Table 3

Estimates of the effect of location-specific scar % on the clinical and echocardiographic endpoints

	Adjusted* estimate of HR
	Septal scar†		Anterior scar‡		Lateral scar§		Inferior scar¶
A: clinical endpoint	Estimate	P value	Estimate	P value	Estimate	P value	Estimate	P value
Time between CRT and death or heart failure admission	1.03	0.002	1.03	<0.001	1.03	0.005	1.02	0.017

	Adjusted* estimate of slope
	Septal scar†		Anterior scar‡		Lateral scar§		Inferior scar¶
B: echocardiographic endpoint	Estimate	P value	Estimate	P value	Estimate	P value	Estimate	P value
Change in EF after CRT	−0.26	0.002	−0.24	<0.001	−0.20	0.044	−0.31	0.008

*Adjusted estimates are from a model where the following covariates were included as predictors (in addition to total scar %): age, gender, presence of left bundle branch block, QRS duration, diabetes mellitus and hypertension.
†Septal scar was calculated as the average of the scar per cent in AHA segments 2, 3, 8, 9 and 14.
‡Anterior scar was calculated as the average of the scar per cent in AHA segments 1, 7 and 13.
§Lateral scar was calculated as the average of the scar per cent in AHA segments 5, 6, 11, 12 and 16.
¶Inferior scar was calculated as the average of the scar per cent in AHA segments 4, 10 and 15.
AHA, American Heart Association; CRT, cardiac resynchronisation therapy; EF, ejection fraction.