A new ultrasonic angioplasty ablation catheter connected to a 19.5 kHz. 25 W transducer was tested in vitro for its ability to disrupt 12-h. 24-h, and 5-day-old whole blood thrombi (n = 45.697 mg +/- 223 mg) and fibrin thrombi (n = 45.338 mg +/- 133 mg), as well as 5-day-old cadaver thrombi (n = 8.270 mg +/- 71 mg) within 10 min. Five of each age were used as control thrombi in which the catheter was moved back and forth without ultrasound emission. The size of ablated thrombus particles was measured by a laser device. The power output at the end of the catheter was assessed calorimetrically. The loss of weight of whole blood thrombi was between 429 (74%) and 524 mg (91%) (p < 0.01, whole blood thrombi vs. control thrombi) and between 302 (85%) and 314 mg (95%) (p < 0.05) for fibrin thrombi, respectively. Thrombus age did not prove to be a highly significant influencing factor. The disruption rate for whole blood thrombi was 0.75 to 1.05 mg/s and for fibrin thrombi 0.69 to 0.7 mg/s. It was only 0.09 mg/s for the cadaver thrombi. 93% of all particles ablated from whole blood thrombus ranged between 0-5 micron, less than 0.2% between 30-150 microns. For fibrin thrombi 69% of all particles were < 10 microns (25% between 10-20 microns). Only 0.02% ranged between 300-600 microns, which was similar for cadaver thrombi. The mean measured power output at the catheter tip was 5.9 W compared to the power output of 25 W at the ultrasound generator.(ABSTRACT TRUNCATED AT 250 WORDS)