ZOLL E Series Monitor Defibrillator Rev R - Manual (Page 122)

E Series Operator’s Guide
A-24 9650-1210-01 Rev. R
Shock Conversion Estimator
Use of a defibrillator shock is currently the best option for terminating ventricular fibrillation and restoring a life
sustaining ECG rhythm [1]. Maintaining blood flow through the heart via cardiopulmonary resuscitation (CPR) has
been shown to improve the chances of a successful defibrillation [1]. The cessation of blood flow through the heart that
occurs when CPR is stopped decreases the likelihood of a successful shock in proportion to the amount of time that
has elapsed without CPR [1]. The repeated use of defibrillator shocks that do not restore a life sustaining rhythm may
cause additional damage to the myocardium and reduce the patient's chances for survival. The use of an accurate
shock outcome predictor can help reduce the duration of CPR interruptions and the number of ineffective (non-
converting) shocks delivered.
Properly performed CPR has been shown to increase blood flow to the heart and increase the neurologically intact
patient survival rate [2]. Following current rescue protocols that alternate periods of CPR and defibrillator shocks, the
rescuer must stop CPR while the defibrillator analyzes the patient's ECG rhythm to determine whether it is shockable.
If the rhythm is non-shockable, the rescuer immediately resumes CPR. When the patient's rhythm is shockable,
however, the rescuer must withhold CPR for an additional period of time while the defibrillator changes, shocks are
delivered, and the outcome evaluated. Should the defibrillator shocks be ineffective, CPR is immediately resumed after
the loss of precious seconds without cardiac blood flow augmentation. If the non-perfusing rhythm can be identified as
unlikely to convert before the delivery of ineffective shocks, non-CPR time can be reduced and there will likely be an
increase in post resuscitation as well as neurologically intact survival. The ability to predict that the current ECG rhythm
will not convert may also help reduce the number of non-converting shocks delivered. This reduction in total shocks
delivered would reduce the damage sustained by the heart during resuscitation. The Shock Conversion Estimator
(SCE) addresses these problems by computing a Shock Prediction Index (SPI) number which measures the
probability that a shockable rhythm will be successfully converted by immediate defibrillation.The SPI number is
directly related to the AMSA measure developed by the Weil Institute of Critical Care Medicine [3].
The Shock Conversion Estimator algorithm was developed and tested using data collected from a registry of ZOLL
AED Pro® and AED Plus® defibrillator field cases. Since the AED Pro and AED Plus defibrillators are first responder
units, all patient records correspond to first responder cardiac arrest situations. The defibrillator shock results from
these cases were annotated as "converted" if a transient return of spontaneous circulation (tROSC) occurred following
the shock. tROSC was defined as post shock ECG rhythms meeting both of the following characteristics:
1. Spontaneous ECG rhythms lasting at least 30 seconds that began within 60 seconds after shock
delivery; and
2. Rhythms exhibiting a heart rate of 40 beats per minute or more.
The post shock rhythm was annotated as "non-converted" if it exhibited any other conversion outcome, e.g. VF, VT,
and asystole.
The total database consisted of 258 patient records containing 586 shocks. The first 109 patient records were used in
the Validation Database which consisted of 251 delivered shocks. The Development Database was constructed from
the remaining patient records, 149 patients, resulting in 535 delivered shocks. The Development Database was used
to develop the algorithm and to establish the threshold SPI values for 95% sensitivity. The Test Database was used to
prospectively validate the performance of the algorithm against the default and other user configurable SPI thresholds.
Figure A-21 on page A-25 presents the sensitivity and specificity curves for the combined datasets. The vertical line
indicates the position of the 7.4 mV-Hz default threshold. 7.4 correlates to a sensitivity and specificity of 95% and 57%,
respectively. Table A-2, “Accuracy Table of SCE Levels and Corresponding SPI Thresholds,” on page A-26 lists the
SCE Level settings (HIGH, MEDIUM, and LOW) and corresponding SPI thresholds, sensitivities, and specificities that
can be configured on the E Series unit. Column 1 is the SPI threshold in mv-Hz. Columns 2 and 3 are the sensitivity
and specificity as described below (expressed in percent).
The preferred treatment for non-converting rhythms may be the delivery of aggressive CPR. The use of the SPI
measure to determine when shock treatments are likely to succeed will help minimize time between the advisory
decision and the start of CPR. Minimizing non-perfusing time during resuscitation is a key contributor to improving
patient outcomes [4].