Since all death is (eventually) sudden and associated with cardiac arr
hythmias, the concept of sudden death is only meaningful if it is unex
pected, while arrhythmic death is only meaningful if life could have c
ontinued had the arrhythmia been prevented or treated. Current classif
ications of death as being arrhythmic or sudden are all biased by the
difficulty of having to decide on the degree of unexpectedness or the
likelihood that life could continue without the arrhythmia. The uncert
ainties are enlarged by the fact that critical data (such as knowledge
of arrhythmias at the time of death or autopsy) are available in only
a few percent of cases. A main problem in using classifications is th
e lack of validation data. This situation has, with the MADIT trial, c
hanged in the case of the Thaler and Hinkle classification of arrhythm
ic death. The MADIT trial demonstrated that arrhythmic death was nearl
y abolished by the implantable defibrillator, indicating that arrhythm
ic death by this classification is meaningful, at least in the populat
ion studied. For future investigations, a call is made for committees
to present data in a way that allows the reader to examine the quality
of the data used for evaluation.