The Obstetrician & Gynaecologist 2007;9:2:116-120
doi: 10.1576/toag.9.2.116.27313
Copyright © 2007 by the Royal College of Obstetricians and Gynaecologists.
Obstetric haemorrhage at Queen Charlotte's and Chelsea Hospital
Sara Paterson-Brown, MA FRCS FRCOG, Consultant Obstetrician and Gynaecologist1
1. Queen Charlotte's and Chelsea Hospital, Du Cane Road, London, W12 OHS, UK Email: spaterson-brown{at}hhnt.nhs.uk
 |
Abstract
|
|---|
Key content:- Obstetric haemorrhage complicates 10% of pregnancies globally.
- There is a 1% case mortality rate.
- It is the leading cause of maternal death worldwide, accounting for over a quarter of all direct deaths.
- In the United Kingdom, death from obstetric haemorrhage is 100 times less common than in developing countries, but this is rising. It is unclear whether more women are bleeding or whether standards of care are falling.
Learning objectives:
- To be aware of the risk factors for obstetric haemorrhage and why they are increasing.
- To learn about how detailed systematic audit of cases of massive obstetric haemorrhage, together with staff feedback, can improve and maintain standards of care.
- To learn how multidisciplinary training in the local workplace can help staff to manage this condition and promote teamwork.
Ethical issues:
- How anonymous, systematic audit of patient details can benefit future patients.
Please cite this article as: Paterson-Brown S. Obstetric haemorrhage at Queen Charlotte's and Chelsea Hospital. The Obstetrician & Gynaecologist 2007;9:116–120.
Keywords caesarean section / maternal mortality / multidisciplinary support / obstetric haemorrhage / oxytocin
|
The national picture in the United Kingdom
|
|---|
Massive haemorrhage is the second most common cause of direct maternal death,1 occurring at a rate of 8.5 per million pregnancies. Why this rate has more than doubled over the last 3 years (Figure 1) is unclear: substandard care, identified in over 70% of cases, was similar in previous reports, so the concern is that more women are experiencing serious bleeding.
The true rate of massive obstetric haemorrhage is variable as definitions differ and estimated blood loss can be inaccurate,2 but it is the most common cause of severe morbidity, occurring in 2–7 per 1 000 pregnancies.1,3 Routine data on intensive care unit (ICU) admissions of obstetric patients is not yet available but, fortunately, the Intensive Care National Audit and Research Centre (ICNARC) has just updated its database to include pregnancy fields. Comprehensive reportage of ICU admissions for this serious condition should, therefore, be available from 2007. Previous reports have shown that haemorrhage is one of the main reasons for ICU admissions in pregnancy, but with different local arrangements in high dependency care provision, rates will vary.4,5 Trapping high dependency care information is extremely difficult but developments in coding are likely to follow on from ‘payment by results’ and may prove to be the catalyst needed to recognise the extent of this morbidity and the workload it demands.
|
The local setting
|
|---|
Queen Charlotte's and Chelsea Hospital (‘Queen Charlotte's’) is a tertiary obstetric unit situated in West London. It combined with the Hammersmith maternity unit in November 2000 and now approximately 5 000 babies are delivered there per annum. Queen Charlotte's serves a community whose birth rate is rising, with almost 60% of pregnancies being from ethnic minorities. Most women are delivered in the labour ward, which has two operating theatres, 19 delivery rooms and a high dependency/recovery area. In addition, approximately 750 women are delivered in a midwife-run birth centre and approximately 30 have a planned home birth. There is also a one-to-one scheme for socially disadvantaged/excluded clients; for example, teenagers and travellers. As in other local units, midwives provide excellent care, but inadequate numbers mean that one-to-one care on the delivery suite is rarely achieved. There is 24-hour anaesthetic cover, 40 hours of committed obstetric consultant cover and junior doctors are rotated through the regional training programme.
Incidence of obstetric haemorrhage
There have been no maternal deaths caused by haemorrhage since detailed systematic audit of all deliveries commenced in 1992. During this time over 60 000 women have delivered in our unit. This is in keeping with national figures. Admissions to the ICU for obstetric haemorrhage are rare (consistently approximately 1 per 1 000 maternities) and are lower than the reported rates quoted above as good use is made of the high dependency area, which is run by labour ward staff with appropriate training and anaesthetic support.
Massive obstetric haemorrhage (MOH), defined as >1 000 ml clinically estimated blood loss, has been systematically audited over the last decade (Figure 2). There is regular training of staff on the tendency to underestimate blood loss (highlighted by the discrepancy between the recorded rates of haemorrhage and transfusion practice), which may partially explain the rise in rates of estimated MOH, as the rate of blood transfusion has remained remarkably stable at approximately 3% (Figure 2). However, to reduce blood usage the unit's policy has changed and includes:
- routinely checking ferritin levels at booking and encouraging appropriate antenatal iron supplementation
- practising active management of the third stage of labour unless women decline
- no longer giving ‘top-up’ blood transfusions to asymptomatic women with a low haemoglobin in the postpartum period.
It is, therefore, likely that there has been a true rise in the incidence of haemorrhage, despite there being no increase in blood transfusion rates.
View larger version (7K):
[in this window]
[in a new window]
|
Figure 2
Massive obstetric haemorrhage and transfusion practice at Queen Charlotte's and Chelsea Hospital (expressed as % of pregnancies)
|
|
|
Are risk factors for haemorrhage increasing?
|
|---|
Rising maternal age
Queen Charlotte's has always served a relatively mature population but, whereas 10% of pregnant women were aged 37 years or older 10 years ago, the figure has now doubled. From previous studies we know that rising maternal age (year by year) is positively associated with complications and interventions, each of which is associated with increased haemorrhage (Figure 3).6 Massive obstetric haemorrhage occurs in 1.5% of those under 37 years compared with 3% in those over 37 years: pregnancies in older women account for between a quarter and half of all major haemorrhages in the unit.
View larger version (12K):
[in this window]
[in a new window]
|
Figure 3
Postpartum haemorrhage >1 000 ml and mode of delivery for singletons and overall rate in multiple pregnancy at Queen Charlotte's and Chelsea Hospital. There was a PPH >1 000 ml in 1.6% of all vaginal deliveries; 4% of women having a CS had a PPH >1 000 ml Br = breech; CS = caesarean section; MP = multiple pregnancy; PPH = postpartum haemorrhage; RF = rotation forceps; SVD = spontaneous vaginal delivery; Tr = traction forceps; V = ventouse
|
|
Multiple pregnancies
Delayed childbearing has made fertility treatment more common. In turn, multiple pregnancies at Queen Charlotte's have risen by 50% over the last decade and appear to be associated with higher rates of obstetric haemorrhage (Figure 3).
Maternal obesity
Rates of maternal obesity are rising but with changes in practice over the last decade for antenatal maternal weight recording, reliable analysis retrospectively is not possible and, so, this is currently being explored prospectively. However, fetal macrosomia has not altered significantly, with approximately 10% of babies weighing at least 4 kg since 1995.
Caesarean section
Caesarean section rates are rising and, therefore, increasing numbers of pregnant women will have had one: 12% in 2005 had a previous caesarean section compared with 7% 10 years ago.
Placenta praevia
This is more common after previous caesarean section: there has been a doubling of the incidence of placenta praevia from 0.4% of pregnancies in our unit up until 1997, rising steadily to 0.8% in 2005.
Morbidly adherent placenta
Morbidly adherent placentas are becoming increasingly frequent: the numbers requiring hysterectomy are shown in Figure 4, but a significant number of these were the result of uterine surgery other than caesarean section (myomectomy and surgical evacuation of products of conception), as illustrated. However, over the last 4 years some cases of placenta praevia accreta resulting from previous multiple caesarean sections have been successfully managed conservatively and, therefore, are not reflected in the numbers for massive haemorrhage or hysterectomy. These are mentioned further below.
View larger version (52K):
[in this window]
[in a new window]
|
Figure 4
Numbers of total abdominal hysterectomies (TAHs), morbidly adherent placentas (MAPs) and morbidly adherent placentas due to previous caesarean section (MAPCS) at Queen Charlotte's and Chelsea Hospital, 1997–2005
|
|
|
The clinical approach to teaching and managing massive haemorrhage
|
|---|
As recommended by the Confidential Enquiries into Maternal Deaths in the United Kingdom, there has been a protocol for the management of massive haemorrhage for many years, which is updated annually (Box 1). Management of haemorrhage is high on the training programme for all staff: theoretical teaching occurs at 6-monthly compulsory training sessions. Clinical knowledge and behaviour are taught and assessed regularly at multidisciplinary emergency scenario sessions, which are attended by all grades of staff (Box 2).
Training in management of MOH stresses the importance of initial resuscitation and rapid assessment, with the aim of limiting blood loss and treating the cause, preserving fertility whenever possible. Conservative measures used include the hydrostatic Rüsch® balloon (Figure 5) and the B-Lynch suture (Figure 6). A cell salvage machine has recently been acquired and is set up for use in women who are at high risk of bleeding.
View larger version (241K):
[in this window]
[in a new window]
|
Figure 6
Example of a B-Lynch suture – a brace suture inserted into a uterus following recurrent atonic episodes after a twin caesarean section
|
|
Access to advanced techniques for haemorrhage control includes intervention radiology (Figure 7) and, more recently, recombinant activated factor VII. Although these are rarely required, they have been of enormous benefit in a small minority of women: approximately four women per year require embolisation and activated factor VII has been used with eight women since introducing it in 2003.
View larger version (93K):
[in this window]
[in a new window]
|
Figure 7
Sequence of radio contrast images illustrating the site of bleeding (arrows), following failed surgical repair of a spiral vaginal tear (vaginal pack in place). The femoral catheter was advanced through the right internal iliac artery (a) to a vaginal branch of the right uterine (b), which was then embolised with subsequent haemostasis (c). (Images kindly provided by James Jackson, Intervention Radiologist, Hammersmith Hospital)
|
|
Antenatal diagnosis of placenta praevia accretas is attempted in all women at risk of this complication, using both colour flow Doppler and magnetic resonance imaging (MRI). Where present, if the woman wants to preserve her uterus, conservative management is offered: the uterine incision is placed away from the placenta and if the placenta fails to separate it is left in place and the uterus closed (Figure 8). Careful follow-up postnatally involves clinical review and serial ß-hCGs twice weekly, together with regular ultrasound scans. The first three women managed conservatively were successful and reported.7 Five more women have since been managed in this way but two have required hysterectomy for haemorrhage – one shortly after surgery and one 6 weeks later. One woman treated successfully has since had another baby with no recurrence of her placenta praevia or accreta.
When postpartum hysterectomy is thought to be necessary, the consultants have a policy of calling in another colleague whenever practicable. Confirming and sharing difficult decisions is felt to be an important part of safe practice and, just as importantly, being seen to do this sets a good example to trainees who should never feel that they, or anyone else, are working in isolation.
Despite clinical impressions to the contrary, postpartum hysterectomy has not shown any obvious increasing trend (Figure 4). This may be partly because of the conservative measures (mentioned above) introduced over this last decade. Since 1995, 28 hysterectomies have been performed in 46 668 pregnancies (0.6 per 1 000). This is slightly higher than the rates of 0.4–0.45 per 1 000 reported recently from regional and national data;8,9 this may be because women are referred from neighbouring hospitals with placenta accreta. In addition a small minority of hysterectomies were undertaken for coexisting malignancy, not haemorrhage.
|
Standards of care and improvements introduced
|
|---|
Because of persistent high rates of substandard care highlighted in the Confidential Enquiries into Maternal Deaths in the United Kingdom, all cases of massive haemorrhage since 1995 have been audited by case note review to identify problems and improve standards of care. This work was originally conducted by registrars working under supervision but it now comes under the scope of risk management (through the Clinical Negligence Scheme for Trusts requiring all massive haemorrhages to be reported) so are reviewed by the consultant lead for the delivery suite. ‘Exemplary practice’, rather than ‘substandard care’ is measured and reported back to staff and this has improved from 40% initially to above 80% within 12 months of commencing the reviews. This improvement has been sustained over the decade. Maintenance of this improvement may be explained by the rapid review and feedback of any problems identified back to the staff involved and by the voicing of general points of learning in the scenario teaching sessions. This approach is simple, personal and effective and requires time and enthusiasm rather than complicated systems.
Over this period, various issues were addressed:
- Initially, poor documentation made detailed assessment of care difficult but this was resolved after being highlighted in training sessions.
- The need to recognise the severity of blood loss — in particular, not to underestimate revealed bleeding and to recognise concealed bleeding — was the main clinical issue and this was addressed through training.2
- Communication problems were identified and resolved:
- A direct line ‘red phone’ was installed between the obstetric theatre and the transfusion laboratory, enabling an anaesthetist to talk directly to transfusion staff. This was a job previously delegated to a junior midwife or a senior house officer and misunderstandings were common.
- Clinicians now specify the time by which they need blood. This has reduced delays, as the staff in the laboratory know the exact degree of clinical urgency.
- The haematologist on call out of hours carries an on-call mobile phone to avoid delays in tracking them down through the switchboard.
|
Ongoing problems and their solutions
|
|---|
Until 2002 approximately 1.4% of all women having caesarean sections returned to the operating theatre for a variety of reasons, most commonly superficial wound problems, but this figure is rising. Last year, 36 women (2.2% of those having caesarean sections) returned, half of them with an atonic uterus that had filled up with blood while in the recovery room. As this develops fairly silently this is an area of serious concern. Whether this is a result of changes in surgical experience or women's characteristics it is not clear from detailed analyses. One possible explanation is that it may be a consequence of the reduction in the dosage of routine intravenous oxytocin given at all caesarean sections, from 10 iu to 5 iu. This followed recommendations in 2001 by the Confidential Enquiries that 10 iu was excessive.10 The maternal death referred to in the Confidential Report involved a cardiac patient, but the recommendation was placed in both the anaesthetic and the cardiac chapters, although particular caution with cardiac patients or cardiovascular compromise was specified. A change in practice was apparent in the UK soon after and, with similar caution being reiterated subsequently by a small randomised trial, this reduced bolus dosage of oxytocin is now widely established.11,12
In response to these increased problems with post-caesarean section uterine atony, the use of oxytocin infusions after caesarean section has increased and various uterotonic regimens are being compared to try to minimise this complication. A randomised trial from Canada13 has provided reassuring evidence that 10 iu of oxytocin, given as a bolus during the third stage of labour in healthy women with an epidural, appears to be safe.14
|
Multidisciplinary support
|
|---|
Dialogue is encouraged both informally, on a daily and case-by-case basis, and formally, through three multidisciplinary discussion forums: the Delivery Suite Management Group, the Directorate's Risk Management Committee and the Directorate's Blood Transfusion Committee.
Sharing ideas on problems encountered and their possible solutions in a formal setting puts weight behind recommendations, which are then minuted and discussed at executive level.
|
Summary
|
|---|
Risk factors for obstetric haemorrhage have increased and the actual incidence at Queen Charlotte's Hospital has also risen over the last decade. Transfusion rates have stayed remarkably stable over this time and the standard of care, which improved when systematic audit was first introduced, has been maintained. Concerns about uterine atony after caesarean section are being addressed by a number of trials across the UK.
If mothers of the future could be fitter, thinner and younger it might be possible to curb caesarean section rates and reduce the incidence of haemorrhage but, realistically, the current trend is likely to continue. Local rehearsals and training must be promoted. In addition, senior personnel need to be increasingly available in person to manage obstetric haemorrhage, as a continuing reduction in the amount of experience gained by junior doctors is inevitable.
|
References
|
|---|
- Confidential Enquiry into Maternal and Child Health. Why Mothers Die 2000–2002. The Sixth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. London: RCOG Press; 2004
- Bose P, Regan F, Paterson-Brown S. Improving the accuracy of estimated blood loss at obstetric haemorrhage using clinical reconstructions. BJOG 2006;113:919–24. doi:10.1111/j.1471-0528.2006.01018.x[Medline]
- Waterstone M, Bewley S, Wolfe C. Incidence and predictors of severe obstetric morbidity: case-control study. BMJ 2001;322:1089–94. doi:10.1136/bmj.322.7294.1089[Abstract/Free Full Text]
- Umo-Etuk J, Lumley J, Holdcroft A. Critically ill parturient women and admission to intensive care: a 5-year review. Int J Obstet Anesth 1996;5:79–84. doi:10.1016/S0959-289X(96)80001-X[Medline]
- Baskett TF, Sternadel J. Maternal intensive care and near-miss mortality in obstetrics. BJOG 1998;105:981–4.
- Rosenthal AN, Paterson-Brown S. Is there an incremental rise in the risk of obstetric intervention with increasing maternal age? BJOG 1998;105:1064–9.
- Jain A, Sepulveda W, Paterson-Brown S. Conservative management of major placenta praevia accreta: three case reports. J Obstet Gynaecol 2004;24 Suppl 1:S63.
- Knight M, Kurinczuk J, Spark P, Harris C, Brocklehurst P. Peripartum hysterectomy: a national case-control study to investigate incidence and risk factors using the UK obstetric surveillance system (UKOSS). J Obstet Gynaecol 2006;Suppl S20
- Eniola OA, Bewley S, Waterstone M, Hooper R, Wolfe CDA. Obstetric hysterectomy in a population of South East England. J Obstet Gynaecol 2006;26:104–9. doi:10.1080/01443610500443196[Medline]
- Lewis G, editor. Why Mothers Die. The Fifth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom, 1997–1999. London: RCOG Press; 2001
- Bolton TJ, Randall K, Yentis SM. Effect of the Confidential Enquiries into Maternal Deaths on the use of syntocinon at Caesarean section in the UK. Anaesthesia 2003;58:277–9.[Medline]
- Pinder AJ, Dresner M, Calow C, Shorten GD, O'Riordan J, Johnson R. Haemodynamic changes caused by oxytocin during caesarean section under spinal anaesthesia. Int J Obstet Anesth 2002;11:156–9. doi:10.1054/ijoa.2002.0970[Medline]
- Davies G, Tessier J, Woodman MC, Lipson A, Hahn P. Maternal hemodynamics after oxytocin bolus compared with infusion in the third stage of labor: a randomized trial. Obstet Gynecol 2005;105:294–9. doi:10.1097/01.AOG.0000148264.20909.bb[Medline]
- Weeks A. A bolus of oxytocin administered during the third stage of labour was not associated with an adverse hemodynamic response. Evidence-Based Obstet Gynecol 2006;8:12–3. doi:10.1016/j.ebobgyn.2006.01.005
TOP
Abstract
The national picture in...
