Study design
This study is designed as a randomised controlled trial to evaluate the effect of CTR integrated with an RPM programme compared to RPM without CTR in CHF patients. Participants will be recruited during hospitalisation for ADHF at Máxima Medical Centre, Eindhoven/Veldhoven, and Catharina Hospital, Eindhoven, The Netherlands. Following hospital discharge, all participants will start with RPM and will be followed by their cardiologist and specialised HF nurse for up-titration to optimal medical therapy (OMT). When a stable situation is achieved (defined as OMT and unchanged HF symptoms for at least two weeks), participants will be randomised to either 18 weeks of comprehensive CTR in addition to RPM (intervention group), or RPM without CTR (control group). All participants will sign informed consent before enrolment. Data will be collected at enrolment (-t2), during the pre-intervention period (hospital discharge until stable HF; -t1), at randomisation (t0), 18 weeks after randomisation (t1), and 6 months after randomisation (t2) (Fig. 1, Table 1). The study protocol was approved by the local Medical Research Ethics Committee (MREC) of Máxima Medical Centre. The trial is registered at the Netherlands Trial Register (NTR) with registration number Trial NL9619.
Table 1. Overview of the assessments during the study period
-t2 = at hospitalisation for acute decompensated heart failure, -t1 = within 2 weeks after hospital discharge, t0 = randomisation and allocation, tx = intervention period, t1 = follow-up visit 1–18 weeks after the intervention started, t2 = follow-up visit 2–6 months after the intervention started.
Functional test: CLET recovery = recovery of the constant-load exercise test, CPET max = maximal cardiopulmonary exercise test, HGS = Handgrip Strength test, PHB = wrist worn device ‘Philips Health Band’, SPPB = Short Physical Performance Battery.
Questionnaire: Acceptance & satisfaction = questionnaire for acceptance and satisfaction of the CTR programme and digital platform, Activity level = subjective activity level question, DS-14 = Type D personality scale, EHFScBS = European Heart Failure Self-Care Behavior Scale, FitMáx = FitMáx©-questionnaire, GAD-7 = Generalised Anxiety Disorder assessment, HADS = Hospital Anxiety and Depression Scale, IPQ = Illness Perception Questionnaire, KCCQ-12 = Kansas City Cardiomyopathy Questionnaire, MLHFQ = Minnesota Living with Heart Failure Questionnaire, MMAS-8 = adjusted Morisky Medication Adherence Scale, MoCA = Montreal Cognitive Assessment, NutriMáx = nutrition questionnaire based on the national nutrition guideline from the Dutch Health Council, PLF = Premorbid Lifestyle questionnaire, PHQ-9 = Patient Health Questionnaire, SNAQ = Short Nutritional Assessment Questionnaire.
Other: Laboratory measurements (e.g. NTproBNP, GDF15).
* All objectives will be assessed in both groups.
Study population
All patients hospitalised for ADHF will be screened for study participation according to the in- and exclusion criteria (Table 2). Patients who are interested will be enrolled before discharge after signing informed consent.
Inclusion criteria:
|
Table 2
Inclusion and exclusion criteria
1) Aged ≥ 18 years
|
2) Diagnosed with congestive heart failure
|
3) Hospitalised primarily for acute decompensated heart failure (ADHF) at the time of inclusion
|
4) Sufficient digital literacy, or caretaker with digital literacy
|
5) Able to speak and read the Dutch Language
|
Exclusion criteria:
|
1) Unable to understand the purpose and procedures of the study
|
2) Unable to walk a distance of 4 meters independently (walking aids are allowed)
|
3) Cardiac rehabilitation programme followed in the previous 12 months
|
4) No internet connection
|
5) Untreated life-threatening cardiac arrhythmia
|
6) Early phase after acute coronary syndrome (latest 3 months)
|
7) Uncontrolled hypertension
|
8) Advanced atrioventricular block
|
9) Severe aortic stenosis
|
10) Up-coming major (cardiac) surgery in 3 months
|
Randomisation, blinding and treatment allocation
Participants are randomly allocated to the intervention or control group (1:1) after reaching stable HF and completing the first measurements. Participants will be replaced when discontinuing from study protocol before randomisation. Randomisation is performed by the investigator using a computerised randomisation system in the web-based database software Castor EDC (Castor Electronic Data Capture, Ciwit BV, Amsterdam, The Netherlands). The computer system uses variable block sizes (2,4) stratified for age (< 75 years or ≥ 75 years) and left ventricular ejection fraction (LVEF ≤ 40%, LVEF 41–49% or LVEF ≥ 50%) for randomisation. The participants, investigator, and other medical professionals will not be blinded for the treatment allocation due to the nature of the intervention. For the assessment of the functional tests (Short Physical Performance Battery and Handgrip Strength test) a blinded investigator will be involved.
Remote Patient Management
Directly after hospital discharge, study participants start with remote patient management (RPM). RPM, also known as telemonitoring, allows health care providers to closely review patient-generated health data, interact with the patient, and initiate clinical treatment if needed 33. RPM is associated with reduced mortality, and admission rate 32. Study participants are asked to daily measure their vital parameters (e.g, heart rate, blood pressure) and weight with medical devices (iHealth Track Blood Pressure monitor, iHealth Lina scale) and manually entry the data on the digital monitoring platform (Mibida BV). In addition to the vital parameters, participants are instructed to answer 5 short HF related questions on the platform (e.g., ‘Do you experience shortness of breath?’). Alerts will be generated by the platform based on predefined and individualised reference values. A specialised cardiac nurse will review the patient-generated health data and alerts during the weekdays, and a cardiac care nurse from the Coronary Care Unit / Cardiology Ward will review the data during evenings and weekends. The patient is instructed to contact the nurse by telephone between defined hours when an alert is generated; for urgent matters the medical professionals are available 24/7. Besides daily monitoring to recognise early clinical decompensation, the specialised nurse and cardiologist are responsible for the up titration to optimal medical therapy (OMT) according to ESC guidelines 16. In a daily meeting with the cardiologist, the nurse will discuss patients with clinically relevant alerts, patients for medication up titration, and patients for remote routine follow-up (2 weeks, and 6 weeks after hospital discharge, and every 3 months). The team will decide what action is needed (e.g. increase in diuretics because of weight gain and oedema). The specialised nurse and cardiologist are primarily responsible for achieving OMT after hospital discharge. All participants will continue with RPM during the entire study period.
Intervention group: Cardiac Telerehabilitation
This multidisciplinary 18-weeks CTR programme includes interventions on physical function and activity, diet, and mental health. Physical, nutritional and psychological goals will be assessed by the specialised nurse at the CTR intake procedure. After the general intake, the patient will be referred for an intake with the physical and occupational therapist, dietician, and psychologist.
Physical training
Patients in the intervention group start with a combined in-hospital intake assessment with the physical and occupational therapist. They will assess the current physical and independence status, the limitations, and discuss personal goals following the latest Dutch CR guideline 34. During this first assessment, the therapist and patient will determine the focus of the following consults and determine the distribution between occupational and physical follow-up. After the first assessment, the first 3 training sessions will be in-hospital, followed by 2 live-video training sessions, and weekly video coaching sessions with occupational or physical therapist. There will be a multidisciplinary evaluation in week 10 with the occupational and physical therapist. The frequency of the coaching sessions after week 10 will be adjusted in response to the evaluation, with the expectation that 2–4 more sessions are needed in the following 7 weeks. After 18 weeks, there will be a final remote assessment with both therapists.
Nutrition intervention
Patients in the intervention group start with an initial video assessment with a dietician to assess and discuss dietary pattern, (unintentional) weight loss, malnutrition, and adherence to sodium and fluid restriction. Before the first consultation, the patients will fill in the NutriMáx-questionnaire based on the nutrition guideline from the Dutch Health Council for providing insight in to the actual nutritional behaviour 35. Personalised nutrition goals will be discussed by the dietician and patient with specific attention for sodium and fluid restrictions. The nutrition behaviour will be followed during the intervention period using a chatbot, which is a web-based/mobile-based conversational dietary assessment tool used for monitoring daily dietary behaviour. There will be 3 individual video consultations with the dietician, and one group-based video consultation with other CTR participants and dietician. Before the final assessment the NutriMáx will be re-assessed.
Psychological intervention
Patients in the intervention group will have a video-based intake assessment with a psychologist for the screening of anxiety and depression symptoms, and the coping strategies used for managing their illness and health. In addition to this assessment, information about psychological status will be obtained from questionnaires (e.g. HADS, GAD-7, PHQ-9, DS-14). The patient and psychologist will determine whether follow-up consultation is needed, and in what form.
Digital platform
The study intervention will be performed using a secured, personalised, and patient-centred digital platform (‘My Flow Coach’, Mibida BV, Eindhoven, The Netherlands). The platform is used for daily monitoring of the RPM programme, and visualising the data from the wrist-worn device (Philips Health Band; PHB, Philips Electronics Nederland B.V., Eindhoven, The Netherlands) directly after hospital discharge. The platform has features that enable patients to:
-
Register and evaluate vital parameters, and HF related complaints for daily monitoring
-
Register and evaluate physical, nutritional, and mental health rehabilitation goals (intervention group)
-
Register treatment modules (intervention group)
-
Upload and review data from the wrist-worn device (e.g. heart rate, steps, active minutes, energy expenditure)
-
Perform video- and chat consultations with health care providers
-
Provide relevant caregivers (e.g. cardiologist, specialised cardiac nurse, physical therapist, occupational therapist, dietician, psychologist, investigator) access to relevant clinical data.
Wrist-worn device
All study participants will receive a wrist-worn device (PHB) at hospital discharge for continuous data collection (e.g. activity counts, heart rate, respiration rate, total energy expenditure, active energy expenditure, steps, and sleep). The PHB provides continues health tracking by measuring the movement and physiological parameters by photoplethysmography technology (PPG). The measurements will be transferred to an application on their mobile phone or tablet via Bluetooth, and to the Philips Actigraphy Server System (PASS). The data will be collected and saved using a study ID (identification) code, and is only accessible by the research team. Participants get access to their raw data in the mobile application and digital platform. They will be asked to wear the wrist-worn device preferably 24 hours a day, but at least during every exercise moment. The data will be collected from hospital discharge until 6 months after randomisation.
Chatbot
The chatbot application (Mibida BV) will monitor the daily dietary behaviour of the intervention group during the intervention period. The patients will regularly receive multiple choice questions from the chatbot about their intake (e.g. ‘Goodmorning, what did you eat this morning?’) (Fig. 2; Chatbot schedule). The chatbot has the intention to trigger ‘nudging’ based on the Nudge Theory 36, which refers to strategically changing the environment to anticipate on altering peoples’ behaviour without forbidding any options. The chatbot is accessible with an application on a mobile phone or tablet for the intervention group during the intervention period. The data entered by the patients will be visible for the dietician on the digital platform.
Control group
Both the control and intervention group will continue to use the RPM programme during the study period, given this is part of regular care, and outpatient appointments with the cardiologist and specialised HF nurse will be planned when needed.
Outcome measures
The primary outcome measure is physical functional capacity at randomisation, 18 weeks and 6 months after randomisation. Secondary outcome measures are recovery after submaximal exercise, maximal exercise capacity, subjective health status and quality of life, personality and behaviour aspects, nutrition behaviour, compliance and acceptance to the intervention, fluctuation of congestive HF biomarkers, and readmission rate.
Physical functional capacity
Physical functional capacity is assessed with the Short Physical Performance Battery (SPPB) functional test. The SPPB is an objective assessment tool to evaluate the lower extremity function in older persons to reflect the physical self-reliance. It is described as a screening tool to detect frailty 37, and a predictor of major adverse health-related events (e.g. disability, hospitalisation and mortality) in elderly patients 38–40. The test consists of three parts: gait speed, standing balance and time to rise from a chair. Each test is scored out of 4 points; 0 corresponds with not able to perform the test, and 4 with best performance. A maximum of 12 points can be scored, 0–3 corresponding with severe physical limitation, 4–9 high risk, and 9–12 low risk for developing new physical limitations.
Maximal exercise capacity and recovery after submaximal exercise
All patients will start the intervention period with a symptom limited maximal Cardiopulmonary Exercise Test (CPET) on a cycle ergometer (Lode Corival, Groningen, The Netherlands) using an individualised ramp protocol aiming of a total test duration of 8–12 minutes. CPET will be used to determine the maximal exercise capacity and the peak workload, to support the prescription of a tailored exercise rehabilitation programme 41,42. The effect of the intervention on maximal exercise capacity, defined as VO2peak, will be evaluated with the validated FitMáx©-questionnaire at randomisation, and follow-up 43. Furthermore, recovery of O2 kinetics (τ-rec) after submaximal exercise will be assessed with a Constant-Load Exercise Test (CLET) at 50% of the peak workload at randomisation and follow-up. The CLET includes 2 min of rest, 2 min of unloaded pedalling, 6 min at 50% of the maximum workload, and a resting period of at least 5 min until reaching or approaching the VO2 baseline value. Submaximal oxygen uptake kinetics are found to be equally related to functional mobility in elderly and HF patients as VO2peak 44. CHF patients consume oxygen at a higher level to their peak oxygen uptake than healthy subjects during activities of daily life (ADL), Spruit et al. found a VO2 of 38–52% of the peak VO2 during ADL 45. Therefore, CLET at 50% of the peak workload is expected to be indicative of ADL activities, and better tolerated and more representative for changes in physical capacity than maximal exercise 46.
Frailty risk screening
Frailty has been defined as a clinical syndrome with declines in multiple physiological systems associated with increased vulnerability to stressors related to adverse outcomes, such as disability, falls, hospitalisation, and mortality 47. Although an universal consensus about an appropriate and accessible screening tool is lacking, the prominent domains are found in the Vigorito frailty assessment tool 48. In this study, a global frailty screening will be made based on this Vigorito frailty assessment tool 49 using the following domains:
- Physical activity and function; evaluated with SPPB and Handgrip Strength test (HGS) at randomisation, and follow-up;
- Malnutrition; evaluated with SNAQ (Short Nutritional Assessment Questionnaire) for malnutrition screening at inclusion;
- Cognitive impairment; evaluated with MoCA (Montreal Cognitive Assessment) at randomisation;
- Comorbidities and medication use; evaluated with the number of medications used at inclusion (hospital discharge);
- Physiological and social status; evaluated with the depression and anxiety screening questionnaires HADS (Hospital Anxiety and Depression Scale), GAD-7 (Generalised Anxiety Disorder assessment) and PHQ-9 (Patient Health Questionnaire) at randomisation, and follow-up.
Subjective health status and quality of life
Health related quality of life (HRQoL) is evaluated with the KCCQ-12 (Kansas City Cardiomyopathy Questionnaire), and MLHFQ (Minnesota Living with Heart Failure Questionnaire). KCCQ-12 is the shorter version of the self-administered KCCQ, and is assessed to measure patients’ perception on their health status. It includes the frequency of HF symptoms, physical and social limitations, and quality of life (QoL) impairment as a result of HF within a 2-week recall period. The MLHFQ is a 21-item, self-administered instrument developed to independently measure the effect of HF on patients’ lives (in the physical, socio-economic and emotional/physiological domain) over the previous 4 weeks. 50. The KCCQ and MLHFQ are both reliable and validated questionnaires responsive to clinical change, however the KCCQ is more strongly correlated with functional status parameters, and MLHFQ more responsive to improvement in physical functional capacity (6MWT) 51,52.
Personality and behaviour aspects
Self-care behaviour and personality characteristics are evaluated with questionnaires to evaluate its influence on the primary outcome. Personality is evaluated with the DS-14 (Type D personality scale), and premorbid behaviour with the PLF (Premorbid Lifestyle questionnaire). Illness perception and self-care behaviour are evaluated using the IPQ (Illness Perception Questionnaire) and EHFScBS (European Heart Failure Self-Care Behavior Scale).
Nutrition behaviour
Nutrition behaviour is assessed with the NutriMáx-questionnaire based on the Dutch dietary guideline of the Dutch Health Council 35. NutriMáx consists of 18 questions that will provide an overview of the nutrition behaviour based on the 15 nutrition categories listed in the national nutrition guideline. The total score is scaled from 0–32, and represents the adaptation to the guideline. A score of 28–32 represents good adaptation, 20–27 moderate adaptation, and 19 or below poor adaptation.
Compliance and physical activity
The data collected by the PHB and stored at PASS is used to determine the compliance and physical activity. Compliance will be evaluated in terms of: (i) the time of wearing the wrist-worn device, (ii) and achievement of the personalised physical goals set by the therapist for the rehabilitation group. Physical activity will be evaluated during the different study phases by change in active and total energy expenditure, and step counts in both groups. Medication adherence and subjective activity level is evaluated with the MMAS-8 (Morisky Medication Adherence Scale) and self-constructed 1-question activity questionnaire.
Acceptance of the intervention and platform
Satisfaction and acceptance of the CTR programme in general, the chatbot application, and digital platform is assessed using a questionnaire based on the 5-point Likert scale. The control group will only evaluate the satisfaction and acceptance of the digital platform.
Readmission rate and other adverse events
The readmission rate and other adverse events are assessed during the study period in both groups. Readmission is defined as a hospitalisation for at least 24 hours. We will differentiate between HF related causes, other cardiovascular causes, and non-cardiovascular causes of readmission. Other adverse events that will be reported are: 1) acute decompensation without hospital admission, 2) myocardial infarction, 3) emergency room visit without hospitalisation, 4) (cardiac) surgery, 5) admission to a nursing home or rehabilitation centre, 6) (cardiovascular) death, and 7) adverse events that might be related to the intervention.
Heart failure biomarkers
HF related biomarkers (NTproBNP and GDF15) are measured to evaluate the effect of CTR on fluctuation of these biomarkers, and the predictive value of these markers on physical functional capacity, readmission, and mortality. N-terminal prohormone of BNP (NTproBNP) is a reliable gold standard diagnostic biomarker in HF, and has high prognostic accuracy for death and HF hospitalisation 53. Growth Differentiation Factor 15 (GDF15) is a less known biomarker, although multiple studies have provided evidence of its prognostic value in CHF patients 54,55. These biomarkers are determined at 6 different moments in the study: (I) the first day of the initial hospital admission, (II) at hospital discharge, (III) 1–2 weeks after hospital discharge, (IV) when ‘stable HF’ is reached, (V) 18 weeks after randomisation, and (VI) 6 months after randomisation.
Statistical analysis
All analyses will be performed according the intention-to-treat principle. Descriptive statistics will be used to present demographic and baseline characteristics. Between- group differences in the primary and secondary endpoints will be analysed by the unpaired T-test for continuous variables and by a chi squared test for categorical variables. A paired T-test will be used to evaluate within-group differences for the primary endpoint.
Sample size calculation
The sample size calculation is based on the primary outcome measurement – physical functional capacity assessed by the SPPB. Assuming that the study population in general corresponds with lower performance (SPPB score ≤ 9), and that CTR will result in an increase of 1.6 points (SD 2.17) as described by Rengo et al., 64 patients are needed to achieve a statistical power of 0.80 56. Previous research found a subjective and quantitative better physical functional capacity with an increase of 0.4–1.5 points 57.