What is known
Palliative care may be helpful in addressing different needs throughout the heart failure trajectory. However, the transition from stable to unstable symptoms and worsening performance state may highlight the need for palliative care. As a patient with stable heart failure decompensates, symptoms may escalate dramatically. However, pharmacological management of acute decompensated heart failure is based largely on expert opinion, as most guidelines relate to chronic stable systolic heart failure.  Clinical decision tools such as the Seattle Heart Failure Model are available to help to estimate the benefit of adding medications or devices in the care of an individual patient. These tools are useful for individualising care decisions in advanced heart failure patients who have multiple comorbidities, and may help identify opportunities for palliative care input. [2-3]
When adjusting medications, dose-benefit balance is a major consideration. Tolerability may be affected by age and comorbidities, and the tendency is to reduce doses in older patients. Nonetheless, therapeutic benefit can still be associated with reduced doses.  Specific adverse effects, especially those related to elderly patients, should be monitored – including postural hypotension, renal dysfunction, electrolyte disturbances, and drug interactions.  Monitoring with ambulatory blood pressure is appropriate to help optimise therapy and may also provide prognostic information  and a six minute walk test is potentially useful as a maximal exercise test in advanced heart failure.  B-natriuretic protein (BNP) or NT-proBNP may also be of value for monitoring effectiveness of treatment in advanced heart failure. 
Opioids for palliation of dyspnoea have been well studied. They are a safe and effective approach for patients with both malignant and non-malignant causes of dyspnoea, although it has not been specifically studied in heart failure. [9-11]
A systematic review has identified evidence to support some specific non-pharmacological options for managing dyspnoea, although the studies have been predominantly done in a COPD population.  Options include breathing training and use of walking aids, (moderate evidence) neuro-electrical muscular stimulation and chest wall vibration (high level evidence). However there is not much evidence for acupuncture / acupressure, distractive auditory stimuli (music) relaxation, fan, counselling and support +/- breathing relaxation training, case management, or psychotherapy.  There is no evidence that relates to symptomatic dyspnoea management specifically in heart failure.  The role of oxygen in treating dyspnoea in non-hypoxic patients is also being studied, however further investigation is needed. 
The benefit of implantable cardiac defibrillators for primary prevention of mortality has not been not established for women with heart failure.  However cardiac resynchronisation therapy (biventricular pacing) has been shown to improve symptoms, exercise tolerance, and health related quality of life,  and to reduce all cause mortality, but not sudden cardiac death. 
Implications for practice
- Evidence based guidance for managing heart failure should be used to optimise treatment. However, in the palliative context treatment must also be individualised for each patient, in terms of the benefits and burdens of whatever is offered, the goals of care and wishes of the patient, and the person’s likely prognosis and comorbidities. As the end of life approaches, the approach to care should be to avoid those treatments which may detract from a peaceful death, and this includes clarifying the patient’s resuscitation status.
- Heart failure treatments used for primary disease management and life prolongation also have an impact on symptoms, and should generally be continued into the palliative phase for as long as they can be tolerated. Palliative care management strategies offer additional approaches to symptom management for many of the problems that occur commonly in advanced heart failure (eg the use of opioids for dyspnoea). Patients with difficult symptoms (physical or psychosocial) can benefit significantly from referral for palliative care.
- Potential treatment options such as non-invasive ventilation and left ventricular assist devices may be part of the spectrum of palliation for some patients, and shared protocols for managing such treatments and negotiating goals of care are needed. The use of implantable cardioverter defibrillators raises the issue of inactivation to prevent “ICD storm” in a dying patient. Again, shared protocols between heart failure and palliative care services are needed.
- Sharing of care between heart failure specialists and palliative care specialists will result in a wider range of treatment options for patients with advanced heart failure, and a broadening of the skills on both sides, with significant improvements in care for patients as a result. The well established model of collaboration between oncology and palliative care specialists in caring for patients with advanced cancer provides an example of how such collaboration can work in practice.