Article Text

Implementation of the NICE guidelines for the primary prevention of mortality from ventricular tachyarrhythmias: implications for UK electrophysiology centres; activity modelling from the UK-HEART study
  1. N P GALL,
  3. A ZAMAN*,
  4. S O'NUNAIN,
  5. K A A FOX,
  6. A FLAPAN,
  7. J NOLAN
  1. King's College Hospital, London
  2. *Freeman Hospital, Newcastle-upon-Tyne
  3. Royal Infirmary, Edinburgh
  4. North Staffordshire Hospital
  5. UK
  1. NP Gall, Department of Cardiology, King's College Hospital, Denmark Hill, London SE5 9RS, UK;nick.gall{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

It is now well established that the implantable cardioverter-defibrillator (ICD) is the most effective treatment for the primary prevention of life threatening ventricular arrhythmia.1 ,2 Despite this, its widespread use in the UK for this indication has been minimal, at least in part because of perceived resource implications. In September 2000, the National Institute for Clinical Excellence (NICE) recommended electrophysiological testing (EPS) for all patients with a history of myocardial infarction who have an ejection fraction of less than 35% and three or more beats of non-sustained ventricular tachycardia (NSVT) on a 24 hour Holter monitor. NICE further recommended that all patients in whom a significant ventricular arrhythmia is induced at EPS should have an ICD implanted. The potential effect of the NICE guidelines on workload in UK electrophysiology centres has not been evaluated.

UK-HEART examined the utility of heart rate variability as a predictor of mortality in heart failure patients.3 Ambulant patients with stable heart failure were enrolled in four UK cardiac centres over a 17 month period (1 December 1993 to 30 April 1995). Patients with diabetes and atrial fibrillation were excluded; otherwise allcomers to cardiology clinics were eligible. All patients underwent echocardiographic estimation of ejection fraction and screening for arrhythmia, including NSVT, using 24 hour ambulatory electrocardiography. The baseline demographic and clinical characteristics of these patients were similar to those in MADIT1 and MUSTT,2 the two primary prevention trials on which the NICE guidelines are based.

A total of 555 patients were entered into UK-HEART, of whom 551 had interpretable 24 hour ECGs. Of these, 57 fulfilled the NICE criteria for EPS. If diabetics and patients with atrial fibrillation had been included this figure would increase to 88 as approximately 35% of heart failure populations suffer from either of these conditions.4 Recent data5 have demonstrated that a second 24 hour tape will identify a further 38% of patients with NSVT, increasing numbers to 142. In MUSTT, sustained ventricular arrhythmia was induced at EPS in 35% of patients. Other smaller studies in similar populations have found equivalent positivity rates.6 If our cohort reacted in the same way there would be 50 positive studies and hence 50 ICD implants.

During the period of recruitment 68 electrophysiological studies for ventricular arrhythmia were performed on patients within the four centres. During an equivalent, contemporary period (1998-2000), 72 studies were performed. Application of NICE guidelines would lead to a further 142 studies, equivalent to three extra per month in each of the three centres performing this procedure. During 1993-95 there were 11 ICDs implanted. Between 1998-2000 there were 79 implanted. The NICE guidelines would mean a further increase of 50, representing one extra ICD per implanting centre per month.

The UK-HEART study was primarily designed to investigate the prognostic utility of measures of heart rate variability in a heart failure population and therefore has some limitations in its application to NSVT. Diabetics and patients with atrial fibrillation were excluded from UK-HEART because of the heart rate variability analysis. Diabetics make up 20% and patients with atrial fibrillation 15% of major heart failure studies. We have assumed that if these patients had been included, our population eligible for further investigation would increase by 35%. Clearly there will be some crossover between the groups and they may have different rates of NSVT. The patients in the study were allcomers to cardiology clinics. Numbers would increase if all primary and secondary care physicians had enrolled patients.

The numbers of investigations and ICDs estimated apply to the first year of guideline implementation. The numbers requiring investigation and treatment the following year, including new diagnoses and those whose disease had progressed, would differ; theoretically fewer than in the first year. The numbers would also fall as some patients eligible for investigation or treatment would be unsuitable for non-cardiological reasons.

Trial acronyms

UK-HEART United Kingdom Heart Failure Evaluation and Assessment of Risk Trial
MADIT Multicenter Automatic Defibrillator Implantation Trial
MUSST Multicenter Unsustained Tachycardia Trial

Overall, these data provide evidence that implementation of the NICE guidelines for the use of ICDs in primary prevention is unlikely to lead to an unmanageable increase in EPS or ICD implantation in UK electrophysiology centres.