Timing of active left ventricular unloading in patients with cardiogenic shock on veno-arterial extracorporeal membrane oxygenation therapy

Aim

To evaluate the impact of timing of active left ventricular (LV) unloading in relation to veno-arterial extracorporeal membrane oxygenation (VA-ECMO) implantation on outcomes of patients with cardiogenic shock (CS).

Methods and results

Data from 421 patients with CS treated with active LV unloading on top of VA-ECMO at 18 tertiary-care centers in 4 countries were collected. Only patients in whom both devices were implanted within 24 hours of each other were considered and patients were stratified by timing of device implantation in early vs. delayed active LV unloading (e.g. active LV unloading before vs. after VA-ECMO). Cox and logistic regression models (adjusted for age, sex, lactate, cardiopulmonary resuscitation (CPR), VA-ECMO assisted CPR and enrollment center) were fitted to evaluate the association between early active LV unloading and 30-day mortality as well as several safety outcomes.

Overall, 310 (73.6%) patients were treated with early active LV unloading. Early active LV unloading was associated with a lower 30-day mortality risk (hazard ratio 0.63, 95% confidence interval 0.46–0.87) and a higher likelihood of weaning from mechanical ventilation (odds ratio 1.25, 95% confidence interval 1.03–1.52), but not with more complications. Importantly, postponing active LV unloading in these patients was associated with higher mortality risk (Figure 1), and lower likelihood of successful weaning from mechanical ventilation.

Conclusion

This exploratory study lends support to the use of early active LV unloading in CS patients on VA-ECMO as a primary treatment strategy, as opposed to a bail-out approach.

Funding Acknowledgement

Type of funding sources: Private company. Main funding source(s): Unrestricted research grant from Abiomed

Temporal trends in incidence, patient characteristics, microbiology and in-hospital mortality in patients with infective endocarditis: a contemporary analysis of 86,469 cases between 2007 and 2019

Background

Infective endocarditis (IE) is characterized by high morbidity and mortality rates, despite recent improvements in diagnostics and treatment. We aimed to investigate incidence, clinical characteristics, and in-hospital mortality in a large-scale nationwide cohort.

Methods

Using data from the German Federal Bureau of Statistics, all IE cases in Germany between 2007 to 2019 were analyzed. Logistic regression models were fitted to assess associations between clinical factors and in-hospital mortality.

Results

In total, 86,469 patients were hospitalized with IE between 2007 and 2019. The mean age was 66.5±14.7 years and 31.8% (n=27,534/86,469) were female. Cardiovascular (CV) comorbidities were common. The incidence of IE in the German population increased from 6.3/100,000 to 10.2/100,000 between 2007 and 2019 (Figure 1). Staphylococcus (n=17,673/86,469; 20.4%) and streptococcus (n=17,618/86,469; 20.4%) were the most common IE-causing bacteria. The prevalence of staphylococcus gradually increased over time, whereas blood culture-negative IE (BCNIE) cases decreased (Figure 2). In-hospital mortality in patients with IE was 14.9%. Compared to BCNIE, staphylococcus and gram-negative pathogens were associated with higher in-hospital mortality. In multivariable analysis, factors associated with higher likelihood of in-hospital mortality were advanced age, female sex, CV comorbidities (e.g., heart failure, COPD, diabetes, stroke), need for dialysis or invasive ventilation, and sepsis.

Conclusions

In this contemporary cohort, incidence of IE increased over time and in-hospital mortality remained high (∼15%). While staphylococcus and streptococcus were the predominant microorganisms, bacteremia with staphylococcusand gram-negative pathogens were associated with higher likelihood of in-hospital mortality. Our results highlight the need for new preventive strategies and interventions in patients with IE.

Funding Acknowledgement

Type of funding sources: None.

Use of mechanical circulatory support in patients with non-ischemic cardiogenic shock

Background

The aim of this study was to evaluate mechanical circulatory support (MCS) for the treatment of non-ischemic cardiogenic shock (CS).

Methods

Data from 1,030 consecutive patients with non-ischemic CS treated with or without MCS at 16 tertiary-care centers were retrospectively collected. The association between MCS and 30-day mortality was assessed in a 1:1 propensity score matched cohort.

Results

MCS was used in 406 (39%) patients. MCS treated patients presented with more severe CS (lactate 5.4 vs. 4.1 mmol/l, systolic blood pressure 80 vs. 83 mmHg, higher SCAI class) and with more disease modifiers (prior cardiac arrest 42.4 vs. 36.1%, mechanical ventilation 78.4 vs. 56.5%). After matching, 272 patients treated with were compared vs. 272 patients treated without MCS. MCS was associated with a lower 30-day mortality (hazard ratio 0.77, 95% confidence interval 0.60–0.98, Figure 1). This finding was consistent through all tested sub-groups except when ejection fraction was considered, indicating an association especially in patients with an ejection fraction ≤20%. Complications occurred more frequently in patients with MCS; e.g. severe bleedings (21.8 vs. 9.2%) and access-site related ischemia (6.6 vs. 0%).

Conclusion

In patients with non-ischemic CS, MCS use was associated with lower 30-day mortality as compared to medical therapy only, particularly in patients with a lower ejection fraction. This provides rationale for randomized trials to validate these findings.

Funding Acknowledgement

Type of funding sources: None.

Impact of left ventricular ejection fraction on mortality and use of mechanical circulatory support in non-ischaemic cardiogenic shock

Background

Evidence in non-ischaemic cardiogenic shock (CS), especially regarding prognostic markers and use of mechanical circulatory support (MCS), is scarce.

Aim

The aim of this study was to evaluate left ventricular ejection fraction (LVEF) as a prognostic marker as well as a factor to guide the use of MCS in non-ischaemic CS.

Methods

In this international observational study, patients with non-ischaemic CS (e.g. caused by severe de-novo or acute on chronic heart failure; but not by acute myocardial infarction) treated with or without MCS from 18 tertiary-care centers in five countries were enrolled. Cox regression models adjusted for age, sex, SCAI class, lactate, prior resuscitation, mechanical ventilation and pH were fitted to evaluate the association between LVEF and 30-day mortality as well as the interaction between MCS use, LVEF and 30-day mortality.

Results

A total of 807 patients were enrolled, of whom 387 (47,9%) were treated with and 418 (52.1%) without MCS; mean age was 63 [interquartile range (IQR) 51.5–72) years, 601 (74.5%) were male, 486 (60.2%) had acute on chronic heart failure, 221 (32.7%) had an ischaemic cardiomyopathy and 277 (34.5%) had prior cardiac arrest. The baseline LVEF was 20 (IQR 15–30) % and baseline lactate was 4.9 (IQR 2.6–8.5) mmol/l.

There was no significant association between LVEF and 30-day mortality risk [hazard ratio (HR) 0.95, 95% confidence interval (CI) 0.74–1.22 if LVEF was considered as a continuous variable; HR 1.09, 95% CI 0.83–1.44 if LVEF was considered as a categorical variable with ≤20% vs. >20%]. However, there was a significant interaction between MCS use and LVEF, indicating a lower 30-day mortality risk with MCS use in patients with a depressed LVEF (HR 0.74, 95% CI 0.52–1.05, interaction-p = 0.04).

Conclusion

In this retrospective, multicenter, international study of patients with non-ischaemic CS, LVEF was not a predictor of 30-day mortality risk. However, we observed a significant interaction between MCS use and LVEF, indicating a lower morality risk with MCS use only in patients with a depressed LVEF. This provides rationale to use LVEF as a parameter to guide MCS therapy in non-ischaemic CS, and calls for a randomized trial on this topic.

Funding Acknowledgement

Type of funding sources: None.

Non-invasive risk prediction based on right ventricular function in patients with pulmonary arterial hypertension

Background

Individual risk assessment in patients with pulmonary arterial hypertension (PAH) is fundamental to improve their outcome. Although right ventricular (RV) dysfunction is a major determinant of outcome in PAH, echocardiographic measures of RV function are poorly represented by current risk models.

Objective

The objective of this study was to identify echocardiographic measures of RV function, which are associated with adverse outcome and to develop a non-invasive, echocardiography-based risk score for PAH patients.

Methods

In 254 patients with PAH we analyzed functional status, laboratory results, pulmonary function and echocardiographic measures. Echocardiographic measures comprised RV chamber diameters, right atrial area, fractional area change (FAC), tricuspid annular plane systolic excursion (TAPSE), 2D RV strain and pericardial effusion. We used Cox regression models to assess the association with the composite endpoint of 5-year all-cause death or lung transplantation. The analyses included a conventional model using only guideline-recommended variables and a model adding significant echocardiographic measures. Based on the final multivariable model a point risk score was derived, indicating the association with the primary outcome.

Results

Median age was 65.5 years, 33.9% were females. During a median follow-up time of 4.18 years 74 patients died (n=63) or underwent lung transplantation (n=11). In univariable analyses low systolic blood pressure (Hazard ratio [HR] 0.99, 95% Confidence Interval [CI] 0.98,1.00), NYHA functional class IV (HR 3.23, 95% CI 1.48,7.07), 6-minute walk distance (HR 1.00, 95% CI 1.00,1.00), NT-proBNP concentrations (HR 1.00, 95% CI 1.00,1.00), renal impairment (HR 0.99, 95% CI 0.98,1.00), reduced diffusion capacity for carbon monoxide (HR 0.99, 95% CI 0.98,1.00), reduced TAPSE (HR 0.90, 95% CI 0.85,0.96) and reduced FAC (HR 0.97, 95% CI 0.94,1.00) were associated with the endpoint. A multivariable, conventional risk model, including NYHA functional class, 6-minute walk distance, NT-proBNP concentrations, pericardial effusion and right atrial area, resulted in a C-Index of 0.539. Adding TAPSE and FAC to this model improved the performance significantly (C-index 0.639, p-value 0.017). This model was translated to a 12-point score with the highest weighting assigned to TAPSE, FAC, pericardial effusion and 6-minute walk distance (Figure).

Conclusion

An easily applicable score integrating non-invasive, echocardiographic parameters of RV function improves prediction of adverse outcome in PAH patients.

Funding Acknowledgement

Type of funding sources: None. Risk prediction chart

Eligibility for sacubitril/valsartan in heart failure across the ejection fraction spectrum: real-world data from the Swedish Heart Failure Registry

BACKGROUND

Randomized controlled trials (RCT) generalizability may be limited due to strict patient selection.

OBJECTIVE

In a real-world heart failure (HF) population, we assessed eligibility for sacubitril/valsartan based on PARADIGM-HF (sacubitril/valsartan effective)/PARAGON-HF [sacubitril/valsartan effective in mildly reduced ejection fraction (EF)].

METHODS

Outpatients from the Swedish HF Registry (SwedeHF) were analysed. In SwedeHF, EF is recorded as <30, 30-39, 40-49 and ≥50%. In PARAGON-HF, sacubitril/valsartan was effective with EF ≤ 57% (i.e. median). We defined reduced EF/PARADIGM-HF as EF < 40%, mildly reduced EF/PARAGON-HF ≤ median as EF 40-49%, and normal EF/PARAGON-HF > median as EF ≥ 50%. We assessed 2 scenarios: (i) criteria likely to influence treatment decisions (pragmatic scenario); (ii) all criteria (literal scenario).

RESULTS

Of 37 790 outpatients, 57% had EF < 40%, 24% EF 40-49% and 19% EF ≥ 50%. In the pragmatic scenario, 63% were eligible in EF < 50% (67% for EF < 40% and 52% for 40-49%) and 52% in EF ≥ 40% (52% for EF ≥ 50%). For the literal scenario, 32% were eligible in EF < 50% (38% of EF < 40%, 20% of EF 40-49%) and 22% in EF ≥ 40% (25% for EF ≥ 50%). Eligible vs. noneligible patients had more severe HF, more comorbidities and overall worse outcomes.

CONCLUSION

In a real-world HF outpatient cohort, 81% of patients had EF < 50%, with 63% eligible for sacubitril/valsartan based on pragmatic criteria and 32% eligible based on literal trial criteria. Similar eligibility was observed for EF 40-49% and ≥50%, suggesting that our estimates for EF < 50% may be reproduced whether or not a higher cut-off for EF is considered.

Presentation characteristics, use of treatments and outcomes in patients with ischaemic vs. non-ischaemic cardiogenic shock

Aim

Evidence on non-ischaemic cardiogenic shock (CS) is scarce. The aim of this study was to investigate differences in presentation characteristics, use of treatments and outcomes in patients with ischaemic vs. non-ischaemic CS.

Methods

Patients with CS admitted to a tertiary care hospital between October 2009 and October 2017 were identified and stratified as ischaemic CS/non-ischaemic CS based on the presence/absence of acute myocardial infarction. Missing data was handled by chained equation multiple imputation. Logistic and Cox regression models were fitted to investigate the association of non-ischaemic CS with presentation characteristics (adjusted for all baseline variables), and use of treatments as well as30-day in-hospital mortality (adjusted for relevant confounders including age, sex, prior cardiac arrest, haemodynamics, pH and lactate).

Results

A total of 978 patients were enrolled in this study; median age was 70 (interquartile range 58, 79) years and 70% were male. 505 patients (43%) had non-ischaemic CS. Patients with non-ischaemic CS were more likely younger and female; were less likely to be active smokers or to have diabetes, but more likely to have chronic renal disease and history of myocardial infarction; and were more likely to present with unfavourable haemodynamics and with mechanical ventilation. Regarding use of treatments, patients with non-ischaemic CS were more likely to be treated with catecholamines [odds ratio (OR) 1.58, 95% confidence interval (CI) 1.11–2.27, p0.01], but less likely to be treated with extracorporeal membrane oxygenation (OR 0.66, 95% CI 0.48–0.92, p=0.02) or percutaneous left ventricular assist devices (OR 0.51, 0.35–0.74, p&lt;0.01). Unadjusted survival probabilities in patients with non-ischaemic vs. ischaemic CS were 36% (95% CI 32–42%) vs. 39% (95% CI 35–45%). After adjustment for multiple relevant confounders, non-ischaemic CS was associated with a significant increase in the risk of 30-day in-hospital mortality (hazard ratio 1.30, 95% CI 1.09–1.55, p&lt;0.01, Figure 1).

Conclusion

In this large study, non-ischaemic CS accounted for almost 50% of all CS cases. Non-ischaemic CS was not only associated with relevant differences in presentation characteristics and use of treatments, but also with a worse prognosis. These findings highlight the need for effective treatment strategies for patients with non-ischaemic CS.

Figure 1

Funding Acknowledgement

Type of funding source: None

281Impact of complications on VA-ECMO support for cardiopulmonary support: analysis of 8,351 adult patients in Germany from 2007 to 2015

Background

Venoarterial extracorporeal membrane oxygenation (VA-ECMO) is increasingly used for treatment of patients with critical cardiopulmonary failure. However, utilization of VA-ECMO support must be carefully weighed against possible complications. Therefore, we investigated the incidence and impact of complications on VA-ECMO support in one of the largest datasets of VA-ECMO therapy.

Material and methods

We analyzed complications and outcomes of all VA-ECMO procedures performed in Germany from 2007 to 2015 by using administrative data from the German Federal Health Monitoring System. For the present analyses all cases treated with VA-ECMO between 2007 and 2015 were identified and selected by the primary procedural code (OPS) for VA-ECMO (OPS code 8852.3).

Results

Among 8,351 patients undergoing VA-ECMO between 2007 and 2015, there were significant changes in complication rates over time such as increase in acute kidney injury (from 35.9% in 2007–2012 to 44.6% in 2013–2015), major bleeding (from 11.3% in 2007–2012 to 19.5% in 2013–2015 and abdominal ischemia (from 4.5% in 2007–2009 to 7.2% in 2013–2015). The incidence of stroke and limb ischemia did not differ over time. Procedure-related and ischemic complications were more frequently observed in non-survivors as compared to survivors (12.2% versus 15.3%, p<0.001) except for major bleeding (20.9% in survivors versus 15.0% in non-survivors, p<0.001). Multivariate analyses retained stroke and acute kidney injury as being significantly associated with 30-day in-hospital mortality, with respective OR [95% CI] of 1.7 [1.0–2.9] and 1.2 [1.1–1.3].

Conclusion

In one of the largest registries, major bleeding and ischemic events are the most common complications on VA-ECMO support. Ischemic complications seem to influence outcome more than bleeding complications. However, only stroke and acute kidney injury were independently associated with higher mortality rates. These findings should be incorporated in risk-benefit stratification when initiation of VA-ECMO and in prevention of complications to avoid additional morbidity and mortality in these critically ill patients.

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.

P6226Association of functional iron deficiency with incident cardiovascular diseases and mortality in the general population

Background

Functional iron deficiency (FID) has been established as a risk factor in patients with cardiovascular diseases (CVD). As opposed to absolute iron deficiency, it reflects stored iron as well as utilized iron and allows for a more accurate evaluation of individual iron status. However, evidence is scant on the relevance of FID to the incidence of CVD in the general population.

Aim

This study aimed to evaluate the association of FID with incident cardiovascular diseases and mortality endpoints in a large population-based cohort.

Methods

FID was defined as either ferritin below 100 μg/L or ferritin between 100 and 299 μg/L and transferrin saturation below 20%. Only individuals free of CVD at baseline from three population-based European cohorts were included. Multivariable-adjusted sex- and cohort-stratified Cox regression analyses were performed to evaluate the association of functional iron deficiency with incident cardiovascular diseases (coronary heart disease, cerebral infarction, heart failure and atrial fibrillation) as well as with all-cause and cardiovascular mortality. Adjustments were performed for sex (as strata), age (as time scale), smoking, total cholesterol, systolic blood pressure, diabetes, body mass index and high-sensitive C-reactive protein.

Results

In total, N=12146 individuals were included in the analysis with a median age of 59.0 years (25thpercentile 45.0, 75thpercentile 68.0), and 45.2% men. Incidence of FID was 64.3%. Median follow-up times were 12.3 to 21.8 years, with an all-cause mortality rate of 18.2% and a cardiovascular mortality rate of 6.2%. Incident coronary heart disease, cerebral infarction, heart failure and atrial fibrillation were observed in 8.7%, 6.5%, 5.9% and 11.7%, respectively.

FID was significantly associated with all-cause mortality (hazard ratio (HR) 1.12, 95% confidence interval (CI) 1.01–1.24, p=0.034), cardiovascular death (HR 1.26, 95% CI 1.03–1.54, p=0.027) and incident coronary heart disease (HR 1.23, 95% CI 1.06–1.43, p<0.01). There was no significant association with the other tested endpoints.

Conclusion

In our analysis of population-based cohorts, FID showed a significant positive association with all-cause as well as cardiovascular mortality and incident coronary heart disease. Further research is needed to validate the role of FID as a cardiovascular risk factor in the general population and to evaluate the impact of iron supplementation on gender and outcome.

P5738Multicenter analysis of left ventricular unloading on top of VA-ECMO for treatment of cardiogenic shock

Background

Veno-arterial extracorporeal membrane oxygenation therapy (VA-ECMO) is increasingly used for treatment of severe cardiogenic shock, although it causes an increase in left ventricular (LV) afterload and might therefore hamper myocardial recovery. Recently, the addition of catheter-based left ventricular assist device (cLVAD) on top of VA-ECMO has been used to unload the LV and to improve outcome measures. However, there is limited data on predictors of outcome in this high-risk population.

Aim

The aim of this study was to evaluate predictors of 30-day survival in a multicentre cohort of severe cardiogenic shock patients treated with cLVADon top of VA-ECMO.

Material and methods

We report on consecutive patients from six tertiary care centers being treated with cLVAD in addition to VA-ECMO for treatment of cardiogenic shock. The primary endpoint is 30-day all-cause mortality. To identify predictors of the primary endpoint, multivariate analysis using an “elastic net” variable selection algorithm was done after imputation of missing variables.

Results

A total of 220 patients treated with cLVAD on top of VA-ECMO were included in the analysis. Of these, 79.1% were male with a median age of 55.5 (25thpercentile 48.0, 75thpercentile 65.6) years. In 60.5% of the patients, acute myocardial infarction was the underlying cause of cardiogenic shock and in 44.6% VA-ECMO was used for refractory cardiac arrest (eCPR). In the multivariable analysis, the following baseline parameters were significantly associated with the primary endpoint: Age (odds ratio of 1.68 per standard deviation), duration of cardiopulmonary resuscitation (OR 2.08 per SD), lactate (OR 1.04 per SD) and time from onset of shock to VA-ECMO (OR 1.30 per SD).

Conclusion and outlook

In this large-scale multicentre analysis of severe cardiogenic shock patients treated with VA-ECMO plus additional cLVAD unloading, age, duration of cardiopulmonary resuscitation, lactate and time from onset of shock to VA-ECMO were significantly associated with 30-day all-cause mortality.

To further investigate this topic, we will evaluate predictors of outcome in distinct patient populations such as acute myocardial infarction vs. acute heart failure and patients without vs. patients with prior cardiopulmonary association.

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.

The predictive value of different equations for estimation of glomerular filtration rate in patients with coronary artery disease - Results from the AtheroGene study

BACKGROUND

Impaired renal function leads to dramatically increased risk for the development and progression of coronary artery disease (CAD). Therefore we aimed to assess the predictive value of different equations for estimated glomerular filtration rate (eGFR) in CAD-patients.

METHODS

From the AtheroGene study 2135 patients were included. eGFR was calculated using the 4-variable Modification of Diet in Renal Disease (4MDRD) equation for serum creatinine (sCr), the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation for sCr and cystatin C (CysC) each alone, and in combination (CysC/sCr). eGFR was assessed regarding the combined outcome of cardiovascular death and non-fatal myocardial infarction and regarding complex CAD represented by a SYNTAX score ≥23. Median follow-up was 4.3years.

RESULTS

Only the CKD-EPI equation using CysC could differentiate between eGFR >90ml/min/1.73m(2) vs. eGFR 60-90ml/min/1.73m(2) according to the occurrence of an endpoint event (log-rank test p=0.009). In the Cox regression analysis only eGFR calculated by CKD-EPI equation for CysC (Hazard ratio per 1 standard deviation (HR) 1.27 (95% CI 1.07-1.50); p=0.007) and for CysC/sCr (HR 1.22 (95% CI 1.02-1.46); p=0.026) were predictive regarding the outcome after adjustment for cardiovascular risk factors and Nt-proBNP. Furthermore, only eGFR calculated by CKD-EPI equation for CysC (odds ratio (OR) 1.57 (95% CI 1.36-1.78); p<0.001) and for CysC/sCr (OR 1.32 (95% CI 1.13-1.53); p<0.001) were significantly associated with a SYNTAX score ≥23.

CONCLUSION

In patients with CAD the CKD-EPI equation for CysC and for CysC/sCr provided the best predictive value regarding the prognosis and the severity of CAD.