One third of patients diagnosed with heart failure demonstrate left ventricular reverse remodelling and recovery of cardiac function following a period of medical therapy. These patients have an excellent long-term prognosis. Whether they need to remain on long-term medical therapy is not clear. The investigators will investigate the safety of therapy withdrawal in patients with a previous diagnosis of heart failure and recovered cardiac function, in a randomised controlled trial.

Patients with a previous diagnosis of heart failure, who have recovered cardiac function, without symptoms, normal serum biomarker concentrations and who are currently still taking heart failure medications, will be recruited. Participants will be randomised to withdrawal of therapy and control arms and will undergo imaging investigations and cardiopulmonary exercise tests at baseline along with serum biomarker tests. Those participants randomised to withdrawal of therapy will have a gradual, supervised reduction of medications. If, at anytime, there are signs of recurring heart failure, medications will be restarted. Participants in the control arm will continue their medical therapy as usual.

Participants will be followed up with further biomarker testing, cardiopulmonary exercise testing and imaging investigations at 6 months.

The primary end-point will be a relapse in heart failure, defined by adverse remodelling, increase in left ventricular size, rise in serum biomarkers or clinical evidence of heart failure as judged by the clinical team.

Inclusion Criteria:

1. Have an index diagnosis of heart failure confirmed by 2 independent operators based on clinical details.
2. Be currently taking at least 1 of the following medications loop diuretic, betablocker, angiotension converting enzyme inhibitor/angiotensin receptor blocker and mineralocorticoid receptor antagonist.
3. Have demonstrated evidence of left ventricular reverse remodelling following the initial diagnosis with subsequent improvement in ejection fraction to >50% and normalisation of left ventricular (LV) volumes.
4. Have no symptoms of heart failure (NYHA Class 1).
5. Low plasma NTproBNP.

Exclusion Criteria:

1. Uncontrolled hypertension.
2. More than moderate valvular disease.
3. Estimated glomerular filtration rate <30mls/min.
4. Atrial/supraventricular/ventricular arrhythmia requiring beta-blockade.
5. Pregnancy.
6. Unstable angina.
7. Age <16 years