| 摘要: | Background and Aim:
Atherosclerosis is a complex and chronic condition that can affect any artery in the body and cause many conditions such as carotid artery disease, coronary heart disease, peripheral artery disease and chronic kidney disease. Mostly, patients find out that they have high stage atherosclerosis only after they experience other cardiovascular disease (CVD) such as heart attack and stroke. Indexes of Carotid B-mode ultrasound (CU) such as carotid Intima Media Thickness (cIMT) and carotid Plaque Score (cPS) are reliable measures to evaluate subclinical atherosclerosis in individuals. Arterial Stiffness Index (ASI) is a test that can evaluate atherosclerosis and Ankle Brachial Index (ABI) is another inexpensive physiologic test that is being used for many years in screening and for diagnosing atherosclerosis.
So we aim to assess the association of different risk factors of atherosclerosis and serum markers with subclinical atherosclerosis indices in community dwelling individuals and non-diabetic individuals (first part a and b). We also evaluate the association and screening values of ASI and ABI in subclinical atherosclerosis (second part) in the present study.
Materials and Methods:
Data of individuals from a prospective cohort community-based study in Taiwan were used in this study. The original study that evaluate CVD risk factors has started in 2005.
Data of 312 individuals that had duplex data and lab data at the same visit for the evaluation of different serum markers in subclinical atherosclerosis in community dwelling individuals were analyzed (firs part a). For determining the serum markers and demographic information value in non-diabetic we excluded diabetics and final 276 individuals were assessed (first part b).
For the second part of the study, data of 208 individuals that had CU together with ASI and ABI measures at the same visit for determining the value of ASI and ABI in subclinical atherosclerosis were analyzed.
Demographic characteristics gathered using a questionnaire. Laboratory data, ASI and ABI, CU was performed using standard methods and devices. Simple and descriptive statistics, Fisher's exact test and ANOVA were used on study variables based on cIMT and cPS groups wherever appropriate. Multiple linear, univariate, multivariate and stepwise logistic regression and Pearson’s correlation tests performed towards the study objectives.
Results:
First part a: Our study results showed a robust relationship of age and cIMT (Pearson’s r=0.55, p<0.001). cIMT is higher in individuals with hypertension and Type 2 Diabetes Mellitus (T2DM) and also in men (p values<0.05). In multiple linear regression Systolic Blood Pressure (SBP), Waist Circumference (WC) and Hemoglobin A1c (HBA1c) was significantly associated with cIMT (p values<0.05). Age (OR=1.078, CI=1.039-1.119), SBP (OR=1.027, CI=1.008-1.047), HBA1c (OR=1.935, CI=1.312-2.852) and Blood Urea Nitrogen (BUN) (OR=1.123, CI=1.031-1.224) were determinants of cIMT and on the other hand age (OR=1.127, CI=1.086-1.169), hypertension (HTN) (OR=2.881, CI=1.533-5.416), HBA1c (OR=1.453, CI=1.082-1.951), Low Density Lipoprotein (LDL) (OR=1.012, CI=1.002-1.022) and High Density Lipoprotein (HDL) (OR=0.971, CI=0.949-0.994) were associated with cPS in separate stepwise logistic regressions.
First part b: As regarding non-diabetic individuals, Body Mass Index (BMI), Fasting Blood Sugar (FBS), Serum Uric Acid (SUA), BUN, Creatinine (Cr), HBA1c, SBP, Diastolic Blood Pressure (DBP), cPS and cIMT was higher in the group with hypertension (p values <0.05). Stepwise logistic regression analysis showed that effects of BUN and HBA1c on cIMT in the previous part (first part a) was due to the diabetic patients, although age (OR=1.078, CI=1.037-1.121) and SBP (OR=1.034, CI=1.012-1.057) is still can be found as determinants of cIMT. WC is also an independent associated factor in non-diabetic individual for high cIMT (OR=1.050, CI=1.008-1.093) in the stepwise logistic regression model. Regarding the stepwise logistic regression for determination of high cPS, results confirmed the independent associations of age (OR=1.125, CI=1.081-1.172), HTN (OR=3.533, CI=1.769-7.056), LDL (OR=1.017, CI=1.006-1.028) and HDL (OR=0.967, CI=0.943-0.992), and once again the effects of HBA1c disappeared for the non-diabetic individuals.
Second part: high measures of ASI (ASI≥61) is significantly correlated to high cIMT (cIMT>0.74 cm) (Spearman r=0.215, p=0.001) but low ABI (ABI≤1.04) is not significantly correlated to high cIMT. On the other hand only low ABI is correlated to high cPS (cPS>2) (Spearman r=0.156, p=0.024). In univariate regression models ASI showed screening values for cIMT (OR=2.842, CI=1.449-5.573, p value=0.002) and ABI showed screening values for cPS (OR=2.064, CI=1.093-3.899, p value=0.025), however multivariate analysis showed that this relationship is not independent of age, sex, HTN and T2DM so the results does not confirm that ASI and ABI to be independent determination of cIMT nor cPS.
Conclusion:
First part a: We can determine the thickness of carotid artery intima media layer by knowing age, SBP, WC and HBA1c of the individual with R 2 = 40.9%. If we are dealing with determination of a thick cIMT, BUN level with omitting WC can provide a model with area under ROC curve (AUC) = 82.8% and can serve as a good model for atherosclerosis evaluation. Knowing the lipid profile (LDL and HDL), HBA1c, age and HTN of an individual, we can obtain AUC = 84.9% for determining a high cPS. This can come to help whenever we need a model to know which person has the higher risk of carotid plaque. First part b: Effects of HBA1c and BUN is prominent on cIMT, when we are dealing with subclinical atherosclerosis in community dwelling individuals (regardless of T2DM status) but dealing with non-diabetic individuals make these measures useless in our assessments. Instead of HBA1c and BUN, when we are dealing with non-diabetic individuals, we should keep the central adiposity index such as WC alongside age and SBP into our considerations for cIMT. Age, HTN, LDL and HDL (but not HBA1c) can be used as atherosclerosis risk assessment (as in cPS) in non-diabetic individuals. Second Part: ASI has value points in screening cIMT, on the other hand, ABI can be used in screening for high cPS individuals. However, because the AUC of these tests for determination of high cIMT and cPS is less than 80%, they cannot be used as reliable screening tests for atherosclerosis evaluation. |
