Response to Letter by Hnid
As pointed out by Dr Hnid, men were significantly more frequently smokers/past smokers as compared with women. As mentioned in the statistic section,1 the multivariate models with the memory score as the dependent variable tested the association with tobacco use, which was not found significant either in the whole study population or in men only. In the discussion, we also reported that “male gender was the strongest independent predictor of PWV in multivariate analysis (data not shown), which may contribute to the differential association of PWV with the cognitive decline regarding gender.” These multivariate models also included tobacco use, which was not found significantly associated with either pulse-wave velocity (PWV) or intima media thickness, or flow-mediated dilation (see subsequently; Tables 1 to 3⇓⇓).
Finally, as stated in the discussion, “In our present study, an association between AS and memory impairment was only observed in men suggesting possible differential effects in hypertensive patients of gender at an early stage of cognitive impairment,” “Prospective studies are required to determine whether such a gender-specific pattern may influence the transition from SMC to dementia,” and “The observational study design does not allow the ascertaining of causality.”
Multivariate Analyses of PWV, Intima Media Thickness, and Flow-Mediated Dilation Predictors
To determine which parameters were associated with PWV, intima media thickness (Table 2), and flow-mediated dilation (Tables 1 and 3⇑), we performed multivariate analyses, including gender, age, body mass index, 24-hour systolic blood pressure, diastolic blood pressure, estimated glomerular filtration rate, microalbuminuria (<30 versus ≥30 mg/24 hours), left ventricular hypertrophy, intima media thickness, PWV (as appropriate), total cholesterol, low-density lipoprotein cholesterol, triglycerides, fasting glucose, diabetes, current or past smoking, folates, von Willebrand, angiotensin-converting enzyme inhibitor, angiotensin receptor blocker, diuretics, calcium channel blockers, β-blockers, other antihypertensive drugs, and anticoagulants using the same statistical procedures described in “Methods” in our article.
A.K.-S. and P.R. contributed equally to this article and are cofirst authors.
Kearney-Schwartz A, Rossignol P, Bracard S, Felblinger J, Fay R, Boivin JM, Lecompte T, Lacolley P, Benetos A, Zannad F. Vascular structure and function is correlated to cognitive performance and white matter hyperintensities in older hypertensive patients with subjective memory complaints. Stroke. 2009; 40: 1229–1236.