Objective: The aim was to produce a model of low output congestive heart failure by rapid pacing in rabbits. Methods: To perform rapid pacing in rabbits, a custom made pacemaker was developed which is light (about 80 g) and can pace at up to 400 beats.min-1 for more than two weeks. A thoracotomy was done and two electrodes were sutured onto the left ventricle. A central venous pressure line was chronically implanted. With the use of this pacemaker, rabbits were paced at 350-400 beats-min-1 for several weeks. Results: Central venous pressure increased from 1.4(SEM 0.2) to 6.4(0.5) mm Hg (p<0.01, n=14). After pacing for 16.1(1.6) d, haemodynamic studies were performed under anaesthesia with thiamylal sodium. Left ventricular end diastolic pressure was higher in the paced rabbits (n=10) than in the control rabbits which underwent sham operation but were not paced (n=6), at -0.6(0.6) v 19.3(2.0) mm Hg (p<0.01). Cardiac output [673(56) v 536(45) ml.min-1, p<0.10] and +dP/dt [1433(97) v 722(51) mm Hg.s-1, p<0.01] were lower in the paced rabbits (n=7-8) than in the control rabbits (n=6). The paced rabbits had more ascites [1.9(1.0) v 45.9(18.9) ml, p<0.05] and pleural effusion [0.4(0.3) v 12.9(6.7) ml, p<0.10] than control rabbits. Plasma noradrenaline was higher in the paced rabbits (n=11) than in the control rabbits (n=7), at 1.59(0.43) v 0.60(0.05) ng.ml-1 (p<0.05). The ratio of wet heart weight or lung weight to body weight was higher (p<0.01) in the paced rabbits than in the control rabbits. Conclusions: Chronic biventricular congestive heart failure can be produced in rabbits by rapid pacing.