. Author manuscript; available in PMC: 2022 Jul 14.

Published in final edited form as: Circ Cardiovasc Imaging. 2021 Jul 14;14(7):e011753. doi: 10.1161/CIRCIMAGING.120.011753

Table 3.

The association between continuous CAC/CFPWV and subtle white-matter injury including sex interaction tests: Framingham Heart Study third-generation cohort (N = 1052)

	Mean FA				lnFW				lnPSMD
Methods	Estimate	95%CI	P value	P value for sex interaction	Estimate	95%CI	P value	P value for sex interaction	Estimate	95%CI	P value	P value for sex interaction
20^th–80^th percentile of lnCAC
Model 1	−0.139	(−0.237, −0.041)	< 0.01	0.020	0.123	(0.023, 0.185)	0.02	0.285	0.130	(0.033, 0.228)	< 0.01	0.756
Model 2	−0.114	(−0.211, −0.018)	0.02	0.043	0.086	(−0.013, 0.185)	0.09	0.326	0.097	(−0.002, 0.196)	0.054	0.628
20^th–80^th percentile of niCFPWV
Model 1	−0.009	(−0.122, 0.104)	0.88	0.045	0.152	(0.038, 0.266)	< 0.05	< 0.05	0.132	(0.019, 0.244)	0.02	0.756
Model 2	−0.003	(−0.126, 0.120)	0.96	0.053	0.124	(−0.002, 0.251)	0.06	0.084	0.102	(−0.024, 0.227)	0.11	0.666

The values were expressed as adjusted mean per 20th-80th percentile higher coronary artery calcium (CAC) in Agatston score (log-transformed), or negative inverse carotid-femoral pulse wave velocity (CFPWV) in meter/sec.

The set of adjusting covariates were sex, age at CT/tonometry, age at CT/tonometry squared, and time between at CT/tonometry and MRI, current smoker (yes/no), SBP (mmHg), diabetes (yes/no), total cholesterol (mg/dL), medications for hypertension (yes/no) and dyslipidemia (yes/no), and total cranial volume (mL) (Model 2)