Global longitudinal strain before cardiac surgery: Improving feasibility, reproducibility, and variability

Area of Science:

• Cardiology
• Echocardiography
• Cardiac Imaging

Background:

• Global longitudinal strain (GLS) is a key predictor of outcomes in patients undergoing cardiac surgery.
• Accurate GLS assessment is crucial for clinical decision-making and surgical timing.
• Poor image quality and nontraceable myocardial segments frequently challenge GLS in this patient group.

Purpose of the Study:

• To evaluate the feasibility, reproducibility, and variation of GLS estimation in patients scheduled for cardiac surgery.
• To compare alternative GLS analysis models accounting for nontraceable segments.
• To determine if modified GLS models can improve feasibility without compromising accuracy.

Main Methods:

• Two-dimensional strain analysis was performed on 250 patients undergoing open-heart surgery.
• Intra- and inter-examiner variability was assessed in 50 patients.
• Feasibility, reproducibility (coefficient of variation), and variation (bias ± LOA) were analyzed for different models, including those with nontraceable segments.

Main Results:

• The standard GLS method (max 1 nontraceable segment/view) was feasible in only 64% of cases (mean GLS -16%).
• Accepting up to three nontraceable segments increased feasibility to 77%, with similar reproducibility and variation.
• A segmented approach (max 1 nontraceable segment per region) improved feasibility to 72% with comparable results.

Conclusions:

• GLS estimation prior to cardiac surgery is limited by moderate feasibility and retest variation, particularly with nontraceable segments.
• Alternative GLS models offer improved feasibility, making GLS a more viable tool in preoperative cardiac assessment.
• These findings support the integration of modified GLS analysis into clinical practice for better patient management.