arznei-telegramm 2002; 33: 1
UNFAVORABLE DATA SUPPRESSED
The life extending effect of angiotensin-converting-enzyme (ACE) inhibitors and beta-blockers in patients with heart failure has been demonstrated in a number of prospective studies (a-t 1994; No. 3: 26-7 and 1999; No. 2: 21-2). Thus, drugs such as enalapril (e.g. PRES) lowered mortality of patients with heart failure of New York Heart Association (NYHA) class II and III from 39.7% to 35.2% within three and a half years (Number needed to treat [NNT]*/3.5 yrs = 22) (1). Mortality among patients of NYHA class IV decreased within six months from 54% to 39% (NNT/6 mo = 7) (2). With additional beta-blockers mortality of patients of NYHA class II and III decreased from 12.8% to 8.4% over a period of 14 months (NNT/1 yr = 27) (3). In patients of NYHA class III and IV, carvedilol (e.g. DILATREND) reduced mortality from 17% to 11% over 10 months (NNT/1 yr = 15; a-t 2001; 32: 62) (4). Today, ACE inhibitors and beta-blockers are considered the standard treatment for heart failure and are well tolerated by most patients.
The role of angiotensin (AT)-II-receptor blockers is, however, still undetermined; whether there is an advantage over, or parity with ACE inhibitors, has not yet been proven. Studies that have been published so far, reveal hints of considerable safety risks: In the ELITE**-II study, more patients died when taking losartan (LORZAAR) than captopril (LOPIRIN etc.). The mortality rate among patients who also took a beta-blocker increased signficantly (a-t 2000; 31: 58-9) (5). The RESOLVD** study had to be terminated prematurely due to an increased mortality and hospitalization rate in patients under candesartan (e.g. BLOPRESS) as compared to enalapril (a-t 1997; No. 12: 127, and 1999; No. 10: 110). Here, half of the participants had taken metoprolol (BELOC etc). An evaluation of the data with regard to the combination therapy with the beta-blocker has not been published.
The Valsartan Heart Failure Trial (Val-HeFT) with valsartan (DIOVAN, PROVAS) was published at the end of 2001 one year after it had first been presented at medical conventions. (6) More than 5000 patients with heart failure of NYHA class II, III and IV, and a left ventricular ejection fraction of less than 40%, took part in this study which was supported by a grant from Novartis Pharmaceuticals. Site monitoring, data collection and data analysis were also performed by Novartis. An independent end-points committee only adjudicated the reports of primary end-points.
In addition to their primary medication - ACE inhibitors (93%), beta-blockers (35%), diuretics and digoxin (LANCOR etc.) - the participants received either up to 160 mg valsartan, twice daily, in slowly increasing dosages, or placebo. Over a period of two and a half years, 19.7% of patients under valsartan had died versus 19.4% in the placebo group. Only the combined end-point of mortality and morbidity was significantly reduced from 32.1% to 28.8%. This result was mainly caused by a lower rate of hospital admissions for heart failure in the valsartan group. However, the total number of admissions did not differ significantly.
Mortality in patients taking both ACE inhibitors and beta-blockers was a significantly increased in the valsartan group. On the other hand, for patients receiving neither an ACE inhibitor nor a beta-blocker, valsartan was supposed to show better results than placebo. Frequency data of these posthoc subgroup analyses are missing.
For the only predefined subgroup - the group of patients taking a beta-blocker, with or without an ACE-inhibitor - no results were reported in the publication. This subgroup overlaps considerably, yet not completely with the group that took beta-blockers plus ACE-inhibitors. We suspect that these data were deliberately suppressed in order to protect the anticipated indication for valsartan in patients not toleranting to ACE-inhibitors, and to avoid a discussion regarding the safety of the combination therapy beta-blocker plus valsartan.
The US-American Food and Drug Administration (FDA) received a considerably larger data pool than the New England Journal of Medicine. According to the data prepared by the manufacturer (7) and the FDA (8), valsartan increased mortality in the pre-defined subgroup of all patients taking beta-blockers significantly from 12.5% to 16.4% (Number needed to harm [NNH]* = 26). Also, in the group of patients taking both an ACE inhibitor and a beta-blocker, mortality rose significantly under additional valsartan from 11.9% to 16.2% (NNH = 23). In the subgroup of all patients receiving an ACE inhibitor, mortality in the valsartan group was 19.9% as compared to 18.8% in the placebo group. Only in the group of patients who did not receive standard treatment, i.e. neither an ACE inhibitor nor a beta-blocker, valsartan lowered mortality significantly from 31.6% to 17% (NNT = 7).
Valsartan increased the number of adverse side effects that required treatment (8). Dizziness, hypotension, hyperkalemia, renal impairment, diarrhea and blurred vision occurred significantly more frequent in the valsartan group (7). The serum potassium concentration rose by at least 20% in 34% of patients in the valsartan group, versus in 21% of patients in the placebo group. A mean change of 50% or more of the serum creatinine and blood urea nitrogen levels occurred in 15% and 47%, respectively, under valsartan, as compared to 6% and 27%, respectively, in the placebo group. The FDA suspects that the advantage of a lower admission rate for heart failure, may have been "offset" by an increased admission rate for adverse drug reactions under valsartan. Alternatively, according to the FDA, there may be a bias in the process of adjudicating hospitalizations as being caused by heart failure(8).
 | The recently published Valsartan Heart Failure Trial (Val-HeFT) with valsartan (DIOVAN, PROVAS) for the treatment of heart failure has been managed and evaluated under considerable participation of the manufacturer. Important data that are unfavorable for valsartan, but essential for the assessment, have been submitted to the FDA, but have been omitted from the publication. |
| In patients with heart failure who in more than 90% of cases take an ACE inhibitor and in 35% of cases a beta-blocker, additionally prescribed valsartan has no effect on overall mortality. The hospitalization rate due to heart failure decreases, while the overall hospitalization rate remains unchanged - possibly, because more hospital admissions are caused by adverse drug related side effects. |
| In patients receiving a beta-blocker, independent of a concommittant therapy with an ACE inhibitor, and in patients who take both a beta-blocker and an ACE inhibitor, valsartan has a significant and clinically relevant adverse effect on mortality. Mortality is increased nonsignificantly in the group of all patients treated with ACE inhibitors. Thus, an AT-II-blocker is contraindicated in patients with heart failure receiving one or both of the standard treatment drugs. Particularly, combining valsartan with a beta-blocker in patients who do not tolerate an ACE inhibitor - a niche indication for valsartan that is probably intended by the manufacturer - appears questionable upon the data presented. |
| Whether treatment with valsartan in patients who take neither an ACE inhibitor nor a beta-blocker is in fact better than placebo, will have to be confirmed by further studies. If 5% of heart failure patients do not tolerate an ACE inhibitor, and of those, another 5% are intolerant of a beta blocker, there would be only 25 out of 10,000 patients for whom valsartan might be indicated. |