Objective To evaluate the effect of CPB with or without cardioplegia on hemodynamics and myocardial function. Methods Thirty NYHA class III patients undergoing mitral valve replacement were randomly divided into three groups of ten each: group Ⅰ received no blood cardioplegia; group Ⅱ received tepid blood cardioplegia solution and group Ⅲ received cold blood cardioplegia solution. The patients were premedicated with pethidine 50mg and scopolamine 0.3mg. Swan-Ganz catheter was inserted via right interval jugular vein into pulmonary artery and radial artery was cannulated under local anesthesia before anesthesia. Anesthesia was induced with midazolam 0.05-0.01 mg.kg-1, fentanyl 15-20?g. kg-1 and pancuronium 0.12mg.kg-1 and maintained with fentanyl, midazolam and pancuromium. Patients in group Ⅰ received no cardioplegic solution. Ascending aorta was not cross-clamped. Body temperature was reduced to 32℃-33 ℃ (naso-pharyngeal T) . The empty heart was beating at 40-60 bmp. In group D and Ⅲ cardioplegic solution (modified St. Thomas solution) was added to blood from oxygenator (in the proportion of 1:4).In group Ⅱ body temperature was reduced to 32℃-33℃.Tepid (32℃ ) hyperkalemia blood cardioplegic solution was infused at 200-250ml.min-1 after ascending aorta was cross-clamped. After cardioplegic arrest hypokalemic blood cardioplegia solution was infused. In group Ⅲ body temperature was reduced to 28℃-29℃. Cold (8℃) hyperkalemic blood cardioplegic solution was infused at 200-250ml. min-1, after ascending aorta was cross-clamped. After cardioplegia arrest cold hypokalemic blood cardioplegia solution was infused every 20 min. Hemodynamic parameters (MAP, MPAP, CO, CI, SVRI, PVRI, LVSWI and RVSWI) were recorded before anesthesia (T0), before CPB (T1), 10, 30, 60 min after termination of CPB (T2-T4 ) and at the end of surgery (T5 ) . Results Demographic data including age, gender, body weight and body surface area were comparable between the three groups. CI in group 1 and Ⅱ was significantly higher at T2 -T5 than that in group Ⅲ . SVRI after CPB in group Ⅰ was significantly lower than that in group Ⅲ . Conclusion The myocardial function after CPB without cardioplegia is better than that after CPB with cold cardioplegia, but is not significantly different from that after CPB with tepid blood cardioplegia.