Effectiveness of late gadolinium enhancement to improve outcomes prediction in patients referred for cardiovascular magnetic resonance after echocardiography

Table 2 Late gadolinium enhancement (LGE) added significant predictive ability for either death or transplant free survival beyond ejection fraction (EF), age, regional wall motion by echocardiography and gender in Cox regression models regardless of how EF was measured

Multivariable Cox regression model:		Adjusted hazard ratio for LGE (95% CI)	NRI (95% CI)	IDI (95% CI)
Outcome	EF modality	Adjusted hazard ratio for LGE (95% CI)	NRI (95% CI)	IDI (95% CI)
Mortality	CMR	3.84 (1.11-13.4)	0.61 (0.30-0.92)	0.017 (0.005, 0.028)
Mortality	echocardiography	4.44 (1.30-15.2)	0.72 (0.40-0.98)	0.024 (0.008, 0.040)
Transplant free survival	CMR	4.02 (1.16-13.9)	0.64 (0.35-0.93)	0.018 (0.005, 0.030)
Transplant free survival	echocardiography	4.79 (1.41-16.3)	0.74 (0.46-0.99)	0.027 (0.009, 0.044)

All models stratified by gender and adjusted for age, regional wall motion abnormalities by echocardiography and ejection fraction (EF) by either echocardiography or CMR. The net reclassification improvement (NRI) and integrated discrimination improvement (IDI) reflect the added performance of the model after LGE data are added.

ISSN: 1532-429X