Indices for Prediction of Coronary Sclerosis with Particular Reference to Apo-B/Apo-AI Ratio

抄録

To determine the most effective index for predicting coronary sclerosis, various lipoproteins and apoproteins were examined. The study subjects included the following four groups: (1) 395 males and 127 females who underwent physical examinations, (2) 49 males and 10 females who were diagnosed as old myocardial infarction confirmed by coronary arteriography, (3) 45 males and 15 females with effort angina, and (4) 49 males and 10 females with vasospastic angina. Our results can be summaried as follows.
1) Using a box-whisker plot indicated that the LDL-C/HDL-C ratio and the Apo-B/Apo-AI ratio adequately reflected the level of coronary sclerosis.
2) An ROC curve demonstrated that the Apo-B/Apo-AI ratio was most accurate in discriminating coronary heart diseases.
3) The study on the effects of background factors on the Apo-B/Apo-AI ratio which used LOWESS and Q-Q plots did not disclose any factors that would affect the ratio except obesity which tended to elevate it.