Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

European Society of Cardiology Congress 2008

Munich, Germany / August 30-September 3, 2008

New clinical information has been generated by results of the two-part GISSI-HF (Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto miocardico-Heart Failure) trial. Despite the fact that neither result is expected to have a major impact on clinical practice, both findings provide insight into the pathophysiology of heart failure (HF) and expand information about the opportunities for treatment when cardiovascular disease (CVD) reaches its end-stage components.

In one part of the study, rosuvastatin was compared to placebo in patients with HF. The inability of rosuvastatin, a highly effective, single-agent LDL-lowering therapy, to improve clinical outcomes in patients with chronic HF (CHF) provides compelling evidence that HF events are not driven by thrombotic processes. In the other part of GISSI-HF, omega 3-polyunsaturated fatty acids (PUFAs) did reduce the risk of events relative to placebo, but the benefits were modest.

“Although statin therapy lowers concentrations of LDL-C, is well tolerated, and seems reasonably safe, it does not produce meaningful improvements in survival in patients with CHF,” according to Dr. Gregg C. Fonarow, Ahmanson-UCLA Cardiomyopathy Center, Los Angeles, California. Commenting in an editorial that accompanied the GISSI-HF results in the New England Journal of Medicine (GISSI-Heart Failure Investigators. N Engl J Med 2008;359:Epub ahead of print), Dr. Fonarow expressed an opinion shared by several commentators here at the European Society of Cardiology (ESC) where the final results were presented simultaneously with publication.

GISSI-HF and CORONA Findings

The expectation of benefit was based on previously published observational studies in HF patients that had associated intensive lipid lowering with improvements in ventricular function, HF status and clinical outcomes, but GISSI-HF is the second large, well-controlled, multicentre prospective trial that was unable to associate lipid lowering with benefit. In the previous trial, called CORONA (Controlled Rosuvastatin Multinational Trial in Heart Failure), 5011 patients with New York Heart Association (NYHA) class II to IV HF with a left ventricular ejection fraction (LVEF) <40% and a history of ischemic CVD were randomized to rosuvastatin 10 mg or placebo. After a median of 33 months of follow-up, there was no significant advantage for the statin for the composite end point, CV mortality or non-fatal coronary events, such as myocardial infarction (MI), or non-fatal stroke.

In GISSI-HF, 4574 patients with NYHA class II to IV HF, irrespective of cause or LVEF, were randomized to rosuvastatin 10 mg or placebo and analyzed (Figure 1).

Figure 1. GISSI-HF Study Design

Patients were followed for a median of 3.9 years, during which time there was no significant difference in the primary end point of time to death and time to death or admission to hospital for CV reasons. Although the authors of this study, like the authors of CORONA, emphasized that rosuvastatin was safe and well tolerated even in patients with advanced HF, they concluded that HF could not be considered an indication for treatment.

“In combination with results of the CORONA study, the results of our data prove that lipid lowering with rosuvastatin 10 mg/day does not improve outcomes in patients with CHF regardless of age, etiology, or level of systolic function,” reported Dr. Gianni Tognoni, Consorzio Mario Negri Sud, Chieti, Italy. “These data suggest that the causes of death from HF are not strongly mediated by the pathogenic mechanisms that rosuvastatin modifies” (Figure 2).

Several past post-hoc analyses of large trials have suggested that patients with HF might benefit from statins, but these analyses are hypothesis-generating only. The results of GISSI-HF once again demonstrate that drawing conclusions about drug efficacy from any type of observational studies or post-hoc analyses is not reliable, irrespective of what statistical adjustments are made.

Analysis of the Findings

The absence of benefit in GISSI-HF is particularly compelling because the study agent was rosuvastatin, which is not only the most effective statin for reducing LDL but particularly strongly associated with many of the pleiotropic actions which might be expected to modify risk of HF progression. These include antifibrotic, anti-oxidant and anti-inflammatory activities which, in turn, have the potential to improve endothelial function, inhibit neurohormonal activation and prevent cardiac arrhythmias. In GISSI-HF, rosuvastatin reduced LDL on average by 32% within the first year and by 27% over the course of three years of follow-up. In a subgroup evaluated for high-sensitivity C-reactive protein (hsCRP), the reductions in this marker of inflammation were significant for rosuvastatin relative to baseline and to placebo (P=0.0195). Despite this activity, there was no substantial protection against events relative to placebo.

One concern expressed by several experts at the ESC is that results of the rosuvastatin arm of GISSI-HF might be misconstrued as demonstrating a weakness for statin therapies. Dr. Philip Poole-Wilson, Professor of Cardiology, Imperial College, London, UK, was asked by the ESC to provide a formal critique of GISSI-HF, and one of his key conclusions from GISSI-HF was that despite the lack of benefit in HF, “patients with coronary heart disease must be given lipid-lowering therapy.”

There are evidence-based and irrefutable findings that rosuvastatin and other statin therapies are highly effective at preventing progression up to the point at which damage to pump function leads to HF, according to Dr. Poole-Wilson. GISSI-HF in no way diminishes the critical need for treatment before the end-stage condition of HF develops. Although Dr. Poole-Wilson conceded that GISSI-HF does not reveal whether it is appropriate to withdraw statin therapy once patients develop HF, he did conclude that the results of both GISSI-HF and CORONA demonstrate that there is no justification for starting statins once HF is already present.

ASTEROID

In ASTEROID (A Study to Evaluate the Effect of Rosuvastatin on Intravascular Ultrasound-Derived Coronary Atheroma Burden), a stratification of effects on plaque regression correlated very strongly with degree of lipid lowering. Although plaque stabilization was observed when LDL levels reached 2.0 mmol/L, plaque regression required even lower levels. As a highly potent lipid-lowering agent, rosuvastatin can often reach these types of reductions with a single pill. It is the absence of benefit with this strategy in GISSI-HF which has provided a strong basis for concluding that lipid lowering is not an effective strategy for reducing HF risk.

Based on the results of GISSI-HF and CORONA, neurohormonal enlargement of the heart, which progressively diminishes pump efficiency and increases the risk of arrhythmias, appears to proceed independently of LDL status. As one of the experts who addressed the ESC, Dr. Eugene Braunwald, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, compared lipid lowering in HF to closing the barn door after the horse escaped.

Although this remark was focused on LDL, such observations may also be relevant to hsCRP in the wake of premature discontinuation of JUPITER (Justification for the Use of Statins in Primary Prevention: an Intervention Trial Evaluating Rosuvastatin). JUPITER was a double-blind clinical trial that randomized 15,000 patients with relatively low LDL but elevated hsCRP to rosuvastatin or placebo. It was stopped in March 2008 by the Data Safety Monitoring Board on the basis of “unequivocal evidence of a reduction in CV morbidity and mortality” among those treated with rosuvastatin compared with placebo. In GISSI-HF, the significant reduction of both LDL and hsCRP suggest that neither is relevant to HF.

“The GISSI-HF study suggests that the opportunity to intervene with lipid lowering is before patients progress to HF,” agreed Dr. Poole-Wilson, who concluded that patients who develop HF could no longer be considered candidates for lipid lowerin
I-HF: Cause of Death

<img2117|center>

The Other Half of GISSI-HF

In the other half of the GISSI-HF study, patients were randomized to 1 g/day of PUFA or placebo. At the end of a median of 3.9 years, there was a modest 9% reduction (hazard ratio 0.91, 95% CI: 0.833-0.998; P=0.041) in the same composite end point employed in the rosuvastatin section of the GISSI-HF trial. The therapy was relatively well tolerated, although there was a slightly greater number of discontinuations for gastrointestinal events in the PUFA group.

“The advantage of PUFA over placebo for both fatal events and hospital admissions for CV events suggests that it is having a favourable although modest effect on the mechanisms that contribute to HF,” reported Dr. Luigi Tavazzi, Hospitals of Care and Research, Cotignola, Italy. Senior author of this portion of the GISSI-HF study, he emphasized that the benefits of PUFA were achieved on top of standard clinical care, which included ACE inhibitors or an angiotensin receptor blocker (ARB) in almost 95% of patients, beta blockers in 65% of patients, and spironolactone in almost 40% of patients.

Given that the benefit was observed without significant adverse effects and on top of standard-of-care therapies, Dr. Fonarow concluded that PUFA “should join the short list of evidence-based life-prolonging therapies for HF.” The same point was made by Dr. Michel Komajda, Université Pierre et Marie Curie, Paris, France, who was invited by the ESC to discuss the results. Although he suggested that the benefits of fish oil in HF “are still a bit of a mystery,” he also suggested that fish oils were an appropriate therapy in patients who could tolerate them because of the benefit observed in GISSI-HF.

Summary

The GISSI-HF study has provided important new information about the management of patients with HF despite the fact that its effects on clinical practice are expected to be modest. In one of the two independent placebo-controlled randomizations, the statin did not have any effect on outcome despite significant reductions in LDL and hsCRP. Although rosuvastatin was well tolerated in the study, the results suggest that events caused by HF are not dependent on thrombotic processes for which intensive lipid lowering is most strongly associated. In the other randomization, fish oils were associated with a significant reduction in clinical events over 3.9 years of follow-up, but data indicate that more than 50 patients must be treated for every one event prevented. Still, for patients who tolerate fish oils, the benefit of this approach is now evidence-based.

Questions and Answers

The following question-and-answer session was conducted with Dr. Jonathan Howlett, Medical Director, Cardiac Transplant and Heart Function Clinic, Queen Elizabeth II Health Sciences Centre, and Associate Professor of Medicine, Dalhousie University, Halifax, Nova Scotia; and Dr. Robert McKelvie, Medical Director, Heart Function Clinic and Medical Diagnostic Unit, Hamilton Health Sciences Corporation, and Professor of Medicine, McMaster University, Ontario.

Q : What do you conclude from the data with rosuvastatin in HF? How did the two trials differ? Should statins be discontinued in HF patients?

Dr. Howlett: It would appear that the CORONA trial speaks to the ischemic systolic HF population while the GISSI-HF trial is more relevant to the non-ischemic HF population in whom the vascular event rate is lower. It is also important to note that these were statin initiation trials, not statin discontinuation trials. Also, unfortunately, the GISSI-HF trial did not have a large component of patients with HF and preserved ejection fraction.   Q: What significance do these findings have for guidelines?

Dr. Howlett: As such, one new guideline might support the use of statins in patients with HF of ischemic etiology where a higher ischemic event rate—for example, stroke or MI—would occur, while not supporting statins where ischemic heart disease is not present.   Q: Statins are critical therapies in high-risk patients, but would you discontinue use of statins in all patients with HF based on the findings?

Dr. Howlett: There is likely not enough evidence from GISSI-HF to substantially alter suggestions for statin use in the 50% of HF patients with preserved systolic function, and as such, previously existing guidelines would, where applicable, provide better guidance.

Q: How do the GISSI-HF results change your management of patients with HF?

Dr. McKelvie: The GISSI-HF study was interesting and doesn’t in my opinion change the management of these patients in any major fashion. The study inclusions were quite specific because they enrolled HF patients who were not taking a statin and evaluated the effects of adding rosuvastatin. This study did not examine the effects of removing a HF patient from statin therapy. 

Q: What do you recommend in regard to statin therapy and HF patients?

Dr. McKelvie: If I was faced with a HF patient with either ischemic or non-ischemic cardiomyopathy who had not previously received statin therapy and had not recently had an acute coronary event—and the patient’s lipid levels were in an acceptable range—then I would not add statin therapy because there does not appear to be any additional benefit. I would not withdraw statin therapy from HF patients who are presently being treated because the drug would have been started for an appropriate reason; for example, following an acute coronary event or for the management of hyperlipidemia. 

Q: So you would not start a statin in a heart failure patient?

Dr. McKelvie: I would start a statin in a HF patient who has an acute coronary event, similar to what would be done in other patients with acute coronary events (e.g. acute MI). So in effect, I wouldn’t start a statin in a stable heart failure patient who, for whatever reason, has not been previously treated with a statin because there is no expectation that this therapy would provide added benefit.  However, I would not stop a statin in a HF patient who has been receiving it long-term and had it previously started for other appropriate reasons. I would start a statin in a HF patient who experiences an acute coronary syndrome and, of course, I would use it to appropriately treat hyperlipidemia in these patients.