the uterine scar has been tested previously under the stress/strain conditions during labor and delivery that were adequate to result in vaginal delivery without prior uterine rupture.

Mercer et al found that the rate of uterine rupture decreased after the first successful VBAC, but that there was no additional protective effect demonstrated thereafter: the uterine rupture rate was 0.87% with no prior VBACs, 0.45% for those with one successful prior VBAC, and 0.43% for those with 2 or more successful prior VBACs (P =.01).^[53] Pooled data from 5 studies indicate an increased uterine rupture rate of 1.4% (1 per 73) in failed VBAC attempts that required a repeat cesarean section in labor.^{[30, 42, 54, 55, 56]}

Interdelivery interval

In a case-control study by Esposito et al, an interpregnancy interval between cesarean delivery and a subsequent pregnancy of < 6 months was nearly 4 times as common among patients who had uterine rupture than in control subjects (17.4 vs 4.7%; OR, 3.92; 95% CI, 1.09-14.3). Among 23 patients who had uterine rupture after a previous cesarean delivery, the mean interpregnancy interval was 20.4 ± 15.4 months compared with 36.5 ± 30.4 months for control patients (P =.01).^[57] Stamilio et al recently confirmed a similar uterine rupture rate of 2.7% in women with an interdelivery interval of < 6 months compared with 0.9% for those having interdelivery intervals of ≥6 months (adjusted OR 2.66, 95% CI, 1.21-5.82).^[58]

Shipp et al similarly found that the risk of symptomatic uterine rupture was increased 3-fold in women with interdelivery intervals of< 18 months when they underwent a TOL after 1 previous cesarean delivery (OR, 3.0; 95% CI, 1.2-7.2).^[59] The authors controlled for maternal age, public assistance, length of labor, gestational age of ³41 weeks, and induction of augmentation of labor with oxytocin.

In additional support of this observation, Bujold et al reported on 1,527 women who underwent a TOL after a single previous low-transverse cesarean delivery, finding that 2.8% of patients who had an interdelivery interval of ≤ 24 months had a uterine rupture compared with 0.9% for those with an interdelivery interval of >24 months (P < .01).^[60] After adjusting for confounding variables, the odds ratio for a uterine rupture during a subsequent TOL was 2.65 for women who had an interdelivery interval of ≤ 24 months compared with women who had a longer interdelivery interval (95% CI, 1.08-5.46).

The authors speculated that a prolonged interpregnancy interval may allow time for the previous cesarean delivery scar to reach its maximal tensile strength before the scar undergoes the mechanical stress and strain with a subsequent intrauterine pregnancy. Interestingly, the authors also observed that the combination of a short interdelivery interval of ≤24 months and a single-layer hysterotomy closure was associated with a uterine rupture rate of 5.6%. This is comparable to the rate of uterine rupture for patients undergoing a TOL with a previous classic midline cesarean scar.

One-layer versus 2-layer hysterotomy closure

In a Canadian study of 1,980 women who underwent a TOL after a single previous low transverse cesarean delivery, Bujold et al found a 4- to 5-fold increased risk of uterine rupture for women who had a previous single-layer uterine closure compared with those having a 2-layer closure. Uterine rupture occurred in 3.1% (15 of 489 cases) of single-layer closure versus 0.5% (8 of 1,491 cases) of double-layer closure (P < .001). Using stepwise multivariate logistic regression, the authors concluded that the OR for uterine rupture in women who had undergone a single previous 1-layer cesarean hysterotomy closure was 3.95 (95% CI, 1.35-11.49) compared with those who had a 2-layer closure.^[61]

Durnwald and Mercer found that 182 patients with single-layer hysterotomy closure did not have an increased rate of uterine rupture, but the rate of uterine windows at subsequent delivery was increased to 3.5% versus 0.7% for those who had a multi-layer closure (P =.046).^[62]

Gyamfi et al reported an 8.6% (3 of 35) rate of uterine rupture in patients with a single-layer closure compared with 1.3% (12 of 913) in those with double-layer closure (P =0.015). Although the single-layer group had a shorter interdelivery interval, the uterine rupture rate remained significantly elevated even when the time interval was controlled for using logistic regression (OR 7.20, 95% CI, 1.81-28.62, P =0.005).^[63]

Multiple prior cesarean deliveries

For women with a history of 2 or more cesarean deliveries, 10 studies published from 1993-2010 showed that the risk of uterine rupture in a subsequent pregnancy ranged from 0.9-6.0% (1 per 17-108 pregnancies). This risk is increased 2- to 16-fold compared to women with only a single previous cesarean delivery. In a study of 17,322 women with scars from cesarean delivery, Miller et al found that, when women underwent a TOL, uterine rupture was 3 times more common with 2 or more scars (1.7%) than with 1 scar (0.6%) [OR, 3.06; 95% CI, 1.95-4.79; P < .001].^[64]

In the largest analysis to date, Macones et al reviewed data from 17 tertiary and community hospitals and found that, in 1,082 women with 2 uterine scars who underwent a TOL, the risk of uterine rupture was increased 2-fold compared with women with only 1 uterine scar (absolute rupture risk 1.8% vs 0.9%; adjusted OR, 2.3; 95% CI, 1.37-3.85).^[65]

In the only study to control for potential confounding variables, Caughey et al concluded that in women who had 2 previous cesarean deliveries who then attempted vaginal birth, the risk of uterine rupture was almost 5 times the risk of those with only 1 previous cesarean delivery (3.7% vs 0.8%; P =.001). The study controlled for several key covariates, including the use of prostaglandin E2 gel, oxytocin induction, oxytocin augmentation, length of labor, and epidural use. They also found that women with a previous vaginal delivery were about one fourth as likely to have a uterine rupture as women without a previous vaginal delivery (OR, 0.26; 95% CI, 0.08-0.88).^[50]

In contrast, Landon et al reported through the MFM Network that there was no significant difference in the uterine rupture rate for women with multiple prior cesarean deliveries versus 1 prior cesarean delivery (0.9% vs 0.7%; P= 0.37).^[66] However, in this study there was a much lower TOL rate of 9% for women with multiple prior cesarean deliveries compared with the 27% rate in the report of Macones et al^[65] and the 73% rate in Miller’s study^[64] . This indicates that there were much more stringent inclusion/exclusion criteria applied by Landon et al, and that this difference may account for the apparent discrepancy in outcomes. Caughey et al did not report the TOL rate in their 12-year data analysis.^[50]

A recent meta-analysis of 17 studies including 5,666 patients undergoing a TOL after 2 or more cesarean deliveries demonstrated a 1.36% uterine rupture rate.^[67] This is similar to the result of our pooled data analysis from 10 studies published from 1993-2010, which shows a 1.81% uterine rupture rate for patients with multiple previous cesarean delivery scars.

A 2004 ACOG guideline suggests that in women with 2 previous cesarean deliveries, only those with a previous vaginal delivery should be considered candidates for a TOL.^[68]

Fetal macrosomia

Elkousy et al found that, in 9,960 women who underwent a TOL after 1 previous cesarean delivery, the risk of uterine rupture was significantly greater for fetuses that weighed >4000 g (2.8%) than in those weighing < 4000 g (1.2%; RR 2.3, P < .001). For women with 1 previous cesarean delivery and no previous vaginal deliveries, the uterine rupture rate was 3.6% for women with a fetal weight of >4000 g compared to women with a fetal weight of < 4000 g (RR 2.3, P < .001).^[69]

Zelop et al reported that the rate of uterine rupture for women delivering neonates weighing >4000 g was 1.6% versus 1% for newborns ≤4000 g, but that the difference was not statistically significant (P =0.24).^[70] More recently, Jastrow et al showed that birth weight was directly correlated with the rate of uterine rupture, with uterine rupture rates of 0.9%, 1.8%, and 2.6% for birth weights of < 3500 g, 3500-3999 g, and ≥4000 g, respectively (P < .05).^[71]

Maternal age

Shipp et al showed that increasing maternal age has a detrimental effect on the rate of uterine rupture. In a multiple logistic regression analysis that was designed to control for confounding factors, the overall rate of uterine rupture in 3,015 women with 1 previous cesarean delivery was 1.1%. The rate of uterine rupture in women older than 30 years (1.4%) versus younger women (0.5%) differed significantly (OR, 3.2; 95% CI, 1.2-8.4).^[72]

Table 1. Absolute Rates of Uterine Rupture for Different Patient Subgroups (Open Table in a new window)

The signs and symptoms of uterine rupture largely depend on the timing, site, and extent of the uterine defect. Uterine rupture at the site of a previous uterine scar is typically less violent and less dramatic than a spontaneous or traumatic rupture because of their relatively reduced vascularity.

The classic signs and symptoms of uterine rupture are (1) fetal distress (as evidenced most often by abnormalities in fetal heart rate), (2) diminished baseline uterine pressure, (3) loss of uterine contractility, (4) abdominal pain, (5) recession of the presenting fetal part, (6) hemorrhage, and (7) shock. However, modern studies show that some of these signs and symptoms are rare and that many may not be reliably distinguished from their occurrences in other, more benign obstetric circumstances (see Table 2).

Table 2. Conditions Associated With Uterine Rupture (Open Table in a new window)

Prolonged, late, or recurrent variable decelerations or fetal bradycardias are often the first and only signs of uterine rupture. Bujold and Gauthier showed that abnormal patterns in fetal heart rate were the first manifestations of uterine rupture in 87% of patients.^[60] In a study by Leung et al, prolonged decelerations in fetal heart rate occurred in 79% of cases and were the most common finding associated with uterine rupture.^[73] Rodriguez et al found that fetal distress was the most common finding associated with uterine rupture, occurring in 78%.^[74] Overall, in 4 studies from 1983-2000, prolonged decelerations of fetal heart rate or bradycardias occurred in 114 (80%) of 143 cases of uterine rupture. In cases that involved the extrusion of the placenta and fetus into the abdominal cavity, prolonged decelerations in fetal heart rate invariably occurred.^{[73, 75, 76, 60]}

Sudden or atypical maternal abdominal pain occurs more rarely than fetal heart rate decelerations or bradycardia. In 9 studies from 1980-2002, abdominal pain occurred in 13-60% of cases of uterine rupture. In a review of 10,967 patients undergoing a TOL, only 22% of complete uterine ruptures presented with abdominal pain and 76% presented with signs of fetal distress diagnosed by continuous electronic fetal monitoring.^[77]

Moreover, in a study by Bujold and Gauthier, abdominal pain was the first sign of rupture in only 5% of patients and occurred in women who developed uterine rupture without epidural analgesia but not in women who received an epidural block.^[60] Thus, abdominal pain is an unreliable and uncommon sign of uterine rupture. Initial concerns that epidural anesthesia might mask the pain caused by uterine rupture have not been verified and there have been no reports of epidural anesthesia delaying the diagnosis of uterine rupture. A guideline from ACOG from 2004 suggests there is no absolute contraindication to epidural anesthesia for a TOL because epidurals rarely mask the signs and symptoms of uterine rupture.^[68]

Phelan et al found that abnormal patterns of uterine activity, such as tetany and hyperstimulation, are often not associated with uterine rupture. In their study, in which monitoring of uterine activity was limited to external tocodynamometry, tetany was defined as a contraction lasting longer than 90 seconds, and hyperstimulation was defined as more than 5 contractions in 10 minutes.^[78] Rodriguez et al found that the usefulness of intrauterine pressure catheters (IUPCs) for diagnosing uterine rupture was not supported. In 76 cases of uterine rupture, the classic description of decreased uterine tone and diminished uterine activity was not observed in any patients, 39 of whom had IUPCs in place. In addition, rates of fetal and maternal morbidity and mortality associated with uterine rupture did not differ with the use of an IUPC compared with external tocodynamometry.^[74]

In 8 reports published from 1980-2002 in which investigators examined the frequency of vaginal bleeding in cases of uterine rupture, vaginal bleeding occurred in 11-67% of cases. In 3 studies, maternal shock from hypovolemia was associated with uterine rupture in 29-46% of cases.^{[2, 5, 79]}

Diagnosis

Because of the short time available to diagnose uterine rupture before the onset of irreversible physiologic damage to the fetus, time-consuming diagnostic methods and sophisticated imaging modalities have only limited use. Therefore, uterine rupture is most appropriately diagnosed on the basis of standard signs and symptoms (see Table 2).

Despite this limitation, various diagnostic techniques have been used to attempt to assess the individual risk of uterine rupture in selected patients. Amniography, radiopelvimetry, and pelvic examination have all proven unsuitable for predicting the risk of uterine rupture in women who desire a TOL after a previous cesarean delivery. In addition, imaging modalities such as CT and MRI have not been clinically useful in diagnosing acute uterine rupture because of the time constraints involved in establishing the diagnosis. Given this limitation, MRI is thought to be superior to CT for evaluating the status of a uterine incision because of its increased soft tissue contrast. All studies of these methods are limited by their retrospective design and their lack of surgical confirmation of true uterine dehiscence.

Several reports have suggested that transabdominal, transvaginal, or sonohysterographic ultrasonography may be useful for detecting uterine-scar defects after cesarean delivery. Rozenberg et al prospectively examined 642 women and found that the risk of uterine rupture after previous cesarean delivery was directly related to the thickness of the lower uterine segment, as measured during transabdominal ultrasonography at 36-38 weeks of gestation. The risk of uterine rupture increased significantly when the uterine wall was thinner than 3.5 mm. Using a 3.5 mm cutoff, the authors had a sensitivity of 88%, specificity of 73.2%, positive predictive value of 11.8%, and a negative predictive value of 99.3% in predicting subsequent uterine rupture.^[80]

In a study of 722 women, Gotoh et al reported that a uterine wall thinner than 2 mm, as determined with ultrasonography performed within 1 week of delivery, significantly increased the risk of uterine rupture. Positive and negative predictive values were 73.9% and 100%, respectively.^[81]

Overview

The consequences of uterine rupture during pregnancy depend on the time that elapses from the rupture until the institution of definitive therapy. Definitive therapy for the fetus is delivery and must generally be accomplished with alacrity to avoid major fetal morbidity and mortality. Conversely, therapy for the mother can generally be supportive and resuscitative until surgical intervention can be undertaken to arrest the often life-threatening uterine hemorrhage.

Several studies have shown that delivery of the fetus within 10-37 minutes of uterine rupture is necessary to prevent serious fetal morbidity and mortality.^{[73, 76, 82, 42, 60]} If proper supportive measures (including fluid resuscitation and blood transfusion), are available to treat the mother, the time for definitive surgical intervention before the onset of major maternal morbidity and mortality may often be substantially longer than that for the fetus.

Therefore, the consequences of uterine rupture may be divided into 2 major categories, depending on whether they apply to the fetus or to the mother (see Table 3).

Table 3. Fetal and Maternal Consequences of Uterine Rupture (Open Table in a new window)

Fetal and Neonatal Consequences of Uterine Rupture

Fetal hypoxia or anoxia

Leung et al found that 5 of 99 neonates (5%) born to women who had uterine ruptures developed neonatal asphyxia (defined as umbilical-artery pH < 7 with seizures and multiorgan dysfunction). No neonate had clinically significant perinatal morbidity when delivery was accomplished within 17 minutes of an isolated and prolonged deceleration of fetal heart rate. If severe late decelerations preceded prolonged deceleration, perinatal asphyxia was observed as soon as 10 minutes from the onset of the prolonged deceleration to delivery.^[73]

In a study by Menihan, 6 of 11 fetuses born after uterine rupture had bradycardias occur between 18-37 minutes prior to delivery. Although the rate of fetal acidosis was high (91%), no permanent neurologic injuries or neonatal deaths occurred.^[76]

In 23 cases of uterine rupture, Bujold and Gauthier found that, even with rapid (< 18-min) intervention between prolonged deceleration in fetal heart rate and delivery, 2 neonates developed hypoxic-ischemic encephalopathies with impaired motor development. They concluded that, though rapid intervention did not always prevent severe metabolic acidosis and serious neonatal disease, it probably did limit the occurrence of neonatal death.^[60]

Fetal acidosis

In 99 cases of uterine rupture, Leung et al found that 43 newborns (43%) had an umbilical-artery pH level of less than 7, and 25 of these newborns had a pH level of less than 6.8. In association with these pH levels, 39 newborns (39%) had 5-minute Apgar scores of less than 7, 12 of whom had 5-minute Apgar scores of less than 3.^[73]

Menihan found that 10 of 11 fetuses (91%) who were born after uterine rupture had an umbilical-artery cord pH level of less than 7.0, and 5 (45%) had 5-minute Apgar scores of less than 7. The most important factor for the development of fetal acidosis was complete extrusion of the fetus and placenta into the maternal abdomen.^[76]

Admission to a neonatal intensive care unit

Menihan found that 8 of 11 newborns (73%) delivered after uterine rupture required admission to the neonatal intensive care unit (NICU).^[76]

Kieser and Baskett found an NICU admission rate for newborns 45% (8 of 18) after uterine rupture.^[83] Landon et al found a similar NICU admission rate of 32% (46 of 144 newborns) after uterine rupture.^[30]

Fetal or neonatal death

In studies reported before 1978, the fetal mortality rate associated with uterine rupture was high. In a review of 33 studies by Schrinsky and Benson, 960 cases of uterine rupture resulted in 620 infant deaths, yielding a perinatal mortality rate of 65%.^[3] Blanchette et al reported that 2 neonates (17%) died among 12 women who had uterine rupture and that 1 of these neonates died after a decision-to-delivery time of only 26 minutes after the acute onset of fetal bradycardia, lower abdominal pain, and vaginal bleeding, which signaled the acute uterine rupture.^[42]

Leung et al reported that 6 perinatal deaths (6%) occurred among 99 patients who had uterine rupture.^[73] In a study by Lydon-Rochelle et al, the perinatal death rate among fetuses in 91 cases of uterine rupture was 5.5% compared with 0.5% in control subjects.^[39] Landon et al reported a perinatal death rate from uterine rupture of 2% (2 of 124) among 19 academic centers in the United States. These studies indicate that the incidence of perinatal death associated with uterine rupture is decreasing in the modern era.^[30]

Maternal Consequences of Uterine Rupture

Severe maternal blood loss or anemia

Cowan et al found that, among 5 patients who developed uterine rupture, mean blood loss was 1,500 mL and great enough to be symptomatic in 3 patients (60%).^[84] In a study by Shipp et al, 25% (7 of 28 women) who had uterine rupture during a TOL after a previous cesarean delivery received a blood transfusion.^[34]

Kieser and Baskett found that 44% (8 of 18 patients) who had a complete uterine rupture required blood transfusion.^[83] Leung et al found that 29% of patients (29 of 99) who had uterine rupture required a blood transfusion.^[73]

Hypovolemic shock

In a study of 93 uterine ruptures by Golan et al, 29% of women who experienced a uterine rupture developed signs and symptoms of hypovolemic shock.^[2] Rahman et al reported that, of 96 women who had uterine rupture, 33 (34%) developed hypovolemic shock.^[5]

These modern rates of maternal shock after uterine rupture appear to be reduced compared with the early rates reported in a 53-year review of the literature by Eden et al; their observed incidence was 46% (11 of 24 cases).^[79]

Maternal bladder injury

Lydon-Rochelle et al reported significant maternal bladder injuries in 8% of women (7 of 91) whose uteri ruptured compared with 240 of 20,004 control patients (1.2%) in whom rupture did not occur (P =.001).^[39] Shipp et al found that bladder injuries occurred in 18% of women (5 of 28) who had a uterine rupture after previous low transverse cesarean delivery.^[34]

In a study by Kieser and Baskett, 17% of women (3 of 18) who developed uterine rupture had a cystotomy.^[83] Leung et al found that 12% (12 of 99) who experienced a uterine rupture had incidental cystotomies at the time of surgery, and 7 more (7%) had either a ruptured bladder or an accidental cystotomy; the combined total urologic injury rate was 19%.^[73]

Need for hysterectomy

In a study from South Africa, 78% of women (261 of 335) who had uterine rupture were treated with hysterectomy.^[4] Flamm et al found that 3 of 39 patients (8%) who developed uterine rupture required hysterectomy.^[85] Kieser and Baskett found that 1 of 18 patients (6%) who developed complete uterine rupture required hysterectomy.^[83] Blanchette et al reported that hysterectomy was necessary in 17% of women (2 of 12) who developed uterine rupture.^[42] Hibbard et al found that 6 hysterectomies (60%) were necessary in 10 women who had a uterine rupture.^[27]

Leung et al reported that 19% of patients (19 of 99) who experienced a uterine rupture required hysterectomy. Thirteen hysterectomies (68%) were performed because the uterus was not deemed repairable, 4 (21%) for irremediable uterine atony, and 1 (5%) because of placenta accreta.^[73]

Maternal death

Maternal death as a consequence of uterine rupture occurs at a rate of 0-1% in modern developed nations, but the mortality rates in developing countries are 5-10%.^{[5, 4]}

The availability of modern medical facilities in developed nations is likely to account for this difference in maternal outcomes. In a South African study from 1976, 22 of 260 women who had pregnancy-related rupture of an unscarred uterus died (mortality rate 8.5%). These deaths could be further subdivided into mortality for women with longitudinal uterine tears (15 of 183 patients [8.2%]), transverse tears (2 of 49 patients [4%]), posterior-wall tears (2 of 16 patients [13%]), and multiple uterine tears (3 of 12 patients [25%]).

Golan et al reported no deaths among 32 mothers who experienced rupture of a scarred uterus compared with 9 deaths among 61 women with an intact uterus (15%).^[2] In a study from Los Angeles in which Leung et al reported on 99 patients with uterine ruptures, 1 woman (1%) died.^[73]

Mokgokong and Marivate noted that the maternal mortality rate associated with uterine rupture largely depends on whether the diagnosis is established before or after delivery; these rates were 4.5% and 10.4%, respectively.^[4]

Treatment

The most critical aspects of treatment in the case of uterine rupture are establishing a timely diagnosis and minimizing the time from the onset of signs and symptoms until the start of definitive surgical therapy. Once a diagnosis of uterine rupture is established, the immediate stabilization of the mother and the delivery of the fetus are imperative.

As a rule, the time available for successful intervention after frank uterine rupture and before the onset of major fetal morbidity is only 10-37 minutes.^{[73, 76, 82, 42, 86]} Therefore, once the diagnosis of uterine rupture is considered, all available resources must quickly and effectively be mobilized to successfully institute a timely surgical treatment that results in favorable outcomes for both the newborn and the mother.

After the fetus is successfully delivered, the type of surgical treatment for the mother should depend on the following factors:

• Type of uterine rupture
• Extent of uterine rupture
• Degree of hemorrhage
• General condition of the mother
• Mother's desire for future childbearing

Uterine bleeding is typically most profuse when the uterine tear is longitudinal rather than transverse. Conservative surgical management involving uterine repair should be reserved for women who have the following findings:

• Desire for future childbearing
• Low transverse uterine rupture
• No extension of the tear to the broad ligament, cervix, or paracolpos
• Easily controllable uterine hemorrhage
• Good general condition
• No clinical or laboratory evidence of an evolving coagulopathy

Hysterectomy should be considered the treatment of choice when intractable uterine bleeding occurs or when the uterine rupture sites are multiple, longitudinal, or low lying.

Because of the short time available for successful intervention, the following 2 premises should always be kept firmly in mind: (1) Maintain a suitably high level of suspicion regarding a potential diagnosis of uterine rupture, especially in high-risk patients, and (2) when in doubt, act quickly and definitively.

Prevention

The absolute risk of uterine rupture in pregnancy is low, but it is highly variable depending on the patient subgroup (see Table 1). Women with normal, intact uteri are at the lowest risk for uterine rupture (1 in 8,434 pregnancies [0.012%]).

The most direct prevention strategy for minimizing the risk of pregnancy-related uterine rupture is to minimize the number of patients who are at highest risk. The salient variable that must be defined in this regard is the threshold for what is considered a tolerable risk. Although this choice is ultimately arbitrary, it should reflect the prevailing risk tolerance of patients, physicians, and of society as a whole. If this threshold is chosen as 1 in 200 women (0.5%) (see Table 1), the categories of patients that exceed this critical value are those with the following:

• Multiple previous cesarean deliveries
• Previous classic midline cesarean delivery
• Previous low vertical cesarean delivery
• Previous low transverse cesarean delivery with a single-layer hysterotomy closure
• Previous cesarean delivery with an interdelivery interval of less than 2 years
• Previous low transverse cesarean delivery with a congenitally abnormal uterus
• Previous cesarean delivery without a previous history of a successful vaginal birth
• Previous cesarean delivery with either labor induction or augmentation
• Previous cesarean delivery in a woman carrying a macrosomic fetus weighing >4000 g
• Previous uterine myomectomy accomplished by means of laparoscopy or laparotomy

If a gravida falls into any of these categories, her risk for uterine rupture is increased to more than 1 in 200, and a clinical management plan should be specifically designed with this increased risk in mind.

Uterine rupture is a rare but often catastrophic obstetric complication with an overall incidence of approximately 1 in 1,536 pregnancies (0.07%). In modern industrialized countries, the uterine rupture rate during pregnancy for a woman with a normal, unscarred uterus is 1 in 8,434 pregnancies (0.012%).

The vast majority of uterine ruptures occur in women who have uterine scars, most of which are the result of previous cesarean deliveries. A single cesarean scar increases the overall rupture rate to 0.5%, with the rate for women with 2 or more cesarean scars increasing to 2%. Other subgroups of women who are at increased risk for uterine rupture are those who have a previous single-layer hysterotomy closure, a short interpregnancy interval after a previous cesarean delivery, a congenital uterine anomaly, a macrosomic fetus, prostaglandin exposure, and a failed previous trial of a vaginal delivery.

Surgical intervention after uterine rupture in less than 10-37 minutes is essential to minimize the risk of permanent perinatal injury to the fetus. However, delivery within this time cannot always prevent severe hypoxia and metabolic acidosis in the fetus or serious neonatal consequences.

The most consistent early indicator of uterine rupture is the onset of a prolonged, persistent, and profound fetal bradycardia. Other signs and symptoms of uterine rupture, such as abdominal pain, abnormal progress in labor, and vaginal bleeding, are less consistent and less valuable than bradycardia in establishing the appropriate diagnosis.

The general guideline that labor-and-delivery suites should be able to start cesarean delivery within 20-30 minutes of a diagnosis of fetal distress is of minimal utility with respect to uterine rupture. In the case of fetal or placental extrusion through the uterine wall, irreversible fetal damage can be expected before that time; therefore, such a recommendation is of limited value in preventing major fetal and neonatal complications. However, action within this time may aid in preventing maternal exsanguination and maternal death, as long as proper supportive and resuscitation methods are available before definitive surgical intervention can be successfully initiated.