I. Remote Patient Monitoring (Telemonitoring) for Heart Failure: What every physician needs to know.
Heart failure is common and its prevalence is increasing. It often causes troublesome symptoms, which may be persistent or recurrent, limit exercise capacity, and impair well-being and quality of life.
Patients who have severe or worsening symptoms are at high risk of hospitalization and death. Treatment with drugs and devices is complex but can stabilize many patients who subsequently have a much reduced rate of hospitalization or death. Ensuring that patients consistently receive the best available treatment at the right dose and that treatment is modified to their changing individual needs is challenging.
Remote patient monitoring or Home Tele-monitoring (HTM) provides a technology that can enable better patient management by (Figure 3):
providing a system that helps ensure that the patient receives optimal treatment tailored to their individual needs
providing the patient with access to education and information that helps them understand their condition and treatment, thereby allowing them to play a greater role in their own care.
monitoring the patient to detect changes in their clinical profile that may require a change in treatment to prevent hospitalization or, alternatively, a preemptive admission to the hospital that may be life-saving.
permitting earlier discharge from the hospital into the HTM community.
processing complex information and filtering it to prevent data overload, while still alerting either patient or clinician to important information.
There are many different types of technology for remote monitoring (Figure 4):
Telephone support by a health professional to check on symptoms and treatment
Voice-interactive systems with backup from a health professional. Information on weight or blood pressure can be entered using a key pad.
Automatic transmission of the responses to symptom questionnaires and of physiologic data (most commonly weight, blood pressure, and heart rate and rhythm) using conventional but networked devices (either at home or through mobile phones) (Figure 1 and Figure 2)
Automatic transmission of data acquired from wearable monitoring technology (and potentially tattoos); typically heart and respiratory rate and temperature
Automatic transmission of data acquired from implanted therapeutic devices, such as pacemakers and defibrillators
Automatic transmission of data acquired from implanted devices with a monitoring-only function, such as pulmonary artery pressure monitoring “chips”
The optimal technology is unknown but voice-interactive systems (low patient compliance) may be inferior. Telephone support appears to reduce hospitalization but may be expensive due to the high cost of staff and has little impact on mortality. HTM using implanted technologies is appealing but these technologies cannot be considered mature.
HTM using symptoms and automated transmission of physiologic data combined with decision support systems that facilitate treatment optimization and avoid information overload have achieved the greatest success so far. Systems that provide the patient with feedback (for example, graphs of weight, blood pressure, and heart rate) are likely to achieve greater patient compliance and engagement, especially if the system gives them insights into why it is important and what they can do about it.
The optimal duration of monitoring is unknown. Different functions of HTM may require different durations. For instance, most of the benefit from education and optimization of guideline-indicated therapy may be derived within 3 to 4 months, whereas monitoring for changes in health status may require lifelong monitoring.
The optimal duration of monitoring may vary from one patient to the next. Stable patients that are already receiving good treatment and are supported by effective conventional monitoring that can detect complications and progression of disease may not benefit from HTM.
HTM is unlikely to work unless supported by a relevant multidisciplinary team/service that is capable of responding appropriately to the information acquired in a timely fashion.
The most important things to remember when considering HTM are:
Is the patient at substantial risk of events? HTM may help low-risk patients by ensuring education and optimization of therapy but is unlikely to reduce hospitalization or mortality in the short or medium term. If the goal is to improve the short- or medium-term outcome, then patients at high risk of events must be enrolled.
If the conventional health care system is already excellent, then HTM probably cannot improve outcome, although it might maintain the quality of care at lower cost. Although the cost of technology can be substantial, the greatest cost in most health care systems is for staff.
Ensuring that staff are used effectively and efficiently makes good economic sense. Monitoring alone does not save lives; it is the changes in management as a consequence of HTM that can improve outcomes. If there is no room for improvement of therapy, then HTM is unlikely to improve outcome.
HTM is best used as a strategy to ensure and maintain a high quality of care (health maintenance). It is much less good at early detection of clinical deterioration (crisis detection) due to the problem of false-positive test results.
Most alerts created by weight gain or changes in heart rate or heart rate variability will turn out to be false alarms. A low threshold of response that leads to small adjustments of treatment to maintain the patient in an ideal “health envelope” rather than a high threshold that leads to potentially large and expensive interventions is likely to lead to better long-term outcomes. The health maintenance strategy is currently more time consuming for staff but increases patient engagement and motivation, and will become more efficient as systems are developed that help patients make many decisions without close support from a health professional.
II. Diagnostic Confirmation: Are you sure your patient needs Home Telemonitoring?
The patient or a caregiver must be willing and able to comply with the technology.
The patient should fulfill one or both of the following criteria:
Patient is in need of education about heart failure and/or titration of therapy that is not preferably or reliably available by other means
Patient is at high-risk of hospitalization or death. This might include:
Current or recent discharge from the hospital after an exacerbation of heart failure
Complex diuretic regimen
Gross elevation of prognostic biomarkers such as NT-proBNP >2,000 pg/ml (BNP >500 pg/ml)
A. History Part I: Pattern Recognition:
Patients with heart failure most suitable for telemonitoring have the following characteristics:
They are willing to try the technology.
They have a partner or caregiver who can remind them what to do and help them do it.
They are at high enough risk of further events to benefit from intervention.
The cause of heart failure is amenable to therapy. Treatment for left ventricular systolic dysfunction is complex but effective.
They are in need of education about heart failure.
They do not have access to well-structured care by other means.
B. History Part 2: Prevalence:
All patients hospitalized for or with heart failure (about 4% to 5% of all emergency hospital admissions among adults) should be considered for HTM. HTM may also be offered to patients with newly diagnosed heart failure to ensure good education and titration of therapy. HTM may also be considered for out-patients with chronic heart failure who require health education, titration of therapy, or who are at high risk of hospital admission.
C. History Part 3:
D. Physical Examination Findings.
E. What diagnostic tests should be performed?
All patients with heart failure should have a standard biochemistry profile done at no more than 4-month intervals to ensure that potassium is in the ideal range (4.1 to 4.9 mmol/L) and to assess renal function so that appropriate therapeutic choices can be made. Measurements should be more frequent when changes to therapy have been made.
Demonstration that natriuretic peptides are increased provides objective evidence of risk and, if they fall with treatment, this is evidence that the disease is being controlled. Whether natriuretic peptides are a useful target for therapy remains controversial.
A standard hematology profile should be done at least annually to detect anemia, which should be investigated and treated as appropriate.
HTM may be able to document heart rhythm on a daily basis. If it cannot, then an ECG should also be done annually or more often.
Symptoms, heart rate and rhythm, blood pressure and weight are essential pieces of information that guide therapeutic decisions. Combined with the above laboratory information and with knowledge of the underlying disease and comorbidities, most therapeutic decisions can be made remotely without requiring a face-to-face encounter.
1. What laboratory studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?
Heart failure always requires objective confirmation of underlying cardiac function. Natriuretic peptides, interpreted in the light of the patients heart rhythm and renal function, are the most robust tools for determining whether the patient has heart failure. Echocardiography is the standard tool for phenotyping heart failure into left ventricular systolic or diastolic dysfunction, valve disease or predominantly right heart disease but is much inferior to natriuretic peptides in detecting heart failure with a normal ejection fraction and in predicting outcome.
2. What imaging studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?
A. Immediate management.
B. Physical Examination Tips to Guide Management.
C. Laboratory Tests to Monitor Response To, and Adjustments in, Management.
As noted above, standard biochemistry and hematology profiles should be done periodically. Natriuretic peptides may be used to monitor disease progression but it is unclear whether they should be used as a therapeutic target.
D. Long-term management.
HTM should be seen as a flexible tool that can be deployed in times of need and suspended or withdrawn when the situation no longer requires it.
The utility of HTM will very much depend on the patient’s view of its usefulness. Patients who are not convinced of its merits will usually stop sending information, at which point it is appropriate to withdraw equipment.
However, many patients are willing to spend 10 minutes of their day contributing to their own health and well-being. Engagement is greatly enhanced by providing feedback (e.g., charts of weight, heart rate, and blood pressure). The support of a multidisciplinary service is also necessary to maintain patient motivation. For patients who engage successfully with HTM, making measurements becomes part of the daily routine akin to brushing their teeth.
E. Common Pitfalls and Side-Effects of Management
Patients who stop sending information need review because:
It may indicate that the patient is not feeling well or has even died. Do not be complacent when data stops arriving! Ask patients to let you know when not to expect data, either because they are going to be away from home or simply because they do not want to send more information.
The patient does not understand the importance of the measurements. For these patients, health education and motivation is appropriate. If this fails, the service should be withdrawn.
The patient is not coping with the technology. Most patients will cope if they are shown what to do 3 to 6 times but some need additional support from informal caregivers or voluntary services.
False alerts are a major potential drawback of HTM. Using HTM mainly as a “crisis detection” strategy is unlikely to succeed because false alerts are likely to far outnumber true alerts, leading to an increase in hospital admissions. The adverse impact of false alerts may be mitigated by experienced staff but can undermine the confidence of both patients and staff in the system, who are then inclined to ignore alerts, some of which may be true positives, or abandon HTM altogether.
The alternative is a health maintenance strategy, which requires more engagement by staff and patients. This strategy attempts to maintain the patient in an optimal state in terms of symptoms and physiologic and biochemical assessments. For instance, if the heart rate in sinus rhythm is regularly >70 bpm, then the dose of beta-blocker could be increased or ivabradine given.
Another example might be a patient with a blood pressure of 90/60 mm Hg who has no features of fluid overload. In this case the dose of diuretic can be reduced. Eventually, decision-support systems will lighten the additional clinical load that this more active approach to HTM creates. Supplementing health maintenance with crisis detection as a secondary feature remains an appropriate strategy.
IV. Management with Co-Morbidities
Common comorbidities of heart failure are atrial fibrillation, either low or high blood pressure, diabetes, renal dysfunction, and chronic lung disease.
HTM is useful for the detection and management of atrial arrhythmias, but ventricular arrhythmias are usually too short lived to be picked up by conventional HTM. Wearable and implanted technologies should be considered if it is important to detect nonsustained arrhythmias.
HTM is useful for the management of either low or high blood pressure. It is unclear, though, if there is a value in HTM of blood glucose or of lung function.
There are no suitable, inexpensive technologies for HTM of renal function as yet. However, data from HTM (heart rate and rhythm, blood pressure, and weight) might be useful in managing comorbidities.
HTM should include patient education modules that can be shared with the patient either through the HTM console or an interactive television monitoring system.
It should not be assumed that the use of the HTM system is intuitively obvious for patients. Being told how to use the system is not the same as being shown how to use it in, preferably in the home setting. Educating informal caregivers is as important as educating the patients.
Ensure that staff are appropriately trained and motivated.
V. Patient Safety and Quality Measures
A. Appropriate Prophylaxis and Other Measures to Prevent Readmission.
HTM will only prevent readmissions and reduce deaths when patients provide information and the information is acted on by patients and health professionals, often supported by decision support tools in the system.
Readmission and death are best prevented by keeping the patient as well as possible (health maintenance) rather than by allowing them to become as sick as possible (crisis management) before doing something.
B. What's the Evidence for specific management and treatment recommendations?
Guidelines generally report that HTM is an area of interest but do not, as yet, recommend this management strategy (ESC Guidelines on Heart failure 2012; 2013 ACCF/AHA Guideline for the Management of Heart Failure). While more data is usually welcome, the lack of consistency of recommendations across the discipline in these guidelines can only be attributed to some advocacy groups being stronger than others rather than based on a consistent evaluation of the evidence.
The evidence in favor of HTM is substantial for patients who have recently been discharged from the hospital after an admission with heart failure, provided that voice-interactive systems that seem to cause patient disengagement are avoided. I believe that it is wrong for guidelines to micromanage recommendations on some issues (for instance, cardiac resynchronization therapy), and extrapolate widely for others (for instance aldosterone antagonists) but to view other interventions with conservative pessimism.
C. DRG Codes and Expected Length of Stay.
Average length of stay for patients admitted with worsening heart failure varies greatly by country but is typically 5 days in the U.S. and 10 days in Western European countries but longer in Eastern Europe. In Japan, the average length of stay exceeds 20 days.
Longer length of stay has been linked to lower mortality subsequent to discharge, presumably due to a better treatment regimen at discharge. HTM could obviate the need to lengthen stay to optimize therapy. Length of stay depends strongly on comorbidities, which in turn depend greatly on age. In the UK, average length of stay is 5 days for patients aged <65 years but 9 days for those aged >84 years.
The average age of patients admitted with heart failure in the U.S. and Western Europe is about 75 years of age for those with left ventricular systolic dysfunction and about 80 years for other heart failure phenotypes. In Eastern Europe and many developing countries, the average age of admission is about a decade younger.
In-patient mortality is higher in Europe, perhaps because of the longer admission but 30-day mortality is similar in Western Europe and the U.S.. In-patient mortality in patients aged <75 years is about 5% but >15% in those aged >75 years in the UK. The 30-day mortality from an admission with heart failure is about 115 in the U.S. and in the UK about 18% overall, ranging from 9% in those aged <75 years to almost 25% in those aged >75 years.
Readmission rates at 30-days are lower in Europe for the same reason and perhaps due to greater patient stability at discharge but seem little affected by age, perhaps because death prevents many readmissions for older patients. In the UK, the 30-day all-cause readmission rate is about 17% (and has been rising as length of stay for the index hospitalization has shortened) but perhaps only one third of those are due to heart failure. In the U.S., about 25% of patients are readmitted by 30 days (2007-2009 data).
There are considerable uncertainty in these data due to the varied ways in which heart failure is coded. In the U.S., the first two diagnostic codes are often used to report heart failure epidemiologic data, whereas in the UK analyses usually focus only on the first code or include all coded positions. In 2012, there were 77,332 deaths or discharges with heart failure in the first diagnostic position but five times that number (386,401) in any diagnostic position. Diagnostic position had little effect on prognosis.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- I. Remote Patient Monitoring (Telemonitoring) for Heart Failure: What every physician needs to know.
- II. Diagnostic Confirmation: Are you sure your patient needs Home Telemonitoring?
- A. History Part I: Pattern Recognition:
- B. History Part 2: Prevalence:
- C. History Part 3:
- D. Physical Examination Findings.
- E. What diagnostic tests should be performed?
- 1. What laboratory studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?
- 2. What imaging studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?
- III. Management.
- A. Immediate management.
- B. Physical Examination Tips to Guide Management.
- C. Laboratory Tests to Monitor Response To, and Adjustments in, Management.
- D. Long-term management.
- E. Common Pitfalls and Side-Effects of Management
- IV. Management with Co-Morbidities
- V. Patient Safety and Quality Measures
- A. Appropriate Prophylaxis and Other Measures to Prevent Readmission.
- B. What's the Evidence for specific management and treatment recommendations?
- C. DRG Codes and Expected Length of Stay.