Type 2 diabetes compromises the value of non-invasively measured augmentation index in predicting the severity of coronary artery disease: a hospital-based observational study.

BACKGROUND

Central hemodynamic indices have been demonstrated to correlate with coronary artery disease (CAD). However, in the context of type 2 diabetes mellitus (DM), this correlation has not been fully illustrated. Therefore, this study was employed to investigate the impact of DM on the correlation between aortic augmentation index and the severity of coronary artery disease.

METHODS

In this study, we analyzed 197 patients who underwent coronary angiography at Anzhen Hospital from September 2015 to January 2016. Central hemodynamics were non-invasively measured with BPro® device (Health STATS, Singapore). The severity of CAD was defined according to SYNTAX scores. Type 2 diabetes was defined according to ADA guidelines. AIx@75 was defined as AIx normalized to a heart rate of 75 bpm. Receiver operating characteristics (ROC) determined the optimal cut-off value of AIx@75 to predict moderate to severe CAD. Multivariate logistic regression analysis evaluated the correlation between central hemodynamic parameters and CAD severity.

RESULTS

Eighty-four (42.6 %) of the studied subjects were diabetic patients. Our findings were that (1) AIx@75 was significantly correlated with SYNTAX. (ROC analyzed AUC: 0.638, 95 % CI 0.555-0.721, p < 0.05). The cut-off value of AIx@75 to predict moderate-to-severe CAD as SYNTAX score more than 22 was 71.45. (2) In non-diabetic patients, correlation analysis revealed that AIx@75, augmentation pressure and peak relative time were significantly correlated with CAD severity (p < 0.05). After adjustment, AIx@75 remained as the only independent predictor of moderate-to-severe CAD (odds ratio 1.099, 95 % CI 1.028-1.176, p < 0.05). (3) In diabetic patients, the correlation between central hemodynamic parameters and the severity of CAD did not exist.

CONCLUSIONS

Aortic augmentation index was significantly related to the severity of CAD and was an independent predictor of severe CAD. However, clinical practitioners should note that its value in DM populations was compromised.