Australia urgently needs a quality improvement approach to emergency laparotomy

A prospective quality improvement register will reduce variations in outcomes and care following emergency laparotomy

A patient presenting to an Australian emergency department with symptoms of acute coronary syndrome will have their presumptive diagnosis, immediate medical treatment and definitive interventional treatment (eg, angioplasty within 90 minutes of arrival) fast tracked by following an agreed, evidence-based pathway that can be monitored for compliance. The 30-day mortality following angioplasty is 2%, but all patients will be admitted to a coronary care unit.1 There are equivalent pathways for patients presenting with stroke.2

Similar evidence-based standards exist for the management of the patient with an acute abdomen and potential sepsis.3 These include a sepsis screen on arrival, blood cultures, antibiotics, lactate testing, an urgent computed tomography scan, a risk assessment to escalate care as appropriate, and post-operative admission to a critical care unit.

However, in Australia there is no established fast-track pathway for patients presenting with an acute abdomen, despite an overall predictive mortality of 8–15%, and double that for patients ≥ 80 years of age.4-10 With hospital staff attention directed elsewhere, the patient with an acute abdomen will wait and, if septic, their mortality will rise by the hour. In the only prospective Australian emergency laparotomy data, less than half of patients had the time of their antibiotic recorded, and in only 13% of these were antibiotics administered within an hour of presentation as recommended.7 Of the highest risk patients, less than half commenced their emergency laparotomy within the standard of 2 hours, and less than two-thirds were admitted to a critical care area post-operatively.4 Australian public hospital administrative data suggest wide inter-hospital variation in practice and outcome (unpublished data).

Meanwhile, outcomes following emergency laparotomy are receiving considerable attention overseas.9,10 This is because emergency laparotomies are common surgical procedures, carrying high 30-day mortality rates, which place significant demands on resources with variable outcomes and standards of care.5,6 Such variation not only suggests that clinical care and outcomes could be improved, but that quality improvement would likely lead to significant cost savings. Mortality rates are substantially higher by 90 days,7,10 suggesting that many emergency laparotomies have limited sustained benefit and that it may be possible to avoid potential futile surgery.

The important steps necessary to deliver the best care for general surgery emergencies were summarised in a seminal United Kingdom report.11 Early studies reflected the limitations of administrative data and the difficulties when translating such data into a clinical setting. In particular, there is no procedure code for emergency laparotomy, which encompasses a wide range of operations with very different outcomes. For these reasons, the UK Emergency Laparotomy Network undertook a prospective study to obtain high quality data. Thirty-five National Health Service hospitals recruited data on 1835 emergency laparotomies,6 finding an overall 30-day mortality of 14.9% (24% in patients aged ≥ 80 years) with a tenfold variation between units.

The Emergency Laparotomy Network prompted the UK government to establish the National Emergency Laparotomy Audit (NELA) in England and Wales.10 The NELA aims to drive quality improvement by collecting and publishing high quality comparative information from all hospitals performing emergency laparotomies. Data collection commenced in 2012 and the third report was published in 2017.10 Many hospitals have since developed their own emergency laparotomy pathways. Improved care, as demonstrated by a reduction in average length of stay of almost 3 days, represented a cost saving of £30 million per annum in bed days alone.10 A key observation was that patients for whom risk assessment was prospectively documented had a lower mortality rate than those for whom it was retrospectively calculated, despite being younger and having lower American Society of Anaesthesiologists physical status grades.

Four hospitals in South East England prospectively measured outcomes before and after the introduction of a bundle of care. The 30-day mortality fell from 15.6% to 9.6%, and at 30 days an additional 5.9 patients were alive for every 100 emergency laparotomies.8 A more recent Danish study noted a fall in 30-day mortality from 21.8% to 15.5% following the introduction of a similar bundle of care.9 Length of intensive care stay fell from 5 to 3 days. Both studies involved substantial quality improvement and education components, and in Denmark the mortality was still falling when the study concluded 2 years after commencement.

However, the Australian and New Zealand health systems differ from those of European countries. The widely dispersed populations of Australia and New Zealand present specific health care delivery challenges, and it would be unwise to extrapolate from overseas results without quality local data.

Although variation in Australian surgical care has been well documented,12 there has been no study on the outcomes or processes related to emergency laparotomies. We analysed Independent Hospital Pricing Authority data (unpublished) relating to abdominal surgery following emergency admissions into public hospitals and found an overall unadjusted in-hospital mortality rate of 5.2% (range, 0–9.3%). There was wide inter-hospital variation in the risk-adjusted standardised mortality ratio and relative length of stay. Length of stay is objective and easy to measure and a short length of stay is a good surrogate marker for uncomplicated care and efficient hospital processes. Use of intensive care varied widely, even in the high risk patients aged ≥ 80 years undergoing procedures such as acute colectomies. There was a more than twofold variation in the national weighted activity unit hospital cost of an emergency laparotomy. However, administrative data are not a substitute for a purpose-designed quality improvement project.

In late 2016, general surgeons in Western Australia undertook a 12-week prospective multi-hospital audit using inclusion and exclusion criteria similar to those of the Emergency Laparotomy Network. Additionally, the audit included compliance with a care bundle and a documented prospective risk assessment, as in the NELA.7 The audit recorded a low 30-day mortality rate (6.6%); however, as almost three-quarters of the emergency laparotomies were performed in a principal referral hospital, the results may not be typical of Australia. It also showed the importance of a documented prospective risk assessment, with the mortality of patients not assessed almost double that of those who had a pre-operative risk assessment.

There is already enthusiastic multispecialty support for a bi-national quality improvement project. The councils of the Royal Australasian College of Surgeons and the Australian and New Zealand College of Anaesthetists have endorsed a business case, an initial working group is determining datasets, definitions are being agreed upon and multi-hospital pilot audits are commencing in both countries. This enthusiasm now needs to be harnessed into a coherent, comprehensive planned project, permitting wider dissemination of the practices that improve outcomes.

This is an ideal project for seeding funding to initiate the development of an online database that, as in the NELA, would provide near real-time data collection and feedback for quality improvement, support data cleansing, analysis and dissemination on individual hospital practice and outcomes, and above all provide widespread education.11 Once established, it would be supported by routine hospital quality and safety activities.

A particular attraction of emergency laparotomy quality improvement is the likely hospital-wide impact. For example, the early identification and treatment of sepsis has been a major driver of improvement,13 and better hospital systems generated by an emergency laparotomy quality improvement project would benefit all patients with sepsis. The setting and documentation of goals of care would improve family and patient satisfaction for high risk patients facing an emergency laparotomy and would help to avoid futile surgery.14 The project requires a multidisciplinary approach involving collaboration between surgeons, anaesthetists, intensivists and emergency physicians. As in the NELA, results would be published by identifiable hospitals so that all could learn from best practice.

A quality improvement project is likely to be highly cost-effective. The NELA costs the equivalent of about A$400 000 per year and, after only 3 years, the reduction in average length of stay led to an estimated cost saving of £30 million per annum in bed days.10 In Australia, a bed saving of one day would represent an estimated saving of $34 million per year (unpublished data).

A recognised limitation of all emergency laparotomy studies to date has been the failure to include patients who presented with an acute abdomen but did not have surgery and subsequently died. Australia is in a unique position to provide a valuable insight into the management of patients at risk of undergoing futile surgery, as all deaths under a surgeon, whether or not an operation was involved, are reviewed by the national surgical mortality audit.15

Patients are rightly demanding provision of high quality outcome data. Australia is many years behind the UK in the provision of such information. The current enthusiasm for a bi-national emergency laparotomy project offers a timely opportunity to remedy this in Australia and New Zealand. While there are clear lessons to be learnt from the NELA, without high quality prospective data there will be no way of knowing whether these lessons have been learnt and then introduced to improve local practice. A patient with an acute abdomen needs to be managed with the same expedited, documented, evidence-based care as a patient with suspected acute coronary syndrome or stroke. It is likely that a bi-national emergency laparotomy quality improvement project will result in a reduction in morbidity and mortality, thereby reducing health care costs.