Association of beta-blockers beyond 1 year after myocardial infarction for patients without heart failure or left ventricular systolic dysfunction and cardiovascular outcomes: nationwide cohort study

Background

Beta-blockers (BB) is an established treatment following presentation with myocardial infarction (MI). However, there is uncertainty as to whether BB use beyond the first year of MI has a secondary preventive role in patients without heart failure and/or left ventricular systolic dysfunction (LVSD).

Purpose

To investigate the association between BB treatment beyond one year after MI for patients without heart failure or LVSD and cardiovascular (CV) outcomes.

Methods

We used data from SWEDEHEART, the national Swedish register for coronary heart disease, to identify patients with MI who were hospitalised between 2005 and 2016. Deterministic linkage of individual patient data was performed with the National Patient Register, the Swedish Prescribed Drug Register, and the National Cause of Death Register. Patients with heart failure or LVSD with left ventricular ejection fraction <50% were excluded. Follow-up started at 1 year after hospitalisation with first MI (index date), when patients were allocated into two groups according to BB treatment. Information about BB treatment at index date and during follow-up was obtained from the Swedish National Prescribed Drug Register. The primary outcome was a composite of all-cause mortality, recurrent MI, unscheduled revascularisation or hospitalisation for heart failure. Secondary outcomes comprised the individual components of the composite outcome, CV death and stroke. Comparison of outcomes between the study groups was performed using Cox and Fine-Gray regression models adjusting for relevant clinical factors after propensity-score weighting. In the main intention-to-treat analysis, patients were censored at end of follow-up (31st December 2017), death or at pre-specified outcome, whichever came first. In supplementary per-protocol analysis, patients were, in addition, censored at the time of first BB discontinuation or switch between treatment arms.

Results

A total of 43,618 patients with MI were hospitalised between 2005 and 2016. Of these, 34,253 (78.5%) were prescribed BB and 9,365 (21.5%) were not on BB treatment at index date 1 year following MI. The median age of the population was 64 years, 25.5% were female, and 36.2% had a STEMI. Median follow-up was 4.5 years. In the intention-to-treat analysis, and after multivariable adjustments and propensity score weighting, BB treatment was associated with a similar rate of the composite CV outcome (hazard ratio [HR] 0.99; 95% confidence interval [CI] 0.93–1.04) compared with no BB treatment. A similar finding was observed when censoring for BB discontinuation or treatment switch during follow-up in a per-protocol analysis (HR 0.98; 95% CI 0.98–1.06). Similar associations were observed for all secondary outcomes (Figure 1).

Conclusions

BB treatment beyond one year after MI for patients without heart failure or LVSD is not associated with a different risk of cardiovascular outcomes compared with patients who do not receive BB.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): The study was financed by grants from the Swedish state under the agreement between the Swedish government and the county councils, the ALF-agreement.

Differences in biomarker concentrations and prediction of long-term outcome in patients with ST-elevation and non-ST-elevation myocardial infarction

Background

Biomarkers offer insights into the pathobiology of myocardial infarction (MI). Differences in biomarker concentrations and its prognostic implications in ST-elevation MI (STEMI) and non-STEMI (NSTEMI) are incompletely understood and may offer insights into tailored treatment.

Purpose

We aimed to characterise differences in activation of pathophysiologic pathways in patients with acute STEMI versus NSTEMI by comparing multiple biomarkers, and investigating their prognostic implications.

Methods

538 STEMI and 544 NSTEMI patients were included in our hospitals between 2008–2014. Blood samples were collected shortly after admission and 92 biomarkers were analysed using Proximity Extension Assay and 87 biomarkers using Multiple Reaction Monitoring mass spectrometry. Adjusted Lasso analysis (penalized logistic regression) was used to select biomarkers that discriminated STEMI from NSTEMI patients. For biomarkers that were identified by the Lasso analysis, concentrations (log2-transformed) were then evaluated in adjusted Cox regressions on associations with death or major cardiovascular events (MACE), a composite of death, hospitalization for MI, ischemic stroke or heart failure.

Results

Eleven biomarkers differed between STEMI and NSTEMI patients (Table) with higher OR (>1.00–1.42) and concentrations in STEMI patients of myoglobin, N-terminal pro-B-type natriuretic peptide (NT-proBNP), serum amyloid A-1 (SAA1) and serum amyloid A-2 (SAA2) protein, ST2 protein, interleukin-6 (IL-6) and chitinase-3-like protein 1 (CHI3L1). NSTEMI patients had lower OR (0.83-<1.00) and higher concentrations of fibroblast growth factor 23 (FGF 23), membrane-bound aminopeptidase P (MAMP), tumour necrosis factor-related activation-induced cytokine (TRANCE) and apolipoprotein C-I. During up to 6.6 years of prognostic follow-up, all of the biomarkers except MAMP displayed significant prognostic information in STEMI and/or NSTEMI patients. However, interaction analyses indicated that all biomarkers were similarly predictive (p>0.05) of death or MACE between MI groups.

Conclusions

In the acute setting, biomarkers indicated greater myocardial dysfunction and inflammation in STEMI patients, whereas NSTEMI patients displayed a heterogeneous and more ambiguous activation of pathophysiologic pathways. These biomarkers were not associated with prognostic differences between STEMI and NSTEMI patients. Therefore, the results do not support treating STEMI and NSTEMI patients differently based on these biomarkers.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Swedish Foundation for Strategic Research. Swedish Association of Local Authorities and Regions. Table 1

The importance of comorbidities for long-term outcome after aortic valve intervention in patients with preserved left ventricular ejection fraction, a nation-wide register study

Background

In Aortic stenosis (AS) cardiovascular comorbidities as well as left ventricular ejection fraction (LVEF) have an impact on postoperative outcome among patients undergoing aortic valve replacement (AVR). The prevalence of heart failure (HF) based on LVEF in patients with severe AS varies. Lately HF with preserved LVEF has gained more attention. The aim is to describe the prevalence and prognostic impact of cardiovascular comorbidities, including HF, in relation to LVEF before AVR in a national cohort of patients with AS.

Methods

Patients >18 years, undergoing AVR due to AS 2008–2014 were identified in the national register for heart diseases, SWEDEHEART. Preoperative LVEF and comorbidities were collected from the register and enriched with data from national patient registries. The outcome events were all cause mortality and hospitalization for HF as the main diagnosis. The cohort was separated by preoperative LVEF status; preserved (>50%) or reduced (≤50%). Outcome events were analysed by Cox regression.

Results

10406 patients, median age 73 (18–96) years whereof 3817 (36.7%) women, were included with a median follow-up of 35 months. In the cohort 15.9%, 73.9% and 10.2% received a mechanical, surgical biological and trans-catheter biological valve prosthesis, respectively. Preserved LVEF was present in 7512 (72.2%). Comorbidities were more frequent in the group with reduced LVEF (p<0.001). Irrespective of LVEF HF influenced outcome negatively (see table).

Conclusion

In patients planned for AVR a history of HF irrespective of LVEF worsen postoperative prognosis and a history of HF seems at least as important as LVEF when predicting long-term outcome. When stratifying patients for AVR with preserved LVEF, comorbidities such as HF and atrial fibrillation should be highlighted, and further research to identify risk factors for a negative postoperative outcome in this group seems important in optimizing the follow-up after AVR.

Funding Acknowledgement

Type of funding source: None

Use of proteomics to identify biomarkers associated with chronic kidney disease and long-term outcomes in patients with myocardial infarction

BACKGROUND

Patients with chronic kidney disease (CKD) have poor outcomes following myocardial infarction (MI). We performed an untargeted examination of 175 biomarkers to identify those with the strongest association with CKD and to examine the association of those biomarkers with long-term outcomes.

METHODS

A total of 175 different biomarkers from MI patients enrolled in the Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies (SWEDEHEART) registry were analysed either by a multiple reaction monitoring mass spectrometry assay or by a multiplex assay (proximity extension assay). Random forests statistical models were used to assess the predictor importance of biomarkers, CKD and outcomes.

RESULTS

A total of 1098 MI patients with a median estimated glomerular filtration rate of 85 mL min^-1 /1.73 m² were followed for a median of 3.2 years. The random forests analyses, without and with adjustment for differences in demography, comorbidities and severity of disease, identified six biomarkers (adrenomedullin, TNF receptor-1, adipocyte fatty acid-binding protein-4, TNF-related apoptosis-inducing ligand receptor 2, growth differentiation factor-15 and TNF receptor-2) to be strongly associated with CKD. All six biomarkers were also amongst the 15 strongest predictors for death, and four of them were amongst the strongest predictors of subsequent MI and heart failure hospitalization.

CONCLUSION

In patients with MI, a proteomic approach could identify six biomarkers that best predicted CKD. These biomarkers were also amongst the most important predictors of long-term outcomes. Thus, these biomarkers indicate underlying mechanisms that may contribute to the poor prognosis seen in patients with MI and CKD.

P5537Biomarkers in addition to clinical characteristics for prediction of peripheral artery disease in patients with recent myocardial infarction

Background

Peripheral artery disease (PAD) is associated with increased risk of cardiovascular events and mortality. PAD is underdiagnosed in patients admitted for myocardial infarction (MI).

Purpose

To evaluate whether biomarkers in addition to clinical characteristics could improve the identification of PAD in MI patients.

Methods

Two prospectively observational studies including unselected patients with recent ST-elevation or non-ST elevation MI were used for the analyses. PAD was defined as abnormal ankle brachial indexes (ABI) score (<0.9 or >1.4) on at least one side. The Proximity Extension Assay (PEA) technique was used to simultaneously analyze 92 biomarkers with association to cardiovascular disease (CVD I and III, Olink Proteomics) in samples early after the acute MI. Random forest was used to identify the biomarkers with higher association to PAD. The discriminating accuracy of adding biomarkers to clinical characteristics including age, sex, diabetes mellitus, hypertension and smoking were analyzed by the c-statistic.

Results

In the REBUS cohort (n=420) the mean age was 67 years, 22.1% were women, 26.4% were smoking, 16.2% had diabetes and 53.3% hypertension. In the VAMIS cohort (n=998) the mean age was 71 years, 33.5% were women, 19.9% were smoking, 18.1% and 49.9% had diabetes and hypertension, respectively. PAD was found in 20.0% and 22.8% of the patients in the REBUS and VAMIS cohort, respectively. Six biomarkers associated with PAD (Tnfrsf 14, tnf r2, tnf r1, GDF-15, IL-6 ra, ctsd) were identified in the REBUS cohort and validated in the VAMIS cohort.

Additional predictive value of the biomarkers compared to the performance of the model with clinical characteristics, as measured by the c-statistic Clinical characteristics (CC) CC + Tnfrsf14 CC + Tnf r2 CC + Tnf r1 CC + GDF-15 CC + IL-6 ra CC + ctsd REBUS (n=420) 0.683 (0.610,0.756) 0.702 (0.632, 0.772) 0.703 (0.633, 0.773) 0.709 (0.640, 0.779) 0.710 (0.640, 0.781) 0.682 (0.608, 0.775) 0.683 (0.610, 0.757) VaMIS (n=998) 0.729 (0.687, 0.770) 0.732 (0.691, 0.733) 0.745 (0.704, 0.785) 0.746 (0.706, 0.787) 0.752 (0.711, 0.792) 0.729 (0.688, 0.770) 0.736 (0.695, 0.777) Tnfrsf 14 = tumor necrosis factor receptor superfamily member 14; Tnf r2 = tumor necrosis factor receptor 2; Tnf r1 = tumor necrosis factor receptor 1; GDF-15 = growth differentiation factor 15; IL-6 ra = interleukin-6 receptor subunit alpha; ctsd = Cathepsin D.

Conclusion

Out of a large panel of biomarkers, tumor necrosis factor and its receptors and GDF-15 were identified associated with PAD in MI patients.

In this underdiagnosed manifestation of atherosclerosis in MI patients, clinical characteristics are strong predictors of PAD, but the cytokines may have an additional value. Further studies focusing on more widespread atherosclerosis in acute coronary syndrome might identify a patient group with benefit of treatment targeting inflammation.

Acknowledgement/Funding

None

Open or endovascular revascularization in the treatment of acute lower limb ischaemia

Background

Consensus is lacking regarding intervention for patients with acute lower limb ischaemia (ALI). The aim was to study amputation‐free survival in patients treated for ALI by either primary open or endovascular revascularization.

Methods

The Swedish Vascular Registry (Swedvasc) was combined with the Population Registry and National Patient Registry to determine follow‐up on mortality and amputation rates. Revascularization techniques were compared by propensity score matching 1 : 1.

Results

Of 9736 patients who underwent open surgery and 6493 who had endovascular treatment between 1994 and 2014, 3365 remained in each group after propensity score matching. Results are from the matched cohort only. Mean age of the patients was 74·7 years; 47·5 per cent were women and mean follow‐up was 4·3 years. At 30‐day follow‐up, the endovascular group had better patency (83·0 versus 78·6 per cent; P < 0·001). Amputation rates were similar at 30 days (7·0 per cent in the endovascular group versus 8·2 per cent in the open group; P = 0·113) and at 1 year (13·8 versus 14·8 per cent; P = 0·320). The mortality rate was lower after endovascular treatment, at 30 days (6·7 versus 11·1 per cent; P < 0·001) and after 1 year (20·2 versus 28·6 per cent; P < 0·001). Accordingly, endovascular treatment had better amputation‐free survival at 30 days (87·5 versus 82·1 per cent; P < 0·001) and 1 year (69·9 versus 61·1 per cent; P < 0·001). The number needed to treat to prevent one death within the first year was 12 with an endovascular compared with an open approach. Five years after surgery, endovascular treatment still had improved survival (HR 0·78, 99 per cent c.i. 0·70 to 0·86) but the difference between the treatment groups occurred mainly in the first year.

Conclusion

Primary endovascular treatment for ALI appeared to reduce mortality compared with open surgery, without any difference in the risk of amputation.

Endovascular may save lives