Skip to main content

Table 3 Regression model to identify association between segmental strain, segement thickness and tissue characteristics

From: Effect of cellular and extracellular pathology assessed by T1 mapping on regional contractile function in hypertrophic cardiomyopathy

 

Segment thickness

BIC

LGE

BIC

Native T1

BIC

Extracellular volume

BIC

Ecc-FT

Univariable

Beta

1.325 (1.200;1.450)

5345.6

9.234 (7.444;11.024)

5592.9

0.039 (0.031;0.047)

5476.3

0.268 (0.168;0.368)

5368.1

P value

<0.001

 

<0.001

 

<0.001

 

<0.001

 

Multivariable 1

Beta

1.317 (1.165;1.470)

5074.1

1.294 (-0.601;3.191)

   

−0.046 (-0.136;0.045)

 

P value

<0.001

 

0.195

   

0.394

 
 

Multivariable 2

Beta

1.219 (1.070;1.368)

5214.3

0.331 (-1.470;2.131)

 

0.015 (0.008;0.022)

   

P value

<0.001

 

0.718

 

<0.001

   

Err-FT

Univariable

Beta

−3.573 (−3.963;−3.183)

7167.8

−25.595 (-30.940;-20.251)

7354.3

−0.118 (−0.142;−0.093)

7177.7

−0.693 (−0.991;−0.394)

6997.0

P value

<0.001

 

<0.001

 

<0.001

 

<0.001

 

Multivariable 1

Beta

−3.532 (-4.011;-3.073)

6758.4

−4.861 (-10.661;0.941)

   

0.201 (-0.080;0.481)

 

P value

<0.001

 

0.101

   

0.161

 
 

Multivariable 2

Beta

−3.193 (-3.659;-2.726)

6986.5

−1.266 (-6.868;4.336)

 

−0.053 (-0.075;-0.030)

   

P value

<0.001

 

0.658

 

<0.001

   

Ecc-SPAMM

Univariable

Beta

1.039 (0.881;1.198)

3651.4

8.525 (6.382;10.667)

3738.8

0.033 (0.024;0.042)

3681.3

0.292 (0.167;0.417)

3601.5

P value

<0.001

 

<0.001

 

<0.001

 

<0.001

 

Multivariable 1

Beta

0.964 (0.782;1.146)

3451.0

2.603 (0.341;4.866)

   

0.061 (-0.055;0.177)

 

P value

<0.001

 

0.024

   

0.303

 
 

Multivariable 2

Beta

0.0849 (0.669;1.030)

3581.4

2.194 (-0.029;4.417)

 

0.017 (0.008;0.025)

   

P value

<0.001

 

0.053

 

<0.001

   
  1. Mixed effects model, with random intercept for each patient assessing the association between Ecc-FT, Err-FT and Ecc-SPAMM and the tissue characteristics of each segment. Multivariable model 1 includes segment thickness, LGE and ECV and model 2 includes segment thickness, LGE and native T1. Values denoted in bold are statistically significant with P<0.05. Beta, regression coefficient with 95% confidence interval; BIC, Bayesian information criterion