# Improvement of Heart Failure Discrimination by the Integration of the Left Ventricle Global Longitudinal Strain

## TL;DR

Adding left ventricle global longitudinal strain (GLS) to standard tests improves the accuracy of diagnosing chronic heart failure in stable patients.

## Contribution

This study demonstrates that integrating GLS significantly enhances heart failure diagnosis beyond traditional clinical and echocardiographic parameters.

## Key findings

- Patients with heart failure had significantly lower GLS values compared to those without.
- Including GLS in logistic models increased the AUC and improved diagnostic reclassification by 19.8% for heart failure.
- GLS was independently associated with heart failure with preserved ejection fraction (HFpEF) and improved its diagnostic accuracy.

## Abstract

Introduction: Clinical diagnosis of chronic heart failure (HF) in ambulatory patients can be difficult. Echocardiography is the most widespread diagnostic imaging technique, although the usefulness of the global longitudinal strain (GLS) of the left ventricle (LV) in this clinical setting is less clear. Methods: We performed a cross-sectional study of stable outpatients and GLS was obtained with an automatic software that uses the three apical planes of the LV. We analyzed the improvement of the diagnostic capacity of including GLS above all the clinical and echocardiographic parameters using reclassification indexes. Results: We included 1362 patients, including 12.9% with HF who presented lower values of ejection fraction (EF) and GLS and worse diastolic function. Most patients (92.8%) with HF had a GLS < −14 as compared to patients without HF (36.1%). LV EF (OR: 0.93) and GLS (OR: 1.27 CI 95% 1.20–1.35) were associated with the presence of HF. The AUC was significantly higher (p < 0.001) in the logistic model that included GLS vs. without GLS, and the reclassification index for GLS was 19.8%. GLS was more affected in patients with HFpEF vs. controls as well as diastolic function parameters. The logistic regression model only identified age (OR: 1.07 95% CI 1.02–1.06) and GLS (OR: 1.29 95% CI 1.21–1.38) as independently associated with the presence of HFpEF. The AUC of the model for the presence of HFpEF with GLS was significantly higher (p < 0.01). The reclassification index for GLS was 38.8%. Conclusions: LV GLS assessment increased the diagnostic discrimination of chronic HF in stable patients.