The NICE draft guidelines for the management of atrial fibrillation need key revisions

The National Institute for Health and Care Excellence (NICE) draft guidelines for the management of Atrial Fibrillation (AF) contain recommendations at odds with available scientific data and recently updated guidelines from learned societies worldwide. In our view, the draft guidelines severely limit patient choice and access to appropriate treatments. [1,2]

Good quality rhythm strips, which use modern technology are increasingly available in the home or at work, including hand-held ECG recorders , and can reliably identify AF. [3] NICE’s insistence on a 12-lead ECG to diagnose frequently paroxysmal arrhythmias such as AF is, in our view, unnecessary and imposes avoidable delays that miss opportunities for timely diagnosis and treatments. 

Patients with AF have higher risk of stroke, often with greater severity and functional impairment. [4,5] Oral anticoagulants (OACs) provide effective prevention, but confer increased risk of major bleeding. NICE is unexpectedly recommending switching from the widely used HAS-BLED score to the ORBIT (Outcomes Registry for Better Informed Treatment of Atrial Fibrillation) score, and say in these draft guidelines that ORBIT is the “most accurate tool to predict the risk of major bleeding” based on calibration. [1] Implementing ORBIT will likely cause confusion as UK healthcare professionals principally use HAS-BLED. Furthermore, HAS-BLED is supported by evidence and prospective trial data and widely validated across all AF patients (whether anticoagulated or not), whereas ORBIT lacks similar robust data and is tested only in selected cohorts already on OACs. [6,7] We urge NICE to retain the HAS-BLED score to predict bleeding risk, but agree that bleeding risk scoring systems should not be used to withhold anticoagulation.

Our greatest concerns about these draft guidelines surround NICE now advocating just two approved “Direct” OACs (DOACs) instead of four.  This impacts all patients with ≥1 stroke risk factor, removing access to once daily medication, withdrawing patient choice, limiting patients’ options for tailored OAC therapy, and effectively rendering two DOACs (edoxaban and rivaroxaban) ineligible for AF. NICE’s draft recommends switching once-daily DOACs or warfarin to twice-daily therapies, even for patients who are stable with their treatment. This introduces unnecessary risks without proven advantages.

By claiming material differences between DOACs, NICE misrepresents publications showing absence of head-to-head data, similarities among DOACs, heterogeneity between DOAC studies affecting outcomes and adverse reactions, and meta-analyses where outcomes in the comparator “warfarin-treated” arms differ substantially. [8-18] Indirect comparisons of heterogeneous studies should not be over-interpreted or used to generate unjustified therapeutic recommendations in national guidelines. [19-22]

Long term preventative treatments, like stroke prevention, involve asymptomatic patients who may not immediately recognise the benefits of prolonged therapy. Poor adherence to medications accounts for ~50% of the shortfall in realising therapeutic goals, while medicating once daily has superior adherence, decreases non-adherence for chronic cardiovascular diseases, is preferred by > 80% patients, and polypharmacy associates with higher risk of stroke and mortality. We urge NICE to retain equivalent recommendations for all DOACs and not remove patient options and choice. [23-33]

Catheter ablation is increasingly utilised to treat AF. Cryoballoon and radiofrequency are most commonly used, accounting for >500,000 ablations worldwide and >92% UK cases annually. Both appear equally effective in large, direct-comparison trials, with possibly fewer adverse events with cryoballoon. [34-36] Cryoballoon involves significantly shorter procedure times, fewer repeat procedures and seems more cost-effective. [37,38] The safety, ease of use, and wide applicability of cryoballoon manifests superiority to drugs, even as first-line therapy. [39,40] By not requiring specialised equipment, cryoablation provides deliverability in district hospitals. [41] It is astonishing that NICE is downgrading cryoballoon, and instead recommending laser-balloon, which accounts for <0.02% of UK ablations. [42] Changing practice wholesale to laser-balloon poses enormous challenges for implementation, including requiring physician-retraining and use of unfamiliar technology presenting significant safety implications. Meanwhile, European guidelines rightly observe, “the choice of energy source should depend on centre availability, operator preference/experience, and patient preference.” [1] Most ablations are performed under conscious sedation, yet NICE recommends general anaesthesia. This offers no major advantages, introduces resource challenges, impacts on turnaround times, and poses additional procedural risks and needless strain on sustainability of many UK centres. [43,44]

Patient-clinician partnerships empower patient choice and individualised care, and we welcome NICE’s recommendation of personalised packages of care and information. However, NICE provides no direction on how, or where, patients access appropriate information and help. We propose NICE signposts patients to organisations such as the AF Association, Arrhythmia Alliance, British Heart Foundation, Anticoagulation Europe and the Stroke Association.

We urge the NICE guideline committee to preserve patient choice, access, and shared decision making. We also urge them to revisit the available evidence and reappraise their decisions regarding diagnosis of AF, choice of bleeding risk score, continued availability of all four DOACs to permit tailoring of OAC based on individual patients’ risk profile and preferences, and maintaining support for cryoballoon catheter ablation, to ensure evidence-based clinical practice recommendations to deliver optimal patient care. The NICE guidelines miss the opportunity to support ongoing efforts for a structured management pathway to provide holistic, integrated care that has been proven in numerous studies to reduce adverse clinical outcomes.

Raj Mattu, consultant cardiologist, Kettering General Hospital, Kettering & University College London, London, United Kingdom.

Dhiraj Gupta, consultant cardiologist, Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart & Chest Hospital, Liverpool, United Kingdom.

Deirdre A Lane, reader in Cardiovascular Health, Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart & Chest Hospital, Liverpool, United Kingdom

Gregory Y H Lip, professor of Cardiovascular Medicine, Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart & Chest Hospital, Liverpool, United Kingdom.

Trudie LobbanPatient Representative, Founder & CEO – Arrhythmia Alliance & AF Association, United Kingdom

Muzahir H Tayebjee, consultant cardiologist, West Yorkshire Arrhythmia Service, Leeds General Infirmary, Leeds, United Kingdom

A John Camm, professor of clinical cardiology, Cardiac Clinical Academic Group, St. George’s University of London, London, United Kingdom.

Not commissioned, Peer reviewed

Competing interests: 

RM: Nil

DG: has received institutional research grants from Boston Scientific, Medtronic and Biosense Webster, and personal advisory fees from Boehringer Ingelheim, Boston Scientific and Abbott

DAL: has received an investigator-initiated educational grant from Bristol-Myers Squibb (BMS); has been a speaker for Boehringer Ingelheim and BMS/Pfizer; and has consulted for BMS, Boehringer Ingelheim, and Daiichi-Sankyo

GYHL: Consultant for Bayer/Janssen, BMS/Pfizer, Boehringer Ingelheim, Verseon and Daiichi-Sankyo.  Speaker for BMS/Pfizer, Boehringer Ingelheim, and Daiichi-Sankyo.  No fees are directly received personally.

TL: Nil

MHT: has received research and educational grants from Medtronic, Biosense Webster and Abbott.

AJC: has received institutional grants from Bayer, Boehringer Ingelheim, Daiichi Sankyo, Pfizer/BMS and personal advisory fees from Bayer, Boehringer Ingelheim, Daiichi Sankyo, Pfizer/BMS, Abbott, Boston Scientific and Medtronic


    1. Hindricks G et al, EurHeart J 2020;0:1-125
    2. Lip GYH et al, CHEST 2018;154(5):1121-1201 doi: 10.1016/j.chest.2018.07.040
    3. Bansal A et al, Journal of Arrhythmia. 2018;34:129–138) doi 10.1002/joa3.12035
    4. Andrew N et al, Neuroepidemiology:2013;40:227–239 DOI: 10.1159/000343667
    5. Ali A et al, J Atr Fibrillation:2016; 8:127 doi/10.1161/STROKEAHA.119.025554 9 
    6. Borre E et al, Thromb Haemos 2018 Dec;118(12):2171-2187 doi: 10.1055/s-0038-1675400
    7. Lip GYH, et al EurHeart J; 2015: 36, 3265–67
    8. Nice [TA275] 27.02.13
    9. Granger C et al, N Engl J Med 2011; 365:981-992 DOI: 10.1056/NEJMoa1107039
    10. N Engl J Med 2013; 369:2093-2104 DOI: 10.1056/NEJMoa1310907
    11. Hijazi Z et al, Circulation. 2014;129:961–970 
    12. Carnicelli AP et  al, Circulation. 2017;135:1273–1275 DOI: 10.1161/CIRCULATIONAHA.116.026714 
    13. Anderson S et al, Clinical Medicine Insights:Therapeutics 2018;10:1-6
    14. Alberts M et al, Stroke. 2020;51:549–555
    15. Yasaka M et al, PLoS ONE 2019;14(2):e0212354
    16. Yang L et al, Am J Cardiol 2020; 126:29− 36 doi: 10.1016/j.amjcard.2020.03.034
    17. Coleman C et al, Current Medical Research and Opinion 2016; 32:12 2047-2053
    18. Ruff CT et al: Lancet; 2014:383:955-62 DOI: 10.1016/S0140-6736(13)62343-0
    19. Mantha S et al, Thromb Haemost. 2012;108(3):476-484 DOI: 10.1160/TH12-02-0093
    20. Lip GYH et al, J Am Coll Cardiol. 2012;60(8):738-746 
    21. Camm AJ et al, Europace. 2018;20(1):1-11
    22. Garlo K et al, Clin J Am Soc Nephrol. 2019 Jan 7; 14(1): 125–136. doi: 10.2215/CJN.06430518
    23. Barber N et al, Annu Rev Nurs Res 2000; 18:48–90 doi: 10.1136/qshc.2003.005926
    24. Jackevicius CA et al, JAMA. 2002;288(4):462–467 DOI: 10.1001/jama.288.4.462
    25. Stuart B et al, Med Care Res Rev 2002; 59(2):123–145
    26. Song X, et al, [abstract]. Value Health. 2010; 13:A170 DOI: 10.1111/j.1524-4733.2010.00725.x
    27. Mancia G et al, J Hypertens, 2007: 25; 1105-1187 doi: 10.1097/HJH.0b013e3281fc975a
    28. Bakhai et al, BMC Cardiovascular Disorders 2013, 13:108
    29. Coleman C et al, Curr Med Res Opin. 2012;28(5):669-680 doi/abs/10.1185/03007995.2012.677419
    30. Caldeira D, et al, Rev Port Cardiol. 2014;33(7-8):431-437 DOI: 10.1016/j.repce.2014.01.014
    31. Srivastava K et al, Patient Preference and Adherence 2013:7 419–434
    32. Mentias A et al, Stroke. 2020; 51:2076–2086 doi: 10.1161/STROKEAHA.120.029541
    33. Lobban T et al, Arrhythmia Alliance Survey 2020; Data on File 
    34. Kuck KH et al, N Engl J Med. 2016;374:2235–2245 DOI: 10.1056/NEJMoa1602014 
    35. Andrade J et al, Circulation 2019;140:1779-1788 DOI: 10.1161/CIRCULATIONAHA.119.042622 
    36. J Interven Car Electrophysiol. 2019;54:237-245 
    37. Kuck KH et al, Eur Heart J 2016;37:2858-65
    38. Chun KRJ et al, J Am Heart Assoc. 2017;6:e006043 DOI: 10.1161/JAHA.117.006043 
    39. Wazni O et al, N Engl J Med 2020; DOI: 10.1056/NEJMoa2029554 
    40. Andrade J et al, N Engl J Med 2020; DOI: 10.1056/NEJMoa2029980
    41. Opel A et al, Europace 2019;21:440-444
    43. Bartoletti S et al, PACE 2019;42:1448-1455  doi: 10.1111/pace.13807 
    44. Creta et al, JCE 2020; doi: 10.1111/jce.14789