A year of treatment with valsartan (Diovan, Novartis) in patients initially in
sinus rhythm after at least one prior episode of atrial fibrillation (AF) failed
to suppress the arrhythmia in what's said to be the first large, randomized,
placebo-controlled test of a secondary-prevention strategy that had once seemed
promising based on less definitive studies [1].

The trial's patients were also required to have a coexisting cardiovascular
disorder or diabetes and be on a stable conventional drug regimen for their
heart disease for at least a month before enrollment.

Many, therefore, were also on varying combinations of ACE inhibitors,
amiodarone, other antiarrhythmics, beta blockers, calcium-channel blockers,
statins, and other agents.

The results of the trial, called Gruppo Italiano per lo Studio della
Sopravvivenza nell'Infarto MiocardicoˇVAtrial Fibrillation (GISSI-AF), are
published in the April 16, 2009 issue of the New England Journal of Medicine.

GISSI-AF reliably suggests that "in patients with frequent episodes and a
long-standing history of atrial fibrillation, the addition of an
angiotensin-receptor blocker [ARB] to standard medical therapy is not useful for
preventing recurrence of the arrhythmia," the report's corresponding author, Dr
Aldo Pietro Maggioni (Associazione Nazionale Medici Cardiologi Ospedalieri
Research Center, Florence, Italy), said to heartwire.

Over one year, AF recurred in 51% of the 722 patients randomized to receive
valsartan at up to 320 mg/day and 52% of the 720 in the placebo group. More than
one AF episode developed in 27% and 28%, respectively. The differences in both
outcomes, co-primary end points of the trial, were solidly nonsignificant
whether adjusted or unadjusted.

Nor were significant valsartan effects seen in secondary end points that
included number of AF recurrences, ventricular rate at first AF recurrence, and
all-cause and CV hospitalization.

In prospectively defined subgroup analyses, Maggioni pointed out, valsartan was
ineffective against AF recurrence whether or not patients were also on ACE
inhibitors, beta blockers, or amiodarone.

The GISSI-AF results probably don't apply to patients with heart failure or LV
systolic dysfunction, who made up only about 8% of the trial's population, he
said.

Also, there's good evidence that ARBs might still be effective at preventing AF
in patients with risk factors for but no history of the arrhythmia. "It's
possible that this kind of drug may be useful in primary prevention," according
to Maggioni.

Most of the trial's patients were taking antiarrhythmics, and a full 88% had
undergone cardioversion for AF within two weeks of randomization, suggesting
that "the substrate for atrial fibrillation was well established in the study
population," observes Dr Anne M Gillis (University of Calgary and Calgary Health
Region, AB) in an accompanying editorial [2].

"Although some data suggest that ACE inhibition may promote the regression of
atrial fibrosis, one wonders whether ARB therapy could have modified the
substrate for atrial fibrillation over the course of the brief follow-up period
of only one year," Gillis writes. "This is a particularly important point, since
approximately half of the primary end points occurred within two months after
randomization. Thus, it may not be surprising that an ARB added no incremental
benefit for the prevention of atrial fibrillation."

Gillis agreed that GISSI-AF results "do not support the adjunctive use of
valsartan for the prevention of atrial fibrillation in patients with a history
of atrial fibrillation who also have hypertension, have required recent
cardioversion, and are receiving established therapies for the prevention of
atrial fibrillation."

There were 10 thromboembolic events in the valsartan group and only two among
controls (p=0.04), despite similar numbers of patients on oral anticoagulation.

"This is probably the play of chance," Maggioni said, given that there were few
such events and no prior studies had suggested there might be such a hazard from
ARBs. In fact, he observed, the number of thromboembolic events in the placebo
group was unusually low and in the valsartan group was consistent with other
studies.

According to the report, the GISSI studies are supported by Associazione
Nazionale Medici Cardiologi Ospedalieri, Florence, and by Istituto di Ricerche
Farmacologiche Mario Negri, Milan, Italy. Maggioni reports receiving research
support and lecture fees from Novartis. Other coauthors report receiving
"institutional research support" or lecture or consulting fees from Novartis.
Maggioni and coauthor Dr Roberto Latini (Istituto di Ricerche Farmacologiche
Mario Negri) "are listed as coinventors on a patent application for the
therapeutic use of valsartan in atrial fibrillation." Gillis reports receiving
consulting fees, lecture fees, and grant support from Medtronic and St Jude
Medical.


GISSI-AF Investigators. Valsartan for prevention of recurrent atrial
fibrillation. N Engl J Med 2009; 360:1606-1617.
Gillis AM. Angiotensin-receptor blockers for prevention of atrial
fibrillation--A matter of timing or target? N Engl J Med 2009; 360:1669-1671.