A multi-country randomised trial of the role of a new telemonitoring system in CHF: the HHH study (Home or Hospital in Heart Failure). Rational, study design and protocol

Abstract

Aims

The HHH study will be a full scale randomized controlled trial in Italy, Poland and UK enrolling 450 CHF patients (LVEF <40%, NYHA cl. II–IV) in 2 arms (usual clinical practice and home-care strategy).

Methods and Results

Objectives will be: (1) to determine if different strategies of home-care telemonitoring, affect hospital admissions, improve patient sense of well-being, and reduce overall costs of medical care; (2) to define the prevalence and the clinical impact of breathing disorders, arrhythmias and abnormalities of HRV in the occurrence of acute periods of instability and when on optimal therapy; (3) to evaluate in the home setting a new system for continuous monitoring of cardio-respiratory signals plus physical activity with teletransmission of recorded data embedded in an interactive voice response based telehealth system.

The 300 'home strategy' patients will be divided into: (1) normal clinical practice supplemented by telephone contacts; (2) as strategy 1 plus periodic telemonitoring of vital signs parameters; (3) as 2° strategy plus periodic 24 h ECG, respiration and physical activity (NICRAM) recording. Enrolment will be over 12 months with a further 12 months follow-up. The primary end-point will be total bed-days occupancy for heart failure in acute medical/surgical beds.

Conclusions

The first patient was enrolled in September 2002. The trial is expected to be completed at the middle of 2005.

Keywords Telemonitoring; Chronic heart failure; Sleep apnoea; Long-term recording

Introduction

All European countries are faced with an ageing population and so despite improved treatment the number of chronically ill patients, such as heart failure (HF) patients, increases with subsequent large escalation in health care costs. This has led to increasing emphasis on improved care in the community (at home) in place of expensive hospital care. Relatively simple solutions such as improved home-monitoring by telephone and community nurses have been demonstrated to be effective,^1–4 but now state of the art of technological developments has made it possible to extend more sophisticated continuous monitoring into the patient's home. The increased costs and perhaps difficulties for patient acceptability dictate that these novel approaches should be rigorously tested against conventional care and also the more technical strategies already in existence.

The multicentre randomised study HHH (Home or Hospital in Heart Failure) has been designed to develop and evaluate a home-telemonitoring (HT) system to deliver care to HF patients outside the hospital. In a preliminary experience conducted in 133 HF patients the telemonitoring system designed for the trial was demonstrated to be both feasible (compliance higher than 80%) and effective with a significant reduction in all clinical events in patients followed by home telemonitoring. Besides controlling vital signs, this study will evaluate in the home setting a new system for long-term non-invasive cardiorespiratory and activity monitoring (NICRAM), suitable to be self-managed by the patient, with transmission of acquired data through standard telephone lines to health care providers. HHH combines the integration of a novel, patient-oriented HT infrastructure with a rigorous clinical outcomes, assessment of patient and local healthcarer empowerment (knowledge and self-care) and cost-effectiveness study.

Consequently, the project is: (a) a technology assessment study and (b) a clinical research trial, across centres of excellence in three European countries.

In the traditional model of health care delivery for chronic heart failure patients, measurement of vital signs such as heart rate, blood pressure and symptoms at home is usually performed by the primary care nurse or by GP. The HHH study, will extend this concept particularly to include respiration, activity, and longer ECG recording periods that will enable power spectral analysis of these variables during rest, sleep and activity. Patient acceptability and the cost-effectiveness of these more complex measures will be assessed.

No previous studies have considered prolonged home-recordings of ECG and respiration. In patients with HF a high prevalence of breathing disorders (approximately 50%) has been reported.⁵ These breathing disturbances consist of periodic oscillations of tidal volume (Periodic Breathing, PB) separated or not by central apnoea (Cheyne–Stokes respiration, CSR). They are not restricted to patients with severe symptomatic ventricular dysfunction or to sleep, but can also occur in compensated heart failure.^6,7 They are associated with marked fluctuations of arterial oxygen saturation, systolic and diastolic pressure and heart rate, with surges in sympathetic discharge. The pattern of periodic breathing plus the presence of episodes of obstructive sleep apnoea may be detrimental in HF patients and lead to increased mortality.⁸ Once CSR is established, apnoea related hypoxemia causes cardiac diastolic dysfunction, and leads to further increases in sympathetic activity which contribute to potential fatal arrhythmia and cardiotoxicity. Moreover, hyperventilation and the increased work of respiratory muscles place an increased demand upon the already reduced cardiac output; patients frequently complain of fatigue and excessive daytime sleepiness which relate to reduced amounts of total and REM sleep in association with sleep fragmentation due to frequent arousals.⁵

Little data has been published on the prevalence of these respiratory abnormalities in prolonged observational periods (more than one night or one day), and no other studies have related the presence of breathing disorders to the occurrence of symptoms and HF destabilisation during the follow-up. Recent research has shown that the distinction between patients with and without CSR is important, because treatment with continuous positive airway pressure CPAP reduced mortality and need for transplant over the next two years in patients with CSR but not those without.⁹ We propose to determine the prevalence, pattern and outcome of these breathing disorders at home to better identify those patients who may benefit more from a specific treatment.

Periodic home-monitoring of the ECG during follow-up will also give important information regarding the possible link between arrhythmia (onset of atrial fibrillation, or brady/tachyarrhythmia) and periods of acute destabilisation. Moreover from analysis of the variability of heart rate (HRV) it is now possible to estimate non-invasively the level of sympathetic activation¹⁰ which is largely involved in the progression of heart failure. Low HRV has been recently shown to stratify risk in post myocardial infarction¹¹ as well as in congestive heart failure¹² although only a few studies have paid attention to the impact of irregular breathing, which may not only contribute to the severity of HF, but can critically alter the measurement of HRV.⁶

The level of physical activity has also been demonstrated to influence the analysis of HRV¹³ and we will develop a method to monitor this by the use of an accelerometer-based system. Continuous recordings of ECG, activity and respiration represent a unique chance to investigate the role of HRV parameters and abnormal respiratory function in estimating prognosis, and will give new insights whether changes in HRV and breathing pattern during follow-up could be a powerful marker of progressive destabilisation in CHF.

Objectives

The specific aims of HHH Study are as follows: (1) to compare in HF patients different strategies of home-care i.e. (i) simple voice contact by telephone plus routine hospital care, (ii) as (i) plus intermittent monitoring of vital signs, (iii) as (ii) plus intermittent continuous monitoring of ECG, activity and respiration (NICRAM) in relation to changes in: hospitalisation for HF, mortality, therapy, quality of life, patient (and carers), medical knowledge, and level of self care; (2) to evaluate in the home setting a new system for long-lasting recording of NICRAM with teletransmission of recorded data totally self-managed by the patient; (3) to define the prevalence and the clinical impact of breathing disorders, arrhythmias and HRV abnormalities during stable periods and acute periods of instability.

Design and protocol

The HHH study has been designed as a multinational, multicentered telemonitoring network, full scale randomised controlled trial. It is supported by E.C. grant (Action line 10.1 Public Health, proposal No. QLGA-CT-2001-02424)

The study will involve three countries: UK, Italy and Poland and will be conducted in 11 centers (see Appendix 1). Calculations suggest that 450 CHF patients in 2 arms (usual clinical practice and home-care strategy) would be required. Patients randomised to home strategy will be divided in 3 groups according to the three different predefined strategies (group 2.1–2.3). It seems appropriate that each country will enrol the same number of patients (n=150) equally distributed in the 2 arms and in the 3 subgroups to avoid any bias in the quality of CHF assistance following usual clinical practice (reference arm) which is expected to be different in different health care systems.

Patients will be randomised in 4 groups: group 1 follow-up according to usual clinical practice plus predischarge NICRAM (blind); group 2.1 follow-up according to usual clinical practice supplemented by monthly monitoring using conventional telephone contacts and pre-discharge NICRAM (blind); group 2.2 follow-up according to usual clinical practice supplemented by monitoring using conventional telephone contact and intermittent telemonitoring of vital signs (see below). Periodic 24-h NICRAM kept blind from clinical management but available for comparative outcomes analysis; group 2.3 follow-up according to usual clinical practice supplemented by monitoring using conventional telephone contact and intermittent telemonitoring of vital signs and 24-h NICRAM available for clinical management.

Enrolment period will be 12 months followed by a follow-up of a further 12 months. Patients will be screened at pre-discharge from hospital, at 6 and 12 months. A clinical assessment, ECG, blood sample determination, Echocardiogram, and QoL questionnaire will be performed at baseline (+ 24-h NICRAM recording), and 12-month follow-up. Before discharge patients will be instructed how to manage the assigned home care strategy (see groups 2.1, 2.2, 2.3). Detailed written information will also be given.

Inclusion criteria and end-points of the study

Patients older that 18 years of age, with symptomatic heart failure (NYHA II–IV) on optimised medical treatment, LVEF≤40% and one episode of decompensation in the previous 12 months will constitute the study population.

Primary endpoints are as follows: (1) bed-days occupancy for heart failure in acute medical/surgical beds (but excluding long-stay wards); (2) composite measure of patient well-being (death, hospital bed days, symptom score and quality of life). Secondary endpoints are: (1) bed-days occupancy for all cardiovascular reasons; (2) bed-days occupancy for all cause; (3) death (all cause); (4) changes in therapy; (5) costs of care and cost-utility; (6) patient satisfaction, self-care and disease knowledge assessment.

In the HHH study some exploratory endpoints have also been designed: (1) occurrence and patterns of CSR including frequency of duration of hypopnoea/apnoea events; (2) effect of abnormality of respiration, depressed heart rate variability and paroxysmal or persistent atrial fibrillation on phases of decompensation.

Telemonitoring systems

Four integrated telemonitoring systems will be compared in the study: (1) a 24-h answering machine which will allow each patient to get in touch with his/her reference hospital at any time and leave a message requiring help or advice (groups 2.1–2.3); (2) a conventional phone call from the managing nurse to the patient to get information about his/her present clinical status (groups 2.1–2.3); (3) an interactive voice response (IVR) system able to automatically handle the periodic transmission of vital signs from the patient at home to the reference hospital (groups 2.2, 2.3); (4) a portable 24-h recording device, handled by the patient at home, able to record cardiorespiratory and physical activity signals (NICRAM) and to transmit them through a telephone line to the collecting center.

The IVR system will automatically receive vital signs from the patient at home. Main features of this system will be: (a) handling of voice messages left by the patients calling the 24-h automatic answering machine; (b) answering patient's phone and asking all relevant questions (patient's ID and vital signs) using digitally recorded messages; (c) saving entered data into the study database. All these functions are automatic and do not require an operator.

Monotoring of vital signs: periodically the patients enrolled in group 2.2 and 2.3 will transmit to the coordination center via the IVR system the following parameters: (1) weight; (2) heart rate; (3) systolic arterial pressure; (4) dyspnoea score; (5) asthenia score; (6) edema score; (7) changes in therapy; (8) blood samples if obtained. All these data will be automatically stored in the data base by IVR system. Each transmitted parameter will be subjected to an automatic range check and to a stability check, based on the rate of change of each parameter over time. Any suspect case will elicit a request for advice from the monitoring physician.

Monitoring cardiorespiratory signals and physical activity at home. Patients of groups 2.2 and 2.3 will be given a portable device able to record continuously ECG, respiration and physical activity for 24-h at home. Recording and transmission of collected data to the analysis centre (the coordinating centre of each country) will be done by the patient. A specific device has been designed for the project (FM, Monza MI); it is a solid-state light-weight Holter-style recorder with built-in signal preprocessing. Within the device, two sensors monitor continuously body position (supine vs standing) and acceleration. The Holter will be self-positioned by the patients and the data after 24 h recording easily transmitted through a dedicated modem to the coordinating center for analysis. The processor within this special dedicated modem (Appel Electronica srl, Torino) is able to dial the reference centre, set-up a standard modem connection, communicate with the recorder processor, start the transfer of data from the recorder to the transmission box and from the latter to the reference centre, check the correctness of the transmission and start again all the operations in case of failure of data transfer. At the reference centre no human intervention is needed to handle the transmission of the data. Transmission management and data storage is automatically performed by a specialised module integrated within the IVR system.

HHH will show if a new long-term telemonitoring system allows early identification of clinical deterioration, earlier optimisation of treatment, better survival, and improves QoL encouraging self care in HF patients. Such telemonitoring systems, if proven to be effective and practical in this protocol, will then be applicable to several other clinical situations (e.g. rehabilitation post-MI and stroke, obstructive sleep apnoea, chronic lung disease).

The first patient was enrolled in September 2002. The trial ultimated enrolment in June 2004 and is expected to be completed in the middle of 2005.

Appendix A: HHH Enrolling Centers

University of Oxford, TMR & Department of Cardiology, John Radcliffe Hospital – Oxford, UK (P. Sleight chairman of the study, P. Johnson, D. Andrews, J. Dwight).

Fondazione S Maugeri Clinica del Lavoro e della Riabilitazione – Montescano, Italy (GD. Pinna, S. Capomolla, MT. La Rovere, R. Maestri).

Institute of Clinical Physiology, CNR – Pisa, Italy (M. Emdin, C. Passino).

Policlinico di Monza – Monza, Italy (A. Mortara, C. Bersano).

Azienda "Ospedialiera Riuniti" – Trieste, Italy (A. Di Lenarda, L. Vitali Serdoz).

IRCCS Policlinico San Matteo – Pavia, Italy (L. Tavazzi, C. Campana).

Clinical Military Hospital – Wroclaw, Poland (P. Ponikowski, K. Nowak, T. Witowski).

Central Hospital of Ministry Affairs & Administration – Warsaw, Poland (Dr R. Gill, A. Pawlak).

Dr A Sokolowski Hospital – Walbryzch, Poland (Dr R. Szelemej, A. Jurczyk).

University of Glasgow – Glasgow, UK (H. Dargie, M. Barlow).

Romford Cardiovascular Research – Romford, UK (H. Kadr, N. Badhusha).