In a consecutive series of 274 AMI cases ventricular tachycardia (VT), defined as three or more ventricular premature beats (VPBs) in succession but not VPBs, has been used as the indication for ventricular fibrillation (VF) prophylaxis. No primary VF occurred, and this fits with the hypothesis of VT as a sufficient indication for prophylaxis against primary VF. Six patients developed complicating VF (preceded by rales or hypotension but not frank pulmonary edema or shock). Four of the six patients (67%) had VT (0-1.5 hours) before VF, while the mean VT incidence of the six corresponding monitoring periods in 247 non-VF patients was 5%. Three of the four VT patients were on lignocaine/procainamide when VF developed. Thus, VT is acceptable as the only warning arrhythmia even in complicating VF but antiarrhythmic drugs do not seem to have the same prophylactic efficacy in complicating VF as in primary VF. Another 21 patients developed VF during shock, frank pulmonary edema or manipulating a pacemaker catheter within the heart.