As part of the development of a model for the study of adverse events,
we have investigated the risk of bradycardia with propofol. A systema
tic search for any type of report, published and unpublished, was made
to review the evidence that propofol increases the risk of bradycardi
a, asystole and death from bradycardic events. Quantitative and qualit
ative analyses of data with different strengths of evidence were perfo
rmed. Sixty-five published and 187 spontaneous reports to drug monitor
ing centres described with different strength of evidence a biological
basis for propofol-induced bradycardia, 1444 bradycardias, 86 asystol
es and 24 deaths. In controlled clinical trials, propofol significantl
y increased the risk of bradycardia compared with other anaesthetics (
number-needed-to-harm 11.3 (95% confidence interval 7.7-21)). In paedi
atric strabismus surgery the number-needed-to-harm was 4.1 (3-6.7). On
e of 660 patients undergoing propofol anaesthesia had an asystole. The
risk of bradycardia-related death during propofol anaesthesia was est
imated to be 1.4 in 100 000. Data from the phase IV study of propofol
did not agree with data from controlled studies. Propofol carries a fi
nite risk for bradycardia with potential for major harm. Study of adve
rse events should be made with systematically searched data and, in co
ntrast with study of efficacy, not restricted to randomized, controlle
d trials.