MULTIVARIATE ANALYSIS OF FACTORS AFFECTING COMMON CAROTID ARTERY INTIMAL MEDIAL THICKNESS
Author Block A Johri, V Kunanithy, D Armstrong, D Brouillard, M Matangi
BACKGROUND: To assess using multivariate analysis the relative importance of traditional vascular risk factors in predicting common carotid artery (CCA) intimal medial thickness (IMT) in both men and women.
METHODS: CARDIOfile, our cardiology database and reporting system was searched for all patients who had blood work, carotid duplex imaging, ankle-brachial index measurement and cardiology consultation within 90 days of each other. For simplicity we only assessed the L) CCA IMT as in our database and consistent with published data the L) CCA IMT is on average thicker. All data were entered into our XLSTAT 2011 program for statistical analysis. Multivariate analysis was performed using multiple linear regression. The unpaired t-test was used to assess differences between mean values. A P value of <0.05 was considered significant for either test.
RESULTS: See Table 1. The mean L) CCA IMT for women (n=127) was 1.11±0.55mm and for men (n=298) 1.24±0.62mm, (P=0.046). Using multivariate analysis and the variables selected we were only able to predict 20.4% in men and 21.2% in women of the actual CCA IMT value.
CONCLUSIONS: Using multivariate analysis, age was the most powerful predictor of CCA IMT in both men and women (P<0.0001). In men ankle-brachial index and HB A1C were significantly associated with L) CCA IMT. In women triglycerides were also an independent predictor of CCA IMT. Clearly as we selected only continuous variables, there are other variables such as smoking, diet and exercise that we did not test. Hence we were only able to account for approximately 20% of the CCA IMT value.
|Males UVA (N=298)||Males MVA (N=298)||Females UNA(N=127)||Females MVA(N=127)|