Article

PDF
Access to the PDF text
Advertising


Free Article !

Archives of cardiovascular diseases
Volume 102, n° 1
pages 29-41 (janvier 2009)
Doi : 10.1016/j.acvd.2008.10.013
Received : 6 October 2008 ;  accepted : 7 October 2008
RESICARD: East Paris network for the management of heart failure: Absence of effect on mortality and rehospitalization in patients with severe heart failure admitted following severe decompensation
RESICARD : réseau Paris-Est pour la prise en charge des patients en insuffisance cardiaque : absence d’effets sur la mortalité et les réhospitalisations chez les patients en insuffisance cardiaque sévère inclus au décours d’une décompensation sévère
 

Patrick Assyag a, Thomas Renaud b, Alain Cohen-Solal c, Magali Viaud d, Henry Krys e, Aleksandra Bundalo d, Pierre-Louis Michel f, Robert Boukobza g, Yann Bourgueil b, Ariel Cohen d,
a 167, avenue Ledru-Rollin, 75011 Paris, France 
b IRDES, 10, rue Vauvenargues, 75018 Paris, France 
c Service de cardiologie, hôpital Lariboisière, 2, rue Ambroise-Paré, 75475 Paris cedex 10, France 
d Service de cardiologie, hôpital Saint-Antoine, AP–HP, université Pierre-et-Marie-Curie, 184, rue du Faubourg-Saint-Antoine, 75571 Paris cedex 12, France 
e 73, boulevard Richard-Lenoir, 75011 Paris, France 
f Service de cardiologie, hôpital Tenon, 4, rue de la Chine, 75970 Paris cedex 20, France 
g Hôpital La-Roseraie, 120, avenue de la République, 93300 Aubervilliers, France 

Corresponding author. Fax: +33 (0) 49 28 28 84.
Summary
Background

Heart failure presents a major public health problem due to its high prevalence and the increasing number of hospital admissions for this condition. A coordinated healthcare network involving general practitioners and cardiologists was set up in the east of Paris in an effort to improve the management and outcomes of patients with severe heart failure.

Aims

To reinforce patient education, improve compliance with medications and identify symptoms requiring treatment modification.

Methods

In this ‘before and after’ study, the control group comprised patients hospitalized for severe heart failure who received conventional management in the year preceding the network set-up. The comparative group consisted of patients hospitalized for severe heart failure who underwent network-led care.

Results

No significant differences were found between rates of first rehospitalization and all-cause mortality at 1 year between control and network groups, or between rates of first hospitalization due to cardiac causes, time to the first event, duration of hospitalization, rates of cardiac death or time to death.

Conclusions

In this non-randomized study, we found no benefit from management according to the RESICARD healthcare network in terms of mortality or hospitalization in patients with severe chronic heart failure.

The full text of this article is available in PDF format.
Résumé
Contexte

L’insuffisance cardiaque (IC) représente un problème de santé publique majeur, lié à sa forte prévalence, et à l’augmentation croissante des hospitalisations. Une coordination de médecins généralistes et de cardiologues de ville et hospitaliers de l’est de Paris s’est constituée dans le cadre d’un réseau ville–hôpital afin de maintenir à domicile les patients insuffisants cardiaques graves.

Objectifs

Renforcer l’éducation du patient, s’assurer de l’observance du traitement et identifier précocement les symptômes justifiant une modification thérapeutique.

Méthodes

La méthodologie utilisée était comparative, non randomisée. Les patients hospitalisés pour IC sévère dans les hôpitaux participants dans l’année précédant l’installation du réseau et répondant aux critères d’inclusion ont constitué le groupe témoin ayant bénéficié d’un traitement conventionnel. Les patients hospitalisés pour IC sévère dans les hôpitaux participants dans l’année suivant l’installation du réseau et répondant aux critères d’inclusion ont constitué le groupe réseau ayant bénéficié d’une prise en charge spécifique.

Résultats

Aucune différence significative n’a été retrouvée concernant les taux de première réhospitalisation et de mortalité à un an toutes causes confondues entre les groupes témoin et réseau. De même, aucune différence significative n’a été retrouvée concernant le taux de première réhospitalisation pour cause cardiaque, le délai moyen de survenue par rapport à l’inclusion et la durée moyenne d’hospitalisation. Aucune différence significative n’a été observée entre les deux groupes concernant les décès de cause cardiaque et leurs délais de survenue par rapport à l’inclusion.

Conclusion

Cette étude prospective non randomisée n’a montré aucun bénéfice à la prise en charge par le réseau de soins RESICARD, en termes de réduction des hospitalisations et de la mortalité, dans l’IC chronique sévère.

The full text of this article is available in PDF format.

Keywords : Heart failure, Mortality

Mots clés : Insuffisance cardiaque, Mortalité


Abbreviations

ACE : Angiotensin-converting enzyme
COACH : Coordinating study evaluating outcomes of advising and counseling in heart failure
GP : General practitioner
LVEF : Left ventricular ejection fraction
NYHA : New York Health Association
OR : Odds ratio
RR : Risk ratio
SMV : Outpatient medical coordination

Background

Heart failure and its management present a major public health concern. The magnitude of this problem is reflected across a wide range of evidence. Heart failure affects more than 4% of the adult population and more than 20% of those aged 65 years and over – who represent a rapidly growing component of the general population [1]. One in three adults over 55 will develop heart failure during their lifetime; the 5-year survival rate is 35%, depending on the diagnosis [2]. This condition is the most common cause of hospitalization in patients over 65 years [3]. More than one-quarter of patients die within 1 year of diagnosis [4] and one-third are readmitted within 6 months of discharge [5]. Hospitalization accounts for two-third of costs related to heart failure [6].

Despite recent progress in our understanding of the pathogenesis of heart failure and the therapies available for its treatment [7], the prognosis for patients remains poor and their quality of life is profoundly affected [8]. Management strategies aimed at improving this situation therefore appear necessary. Multidisciplinary management for patients with heart failure after their discharge from hospital has been tested with success in Anglo-Saxon countries and in Sweden [9], reducing effectively both the number of readmissions and, to a lesser extent, heart failure-related mortality. In addition, recent European recommendations on the management of chronic heart failure [7] recommend the creation of professional healthcare groups to reduce both heart failure-related symptoms and hospital admissions (class IA) and to improve survival rates (class IB).

Since observations and potential solutions described in the international literature were considered to be applicable to France, a group of hospital- and community-based cardiologists decided to create a healthcare network for the management of patients with severe heart failure in the Île-de-France area. These primary care physicians (GP and cardiologists) and health service and private hospital cardiologists collaborated to build a healthcare network, RESICARD, covering the “bassin de Vie No. 1” (i.e., the 11th, 12th, 19th and 20th districts located on the east side of Paris).

The primary aim of this network was to improve the quality of life of patients with heart failure by reducing midterm mortality and number of hospital readmissions. In practical terms, the aims of RESICARD were to encourage closer cooperation between hospital and community care environments, improve coordination between GP and cardiologists and provide educational programmes targeting both patients and their families. This approach complied with the principles of Good Medical Practice and ongoing efforts to improve the healthcare/expenditure ratio, particularly by reducing hospitalization-related costs.

Methods
Population

Patients were recruited from three Paris state university hospitals (Saint-Antoine, Beaujon and Tenon) and one private hospital (La-Roseraie). Patients were recruited at the end of their hospitalization for acute cardiac decompensation. Inclusion criteria were as follows:

age greater than 18 years;
presence of criteria for severe heart failure as defined by the European Society of Cardiology [10] in patients hospitalized in a cardiology intensive care unit;
willingness to be monitored as stipulated in the protocol and to be followed-up by a GP and a cardiologist on an outpatient basis.

The exclusion criteria were as follows:

recent transmural infarction (<1 month) resulting in congestive heart failure;
no history of congestive heart failure;
valvular heart disease; reversible myocardial ischaemia (planned revascularization over the short term) or obstructive cardiomyopathy whose aetiological treatment corrected the heart failure;
serious renal or hepatic disease;
NYHA functional class I or II at admission;
any life-threatening or functionally incapacitating disease such as cancer or chronic pulmonary disease;
patients on a waiting list for a heart transplant;
multisite pacemaker;
long-term mechanical circulatory assistance;
scheduled discharge to a long-stay convalescent home;
personal or psychological criteria such as the patient’s refusal to participate, lifestyle incompatible with regular monitoring, inability to understand and answer the questionnaire;
participation in another protocol.

Patient recruitment

In this ‘before and after’ study, patients were recruited at participating hospitals in two consecutive stages:

January 2001 to February 2002: inclusion of hospitalized patients in the year preceding creation of the healthcare network. This ‘control’ group was to receive conventional care;
February 2002 to July 2003: inclusion of hospitalized patients in the year following creation of the healthcare network. This ‘network’ group was to receive network-led care (therapeutic intervention group).

Patients meeting the inclusion criteria were informed about the working hypothesis, and two groups of patients, treated according to the principles of Good Medical Practice, were considered. After discussion with the methodology specialists, it became obvious that randomization of patients would be difficult [11] because of the risk of a protocol bias arising during the appointment with the patient’s physician. We therefore decided to apply a comparative methodology, using a ‘before and after’ approach in terms of creation of the network.

Function of the healthcare network

This network was created as stipulated in legislation-governing networks with an administrative council, a steering committee and a scientific committee. It was financed by the Fond d’aide à la qualité des soins de ville (Community Care Quality Fund or FAQSV), the Union régionale des médecins libéraux d’Île-de-France (URML Île-de-France Private Doctors Regional Union) and, to a very limited extent, by the pharmaceutical industry but only for continuous medical training meetings, brochures and information videos provided to patients.

The monitoring scheduled, and in part carried out by the network, involved 150 GP and 25 community cardiologists for each region (East Paris), all of whom signed an agreement. Patients hospitalized for severe heart failure (stage III–IV) in one of the institutions taking part in the study were offered specific care after their discharge. This consisted initially of:

multidisciplinary management:
an educational campaign for patients and their families, which started at the hospital (patients were shown a film and given printed documents) and was to be continued at the doctor’s office, with more information to be handed to the patient and stored in his or her personal treatment diary,
personalized advice on diet,
a meeting with a social worker;

medical care:
medications prescribed were left to the discretion of the doctor, who was guided by current prescription guidelines and Good Medical Practice (national and international recommendations, consensus, etc.),
immediately upon their discharge, patients were enrolled in the ‘network arm’ of the study (therapeutic-intervention group), illustrated in Figure 1 and were given more regular and prearranged appointments with their GP and cardiologist than they would have normally (control group, standard treatment). Patients were asked to attend a first appointment with the GP within 72hours of discharge from hospital. The protocol stipulated that they visit their GP every 2 weeks until the third month and once or twice each month thereafter depending on their clinical condition. Patients were requested to visit their cardiologist in the first week after discharge from hospital, followed by a monthly appointment for 3 months, then every 3 months, again depending on their clinical condition. Special care was made for each of these appointments at which patients were advised on how best to manage their disease. When necessary, additional appointments were arranged for treatment adjustment, managing complications, physiotherapy and advice on nursing care;

availability of SMV. The role of the SMV was to coordinate the network globally and to manage patients individually. The SMV could be contacted by patients or physicians for information and feedback. The SMV was also responsible for contacting the patients regularly by phone. The SMV consisted of an assistant doctor who had received training in education measures for patients with heart failure and whose role was to work closely with a nurse to ensure that the patient’s educational goals were met. Together, they provided patients with educational information support and dietary advice, and recorded patients’ weights, heart rates and blood pressures on a weekly basis. In the event of deterioration in a patient’s clinical condition (‘alarm signals’) or an emergency, they informed the GP and/or cardiologist. They answered the patient’s questions and assessed their level of participation in the network and satisfaction with the care provided. The team also included a social worker and dietician. The aims of the SMV were to collect data about the patients in the network, optimize coordination between GP and cardiologists for patient-education purposes, guide patients with social problems and run a medical telephone helpline open 24/7, through which patients could remain in permanent contact with their network doctor. All of the care received by patients was recorded in their treatment diary, which they kept and handed over to the healthcare provider at the different appointments.



Figure 1


Figure 1. 

Network overview.

Zoom

Follow-up and assessment criteria

The control group consisted of patients who were willing to take part in the study, with assessment being an integral part of conventional treatment, in line with Good Medical Practice. Clinical and sociodemographic information was recorded at the time of inclusion (hospital discharge), then at regular intervals via phone calls with the patients and/or their physician. Each time the patient was readmitted, further information was recorded. The patients were not given any planned appointments or specific examinations as part of the assessment. They received care according to usual and proven Good Medical Practice and their treatment was optimized at the discretion of their physician and was recorded. Conversely, they were not given any specific educational materials.

Patients in both groups provided their informed consent and follow-up lasted for 12 months.

Assessment criteria

The primary endpoints (primary assessment criteria) were:

onset of major cardiac events, defined as readmission or death from cardiovascular causes during follow-up;
interval between inclusion and onset of a major cardiac event.

The secondary endpoints (evaluation criteria) were all-cause death, total number of hospital readmissions, number of readmissions for heart failure and duration of hospitalization (days in hospital avoided).

Follow-up

Patients were evaluated on the basis of a data grid, making it possible to monitor changes in the objectives mentioned above. The evaluation was carried out at 6 months and at 12 months. The evaluation grid listed the patient’s condition at the time of the evaluation, the care required since the last evaluation and a quality-of-life score.

Statistical analysis

The statistical analysis included:

comparisons of baseline data and of major events using χ2 tests. All percentages were compared with non-parametric exact tests;
comparisons of changes in clinical parameters: survival curves were created and analysed using the Kaplan-Meier method and a log-rank test;
evaluation of the effects on the primary endpoints of individual management types (network, non-network) and various demographic and clinical variables, using Cox semiparametric proportional hazard models.

Results
Baseline characteristics of study population

A total of 454 patients were enrolled at the end of the recruitment period, 25 of whom were excluded immediately: 15 at the beginning of the study and 10 who were either excluded from the protocol or lost to follow-up. The final study population therefore comprised 429 patients: 211 were enrolled during the control phase (control group, conventional care) and 218 during the network phase (network group, therapeutic intervention).

Sociodemographic characteristics and heart-failure profiles

The sociodemographic characteristics and heart-failure profiles of the patients are shown in Table 1. There was no significant difference in the mean age of control and network groups (73.8±13.4 versus 72.1±14.3 years, respectively). A higher proportion of patients in both groups were men (56.9% versus 59.6%, respectively). No significant difference in inclusion site was observed between the two groups, with most patients being enrolled at the Tenon and Saint-Antoine university state hospitals. Most patients in both groups were independent of each other and had no history of heart failure in the year preceding their inclusion (control group 79.7% versus network group 84.8%). Conversely, patients in the network group showed a significant improvement in their functional status after discharge versus preadmission compared to the patients in the control group (p <0.001), as evidenced by the three-class drop in NYHA functional class observed in 20.7% of cases versus 1.9% of controls and the two-class drop in 44.6% of cases versus 38.8% of controls. Indeed, 9.8% of patients in the network group versus 24.2% in the control group were in NYHA class III at discharge.

Ischaemic heart disease was the most common cause of heart failure in both groups but with a significantly greater occurrence difference in the network group (53.2% versus 42.7%, p =0.028). Conversely, dilated cardiomyopathy was more common in the control group than in the network group (30.3% versus 21.6%, p =0.038). At least one triggering factor for the initial episode of heart failure was found in both groups (68.6% versus 70.6%). Dysrrhythmia was the most common triggering factor in the control group whereas poor treatment compliance followed by dysrrhythmia were the most common factors in the network group. Poor compliance with treatment at admission was more frequently observed in the network versus the control group (Table 1).

Clinical, paraclinical and therapeutic characteristics

No significant differences were observed between groups for the results of the clinical examinations performed on admission and at discharge. An overall improvement in the clinical condition of patients was observed during the initial period in hospital, with alleviation of the physical signs of congestive heart failure in both groups (Table 2). Conversely, the prescribed discharge medications varied between groups, with beta-blockers (64.7% versus 48.8%, p =0.0009), ACE inhibitors (80.7 versus 67.8%, p =0.002), potassium-sparing diuretics (36.2 versus 19.9%, p =0.0002) and statins (34.9 versus 24.5%, p =0.02) prescribed more frequently in the network group.

No significant difference was observed between groups in terms of the conventional laboratory test profile obtained on admission and at discharge. In both groups, the mean concentration of serum creatinine was approximately 123μmol/L at discharge. Similarly, no significant differences were found in the electrocardiographic anomalies observed. Chest X-rays taken at admission showed cardiomegaly, with a cardiothoracic index greater than 0.6 in most patients in both groups, more frequent alveolar oedema in the control group (74.7% versus 57.7%; p =0.003) and, by contrast, more frequent pleural effusion in the network group (37.4 versus 27.1%; p =0.026). Approximately one-half of the patients enrolled in the study had a LVEF between 20% and 40% on the echocardiogram. LVEF was preserved (>50%) in one-fifth to one-quarter of patients, with no significant difference observed between the two groups.

Primary endpoints

No significant differences were observed between the control or network groups in the number of deaths from cardiovascular causes (9.9% versus 8.6%, respectively) or in the mean time of onset after inclusion (158.8±92.8 versus 167.3±102.4 days, respectively) (Table 3). This was also the case for the numbers of first readmissions for cardiovascular causes (27.7 versus 31.1%, respectively) and the mean time for readmission after inclusion (96.9±79.5 versus 107.5±94.7 days, respectively) (Table 4). Finally, the mean duration of hospitalization for cardiovascular causes was comparable in both groups (12.7±11.7 versus 13.3±13.4 days, respectively) (Table 4).

Secondary endpoints

No significant difference was observed between groups in the overall 1-year mortality rate, irrespective of cause (control group 21.6% versus network group 18.6%) (Table 3). Similarly, all-cause readmission rates were also comparable (Table 4). In addition, no significant differences were observed in the number of deaths from heart failure and their time of onset calculated from the date of inclusion (Table 3). The rate of first readmissions for heart failure, the interval between discharge and readmission, and the duration of hospitalization were comparable (Table 4).

Major cardiac event-free survival (death from cardiac cause or readmission for cardiac reasons)

The network approach did not impact on the interval between inclusion and the onset of a major cardiac event, or on the probability of such an event occurring by itself (Figure 2).



Figure 2


Figure 2. 

Kaplan-Meier curve showing event-free survival in control and network groups.

Zoom

Survival curves plotted for clinical variables known to be indicators of a poor prognosis (discharge NYHA class, existence of ischaemic heart disease and number of episodes of heart failure in the preceding year) confirmed their predictive value for major cardiac events (Figure 3). Dual stratification of the survival curves by inclusion group, and these indicators taken individually, showed that the network did not exert a protective effect against the onset of a major cardiac event as a function of these three parameters (Figure 3A–C).



Figure 3


Figure 3. 

Kaplan-Meier curves showing event-free survival in controls and network groups. A. NYHA class; B. Existence or not of ischemic heart disease; C. History of heart failure episodes; D. Intensity of RESICARD monitoring: as a function of frequency of SMV calls. E. As a function of number of GP and cardiologist appointments.

Zoom

The survival curves were stratified using the two RESICARD monitoring intensity indicators with the aim of distinguishing, within the network group, the patients monitored as stipulated in the protocol and poorly or unmonitored patients, thus refining the ‘network effect’ analysis. The first indicator was SMV monitoring, for which three groups of patients were studied: controls, those insufficiently monitored by the SMV (<1.5 calls per month) and those correctly monitored by the SMV (>1.5 calls per month) (Figure 3D). The second RESICARD monitoring indicator compared patients who never visited their GP or cardiologist (n =40), those who rarely or only occasionally attended appointments (GP<0.4 times per month, cardiologist>0.2 times per month) and those who saw their healthcare provider regularly (GP>0.4 times per month, cardiologist>0.2 times per month) (Figure 3E). The two curves obtained show that intense monitoring by the SMV and more frequent medical appointments (GP and cardiologist) exert a marginally protective effect on the onset of major cardiac events, but this trend was not statistically significant.

Factors associated with onset of major cardiac events and cardiac mortality at 1 year

In multivariable analysis using Cox’s semiparametric proportional modelling, after adjustment for cardiovascular risk factors (age, sex, hypertension and smoking), presence of ischaemic heart disease, NYHA functional class at discharge, LVEF and history of heart failure, the only independent parameters significantly associated with the onset of major cardiac events were history of heart failure (p =0.05), other heart disease (p =0.0004) and NYHA class at discharge (p =0.0033). Global management via the network (p =0.23), intensity of monitoring received via the SMV (p =0.9) and frequency of appointments with the GP or cardiologist (p =0.3) had an impact on the primary endpoints, time to onset between inclusion and major cardiac events.

In a second multivariable analysis adjusting for age, sex, NYHA class at discharge, history of cardiac disease, hospital at which the patient was enrolled, type of heart disease and treatments at discharge (diuretics, ACE inhibitors, angiotensin II receptor antagonists, beta-blockers), the only independent parameters associated with cardiac mortality at 1 year were history of at least one episode of heart failure in the preceding year (OR 2.9, 95% confidence interval [95% CI] 2.2–15.7; p=0.02) and NYHA class II or IV at discharge (OR 2.9, 95% CI 1.1–7.3; p=0.02). Management by the network (OR 1.02, 95% CI 0.4–2.3; p=0.9), intensity of SMV follow-up and frequency of appointments were not independent predictors for cardiac mortality (OR 0.7, 95% CI 0.2–3.1; p=0.7 and 0.9, 0.3–2.8; p=0.8, respectively).

Discussion

This prospective, non-randomized study examined a collaborative network of different healthcare professionals (health system and private practice) who followed a multidisciplinary outpatient approach to the management of acute heart failure. The results do not concur with the literature, since no benefit was found for patients managed by the RESICARD network in terms of readmissions or mortality.

Several studies have demonstrated the benefits of multidisciplinary care for patients with chronic heart failure after discharge [12, 13, 14, 15, 16, 17, 18, 19, 20]. While most report favourable results, not all have shown a positive effect [21, 22, 23, 24]. For example, the randomized multicentre COACH study [24], which included one of the largest series of patients described since the advent of extrahospital heart failure management networks (n =1024), did not show that this management approach had a significant effect on the combined primary endpoint, readmissions and mortality subsequent to heart failure when compared to the control group receiving conventional care. Thus, in view of the disparities in the results published, it appears necessary to analyse the different factors that could explain the lack of significant differences in our network experience with regard the incidence of events in the control and network populations.

Network strategy and implementation

The findings from studies reporting positive results for multidisciplinary network management of discharged patients vary depending on the event examined (mortality or readmission) and the type of programme applied. A recent meta-analysis [9] explored the results of 29 randomized trials, with the aim of stratifying the benefits obtained as a function of the strategy adopted, and showed that follow-up by a multidisciplinary specialist team in a medical setting (doctor’s office or clinic) or via home visits led to a significant reduction in mortality (RR 0.75, 95% CI 0.59–0.96) and readmissions for heart failure (RR 0.74, 95% CI 0.63–0.87) or for all causes combined (RR 0.81, 95% CI 0.71–0.92). Encouraging self-care and education through a telephone advisory system or nurse-led information programme during home visits led to a significant reduction in readmissions for heart failure (RR 0.66, 95% CI 0.52–0.83) and all readmissions (RR 0.73, 95% CI 0.57–0.93) but did not have any impact on mortality (RR 1.14, 95% CI 0.67–1.94). Finally, using telephone calls to encourage patients to see their doctor if their condition worsened led to a significant decrease in the number of readmissions for heart failure (RR 0.75, 95% CI 0.57–0.99) but did not have a significant effect on mortality (RR 0.91, 95% CI 0.67–1.29) or all-cause readmissions (RR 0.98, 95% CI 0.80–1.20). In spite of these nuances, the meta-analysis remains on the whole positive and in favour of a multidisciplinary approach, irrespective of the strategy employed. Nonetheless, these results suggest that physician expertise, a carefully structured multidisciplinary teamwork and long and detailed preparation before implementation are required for this kind of outpatient management programme to be of optimal benefit to patients. Training qualified staff and developing links amongst the various healthcare professionals involved is a primary objective and one that is indeed difficult to achieve [25].

The strategy adopted in RESICARD aimed to be ‘optimal’ in that after discharge, patients were given a series of appointments with their GP and cardiologist at regular and tailored intervals over a 3-month period and were then monitored by the multidisciplinary SMV team. In our analysis, stratification of survival curves as a function of the two monitoring intensity indicators (Figure 3D and E) showed that intense monitoring by SMV and more frequent medical appointments (GP and cardiologist) had a protective effect on the onset of major cardiac events, although this trend was not statistically significant. Should precedence have been given to home visits rather than to phone calls, as was the case in the studies reporting optimal results? This assertion remains a matter of debate.

Questions that could arise regarding McAlister et al.’s meta-analysis [9] concern the fact that although the indirect comparisons of the studies analysed were adjusted, evaluation of the relative benefits of the different parts of each type of management plan could be biased by the absence of direct comparison. For example, one study [26] reported that a daily phone call to patients supervised by a nurse was more effective and less costly than home visits by nurses who lacked specialization in heart failure. Moreover, the recent COACH study [24] tested two levels of monitoring intensity (moderate: appointment with the cardiologist+contact with a nurse; intensive: appointment with a cardiologist+contact with a nurse+home visits+multidisciplinary team comprising dietician, physiotherapist and social worker) versus a control group receiving conventional care (cardiologist appointment) and found no difference in the onset of adverse events as a function of the strategy applied (control group 42%; moderate monitoring group 41%, hazard ratio 0.96, p =0.73; intensive monitoring group 38%, hazard ratio 0.93, p =0.52). Thus, direct comparison with other types of management plans (or monitoring intensity) appears to be next step requiring investigation.

Time dedicated to learning within the network

Three elements appear fundamental to the success of a multidisciplinary management program. The first of these – availability of nursing staff who have received specialized training in the management of an elderly population presenting with several comorbid conditions and who are sometimes difficult to deal with owing to cognitive impairment, instability and/or psychological difficulties – is the cornerstone of all management programs. The second element is the time and effort invested by staff in educating patients and their families about the disease, what triggers it and the importance of complying with treatment and diet plans, making it necessary to preselect patients considered most likely to benefit from a network-care system. The third and last element is the willingness of patients to consult healthcare professionals trained in heart failure care. The meta-analysis by McAlister et al. [9] found that 27 of 29 studies were conducted in high-risk cardiac failure patients (recently discharged from hospital), and that all the programmes were run by trained teams from specialist teaching hospitals and centres. The authors of this meta-analysis also highlighted the fact that the benefits observed in their study could not with certainty apply to low-risk patient groups, especially if the care programmes concerned were run by teams that had not received the requisite training. However, the RESICARD network differs in several ways: it was designed for all patients hospitalized for acute heart failure in an intensive care unit but without preselection on the basis of potential compliance; the specialist staff was located in the community (and not in hospitals) and the patient monitoring programme meant that doctors in the outpatient sector were involved heavily. Moreover, as part of the RESICARD network, healthcare providers were obliged to take on new roles and learn new tasks such as patient education, and this was a lengthy process. This bias is not quantifiable but could explain, at least in part, the neutral network effect observed.

Optimal medical treatment and monitoring

Large studies evaluating current medical practices in Europe for heart failure continue to show that evidence-based medicine is still not sufficiently available [27]. In recent meta-analyses [9, 28, 29, 30] evaluating the benefits of network management after discharge, few studies report the frequency of ACE inhibitor or beta-blocker prescription, making it impossible for the authors of meta-analyses to conclude as to the specific and additional benefits of such programmes, since the benefits observed could simply be attributed to optimal guideline-based practices or patient compliance with treatment. In RESICARD, a high proportion of patients in both control and network groups were prescribed beta-blockers (48.8% versus 64.7%, respectively) and ACE inhibitors (67.8% versus 80.7%) when they were discharged home, with significant differences in favour of the network group (Table 2). We cannot therefore rule out the possibility that the negative result obtained in this study could be related to the fact that patients in both groups were receiving optimal treatment after hospital discharge. For obvious reasons, it is difficult to obtain and measure a network-specific effect in a group receiving optimal medical treatment.

This hypothesis is supported by the relatively low incidence of events measured in the control group in our study, which was comparable to that reported in the COACH study [24], indicating that patients were managed in a satisfactory manner before RESICARD was set up.

In the initial studies [9], patients in the control group did not receive any specialist follow-up, which could explain the higher incidence of events. In our study as well as in the COACH study [24], patients in the control group were conventionally monitored by a cardiologist, which might improve the likelihood of receiving optimal management.

Heart failure due to systolic dysfunction versus preserved LVEF

In our study, about 60% of patients in both groups had an LVEF of less than 40% (p =NS). LVEF was preserved (LVEF>50%) in about one-quarter of cases in both groups and between 40% and 50% in the remainder. Similarly, patients with both altered and preserved LVEF were enrolled in the COACH study [24], with a mean LVEF of 34%. Very few meta-analyses reporting the positive effects of multidisciplinary management programmes mention whether the studies analysed included both heart failure patients with systolic dysfunction and those with preserved LVEF [28, 29]. In the meta-analysis by Holland et al. [28], only 15 of the 30 studies included reported the baseline LVEF, with values varying from 22% to 43%. Similarly, in the six-study meta-analysis by Phillips et al. [29], LVEF was between 27% and 41%, with a mean of 34%. In view of the lack of precision regarding LVEF and given the prognosis for heart failure with preserved LVEF, we cannot rule out the possibility that the positive results obtained in these studies could be related to recruitment bias.

Study population: limitations associated with non-randomization

In this unusual context, in which a medical practice rather than a medication is tested, it appeared difficult to propose randomization at discharge. We did consider patient randomization (testing the protocol) but felt that we would encounter a protocol bias given that it would be difficult for cardiologists to monitor patients in the conventional and therapeutic-intervention groups. Randomization of physicians also appeared difficult. This led us to adopt a ‘before and after’ approach, particularly since the initial meta-analysis [31] of the first non-randomized multidisciplinary management studies, which also used the same method as ours (ok?), had reported convincing results. Seven non-randomized studies were analysed; the aims of the programme were to improve compliance with recommended optimal treatments, reinforce patient education and optimize monitoring. The follow-up period was in excess of 3 months. Five studies reported a significant improvement in the functional status of patients, their aerobic capacity and quality of life. Six studies showed a 50–80% reduction in the risk of readmission, leading the authors of the meta-analysis to suggest a potential beneficial effect of this type of management on readmission rates. However, non-randomization is the primary limitation of prospective observational studies in general and of the study we conducted in particular, since it makes it impossible to exclude the presence of unknown biases, making interpretation of the results speculative.

Conclusions

Large-scale heart failure management programmes exist in many countries. Recent recommendations encourage the development of specialist outpatient management programmes with the objective of improving symptoms, reducing hospitalizations and increasing survival rates. Our non-randomized, prospective study did not show that inclusion in the RESICARD healthcare network had a beneficial effect on the number of hospitalizations or the mortality rate in this patient group. However, it did suggest that intensive monitoring by an SMV (community medical coordination) and attending appropriate medical appointments did have a slightly protective (although not significant) effect against the onset of major cardiac events. Although the results were negative, this study allowed us to explore the difficulties related to setting up a heart failure management network involving a variety of healthcare professionals and in training qualified staff dedicated to this task. While the literature has shown that factors play a key role in the success of these programmes, the elements truly responsible for the favourable effects have yet to be identified. To solve this problem, additional studies measuring monitoring and management intensity, in particular, are required.

References

Cleland J.G., Swedberg K., Poole-Wilson P.A. Successes and failures of current treatment of heart failure Lancet 1998 ;  352 : SI19-SI28
Gysele S. Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure Eur Heart J 2004 ;  25 : 1614-1619
Massie B.M., Shah N.B. Evolving trends in the epidemiologic factors of heart failure: rationale for preventive strategies and comprehensive disease management Am Heart J 1997 ;  133 : 703-712 [cross-ref]
Jong P., Vowinckel E., Liu P.P., Gong Y., Tu J.V. Prognosis and determinants of survival in patients newly hospitalised for heart failure: a population-based study Arch Intern Med 2002 ;  162 : 1689-1694 [cross-ref]
Krumholz H.M. Predictors of readmission among elderly survivors of admission with heart failure Am Heart J 2000 ;  139 : 72-77 [cross-ref]
Ryden-Bergsten T., Andersson F. The health care costs of heart failure in Sweden J Intern Med 1999 ;  246 : 275-284 [cross-ref]
Swedberg K., Cleland J., Dargie H., and al. Task force for the diagnosis and treatment of chronic heart failure of the european society of cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): the task force for the diagnosis and treatment of chronic heart failure of the European Society of Cardiology Eur Heart J 2005 ;  26 (11) : 1115-1140
MacIntyre K., Capewell S., Stewart S., Chalmers J.W., Boyd J., Finlayson A., and al. Evidence of improving prognosis in heart failure: trends in case fatality in 66,547 patients hospitalized between 1986 and 1995 Circulation 2000 ;  102 : 1126-1131
McAlister F.A., Stewart S., Ferrua S., McMurray J.J. Multidisciplinary strategies for the management of heart failure patients at high risk for admission: a systematic review of randomized trials J Am Coll Cardiol 2004 ;  44 : 810-819 [cross-ref]
The task force in heart Failure of the European Society of Cardiology. Guidelines for the diagnosis of heart failure. Eur Heart J 1995;16: 741–51.
Gallo C., Perrone F., De Placido S. Informed versus randomised consent to clinical trials Lancet 1995 ;  346 : 1060-1064 [cross-ref]
Cline C.M., Israelsson B.Y., Willenheimer R.B., Broms K., Erhardt L.R. Cost effective management programme for heart failure reduces hospitalisation Heart 1998 ;  80 : 442-446
Jaarsma T., Halfens R., Huijer Abu-Saad H., Dracup K., Gorgels T., van Ree J., and al. Effects of education and support on self-care and resource utilization in patients with heart failure Eur Heart J 1999 ;  20 : 673-682 [cross-ref]
Blue L., Lang E., McMurray J.J., Davie A.P., McDonagh T.A., Murdoch D.R., and al. Randomised controlled trial of specialist nurse intervention in heart failure BMJ 2001 ;  323 : 715-718 [cross-ref]
Kasper E.K., Gerstenblith G., Hefter G., Van Anden E., Brinker J.A., Thiemann D.R., and al. A randomized trial of the efficacy of multidisciplinary care in heart failure outpatients at high risk of hospital readmission J Am Coll Cardiol 2002 ;  39 : 471-480 [cross-ref]
Krumholz H.M., Amatruda J., Smith G.L., Mattera J.A., Roumanis S.A., Radford M.J., and al. Randomized trial of an education and support intervention to prevent readmission of patients with heart failure J Am Coll Cardiol 2002 ;  39 : 83-89 [cross-ref]
Harrison M.B., Browne G.B., Roberts J., Tugwell P., Gafni A., Graham I.D. Quality of life of individuals with heart failure: a randomized trial of the effectiveness of two models of hospital-to-home transition Med Care 2002 ;  40 : 271-282 [cross-ref]
Strömberg A., Mårtensson J., Fridlund B., Levin L.A., Karlsson J.E., Dahlström U. Nurse-led heart failure clinics improve survival and self-care behaviour in patients with heart failure: results from a prospective, randomised trial Eur Heart J 2003 ;  24 : 1014-1023
Rich M.W., Gray D.B., Beckham V., Wittenberg C., Luther P. Effect of a multidisciplinary intervention on medication compliance in elderly patients with congestive heart failure Am J Med 1996 ;  101 : 270-276 [cross-ref]
Rich M.W., Beckham V., Wittenberg C., Leven C.L., Freedland K.E., Carney R.M. A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure N Engl J Med 1995 ;  333 : 1190-1195 [cross-ref]
Ekman I., Andersson B., Ehnfors M., Matejka G., Persson B., Fagerberg B. Feasibility of a nurse-monitored, outpatient-care programme for elderly patients with moderate-to-severe, chronic heart failure Eur Heart J 1998 ;  19 : 1254-1260 [cross-ref]
Veterans affairs cooperative study group on primary care and hospital readmissionWeinberger M., Oddone E.Z., Henderson W.G. Does increased access to primary care reduce hospital readmissions? N Engl J Med 1996 ;  334 : 1441-1447 [cross-ref]
Doughty R.N., Wright S.P., Pearl A., Walsh H.J., Muncaster S., Whalley G.A., and al. Randomized, controlled trial of integrated heart failure management: the Auckland Heart Failure Management Study Eur Heart J 2002 ;  23 : 139-146 [cross-ref]
Jaarsma T., van der Wal M.H., Lesman-Leegte I., and al. Effect of moderate or intensive disease management program on outcome in patients with heart failure: coordinating study evaluating outcomes of advising and counseling in heart failure (COACH) Arch Intern Med 2008 ;  168 : 316-324 [cross-ref]
Konstam M.A., Lindenfeld J., Pina I.L., Packer M., Lazar R.M., Warner Stevenson L., and al. Key issues in trial design for ventricular assist devices: a position statement of the Heart Failure Society of America J Card Fail 2004 ;  10 : 91-100 [cross-ref]
Benatar D., Bondmass M., Ghitelman J., Avitall B. Outcomes of chronic heart failure Arch Intern Med 2003 ;  163 : 347-352 [cross-ref]
Lenzen M.J., Scholte op Reimer W.J., Boersma E., Vantrimpont P.J., Follath F., Swedberg K., and al. Differences between patients with a preserved and a depressed left ventricular function: a report from the EuroHeart Failure Survey Eur Heart J 2004 ;  25 : 1214-1220 [cross-ref]
Holland R., Battersby J., Harvey I., Lenaghan E., Smith J., Hay L. Systematic review of multidisciplinary interventions in heart failure Heart 2005 ;  91 : 899-906 [cross-ref]
Phillips C.O., Singa R.M., Rubin H.R., Jaarsma T. Complexity of program and clinical outcomes of heart failure disease management incorporating specialist nurse-led heart failure clinics. A meta-regression analysis Eur J Heart Fail 2005 ;  7 : 333-341 [cross-ref]
Jovicic A., Holroyd-Leduc J.M., Straus S.E. Effects of self-management intervention on health outcomes of patients with heart failure: a systematic review of randomized controlled trials BMC Cardiovasc Disord 2006 ;  6 : 43 [cross-ref]
Philbin E.F. Comprehensive multidisciplinary programs for the management of patients with congestive heart failure J Gen Intern Med 1999 ;  14 : 130-135 [cross-ref]



© 2009  Published by Elsevier Masson SAS.
EM-CONSULTE.COM is registrered at the CNIL, déclaration n° 1286925.
As per the Law relating to information storage and personal integrity, you have the right to oppose (art 26 of that law), access (art 34 of that law) and rectify (art 36 of that law) your personal data. You may thus request that your data, should it be inaccurate, incomplete, unclear, outdated, not be used or stored, be corrected, clarified, updated or deleted.
Personal information regarding our website's visitors, including their identity, is confidential.
The owners of this website hereby guarantee to respect the legal confidentiality conditions, applicable in France, and not to disclose this data to third parties.
Close
Article Outline