1 In the last decade, a number of clinical observations have questioned the ecacy of certain class I antiarrhythmic drugs against ischaemia-induced ventricular ®brillation. The eects of three drugs of this class, disopyramide (Ia), lignocaine (Ib) and¯ecainide (Ic) on the vulnerability to ®brillation during experimental ischaemia were investigated. 2 The study was carried out in anaesthetized, open-chest pigs (n=8 for each of the drugs, in addition to the control group, n=6). Vulnerability to ®brillation was evaluated by measuring electrical ®brillation threshold (EFT) by means of stepwise increased intensity of wide (100 ms) diastolic impulses applied to the ischaemic tissue at a 180 beats min 71 rate. Monophasic action potential (MAP) duration and conduction time in the ischaemic region were also measured. 3 EFT determinations were performed before and during periods of ischaemia induced by complete occlusion of the left anterior descending coronary artery near its origin. Ischaemic periods of increasing duration (30, 60, 90, 120, 150 s) were induced to determine the electrophysiological changes, of EFT especially, leading to ®brillation. 4 In the absence of ischaemia, all three drugs, administered by intravenous route (1 mg kg 71 plus 0.04 mg kg 71 min 71 ) increased EFT to a similar extent (from approximately 7 to 10 mA), despite a 25% prolongation of conduction time. 5 During ischaemia, none of the drugs prevented the fall in EFT towards 0 mA, resulting in spontaneous ®brillation. After 30 s of ischaemia, they no longer had any capacity for raising EFT and, after 60, 90 and 120 s of ischaemia, the decrease in EFT was exacerbated. This accelerated reduction in EFT shortened the time to onset of ®brillation (after 120 s of ischaemia, 62.5% of ®brillations with ecainide instead of 12.5 under control conditions, 75% instead of 25 with lignocaine and 50% instead of 25 with disopyramide). The reduction in MAP duration due to ischaemia was also signi®cantly accelerated (at 60 s, 178+5 ms instead of 192+4 with¯ecainide, 175+3 ms instead of 194+5 with lignocaine and 180+5 ms instead of 196+3 with disopyramide) and the slowing of conduction was made worse (prolongation of conduction time by 70% instead of 50). 6 In conclusion, the anti®brillatory properties normally manifested by these drugs are ®rst suppressed, then inverted by ischaemia, depending on oxygen debt varying with severity and duration of ischaemia.