Background
Both heparin and bivalirudin are antithrombotic agents.
They are administered to patients in the cardiac catheter
lab during their emergency treatment for STEMI. Both are
in widespread and routine clinical use around the world.
There is debate as to whether one has advantage over the
other. HEAT-PPCI was a randomised controlled trial
(RCT) designed to inform this debate.
A well-designed and well-conducted RCT provides
robust evidence about the effectiveness of treatment. In
RCTs, the study population is divided into two (or
more) groups. The groups receive comparable, routine
care—except for the treatment under investigation. A
key feature is that the treatment allocation is random.
Depending on the design of the study, the patients (and
indeed the treating physician), may or may not have the
treatment allocation revealed to them at the time. In an
open-label study, both the clinical team and research
participants are aware of the treatment allocation. By
comparison, trials may be single-blinded, where the
treatment allocation is concealed from participants; or
double-blinded, where researchers are also unaware
(Akobeng, 2005). In routine research practice, patients
agreeing to participate in RCTs will have accepted that
they will be randomised to one of the groups. They will
have taken the time to consider the trial, studied the
documentation and discussed any concerns with
researchers. Informed consent will be taken only after
this process has occurred. This is not possible in
emergency situations, such as in the STEMI population
under investigation in HEAT-PPCI.
Although treatments in RCTs are randomly allocated,
this does nothing to ensure that the results observed in
such trials are generalisable to the wider population. For
a variety of reasons, almost all medical trials tend to
recruit just a small proportion of all the patients who
actually have the medical condition under consideration.
Women, older people and those from lower
socioeconomic groups are some of those often underrepresented
in clinical trials (Britton et al, 1999). This
means that the results of some trials may not help guide
treatment in the ‘real-world’ population