To investigate the association of carotid intima-media thickness (CIMT) with abnormal glucose and central obesity.

A total of 10 207 community residents 40 years of age or older who attended the Chinese patients with Type 2 Diabetes Cancer Risk Epidemic Studies from August 16, 2011 to December 10, 2011. There were 9 556 cases elected according to inclusion and exclusion criteria. Waist, waist-to-hip ratio, body mass index and blood pressure were measured. Fasting plasma glucose, postprandial 2 hours blood glucose, glycosylated hemoglobin, high-density lipoprotein cholesterol, low density lipoprotein cholesterol, total cholesterol, triglycerides and blood uric acid were tested. CIMT was evaluated by neck vascular color Doppler ultrasonography. Respectively compare males and females carotid intima-media thickness in the quartile groups which made by waist circumference. The residents were divided into four groups: normal group (n=955), central obesity group (n=3 919), abnormal glucose group (n=465) and central obesity and abnormal glucose group (n=4 217). Single factor analysis of variance was used to compare the CIMT of each group. Pearson correlation analysis was performed for the correlation between waist circumference, HbA_1c and other indicators and CIMT. The optimal cutoff point of carotid endarteral thickening screening was analyzed by using receiver operating characteristic (ROC) curve.

(1) The CIMT values were (0.95±0.18), (1.00±0.17), (1.02±0.19), and (1.06±0.17) cm for normal group, central obesity group, abnormal glucose group, and central obesity and abnormal glucose group, respectively. CIMT level of central obesity abnormal glucose group was statistically higher than other three groups (F=156.57, P<0.001). (2) CIMT were positively correlated with waist circumference and HbA_1c (r=0.209, r=0.186, P<0.001) with Pearson test. Multiple stepwise regression showed that there was a positive correlation between CIMT and waist circumference and HbA_1c after adjusted for age, fasting plasma glucose, systolic blood pressure, LDL-C and uric acid (P<0.01). HDL-C and diastolic blood pressure were negatively correlated with CIMT (P<0.001). The areas under the receiver operating characteristic curve for predicting CIMT thickening with waist circumference were 0.615 and 0.604 in male and female, which was statistically significant (P<0.001). When waist circumference was 92.3 cm in male and 89.5 cm in female, the sensitivity was 47.7% and 57.2%, while as specificity was 68.9% and 57.6%, the Youden index was 0.166 and 0.148. The areas under the ROC curve for waist circumference combined HbA_1c as a predictor of abnormal CIMT were 0.659 and 0.642 in male and female with sensitivity at 57.3% and 37.7% and specificity 63.3% and 41.0% respectively.

The CIMT values are significantly increased in the central obesity group and abnormal glucose group. There is significant correlation between waist circumference and CIMT, waist circumference and HbA_1c are the independent risk factors for atherosclerosis. Waist circumference can be used to screen high-risk group with CIMT thickening.

Obesity; Atherosclerosis; Impaired glucose; Intima-media thickness