The risk of cardiovascular death in elderly patients with possible heart failure. Results from a 6-year follow-up of a Swedish primary care population

Generalizability of Cardiovascular Disease Clinical Prediction Models: 158 Independent External Validations of 104 Unique Models

Background:

While clinical prediction models (CPMs) are used increasingly commonly to guide patient care, the performance and clinical utility of these CPMs in new patient cohorts is poorly understood.

Methods:

We performed 158 external validations of 104 unique CPMs across 3 domains of cardiovascular disease (primary prevention, acute coronary syndrome, and heart failure). Validations were performed in publicly available clinical trial cohorts and model performance was assessed using measures of discrimination, calibration, and net benefit. To explore potential reasons for poor model performance, CPM-clinical trial cohort pairs were stratified based on relatedness, a domain-specific set of characteristics to qualitatively grade the similarity of derivation and validation patient populations. We also examined the model-based C-statistic to assess whether changes in discrimination were because of differences in case-mix between the derivation and validation samples. The impact of model updating on model performance was also assessed.

Results:

Discrimination decreased significantly between model derivation (0.76 [interquartile range 0.73–0.78]) and validation (0.64 [interquartile range 0.60–0.67], P<0.001), but approximately half of this decrease was because of narrower case-mix in the validation samples. CPMs had better discrimination when tested in related compared with distantly related trial cohorts. Calibration slope was also significantly higher in related trial cohorts (0.77 [interquartile range, 0.59–0.90]) than distantly related cohorts (0.59 [interquartile range 0.43–0.73], P=0.001). When considering the full range of possible decision thresholds between half and twice the outcome incidence, 91% of models had a risk of harm (net benefit below default strategy) at some threshold; this risk could be reduced substantially via updating model intercept, calibration slope, or complete re-estimation.

Conclusions:

There are significant decreases in model performance when applying cardiovascular disease CPMs to new patient populations, resulting in substantial risk of harm. Model updating can mitigate these risks. Care should be taken when using CPMs to guide clinical decision-making.

Limited value of NT-proBNP as a prognostic marker of all-cause mortality in patients with heart failure with preserved and mid-range ejection fraction in primary care: A report from the swedish heart failure register

Aim: The prognostic value of natriuretic peptides in the management of heart failure (HF) patients with ejection fraction (EF) <40% is well established, but is less known for those with EF ≥40% managed in primary care (PC). Therefore, the aim of this study is to describe the prognostic significance of plasma NT-proBNP in such patients managed in PC.Subjects: We included 924 HF patients (48% women) with EF ≥40% and NT-proBNP registered in the Swedish Heart Failure Registry. Follow-up was 1100 ± 687 days.Results: One-, three- and five-year mortality rates were 8.1%, 23.9% and 44.7% in patients with EF 40-50% (HFmrEF) and 7.3%, 23.6% and 37.2% in patients with EF ≥50% (HFpEF) (p = 0.26). Patients with the highest mean values of NT-proBNP had the highest all-cause mortality but wide standard deviations (SDs). In univariate regression analysis, there was an association only between NT-proBNP quartiles and all-cause mortality. In HFmrEF patients, hazard ratio (HR) was 1.96 (95% CI 1.60-2.39) p < 0.0001) and in HFpEF patients, HR was 1.72 (95% CI 1.49-1.98) p < 0.0001). In a multivariate Cox proportional hazard regression analysis, adjusted for age, NYHA class, atrial fibrillation and GFR class, this association remained regarding NT-proBNP quartiles [HR 1.83 (95% CI 1.38-2.44), p < 0.0001] and [HR 1.48 (95% CI 1.16-1.90), p = 0.0001], HFmrEF and HFpEF, respectively.Conclusion: NT-proBNP has a prognostic value in patients with HF and EF ≥40% managed in PC. However, its clinical utility is limited due to high SDs and the fact that it is not independent in this population which is characterized by high age and much comorbidity.Key pointsIt is uncertain whether NT-proBNP predicts risk in heart failure with preserved ejection fraction (EF > 40%, HFpEF) managed in primary care.We show that high NT-proBNP predicts increased all-cause mortality in HFpEF-patients managed in primary care.The clinical use is however limited due to large standard deviations, many co-morbidities and high age.Many of these co-morbidities contribute to all-cause mortality and management of these patients should also focus on these co-morbidities.

Plasma chromogranin A is a marker of death in elderly patients presenting with symptoms of heart failure

Cardiovascular risk assessment remains difficult in elderly patients. We examined whether chromogranin A (CgA) measurement in plasma may be valuable in assessing risk of death in elderly patients with symptoms of heart failure in a primary care setting. A total of 470 patients (mean age 73 years) were followed for 10 years. For CgA plasma measurement, we used a two-step method including a screening test and a confirmative test with plasma pre-treatment with trypsin. Cox multivariable proportional regression and receiver-operating curve (ROC) analyses were used to assess mortality risk. Assessment of cardiovascular mortality during the first 3 years of observation showed that CgA measurement contained useful information with a hazard ratio (HR) of 5.4 (95% CI 1.7-16.4) (CgA confirm). In a multivariate setting, the corresponding HR was 5.9 (95% CI 1.8-19.1). WHEN ADDING N-TERMINAL PROBNP (NT-PROBNP) TO THE MODEL, CGA CONFIRM STILL POSSESSED PROGNOSTIC INFORMATION (HR: 6.1; 95% CI 1.8-20.7). The result for predicting all-cause mortality displayed the same pattern. ROC analyses in comparison to NT-proBNP to identify patients on top of clinical variables at risk of cardiovascular death within 5 years of follow-up showed significant additive value of CgA confirm measurements compared with NT-proBNP and clinical variables. CgA measurement in the plasma of elderly patients with symptoms of heart failure can identify those at increased risk of short- and long-term mortality.

Adiponectin through its biphasic serum level is a useful biomarker during transition from diastolic dysfunction to systolic dysfunction - an experimental study

Background

Adiponectin is reported to relate with cardiovascular diseases, we sought to examine whether adiponectin is associated with disease progression of heart failure from hypertension in rats in comparison with other known biomarkers and echocardiographic parameters. Spontaneously hypertensive rats (SHR, n = 35), aged 1 month, were used and followed up to 18 months. High frequency echocardiography was performed both at baseline and every 3 months thereafter. Moreover, serum levels of N-terminal pro-natriuretic peptide (NT-proBNP) and interleukin-6 (IL-6) as well as serum level and tissue expression of adiponectin were determined at the same time as echocardiography.

Results

The results clearly demonstrated time-dependent progression of hypertension and heart dysfunction as evidenced by gradually increased left ventricular mass index, NT-proBNP, IL-6 as well as gradually decreased cardiac function as assessed by echocardiography. Meanwhile, tissue and serum adiponectin decreased from 3 months and reached plateau until 12 months in parallel with decreasing of cardiac diastolic function. Thereafter, adiponectin levels increased prior to occurrence of systolic dysfunction. Adiponectin concentration is inversely related with NT-proBNP, IL-6 and E/E′ (correlation coefficient (r) = −0.756 for NT-proBNP, p < 0.001, -0.635 for IL-6, p = 0.002, and −0.626 for E/E′, p = 0.002, respectively) while positively correlated with E/A and E′/A′ (r = 0.683 for E/A, p = 0.001, 0.671 for E′/A′, p = 0.001, respectively). No difference for adiponectin distribution among visceral adipose tissues was found.

Conclusion

Adiponectin through its biphasic serum level is a useful biomarker during transition from diastolic dysfunction to systolic dysfunction.

N-Terminal Pro–Atrial Natriuretic Peptide Measurement in Plasma Suggests Covalent Modification

BACKGROUND

The N-terminal fragment of cardiac-derived pro–B-type natriuretic peptide is a glycosylated polypeptide. It is unknown whether N-terminal pro–atrial natriuretic peptide (proANP) fragments are also covalently modified. We therefore evaluated the clinical performance of 2 distinctly different proANP assays on clinical outcome.

METHODS

We examined 474 elderly patients with symptoms of heart failure presenting in a primary healthcare setting. Samples were analyzed with an automated immunoluminometric midregion proANP (MR-proANP) assay and a new processing-independent assay (PIA) developed in our laboratory. The results were compared with Bland–Altman plots, and clinical performance was assessed by generating ROC curves for different clinical outcomes.

RESULTS

Despite linear regression results indicating a good correlation (r = 0.85; P &lt; 0.0001), the PIA measured considerably more proANP than the MR-proANP assay (mean difference, 663 pmol/L; SD, 478 pmol/L). In contrast, the clinical performances of the 2 assays [as assessed by the area under the ROC curve (AUC)] in detecting left ventricular dysfunction were similar [proANP PIA, 0.71 (95% CI, 0.63–0.79); MR-proANP assay, 0.74 (95% CI, 0.66–0.81); P = 0.32]. The prognostic ability to report cardiovascular mortality during a 10-year follow-up revealed AUC values of 0.66 (95% CI, 0.60–0.71) for the proANP PIA and 0.69 (95% CI, 0.63–0.74) for the MR-proANP assay (P = 0.08, for comparing the 2 assays).

CONCLUSIONS

Our data suggest that N-terminal proANP fragments in patient plasma differ from the calibrator peptides used but that the difference does not affect ROC curves in an elderly cohort of patients with mild to moderate heart failure. We suggest that human N-terminal proANP fragments can be covalently modified.

Baseline low-density lipoprotein cholesterol levels and outcome in patients with heart failure

The incidence of heart failure (HF) is constantly increasing in the Western world. Treatment with statins is well established for the primary and secondary prevention of cardiac events by lowering low-density lipoprotein (LDL) cholesterol levels. There are conflicting reports on the role of LDL cholesterol as an adverse prognostic predictor in patients with advanced HF. The aim of this study was to investigate the association between LDL cholesterol levels and clinical outcomes in 297 patients with severe HF (average New York Heart Association class 2.8). The mean follow-up period was 3.7 years (range 8 months to 11.5 years), and 37% of the patients died during follow-up. The mean time to first hospital admission for HF was 25 +/- 17 months. The study group was divided according to plasma LDL level < or =89, >89 to < or =115, >115 mg/dl. Patients with the highest baseline LDL cholesterol levels had significantly improved outcomes, whereas those with the lowest LDL cholesterol levels had the highest mortality. When analyzed with respect to statin use, it emerged that the negative association between LDL cholesterol level and mortality was present only in the patients with HF who were treated with statins. In conclusion, lower LDL cholesterol levels appear to predict less favorable outcomes in patients with HF, particularly those taking statins, raising questions about the need for aggressive LDL cholesterol-lowering strategy in patients with HF, regardless of its cause.

Usefulness of peak exercise oxygen consumption and the heart failure survival score to predict survival in patients >65 years of age with heart failure

Peak exercise oxygen consumption (Vo(2)) and the Heart Failure (HF) Survival Score (HFSS) were developed in middle-aged patient cohorts referred for heart transplantation with HF. The prognostic value of Vo(2) in patients >65 years has not been well studied. Accordingly, the prognostic value of peak Vo(2) was evaluated in these patients with HF. A retrospective analysis of 396 patients with HF >65 years with cardiopulmonary exercise testing was performed. Peak Vo(2) and components of the HFSS (presence of coronary artery disease, left ventricular ejection fraction, heart rate, mean arterial blood pressure, presence of intraventricular conduction defects, and serum sodium) were collected. Follow-up averaged 1,038 +/- 983 days. Outcome events were defined as death, implantation of a left ventricular assist device, or urgent transplantation. Patients were divided into risk strata for peak Vo(2) and HFSS based on previous cut-off points. Survival curves were derived using Kaplan-Meier analysis and compared using log-rank analysis. Survival differed markedly by Vo(2) stratum (p <0.0001), with significantly better survival rates for the low- (>14 ml/kg/min) versus medium- (10 to 14 ml/kg/min), low- versus high- (<10 ml/kg/min), and medium- versus high-risk strata (all p <0.05). Survival also differed markedly by HFSS stratum (p <0.0001), with significantly better survival rates for the low- (> or =8.10) versus medium- (7.20 to 8.09), low- versus high- (< or =7.19), and medium- versus high-risk strata (all p <0.0001). In conclusion, peak Vo(2) and the HFSS were both excellent parameters to predict survival in patients >65 years with HF.

Cystatin C and NT-proBNP, a powerful combination of biomarkers for predicting cardiovascular mortality in elderly patients with heart failure: results from a 10-year study in primary care

AIMS

Heart failure (HF) is common among the elderly patients. It is essential to identify those at high risk in order to optimize the use of resources. We aimed to evaluate whether a combination of two biomarkers might give better prognostic information about the risk of cardiovascular (CV) mortality in patients with symptoms associated with HF, compared with only one biomarker.

METHODS AND RESULTS

Four hundred and sixty-four primary health-care patients (mean age 73 years, range 65-87) with symptoms of HF were examined. All patients were evaluated using Doppler echocardiography and blood samples, including measurement of cystatin C and NT-proBNP. The patients were followed over a 10-year period. Patients with serum cystatin C levels within the highest quartile had almost three times the risk (HR: 2.92; 95% CI: 1.23-4.90) of CV mortality compared with those patients who had levels within the first, second, or third quartiles. If, at the same time, the patient had a plasma concentration of NT-proBNP within the highest quartile, the risk increased to >13 times (HR: 13.61; 95% CI: 2.56-72.24) during 10 years of follow-up or >17 times (HR: 17.04; 95% CI: 1.80-163.39) after 5 years of follow-up.

CONCLUSION

Combined analysis of cystatin C and NT-proBNP could provide important prognostic information among elderly patients in the community with symptoms of HF.

Impact of Epitope Specificity and Precursor Maturation in Pro-B–Type Natriuretic Peptide Measurement

Background: Cardiac-derived natriuretic peptides are sensitive plasma markers of cardiac dysfunction. Recent reports have disclosed a more complex molecular heterogeneity of B-type natriuretic peptide precursor (proBNP)-derived peptides than previously suggested. In this study, we examined the impact of epitope specificity and precursor maturation on plasma measurement of proBNP-derived peptides.

Methods: We compared 2 assays, N-terminal proBNP and proBNP 1–76, in a randomly collected set of human plasma specimens (n = 370). Additionally, we evaluated the clinical performance of 4 assays with different epitope specificities in a cohort of elderly patients presenting with symptoms associated with heart failure (n = 415).

Results: Comparison of N-terminal proBNP with proBNP 1–76 measurement in plasma revealed a high correlation on regression analysis (r2 = 0.91, P &lt; 0.0001). Nevertheless, the proBNP 1–76 assay measured lower concentrations in the high range than the N-terminal proBNP assay. Correlations between assay measurements in a clinical setting were comparable for all the assays (r2 approximately 0.57–0.83), and ROC analyses revealed area-under-the-curve values ranging between 0.77 and 0.81 for identifying reduced left ventricular ejection fraction. In parallel, all assays displayed comparable abilities in predicting long-term mortality.

Conclusions: Our results reveal marked assay differences in analytical assay comparison, contrasting the overall comparable clinical performance in cardiovascular diagnostics or prognosis in the elderly.

Quality-adjusted life year weights among elderly patients with heart failure

BACKGROUND

When assessing health-related quality of life (HRQoL) in elderly patients with heart failure (HF), the process of obtaining quality-adjusted life year (QALY) weights is generally complicated and time-consuming.

AIM

To evaluate whether information regarding HRQoL and QALY weights can be derived directly from the established and widely used New York Heart Association (NYHA) functional classification system.

METHODS

NYHA functional status was assessed independently both by the individual patients and by the examining cardiologist in 323 elderly patients with symptoms of HF recruited from primary care. HRQoL was evaluated using the SF-36 questionnaire and a time trade-off (TTO) scenario. The TTO technique generates direct QALY weights.

RESULTS

Both the TTO technique and SF-36 values demonstrated a statistically significant correlation with NYHA functional status. The TTO values also correlated with all SF-36 dimensions. Increasing impairment was associated with statistically significant drops in both SF-36 values and TTO-based QALY weights. For patients in NYHA classes I-IV the QALY weights were 0.77, 0.68, 0.61, and 0.50, respectively. Thus in elderly patients, symptoms of HF have a major impact on perceived quality of life.

CONCLUSION

The results of the present study show that QALY weights, an important instrument in the health economic evaluation of treatment strategies, can be derived directly from NYHA classification in elderly HF patients.

Can NT-proBNP predict risk of cardiovascular mortality within 10 years? Results from an epidemiological study of elderly patients with symptoms of heart failure

BACKGROUND

Heart failure has a serious prognosis. However, among elderly patients the panorama of concomitant diseases makes it difficult to implement the results from epidemiological studies. The aim of this study was to evaluate the influence of different clinical variables on cardiovascular mortality during a long-term follow-up.

METHODS AND RESULTS

In all, 474 elderly patients (age 65-82 years) in primary health care were evaluated and followed during a 10 year period. All patients had symptoms associated with heart failure and were examined by a cardiologist. Blood samples including NT-proBNP were analyzed, and ECG and Doppler echocardiography were assessed. Both the systolic and diastolic function was evaluated. Functional capacity was evaluated according to the NYHA classification. During the 10 years of follow-up those with the highest quartile of plasma concentration of NT-proBNP had almost four times increased risk of cardiovascular mortality. Impaired systolic function, diabetes and reduced functional capacity were all markers of increased risk of cardiovascular mortality. All variables were also evaluated after 5 years, with higher risk ratios for a majority of variables.

CONCLUSION

In this study 474 patients with symptoms of heart failure were followed during 10 years. High plasma concentration of NT-proBNP could predict almost four times increased risk of cardiovascular mortality up to 10 years. Also, impaired cardiac function according to echocardiography, and reduced functional capacity as well as diabetes all had influence on risk of cardiovascular mortality up to 10 years.

Reference intervals and decision limits for B-type natriuretic peptide (BNP) and its precursor (Nt-proBNP) in the elderly

BACKGROUND

Elderly patients have the highest prevalence of heart failure (HF). The aims of the study were to establish a reference interval for B-type natriuretic peptide (BNP) and (Nt-proBNP) in elderly people, and to identify clinically relevant decision limits based on long-term outcome.

METHODS

Plasma concentrations of BNP and Nt-proBNP were measured from two elderly populations: 218 healthy subjects (mean age 73 years, population I), and 474 patients (mean age 73 years, population II) with symptoms associated with HF. Study population II was followed for 6 years with registration of all cardiovascular mortality.

RESULTS

An association between both BNP and Nt-proBNP concentrations and age was found. The upper limit for the reference intervals in the healthy elderly (population I) was: BNP< or =28 pmol/L (< or =97 ng/L), and Nt-proBNP< or =64 pmol/L (< or =540 ng/L). Based on cardiovascular mortality, decision limits for BNP (approximately 50 pmol/L, approximately 170 ng/L) and Nt-proBNP ( approximately 200 pmol/L, approximately 1700 ng/L) (population II) were determined.

CONCLUSIONS

Besides establishing reference intervals for BNP and Nt-proBNP in an elderly population, a higher clinically relevant decision limit for BNP and Nt-proBNP was identified, indicating additive prognostic information of the peptides on top of measurements by echocardiography. Therefore, both reference intervals and decision limits should be included in clinical practice.

Association between cardiac disorders and a decades-previous history of diphtheria

A long, prior history of diphtheria is common among middle-aged and elder European adults. The aim of the present study was to determine whether the risk of reduced ventricular function and impaired intraventricular conduction is increased in individuals with a history of diphtheria. A study population of 2,480 subjects (1,222 women) aged 45 years or older who were recruited for the Study of Health in Pomerania were available for the present analyses. Left ventricular function was assessed by echocardiography. Intraventricular conduction blocks were diagnosed using electrocardiograms. Multivariable analyses revealed that individuals with a history of diphtheria had neither an increased odds for reduced fractional shortening (OR 1.21, 95% CI 0.69-2.11; p=0.51) nor an increased odds for intraventricular conduction blocks (OR 0.90, 95% CI 0.55-1.46; p=0.67). However, regression models revealed two-way interactions between the exposure variable and hypertension with respect to both endpoints. A history of diphtheria increased the odds for both endpoints in normotensive but not in hypertensive individuals. The findings show that a history of diphtheria several decades previously in a patient is a risk marker for reduced cardiac function and impaired intraventricular conduction in individuals at low risk for these disorders.