were roughly equipotent with respect to systemic
arterial pressure, and there were only minor differences
in the use of open-label norepinephrine,
most of which were related to early termination
of the study drug and a shift to open-label norepinephrine
because of the occurrence of arrhythmias
that were difficult to control. Doses of openlabel
norepinephrine and the use of open-label
epinephrine and vasopressin were similar between
the two groups. Second, we used a sequential design,
which potentially allowed us to stop the
study early if an effect larger than that expected
from observational trials occurred; however, the
trial was eventually stopped after inclusion of more
patients than we had expected to be included on
the basis of our estimates of the sample size.
Accordingly, all conclusions related to the primary
outcome reached the predefined power.
In summary, although the rate of death did
not differ significantly between the group of patients
treated with dopamine and the group treated
with norepinephrine, this study raises serious concerns
about the safety of dopamine therapy, since
dopamine, as compared with norepinephrine, was
associated with more arrhythmias and with an
increased rate of death in the subgroup of patients
with cardiogenic shock.