Early thoughts on ‘Why Mothers Die 2000–2002’

Obstetric anaesthetists, together with many general anaesthetists with only a peripheral interest in obstetrics, eagerly await the publication of each triennial report into maternal deaths in the UK, to see what new messages it has for obstetric anaesthesia. The latest report, published on 12th November, ‘Why Mothers Die 2000–2002’ [1] is no exception. It is the first report of the new Millennium and marks the 50th Anniversary of the first triennial report, which covered the period 1952–54. It is also the first published under the auspices of the Confidential Enquiry into Maternal and Child Health (CEMACH), newly formed from the ashes of the Confidential Enquiry into Maternal Deaths (CEMD) and the Confidential Enquiry into Stillbirths and Deaths in Infancy (CESDI), the new organisation being under the umbrella of the National Institute of Clinical Excellence (NICE). In this triennium there were 391 deaths reported, of which 106 were classified as Direct (death caused by pregnancy occurring within 42 days of delivery), 155 as Indirect, (death from a pre-existing condition aggravated by pregnancy), 36 as Coincidental (death unrelated to pregnancy) and 94 as Late (death occurring between 42 days and 1 year after delivery). This continues the trend of the last triennia where there were more indirect than direct deaths. As in the previous Report, there is an attempt to evaluate the part that social factors play in maternal death. As an example, women who were unemployed with an unemployed partner were up to 20 times more likely to die than women from more advantaged groups. Mortality was higher in women from poor communities or from a minority ethnic group. Late booking or poor attendance at antenatal clinic was recognised as a major risk factor. Other risk factors included obesity, being a victim of domestic violence, and substance abuse. Thus, maternity services should be approachable and flexible enough to meet the needs of all women including the vulnerable and hard to reach. Health education should be targeted at these socially excluded groups and maternity services should be such that all women are motivated to re-attend throughout pregnancy. Venous thrombosis and thromboembolism remains the leading cause of direct deaths with an unchanged overall incidence, although the incidence of death from thromboembolism continues the decline noted in the last triennium, probably the result of improved thromboprophylaxis. That said, the Report highlights substandard practice from a failure to recognise risk factors, and slow introduction or inadequate dosage of thromboprophylaxis and treatment. The mortality rate from eclampsia and pre-eclampsia remains unchanged from the last Report [2]. There remain incidences of inadequate treatment of hypertension with subsequent intracranial haemorrhage, often because a consultant obstetrician was involved late. The message from this is that all healthcare professionals involved in the care of the pregnant mother should be alert to the warning signs of hypertension, such as continued headache, and have a low threshold for measuring blood pressure and analysing urine for protein. Severe and life threatening hypertension should result in early involvement of consultants in obstetrics and critical care and the prompt introduction of antihypertensive therapy. Catastrophic haemorrhage remains a major problem with a reversal of the reduction seen in the last Report [2]. In this triennium, there were 17 deaths due to major haemorrhage compared to seven deaths reported in 1997–9; most of this increase was due to a striking rise in deaths from postpartum haemorrhage. The Report draws attention to the fact that women at known high risk of haemorrhage are still being delivered in isolated units, ill-equipped to manage sudden, life-threatening emergencies. The association between placenta praevia and a uterine scar from previous Caesarean section is once again highlighted. As might be expected, death from major haemorrhage is further discussed in the Anaesthetic Chapter and the responsibility of five of the 17 haemorrhage deaths lay in part with substandard anaesthetic care. Fit and healthy young women initially tolerate haemorrhagic hypovolaemia and there is often a delay in recognising the severity of haemorrhage. It is clear that the management of major obstetric haemorrhage remains a challenge and requires close teamwork, in which anaesthetists can play a major part. The Report recommends early consultant input from obstetrics, anaesthetics, and critical care. In addition, as the rapid development of coagulation failure makes surgical treatment both difficult and ineffective, its management requires early consultant haematology advice. For prompt restoration of circulating volume, the availability of blood, blood products and rapid infusion equipment is essential, as is invasive arterial and central venous pressure monitoring. Further guidance to adequate restoration of circulating volume can be provided by arterial blood analysis of acid base balance and lactate. Successful management of major obstetric haemorrhage requires efficient multidisciplinary teamwork and is best guided by a protocol that is regularly practiced. The number of women dying from diseases aggravated by childbirth is now greater than that directly caused by pregnancy. Deaths from psychiatric causes are the leading cause of maternal death, followed by cardiac disease. A recurring cause of substandard care in these patients is the lack of coordinated multidisciplinary care. Ideally, this should commence pregnancy with accurate expert counselling of the likely risks of pregnancy. Obstetric consultant care should begin early in pregnancy with appropriate input from consultants from other specialties including anaesthesia. Referral to a tertiary centre may be necessary but does not preclude delivery locally. A clear plan for delivery should be formulated from the expert advice and documented and communicated to all likely to be involved in delivery care. The chapter on anaesthesia-related death is startling. Obstetric anaesthetists have been comforted in the past by the dramatic reduction in anaesthetic-related maternal deaths that have occurred since the early 1980s. In the 1982–84 triennia [3] there were 19 anaesthetic deaths, while the subsequent four reports from 1985–1996 [4-7] identified the same total number of deaths in 12 years. The dramatic improvement has been attributed to a number of factors, including improved training and education, increased use of regional anaesthesia in place of general anaesthesia, and the adoption of effective antacid prophylaxis. This improvement is even more impressive when we consider the progressive rise in Caesarean delivery. In the last triennium [2], there were three anaesthetic deaths reported, two related to general and one to regional anaesthesia. In this triennium, there has been an increase to six reported anaesthetic deaths, all related to general anaesthesia. It is important to note that there was, in addition, a late death where the initial event occurred during the 1997–99 triennium and should perhaps be included in that Report's figures, increasing the deaths to four. A rise of anaesthetic-related deaths from four to six is not of statistical significance and should not cause undue concern but what is worrying is the rise in deaths due to poor airway management. Also, as the Report points out, with the decrease in general anaesthesia in favour of greater use of regional anaesthesia, the six deaths reported in this triennium suggests a risk of one death per 20 000 maternal general anaesthetics, a rate similar to that in the 1982–84 Report. Two deaths and the one late death were due to undiagnosed oesophageal placement of the tracheal tube leading to hypoxia and cardiac arrest. In only one of these three cases was capnography used and the tubing to the capnograph became blocked by gastric secretions. In all three cases, auscultation of breath sounds suggested tracheal and not oesophageal intubation. It is also interesting to note that all three of these cases were carried out by anaesthetic senior house officers (SHOs) working in isolation and unable to receive help from a more experienced anaesthetist who might have been able to recognise and correct the misplaced tracheal tube. In one of these cases, there was a failure of equipment checking: the patient rapidly desaturated because pre-oxygenation had been ineffective, as the anaesthetist had not noticed that the fresh gas outlet had not been reconnected following a previous case under regional anaesthesia. This highlights the importance of checking anaesthetic equipment before each case [8]. Furthermore, the problem can be avoided by the use of a separate oxygen source for administering supplemental oxygen during regional anaesthesia. Two deaths, one a Caesarean section and the other a mid-trimester termination of pregnancy, were the result of inadequately managed hypoventilation. Both occurred in isolated sites, which the assessors felt contributed to the substandard care. In addition, one patient was obese and in one of the cases capnography was not used. There was one death due to gastric aspiration following a failed intubation in an obese patient. It was unclear whether she had received antacid prophylaxis. The final death was as a result of unsuccessful resuscitation from anaphylaxis shortly after induction of anaesthesia for evacuation of products of conception. The diagnosis was confirmed by elevated serum tryptase indicating mast cell degranulation. The Report further identifies 20 maternal deaths in which poor anaesthetic care contributed. The aspects that are highlighted by these deaths include poor multidisciplinary cooperation and communication, lack of appreciation of the severity of illness and inadequate management of severe haemorrhage. What are we to learn from these deaths? It is both disappointing and worrying that failure to detect oesophageal intubation has re-emerged as a cause of maternal death. It was inexcusable in the triennium 1985–87 [4] where it was the cause of three deaths, but at least that was at a time when capnography was not so readily available as today and not part of mandatory minimal monitoring [9]. The blame has to lie with a lack of awareness by our trainees that clinical signs of oesophageal intubation are unreliable and that correct tube placement should always be confirmed by capnography. Capnography is also a useful indicator of adequate ventilation, a feature in two further deaths. These cases lend further voice to the concern that trainees are not getting sufficient experience in managing the airway. The continued reduction in trainee hours and the increasing use of the laryngeal mask airway are some of the underlying reasons for this concern. If a decrease in experience amongst trainees is a problem, we need to explore alternative ways to improve airway skills. There are other training-related issues. There is a suggestion in the Report that maternal cardiac resuscitation is not well carried out, that failed intubation drills are poorly performed and that trainees of all specialties are poor at identifying the severity and immediacy of the patient's condition, thus not involving senior staff in time. These deficits can be addressed by targeted training, the use of simulators, the teaching and practice of drills as well as formal assessment of the necessary competencies. The Royal College of Anaesthetists has led the way in competency-based training [10] and now all novice anaesthetic trainees have their basic anaesthetic competency assessed before being allowed to work under indirect supervision, usually after 3–4 months. In the UK, trainees are not allowed to be on call for delivery suite until they have completed one year of training [11]. In view of the fact that the majority of the anaesthetic deaths involved SHOs working alone, should we only allow specialist registrar or experienced career grade anaesthetists to be responsible for anaesthesia on delivery suites? At a time of reduced trainee hours, this may not be possible given the current configuration of obstetric services. If this is the case, we must ensure that obstetric anaesthetic trainees are trained and formally assessed as having the relevant competencies to manage the maternal airway, obstetric emergencies and the critically compromised pregnant mother before going on the obstetric anaesthetic night rota. What about the problem of isolated maternity sites? Should we continue to deliver obstetric anaesthesia in isolated units without an adequate range of backup services? Such sites are criticised in several of the chapters in the Report. Reconfiguring services to phase out some of these sites is a long-term goal but, in the meantime, every effort should be made to minimise risk at isolated sites. This includes patient risk assessment and selection, staffing with experienced personnel, and clear protocols for managing patients including transfer at the earliest signs of unexpected complications. The Report once again identifies a lack of timely involvement of critical care services. There are many reasons for this, with lack of provision, delayed communication with intensivists, and delayed recognition of severity of illness amongst them. Improved training of maternity staff in recognising ill patients, perhaps supplemented by early warning severity of illness scores, would seem a sensible way forward. Breaking down any reluctance to seek help from Intensive Care should be overcome by the increasing development of critical care outreach services [12] aimed at increasing Intensive Care staff involvement in patient management before they need admission to ICU. It is refreshing to see the Report acknowledge the devastating effect that a maternal death can have on the staff involved. Staff require individual support and can benefit from counselling. This may be provided locally but alternatively the Report informs us that the Joint Committee on Good Practice is establishing a counselling service and can be contacted through the Royal College of Anaesthetists or the Association of Anaesthetists of Great Britain and Ireland. There are undoubtedly many lessons that can be learnt from the triennial reports of maternal deaths, but it has been said that death is the tip of the morbidity iceberg and there are many other obstetric anaesthetic challenges. For instance, there are few maternal deaths related to regional anaesthesia but that does not mean that its conduct is perfect and without morbidity. It is therefore important when contemplating actions to minimise anaesthetic risk on the Delivery Suite that we do not solely focus on those that cause mortality.