TABLE 2.

Multivariable-Adjusted Associations of BMI and CFR With Adverse Events

					Model Statistics
Sequential Models for Composite Events	BMI* Hazard Ratio (95% CI)	p Value	CFR† Hazard Ratio (95% CI)	p Value	Likelihood Ratio Chi-Square	p Value‡	Harrell’s C-Index
Multivariable model 1: demographic factors	–	–	–	–	59.20	<0.001	0.65
Multivariable model 2: model 1 + clinical factors	–	–	–	–	108.67	<0.001	0.71
Multivariable model 3: model 2 + BMI	1.31 (1.05–1.64)	0.018	–	–	114.02	<0.001	0.71
Multivariable model 4: model 3 + CFR	1.20 (0.95–1.50)	0.125	1.95 (1.41–2.69)	<0.001	132.17	<0.001	0.74

Model 1: adjusted for demographic factors including age, sex, and race. Model 2: adjusted for demographic and clinical factors including age, sex, race, hypertension, diabetes, peripheral vascular disease, chronic obstructive pulmonary disease, tobacco use, atrial fibrillation, estimated glomerular filtration rate, b-blocker use, and left ventricular ejection fraction. Model 3: adjusted for demographic and clinical factors as in model 2, and body mass index (BMI). Model 4: adjusted for demographic, clinical factors and BMI as in model 3, and coronary flow reserve (CFR). No significant interaction was observed between CFR and BMI.

^*Per +10 U.

^†Per −1 U.

^‡For likelihood ratio test between sequential models; the first value is for comparison of multivariable model 1 to the univariate analysis of BMI.

CI = confidence interval.