Abstract
BACKGROUND: The vascular complications of outpatients with diabetes at ordinary hospitals vary. Ischemic heart disease is barely predictable after treatment using previously reported therapeutic indices. We developed a simple and noninvasive screening method to evaluate the possibility of ischemic heart disease in patients with diabetes. METHODS: Five years of clinical data from 337 outpatients (196 males and 141 females) with diabetes were analyzed. Twenty-three males and 14 females had ischemic heart disease. We examined the possibility of predicting ischemic heart disease after analyzing this population. The analyzed laboratory data included the following: minimum value of right or left ankle-brachial indices (ABI), maximum value of right or left pulse wave velocities (PWV), aortic calcification diagnosed on plain chest radiographs, plaque score (PS), maximum value of intima media thickness at the cervical artery (IMT), electrocardiographic (ECG) ischemic changes (including ST-T changes or abnormal Q waves, which were re-examined by a cardiologist), HbA1c, low-density lipoprotein cholesterol (LDL-C), uric acid (UA), urine albumin, age, sex, disease duration, and body mass index. All data were subjected to multivariate logistic regression analyses. RESULTS: The presence of ECG ischemic changes, aortic calcification, minimum ABI, maximum IMT, LDL-C, and UA were evaluated in multivariate logistic regression analysis with the onset of ischemic heart disease. The receiver operating characteristic curve indicated an area under the curve of 0.879 (0.820 - 0.938; P = 0.00). CONCLUSIONS: Ischemic heart disease could be predicted in patients with diabetes using a combination of results from conventional physical and laboratory tests.