Many clinical trials unintentionally include patients with a low risk of th
e trial endpoints. PRIME II (The Second Perspective Randomised study of Ibo
pamine on Mortality and Efficacy) was a large international randomised doub
le blind trial comparing the addition of ibopamine or placebo to the therap
y of patients with advanced heart failure. The trial was stopped prematurel
y because ibopamine was associated with an increased fatality rate, but the
protocol achieved its objective of including high-risk patients. Here we d
escribe the protocol details that enabled patients with the desired degree
of risk to be included. We also amplify our definition of mode of death. Th
e PRIME II protocol was designed with the intention that patients in the pl
acebo group would have an annual fatality rate of 20%. Since the study was
to be conducted in some 200 centres in 13 European countries, the inclusion
criteria had to be simple and flexible, allowing for different clinical pr
actice. The inclusion criteria, together with the use of simple investigati
ons (which did not have to include angiographic or radionuclide ventriculog
raphy) are described. The annual fatality rate in the placebo group was jus
t over 20%. Six categories of mode of death were used, but while they were
reasonably easy to apply they did not reveal the reason for the unexpected
adverse effect of ibopamine. The inclusion and exclusion criteria used for
PRIME II, and the definitions of mode of death, were effective. The PRIME I
I protocol can be used as a model for future heart failure studies. (C) 199
9 European Society of Cardiology. All rights reserved.