LVEF improvement in HFrEF reflects progressive atrial and neurohormonal reverse remodelling rather than favourable baseline atrial mechanics, and is associated with lower mortality and fewer heart failure hospitalisations.
Key Findings
Results
LVEF improvement occurred in 45.7% of HFrEF patients followed for 24 months.
175 HFrEF patients were followed prospectively for 24 months with serial echocardiography, left atrial (LA) strain and 16 biomarkers.
LVEF improvement was defined as an increase of at least 10 points to a value above 40%.
80 patients (45.7%) met this definition.
Results
Improvers differed from non-improvers in clinical, structural and biomarker characteristics at baseline.
Improvers were more often women and had non-ischaemic aetiology.
They had smaller LV volumes, better LV longitudinal strain and shorter atrial conduction time.
They had lower ANP, relaxin-2 and COL4A1.
Results
Baseline LA volume and LA strain did not differ between improvers and non-improvers.
Favourable baseline atrial mechanics did not distinguish patients who later improved.
This contrasts with the differences seen in LV volumes, LV longitudinal strain and atrial conduction time.
Results
An exploratory eight-variable model discriminated improvers from non-improvers, but performance fell after correction for optimism.
The apparent AUC was 0.93.
The optimism-corrected AUC was 0.84.
The model was described as exploratory.
Results
NT-proBNP and LA reservoir strain diverged progressively between improvers and non-improvers over follow-up.
Serial echocardiography and biomarker measurements over 24 months showed progressively widening differences between groups.
The authors interpret this as progressive atrial and neurohormonal reverse remodelling.
Results
LVEF improvement was associated with lower mortality and fewer heart failure hospitalisations.
Mortality: HR 0.25 (95% CI 0.08-0.75).
Heart failure hospitalisation: HR 0.38 (95% CI 0.18-0.78).
Conclusions
The authors conclude that LVEF improvement reflects progressive atrial and neurohormonal reverse remodelling rather than favourable baseline atrial mechanics.
Baseline LA volume and strain did not differ between groups, while NT-proBNP and LA reservoir strain diverged over time.
Integrated atrial mechanics-biomarker profiling was used to characterise the substrate of LVEF improvement.
What This Means
This study followed 175 people with heart failure and weakened pumping function for two years, repeatedly measuring heart ultrasound, the stretch of the left atrium (the upper heart chamber), and 16 blood markers. About 46% of patients (80) had a meaningful recovery in pumping function, defined as an increase of at least 10 points to above 40%.
Patients who recovered were more often women and had heart failure not caused by blocked arteries. They tended to have smaller heart chambers, better heart muscle shortening, quicker electrical conduction in the atrium, and lower levels of several blood markers. Atrial size and stretch at the start did not differ between the groups. Over time, though, the atrial stretch and the stress marker NT-proBNP diverged, with recovering patients improving more. Recovery was linked to lower risk of death (HR 0.25) and fewer hospital stays for heart failure (HR 0.38).
This research suggests that recovery of heart function is accompanied by gradual healing of the atrium and hormone systems rather than being predictable from how the atrium works at the start. A combined model of eight measures separated recoverers from non-recoverers fairly well (AUC 0.84 after correction), but it was exploratory and would need to be tested in other groups of patients before any practical use.