Objective To test the accuracy of American College of Cardiology (ACC)/American Heart Association (AHA) guidelines for preoperative cardiovascular evaluation for noncardiac surgery in Chinese patients and to compare ACC/AHA guidelines with Goldman index and Lee index. Methods From January to December 2003, all patients aged ≥70 yr or patients aged 40-69 yr with a history of cardio- and cerebro-vascular disease, abnormal ECG or diabetes scheduled for noncardiovascular surgery were included in this study. A total of 1 248 patients were eligible. Their clinical data including demographic data, history of cardiovascular disease, routine physical examination and laboratory tests, the scheduled surgery and type of anesthesia were collected. The patients were then evaluated for cardiac risk and classified according to ACC/AHA guidelines (high, moderate, low and no risk), Lee index (class I -IV ) and Goldman index (class I - III ). The cardiac risk of the scheduled surgery was then stratified according to ACC/AHA guidelines. The patients were followed up until discharged from hospital. Cardiac events were defined as cardiac death, myocardial infarct, myocardial ischemia, minor myocardial cell injury, ventricular dysfunction and serious arrhythmia. Likelihood ratio of the 3 methods was calculated. Risk factors for adverse events were identified by univariate analysis and multivariate Logistic regression analysis. Results Of the 1 248 patients 694 were male and 554 female. Their age ranged from 40-102 years (mean age 65.4 yr). 44.7 % of the patients were aged ≥ 70 years. High risk operation accounted for 6.3 % and emergency operation 7.9% . One patients died of cardiac event and ten patients of other causes. Seventy-three perioperative cardiac events occurred in 53 patients. The morbidity rate was 4.2% . Goldman index and ACC/AHA cardiac risk stratification were correlated with adverse cardiac outcomes ( P