Within the last years, a variety of different energy sources has been investigated to test their feasibility for catheter ablation of myocardial tissue. This report summarizes our experience of the use of radiofrequency alternating current (500 kHz, unipolar mode) for coagulation of ventricular myocardium in canine experiments. Under standardized in vitro conditions, we found a significant correlation between actually delivered radiofrequency energy and assessed myocardial necrosis (r = 0.87). However, this did not hold for percutaneous application of radiofrequency alternating current to the beating dog heart (r = 0.32). In the intact dog heart, the size of induced lesions paralleled catheter contact pressure which was varied until ST-egment elevation was either 0-2 mV or 5-7 mV. However, no statistical significant differences in either calculated energy or tissue impedance were observed. Under in vivo conditions, a significant improvement in the predictability of the resulting size of lesions was observed when catheter tip temperature, measured via a built-in Ni/CrNi thermoelement, was monitored (r = 0.07). Changes in tip temperature during coagulation also indicated the quality of catheter contact, catheter damage and the appearance of carbonization at the tip of the ablation catheter. Total perforation of the myocardial wall and proarrhythmogenic effects were only rarely observed. In conclusion, catheter coagulation of myocardial tissue using radiofrequency energy can be considered as safe and effective. Since changes in catheter tip temperature occurring during coagulation were found to predict the extent of induced tissue necrosis, the development of temperature controlled radiofrequency devices seems promising and necessary.