A new tool for estimating left ventricular ejection fraction derived from wall motion score index.

Background: Radionuclide angiography (RNA) and echocardiography (biplane Simpson method) are the most accepted techniques for left ventricular ejection fraction (LVEF) assessment. A new method to evaluate LVEF based on the regional wall motion assessment of the LV was attempted.

Objective: To develop a simple method for LVEF estimation using wall motion score index (WMSI) with transthoracic echocardiography (TTE).

Methods: Two hundred and forty-three patients with abnormal LV contractility had TTE and RNA performed less than three days apart. The WMSI was calculated in all patients using the 16-segment model as proposed by the American Society of Echocardiography. For the first 150 patients, a correlation between LV WMSI and RNA EF was established to create a regression equation. This regression equation (RNA LVEF=92.8-25.8xWMSI) was used on 93 consecutive patients to compare this equation with RNA EF. From the total cohort (243 patients), three subgroups were studied specifically: atrial fibrillation (AF) (n=50 patients), dyskinesia (DK) (n=40 patients) and aneurysm (AN) (n=42 patients).

Results: Correlation between RNA EF and WMSI in the first 150 patients was r=0.82. In the second group of 93 consecutive patients, WMSI EF derived from the estimated regression equation correlated well with RNA EF (r=0.86). Correlation remained high in the three subgroups: AF (r=0.87), DK (r=0.87) and AN (r=0.80). In the 111 patients without DK, AN or AF correlation between RNA and the studied method was even higher (r=0.91). In a random subgroup of 54 patients, RNA was compared with the biplane Simpson method (49 of 54 patients, r=0.82). In the same subgroup of 54 patients, the score was modified to allow for mild hypokinesia (score=1.5) and severe hypokinesia (score=2.5) (54 of 54 patients, r=0.83).

Conclusion: LVEF assessment by this new simple mathematical model using the WMSI is feasible and easy to use during routine TTE. It has excellent correlation with other methods such as biplane Simpson and RNA.