GISSI-HF (Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico – Insufficienza cardiaca) provides support for the use of fish oil supplements in patients with symptomatic heart failure (HF), but indicates no benefit of statin therapy in this patient population. Fish oil supplements reduced the risk of all-cause death or hospitalization for cardiovascular causes, while treatment with rosuvastatin had no affect on these outcomes.

The GISSI-HF Program included two nested studies designed to evaluate n-3 polyunsaturated fatty acids (PUFA) and statins among patients receiving optimal medical therapy for HF. Study investigators Luigi Tavazzi, MD, Policlinico San Matteo di Pavia, Italy, presented the n-3 PUFA results, and Gianni Tognoni, MD, Consorzio Mario Negri Sud, Italy, reported the rosuvastatin findings which were simultaneously published online in The Lancet.

GISSI-HF (NCT00336336) enrolled 6,975 patients with chronic New York Heart Association (NYHA) class II-IV HF, regardless of etiology and with any baseline left ventricular ejection fraction. Patients were randomly assigned to treatment with n-3 PUFA 1 g daily or placebo, and those with neither a clear indication or contraindication to statin therapy (n=4,574) were also randomly assigned to treatment with rosuvastatin 10 mg daily or placebo. The two primary endpoints were all-cause mortality and all-cause mortality or hospitalization for cardiovascular events.

Benefits with Fish Oil Supplements

After a median of 3.9 years, 27.3% of patients in the n-3 PUFA group and 29.1% of patients in the placebo group died from any cause (p=0.041). After adjusting for recent HF hospitalization, prior pacemaker implantation, and the presence of aortic stenosis, treatment with n-3 PUFA reduced the relative risk (RR) of death by 9% (HR=0.91; 95% CI 0.83 to 0.99).

Fish oil supplementation also reduced the RR of death or hospital admission for cardiovascular events by 8%, from 59.0% in the placebo group to 56.7% in the n-3 PUFA group (p=0.009). In addition, according to sub-group analyses, patients treated with n-3 PUFA were less likely than placebo recipients to be hospitalized for ventricular arrhythmia (2.8% vs 3.8%; p=0.013).

Low-dose n-3 PUFA was very well tolerated, with no excess in treatment discontinuations (28.7% vs 29.6% in the placebo group) or adverse events (2.9% vs 3.0%). A similar number of patients in the n-3 PUFA group (n=96) and the placebo group (n=92) reported gastrointestinal problems, the most common adverse event.

Neutral Effects of Statin Therapy

Treatment with rosuvastatin did not improve prognosis in the GISSI-HF trial. Patients in the rosuvastatin and placebo groups were equally likely to reach the endpoints of all-cause mortality (28.8% and 28.1%, respectively; p=0.660) and all-cause mortality or CV hospitalization (57.1% vs 56.1%; p=0.594). Moreover, there were no differences between the treatment groups in the endpoint components of CV mortality, sudden cardiac death, CV hospitalization, fatal and nonfatal myocardial infarction (MI), or fatal and nonfatal stroke.

Rosuvastatin failed to alter mortality despite clear pharmacologic activity, including beneficial effects on low-density lipoprotein cholesterol (LDL-C) and high-sensitivity C-reactive protein (hsCRP), a marker of inflammation. Whereas LDL-C levels remained relatively steady in the placebo group (increasing by 7% at year 1 and decreasing by 2% at year 3), LDL-C levels dropped by 32% at year 1 and remained reduced by 27% compared to baseline at year 3 in the rosuvastatin group (p<0.001). After 3 months of therapy, the mean hsCRP level dropped by 4.6% in the placebo group and 16.6% among those receiving rosuvastatin (p=0.020).

GISSI-HF joins the CORONA (Controlled Rosuvastatin Multinational Trial in HF) trial in showing that statin therapy does not lessen the risk of death in patients with HF [Kjekshus J et al. N Engl J Med 2007]. Compared with GISSI-HF, CORONA evaluated an older group of patients (mean age, 68 years vs 73 years) with a greater prevalence of ischemic HF (40% vs 100%) and more severe disease (37% vs 63% NYHA III/IV). However, the findings of GISSI-HF mirror those of CORONA: rosuvastatin has a neutral effect in HF patients.

Clinical Implications

The role of statins in HF, and the clinical implications of GISSI-HF in particular, will be a source of debate within the cardiology community, said Philip A. Poole-Wilson, MD, Imperial College, London, UK. Given the findings of CORONA and GISSI-HF, Prof. Poole-Wilson argued that patients with symptomatic HF (NYHA II-IV) should not be started on statin therapy, and those who are already taking statins should have these agents withdrawn. For patients who have structural or functional defects but no HF symptoms (NYHA I), the clinical consequences of GISSI-HF are less clear.

Prof. Tognoni had a more conservative interpretation of the GISSI-HF findings. Although statins should not be given to patients with HF of non-ischemic etiology, physicians should carefully consider the benefits of continuing or discontinuing statin therapy in patients with ischemic HF, he said. For example, discontinuing statin therapy may improve compliance with other concurrent, evidence-based treatments.

Overall, the neutral findings of GISSI-HF do not diminish the valuable role of statins in CV risk reduction. "Patients with coronary heart disease without HF must be started and maintained on statin therapy," Prof. Poole-Wilson concluded.