Why women bleed and how they are saved: a cross-sectional study of caesarean section near-miss morbidity

Background

Maternal deaths from ‘bleeding during and after caesarean section’ (BDACS) have increased in South Africa, and have now become the largest sub-cause of deaths from obstetric haemorrhage. The aim of this study was to describe risk factors and causes of near-miss related to BDACS and interventions used to arrest haemorrhage and treat its effects.

Methods

Cross-sectional prospective study in 13 urban public hospitals in South Africa, from July to December 2014.

Results

There were 93 cases of near-miss related and 7 maternal deaths related to BDACS. The near-miss rate was 2.1/1000 live births, and the case fatality rate was 3.5/10 000 caesarean sections. Associated near-miss risk factors were previous caesarean section in 60% of multiparas, pre-operative anaemia (55%), abruptio placentae (20%) and placenta praevia and/or accreta (20%). Atonic uterus (43%) was the most frequent anatomical cause of bleeding for near-miss, followed by surgical trauma (29%). The median duration of the operations resulting in near-miss was 90 min, with 81% noted as difficult by the surgeon. Interventions in cases of near-miss included second-look laparotomy (46%), hysterectomy (41%), B-Lynch brace suture (9%), intensive care unit admission (32%) and red cell transfusion ≥3 units (21%).

Conclusion

Cases from maternal near-miss from BDACS were frequently associated with pre-operative risk factors. Extensive life-saving interventions were required during and after the operations. An important factor in initiating the sequence of interventions is the realisation by the surgeon that the caesarean section is difficult, so that the progression from uneventful operation to near-miss to death can be arrested.

Babies who die from labour-related intrapartum hypoxia: a confidential enquiry in South African public hospitals

Seventeen hospitals, from a range of health-care environments, participated in confidential enquiries of perinatal deaths resulting from labour-related intrapartum hypoxia. There were 102 deaths, including 22 stillbirths and 80 neonatal deaths. The mean birthweight was 3021 g. The active phase of the first stage of labour was prolonged beyond 12 h in six cases, and oxytocin was used for induction or augmentation in 10 women. Fetal heart decelerations were detected in 39 (49%) of the babies that went on to die in the neonatal period, and meconium passage was evident in 50 (63%). There were six breech presentations, and seven cases of cord prolapse. The majority of these deaths occurred in low-risk women with apparently uncomplicated labour. There appears to be a failure to detect or respond to evidence of fetal distress. Intrapartum care for all women in labour requires close attention to detail in monitoring fetal health.

Natural acquired humoral immunity against serotype-specific group B Streptococcus rectovaginal colonization acquisition in pregnant women

Group B Streptococcus (GBS) rectovaginal colonization in pregnant women is associated with invasive GBS disease in newborns, preterm delivery and stillbirths. We studied the association of GBS serotype-specific capsular polysaccharide (CPS) antibody on new acquisition and clearance of rectovaginal GBS colonization in pregnant women from 20 weeks until 37 to 40 weeks' gestation. Serum serotype-specific CPS IgG antibody concentration was measured by multiplex enzyme-linked immunosorbent assay and opsonophagocytic activity (OPA) titres. Rectovaginal swabs were evaluated for GBS colonization, using standard culture methods and serotyping by latex agglutination, at five to six weekly intervals. Higher serotype III CPS antibody concentration was associated with lower risk of rectovaginal acquisition of serotype III during pregnancy (p 0.009). Furthermore, serotype-specific OPA titres to Ia and III were higher in women who remained free of GBS colonization throughout the study compared to those who acquired the homotypic serotype (p <0.001 for both serotypes). Serum CPS IgG values of ≥1μg/mL for serotype V and ≥3μg/mL for serotypes Ia and III were significantly associated with protection against rectovaginal acquisition of the homotypic serotype. A GBS vaccine that induces sufficient capsular antibody in pregnant women, including high OPA titres, could protect against rectovaginal colonization during the latter half of pregnancy.

Declining maternal mortality in the face of persistently high HIV prevalence in a middle-income country

OBJECTIVE

To estimate maternal mortality ratio (MMR) and determine maternal death causes and trends in Greater Soweto, Johannesburg, South Africa.

DESIGN

Cross-sectional study.

SETTING

Chris Hani Baragwanath Maternity Hospital (CHBMH) in Greater Soweto.

POPULATION

Maternal deaths at CHBMH.

METHODS

Record review of maternal deaths from 1997 to 2012, using hospital death records, with denominator data from the district health information system and the hospital.

MAIN OUTCOME MEASURES

Maternal mortality ratio per 100,000 live births, and causes of death classified as in the South African confidential enquiries.

RESULTS

There were 479 deaths, with a peak MMR of 139 in 2004 and a decline to 86 in 2012. Of 332 women tested, 245 (74%) were HIV-infected. Nonpregnancy-related infection (40%) was the most frequent cause of death, followed by hypertension (16%) and obstetric haemorrhage (13%). HIV infection rates in these groups were 92%, 30% and 61%, respectively. Previous caesarean section was associated with obstetric haemorrhage death (odds ratio [OR] 3.2, 95% confidence interval [95% CI] 1.7-6.0), maternal age ≥35 years with hypertension death (OR 2.2, 95% CI 1.2-3.7) and antenatal anaemia with nonpregnancy-related infection death (OR 4.0, 95% CI 2.3-6.9), compared with other causes of death.

CONCLUSION

There is evidence of a decline in MMR since HIV treatment for pregnant women was introduced in 2004. Previous caesarean section, advanced maternal age, and anaemia were associated with death from obstetric haemorrhage, hypertensive disorders of pregnancy and nonpregnancy-related infections, respectively. MMR may be further reduced with accelerated initiation of HIV treatment during pregnancy.

The value of routine mid-trimester ultrasound in low-risk pregnancies at primary care level

This study investigated the effect of routine second-trimester ultrasound scanning on obstetric management and pregnancy outcomes. This was an open cluster, randomised, controlled trial. Clusters of women with low-risk pregnancies presenting in the second trimester were randomised to receive an ultrasound scan followed by usual antenatal care, or to an unscanned control group undergoing conventional antenatal care only. Out of the 962 women randomised, follow-up was successful for 804 (83.6%), with 416 allocated to the ultrasound scan group and 388 controls. There were no significant differences between the ultrasound scan group and the control group in terms of prenatal hospitalisation, mode of delivery, miscarriage, perinatal mortality rate and low birthweight rate. Ultrasound dating was associated with a lower rate of induction of labour for post-term pregnancy (1.4% vs. 3.6%; P=0.049). However, ultrasound scanning in low-risk pregnancies was not associated with improvements in pregnancy outcome. Opsomming Hierdie studie het die effek van roetine mid-trimester ultraklankskandering op swangerskapsorg en –uitkomste ondersoek. Dit was ’n oop tros, lukrake, beheerde proef. Groepe vroue met laerisikoswanger- skap in die midtrimester is lukraak toegewys vir ’n ultraklank-skandering, gevolg deur voorgeskrewe voorgeboor-tesorg, of vir ’n kontrolegroep wat voorgeboortesorg volgens nasionaal voorgeskrewe protokol sonder skandering ontvang het. Van die 962 vroue wat aan die steekproef deelgeneem het kon data vir 804 (83.6%) suksesvol opgevolg word, met 416 in die ultraklankgroep en 388 in die kontrolegroep. Geen beduidende verskille is tussen die twee groepe gevind ten opsigte van voorgeboorte-hospitalisasie, geboortemetode, miskraamstatistiek, perinatale komplikasies of laegeboortegewig nie. Ultraklankdatering van swangerskappe is met minder kraaminduksie (1.4% teen 3.6%; P=0.049) vir natrimesterswangerskap geassosieer. Roetine ultraklankskandering in laerisikoswangerskap het egter geen verbetering in swangerskapsuitkomste te weeg gebring nie.

Accuracy of cervical assessment in the active phase of labour

OBJECTIVE

To determine accuracy of clinicians in estimating cervical dilatation during the active phase of labour and how this is affected by clinician experience and obstetric factors.

DESIGN

Prospective, cross-sectional, comparative study.

SETTING

Chris Hani Baragwanath Hospital labour ward, Johannesburg, South Africa.

POPULATION

Women at term in the active phase of labour, with vertex presentations and live fetuses.

METHODS

The researcher performed cervical assessment immediately after the clinician on duty. The researcher and clinician were unaware of each other's findings. The researcher, used as the standard, was an experienced obstetric consultant, and the clinicians were hospital consultants and registrars at various levels of training. Accuracy was defined as agreement of the clinician's cervical dilatation estimate with that of the researcher. Multivariate logistic regression analysis was carried out to determine independent predictors of inaccuracy.

MAIN OUTCOME MEASURE

Agreement in estimation of cervical dilatation between the researcher and the clinicians.

RESULTS

Examinations were performed on 508 women. The researcher and clinicians agreed on the dilatation in 250 instances (49.2%) and differed by 2 cm or more in 56 (11.0%) (kappa = 0.40, 95% CI 0.34-0.45). Accuracy was greater at low (3-4 cm) and high (8-10 cm) dilatations. Reduced accuracy was associated with decreasing clinician experience and with lower stations of fetal head.

CONCLUSION

This is the first study to investigate accuracy of cervical assessment in parturient women. Results were similar to those found in studies that used models, with about 90% of estimations accurate to within 1 cm.

Complications associated with cesarean section in the second stage of labor

OBJECTIVE

To determine maternal and neonatal complications associated with cesarean section done in the second stage of labor.

METHOD

Cohort study comparing cesarean sections done in the second stage of labor (cases) with those done for poor progress in the first stage (controls). Only singleton cephalic live pregnancies at 36 weeks or more, without previous cesarean section, were included.

RESULT

There were 39 cases and 39 controls. Cesarean section in the second stage of labor took significantly longer (median 45 vs. 30 min; P<0.001), and was associated with more frequent postoperative pyrexia (10 vs. 2; P=0.012). There were more neonatal admissions in the case group (17 vs. 3; P<0.001). Hypoxic ischemic encephalopathy was more frequent in infants following second-stage cesarean section (8 vs. 1; P=0.013), as was subaponeurotic hemorrhage (6 vs. 0; P=0.012).

CONCLUSION

Cesarean section in the second stage of labor is associated with significant intraoperative and neonatal morbidity.

Pregnancy outcomes in urban black South African women aged 35 years and older

We measured maternal and fetal outcomes for black South African women of age 35 years and above, compared with younger controls. The study was carried out at the Chris Hani Baragwanath Maternity Hospital and the Soweto Community Health Centres, that provide maternity services for urban black pregnant women. This was a retrospective pregnancy cohort, comparing pregnant women aged 35 and above with a control group aged 20-29 years. Early pregnancy losses (<20 weeks) were excluded. There were 1047 women in each arm of the study. Frequencies of hypertension (22.8%), gestational diabetes (0.9%), and prelabour rupture of the membranes (6.9%) were significantly higher for older women. Perinatal mortality rates (42/1000 vs. 33/1000) did not differ significantly. Antepartum admission, induction of labour, caesarean section and neonatal admission were required more frequently in older women. We conclude that pregnancy at age 35 years and above is often difficult and expensive, requiring hospital based interventions to maintain a low perinatal mortality rate.

Does human immunodeficiency virus infection protect against preeclampsia-eclampsia?

OBJECTIVE

In view of recent suggestions that human immunodeficiency virus (HIV) infection may protect against preeclampsia, this study was done to evaluate whether untreated HIV-positive pregnant women have a lower rate of preeclampsia-eclampsia than HIV-negative women.

METHODS

Subjects for this study were pregnant women from Soweto, South Africa, who gave birth from March to December 2002 at midwife-run clinics or at the Chris Hani Baragwanath Hospital and in whom the HIV status was known. A sample size calculation indicated that 2,588 subjects would be required to show statistical significance at P <.05 with a power of 80% for a reduction in the rate of preeclampsia from 8% to 5% with HIV seropositivity, assuming an HIV seroprevalence rate of 30%. Data collection was by record review from randomly selected patient files and birth registers.

RESULTS

In the total sample of 2,600 women, 1,797 gave birth at the hospital and 803 at the midwife-run clinics. The HIV seroprevalence rate was 27.1%. Hypertension was found in 17.3% of women, with 5.3% having preeclampsia-eclampsia. The rates of preeclampsia-eclampsia were 5.2% in HIV-negative and 5.7% in HIV-positive women (P =.61). CD4 count results were available for only 13 women (0.5%).

CONCLUSION

Human immunodeficiency virus seropositivity was not associated with any reduction in the risk of developing preeclampsia-eclampsia.

Where there is no ultrasound: delay and failure of antenatal diagnosis of twin pregnancy

We investigate the problem of late diagnosis of twin pregnancy in Soweto, South Africa, where routine antenatal ultrasound is not available. One hundred consecutive pairs of twins were studied, using the notes of mothers who delivered twins at Chris Hani Baragwanath Hospital and the referring Soweto clinics. A positive history was found in 31 mothers (22 family history, nine previous history of twins). Six mothers did not attend for antenatal care. Twenty-five twin pregnancies were discovered at delivery, 15 of them in the second stage of labour, and 27 were diagnosed accidentally in the third trimester. Only 15 pregnancies were referred specifically for suspicion of twin pregnancy. Most twin pregnancies are detected only in the third trimester or at delivery. Until routine ultrasound is available to all pregnant women, the teaching of antenatal care in South Africa must give emphasis to clinical suspicion of twin pregnancy.

Intrapartum-related birth asphyxia in South Africa--lessons from the first national perinatal care survey

BACKGROUND

The recent amalgamation of data by users of the Perinatal Problem Identification Programme (PPIP) throughout South Africa has culminated in the publication of the Saving Babies report.

OBJECTIVES

To determine the absolute rate of death from intrapartum-related birth asphyxia, and the contribution of intrapartum-related asphyxia to total perinatal mortality in South African hospitals, and to identify the primary obstetric causes and avoidable factors for these deaths.

METHODS

The amalgamated PPIP data for the year 2000 were obtained from 27 state hospitals (6 metropolitan, 12 town and 9 rural) in South Africa. In PPIP-based audit, all perinatal deaths are assigned primary obstetric causes and avoidable factors, and these elements were obtained for all deaths resulting from intrapartum-related birth asphyxia.

RESULTS

There were 123,508 births in the hospitals surveyed, with 4,142 perinatal deaths among infants > or = 1,000 g, giving a perinatal mortality rate of 33.5/1,000 births. The perinatal mortality rate from intrapartum-related birth asphyxia was 4.8/1,000 births. The most frequent avoidable factors were delay by mothers in seeking attention during labour (36.6%), signs of fetal distress interpreted incorrectly (24.9%), inadequate fetal monitoring (18.0%) and no response to poor progress in labour (7.0%). The perinatal mortality rates for metropolitan, town, and rural areas were 30.0, 39.4 and 30.9/1,000 births respectively. The contribution of intrapartum-related birth asphyxia to perinatal mortality in these areas was 10.8%, 16.7% and 26.4% respectively.

CONCLUSION

The high rates of perinatal death from intrapartum-related birth asphyxia in South Africa are typical of those in underdeveloped countries, with the most serious deficiencies in rural areas. Most of these deaths are avoidable and the reduction of these rates presents an important challenge to providers of perinatal care in this country. Areas worthy of research and action include provision of mothers' waiting facilities in rural regions, improvements in fetal monitoring, partogram-based labour management, and the establishment of midwifery staffing norms for South African labour units.

Legal termination of pregnancy among teenagers and older women in Soweto, 1999-2001

BACKGROUND

Legal termination of pregnancy (TOP) was introduced in South Africa in 1996. No data are available to relate the numbers of TOPs to the total number of pregnancies in specific health regions. The level of use of TOPs by women of different age groups is not known.

OBJECTIVE

To determine the proportion of pregnancies that end in TOP, with special reference to maternal age, and to measure trends in use from 1999 to 2001.

SETTING

Greater Soweto, Orange Farm and Lenasia, a densely populated urban health region served by Chris Hani Baragwanath Hospital and comprehensive primary care reproductive health services.

METHODS

Two cross-sectional studies performed in 1999 and 2001, counting all pregnancies managed in state-run health services, including legal terminations, spontaneous miscarriages, ectopic pregnancies and deliveries.

RESULTS

There were 5,412 pregnancies in the study period (9 weeks) in 1999, and 5,316 in the study period (8 weeks) in 2001. The TOP rates decreased from 16.1% to 13.6% (P = 0.20). The TOP rates for teenagers decreased from 22.3% to 16.3% (P = 0.006), but were higher than those for older women (15.2% in 1999 and 13.2% in 2001, P = 0.006 and 0.028 respectively). TOP rates for teenagers 13-16 years decreased from 28.0% to 23.0% (P = 0.44), and rates for older teenagers declined from 21.0% to 14.9% (P = 0.008). In 2001, 16.2% of women aged 35 and above underwent TOP, compared with 12.7% of women aged 20-34 years (P = 0.014).

CONCLUSION

Use of TOP services was highest in women at the extremes of reproductive age. There was a significant decline in TOP rates among older teenagers between 1999 and 2001. These data, from a comprehensive urban reproductive health service, provide a benchmark for comparison elsewhere and in the future.

Maternal and fetal risks associated with prolonged latent phase of labour

A pregnancy cohort study was undertaken to assess the fetal, maternal and obstetric risks associated with the prolonged latent phase of labour. One hundred and fifty cases (latent phase >8 hours) were compared with 100 controls (latent phase <8 hours). Data were collected from hospital records. There were more nulliparas in the case group (73% vs. 39%; P<0.0001). Cases required oxytocin augmentation (62% vs. 17%; P<0.0001) and caesarean section (29% vs. 6%; P<0.0001) more frequently than controls. Thick meconium staining of the liquor was more frequent in cases (15% vs. 5%; P<0.05), as were 5-minute Apgar scores less than 7 (17% vs. 3%; P<0.001) and admission to the neonatal unit (22% vs. 1%; P<0.0001). Prolonged latent phase is associated with increased risks for obstetric intervention and poor fetal outcome. The optimal management of prolonged latent phase is still unknown and requires further study by means of randomised controlled trials.

Vaginal delivery of monoamniotic twins with umbilical cord entanglement. A case report

BACKGROUND

Monoamniotic twinning is frequently complicated by umbilical cord entanglement and fetal death. Should vaginal delivery take place, this may present as an acute intrapartum emergency.

CASE

A 26-year-old woman, gravida 3, para 2, presented in the second stage of labor and gave birth to a macerated, stillborn infant weighing 1,340 g. At the time of delivery it was not known that this was a twin pregnancy. Delivery was achieved only after division of a cord around the neck; it turned out to be that of a live, second twin. This infant, weighing 2,530 g, was delivered by rapid breech extraction and made a satisfactory recovery.

CONCLUSION

When multiple pregnancy or monoamniotic multiple pregnancy has not been excluded, a nuchal cord might be that of an undiagnosed second twin.

Routine heart and lung auscultation in prenatal care

OBJECTIVE

To determine the value of routine heart and lung auscultation (HLA) in pregnant women.

METHODS

Descriptive study by record review of maternal case-files.

RESULTS

Of 3191 mothers, there were 24 with heart disease (0.8%) and 53 with lung disease (1.7%). On routine HLA, only two new cases of heart disease, both with asymptomatic valvular conditions, were detected, and none of lung disease.

CONCLUSION

Routine chest auscultation is of very little value in prenatal care.

Unbooked mothers and their babies--what causes the poor outcomes?

OBJECTIVES

To compare perinatal outcome between booked and unbooked mothers of similar gestational age using birth weight as a proxy for gestational age.

DESIGN

91 unbooked mothers were obtained from a pool of 2,054 mothers from the Chiawelo district in Soweto delivered consecutively at the local community clinic or at Baragwanath, the referral hospital. For each of these, a single booked control from the same pool was found, matched to within 50 g birth weight. Detailed information from case notes was used to compare the two groups and their perinatal outcomes.

RESULTS

Unbooked mothers tended to be at lower obstetric risk. None of the unbooked versus 10 of the booked mothers had a poor obstetric history (2 previous abortions or a previous perinatal death). There was a trend to more twin pregnancies (5 v. 1) and more previous caesarean sections (12 v. 8) in the booked mothers. Perinatal mortality rates were not significantly different (unbooked 258/1,000; booked 268/1,000). Perinatal morbidity, as assessed on the basis of need for neonatal admission and ventilation, was similar in the two groups.

CONCLUSION

Booked mothers were at higher risk, with worse obstetric histories, which probably influenced their decisions to book for antenatal care. Unbooked mothers seemed to be at lower risk, often presenting 'unbooked' because of preterm complications. Many of the latter would probably have booked had their pregnancies advanced uneventfully. As such, unbooked status should not be regarded as a high-risk, poor prognostic category. These mothers and their babies should therefore receive the same perinatal care as booked patients.

Previous birth outcome of antenatal clinic attenders in northern KwaZulu--perinatal and infant mortality rates

A questionnaire survey was undertaken among pregnant women presenting for the first time to antenatal clinics attached to Mosvold Hospital in rural northern KwaZulu. They were asked details of the outcome of their previous pregnancies and the survival of their last-born children. Of 2,388 mothers interviewed, 1,795 had given birth previously. Of these, 41% had had their last deliveries at home, 47% in hospital and 10% at clinics. The perinatal mortality rate was 38/1,000, with no significant differences between home and hospital or clinic births. The infant mortality rate for live-born children was 62/1,000. Fifty per cent of child deaths occurred at home. The study methodology was easily applied and provided an acceptable alternative technique for measuring child mortality rates, which are so difficult to obtain in rural areas in southern Africa.

The effect of continuous low-dose epidural analgesia on uterine work during the active phase of the first stage of labour

Fifty-six patients admitted to hospital in normal spontaneous labour were given a continuous epidural infusion of 10 ml (12.5 mg) of 0.125% bupivacaine during the 1st stage of labour. Uterine contractions were monitored with an intra-uterine pressure transducer and uterine work expressed both in Alexandria units and the area under the curve of the intra-uterine pressure recording summated over 15 minutes. Variables that have been shown to affect uterine contractions were strictly excluded. Our study showed that continuous low-dose epidural analgesia had no significant effect on uterine work during the active phase of the 1st stage of labour when cervical dilatation was 5 cm or more. If cervical dilatation was between 3 and 4 cm there was a 2.9% reduction. This reduction in uterine work was also not significant and in all patients spontaneously returned to its previous level within 91 minutes.

Vaccination status of children aged 12-23 months in the Mosvold health ward of KwaZulu

Using the random cluster sampling technique of the EPI, 210 children of 12-23 months were selected and their vaccination records were examined. Fifty-one (24%) had no vaccination record and 32 of them (15%) were never vaccinated. Using a 'worst' and 'best' estimate analysis it was found that 74-83% had had BCG, 47-56% had had 3 doses of DPT, 48-57% had had 3 doses of polio and 47-56% had had 1 dose of measles vaccine. Between 28-31% of those who received a first dose of DPT or polio vaccine did not receive a third dose. This 'drop-out' effect increased with increasing distance between children's homes and the nearest clinic, as did vaccination coverage in general. As a result of this study, several measures will be undertaken to improve health care delivery in this area. These include the organisation of a mass immunisation campaign later this year, the initiation of more mobile clinics and the training of community health workers.