Sudden death (SD) has a major impact on mortality (M) in patients with left ventricular systolic dysfunction (SyD). In GISSI-Prevenzione, treatment with n-3 polyunsaturated fatty acids (PUFA) reduced M and SD in post-MI patients, but their effect in patients with SyD is unknown.

11,323 patients with prior MI and NYHA class < or = II were recruited. After excluding patients with no ejection fraction (EF) measurement (1684), and those with missing data (n=9), 9630 patients were available for analysis. Multivariate Cox regression adjusted models were fitted.

Compared to patients with EF > 50%, SyD patients had higher M (12.3% vs. 6.0%) and SD (3.4% vs. 1.4%) rates. PUFA reduced M similarly in patients with (RR 0.76 (0.60-0.96) P=0.02) and without SyD (RR 0.81 (0.59-1.10) P=0.17) (heterogeneity tests P=0.55). In contrast, the effect on SD was markedly asymmetrical: PUFA produced a marked reduction (RR 0.42 (0.26-0.67) P=0.0003) of risk in SyD patients whereas the effect was less evident (RR 0.89 (0.41-1.69) P=0.71) in patients with EF > 50% (heterogeneity tests P=0.07). There was a significant increase in SD with worsening EF (P test for trend=0.02), the benefit on SD in patients with EF < or = 40% being 4-fold higher than in those with EF > 50%.

Increasing SyD is associated with elevated risk of SD and with increasing benefit from PUFA. The effect of PUFA on SD reduction was greater in patients with SyD. Prospective trials testing the effect of PUFA in populations with SyD are required.