After adjusting for demographic and vascular risk factors (Model 2), the association remained marginally significant (0.0046 vs. 0.0034 mL/TIV, p = 0.051), suggesting a potential independent effect of AF on cerebral white matter changes.
Although attenuated in the fully adjusted infarct model, the association remained of borderline significance (Model 4: 0.0078, p = 0.059).
0.0041 mL/TIV), although the differences did not reach conventional levels of significance ( p = 0.071 and p = 0.285 in Models 2 and 4, respectively).
Similarly, lobar CMBs showed a trend toward increased odds in men with AF in the unadjusted model (OR = 2.4, 95% CI: 0.9–6.3, p = 0.089), but this association was weakened and lost statistical significance in adjusted models (Model 4: OR = 1.2, 95% CI: 0.3–4.2, p = 0.759).
In terms of lifestyle factors, current smoking and alcohol risk consumption (>98 g/week) were numerically higher in the AF group (10.6% and 37.6%, respectively) than in non-AF individuals (6.7% and 32.6%), though these differences did not reach statistical significance ( p = 0.169 and p = 0.358, respectively).
Nevertheless, given the modest sample size and the borderline level of statistical significance, these findings should be interpreted cautiously and regarded as hypothesis-generating, warranting confirmation in larger longitudinal studies with detailed perfusion imaging.