Table 3

Multivariable linear regression models to test the relationship between the maximum rise of NT-proBNP and other clinical variables

Group	Variable	Coefficient (β)	SE	95% CI	P value
All patients	Intercept	4.43	0.01	4.41 to 4.45	<0.001
r²=0.17; adjusted r²=0.16
	LVEF <45%	−0.30	0.04	−0.38 to −0.21	<0.001
	SAVR	0.21	0.04	0.13 to 0.30	<0.001
	PHF	0.20	0.04	0.11 to 0.28	<0.001
All-SAVR	Intercept	4.38	0.02	4.34 to 4.41	<0.001
r²=0.17; adjusted r²=0.16
	Age	0.27	0.05	0.17 to 0.37	<0.001
	LVEF <45%	−0.20	0.05	−0.30 to −0.10	<0.001
	PHF	0.17	0.06	0.06 to 0.28	0.002
	CK-MB D1	0.12	0.05	0.01 to 0.23	0.03
TAVI	Intercept	4.29	0.03	4.23 to 4.35	<0.001
r²=0.22; adjusted r²=0.20
	LVEF <45%	−0.37	0.09	−0.55 to −0.18	<0.001
	CK-MB D1	0.23	0.09	0.04 to 0.41	0.02

CK-MB D1, creatine kinase MB the 1st postoperative day; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-B-type natriuretic peptide; PHF, postoperative heart failure; r², the percentage of the response variable variation that is explained by the model; SAVR, surgical aortic valve replacement.