Radiofrequency coagulation of ventricular myocardium: improved prediction of lesion size by monitoring catheter tip temperature.

To assess the importance of voltage, current, impedance and catheter tip temperature for the prediction of the size of tissue injury induced by transcatheter radiofrequency application, radiofrequency pulses (500 kHz) were delivered both in vitro and in vivo to isolated ventricular preparations and the intact canine heart, respectively. Radiofrequency coagulations were performed using unipolar electrode configuration. Besides measurements of current and voltage which were used to calculate the delivered power and tissue impedance, the catheter tip temperature was monitored during each application using specially designed 6F USCI catheters with a built-in nickel/chromium-nickel thermoelement. Lesion dimensions were measured and the correlation between lesion volume and delivered radiofrequency energy, maximum changes in catheter tip temperature and the integral of the temperature curve were calculated. First, in a pilot in vitro investigation, 50 radiofrequency coagulations (3.2 W-22.4 W, pulse duration 10 s) were performed in ventricular preparations from freshly excised dog hearts. The correlation between applied radiofrequency energy and lesion volume was 0.87; the correlation between maximal catheter tip temperature and lesion volume was 0.82; the correlation between temperature integral and lesion volume was 0.9. In the intact dog heart, 44 radiofrequency pulses were delivered to the left and right ventricular endocardium in 12 anaesthetized dogs (exposure time: 10 s). Delivered power ranged between 5.6 W and 24.6 W; tissue impedance varied between 92 omega and 364 omega; lesion volume measured 0-273 mm3; developed peak temperatures ranged from 16.25 degrees C to 196 degrees C. The calculated integral beneath temperature curves measured 126-1971 degrees C.s. The correlation between applied radiofrequency energy and lesion volume was 0.32; the correlation between maximal catheter tip temperature and lesion volume was 0.61. Temperature integral correlated best with the assessed volume of myocardial necrosis (r = 0.7). No significant arrhythmogenic or haemodynamic side-effects were observed. Macroscopic examination showed a central depression surrounded by a zone of homogeneous coagulation. Vaporization and crater formation up to a depth of 4 mm were observed following three radiofrequency discharges. In two of these cases, rapid changes and oscillation of catheter tip temperature were observed. Thus, monitoring of catheter tip temperature during radiofrequency energy application improves the prediction of lesion size. In addition, temperature monitoring might improve the safety of the procedure with respect to the risk of perforation.