56 Monitoring

Heart failure (HF) is a chronic condition with the risk of episodic deterioration (‘decompensation’). Aims of therapy include stabilization of the syndrome, improvement in the prognosis and quality of life, and the avoidance of hospitalization. Early detection of deterioration is a key component of HF management programmes, with the goal of adjusting therapy to restabilize the syndrome rapidly and to avoid the need for emergency hospitalization.

HF management programmes were developed to help ensure patients received appropriate multiprofessional input to their treatment, education, and monitoring. They were designed to help fill the gap that often existed between discharge from hospital and the traditional outpatient clinic review. Many patients decompensated during this high-risk period, resulting in readmission to hospital. However, despite such programmes becoming more widespread, perhaps as many as 30% of patients are still readmitted within 12 months of hospital discharge,¹ with some of the admissions preventable with better monitoring.

This chapter discusses the key elements of monitoring of patients with HF, including what and who should be monitored, the frequency of monitoring, and the different methods of monitoring.

Professional guidelines suggest that all patients recently hospitalized with HF (and other ‘high-risk’ patients) should enter a HF management programme.^2,3 Only around 20% of such patients in England do so at the present time.⁴ Many of the key components of such a programme relate to monitoring (Box 56.1). Meta-analysis suggests that the various types of disease management programmes are similarly effective in terms of reducing mortality and rehospitalizations.⁵

Monitoring can take many forms: supported self-monitoring, clinic attendance, home visits, or remote monitoring. Most modern programmes combine several different approaches, tailoring the care to the needs of the patient and family, and taking account of the local resources and expertise available in both primary and secondary care. Such a tailored approach, matching the model of care to the severity of the condition, is in keeping with current health care policy in the developed world: most attention is focused on those with more complex needs (‘case management’), built on a platform of good disease management in general and with increasingly expert patients who are supported to self-care (Fig. 56.1).⁶

Elderly patients, or those with multiple co-morbidities, will have more complex healthcare needs that place them at higher risk of hospitalization. They benefit from more intensive case management.

A recent randomized study from the Netherlands confirms the value of a more tailored approach, reporting that more intensive management for all patients is not necessarily better than a less intensive approach.⁷

Historically, patients with HF were not taught to self-monitor but were reviewed periodically by a doctor working in either primary or secondary care. Even in the high-risk period after hospital discharge it was traditional to arrange hospital clinic review some weeks after discharge. At that visit the doctor would assess the patient and determine if changes to treatment were required. In the United Kingdom, with very few physicians with a special interest in the condition, many patients would be discharged back to primary care review alone. Such patients might perhaps be reviewed in other clinics, due to comorbidities such as diabetes, where the HF syndrome might or might not be reassessed. It is little surprise that this model of monitoring was associated with poor outcomes, including a high rate of emergency rehospitalizations. Such a model is now considered outdated and substandard.

There is much that the patient and their family can do to monitor how well the HF syndrome is controlled. Professional monitoring is likely to be supplemental to this self-monitoring. Self-monitoring should facilitate self-management, where a patient adjusts their therapy depending on the control of the HF syndrome. Typically this would involve a patient adjusting the dose of diuretic depending on changes in their weight.

Where they wish, patients and their families should be helped to become involved in monitoring their HF. They will need advice on how to monitor symptoms of HF easily and on the significance of any changes in symptoms, weight, or other measurements such as blood pressure or pulse. Simple management strategies and information on when and where to seek professional help can then help the patient self-manage.

Patients experience a variety of symptoms, which complicates their ability to recognize the importance of symptoms and to identify their cause as HF-related.^8,–10 Older age, depression, and cognitive dysfunction may decrease self-care ability.

To be effective, self-monitoring requires the local HF service to be easily accessible to the patient and their family/carer. Although most HF services provide a telephone advice line (at least during working hours Monday through Friday), a proportion of patients will be reluctant to use this for fear of interrupting the professional and uncertainty regarding the significance of their symptoms. Telemonitoring can be useful in this situation and patients may develop expertise through the timely feedback provided by the monitored data and from the health professional contact triggered by abnormal results.

Heart failure typically affects elderly people, and their limited mobility and lack of social support may make hospital clinic attendance problematic. Home visits can bridge the gap, but are costly in terms of travel time for the health professional, limiting the case load that a specialist nurse can take on. Telehealth (‘healthcare at a distance’) has the potential to widen access to high-quality care and to provide this care closer to home than the traditional, usually hospital-based, model can. Telemonitoring—the remote monitoring of patients using information technology—has developed rapidly in the past decade.

There are several types of remote monitoring, from simple to complex:

The UK National Institute for Health and Clinical Excellence (NICE) has made recommendations for monitoring of HF (Box 56.2).² These recommendations are based on consensus on what constitutes good clinical practice, rather than on specific randomized controlled trials. The ideal variable to monitor would be simple and convenient to measure, reproducible, sensitive to changes in the control of HF, specific for this condition, and would change rapidly enough to give an early warning of decompensation. No such variable has been identified. In practice, several measurements are relied on to detect decompensation. The patient’s history, signs, and symptoms, supplemented where possible with physiological data, are combined with skilled clinical interpretation.

The typical signs and symptoms of clinical deterioration of chronic

HF are increasing fluid retention, breathlessness, and effort intolerance, but can include less specific symptoms such as fatigue, cough, and poorer cognition. In many, but not all, patients and episodes, deterioration is gradual and it should be possible to detect such deterioration if the patient or HF team is monitoring the clinical condition. Many patients are aware of symptoms for several weeks but only seek professional help when they became intolerable.¹¹

In some patients, and in some episodes, the deterioration may be abrupt and thus could not be detected earlier by closer monitoring. Such deterioration might occur as a result of sudden changes in cardiac rhythm (such as the onset of atrial fibrillation) or incidental infection (particularly chest or urinary tract infection).

Some telemonitoring systems ask patients questions about a range of symptoms which can be useful in identifying deterioration (particularly if combined with daily weight monitoring). From our experience in the Home-HF study,¹² patients often find it difficult to say if any specific symptom is better or worse than the day before. Changes are often subtle and noticed at different time points during the day, making it difficult to decide if they relate to a change in the overall condition or to differing activities.

The New York Heart Association (NYHA) classification of HF¹³ provides some structure to symptom monitoring by health care professionals by grading the severity of HF according to a broad classification of functional limitation. It is relatively easy to score but its value is limited by poor sensitivity to small change and poor reproducibility among clinicians.¹⁴ There is also limited agreement between patients and their clinicians.¹⁵

Patients tend to measure change in their symptoms by the effect on their activities of daily living rather than more abstract concepts: they may notice a change in their ability to go out shopping, tidy the garden, or walk to their friend’s house, and so behavioural questioning may be more effective at monitoring change in symptoms.

The six-minute walk test is a more reliable tool to monitor changes in functional capacity,¹⁶ but is rarely done in routine practice. However, it is of less value where functional capacity is limited by comorbidity, such as osteoarthritis.

An early sign of worsening HF in many patients is increasing fluid retention, and in theory daily monitoring of weight using accurate scales should be useful in identifying decompensation early. Such an approach is standard practice in HF management programmes, and is recommended in professional guidelines.^2,3,17

There is some evidence that the speed of weight gain is more sensitive and specific for HF decompensation than absolute weight change, with an increase of more than 2 kg over a period of 72 h being clinically significant.³ For more effective monitoring, the weight increase should be the weight above ‘dry’ weight; that is, the steady weight achieved following any change in diuretic therapy and where there are no signs of fluid overload. To allow for normal weight change, ‘dry’ weight should be recalculated periodically (every 1–2 months). This is particularly important after discharge from hospital, when patients generally feel better and start to eat more and therefore may put on muscle or fat rather than merely fluid weight.

The significance of an increase in body weight of 2 kg depends upon patient size, and such a weight increase in a patient weighing only 40 kg is likely to be more significant than in someone of 120 kg. Attempts have been made to correct for patient size, with different patients being given different guidelines for identifying possible increasing fluid retention.

Despite the widespread use of weight monitoring, its accuracy is limited. In an attempt to establish the sensitivity and specificity of weight gain in correctly identifying clinical deterioration, Lewin and colleagues¹⁸ recruited patients with established HF, predominately (70%) left ventricular systolic dysfunction (LVSD), and compared the patients’ daily weight diary against clinical examination. They concluded that a weight gain of more than 2 kg over 48–72 h demonstrated good specificity (97%) but poor sensitivity (9%) for predicting clinical deterioration. A weight increase of more than 2% above dry weight had a similar specificity (94%) with only marginal improvement in sensitivity (17%). Although such weight increase is highly specific for decompensation, these results suggest that the lack of such a change cannot be taken to exclude decompensation. Attention must therefore also be directed at assessment of symptoms or other physiological measurements that may reflect overall HF status. Patients who are able to self-care should also be aware that while weight monitoring is useful, they should seek help if they notice an increase in symptoms regardless of any weight change.

Data from a home telemonitoring study in patients with left ventricular ejection fraction (LVEF) less than 40% also showed that relying only on weight monitoring is unlikely to be of great value in detecting decompensation, even when using more complex calculations of weight change. Weight gain started around 14 days prior to a hospital admission with worsening HF, but substantial weight gain was seen in only around 20% of patients.¹⁹ Continuous weight monitoring demonstrated that weight gain starts much earlier than expected, may be considerably less than that suggested in guidelines, and that patients may present with worsening HF without any weight increase.

It may be that the cause of decompensation influences the number of days over which weight gain occurs. For example, weight gain resulting from worsening left ventricular function or nonadherence to angiotensin converting enzyme (ACE) inhibitors or β-blockers may be more insidious, whereas infection, arrhythmias, or noncompliance with diuretic medication may cause a more rapid increase in volume and may even precipitate HF decompensation before any change in weight is noticed.

For weight monitoring to be reliable it requires consistency in its monitoring; patients should be encouraged to weigh themselves at the same time each morning, wearing similar clothing. Although these suggestions appear straightforward, patients are likely to monitor their weight at a time they find convenient and in clothing that may differ in weight from one day to the next. This does not necessarily imply inadequate knowledge or understanding of weight monitoring, but reflects the everyday reality of living with HF and the attempt to integrate self-monitoring into their daily lives. In some cases, although monitoring is done very carefully, instructions on when to inform the HF team about changes are not followed, often because the patient or family do not feel the weight change is important or they do not wish to ‘bother’ the professionals.

Another limitation of relying on weight monitoring alone to determine fluid status is that patients with advanced HF may develop cachexia and a stable weight over time may mask volume overload combined with reducing muscle and fat mass.

Despite the above limitations, daily weight measurement is easy to monitor and is considered the cornerstone of daily monitoring in most HF management programmes. As argued above, it is best combined with monitoring of other variables, including symptoms.

In many patients, increasing fluid retention may be most marked in the legs. Monitoring of ankle swelling is a routine component of clinical monitoring for HF deterioration. However, perhaps as much as 5 L of excess fluid is present before oedema is noticed by most patients. Patients frequently find it difficult to identify when their ankles are more swollen than usual and may only notice change when it is so severe that they no longer can wear their normal shoes.

In the patient with venous insufficiency, it may be difficult to distinguish dependent oedema resulting from HF from the normal state. Often the oedema is asymmetrical, particularly if the patient has had venous harvesting for bypass surgery, and patients often concentrate on the better leg, denying that fluid is building up.

In some patients, particularly those with failure of the right ventricle or tricuspid regurgitation, accumulation of fluid may be more marked in the abdomen than in the legs. Patients may notice an increase in girth, with their clothes or belt becoming tighter, or increased pulsation in their neck as the venous pressure increases. In our experience, it is rare for patients to identify decompensation early through these signs.

Detection of significant hypotension (including postural hypotension) is important in the clinical review of a patient. It may point towards intravascular fluid depletion due to excessively high diuretic usage, or a low cardiac output, or sepsis, or inappropriately high dosage of neurohormonal antagonists. For some patients, the control of hypertension is also critical to the control of the HF syndrome. Many patients monitor their own blood pressure at home, and it can be easily monitored remotely.

Arrhythmia is common in patients with HF. Ventricular arrhythmia may be life-threatening and atrial arrhythmia (such as atrial fibrillation) may cause symptom deterioration. Both complicate management decisions. Heart rate monitoring (supplemented by rhythm monitoring if possible and appropriate) may help identify problems that requires further investigation.

Single-lead ECG monitoring has been added to external telemonitoring equipment. However, it increases the complexity of monitoring and to date there is no evidence of additional benefit.^20,21 Heart rate monitoring alone is easier and may be sufficient in many cases. Where devices such as cardiac resynchronization therapy (CRT) or a dual-chamber implantable cardioverter-defibrillator (ICD) are implanted, device interrogation can provide potentially continuous information (but more often daily or less frequently) on both heart rate and rhythm.

Renal function is often impaired in patients with HF, and deterioration is associated with a poor prognosis. Intercurrent illness (such as infection) can have a profound effect on renal function, as can the introduction and up-titration of therapies such as ACE inhibitors, angiotensin receptor blockers, β-blockers or aldosterone antagonists. Regular monitoring of serum biochemistry (in particular serum potassium, urea, and creatinine) is important and structured follow-up can ensure this is not missed. Biochemistry should be checked within 2 weeks of the introduction or up-titration of new medication and otherwise at a minimum of 6-monthly intervals. Intercurrent illness should trigger a check of HF status, including renal function. The introduction of drugs such as nonsteroidal anti-inflammatory drugs (NSAIDs), either prescribed or obtained over the counter, can markedly impair renal function in patients with HF.

Other blood tests may be required as part of routine monitoring: anaemia is increasingly common as HF advances, as is diabetes, so a check on the full blood count and plasma glucose should be made periodically. Thyroid and liver function may be deranged by amiodarone. Urate is often high in patients with HF on diuretic therapy, and repeated attacks of acute gout may occur if allopurinol is not used.

Several randomized trials have assessed whether the serial measurement of plasma B-type natriuretic peptide (BNP) or NT-proBNP might aid the management of patients with HF. Early studies suggested the clinical outcome was better if a target BNP or NT-proBNP was used to titrate diuretics and ACE inhibitors, but a more recent and larger randomized trial found little evidence of benefit, except in a post-hoc subgroup of those aged less than 75 years.²² The 2010 partial update to the NICE guidance on the management of HF suggests that serial measurement of plasma BNP should be considered for certain patients being followed up in specialist practice, particularly if they have been recently hospitalized or there have been problems in up-titrating their drug therapy.²³

The degree to which plasma natriuretic peptides fall during a hospitalization for HF is a good indicator of the subsequent risk of readmission, and might be used to target closer follow-up for some patients when the plasma level fails to fall during admission.²⁴ Near-patient testing of plasma BNP is advancing rapidly, and it is likely that a home-based test will be available in the near future. Near-patient testing of electrolytes and creatinine is unlikely to become available in the near future.

Monitoring adherence to medication is problematic but it is estimated that between 20% and 50% of patients mismanage their medication.^25,26 Patients with HF are likely to have multiple comorbidities, cognitive decline, social isolation, and depression, all of which adversely affect their ability to adhere to medication regimes.

Patients may choose not to comply with medication from concern about side effects, or they may simply forget; they may be unable to refill the drug prescription, or be unsure how to take the medication or of its value. In some countries, there may be economic constraints. Nonadherence with medication has been implicated in possibly as many as 30% of hospital admissions with HF.²⁷ Electronic monitoring of medication adherence is possible and may provide insights into particular issues. Some telemonitoring systems also provide reminders to patients regarding their medication.

Weight loss is a poor prognostic factor in HF. There is a variety of contributors to weight loss including inadequate dietary intake, loss of appetite, and malabsorption of nutrients.

The simplest method of monitoring nutritional status is through recording the body weight and the European Society of Cardiology (ESC) guidance states that a decrease of more than 6% of total body weight within 6 months is suggestive of cachexia.³ Monitoring of appetite and interest in eating can easily become part of regular monitoring and may identify potential malnutrition.

As many as 40% of patients with HF may become moderately depressed or anxious.^28,29 Depression is an independent risk factor for mortality,³⁰ increases symptom reporting,²⁹ decreases quality of life, and (alongside anxiety) can decrease self-care ability.

Cognitive decline through age, depression, or chronic illness is also a potential influence on patient outcome and is likely to make self-care more difficult. There are scoring tools to quantify decline in cognitive function, but they may have limited value in elderly people or in routine clinical practice.

Some of the newer models of implantable CRT devices include features with the potential to monitor and identify preclinical signs of worsening HF remotely. For example, intrathoracic impedance can be measured between the right ventricular lead and the generator of a pacing device. Accumulating intrathoracic fluid decreases the impedance. Evidence for the value of intrathoracic impedance monitoring in detection of worsening HF is emerging, although it is unlikely to be sensitive enough on its own to aid decision-making.^31,–33 False-positive results may trigger an increase in health care utilization.

Other devices include sensors capable of monitoring circulatory pressures (including right ventricular, pulmonary artery and left atrial pressures), heart rate variability,^34,35 and mean daily physical activity with movement detectors.³⁵ The feasibility of using trends in these variables to assist early intervention and prevent overt clinical deterioration has been established, but results of larger studies are awaited to determine their exact role.³⁶ Currently the strongest evidence exists for monitoring pulmonary artery pressure, which is discussed further below.

In the future, it is likely that data from devices will be combined with traditional indicators such as vital signs, weight change, and symptoms to indicate clinical status and detect worsening HF more reliably. Patients with advanced HF are most likely to benefit.

The DIAL trial³⁷ enrolled 1518 stable outpatients in Argentina, and used a centralized call centre together with protocol-driven phone calls focusing on symptom monitoring, adjustment of medication, and patient education. All phone calls were nurse-initiated: initially at 2-weekly intervals, after which the frequency was based on the clinical condition. The study population was relatively young with a mean age of 65 years and 80% had LVSD. There was a statistically significant 20% reduction in the number of patients who died or were admitted to hospital with HF, largely attributable to the reduction in the number of patients admitted with worsening HF. Mortality was unaffected. Importantly, patients were generally on evidence-based medication at randomization but more patients in the control arm stopped taking their medication or reported nonadherence with other lifestyle advice, demonstrating the benefit of regular monitoring on adherence with treatment of proven benefit.

Other studies of telephone monitoring in different health care settings provide further evidence of benefit and when combined in a meta-analysis indicate a significant reduction in HF-related hospital admission, a trend towards a reduction in mortality, but no effect on all-cause admission.³⁸ The limitations of such models are that the phone calls are primarily initiated by the professional at preset times (usually protocol-driven) and they are thus unable to detect more rapid changes in the condition.

Telemonitoring uses patient-initiated remote electronic monitoring and involves the transmission of information from the patient (usually in their home) to the health professional (Fig. 56.2). In most cases, data are transmitted using a domestic telephone line; manufacturers are developing equipment with wireless transmission. Telemonitoring enables frequent (usually daily) assessment of clinical variables by a health professional, early identification of clinical deterioration, and early patient management (either remotely or through early recall to clinic for review). There is no consensus as to which variables provide the most useful data. There is a balance between accurate patient assessment and overburdening the patient.

The Trans-European Network-Home-Care Management System (TEN-HMS) study randomized 426 patients (mean age 67 years) with LVSD (LVEF 〈40%) in NYHA class II–IV to home telemonitoring, nurse telephone support, or usual care. This multicentre study recruited patients from 16 centres and 3 European countries. There was no statistically significant effect on HF-related hospitalization or average length of stay for such admissions. Although it was not a primary outcome, there was a statistically significant reduction in mortality in the telemonitoring group compared to usual care. A greater number of patients in the telemonitored group was ultimately on optimal drug therapy than in the usual care group.²¹

We have reported the results of a randomized trial of home telemonitoring (Home-HF)in a typical elderly HF population (mean age 71, 45% 〉75 years), which included those with either preserved or impaired left ventricular systolic function. We compared 6 months of daily telemonitoring with specialist HF care in 182 patients discharged from three district hospitals in west London, United Kingdom. We found no difference in the primary outcome of all-cause hospitalization, but did find a significant decrease in the proportion of hospitalizations for HF that were as an emergency (usual care 81% vs telemonitoring 36%, p = 0.01). We also found a reduction in the number of emergency department attendances and secondary care clinic visits with telemonitoring.¹²

A recent meta-analysis including around 800 patients suggested that telemonitoring was associated with a 38% reduction in risk of all-cause mortality with no effect on all-cause hospitalization.³⁹ The reduction in mortality may not be seen in a more elderly population, but the technology can enable specialist services to care for more patients, with an improvement in patient experience. Home-based technology is generally easy to use, and studies consistently report very high levels of usage (〉80%). Systems are likely to be adopted widely in European health care practice, and will lead to a major change in the way HF clinicians interact with patients.⁴⁰

Rapid advances in health care technology have led to the possibility of remotely monitoring a large number of physiological variables using implanted therapeutic devices, such as CRT or ICD. Using either a bedside monitoring device or a portable ‘interrogator’ placed over the device, the patient can be remotely monitored, with the data being transferred to a central station. If patients feel unwell or have concerns about their heart or device function, they can trigger a download from the device to the health professional. Using a web-based system, the professional can monitor device performance and a number of physiological variables.

Such remote monitoring reduces rather than eliminates all face-to-face contact, and an annual face-to-face clinic visit is recommended for clinical examination.⁴¹ A recent audit of 97 patients at our centre demonstrated a 70% reduction in face-to-face pacing clinic visits with the introduction of remote monitoring. In 84% of the remote scheduled pacing checks, no further action was required.⁴² There are early indicators that remote monitoring of implantable devices considerably increases the number of patients who can be safely cared for by one practitioner and extends the reach of effective monitoring programmes. Good communication between the electrophysiology department, pacing technologist, and HF team is essential.

How often data should be examined is unclear, as is the value of individual physiological variables monitored (discussed earlier in chapter). Large clinical trials are under way to test the clinical value of such monitoring in terms of patient outcome and health care expenditure.³⁶

Currently, the strongest evidence exists for monitoring pulmonary artery pressure. A randomized controlled study of the addition of data from an implanted haemodynamic monitor to usual clinical review reported a statistically nonsignificant 21% reduction in HF hospitalizations or urgent care visits requiring intravenous therapy in the 134 patients with advanced HF in the active arm, compared with 140 randomized to the control arm.⁴³ An algorithm indirectly estimated pulmonary artery pressure from measured right ventricular pressure in this device. Directly measured pulmonary artery pressure appears to be more useful, with recent data (as yet unpublished) from the CHAMPION study reporting a 30% reduction in HF hospitalization over 6 months in patients with an implantable pulmonary artery pressure monitoring device in NYHA class III.

It is often not possible, or indeed necessary, to follow all patients with the same level of intensity. The period of greatest vulnerability remains the period immediately following hospital discharge. Only a minority of patients are readmitted, but it is best to schedule a review soon after hospital discharge and preferably within 10 days. A major difficulty in establishing the frequency of subsequent follow-up is the substantial difference in the intensity of interventions in the studies demonstrating benefit from management programmes, with follow-up varying from a single home visit⁴⁴ to weekly home visits for older or unstable patients.⁴⁵ The positive outcomes of the early studies were largely derived from trials in patients with LVSD where ‘usual’ care was very poor. The generalizability to current patients, particularly where ‘usual’ care has improved, is unclear.

The COACH study (Coordinating study evaluating Outcome of Advising and Counselling), a multicentre study of over 1000 patients in the Netherlands, was designed to establish the optimal frequency of professional monitoring in a more general HF population.⁷ Patients were randomized to one of three groups: (1) outpatient cardiologist clinic visit within 2 months of discharge and 6-monthly thereafter; (2) the addition of nine outpatient nurse visits; or (3) a total of four visits to the cardiologist, and twenty contacts with the nurse (with the greatest intensity within the first month of hospital discharge), and multidisciplinary team contact for lifestyle advice. Patients were followed up for a period of 18 months. The study reported no difference in the risk of death or HF-related hospital admission between the groups. There was also no difference in the total number of days lost to death or all-cause hospitalization. Importantly, there was a substantial increase in contact with the secondary care HF team in all groups and this was frequently triggered by an increase in symptoms. The study has been interpreted as providing evidence that HF programmes need to be adaptive to patient need, pointing to the need for an open access follow-up service, with varying levels of support.

A 6-monthly clinical review, including a check of symptom control, cardiac rhythm, medication, and blood chemistry may be sufficient when the patient’s condition is stable. An annual review by a cardiologist may also be useful. Patients with advanced HF, those undergoing therapy changes, and those with a history of frequent decompensation will benefit from more frequent monitoring. Patients who have multiple comorbidities or renal failure, are elderly, or are on multiple medications may also benefit from more frequent monitoring, but this must be done carefully to prevent confusion from too many visits and too many changes to medication.

Patients with implanted devices such as CRT, ICD, or combined CRT-D will require technical checks on the devices.⁴¹ Patients with long-term ventricular assist devices will have follow-up organized by the implanting centre, as will patients after heart transplantation.

Current UK government policy promotes health care that is easily accessible and adaptive to patient need, with a focus on the delivery of care in the community.⁶ A secondary care specialist is usually involved with the initial introduction of medication, but optimization can be safely carried out in primary care. This process is best guided by a management plan, shared between primary and secondary care. Once the patient is on optimized medication, the GP and community HF nurse can continue to monitor the majority of patients with only an annual review by a consultant cardiologist (either in secondary care or a satellite community clinic). Patients with more complex needs, such as elderly people or those with comorbidities, may benefit from care delivery by a case manager or care coordinator (often a community matron). Based in primary care, the care coordinator is well placed to liaise regularly with the multidisciplinary community health care team and social care services to ensure the right services are offered at the right time.

Cardiologists and specialist nurses have played a central role in HF management in most studies and are key members of the HF team in clinical practice. Not all patients need to be seen at all time points by a secondary care specialist. Where primary care services are well-developed, the GP and community nurse should be involved in the routine monitoring of the stable patient. Monitoring must be incorporated into the activities of the multidisciplinary team,⁴⁶ with timely (and accurate) communication between primary, secondary, and tertiary care being vital. Where monitoring also involves technical checks on implanted devices, communication must also be good between those responsible for the technology and the HF multidisciplinary team.

Education and training underpin effective monitoring. Subspecialist HF training is now recognized in the United Kingdom but there are differences in medical specialist training across Europe: HF is not recognized as a subspecialty in all countries. The European Heart Failure Curriculum for Nurses has attempted to provide some standardization, but HF education is usually local and country specific. In the United Kingdom, nurses and some allied professionals (such as pharmacists) obtain further qualifications to expand their role to include the prescription of medications and so manage their own patient caseload.⁴⁷ Training in remote monitoring is essential, as it requires decision-making in response to different cues than is usual in traditional face-to-face clinical contacts. There is the risk that more data leads to more frequent, but not necessarily effective, changes in the management plan and increased health care utilization. Remote monitoring cannot replace more traditional care, but should be integrated into a multiprofessional disease management programme, enabling more patients to access high-quality care.

Monitoring of patients with HF happens in a variety of settings: primary, secondary, and tertiary care. The level of complexity varies from patient to patient, and varies during the course of an individual’s illness. Final decisions will be needed on when to stop monitoring for the active control of HF and move towards monitoring for symptom relief and end-of-life care (see Chapter 55). Organizations need to develop systems that are adaptive and individualized; both adding and removing components as needs change.

The management of HF is complex and patient expectations for care are increasing. A variety of new monitoring systems is becoming available. Many enable remote monitoring, which increases patient convenience and improves the experience of care. The systems also increase the amount of data that can be monitored and may complicate decision-making. The evidence base for telemonitoring is increasing, but better guidance is needed to help identify patients most likely to benefit from intensive monitoring and to manage the increased data effectively. Although the health care professional is important, the patient and their family are central to the success of monitoring.