Summary
Background
The prevalence and incidence of heart failure (HF) in elderly patients are increasing worldwide. Management of HF with reduced ejection fraction (HF-REF) in patients aged 80 years or more follows international guidelines, despite the lack of a dedicated study in this frail population.
Aims
To determine whether optimized management of HF-REF in patients aged 80 years or more can improve quality of life at 6 months.
Methods
Patients aged 80 years or more hospitalized for acute HF-REF were randomized prospectively into an optimized group or a control group (usual care). All patients benefitted from the same in-hospital management. Optimized group patients were also managed at 3, 6 and 9 weeks, and 3, 6, 9 and 12 months after initial hospitalization, to optimize HF-REF treatment. The primary endpoint was quality of life at 6 months.
Results
The trial was stopped prematurely, according to prespecified rules and an independent data monitoring board, after 34 patients were included ( n = 17 in each group). There was no difference in quality of life at baseline and at 6 months between the two groups ( P = 0.14 and 0.64, respectively), although a significant improvement was observed between baseline and 6 months in the optimized group compared with the control group: −20.2 ± 25.2 ( P = 0.01) versus −9.9 ± 19.0 ( P = 0.19). Mortality at 12 months was lower in the optimized group (17.7% vs 47.1%; P = 0.03). There was no increase in acute renal failure, hyperkalaemia or falls in the optimized group ( P = 0.49, 1 and 1, respectively).
Conclusions
Optimizing the management of HF-REF in patients aged 80 years or more, according to the modalities of the HF80 study, seems to be both effective and safe.
Résumé
Contexte
La prévalence et l’incidence de l’insuffisance cardiaque (IC) chez les patients âgés augmentent. La prise en charge de l’IC à fraction d’éjection du ventricule gauche (FEVG) altérée des patients de plus de 80 ans suit les recommandations internationales malgré l’absence d’étude dédiée.
Objectif
Déterminer si l’optimisation de la prise en charge de l’IC à FE altérée des octogénaires est bénéfique.
Méthodes
HF80 pilot study est une étude pilote, randomisée, prospective. Les patients octogénaires hospitalisés pour IC avec une FEVG altérée ont été randomisés dans un groupe optimisé (avec prise en charge optimisée) ou dans un groupe contrôle. La qualité de vie (QdV) à 6 mois (Questionnaire de Minnesota) était le critère de jugement principal.
Résultats
L’étude a été stoppée prématurément suivant les règles pré-établies par la comité de suivi de l’étude après l’inclusion de trente-quatre patients ( n = 17 dans chaque groupe). Il n’y avait pas de différence de QdV à baseline et à 6 mois entre les 2 groupes ( p = 0,14 et 0,64, respectivement) mais une amélioration significative dans le groupe optimisé à la différence du groupe contrôle : –20,2 ± 25,2 ( p = 0,01) vs –9,9 ± 19,0 ( p = 0,19). La mortalité à 12 mois était moindre dans le groupe optimisé (17,7 % vs 47,1 % ; p = 0,03). Il n’y a pas eu d’augmentation d’insuffisance rénale aigüe, d’hyperkaliémie et de chute dans le groupe optimisé ( p = 0,49, 1 et 1, respectivement).
Conclusions
L’optimisation de la prise en charge de l’IC à FEVG altérée, chez les octogénaires, selon les modalités d’HF80 Pilot Study, semble être efficace et sûre.
Background
Chronic heart failure (HF) represents an epidemic in developed countries. Its prevalence increases exponentially with age because of better management of cardiovascular diseases (e.g. ischaemic heart disease, hypertension) , improved HF prognosis, and an ageing population . The prognosis in this group of patients is poor, with a 1-year mortality of approximately 30% .
Numerous clinical studies have highlighted the benefits of drug therapy and resynchronization therapy in chronic HF patients with reduced left ventricular ejection fraction (HF-REF), ultimately leading to the publication of guideline recommendations . Unfortunately, patients aged 80 years or more were not included in these studies. Hence, no specific guidelines have been proposed in these high-risk patients. Only very sparse studies and subgroup analyses have shown the effectiveness and safety of the use of these therapies among younger patients (aged > 70 or 75 years). We recently showed, in a ‘real life’ analysis, that a combination of HF-REF drugs was effective in reducing mortality in patients aged 80 years or more . Moreover, several registers have revealed an underuse of these therapies in these fragile patients , probably as a result of a fear, on the part of clinicians, of inducing iatrogenic complications.
The main objective of the present study was to assess, in a randomized controlled trial, the impact of optimizing the management of HF-REF on quality of life (QoL) at 6 months in chronic HF patients aged 80 years or more. The secondary objective was to determine the effectiveness and safety of this optimization at 12 months.
Methods
The design and rationale of this prospective, single-centre, randomized controlled trial ( NCT01437371 ) have been described in detail .
Study population
White patients aged 80 years or more and managed for an episode of acute HF were included in the course of their hospitalization. These patients had to present with HF-REF (left ventricular ejection fraction ≤ 35%). Subjects were excluded if they presented with dementia, displayed reduced mobility, were already included in a study to optimize the treatment of HF, or had a treatable aetiology of chronic HF or a creatinine clearance < 30 mL/min/1.73 m 2 , according to the Modification of Diet in Renal Disease (MDRD) equation.
The local ethics committee (Comité de Protection des Personnes Sud Est 6, reference AU 889) and the National Security Agency of Medicines and Health Products (reference 2011-A00305-36) approved the study. Written information was given to patients, including conditions of anonymity and withdrawal from the study, and their written consent (recorded by the promoter in a dedicated archives room) was obtained. The local ethics committee and the National Security Agency of Medicines and Health Products approved this consent procedure. The study was monitored by an independent data safety monitoring board (three clinicians specialized in HF and one statistician specialized in QoL).
Endpoints
The primary endpoint was assessment of QoL in patients aged 80 years or more using the Minnesota Living with Heart Failure Questionnaire (MLHFQ) at 6 months. Secondary outcomes were QoL (12-Item Short Form Survey [SF-12] and MLHFQ), functional capacity, mortality, the composite of mortality and hospitalization for HF, the composite of mortality and all-cause hospitalization, and secondary events at 12 months.
Study design
Patients were recruited and randomized (1:1 ratio, by minimization) into two groups: the usual care group, in which the management of consultations (frequency and modality) was left to the discretion of the patient’s general practitioner and cardiologist; and the optimized management group, in which face-to-face consultations were set at fixed time-points (3, 6 and 9 weeks, and 3, 6, 9 and 12 months after the initial hospitalization) and carried out by the investigation’s cardiologist in order to perform titration of treatment (the higher tolerated doses were targeted for medications).
Stratification was performed for sex (male/female), history of hospitalization for HF (yes/no) and lifestyle before admission (at home/institution).
Identical assessments were performed at baseline and at 12 months for all patients: clinical examination; QoL questionnaires; a 6-minute walking test ; an electrocardiogram; transthoracic echocardiography ; a blood test (haemoglobin, serum potassium, serum sodium, liver evaluation, creatininaemia [MDRD], N-terminal pro-B-type natriuretic peptide, nutritional status) and a Mini-Mental State Examination . During each scheduled consultation in the optimized management group, several variables were recorded to optimize the management of HF: a clinical examination; a blood test; and an electrocardiogram. Adverse events were collected (cardiovascular and all-cause deaths and hospitalizations, hospitalization for worsening renal function, kalaemia > 5.5 mmol/L and > 6 mmol/L).
Patients in both groups each had a ‘log book’ during the 12 months for recording therapeutic adaptations.
Treatment dose was qualified according to three categories: high (highest/target dose recommended, e.g. bisoprolol 10 mg once daily or ramipril 5 mg twice daily or spironolactone 50 mg once daily); medium (>50% of the target dose, e.g. bisoprolol 5 or 7.5 mg once daily or ramipril 2.5 or 3.75 mg twice daily or spironolactone 25 mg once daily); and low (<50% of the target dose, e.g. bisoprolol 1.25 or 2.5 mg once daily or ramipril 1.25 once daily or twice daily or spironolactone 12.5 mg once daily).
Statistical analysis and considerations
As described previously , sample size estimation, based on the study by Morcillo et al. , was fixed at 40 patients by randomization arm to show a difference of 20 points ( σ = 25) with regard to MLHFQ score (two-tailed type I error = 5%, statistical power = 90%, rate of loss to follow up = 15%). An interim analysis was planned in order to conclude on efficacy, safety or futility. For 34 assessable patients (17 per group), a difference in QoL score between the two arms would be considered significant for an adjusted α equal to 0.003 (Lan-DeMets approach; East© software, Cytel, Cambridge, MA, USA). A multidisciplinary committee was established, with the power to stop the study if appropriate, according to clinical relevance and statistical results.
Statistical analysis was performed on intention-to-treat using Stata 13 software (StataCorp LP, College Station, TX, US). The tests were two-sided, with a type I error set at α = 0.05 (except for aim outcome because of interim analysis). Baseline characteristics are presented as mean ± standard deviation or median [interquartile range] for continuous data (assumption of normality assessed using the Shapiro–Wilk test) and as number of patients and associated percentages for categorical variables. Comparison of patient characteristics between the randomization groups was performed using the chi-squared test or Fisher’s exact test for categorical variables (followed by a Marascuillo post hoc test upon reaching significance), and by Student’s t test or the Mann–Whitney test for quantitative variables (homoscedasticity verified using the Fisher–Snedecor test). The primary analysis was completed using recommendations proposed by Vickers and Altman with analysis of covariance considering QoL score at baseline. Comparisons in paired situations (intragroup evolution of QoL) were performed using the paired Student’s t test or the Wilcoxon test when appropriate. Longitudinal analyses regarding QoL were conducted using linear mixed models with random intercept, testing for group and time effects and their interactions. Censored data were estimated using the Kaplan–Meier method and compared between groups using usual log-rank test or Wilcoxon–Breslow–Gehan test, according to their implementation assumptions. As proposed by some statisticians, we chose to report all individual P values for secondary outcomes without any mathematical correction for distinct tests comparing the two arms . A per-protocol analysis was proposed to take into account attrition bias, because two patients in the optimized management group did not complete the follow-up according to the study protocol.
Results
According to the statistical design and sample size estimation, an interim analysis was performed for 34 assessable patients. As a result of the number of adverse events (significant mortality in the usual care group), the study was terminated prematurely on the decision of the local ethics committee and the French National Agency of Medicine and Health Products Safety, based on the proposition of the independent data monitoring board ( Fig. 1 ).
Characteristics at baseline ( Table 1 )
Thirty-four patients were enrolled between October 2011 and May 2013, with a mean age of 84.1 ± 3.2 years in the usual care group versus 84.1 ± 3.5 years in the optimized management group. Seventeen patients were randomized to each group.
| Usual care | Optimized management | |
|---|---|---|
| ( n = 17) | ( n = 17) | |
| Demographics | ||
| Age (years) | 84.1 ± 3.2 | 84.1 ± 3.5 |
| Female sex | 5 (29) | 7 (41) |
| First hospitalization for HF | 9 (53) | 10 (59) |
| Body mass index (kg/m 2 ) | 24.8 ± 3.6 | 23.4 ± 3.6 |
| Vital signs | ||
| Heart rate (bpm) | 87 ± 21 | 83 ± 17 |
| Systolic blood pressure (mmHg) | 120 ± 17 | 121 ± 13 |
| Diastolic blood pressure (mmHg) | 70 ± 6 | 73 ± 13 |
| NYHA class | ||
| I | 0 (0) | 0 (0) |
| II | 2 (12) | 3 (18) |
| III | 7 (41) | 7 (41) |
| IV | 8 (47) | 7 (41) |
| Medical history | ||
| Hypertension | 14 (82) | 10 (59) |
| Diabetes | 10 (59) | 3 (18) |
| Current smoker | 1 (6) | 2 (12) |
| Primary heart failure | 9 (53) | 10 (59) |
| Coronary artery disease | 11 (65) | 8 (47) |
| Hypertensive heart disease | 1 (6) | 1 (6) |
| Dilated cardiomyopathy | 5 (29) | 10 (59) |
| Other cardiomyopathy | 1 (6) | 2 (12) |
| Atrial fibrillation/flutter | 4 (24) | 5 (29) |
| Stroke | 5 (29) | 4 (24) |
| Chronic obstructive pulmonary disease a | 3 (18) | 2 (12) |
| Cancer | 3 (18) | 2 (12) |
| Kidney disease b | 10 (59) | 8 (47) |
| Anaemia c | 6 (35) | 4 (24) |
| Severe anaemia d | 1 (6) | 3 (18) |
| Paraclinical variables | ||
| Left ventricular ejection fraction (%) | 26.7 ± 5.5 | 27.8 ± 7.1 |
| NT-proBNP | 11,493 [4910–20,634] | 12,929 [6778–20,446] |
| Serum creatinine (μmol/L) | 112 ± 33 | 109 ± 36 |
| GFR (mL/min) | 59.4 ± 23.1 | 58.5 ± 18.6 |
| Serum potassium (mmol/L) | 4.3 ± 0.7 | 3.9 ± 0.6 |
| Serum sodium (mmol/L) | 136 ± 5 | 140 ± 2 |
| Haemoglobin (g/dL) | 12.9 ± 1.3 | 13.6 ± 2.1 |
| Sinusal rate | 11 (65) | 11 (65) |
| Left bundle branch block | 7 (41) | 7 (41) |
| Supraventricular tachycardia | 4 (24) | 5 (29) |
| 6-minute walk test | 134 ± 121 | 194 ± 148 |
| Quality of life | ||
| MLHFQ (total) | 38.1 ± 18.7 | 47.7 ± 23 |
| MLHFQ (physical) | 19.2 ± 11.2 | 24.2 ± 12.4 |
| MLHFQ (emotional) | 8.1 ± 6 | 9.6 ± 6.3 |
| SF12 (physical) | 31.9 ± 11.3 | 31.3 ± 11.4 |
| SF12 (mental) | 49.2 ± 15.9 | 44.4 ± 18.5 |
| Medication/device at baseline | ||
| ACE inhibitor or ARB | 8 (53) | 7 (47) |
| Beta-blocker | 8 (53) | 8 (53) |
| Mineralocorticoid receptor antagonist | 2 (13) | 3 (20) |
| Ivabradine | 0 (0) | 0 (0) |
| Loop diuretic | 14 (93) | 15 (100) |
| Implantable CRT | 1 (6) | 1 (6) |
a Diagnosed by a pulmonologist.
b Defined by creatinine clearance < 60 mL/min/1.73 m 2 according to the Modification of Diet in Renal Disease equation.
c Defined by haemoglobin < 13 g/dL in men and 12 g/dL in women.
d Defined by haemoglobin < 11.5 g/dL in men and 10.5 g/dL in women.
Characteristics at baseline are presented in Table 1 . There were no significant differences between the two groups, except for diabetes (59% in the usual care group vs 18%) and serum sodium (136 ± 5 mmol/L in the usual care group vs 140 ± 2 mmol/L in the optimized management group at baseline; however, no difference was observed at discharge: 137 ± 5 mmol/L in the usual care group vs 137 ± 3 mmol/L in the optimized management group).
No difference was observed between both groups at 12 months in terms of left ventricular ejection fraction (41.4 ± 15% in the usual care group vs 40.4 ± 13% in the optimized management group; P = 0.79) and N-terminal pro-B-type natriuretic peptide concentration (2032 ± 2079 ng/L in the usual care group vs 4033 ± 4577 ng/L in the optimized management group; P = 0.27).
Quality of life ( Table 2 , Fig. 2 )
At baseline and at 6 months, there was no difference in QoL between the two groups ( P = 0.14 and P = 0.64, respectively). The improvement from baseline to 6 months was significant in the optimized management group (−20.2 ± 25.2; P = 0.01) in contrast to the usual care group (−9.9 ± 19.0; P = 0.19). There was a significant difference in the evolution of mental state, assessed with the SF-12 questionnaire, in favour of the optimized management group ( P = 0.04). There was an improvement in New York Heart Association class in both groups between discharge and 12 months ( Fig. 3 ).
| Usual care | Optimized management | P a | |
|---|---|---|---|
| ( n = 8) | ( n = 13) | ||
| MLHFQ (total) | |||
| Baseline | 36.1 ± 20.4 | 44.3 ± 25.1 | 0.33 |
| Month 6 | 26.3 ± 21.2 | 24.2 ± 24.4 | 0.64 |
| Evolution | −9.9 ± 19 | −20.2 ± 25.2 | 0.56 |
| MLHFQ (physical) | |||
| Baseline | 19 ± 12.1 | 22.3 ± 13.2 | 0.43 |
| Month 6 | 13.8 ± 10.8 | 13.5 ± 13.1 | 0.77 |
| Evolution | −5.3 ± 10.8 | −8.8 ± 11.9 | 0.56 |
| MLHFQ (emotional) | |||
| Baseline | 7.4 ± 5.5 | 8.5 ± 6.3 | 0.74 |
| Month 6 | 9.3 ± 9.2 | 5 ± 5.4 | 0.35 |
| Evolution | +1.9 ± 9.6 | −3.5 ± 7.2 | 0.19 |
| SF-12 (physical) | |||
| Baseline | 30.1 ± 13.2 | 32.3 ± 10.6 | 0.64 |
| Month 6 | 40.5 ± 11 | 37.2 ± 12.2 | 0.51 |
| Evolution | +10.4 ± 12.6 | +4.9 ± 11.7 | 0.51 |
| SF-12 (mental) | |||
| Baseline | 55.6 ± 18.3 | 46 ± 17.6 | 0.26 |
| Month 6 | 43 ± 12.8 | 51.3 ± 11.5 | 0.17 |
| Evolution | −12.5 ± 19.5 | +5.3 ± 20.8 | 0.04 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
