Association is not causation: treatment effects cannot be estimated from observational data in heart failure.

Aims: Treatment 'effects' are often inferred from non-randomized and observational studies. These studies have inherent biases and limitations, which may make therapeutic inferences based on their results unreliable. We compared the conflicting findings of these studies to those of prospective randomized controlled trials (RCTs) in relation to pharmacological treatments for heart failure (HF). Methods and results: We searched Medline and Embase to identify studies of the association between non-randomized drug therapy and all-cause mortality in patients with HF until 31 December 2017. The treatments of interest were: angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, beta-blockers, mineralocorticoid receptor antagonists (MRAs), statins, and digoxin. We compared the findings of these observational studies with those of relevant RCTs. We identified 92 publications, reporting 94 non-randomized studies, describing 158 estimates of the 'effect' of the six treatments of interest on all-cause mortality, i.e. some studies examined more than one treatment and/or HF phenotype. These six treatments had been tested in 25 RCTs. For example, two pivotal RCTs showed that MRAs reduced mortality in patients with HF with reduced ejection fraction. However, only one of 12 non-randomized studies found that MRAs were of benefit, with 10 finding a neutral effect, and one a harmful effect. Conclusion: This comprehensive comparison of studies of non-randomized data with the findings of RCTs in HF shows that it is not possible to make reliable therapeutic inferences from observational associations. While trials undoubtedly leave gaps in evidence and enrol selected participants, they clearly remain the best guide to the treatment of patients.

Prevalence of Coronary Artery Disease and Coronary Microvascular Dysfunction in Patients With Heart Failure With Preserved Ejection Fraction.

Importance: Coronary artery disease (CAD) and coronary microvascular dysfunction (CMD) may contribute to the pathophysiologic characteristics of heart failure with preserved ejection fraction (HFpEF). However, the prevalence of CAD and CMD have not been systematically studied. Objective: To examine the prevalence of CAD and CMD in hospitalized patients with HFpEF. Design, Setting, and Participants: A total of 106 consecutive patients hospitalized with HFpEF were evaluated in this prospective, multicenter, cohort study conducted between January 2, 2017, and August 1, 2018; data analysis was performed from March 4 to September 6, 2019. Participants underwent coronary angiography with guidewire-based assessment of coronary flow reserve, index of microvascular resistance, and fractional flow reserve, followed by coronary vasoreactivity testing. Cardiac magnetic resonance imaging was performed with late gadolinium enhancement and assessment of extracellular volume. Myocardial perfusion was assessed qualitatively and semiquantitatively using the myocardial-perfusion reserve index. Main Outcomes and Measures: The prevalence of obstructive epicardial CAD, CMD, and myocardial ischemia, infarction, and fibrosis. Results: Of 106 participants enrolled (53 [50%] women; mean [SD] age, 72 [9] years), 75 had coronary angiography, 62 had assessment of coronary microvascular function, 41 underwent coronary vasoreactivity testing, and 52 received cardiac magnetic resonance imaging. Obstructive epicardial CAD was present in 38 of 75 participants (51%, 95% CI, 39%-62%); 19 of 38 (50%; 95% CI, 34%-66%) had no history of CAD. Endothelium-independent CMD (ie, coronary flow reserve <2.0 and/or index of microvascular resistance ≥25) was identified in 41 of 62 participants (66%; 95% CI, 53%-77%). Endothelium-dependent CMD (ie, abnormal coronary vasoreactivity) was identified in 10 of 41 participants (24%; 95% CI, 13%-40%). Overall, 45 of 53 participants (85%; 95% CI, 72%-92%) had evidence of CMD and 29 of 36 (81%; 95% CI, 64%-91%) of those without obstructive epicardial CAD had CMD. Cardiac magnetic resonance imaging findings included myocardial-perfusion reserve index less than or equal to 1.84 (ie, impaired global myocardial perfusion) in 29 of 41 patients (71%; 95% CI, 54%-83%), visual perfusion defect in 14 of 46 patients (30%; 95% CI, 19%-46%), ischemic late gadolinium enhancement (ie, myocardial infarction) in 14 of 52 patients (27%; 95% CI, 16%-41%), and extracellular volume greater than 30% (ie, diffuse myocardial fibrosis) in 20 of 48 patients (42%; 95% CI, 28%-56%). Patients with obstructive CAD had more adverse events during follow-up (28 [74%]) than those without obstructive CAD (17 [46%]). Conclusions and Relevance: In this cohort study, 91% of patients with HFpEF had evidence of epicardial CAD, CMD, or both. Of those without obstructive CAD, 81% had CMD. Obstructive epicardial CAD and CMD appear to be common and often unrecognized in hospitalized patients with HFpEF and may be therapeutic targets.

Effect of digoxin in patients with heart failure and mid-range (borderline) left ventricular ejection fraction.

AIMS: To evaluate the effects of digoxin in patients with the newly described phenotype of heart failure (HF) and mid-range ejection fraction (HFmrEF), attributed to mild left ventricular systolic dysfunction. METHODS AND RESULTS: We carried out a retrospective analysis of the Digitalis Investigation Group (DIG) trial which had 7788 patients available for analysis with a left ventricular ejection fraction (LVEF) ranging between 3% and 85%. We compared the effect of digoxin to placebo in three mutually exclusive groups of patients defined by LVEF category: <40% (HF with reduced LVEF, HFrEF, n = 5874), 40-49% (HFmrEF, n = 1195) and ≥50% (HF with preserved LVEF, HFpEF, n = 719). The primary outcome was the composite of cardiovascular death or HF hospitalisation. Patients with HFmrEF resembled patients with HFrEF, more than those with HFpEF, with respect to age, sex and aetiology but were more like HFpEF patients with respect to blood pressure and the prevalence of hypertension. Event rates in patients with HFmrEF were similar to those in HFpEF and much lower than in HFrEF. Digoxin reduced the primary endpoint in patients with HFrEF, mainly due to reduced HF hospitalisation: the digoxin/placebo hazard ratio (HR) for HF hospitalisation was 0.71 [95% confidence interval (CI) 0.65-0.77]. The digoxin/placebo HR for HF hospitalisation in patients with HFmrEF was 0.80 (95% CI 0.63-1.03) and 0.85 (95% CI 0.62-1.17) in those with HFpEF. The digoxin/placebo HR for the composite of HF death or HF hospitalisation was 0.74 (95% CI 0.68-0.81) in HFrEF, 0.83 (95% CI 0.66-1.05) in HFmrEF and 0.88 (95% CI 0.65-1.19) in HFpEF. CONCLUSIONS: In this study, event rates in patients with HFmrEF were closer to those in HFpEF than HFrEF. Digoxin had most effect on HF hospitalisation in patients with HFrEF, an intermediate effect in HFmrEF, and the smallest effect in HFpEF.

Falling Cardiovascular Mortality in Heart Failure With Reduced Ejection Fraction and Implications for Clinical Trials.

OBJECTIVES: This study examined the trends in the relative contributions of cardiovascular and noncardiovascular mortality to total mortality according to use of beta-blockers in clinical trials of patients with heart failure with reduced ejection fraction (HF-REF). BACKGROUND: With the increasingly widespread use of disease-modifying therapies, particularly beta-blockers, in HF-REF, the proportion of patients dying from cardiovascular causes is likely to be decreasing. METHODS: In a systematic review, 2 investigators independently searched online databases to identify clinical trials including >400 patients with chronic heart failure published between 1986 and 2014 and that adjudicated cause of death. Trials were divided into 3 groups on the basis of the proportion of patients treated with a beta-blocker (<33% [low], 33% to 66% [medium], and >66% [high]). Percentages of total deaths adjudicated as cardiovascular or noncardiovascular were calculated by weighted means and weighted standard deviations. Weighted Student t tests were used to compare results between groups. RESULTS: Sixty-six trials met the inclusion criteria with a total of 136,182 patients and 32,140 deaths. There was a sequential increase in the percentage of noncardiovascular deaths with increasing beta-blocker use from 11.4% of all deaths in trials with low beta-blocker use to 19.1% in those with high beta-blocker use (p < 0.001). CONCLUSIONS: In trials of patients with HF-REF, the proportion of deaths adjudicated as cardiovascular has decreased. Cardiovascular mortality, and not all-cause mortality, should be used as an endpoint for trials of new treatments for HF-REF.

The validity and reliability of 1-Hz and 5-Hz global positioning systems for linear, multidirectional, and soccer-specific activities.

PURPOSE: The study aimed to analyze the validity and reliability of commercially available nondifferential Global Positioning System (NdGPS) devices for measures of total distance during linear, multidirectional and soccer-specific motion at 1-Hz and 5-Hz sampling frequencies. METHODS: Linear (32 trials), multidirectional (192 trials) and soccer-specific courses (40 trials) were created to test the validity (mean ± 90% confidence intervals), reliability (mean ± 90% confidence intervals) and bias (mean ± 90% confidence intervals) of the NdGPS devices against measured distance. Standard error of the estimate established validity, reliability was determined using typical error and percentage bias was established. RESULTS: The 1-Hz and 5-Hz data ranged from 1.3% ± 0.76 to 3.1% ± 1.37 for validity and 2.03% ± 1.31 to 5.31% ± 1.2 for reliability for measures of linear and soccer-specific motion. For multidirectional activity, data ranged from 1.8% ± 0.8 to 6.88% ± 2.99 for validity and from 3.08% ± 1.34 to 7.71% ± 1.65 for reliability. The 1-Hz underestimated some complex courses by up to 11%. CONCLUSIONS: 1-Hz and 5-Hz NdGPS could be used to quantify distance in soccer and similar field-based team sports. Both 1-Hz and 5-Hz have a threshold beyond which reliability is compromised. 1-Hz also underestimates distance and is less valid in more complex courses.