Patient population

Selection of centres has been described in detail previously . Briefly, centres had to have fulfilled the requirements for participation in the I-CARE programme , thereby validating their specialization in CHF management. Training was completed for 110 centres in 2007 and for more than 220 in 2010. Among the first 110 trained centres throughout France, 61 (55.5%) volunteered to contribute patients to the ODIN cohort.

Patients were enrolled prospectively between 2007 and 2010; during this period the therapeutic guidelines for CHF remained largely stable . Enrolment in the study was consecutive. To increase the external validity, exclusion criteria were deliberately kept to a minimum: patients were excluded only if they attended sites not participating in the I-CARE programme or if they declined to participate.

All patients received usual care, which consisted of care management according to European guidelines . Medical therapy was adjusted as judged necessary by the investigator. Patients completed a full education programme in addition to receiving usual care.

Four patient groups were defined according to age: < 60 years (Age 1); 60 to < 70 years (Age 2); 70 to < 80 years (Age 3); and ≥ 80 years (Age 4).

The initial clinical questionnaire did not take into account the pathophysiological type of CHF ( i.e . systolic or diastolic CHF); however, left ventricular ejection fraction (LVEF) was available in 92.3% of the patients, with a mean value of 39.8 ± 14.2% (range 6–88%). The threshold for defining LVEF groups has been discussed in European and American heart failure guidelines . We chose the usual 45% threshold for defining reduced LVEF and the 50% threshold for defining preserved LVEF. In accordance with the American guidelines , we preferred to create a borderline group to see if patients with an LVEF between 45 and 50% presented a similar profile to one of the other usual LVEF groups. It was therefore decided to create three groups according to LVEF: < 45% (heart failure with reduced ejection fraction, HF-REF); 45–50% (heart failure with borderline ejection fraction, HF-BEF); and > 50% (heart failure with preserved ejection fraction, HF-PEF).

Data collection

The investigation conformed to the principles outlined in the Declaration of Helsinki. The study was approved by the Commission nationale informatique et liberté – as required by French law for any patient cohort study in France – and the institutional independent ethical committee of the French Society of Cardiology. All patients gave informed consent before their inclusion in the cohort. Declarations of inclusion were made by centres on both nominative 1-page and medical 2-page record forms for each patient after being given an anonymous number by the centre; these were mailed separately to the French Society of Cardiology. Data were then recorded on two types of computerized case record forms according to the procedure requested and approved by the Commission nationale informatique et liberté.

Data were collected on the main cardiovascular classes of drugs administered by the oral route: all recommended CHF classes (angiotensin-converting enzyme inhibitors [ACEIs], beta-blockers, angiotensin-receptor blockers [ARBs], diuretics, mineralocorticoid-receptor antagonists [MRAs], digoxin and nitrates) and all other main non-CHF cardiovascular classes usually prescribed for the treatment of CHF causes or complications (amiodarone, antiplatelet therapies [aspirin and clopidogrel], anticoagulant drugs [antivitamin K therapies only], calcium blockers and statins). Furthermore, for all main CHF classes, the types of drugs used most frequently were listed with their daily dosage (perindopril and ramipril for ACEIs; bisoprolol, carvedilol, metoprolol and nebivolol for beta-blockers; candesartan and valsartan for ARBs; furosemide for loop diuretics; and eplerenone and spironolactone for MRAs) .

Follow-up data were collected through contact with the attending physicians, the patients or their families. If missing, vital status was assessed from the registries of the patients’ birthplaces. Only 4.9% ( n = 158) of patients were lost to follow-up.

Statistical analysis

All continuous variables are described as means ± standard deviations, except for time variables, which are expressed as medians [interquartile ranges]. All categorical variables are described with absolute and relative frequency distributions. Comparisons between groups used one-way analysis of variance and unpaired t tests for continuous variables and χ ² tests for discrete variables. The Bonferroni test was used for comparisons between LVEF groups. Survival curves were generated by the Kaplan-Meier method and compared with log-rank tests. Prognostic factors for mortality, including age-defined groups, LVEF-defined groups, risk factors, heart failure aetiology, biological variables and main CHF drug treatments, were analysed with bivariate and multivariable Cox models. Biological variables were categorized into quartiles for serum creatinine (< 9.5 mg/L; 9.5–11.5 mg/L; > 11.5–15.0 mg/L; > 15 mg/L) or divided according to their median value for serum glucose (≤ and > 0.99 g/L) and serum haemoglobin (≤ and > 13.0 g/L). Serum natriuretic peptides were organized as non-decompensated (defined as B-type natriuretic peptide ≤ 400 pg/mL and/or N-terminal pro-B-type natriuretic peptide ≤ 450 pg/mL if the patient was aged < 50 years, ≤ 900 pg/mL if aged between 50 and 75 years and ≤ 1800 pg/mL if aged > 75 years) or decompensated levels. For multivariable models, a stepwise variable selection with sle = 0.1 and sls = 0.05 was applied. Results are expressed as hazard ratios with a 95% confidence interval.

All statistical analyses were performed using SPSS version PASW 18.0 software. For all tests, P < 0.05 was considered significant.

Cardiovascular medications in the whole population

The ODIN population has been detailed previously . In summary, a total of 3248 patients were included in the ODIN cohort, 11 of whom were declared twice ( i.e . by two different centres). For these patients, the first declaration was considered as the inclusion in the cohort and the second declaration was considered as a follow-up visit. The ODIN cohort therefore comprised 3237 patients with a mean age of 67.6 ± 14.2 years (range 16–97 years); 69.4% were men. Baseline clinical and biological main variables according to age-defined and LVEF-defined groups are reported in Table 1 . For CHF-dedicated medications, the large majority of patients received blockers of the renin-angiotensin system (69.9% received ACEIs and 20.3% received ARBs, which were always prescribed for ACEI intolerance), beta-blockers (80.6%), and loop diuretics (80.1%). Fewer patients were treated with MRAs (32.7%), digoxin (12.6%) and nitrates (8.7%).

Table 1

Main patient characteristics according to age-defined and LVEF-defined groups.

	Age 1	Age 2	Age 3	Age 4	P	HF-REF	HF-BEF	HF-PEF	P
Women	26	24	30	44	< 0.001	24	35	45	< 0.001
Age (years)	50.2 ± 8.1	64.9 ± 3.0	75.3 ± 2.9	84.7 ± 3.3	< 0.001	65.6 ± 13.9	68.6 ± 14.2	72.3 ± 12.9	< 0.001
Ischaemic heart disease	38	48	53	50	< 0.001	52	45	34	< 0.001
DCM	37	28	19	12	< 0.001	32	20	11	< 0.001
Hypertension	36	56	65	67	< 0.001	49	57	71	< 0.001
Diabetes	25	39	43	27	< 0.001	31	35	38	0.006
Hypercholesterolaemia	46	58	58	43	< 0.001	53	52	47	ns
Smoking	55	44	34	22	< 0.001	45	39	31	< 0.001
NYHA III/IV class	13.5	17.0	22.7	36.5	< 0.001	22.5	19.1	17.2	0.010
LVEF (%)	36.0 ± 12.3	38.3 ± 13.9	41.2 ± 14.7	44.8 ± 14.7	< 0.001	31.1 ± 7.4	47.2 ± 2.3	62.0 ± 7.0	< 0.001
Atrial fibrillation	9.3	20.5	36.6	48.9	< 0.001	22.5	30.3	40.7	< 0.001
LBBB	21.8	27.4	26.6	23.7	ns	29.9	17.8	13.8	< 0.001
Biventricular pacemaker	5.7	11.7	11.2	5.2	< 0.001	11.5	5.4	1.7	< 0.001
ICD	16.8	19.1	11.7	1.7	< 0.001	18.9	3.9	1.3	< 0.001
BMI (%)	27.8 ± 6.2	28.1 ± 5.8	27.6 ± 5.5	25.3 ± 4.2	< 0.001	26.8 ± 5.4	27.8 ± 5.9	28.7 ± 6.3	< 0.001
Serum glucose (g/L)	1.07 ± 0.43	1.13 ± 0.42	1.13 ± 0.38	1.09 ± 0.42	0.028	1.07 ± 0.38	1.09 ± 0.38	1.16 ± 0.44	0.002
Serum creatinine (mg/L)	11.2 ± 5.3	13.1 ± 6.2	13.9 ± 5.8	14.4 ± 6.2	< 0.001	12.8 ±5.7	13.3 ± 7.2	13.3 ± 5.8	ns
Serum haemoglobin (g/L)	13.5 ± 1.9	13.1 ± 1.9	12.6 ± 1.8	12.4 ± 1.8	< 0.001	13.1 ± 1.9	12.7 ± 1.9	12.7 ± 1.9	< 0.001
BNP (pg/mL)	415 ± 582	574 ± 874	816 ± 1374	1158 ± 1404	< 0.001	764 ± 1259	565 ± 735	550 ± 708	0.003
NT-BNP (pg/mL)	2370 ± 3831	2875 ± 5149	4115 ± 5420	6047 ± 7003	< 0.001	4323 ± 6058	3803 ± 5697	2504 ± 3986	0.003
Follow-up deaths	9.4	14.5	25.9	37.4	< 0.001	20.3	22.8	20.5	ns

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Real-time three-dimensional foetal echocardiography using a new transabdominal xMATRIX array transducer
2. Predictors of 6-month poor clinical outcomes after transcatheter aortic valve implantation
3. “Squatting”
4. Giant thrombosed aneurysm of the right coronary artery

Stay updated, free articles. Join our Telegram channel

Join