Resting heart rate and incident atrial fibrillation: A stratified Mendelian randomization in the AFGen consortium

Background

Both elevated and low resting heart rates are associated with atrial fibrillation (AF), suggesting a U-shaped relationship. However, evidence for a U-shaped causal association between genetically-determined resting heart rate and incident AF is limited. We investigated potential directional changes of the causal association between genetically-determined resting heart rate and incident AF.

Method and results

Seven cohorts of the AFGen consortium contributed data to this meta-analysis. All participants were of European ancestry with known AF status, genotype information, and a heart rate measurement from a baseline electrocardiogram (ECG). Three strata of instrumental variable-free resting heart rate were used to assess possible non-linear associations between genetically-determined resting heart rate and the logarithm of the incident AF hazard rate: <65; 65–75; and >75 beats per minute (bpm). Mendelian randomization analyses using a weighted resting heart rate polygenic risk score were performed for each stratum.

We studied 38,981 individuals (mean age 59±10 years, 54% women) with a mean resting heart rate of 67±11 bpm. During a mean follow-up of 13±5 years, 4,779 (12%) individuals developed AF. A U-shaped association between the resting heart rate and the incident AF-hazard ratio was observed. Genetically-determined resting heart rate was inversely associated with incident AF for instrumental variable-free resting heart rates below 65 bpm (hazard ratio for genetically-determined resting heart rate, 0.96; 95% confidence interval, 0.94–0.99; p = 0.01). Genetically-determined resting heart rate was not associated with incident AF in the other two strata.

Conclusions

For resting heart rates below 65 bpm, our results support an inverse causal association between genetically-determined resting heart rate and incident AF.

The 3S-AF scheme, rather than the 4S-AF scheme, predicts progression in patients with paroxysmal atrial fibrillation: data from RACE V study

Funding Acknowledgements

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): support from the Netherlands Cardiovascular Research Initiative: an initiative with support of the Dutch Heart Foundation, CVON 2014-9: Reappraisal of Atrial Fibrillation: interaction between hyperCoagulability, Electrical remodelling, and Vascular destabilisation in the progression of AF (RACE V).

Purpose

To assess whether the 4S-AF scheme predicts AF progression, cardiovascular hospitalizations and mortality in patients with self-terminating paroxysmal AF.

Methods

We analysed well-phenotyped patients with paroxysmal AF from the Reappraisal of Atrial Fibrillation: Interaction between HyperCoagulability, Electrical remodelling, and Vascular Destabilisation in the Progression of AF (RACE V study). From the 417 patients included in RACE V, 341 (82%) had echocardiography available. Patient had continuous monitoring with implantable loop recorders or pacemakers. Primary endpoint of RACE V was AF progression, defined as (1) progression to persistent or permanent AF, or (2) progression of PAF with &gt;3% burden increase. Median follow-up was 2.2 (1.6-2.8) years. Patients were given a score based on the components of the 4S-AF scheme (St, stroke=1; Sy, symptoms=2; Sb, Severity of burden=2; Su, Substrate=5) to a total maximum of 10 points (table 1). Left atrial fibrosis was not evaluated in our patients and therefore not included into the score. A score of zero (0) in the AF burden domain was given to all patients due to the presence of paroxysmal AF in all. A modified 4S-AF scheme was designed by eliminating the symptom domain, resulting in a 3S-AF scheme. Logistic regression was performed to assess AF progression and the composite endpoint of cardiovascular hospitalizations and mortality, C-statistic to assess prediction of the score, for both using the 4S-AF and the modified 3S-AF scheme.

Results

Mean age was 65 (IQR 58-71) years, 149 (44%) were women, 103 (48%) had heart failure (HFrEF 6 [2%]; HFpEF 97 [46%]), 276 (81%) had hypertension, 38(11%) had coronary artery disease and 162(48%) atherosclerosis (Table 2, Panel A). Based on the 4S-AF scheme, patients had an average score of 4.5±1.3, the majority had a score under 5 (n=272, 80%), 20% of the score was explained by the S1 domain (stroke), 16% of the score was explained by the Sy domain (symptoms), and 64% of the score was explained by the Su domain (substrate). The score points from the 4S-AF scheme did not predict the risk of AF progression (OR 1.08 95%CI 0.84 – 1.39, C-statistic 0.53) nor the composite endpoint (OR 0.79 95%CI 0.53 – 1.20, C-statistic 0.42, Table 2, Panel B). However, when excluding the Sy domain (symptoms) from the scheme, the 3S-AF scheme, it predicted the risk of progression (OR 1.54 95%CI 1.12 – 2.18, C-statistic 0.61, Table 2, Panel B).

Conclusion

In paroxysmal AF patients the 4S-AF scheme does not predict AF progression nor the composite endpoint cardiovascular hospitalizations and mortality. Although symptoms are important for choosing the treatment strategy, they may be less relevant to determine AF progression, cardiovascular hospitalization and mortality. To assess progression, the 3S-AF scheme may be more appropriate.

Risk factors, subsequent disease onset and prognostic impact of myocardial infarction and atrial fibrillation

Background

Myocardial infarction (MI) is a known risk factor for incident atrial fibrillation (AF), while AF frequently complicates acute MI. Although both diseases share common cardiovascular risk factors, the direction and strength of the association of the risk factors with disease onset, subsequent disease incidence and mortality are not completely understood.

Purpose

Our goal was to define the temporal relationship of MI and AF and the association of cardiovascular risk factors with disease incidence in order to determine whether common clinical risk factors show different associations with incident MI or AF. We further aimed to investigate predictors of subsequent disease onset and the impact of subsequent disease diagnosis on mortality.

Methods

In pooled multivariable Cox regression analyses we examined temporal relations of disease onset and identified predictors of MI, AF and subsequent all-cause mortality in 108,363 individuals (median age 46.0 years, 48.2% men) free of MI and AF at baseline from six European population-based cohorts.

Results

Over a maximum follow-up of 10.0 years 3558 (3.3%) individuals were diagnosed exclusively with MI, 1922 (1.8%) with AF but no MI, and 491 (0.5%) individuals developed both MI and AF. Association of male sex, systolic blood pressure, antihypertensive treatment and diabetes mellitus appeared to be stronger with incident MI than with AF, whereas increasing age and body mass index showed a higher risk for incident AF. Total cholesterol and daily smoking were significantly related to incident MI but not AF. The combined population attributable fraction of the cardiovascular risk factors was over 70% for incident MI, whereas it was only about one quarter for incident AF. Subsequent MI after incident AF (hazard ratio1.68, 95% CI 1.03–2.74) and subsequent AF after MI (hazard ratio 1.75, 95% CI 1.31–2.34) both significantly increased overall mortality risk.

Conclusions

Subsequent diagnosis of MI and AF was associated with a significant increase in mortality, irrespective of the first event. We found different associations of common cardiovascular risk factors with incident MI and AF indicating distinct pathophysiological pathways in disease development.

Funding Acknowledgement

Type of funding sources: Public grant(s) – EU funding. Main funding source(s): European Union's Horizon 2020 research and innovation programme (grant agreement No 847770, AFFECT-EU) European Union's Horizon 2020 research and innovation programme (grant agreement No 648131)

Risk prediction of atrial fibrillation and its complications in the community using high-sensitivity cardiac troponin I: results from the BiomarCaRE Consortium

Aims

Atrial fibrillation (AF) is becoming increasingly common and is associated with serious complications. Traditional cardiovascular risk factors (CVRF) do not explain all AF cases. Blood-based biomarkers reflecting cardiac injury may help close this gap. High-sensitivity troponin I (hsTnI) has emerged as a potential predictor.

Methods

We investigated the predictive ability of hsTnI for incident AF in 29,227 participants (median age 52.6 years, 51.2% men) across four different European community cohorts of the Biomarkers for Cardiovascular Risk Assessment in Europe (BiomarCaRE) consortium in comparison to CVRF and established biomarkers (high-sensitive C-reactive protein (hsCRP), N-terminal pro B-type natriuretic peptide (NT-proBNP)).

Results

During a median follow-up of 13.8 (lower and upper quartiles 4.5, 21.3) years, 1,509 (5.2%) participants developed AF. Those in the highest fourth of hsTnI values at baseline (≥5.1 ng/L) had a 2.71-fold (95% confidence interval (CI) 2.31, 3.17; P&lt;0.01) risk for developing AF compared to those in the lowest fourth (≤2.1 ng/L). In multivariable-adjusted Cox proportional hazard models no statistically significant association was seen between hsTnI and AF, whereas NT-proBNP (hazard ratio (HR) per two-fold increase in NT-proBNP 1.64; 95% CI 1.56, 1.72; P&lt;0.001) as well as hsCRP (HR ratio per two-fold increase in hsCRP 1.05; 95% CI 1.01, 1.10; P=0.01) were statistically significantly related to incident AF. Inclusion of hsTnI did not improve model discrimination over CVRFs (C-index CVRF 0.7914 vs. C-index CVRF, hsTnI 0.7927; 95% CI −0.0004, 0.0031; P=0.130). Higher hsTnI concentrations were associated with AF complications such as stroke (HR 1.25; 95% CI 1.03, 1.51; P=0.02), heart failure (HR 1.27; 95% CI 1.12, 1.44; P&lt;0.001) and cardiovascular events (HR 1.24; 95% CI 1.08, 1.42; P&lt;0.001) as well as overall mortality (HR 1.15; 95% CI 1.05, 1.25; P&lt;0.001) in those who were diagnosed with AF.

Conclusion

hsTnI as a biomarker of myocardial injury does not improve prediction of AF incidence beyond classical CVRFs. However, it is associated with AF complications and mortality after AF onset probably reflecting underlying subclinical cardiovascular impairment.

Funding Acknowledgement

Type of funding sources: Public grant(s) – EU funding. Main funding source(s): European Union Seventh Framework Programme (FP7/2007-2013

Atrial fibrillation detected at screening is not a benign condition - a comparison of clinical outcomes in screen-detected vs. hospital-detected atrial fibrillation

Funding Acknowledgements

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Dutch Heart Foundation

Background

Screening for atrial fibrillation (AF) improves detection of AF. However, it is unknown whether AF detected at screening carries risks similar to clinically detected AF, and if it should be treated similarly.

Purpose

We aimed to compare clinical outcomes in individuals with screen-detected vs. hospital-detected incident AF.

Methods

We studied 8265 individuals (mean age 49 ± 13 years, 50% women) without prevalent AF from the population-based PREVEND (Prevention of Renal and Vascular End-Stage Disease) cohort study. By design, 70% of PREVEND participants had urinary albumin concentration ≥10 mg/l. AF was considered screen-detected when first detected on a 12-lead electrocardiogram (ECG) during one of the PREVEND study visits, and hospital-detected when first detected on a hospital ECG. Using Cox regression models with screen-detected and hospital-detected AF as time-varying covariates, we studied the association of screen-detected vs. hospital-detected AF with mortality, incident heart failure (HF), and incident cardiovascular (CV) events.

Results

During a mean follow-up of 9.7 years, 265 participants (3.2%) developed incident AF (mean age 62 ± 9 years, 30% women, 65% hypertension, 23% obesity, 9% diabetes, 15% history of myocardial infarction, 3% history of stroke, 2% prevalent HF). Of all incident AF cases, 60 (23%) were screen-detected and 205 (77%) hospital-detected. Baseline characteristics were generally comparable between participants with screen-detected and hospital-detected AF. A larger proportion of incident AF was screen-detected in men (26%) compared to women (15%). In univariabe analysis, both screen-detected and hospital-detected AF were strongly associated with death, incident HF, and incident CV events. After multivariable adjustment, hospital-detected AF was significantly associated with death (HR 2.95, 95% CI 2.18-4.00), incident HF (HR 3.98, 95% CI 2.49-6.34), and incident CV events (HR 1.92, 95% CI 1.21-3.06). Screen-detected AF was significantly associated with death (HR 2.21, 95% CI 1.09-4.47) and incident HF (HR 4.90, 95% CI 2.28-10.57), but not with incident CV events (HR 1.12, 95% CI 0.46-2.71).

Conclusions

In a population-based cohort enriched for microalbuminuria, almost a quarter of incident AF cases was first detected through ECG screening. Compared to hospital-detected AF, screen-detected AF was similarly associated with adverse outcomes. Although randomised trials are needed, this study highlights that AF screening may help decrease the general burden of CV disease.

Multi-omics approach to post-CABG renal impairment

Background

Renal impairment is a common complication after CABG (coronary artery bypass graft) surgery associated with an adverse outcome.

Purpose

To further characterize the molecular framework of the disease through omics analyses.

Methods

In N=165 CABG patients we performed multi-omics-analyses in preoperatively collected blood and tissue samples as well as 991 creatinine measurements. We used multivariable mixed-model regression analyses to analyse post-operative creatinine increase and to find common genetic polymorphisms, transcripts, metabolites and/or proteins associated with changes in postoperative creatinine increase. Multiple testing was accounted for by setting a 5%-limit on the false discovery rate (FDR) using the Benjamini-Hochberg procedure.

Results

Post-operative increase of log transformed creatinine was 0.035 (8%); 95% confidence interval (CI) 0.025, 0.045; P&lt;0.001. We identified 55 gene expressions and two proteins associated with post-CABG renal impairment. On the metabolomic and single nucleotide point mutation (SNP) level, no relevant targets were found. The three most important identified gene expressions were MIR3202.1 (beta of log transformed creatinine increase per standard deviation gene expression increase −0.034; 95% CI: −0.048, 0.020; P&lt;0.001), LOC105374386 (−0.032; 95% CI: −0.046, 0.019; P&lt;0.001) and maternal embryonic leucine zipper kinase (MELK) (−0.022; 95% CI: −0.032, 0.013; P&lt;0.001). Expression of all three was associated with a lower risk of post-CABG renal impairment. The same applies to the identified protein CAPRIN2 (−0.042; 95% CI: −0.062, 0.022; P&lt;0.001), while expression of the protein TUBB6 was associated with a higher risk (0.033; 95% CI: 0.017, 0.048; P&lt;0.001).

Conclusions

In an integrated approach we identified omics-biomarkers for the prediction of renal impairment after CABG surgery. The underlying pathophysiological associations of these genes and proteins are not fully understood. MELK might be an interesting target for further investigations, as it plays a prominent role in cell cycle control, cell proliferation, apoptosis, cell migration and cell renewal. Our results may help to better identify individuals at risk and lay the methodological groundwork for further omics analyses.

Funding Acknowledgement

Type of funding source: None

Atrial fibrillation risk factor burden and disease onset across age decades

Background

Although a number of risk factors have been associated with the progression of atrial fibrillation (AF), there is limited knowledge of their relevance for AF in relation to age.

Purpose

We examined whether the association between modifiable risk factors and AF differed between age decades.

Methods

Data were derived from five European cohorts from Denmark, Finland, Italy, Sweden, and Norway. In total, 66 951 individuals (49.1% men) aged ≥40 years (mean baseline age 53.5 years) and without prevalent AF were followed for incident AF, with the follow-up truncated at 10 years. Data on risk factors (body mass index [BMI], hypertension [systolic blood pressure ≥140 mmHg and/or use of antihypertensive medication], diabetes mellitus, myocardial infarction [MI] event before baseline examinations, daily smoking, and alcohol consumption) were available from the baseline examinations. Stratification into age decades was based on age at baseline examination. Furthermore, the participants were followed for events of stroke or mortality after AF diagnosis. Mortality, stroke, and AF outcomes were derived from national registers and hospital discharge registers. All analyses were adjusted for AF risk factors.

Results

The incidence of AF increased from 0.9 per 1000 person-years at the age of 40 to &lt;50, to 17.7 at the age of ≥70 years. Multivariable-adjusted Cox models showed that BMI, hypertension, alcohol consumption, and history of MI were associated with increased risk of AF across age decades (p&lt;0.05). Of these, the risk of AF associated with BMI and an MI event before baseline examinations differed across age decades. For each 5 units increase in BMI, risk of AF increased with 40% (95% confidence interval 17–68%) at the age of 40 to &lt;50, falling to 17% (6–29%) at the age of ≥70 years (p=0.08 for difference between age decades 40 to &lt;50 and ≥70). Participants with a history of MI showed decreased risk of AF with ageing, from a hazard ratio (HR) of 5.53 (2.85–10.73) in the 40 to &lt;50 age group to a HR of 1.41 (1.11–1.79) at the age of ≥70 (p&lt;0.001). Daily smoking and prevalent diabetes mellitus were in general not associated with AF. The multivariable-adjusted associations between new-onset AF and the succeeding risk of stroke and mortality increased with age, showing a 1.6 to 2.6-fold increase in risk of death at ages ≥60 years and two-fold increased risk of stroke in participants aged ≥70 years (p≤0.001).

Conclusion

The relative importance of modifiable risk factors on incident AF do not vary across age decades, with a few exceptions; BMI and a history of MI were stronger risk factors for AF at younger ages. Thus, preventive measures should target risk factors rigorously, in particular obesity. New-onset AF was associated with increased risk of stroke and mortality only at older ages, emphasizing the importance of adequate patient management in the older and oldest old.

Funding Acknowledgement

Type of funding source: None

Assessment of causality of natriuretic peptides and atrial fibrillation and heart failure – a Mendelian randomization study in the FINRISK cohort

Background

Natriuretic peptides are extensively studied biomarkers for atrial fibrillation (AF) and heart failure (HF). Their role in the pathogenesis of both diseases is not entirely understood and in previous studies several single nucleotide polymorphisms (SNPs) at the NPPA-NPPB locus associated with natriuretic peptides have been identified.

Purpose

We investigated whether a causal relationship exists between natriuretic peptides and AF as well as HF using a Mendelian randomization approach.

Methods

N-terminal pro B-type natriuretic peptide (NT-proBNP) (N=6669), B-type natriuretic peptide (BNP) (N=6674) and mid-regional pro atrial natriuretic peptide (MR-proANP) (N=6813) were measured in the FINRISK 1997 cohort. Thirty common SNPs related to NT-proBNP, BNP and MR-proANP were selected from prior studies. We performed six Mendelian randomizations for all three natriuretic peptide biomarkers and for both outcomes, AF and HF separately. Polygenic risk scores (PRS) based on multiple SNPs were used as the genetic instrumental variable in Mendelian randomizations.

Results

PRS were significantly associated with the three natriuretic peptides. PRS were not significantly associated with incident AF nor HF. Most cardiovascular risk factors showed significant confounding percentages, but no association with PRS. A causal relation, other than a weak one, is unlikely.

Conclusion

In our Mendelian randomization approach, based on common genetic variation at the NPPA-NPPB locus, associations of the common polymorphisms with natriuretic peptides and the protein biomarkers themselves with incident disease could be confirmed. A strong causal relationship between natriuretic peptides and incidence of AF as well as HF was ruled out. Therapeutic approaches targeting natriuretic peptides will therefore very likely work through indirect mechanisms.

Comparison of hazard ratios

Funding Acknowledgement

Type of funding source: Public grant(s) – EU funding. Main funding source(s): European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme, German Ministry of Research and Education

P6310Risk predictors of cardiac allograft vasculopathy after heart transplantation: results from the United States Organ Procurement and Transplantation Network

Background

Cardiac allograft vasculopathy (CAV) is a major long-term complication after heart transplantation leading to chronic graft failure and increased mortality.

Purpose

The aim of this study was to determine recipient- and donor-related risk factors for the development of CAV in patients after heart transplantation.

Methods

Overall, data from 34,994 heart transplant recipients prospectively enrolled from July 2004 to March 2015 in the Organ Procurement and Transplantation Network (OPTN) were analyzed. Patients aged <18 years and those without information about CAV and re-transplantation were excluded. Multivariable-adjusted analyses were performed to identify recipient- and donor-related risk factors for new-onset CAV. The mean follow-up time was 66.8 months. Analyses are based on OPTN data as of March 6, 2017.

Results

Of 34,994 patients after heart transplantation, 12,668 (36.2%) patients developed CAV. Mean age was 52±12 years for the recipients (76.1% men) and 31±12 years for the donors (71.0% men), respectively.

In recipients, male sex (hazard ratio [HR] 1.14, 95% confidence interval [CI] 1.09–1.19, p<0.001), African American ethnicity (HR 1.11, 95% CI 1.06–1.17, p<0.001), body mass index (BMI) (HR per 5 kg/m2 increase 1.08, 95% CI 1.06–1.11, p<0.001) and smoking (HR 1.07, 95% CI 1.01–1.13, p=0.03) were associated with incident CAV. Moreover, recipients with ischemic (HR 1.30, 95% CI 1.09–1.55, p=0.003) and hypertrophic cardiomyopathy (HR 1.26, 95% CI 1.02–1.57, p=0.03) had a higher risk for new-onset CAV than patients with other cardiomyopathies.

In donors, age (HR 1.11, 95% CI 1.10–1.11, p<0.001), male sex (HR 1.28 95% CI 1.22–1.34, p<0.001), BMI (HR per 5 kg/m2 increase 1.04, 95% CI 1.02–1.05, p<0.001), smoking (HR 1.09, 95% CI 1.04–1.13, p<0.001), diabetes (HR 1.21 95% CI 1.09–1.36, p<0.001) and arterial hypertension (HR 1.13, 95% CI 1.07–1.20, p<0.001) were associated with new-onset CAV. Contrarily, African American (HR 0.93, 95% CI 0.88–0.98, p=0.007) and Hispanic ethnicity (HR 0.94, 95% CI 0.89–0.99, p=0.03) seemed to be protective.

Conclusion

Both recipient and donor male sex as well as the classical cardiovascular risk factors BMI and smoking were associated with incident CAV. On the donor side, additionally, diabetes and arterial hypertension were related to new-onset CAV. Diverse ethnicities were differentially related to new-onset CAV. Further studies are needed to clarify whether modification of cardiovascular risk factors as well as improved donor selection will reduce CAV burden.

P3820Differential associations of common risk factors and biomarkers with atrial fibrillation and heart failure and their ability to predict sequential disease onset and mortality

Background

Although atrial fibrillation (AF) and heart failure (HF) have a similar cardiovascular risk profile, the differential associations of the risk factors with both disease are incompletely understood.

Aim

The aim of this study was to understand whether common clinical risk factors and cardiovascular biomarkers show different associations with incident AF and HF, and to investigate predictors of sequential disease onset and mortality.

Methods

In 58693 individuals free of AF and HF from European population-based cohorts, pooled multivariable Cox regression analysis was used to find predictors for AF, HF and all-cause mortality. P-values for differences between Hazard Ratios (HR) of risk factors for AF and HF were estimated using bootstrapping with 5,000 replications. When AF and/or HF were used in Cox regressions as explanatory variables, they were included as time-dependent variables.

Results

Median age was 50.5 years, 49.3% were men. Median follow-up time was 13.8 years with an all-cause mortality rate of 15.7%. Incident AF and HF was present in 5.0% and 5.4% of the participants, with 1.8% showing a sequential disease onset.

In multivariable-adjusted models we observed stronger associations of body mass index (HR of 1.32 (95% CI 1.25–1.39) vs. 1.42 (95% CI 1.36–1.49), p=0.02), smoking (HR of 1.21 (95% CI 1.08–1.33) vs. 2.11 (95% CI 1.90–2.32), p<0.01) and antihypertensive medication (HR of 1.21 (95% CI 1.10–1.35) vs. 1.43 (95% CI 1.27–1.59), p<0.01) with incident HF than with incident AF.

Total serum cholesterol (HR of 1.10 (95% CI 1.06–1.15), prevalent diabetes (HR of 3.46 (95% CI 2.60–4.32), high-sensitive C-reactive protein (HR of 1.12 (95% CI 1.08–1.16)) and glomerular filtration rate (HR of 0.92 (95% CI 0.85–1.00) were significantly related to incident HF but not AF.

Age (HR of 1.54 (95% CI 1.47–1.61) vs. 1.54 (95% CI 1.47–1.62), p=0.95), male sex (HR of 2.87 (95% CI 2.42–3.33), p=0.13), prevalent myocardial infarction (HR of 1.65 (95% CI 1.26–2.04) vs. 1.75 (95% CI 1.36–2.11), p=0.73) and NT-proBNP (HR of 1.59 (95% CI 1.50–1.68) vs. 1.60 (95% CI 1.51–1.69), p=0.86) showed comparable associations with both diseases.

Age, male sex, body mass index, total serum cholesterol, prevalent diabetes and NT-proBNP were all predictors of sequential disease onset after multivariable adjustment.

In models including cardiovascular risk factors and NT-proBNP, the time-varying covariates incident AF and HF showed a strong association with all-cause mortality, with HR of 2.2 (95% CI 1.9–2.5) and 10.7 (95% CI 9.1–12.6), respectively. Sequential disease onset further increased the hazard ratio to 15.1 (95% CI 11.6–19.5).

Conclusion

In our pooled analysis of population-based cohorts, new-onset AF and HF showed different associations with common cardiovascular risk factors and biomarkers. Although both diseases significantly increased mortality, the highest risk was observed in individuals with sequential disease onset.

P4796Risk prediction of atrial fibrillation in the community combining biomarkers and genetics

Background

Classical cardiovascular risk factors (CVRF), biomarkers and genetic variation have been suggested for risk assessment of atrial fibrillation (AF).

Purpose

To evaluate their clinical potential, we analysed their individual and combined effectiveness in AF prediction.

Methods

In N=6945 individuals of the FINRISK 1997 cohort, we assessed the predictive value of CVRF, N-terminal pro B-type natriuretic peptide (NT-proBNP) and 145 recently identified single nucleotide polymorphisms (SNPs) for incident AF.

Results

Over a median follow-up of 17.8 years, N=551 participants (7.9%) developed AF. In multivariable-adjusted Cox proportional hazard models, NT-proBNP (hazard ratio (HR) per standard deviation (SD) 1.90, 95% confidence interval (CI): 1.71–2.11, P<0.001) and the polygenic risk score (PRS) (HR per SD 1.66, 95% CI: 1.51–1.84, P<0.001) were significantly related to incident AF. The discriminatory ability improved asymptotically with increasing numbers of SNPs. Compared to a clinical model, AF risk prediction was significantly improved by addition of NT-proBNP and the PRS. The C-statistic for the combination of all CVRF, NT-proBNP and the PRS reached 0.82 compared to 0.77 for CVRF only (P<0.001). Comparing the highest versus lowest quartile, age remained the strongest risk factor with a 15-fold increased risk of AF. The highest quartiles of NT-proBNP and the PRS both showed an approximately 3-fold increased AF risk compared to the lowest quartiles.

C-Index for AF prediction

Conclusions

The PRS and the established biomarker NT-proBNP predicted incident AF comparably. Both provided incremental predictive value over standard clinical variables. Further improvements for the PRS are likely with the discovery of additional SNPs.

Acknowledgement/Funding

European Research Council, German Ministry of Research and Education, DZHK, European Union Seventh Framework Programme, CHANCES, THL

Comment on 'Empirical versus modelling approaches to the estimation of measurement uncertainty caused by primary sampling' by J. A. Lyn, M. H. Ramsey, A. P. Damant and R. Wood

Recently, Lyn et al. (Analyst, 2007, 132, 1231) compared two ways of estimating the standard uncertainty of sampling pistachio nuts for aflatoxins--a modelling method and an empirical method. Their case study used robust analysis of variance (RANOVA) to derive the uncertainty estimates, highlighting a substantial difference between the two: the estimate of sampling uncertainty derived from the modelling method was six-fold greater than that using the empirical approach (cf. 136% and 22.5%, respectively, when expressed as relative standard deviations (RSDs) at 68% confidence). A further analysis of this case study is reported here and suggests that the estimation uncertainty during RANOVA in the empirical approach could account for this difference.