Echocardiographic measurements
The device size was correlated significantly with all the maximal ASD diameter by 2D-TTE (R 0.64, P ≤0.001), by 2D-TEE (R 0.78, P ≤0.001) and 3D-TEE (R 0.80 P ≤0.001). The correlations between maximal ASD diameters by echocardiography and device size are reported in Table (2). The strongest correlation was noted with 3D TEE maximal diameter R 0.80. Mean device size 24.04 ± 6.53, Mean 3D TEE maximal diameter 19.52±6.01, and mean TEE maximal diameter 17.42±5.86.Maximal ASD diameters by 3D-TEE were larger than maximal ASD diameters by 2D-TEE and by 2D-TTE The mean difference was 2 mm P ≤ 0.001 , and 2.25 mm , P 0.008).
The device size was larger than all echocardiographic ASD diameters. The mean difference between Device size and 3D-TEE maximal diameter, 2D-TEE maximal diameter, and 2D-TTE maximal diameter was 4.0 mm P ≤0.001, 6 mm, P ≤ 0.001- and 6.5-mm P ≤ 0.001 respectively. (Table 4).
With multiple linear regression analysis, one formula was suggested to predict device size using the different echocardiographic measurements. The suggested formula was y = c + b*x where y = estimated dependent variable score (device size) c = constant, b = beta coefficient, and x = score on the independent variable .This model was more predictive with 3D maximal diameter with R 72%.The values for this suggested formula are presented in (Table 3) Using the 3D maximal diameter (Device size = 6.9 + 0.87 x 3D max diameter ) R2 = 0.72 P ≤ 0.001.Using 2D TEE maximal diameter ( Device size =8.9 +0.87 x 2D TEE max diameters ) R2=0.69 P ≤ 0.001.However, all these formulas needs to be validated on a larger number of patients.
Using the first and second formulas there was no significant difference between the suggested device and the actual device. Mean difference was 23.97±5.31 for equation 1 and 24.03±5.08 for equation 2 with P 0.919 and 0.984 respectively.