376
PREHOSPITAL EMERGENCY CARE OCTOBER / DECEMBER 2002 VOLUME 6 / NUMBER 4
second case there was confusion as to the presence of
a pulse.
the physical design of the nursing home may some-
how contribute to delay. Regarding this last possibili-
ty, motion artifact that occurred during repositioning
of the patient (from bed to a firmer, potentially more
accessible area) contributed to the delay in two of the
three delayed nursing home cardiac arrests. Finally,
the qualitative assessment of delay reinforces the
importance of standing clear of the patient during
analysis and maintaining AED upkeep, and suggests
that the EMT be prepared to overcome uncommon
obstacles to pad application that might not be part of
the typical training preparation.
DISCUSSION
In an EMS system with a mature experience in AED
operation, we found that the average interval from
EMT on-scene recognition of a cardiac arrest victim to
first shock was slightly less than 1 minute and that the
vast majority (94%) of patients were shocked within 90
seconds. Delayed shock was associated with unwit-
nessed arrests and nursing home location. In addition,
qualitative assessment of delay identified a mix of
arrest, patient, and operator factors that may be
important.
LIMITATIONS
The interval from collapse to shock is the critical
determinant in resuscitation of VF cardiac arrest.
Although the time required for EMS to arrive at the
scene is often a major portion of this interval, once on
the scene timely therapeutic action is clearly indicat-
This study has several limitations. In the King County
EMS system, AED application and defibrillation is the
highest priority in the treatment of cardiac arrest.
Other settings may have other protocols (e.g., 90 sec-
onds of CPR prior to AED application) that may limit
5
12
ed. In this cohort, patients typically received a shock
the relevance of the findings of this study.
within 1 minute from EMT on-scene recognition of a
cardiac arrest victim, and nearly all within 90 seconds.
The durations in this investigation are similar to the
simulated arrest experience of EMS professionals,
modestly less than AED-naïve young persons in sim-
ulated arrest, and considerably less than the intervals
in simulated arrests among older adults trained in
AED operation. Of note, in many training scenarios,
the time-to-shock interval includes the initial patient
assessment (responsiveness, airway, respirations, and
pulse check). In this study, the AED was not turned on
until the event was determined to be a cardiac arrest.
Thus, the time-to-shock intervals in this study would
have been slightly longer if an initial patient assess-
ment interval had been included. Nonetheless, one
interpretation of the results is that a reasonable goal
for training and the field is 60 seconds with a mini-
mum standard of 90 seconds. Although the EMS care
of every cardiac arrest patient should be reviewed in
an effort to improve patient outcomes, those in whom
time to shock requires >90 seconds may deserve spe-
cial attention.
Approximately one-third of eligible cardiac arrests
had an incomplete AED record. These excluded cases
may somehow have been different with respect to
EMT AED application and shock, though they pos-
sessed similar event, EMS, patient demographic, and
survival characteristics when compared with the
study cohort. During the study period, EMTs did not
standardly submit an electronic AED recording if
paramedics and EMTs arrived on scene simultaneous-
ly. Thus, we were unable to assess the potential influ-
ence of paramedics on time to shock by EMT AED.
This was a retrospective analysis that used electronic
and written records to assess time to shock and inves-
tigate potential causes of delay. An investigator was
not present on scene to observe the care of each arrest.
Consequently, though we feel the intervals truly
reflect the timing of care, the causes of delay required
interpretation and should be viewed cautiously.
Moreover, only a relatively small number of cases had
a delayed time to shock. Thus, we had limited power
to detect associations, and some characteristics impor-
tant in EMT time to shock may not have been identi-
fied.
8
,9
Our findings provide insights that may be useful in
limiting delays. It is plausible that additional time
elapsed in the unwitnessed arrest while the EMT
attempted to establish the circumstances of the event.
Event circumstances, however, are generally unlikely
to alter the initial EMS care of airway, breathing, cir-
culation, and defibrillation. The association between
delay and nursing home location may be explained by
several possibilities: 1) additional outside medical per-
sonnel at the nursing home may hinder or distract the
EMT care process; 2) given end-of-life preferences and
the need for documentation among nursing home res-
idents, delay may occur while the EMS personnel con-
firm the patient’s resuscitation preferences1 ; or 3)
CONCLUSIONS
The findings of this study suggest that a 1-minute goal
and a 90-second minimum standard for time to first
shock are appropriate for EMT AED defibrillation in
the field. The EMT should be cognizant of delays in
attempted defibrillation and whether these delays are
necessary and/or modifiable. Such efforts may aid in
improving the outcome of cardiac arrest.
The authors thank the EMTs, paramedics, and emergency dispatch-
ers of King County, Washington, for their ongoing excellence in the
care of cardiac arrest, as well as Dr. Richard Cummins, for his fore-
0,11