In March 2020, as the covid-19 pandemic was escalating in the UK, the National Institute for Health and Care Excellence (NICE) published its delayed National Guideline (NG156) on the management of people with Abdominal Aortic Aneurysm [AAA]. [1] The Guideline is the product of four years’ work by NICE, its technical team, and its AAA Guideline Development Committee [GDC] (to which the authors were appointed). The development of this guidance has been a drawn-out process and has caused much debate.

NICE’s draft guidance was published for stakeholder comment in May 2018. It made fifty recommendations about the comprehensive management of people with AAA. Controversially, it recommended against the use of Endovascular AAA repair [EVAR] for the repair of unruptured AAA in most circumstances. 

EVAR is associated with approximately one-third of the perioperative mortality of open surgical repair [OSR] and substantially less perioperative morbidity. [2] On this basis, EVAR has been enthusiastically adopted by vascular specialists. However, the perioperative survival benefit of EVAR (versus OSR) is progressively eroded so by four years survival is equal and beyond eight years survival is worse with EVAR. [3] Further interventions to the device are needed in nearly a quarter of patients. [3] That EVAR might be an alternative where OSR is high-risk was dealt a major blow by the results of a randomised controlled trial [RCT] demonstrating no survival benefit versus conservative management in this comorbid cohort. [4] EVAR is cost-ineffective (compared with OSR) for people fit for OSR and is unlikely to be cost-effective for many people unfit for it. [5] In most circumstances, the opportunity cost of undertaking EVAR is well above conventional NHS “willingness to pay” thresholds. The corollary of this is that undertaking EVAR causes net harm by diverting NHS resource from where it could be more effectively spent. NICE has rejected many (mainly drug) technologies on the basis of cost-effectiveness. 

Despite accumulating and concordant long-term data confirming these clinical and health-economic conclusions, vascular specialists and medical device industry partners have enthusiastically developed and promoted EVAR. The anatomical limitations of the technology have been expanded (“complex EVAR”). The recommendation in the draft guidance that EVAR should not be used therefore represented a threat to status-quo management of AAA. 

Consultation on the draft guidance was undertaken by NICE between May and June 2018. Six hundred and thirty five pages of stakeholder feedback was received from UK and international professional organisations involved in the management of people with AAA and in the accreditation and training of vascular specialists. [6] Feedback was also received from several UK vascular units; health technology industry organisations; parliamentary special interest groups; the AAA screening programme; vascular specialist commissioners and academic groups involved in AAA research. No direct feedback was received from organisations representing people with AAA. Themes raised by stakeholders were type, generalisability and contemporary relevance of the evidence that was assessed in the development of the guideline, the applicability of generic evidence to subgroups (especially women), training, implementation issues, and the removal of choice in shared decision making. 

This feedback prompted NICE to do supplementary analysis and the GDC to reconsider their recommendations. A systematic review of observational cohort data did not provide evidence to undermine the conclusions of the RCTs, but the GDC reevaluated the generalisability of some of the RCT and health-economic data and reconsidered arguments about choice. [7] Iterative amendments to the draft guidance were made from July 2018 and a settled text was unanimously agreed by the GDC in September 2019. The recommendation against EVAR was removed but explicit discussion about the uncertainty of its benefits and harms was recommended during shared decision making with patients. Priorities for future research were highlighted, particularly for “complex EVAR” which has not been subject to rigorous assessment in an RCT.

Unexpectedly, the NICE Board rejected the GDC’s revisions, citing implementation issues and in particular the increase in critical care capacity which would be required should there be a wholesale switch from EVAR to OSR. Approximately 60% of UK AAA repairs were performed with EVAR in 2018, with some centres using it almost exclusively. [8] Additionally, NICE claimed that the revised guideline remained unacceptable to clinicians involved in the care of people with AAA, though it is unclear how it arrived at this assertion as the revised guidelines remained embargoed—there are no vascular specialists on the Board. Instead NICE’s Board wrote their own recommendations about the repair of unruptured AAA, recommending either OSR or EVAR effectively as equivalent, and diluting the requirement to undertake “complex EVAR” only within an RCT to undertaking it within “special arrangements for consent and for audit and research.” 

The vascular surgery and interventional radiology community must take some responsibility for the discrepancy between the evidence base and accepted practice in AAA repair. This placed NICE in a difficult position. Despite this it is unclear how NICE justified its Board’s revision of the GDC’s recommendations, particularly in light of the clear cost effectiveness arguments and standing advice to GDC committee members on the centrality (albeit not exclusive centrality) of this in making recommendations. NICE has refused to discuss whether the Board sought additional advice from external third parties, what influence they had, how their conflicts of interest were managed, how they were chosen or how much of NICE’s technical analyses they were familiar with.

The NICE guideline on the management of AAA contains many positive recommendations and will improve the management of people with AAA. However, the guideline fails to offer any meaningful guidance for the issue of how (and if) to repair unruptured AAA. NICE has created an anodyne document that merely reflects current practice, is difficult to use and easy to ignore. In doing so NICE has chosen to follow rather than lead.

However, NICE’s U-turn has wider relevance than just for people with AAA. The legitimacy and authority of NICE guidance derives largely from the openness and transparency of its processes within a clear published philosophical, economic and ethical framework. [9] In the creation of alternative recommendations NICE has rejected the unanimous conclusions of its independent GDC and bypassed its own published processes. It has then failed to be open and transparent in describing the alternative processes and principles by which this occurred, or the details of the individuals directly or indirectly involved. While NICE has asserted its right to over-rule a GDC it has (to the authors’ knowledge) never previously done so. [10] This behaviour undermines NICE’s legitimacy and authority and calls into question its reputation for scientific rigour, independence, and procedural justice. If NICE cannot (or will not) perform its function to make evidence-based recommendations using its charter principles then what is it for?

Christopher Hammond, Consultant vascular radiologist, Leeds. Member, NICE AAA GDC, 2014-2020

Andrew Bradbury, Sampson Gamgee Professor of Vascular Surgery, University of Birmingham, Chair, NICE AAA GDC, 2014-2020

On behalf of all the members of the NICE AAA GDC, 2014-2020

Competing interests: None declared

References

1. NICE guideline [NG156]. Abdominal aortic aneurysm: diagnosis and management.
2. Available at: https://www.nice.org.uk/guidance/ng156. Accessed 01.11.20
3. Powell JT, Sweeting MJ, Ulug P et al. Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years. Br J Surg 2017;104:166 – 178.
4. Patel R, Sweeting MJ, Powell JT et al. Endovascular versus open repair of abdominal aortic aneurysm in 15-years’ follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial. Lancet 2016;388:2366 – 2374.
5. Sweeting MJ, Patel R, Powell JT et al. Endovascular Repair of Abdominal Aortic Aneurysm in Patients Physically Ineligible for Open Repair. Ann Surg 2017;266:713-719.
6. Patel R, Powell JT, Sweeting MJ et al. The UK EndoVascular Aneurysm Repair (EVAR) randomised controlled trials: long-term follow-up and cost-effectiveness analysis. Health Technology Assessment 2018;22:1 – 132.
7. NICE NG156 Consultation on draft guidance – stakeholder comments table. Available at: https://www.nice.org.uk/guidance/ng156/documents/consultation-comments-and-responses-2 Accessed 01.11.2020
8. NICE NG156 Evidence review K2: Observational evidence on the effectiveness of endovascular aneurysm repair compared with open surgical repair of unruptured abdominal aortic aneurysms. 2020. Available at: https://www.nice.org.uk/guidance/ng156/evidence/k2-observational-evidence-on-the-effectiveness-of-endovascular-aneurysm-repair-compared-with-open-surgical-repair-of-unruptured-abdominal-aortic-aneurysms-pdf-255167681367 Accessed 01.11.2020
9. VSQIP Homepage. https://www.vsqip.org.uk. Accessed 01.11.20
10. NICE. Our Charter. Available at: https://www.nice.org.uk/about/who-we-are/our-charter Accessed 01.11.20
11. Chrisp, P. Comment on: Endovascular treatment of abdominal aortic aneurysm. A NICE U-turn. Br J Surg 2020;107:e281 doi: 10.1002/bjs.11717