Uterine rupture is an important cause of both maternal and fetal death worldwide although advances in obstetric management have reduced adverse outcomes to a minimum in developed countries. The incidence varies widely, but most authors suggest figures of 1 : 1200–1500 deliveries. The maternal mortality can be high, with haemorrhage and its complications being the most important contributory factors. Nevertheless, only six maternal deaths occurred from uterine rupture in the United Kingdom between 1985 and 1987 in relation to 2.3 million deliveries. In no mother was death due to rupture of a caesarean section scar, although one patient had had a previous caesarean section.

Often a distinction is made between rupture of a scarred uterus and ‘spontaneous’ rupture where there is no previous scar. However, the term ‘spontaneous’ is misleading in that such events are usually associated with obstetric trauma during labour (Table 1) 426.

Previous caesarean section is the most common factor, and while rupture is reported in only 0.3 to 0.7 per cent of labours in the presence of a previous lower-segment scar, it occurs in up to 9 per cent of labours following a classical uterine scar.

The number of previous lower-segment scars appears not to increase the risk of subsequent rupture. A 3 per cent incidence of occult or subclinical ruptures or dehiscences has been reported when an elective repeat caesarean section was performed. In a similar group, 0.5 per cent ruptures and 1.9 per cent dehiscences were reported. Provided the uterine scar does not extend into the upper segment, it is insignificant with regard to rupture whether the lower-segment incision is transverse or vertical.

A lower-segment scar typically ruptures only during labour, particularly in the late first or second stage when uterine activity reaches its peak. Upper segment scars, in contrast, may rupture prior to labour, when the presentation mimics closely that of abruptio placentae.

Other uterine scars (see Table 1 426) have been associated with rupture in pregnancy or in labour. Patients with such scars require skilled obstetric management.

Less commonly, ruptures may be associated with obstetric trauma from instrumental deliveries or intrauterine manipulations; with abdominal trauma; and with uterine anomalies, such as a pregnancy within a uterine cornu. The latter may mimic ectopic pregnancy in presentation.

A combination of factors could occur, with rupture occurring in a scarred uterus, which may or may not involve the previous scar.

The presentation may vary from dramatic, with collapse of the mother, to vague symptoms and signs with the diagnosis being delayed by many hours. Furthermore, routine exploration of previous scar sites, either digitally via the cervix after vaginal delivery or directly at caesarean section, may reveal hitherto unsuspected weaknesses or ruptures in the scar. Such occult defects may be divided into ‘rupture’ requiring surgical repair or intervention, or ‘dehiscence’ not requiring intervention. While the classical features are listed in Table 2 427, some or all may be absent.

Signs such as scar tenderness are unreliable, and suprapubic tenderness is common in many labours. Fears that epidural analgesia would mask symptoms of scar rupture are probably unfounded as features other than tenderness often predominate.

Antenatal presentation of uterine rupture can often be confused with that of abruptio placentae. The presentation is not necessarily dramatic initially, and the diagnosis should be considered when mothers with a scarred uterus present with vaginal bleeding or abdominal pain. Maternal shock and fetal compromise or death are common, with fetal survival being rare.

To detect uterine rupture as early as possible, meticulous observation is essential, particularly in mothers with previous uterine surgery. Regular assessment of the strength of uterine contractions and observations of maternal pulse and blood pressure, fetal heart rate pattern, and the progress of labour are all facilitated by the routine use of a partogram. Vaginal bleeding should be noted promptly. Continuous electronic fetal monitoring is advisable where facilities exist to detect early signs of fetal compromise. Intrauterine pressure monitoring is of value in confirming delayed progress in the late first stage despite good uterine activity. Unfortunately, no consistent pattern of uterine pressure presages uterine rupture.

In many cases, the diagnosis is delayed until several hours after delivery. These cases are often associated with a more serious outcome in association with severe haemorrhage. Continued close observation in the postdelivery period is essential, therefore, for mothers at risk of rupture.

In some areas some symphysiotomy has reduced the risk of uterine rupture by avoiding a caesarean scar. In those mothers with a previous caesarean section, it is probably wise not to allow a subsequent delivery should an upper-segment incision have been made, a more common event nowadays with caesarean sections being performed for very preterm deliveries. In multipara, oxytocin should be used only after careful clinical consideration. Early recourse to caesarean section rather than attempting complicated operative vaginal deliveries has become evermore established in obstetric practice in developed countries.

Phelan et al. allowed a trial of labour in patients with more than one previous caesarean section and achieved successful vaginal delivery in 82 per cent of mothers with one previous caesarean section (1348/1637 cases), 72 per cent after two caesarean sections (107/149 cases), and 90 per cent after three (9/10 cases). Those least likely to succeed in a vaginal delivery were those in whom the previous caesarean section had been for failure to progress in labour due to cephalopelvic disproportion.

The factors influencing management are indicated in Table 4 429. It will be influenced by whether the rupture occurs antenatally, during labour, or after delivery. The severity of injury depends on whether the rupture involves the placental site, cervix, bladder, or ureters. The risk of bladder involvement is strongly related to the number of previous caesarean sections as the tissues are so often densely adherent. Ruptures not involving a previous uterine scar are usually more dramatic, requiring urgent measures to avoid serious sequelae from hypotension and haemorrhage.

Resuscitation and stabilization of the patient's general condition is vital (Table 5) 430, with blood transfusion being invariably required. Undue delay before surgery can worsen the condition of an already compromised patient, particularly if disseminated intravascular coagulation occurs.

If the rupture follows prolonged obstructed labour, the mother may be dehydrated, ketotic and acidotic. Marked biochemical disturbances must be tackled in these patients. The health of the mother must always take priority over the well-being of the fetus, although, ideally, one aims to deliver a live fetus from a well mother.

Skilled general anaesthesia is mandatory and will permit greater control of the patient's overall haemodynamic condition. Experienced help must be sought as the operative procedures during laparotomy can be technically very difficult.

Control of haemorrhage is a priority and manual compression of the aorta or ligation of one or both internal iliac arteries can be useful. Because of the enormous collateral circulation, both iliac arteries may need ligation. Care should be taken to identify the ureters properly. In the face of brisk haemorrhage, closure or clamping of the edges of the defect temporarily, prior to any definitive procedure, will reduce the blood loss.

Closure of the defect with sterilization may be the procedure of choice if the patient is severely shocked. The patient's wishes must be considered when choosing an appropriate procedure, as several series have reported pregnancies following conservative surgery (repair of the defect only). The patient must be counselled subsequently on the risk of repeat rupture, and she may then elect for sterilization.

Hysterectomy is often the procedure of choice in view of the risk involved with subsequent pregnancies. At hysterectomy, the tissues must be handled gently as they are very vascular and oedematous and the pelvic ligaments are lax. The greatly enlarged vessels should be carefully clamped, and it is essential to double-tie all pedicles. The risk of tearing the bladder wall during dissection should not be underestimated and can be reduced by sharp dissection, keeping to the midline. This will also reduce the risk of trauma to the venous plexuses which lie laterally behind the bladder base.

Subtotal hysterectomy is appropriate when safe identification of the ureters is impossible. As identification of the lower edge of the dilated cervix can be difficult, part of the cervix may be left behind. The subtotal operation may be the procedure of choice if the surgeon is inexperienced.

Both the pelvis and the rectus sheath should be drained, especially in the presence of infection or where extensive dissection of adherent tissues was necessary. Wide-bore suction drains are helpful with the pelvis. Large subrectus haematomas develop easily if disseminated intravascular coagulation supervenes. An indwelling urinary catheter is mandatory, both to assist in monitoring fluid balance and in detecting haematuria, which may signify unrecognized bladder injury. Prior to the end of the abdominal procedure, examining the patient in the lithotomy position can assist in diagnosing whether the entire cervix has been removed, as well as excluding associated vaginal injuries or lacerations. Antibiotic prophylaxis with broad-spectum agents is advised. The risk of sepsis is particularly high when membranes have been ruptured for a long time.

This should include observation for signs of infection, careful monitoring of fluid balance, renal function, and coagulation status, and adequate blood replacement. Central venous pressure monitoring can be invaluable.

The hysterectomy specimen will permit identification of abnormal trophoblast invasion at the site of the rupture and the full extent of the trauma. It also serves as a check that all the cervix has been removed.

Mortality and morbidity results largely from haemorrhage, impaired renal perfusion due to hypovolaemia, and infection. In one large series over half of the ruptures occurred in the presence of previous uterine scarring, the majority of which had mild clinical features with little bleeding from the avascular scar site, and were repaired conservatively. When more extensive lacerations have developed, or in cases where rupture was traumatic or spontaneous, 85 per cent of such patients required immediate hysterectomy. The overall maternal mortality was 3 per cent, with all deaths occurring in patients who ruptured a previously intact uterus. Fetal mortality was 27.8 per cent, with death being three times more common in rupture of an unscarred uterus. Furthermore, infant morbidity was relatively high amongst those surviving a traumatic uterine rupture; central nervous system or peripheral nerve injuries or respiratory distress symptoms were the most common sequelae.

In one study a good outcome was found, with only a single repeat rupture occurring in 22 subsequent pregnancies following conservative surgery. Some patients even achieved vaginal deliveries. Only two out of 28 patients had involuntary infertility after the procedure. As fear of subsequent pregnancies can be an important factor, conservative surgery should only be performed when the couple actively desire further children. Subsequent pregnancy after conservative surgery should usually be managed by elective caesarean section in the later third trimester, with elective sterilization when the couple's family is complete.

Uterine rupture remains a serious and life-threatening event. The increasing use of caesarean section for delivery in the Western world will lead to an increased number of mothers at risk of scar rupture in subsequent pregnancies. Maternal and fetal outcome is worse following rupture of an unscarred uterus, often in association with the use of oxytocics. If the morbidity and mortality are to be kept to a minimum, the management of uterine rupture should involve careful collaboration between obstetricians, anaesthetists, and the intensive care unit team.

Aguero D, Kizer S. Obstetric prognosis of the repair of uterine rupture. Surg Gynecol Obstet 1968; 124: 528–30.

Arulkumaran S, Gibb DMF, Ingemarsson I, Kitchener HC, Ratnam SS. Uterine activity during spontaneous labour after previous lower-segment caesarean section. Br J Obstet Gynaecol 1989; 96: 933–8.

Arulkumaran S, Ingemarsson I, Ratnam SS. Oxytocin augmentation in dysfunctional labour after previous caesarean section. Br J Obstet Gynaecol 1989; 96: 939–41.

Beckley S, Gee H, Newton JR. Scar rupture in labour after previous lower uterine segment caesarean section; the role of uterine activity measurement. Br J Obstet Gynaecol 1991; 98; 265–9.

Department of Health. Report of confidential enquiries into maternal death in the United Kingdom 1985–87. London: HMSO, 1991; 68.

Dewhurst CJ. The ruptured caesarean section scar. J Obstet Gynaecol Br Commonwlth 1957; 64: 113–118.

Garnet JD. Uterine rupture during pregnancy: an analysis of 133 patients. Obstet Gynecol 1964; 23: 898–905.

Merrill BS, Gibbs CE. Planned vaginal delivery following caesarean section. Obstet Gynecol 1978; 52: 50–2.

O'Connor RA, Gaughan B. Pregnancy following simple repair of the ruptured gravid uterus. Br J Obstet Gynaecol 1989; 96: 942–4.

Phelan JP, Clark SL, Diaz F, Paul RH. Vaginal birth after caesarean. Am J Obstet Gynecol 1987; 157: 1510–15.

Reyes-Ceja L, Cabrera R, Insfran E, Herrera-Cassa F. Pregnancy following previous uterine rupture. Obstet Gynecol 1969; 34: 387–9.

Sheth SS. Suturing the tear as treatment in uterine rupture. Am J Obstet Gynecol 1969; 105: 440–3.

Shrinsky DC, Benson RC. Rupture of the pregnant uterus: a review. Obstet Gynaecol Surv 1978; 33: 217–32.