Response on Rapid Review to Inform Policy Guidance on Welsh Respiratory ECMO Provision

Clinical Effectiveness of ECMO

The review ¹⁰ has concluded that the ‘evidence for the clinical benefit of ECMO over mechanical ventilation cannot be extended to all patients receiving ECMO for respiratory failure’. However, this conclusion is not entirely clear and may be open to different interpretations by policymakers. Previously published randomized clinical trials on respiratory ECMO, such as the EOLIA ¹¹ (ECMO to rescue acute lung injury in severe ARDS) study, involved late cross-over of patients, with some in the control arm ultimately receiving veno-venous ECMO (VV-ECMO). This likely contributed to the non-significant absolute mortality difference and contributed to trial being stopped early on the grounds of statistical futility. ¹¹ A subsequent Bayesian analysis indicated a high posterior probability of survival benefit with ECMO. ¹² Also, consensus group and leading experts recommend the use of ECMO with severe respiratory failure refractory to conventional measures. ¹³ Additionally, the review on clinical effectiveness would benefit from incorporating the recent network meta-analysis by Sud et al, ⁹ which compared VV-ECMO with mechanical ventilation and concluded that VV-ECMO reduced mortality. Furthermore, Whebell et al ¹⁴ retrospectively analysed COVID-19 VV-ECMO outcomes in two UK ECMO centres and found that VV-ECMO led to an absolute risk reduction in mortality of 16.3% to 18.2% compared to conventional management. Including above clinical evidence would provide a more comprehensive perspective on the effectiveness of ECMO therapy in patients admitted with severe acute respiratory failure.

Centre Size on Patient Outcomes

All five studies included in the rapid review have inherent limitations.^15-19 Incorporating findings from the study by Atacan et al, ²⁰ which examined the relationship between ECMO centre volume and patient outcomes, could provide a more balanced perspective. Atacan et al analysed long-term mortality and disability outcomes across three ECMO subgroups—VV-ECMO for respiratory failure, veno-arterial (VA-ECMO) for cardiogenic shock, and extracorporeal cardiopulmonary resucitation (E-CPR)/VA-ECMO for refractory cardiopulmonary resuscitation—and found no significant difference in six-month mortality or new disability between high- and low-volume ECMO centres. Their study, conducted across two large countries with geographical distributions similar to Wales, concluded that centre size does not impact clinical outcomes, provided that both small and large centres collaborate and follow standardized education, training, and evidence-based care protocols. Additionally, the review has cited position statements on the organization of ECMO programs for acute respiratory failure and cardiac failure but has drawn incorrect inferences. The position statement for adult patients states that ‘the annual volume for the entire centre should be at least 20 ECMO cases per year, with a minimum of 12 VV-ECMO cases for acute respiratory failure (ARF)’. ²¹ Similarly, the position statement for cardiac ECMO programs recommends ‘a reasonable goal of at least 30 adult ECMO cases per year, with a substantial proportion being for cardiac failure’. ²² It is important to note that these 30 annual adult ECMO cases can be combination of respiratory and cardiac ECMO and not just isolated cardiac cases.

Healthcare Professional Delivery of ECMO

Nurse-delivered bedside ECMO care is cost-efficient if more than 10 patients ²³ receive treatment annually or the number of ECMO days is between 92 and 155 per year per centre. ²⁴ As per the information shared by National Health Service England (NHSE) only up until the year 2021/2022, the average number of annual ECMO bed days on respiratory ECMO alone for Welsh patients was ∼300 from 2017–2022, and it varied from 230–342 patients (Table 1). The annual number of bed days for cardiac ECMO patients were not shared by NHSE. There is already a precedent of nurse-delivered bedside ECMO care in England and Scotland ECMO centres, and average annual ECMO bed days for Welsh patients suggest that it has the potential to achieve similar efficiency locally in Wales.

OPEN IN VIEWER

Table 1.

As per FOI Data Shared by the NHSE Specialized Commissioning Manager Showcasing the Annual Number of respiratory ECMO Days and Non-ECMO critical care bed Days for Welsh Patients From 2017–22.

Financial year	Total VV-ECMO length of stay (days)	Total critical care days (at the ECMO centre) post VV-ECMO
2017/18	232	148
2018/19	298	232
2019/20	316	193
2020/21	342	265
2021/22 (Incomplete year)	226	195

Cost-Effectiveness

There have been many publications on the cost-effectiveness of ECMO internationally across multiple healthcare settings.^25,26 A very large pragmatic work looking into cost-effectiveness was recently performed by Oude Lansink-Hartgring et al. ²⁷ They performed a cost-effective analysis looking into health-related quality of life post-ECMO therapy, as well as cost per quality-adjusted life years (QALY) in 428 ECMO patients over two years from 2017–19. The costs included were not only hospital costs but also societal costs such as follow-up, general practice visits, specialist care visits and work absenteeism costs. Patients were divided into three subgroups based on the type of ECMO therapy (Respiratory, Cardiac or E-CPR), and all patients nearly achieved pre-illness baseline QALY one-year post-ECMO therapy. The average cost per QALY across all three subgroups was below the UK National Institute for Health and Care Excellence (NICE) cost-effectiveness threshold. ²⁸

Apart from four specific rapid review ¹⁰ themes, policymaking will also benefit from a critical appraisal of the following areas to inform ECMO service provision in Wales.

1. Equity of access to ECMO therapy for the Welsh population as per NHS constitution core values and principles. ²⁹

Internationally, ECMO therapy is now a core and standard organ support tool to provide tertiary critical care, respiratory and cardiac services within a network of hospitals and a key component for most cardio-respiratory pathways. Two hundred and eighty-one centres within Europe ³³ come under European chapter of the extracorporeal life support organization (EuroELSO) (Figure 2), with some centres even serving a population of less than a million against a position statement recommendation of a minimum 2–3 million population for starting a new standalone centre. ²¹ Policymakers also need to consider the annual ECMO financial outflow from the Welsh healthcare budget, which could be used to invest locally in incomplete and partially developed tertiary cardio-respiratory pathways as well as to strengthen and futureproof Welsh NHS workforce as per the strategic government and ministerial priorities.

OPEN IN VIEWER

Figure 2.

Number of adult and paediatric ECMO centres across Europe and beyond. ¹⁶

As per the current Welsh ECMO commissioning pathway, ³⁴ patients are referred to England ECMO centres, which are located geographically distant from the Welsh hospital network (Figure 3). Pathway for respiratory or VV-ECMO is relatively less time-critical, as other rescue strategies like prone ventilation could be instituted until the arrival of an ECMO team from English centres. In comparison, ECMO therapy is required within a few hours in cardiogenic shock (time-critical VA-ECMO for cardiogenic shock) ³⁵ and within one hour in patients presenting with refractory cardiac arrest (super-time-critical VA-ECMO for refractory cardiopulmonary resuscitation).^5,36

OPEN IN VIEWER

Figure 3.

UK map showing multiple ECMO centres in England (>20) and Scotland (3), and no ECMO centre in Wales. ¹

The trajectory, clinical evidence, and incidence of patients requiring timely VA-ECMO for cardiac etiologies^1-8,35,36 are increasing both internationally and within the UK. Policy decisions should consider including the indications for time-critical therapies alongside ECMO use for respiratory failure. The ability to provide timely ECMO within Wales has the potential to also reduce costs associated with current inefficient and incomplete cardio-respiratory pathways and will facilitate future developments, such as uncontrolled donation after circulatory death^37,38, The lack of access to time-critical ECMO for select indications will continue to negatively impact patient outcomes, limiting timely access to effective treatment pathways. It also places additional burdens on patients and their families, who may have to travel further for care. Since 2011, the development of ECMO centre in Wales has been overlooked.^34,39 If this trend continues, the healthcare gap between Wales and both local and international counterparts will widen, affecting patient outcomes and limiting the clinical team’s ability to deliver sustainable, value-based, and cost-effective cardio-respiratory tertiary care services. ³⁰ In summary, policy decisions on ECMO could benefit from a data-driven approach that considers a broader range of perspectives. The rapid review on ECMO use for respiratory failure is constrained by the inclusion of a limited selection of clinical evidence and would benefit from including recent clinical evidence, as well as incorporating a wider range of ECMO indications and clinical data on ECMO trends over the past 5-10 years. These adjustments could make a meaningful contribution to the availability of ECMO therapy in Wales locally and support a more informed policy decisions.