Ambulatory patients treated at a multidisciplinary HF clinic between May 2006 and May 2013 were consecutively included in the study. The referral inclusion criteria and blood sample collection have been described elsewhere.13 In brief, patients were referred to the HF clinic by cardiology or internal medicine departments and, to a lesser extent, from the emergency or other hospital departments. The principal referral criterion was HF according to the European Society of Cardiology guidelines irrespective of etiology and at least 1 hospitalization for HF. Amyloid-beta 1-40 peptide and all other biomarkers were analyzed from the same blood sample, which was stored −80°C, without previous freeze-thaw cycles. All samples were obtained between 9:00 am and 12:00 pm. All participants provided written informed consent and the study was approved by the local ethics committee. All study procedures were conducted in accordance with the ethical standards outlined in the Helsinki Declaration of 1975, revised in 1983.

Cognition was evaluated using the Pfeiffer questionnaire, a short mental status survey, in 802 patients at baseline (within 6 months of blood sampling) and in 405 patients during follow-up (median 3.4 years [2.1-6.1]). The test was considered diagnostic of cognitive impairment if the score was > 3. The patients’ educational levels were considered for scoring (± 1 scoring points relative to educational level), as previously reported.14

All patients were followed up at regular intervals, with additional visits, as required, in cases of decompensation.13 Patients who did not attend regular visits were contacted by telephone. The primary outcomes were all-cause death, cardiovascular (CV) death, and HF-related death. A death was considered of CV origin if it was caused by HF (decompensated HF or treatment-resistant HF, in the absence of another cause); sudden death (unexpected death, witnessed or not, of a previously stable patient with no evidence of worsening HF or any other cause of death); acute myocardial infarction (death directly related with acute myocardial infarction, whether due to mechanic, hemodynamic, or arrhythmic complications); stroke (associated with recent acute neurologic deficit); procedural (postdiagnostic or posttherapeutic death); or other CV causes (eg, rupture of an aneurysm, peripheral ischemia, or aortic dissection). Fatal events were identified from the clinical records of patients with HF, hospital wards, the emergency room, primary care physicians, or by contacting the patient's relatives. Furthermore, data were verified from the databases of the Catalan and Spanish Health Systems. Adjudication of events was performed individually by 2 of the authors (M. Domingo and J. Lupón).13

Categorical variables are expressed as percentages. Continuous variables are expressed as means ± standard deviation or medians [25-75 percentiles] according to normal or nonnormal distributions, respectively. Normality was assessed with normal Q-Q plots. Amyloid-beta 1-40 peptide values were log-transformed for analyses. Differences in Aβ40 concentrations among sexes, cognitive function, and HF etiology were assessed with the Mann-Whitney U test or Kruskal-Wallis test. Statistical differences (P value for trend) in Aβ40 concentrations across age strata (≤ 50, > 50 to ≤ 60, > 60 to ≤ 70, > 70 to ≤ 80, and > 80 years), New York Heart Association (NYHA) functional class (I, II, and III-IV), estimated glomerular filtration rate (eGFR) by Chronic Kidney Disease Epidemiology Collaboration formula (≥ 60, 30 to < 60, and < 30mL/min/1.73 m2), and body mass index (< 20.5, 20.5 to < 25.5, 25.5 to < 30, and ≥ 30kg/m2) were determined using the Spearman test. Correlations with Aβ40 were assessed with the Pearson test or Spearman's r ho (Pfeiffer questionnaire) as needed. Statistical differences between groups among categorical variables were assessed using the chi-square test. A multivariable lineal regression analysis was also performed to ascertain independent associations with Aβ40 levels.

To determine the independent predictive value of Aβ40 in HF, univariate and multivariable Cox regression analyses (backward step method) were executed for the defined endpoints. In all analyses involving CV and HF-related death, a competing risk strategy using the Gray method was adopted, considering non-CV as the competing event for CV death and other CV death and non-CV death for HF-related death. In multivariable analyses, a model with relevant clinical predictors (age, sex, ischemic etiology of HF, left ventricular ejection fraction, NYHA functional class, presence of diabetes mellitus, hemoglobin, serum sodium, eGFR, and treatment with beta-blockers and angiotensin-converting enzyme inhibitor or angiotensin receptor blocker plus N-terminal pro-B-type natriuretic peptide) was developed and Aβ40 was included as an independent covariate. The proportionality and linearity assumptions were tested. To fulfill the assumption of linearity, the logarithmic functions of Aβ40 and N-terminal pro-B-type natriuretic peptide were used in the Cox models. Kaplan-Meier survival curves for all-cause death and cumulative incidence curves for CV and HF-related death were plotted. P values were obtained using log-rank and Gray method respectively

Comprehensive measurements of performance were used (discrimination and calibration) and the incremental value of adding Aβ40 to the model was explored by reclassification and discrimination improvement.

Statistical analyses were performed using the R (version 3.2.3) statistical package (Foundation for Statistical Computing, Vienna, Austria). A 2-sided P < .05 was considered significant.

Blood Aβ40 levels showed a weak correlation with Pfeiffer score both at baseline (Rho = 0.10, P = .004) and during follow-up (Rho = 0.10, P < .05). When the Pfeiffer questionnaire was used, cognitive impairment was present in 23 patients (2.9%) at baseline and in 9 patients (2.2%) during follow-up. No differences in blood Aβ40 levels were observed between patients with and without cognitive impairment (Pfeiffer questionnaire scoring adjusted for educational level) at baseline (P = .97) or during follow-up (P = .20) (Figure 3). If the patient's educational level was not considered for scoring, cognitive impairment was diagnosed in 29 patients (3.6%) at baseline and in 14 patients (3.6%) during follow-up. Again, no differences in blood Aβ40 levels were found between patients with and without cognitive impairment at baseline (P = .80) or during follow-up (P = .55).

Figure 3.

Box plots of Aβ40 concentration and cognitive impairment assessed by the Pfeiffer questionnaire. Aβ40, amyloid-beta 1-40 peptide.

(0.12MB).

Amyloid-beta 1-40 peptide levels were significantly higher in patients with all-cause death (median 586.2 pg/mL [405-851.9] vs 458.9 pg/mL [318.1-613.2], P < .001); CV death (median 605.2 pg/mL [418.3-892.7] vs 486.9 pg/mL [348-682], P < .001); and HF-related death (median 635.5 pg/mL [449.1-923.8] vs 496.7 pg/mL [353.7-718], P < .001). The diverging curves of survival and cumulative incidence of CV and HF-related death relative to Aβ40 median values are shown in Figure 4.

Figure 4.

Survival curves relative to median blood Aβ40 concentration. A: Kaplan-Meier survival curves for all-cause death. B: cumulative incidence of cardiovascular death, taking into account other causes of noncardiovascular of death as competitive risk event. C: cumulative incidence of heart failure-related death, taking into account other cardiovascular and noncardiovascular causes of death as competitive risk event. Aβ40, amyloid-beta 1-40 peptide; HR, hazard ratio.

(0.25MB).

In the univariate analyses, Aβ40 levels, as a continuous variable, were significantly associated with all-cause death (hazard ratio [HR], 1.51; 95% confidence interval [95%CI], 1.37-1.66; P < .001), CV death (HR, 1.42; 95%CI, 1.25-1.62; P < .001) and HF-related death (HR, 1.41; 95%CI, 1.17-1.69; P < .001). In multivariable analyses, the addition of Aβ40 to a model that included relevant clinical predictors plus N-terminal pro-B-type natriuretic peptide was significantly associated with all-cause death (HR, 1.22; 95%CI, 1.10-1.35; P < .001), and CV death (HR, 1.18; 95%CI, 1.03-1.36; P = .02), but not with HF-related death (HR, 1.13; 95%CI, 0.93-1.37; P = .22) (Table 2). After addition of the inflammatory biomarkers high-sensitivity C-reactive protein and ST2 in the model, Aβ40 remained also significantly associated with all-cause death (P = .001) and CV death (P = .04), but not with HF-related death (Table 1 of the supplementary material).

The model that included Aβ40 showed better global goodness-of-fit in all-cause (P < .001) and CV death (P = .02) and reclassification measured by continuous net reclassification index in all-cause (P = .001), CV (P < .001) and HF-related death (P = .01) (Table 2 of the supplementary material).

Table 3 of the supplementary material shows clinical characteristics and treatment relative to 2 periods of inclusion (2006-2009 vs 2010-2013). Although several differences were found between the 2 periods, there was no interaction between Aβ40 and the period of inclusion, and HR were very similar in separate multivariable analyses (HR, 1.29; [1.16–1.44] in the first period and HR, 1.32; [0.99–1.75] in the second).

Table 4 of the supplementary material and Table 5 of the supplementary material show multivariable survival analyses only in those patients with left ventricular ejection fraction < 50%. Amyloid-beta 1-40 peptide remained independently associated with all-cause death. The Figure of the supplementary material shows survival curves in that subgroup of patients relative to Aβ40 median values. Amyloid-beta 1-40 peptide blood levels above the median were significantly associated with higher all-cause mortality and higher cumulative incidence of CV and HF-related death than in the whole cohort.

Although the study included a real-life HF population with different etiologies, the prominence of men with HF and reduced ejection fraction suggests that our results need to be further validated and should not be extrapolated to the general population with HF. In the near future, with the likely widespread use of sacubitril/valsartan in patients with HF and reduced ejection fraction, the prognostic value of Aβ40 may change. Alternative screening tests for cognitive impairment in older people are available. Nevertheless, the Pfeiffer questionnaire remains valuable27,28 and, in this study, it served as a proof of principle to explore the absence of an association between cognitive impairment and serum Aβ40. Further studies are required to assess whether the addition of a senescence biomarker such as Aβ40, together with conventional HF biomarkers, may provide further insight in HF prognosis, particularly in a real-life scenario with increasingly aging and co-morbid HF patients.