EASI ECG Monitoring - DOC

Document Sample
EASI ECG Monitoring - DOC Powered By Docstoc
					Centre for Clinical Nursing Research
Deakin University – Epworth Hospital

    EASI ECG Monitoring vs
    Traditional 12-Lead ECG.
    A Review of the Literature

Prepared by Bernice Redley
May 2005

                                 Centre for Clinical Nursing Research
                                 Deakin University – Epworth Hospital

Table of Contents

What is EASI ECG Monitoring? ................................................................................... 3
   Advantages of the EASI ECG monitoring system ..................................................... 4
   Disadvantages of the EASI ECG monitoring System................................................ 4
   Comparisons between the EASI and Mason-Likar 12 lead ECG systems ................ 5
Summary of the Research .............................................................................................. 6
      Methods.................................................................................................................. 6
      Detection of acute ischaemia and Myocardial Infarction (MI) .............................. 6
      Lead amplitudes ..................................................................................................... 7
      Conduction and rhythm disturbances..................................................................... 7
      Misdiagnosis .......................................................................................................... 7
   Limitations ................................................................................................................. 8
   Summary .................................................................................................................... 8
   APPENDIX 1. Summary Table by Author ................................................................ 9
   APPENDIX 2. Summary Table by Parameter ......................................................... 13
   References ................................................................................................................ 19

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                        2
Author: Bernice Redley
                             Centre for Clinical Nursing Research
                             Deakin University – Epworth Hospital

What is EASI ECG Monitoring?
EASI is a method of continuous electrocardiogram (ECG) monitoring that is an
alternative to both the commonly used 5-electrode lead bedside monitoring system
and the traditional 10-electrode Mason-Likar 12–lead ECG system. The EASI lead
configuration enables continuous 12 lead ECG ambulatory monitoring using only 5
electrodes. The EASI 12 lead ECG is derived from this reduced lead set using a
method described by Dower et al. [1-3].
The EASI 12-lead ECG is derived from a set of 5 leads; 4 recording electrodes and
one grounding electrode. The placement of these leads is as follows [4].
E: Lower extreme of the sternum
A: Left mid-axillary line, same transverse line as E
S: Sternal manubrium
I: Right mid-axillary line, same transverse line as E
G: Fifth electrode is the ground and can be placed anywhere on the torso
                              Diagram of EASI electrode placement

                                           [4 p. 180]

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                            3
Author: Bernice Redley
Advantages of the EASI ECG monitoring system
 Easy to apply. The EASI is a 5 electrode system that provides a 12-lead ECG,
  making it easier to rapidly and accurately apply the leads in stressful situations [4].
 Time saving for staff. As a continuous 12-lead ECG is produced, there is no need
  to perform traditional intermittent 12-lead ECGs.
 Continuous 12-lead ECG. The reduced set of leads provides continuous 12-lead
  ECG monitoring as opposed to a snapshot 10-15 second view ECG with the
  traditional ECG [5].
 Better patient comfort. The reduced set of EASI leads is less bulky than a 10-
  electrode Mason-Likar system, improving patient comfort, simplifying application
  and maintenance for continuous monitoring [5].
 Less waveform interference. With less leads, the EASI system is less susceptible
  patient movement, signal interference or myeloelectric noise making it more
  suitable and sensitive (than the 10 electrode system) for continuous ambulatory
  ECG monitoring [6-8].
 Better serial comparisons. EASI derived ECGs have better reproducibility for serial
  comparisons [6, 8, 9].
 More leads viewed continuously. EASI provides a continuous 12-lead ECG
  allowing detection of transient myocardial ischaemia across multiple leads, while
  the more common 5-lead ambulatory monitoring system does not allow monitoring
  of all the precordial leads at the same time (usually lead II and V1 only)[5, 10-12].

Disadvantages of the EASI ECG monitoring System
 Unfamiliar to staff. The EASI system is unfamiliar to clinicians so training in this
  system may be required [13]
 Differs to traditional method. Some leads from the EASI derived ECG have been
  found to differ slightly to the standard ECG, requiring consideration during
  interpretation. These differences are outlined below.
 Placement error. Lead placement error may lead to discrepancies [4].
 Less sensitive in detecting some conditions. Due to lower amplitudes in some leads,
  it is difficult to detect chamber enlargement (atrial and ventricular hypertrophy )
  with EASI ECG [7, 12], though similar problems also exist with the standard ECG.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                 4
Author: Bernice Redley
Comparisons between the EASI and Mason-Likar 12 lead ECG
 Duration and amplitude of P, QRS, ST and T waves may differ between the EASI
  derived and standard ECG systems [5, 7, 9, 14], with the EASI system
  demonstrating less amplitude. It is important that this is considered during analysis
  and serial comparisons are made using the same system.
 Frontal plane transition zone at V2-V3 in EASI and V3-V4 in traditional [7].
 The correlation between EASI ECG and the standard ECG complex interpretations
  reached less than 90% agreement in some leads (particularly III, aVL, aVF, V4 and
  V5) [9].
 High levels of correlation have been found between the EASI and the standard
  ECG complexes in leads I, aVR, V1, V2, V3, and V6 [9].
 Greater than 90% agreement for detection of small (100 V) ST segment changes
  indicating acute ischaemia [12, 13]
 Greater than 90% agreement for detection of prior myocardial infarction (Q wave,
  isolated ST-T) [9, 12, 13].
 Similar rates of false diagnosis or misinterpretation of ECGs by clinicians with both
  systems [12, 13]
 100% agreement for rhythm identification [7, 12].
 95-100% agreement for conduction problems (Bundle branch blocks, fascicular
  blocks) [7, 12].
 Differences in axis of up to 30 degrees between the two systems [7, 12].

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                              5
Author: Bernice Redley
Summary of the Research
Several studies have compared derived EASI 12-lead ECGs with the standard Mason-
Likar 12 lead ECG. Overall the two systems are similar for many of the parameters
examined, but some differences also exist. A summary of these studies is provided

The strongest study design simultaneously collected the EASI derived and standard 12
lead ECGs continuously across a large population of subjects (540 patients) from
acute clinical areas (Emergency and cardiac catheter), then used a computer analysis
program to compare the recordings [12]. The authors then used a clinical expert to
examine the accuracy of the ECG interpretations in collaboration with clinical data
such as cardiac enzymes, echocardiography, patient history and clinical examination.
Use of computer analysis and simultaneous data collection were strengths as slightly
greater variations within and between EASI and standard ECG recordings have been
associated with human interpretations [7]; or ECGs that were not collected
simultaneously [4, 14]. Variations have also been noted between studies that used data
from healthy subjects in laboratory settings [4, 6] and those conducted in real clinical
circumstances with patients with variable diagnoses requiring cardiac monitoring [7,
9, 12-14]. Studies using patient population provide more useful clinically relevant
information. Few studies incorporated clinical findings to confirm the diagnosis
derived from the ECG [12, 14].

Detection of acute ischaemia and Myocardial Infarction (MI)
The rate of agreement between EASI ECG and the standard 12-lead ECG for the
detection of acute ischaemia was high in the majority of reviewed studies [9, 11-16].
Some ECG differences were detected between different types of ischemia. Both
inflation induced ischaemia during cardiac catheterisation and spontaneous ischaemic
events have been examined.
Rautaharju et al. [14] reported 85% agreement between EASI and standard ECGs in
the detection of inflation induced ischaemia during cardiac catheterisation. These
authors used 100 V (which is less than usually considered clinically significant) as
the ST segment threshold. The use of independent electro-cardiographers reading the
ECGs may have contributed to the lower agreement for ischaemic changes in this
study. Studies that used computer program analysis reported much higher agreement
(>99%) between EASI derived and standard ECGs for the detection of ischaemia [12,
13] than studies that used clinical experts with or without computer assistance [7, 14].
Drew et al. [12] reported the EASI system accurately detected 93% of acute transient
myocardial ischaemic events, while the leads routinely used for bedside monitoring
(Leads II and V1) detected only 42%. In this study, discrepancies between the EASI
and standards ECG were found in 17 of 238 ST events. Review by an expert clinician
revealed the EASI system was less sensitive in detecting low voltage ST segment
changes. None of these 17 discrepancy events were associated with chest pain, none
involved large changes (>200 V) and none involved the need for clinical
intervention. These findings suggest that these differences were not clinically

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                               6
Author: Bernice Redley
Similar findings were experienced in the detection of ECG changes suggesting old
MI. Comparisons of EASI derived and standard ECGs revealed 95% agreement for
anterior and 92% agreement for inferior Q-wave MI in a large study sample [12]. The
EASI system had greater sensitivity than the standard system in detecting these
changes (59% vs 55%) but slightly less specificity (95% vs 99%).

Lead amplitudes
Differences in the amplitudes of several leads have been consistent findings in studies
comparing the EASI derived ECG with the standard 12-lead. Most commonly
discrepancies were detected in V3-V4 [4, 7, 9, 13] or isolated ST-T segment changes
of low amplitude [12, 14]. Only minor differences of 14-30 degrees in frontal plane
axis were detected [7, 12]. These small differences are not usually clinically
Low amplitudes were also associated with low sensitivity in detecting chamber
enlargements using the both methods of ECG acquisition. These difficulties were
similar with both the EASI derived and standard ECG, except in the cases of left
ventricular hypertrophy where the EASI derived ECG had higher specificity than the
standard ECG [12].
EASI derived ECG have shown lower amplitude Q waves in V6, smaller amplitude R
wave in V1, V2 and V4, and less ST amplitude in V4 and V5 [14]. Horacek [9]
compared ECGs collected from clinical settings revealing less than a 90% correlation
between EASI derived and standard ECG in leads III, aVL for normal and post MI
ECGs, with Q waves, non-Q waves and history of VT.

Conduction and rhythm disturbances
Comparisons between the EASI derived and standard ECGs consistently demonstrate
high levels of agreement (95%-100%) in the detection of conduction problems such as
pre-excitation patterns, bundle branch blocks, fascicular blocks [7, 12, 16] and rhythm
disturbances such as VT and SVT [9, 12, 16]. Minor deviations of 0-2ms in the
cardiac intervals for the PR, QRS, QT and QTC have been detected. While the
observed difference of 1ms in QRS was statistically significant it was not clinically
significant [12]. Again the clinical significance of these findings must be questioned.

Similar rates of misdiagnosis have been reported for the two types of ECGs with the
majority related to old MI changes. Drew et al. [7] reported two false positives and
four false negative in EASI derived ECGs from patients with prior myocardial
ischaemia, while six false negatives were reported from the standard ECGs from
patients with a prior MI. Similarly Rautaharju [14] reported 10% of ECGs had
significant differences between the EASI derived and standard method with 3.4%
involving Q-wave MI or ST changes in lead II. However, in this study none of the
new changes were identified as old, and none of the Q-wave MI’s were missed using
the EASI derived ECG for interpretation. Drew [13] found only one false negative
from 207 derived ECGs and in her 1999 study, two ECGs were misdiagnosed using
both the EASI derived and standard ECG methods.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                               7
Author: Bernice Redley
There are several variables to consider that pose limitations to these study findings.
The clinical significance of these limitations must be considered when deciding about
the appropriateness of the EASI methods for continuous ECG monitoring.
Various computer programs (Phillips, Marquette, Montana) and equipment were used
to collect ECG data across the studies. The types of electrode, the method used for
placement of ECG electrodes, the expertise of the person applying the electrodes and
the timing of data collected are all independent variables not addressed in many
studies. It has been established that slight variations in the placement of the
electrodes may increase error when making ECG comparisons [8]. The influence of
these factors on the quality of the recordings and subsequent comparisons is unknown.
The clinical conditions under which data were collected also varied. Laboratory
conditions and healthy subjects differed to clinical conditions and ill patients. Finally,
few studies confirmed the ECG diagnosis using other clinical diagnostic criteria,
suggesting that the error in patient diagnosis may have been higher with both methods
of ECG data acquisition.

EASI derived ECGs are comparable to the standard ECG for the diagnosis of wide
complex tachycardias [9, 12, 16] and myocardial ischaemia [9, 11-16]. The
reproducibility of the EASI derived ECG makes it more useful in serial comparisons
over time. The EASI derived system for continuous ambulatory monitoring has
advantages over the standard 5-lead monitoring systems and continuous 12 lead
monitoring systems with the benefits of both. The EASI system allows continuous 12
lead monitoring instead of the limited II and V1 capability of the traditional 5-
electrode method. The EASI system is less susceptible to noise and interference than
the bulky 10-electrode 12-lead system. These benefits enable significant
improvements in the detection of spontaneous ischemia during continuous ambulatory
While there are differences between ECG data collected using the EASI derived and
standard 12-lead ECG methods, the clinical significance of these changes is
questionable. The lower amplitude of some leads in the EASI derived ECG may be
overcome by adjusting the thresholds used to determine clinical significance and
training staff in interpretation. Similar limitations in terms of clinical diagnosis and
variation in interpretation by clinicians affect both methods of 12-lead ECG

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                   8
Author: Bernice Redley
APPENDIX 1. Summary Table by Author
AUTHOR                   TITLE                STUDY DESIGN              Number of                 FINDINGS
Drew [12]                Accuracy of the      Two methods of            All ED patients over 2    EASI and standard ECG compared for cardiac
                         EASI 12 lead ECG     continuous ECG            year period. 540          rhythm, cardiac intervals, QRS axis, chamber
                         compared to the      monitoring                patients enrolled, 426    enlargement-hypertrophy, BBB and fascicular
                         standard 12 lead     simultaneously from       with acute coronary       blocks and prior MI. 100% agreement on
                         ECG for              ED admission or from      syndromes. Some not       rhythm identification 84-89% agreement on
                         diagnosing cardiac   catheter based            diagnosed with acute      chamber enlargement, 997% agreement on R &
                         abnormalities        intervention. Computer    ischaemia and ruled       L BBB, 97-98% agreement on fascicular
                                              analysis program used     out.                      blocks, and 92-95% agreement on prior MI.
                                              to compare ECGs.
                                              examined and diagnosis
                                              made by specialist and
                                              clinical and echo
Drew [13]                Comparison of        Compared standard and     207 patients with timed   Agreement regarding ischaemia in 150 of 151
                         standard and         derived continuous        measurements (derived     patients with both methods.
                         derived 12 lead      ECG                       analysed every 20 secs,   Different time ratio for measurement may have
                         ECG for diagnosis                              standard continuously),
                                              Threshold of >1 lead of                             led to error in one care but may also have led to
                         of coronary                                    151 during procedure
                                              >100 V, 80 ms post j                               agreement in others.
                                              point used to define
                         induced myocardial

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                              9
Author: Bernice Redley
AUTHOR                   TITLE                  STUDY DESIGN             Number of                FINDINGS
Drew [16]                Comparison of a        Compared EASI with       649 patients (CP in ED   Identical for BBB, LAH, RVH, prior MI. 99.2%
                         new reduced lead       Interpolated standard    = 509 and Tachycardia    agreement of ischaemia in the ED. Appears to
                         set ECG with           limb leads and V1 and    in EP Lab = 140).        have higher agreement than EASI model.
                         standard ECG for       V5                                                Applied to patients by expert.
                         diagnosing cardiac
                         arrhythmias and
Drew [7]                 Comparison of a        Compare during EP        64 episodes of wide      9 pts had LVH, only 6 with evidence on derived
                         vectorcardiography     studies. Simultaneous    QRS complex              ECG. 6 pts with WPW identical delta wave
                         derived 12-lead        recordings and two       tachycardia in 49        morphology, 45 pts with prior MI 92%
                         ECG with the           expert investigators     patients                 agreement Lower voltages present in V3 and
                         conventional ECG       independently                                     V4 of derived ECG. QRS morphology
                         during wide QRS        compared using                                    dissimilar in 35% of pts in V3 and 51% in V4.
                         complex                standard criteria                                 (early transition of the RV rS to the LV qR in
                         tachycardia, and its   including QRS pattern,                            derived). Transitional zone generally appeared
                         potential              sequence, width,                                  in V3 or V4 or the ECG appeared in V2-V3 of
                         application for        morphology and                                    the derived ECG. QRS voltage less in the
                         continuous bedside     voltages.                                         precordial leads of the derived ECG affecting
                         monitoring.                                                              Dx of LVH. Derived ECG had false positive in
                                                                                                  2 cases prior MI in young person with no
                                                                                                  history, 2 false negative of inferior MI.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                       10
Author: Bernice Redley
AUTHOR                   TITLE                 STUDY DESIGN             Number of                  FINDINGS
Horacek [9]              Diagnostic                                     290 normal subjects        V6 best derived lead with highest agreement
                         accuracy of derived                            and 497 with prior MI      closely followed by V1 and V2. V3-V5 not
                         compared to                                    (36 with non Q, 282 Q-     reproduced well in the MI subgroups as they are
                         standard 12-lead                               wave, 179 with history     in normal subjects. Worse as the structural
                         ECG                                            of VT)                     damage to the myocardium increases the
                                                                                                   diagnostic performance markedly increases.
Rautaharju [14]          Comparability of      Read by two              40 patients recorded       No significant differences between ECG
                         12-lead ECGs          independent expert       prior to and at peak       readers in acute ischaemia. Suggest need for
                         derived from EASI     ECG readers and          inflation during PCTA,     modified prior MI criteria and ST thresholds for
                         leads with standard   analysis program.                                   AMI specific for EASI. Significant differences
                                                                        382 with old MI ECGs
                         12-lead ECG in                                                            within readers and the program
                                               ECGs printed side by
                         classification of                              472 with non MI.
                                               side in random order.                               10% with clinically significant differences,
                         AMI and old MI        Readers blinded.                                    3.4% classed as q-MI and 11 as isolated ST-T
                                               Coding forms used to                                evolution
                                               classify the ECGs
Sejersten [4]            The relative          Compared waveforms       20 paramedics              EASI and paramedic waveforms were equally
                         accuracies of ECG     from ‘Gold Standard’     collected data on each     accurate in 47%, paramedic more accurate in
                         precordial lead       ECG with paramedic       other. Paired to act as    31% and EASI in 22% when compared to Gold
                         waveforms derived     applied and EASI         experimental technician    Standard. Significant difference in paramedic
                         from EASI leads       derived ECGs. First      and study participant. 3   placement of leads mean 30mm misplacement.
                         and those acquired    acquired from            ECGs on each. 720          Review of discrepancies revealed that EASI
                         from paramedic        paramedic the two        comparisons of             more accurate for ST changes in V4 & V5. and
                         applied standard      simultaneously by        precordial leads.          T in V6. Paramedic more accurate Q amplitude
                         leads                 technician. Difference                              in V6, R amplitude in V1, V2, V4 and ST
                                               threshold of 10 V.                                 deviation in V1

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                           11
Author: Bernice Redley
AUTHOR                   TITLE               STUDY DESIGN              Number of               FINDINGS
Welinder [6]             Comparison of       Compared baseline         20 healthy volunteers   EASI ‘winner’ significantly more often in
                         signal quality      wander and myoelectric    with simultaneous       treadmill and supine to L in precordial.
                         between EASI and    noise amplitudes of       measurements            Traditional ‘winner’ significantly more often in
                         Mason-Likar 12-     EASI and traditional.                             Supine to R limb and precordial leads
                         Lead ECG during     Simultaneous
                         physical activity   recordings. Inter-rater
                                             not addressed. Range
                                             thresholds (?valid)
                                             established for noise.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                          12
Author: Bernice Redley
APPENDIX 2. Summary Table by Parameter
Parameter            Drew (1992)           Rautaharju,       Welinder,   Drew (1997)       Sejersten,   Drew (1999)                    Horacek
                                           (2002)            (2004)                        (2003)                                      (2000)
Acute                                      Inflation                     99% agreement                  138 patients with 238 ST
Ischaemia                                  induced                       in Cx artery                   events. 26 had acute ST
                                           ischaemia using               peak ST in V3                  elevation patterns. 63
                                           100 V as                     for derived and                angioplasty induced
                                           threshold 85%                 V3 or V4 for                   ischaemia and 150
                                           agreement,                    standard                       spontaneous ischaemia. 89%
                                           27/40 positive                                               agreement, the (14 standard
                                           and 7/40                                                     and 3 EASI) events with
                                           negative                                                     discrepancy were not
                                                                                                        associated with chest pain
                                                                                                        and none involved large
                                           differences in
                                           New Q MI                                                     changes (>200V). EASI
                                                                                                        detected 93% of ischaemic
                                           6.1% difference                                              events while the leads
                                           in ST-T                                                      routinely used for bedside
                                           segments                                                     monitoring (II and V1)
                                                                                                        detected 42%.
                                                                                                        EASI low amplitude ST
                                                                                                        changes ? clinical
                                                                                                        significance, none resulted
                                                                                                        in clinical interventions or
                                                                                                        poor outcomes..

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                         13
Author: Bernice Redley
Parameter            Drew (1992)           Rautaharju,        Welinder,   Drew (1997)   Sejersten, (2003)      Drew (1999)                Horacek
                                           (2002)             (2004)                                                                      (2000)
Prior MI (Q-         92%                   85% agreement                                                       Agreement 95% for          Correlations
wave)                agreement             with computer,                                                      anterior and 92% for       between
                                           57% agreement                                                       inferior. In prior         EASI and
                                           for isolated ST-                                                    inferior MI, EASI had      standard less
                                           T in old MI                                                         greater sensitivity than   than 90% in
                                                                                                               standard (59% vs 55%)      leads II, III,
                                                                                                               but lower specificity      aVL, aVF,
                                                                                                               (95% vs 99%)               V4 and V5
                                                                                                                                          for q wave
Normal ECG                                 80% agreement                  98%           Similar deviations
                                           with computer                  agreement     between paramedic
                                                                                        and EASI.
Lead                 Transition                                                         EASI-Gold                                         Correlations
differences          zone V2-V3                                                         compared with                                     less than
                     in EASI and                                                        Paramedic-Gold                                    90% in III,
                     V3-V4 in                                                           revealed significant                              aVL for
                     traditional                                                        differences; inc Q                                normal and
                                                                                        amp in V6, smaller                                post MI
                                                                                        R amp in V1, V2                                   patients with
                                                                                        and V4. Less ST in                                no Q, Q and
                                                                                        V1, inc ST in V4                                  VT
                                                                                        and V5 and inc T
                                                                                        amp in V6

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                              14
Author: Bernice Redley
Parameter            Drew (1992)           Rautaharju,        Welinder,   Drew (1997)     Sejersten, (2003)   Drew (1999)               Horacek
                                           (2002)             (2004)                                                                    (2000)
Axis                 Frontal plane                                                                            Mean differences in P
                     Axis in                                                                                  wave and QRS and T-
                     Precordial                                                                               wave axis were 21
                     leads only                                                                               degrees, t-wave axis
                     Difference 19                                                                            differed more between
                     deg. 95% CI                                                                              the two ECG types
                     of 14-24 deg.                                                                            mean 30 degrees.
Conduction           100%                                                                                     Agreement for BBB
                     agreement on                                                                             and fascicular blocks
                     BBB and                                                                                  ranged from 95-99%.
                     fascicular                                                                               EASI closer to expert
                     blocks                                                                                   in RBBB.
False                EASI False            10% clinically                 No false                            2 misdiagnosed by
diagnosis            positive in 2         significant                    positives and                       both the EASI and
                     and false             differences B/W                1 false                             standard 12 lead due to
                     negative in 4         EASI and                       negative with                       low amplitude p
                     with prior MI         traditional 3.4%               derived ECG                         waves.
                     and traditional       q-MI and 11 ST
                     False negative        changes. None
                     in 6 pts with         of normal
                     prior MI              classed as old
                                           MI and none of
                                           Q-wave missed.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                          15
Author: Bernice Redley
Parameter            Drew (1992)           Rautaharju,   Welinder,   Drew (1997)   Sejersten,   Drew (1999)                 Horacek
                                           (2002)        (2004)                    (2003)                                   (2000)
Cardiac              100% agreement                                                             100% agreement              Correlations
rhythm               in diagnosis and                                                                                       less than
                     misdiagnosis for                                                                                       90% for
                     Wide complex                                                                                           leads II, III,
                     tachycardia                                                                                            aVL, aVF,
                     Difference in axis                                                                                     V3, V4, V5
                     17 deg with CI                                                                                         between
                     95% 12-22 deg.                                                                                         EASI and
                                                                                                                            patients with
                     SVT Vs VT
                     criteria agreement
                     in V1, V2, V6
Chamber              Lower V3-V4                                                                177 evaluated for atrial
enlargement          voltages, leading                                                          enlargement using Echo.
                     to inability to                                                            RAH in 17 patients and
                     detect LVH in 1/3                                                          LAH in 46 patients, RVH
                     of patients with                                                           in 15 and LVH in 68.
                     wide complex                                                               Agreement was between
                     tachycardia 39%                                                            84-98%. Both methods
                     in V3 and 55% in                                                           had low sensitivity but
                     V4                                                                         high specificity. For LVH
                                                                                                EASI had higher
                                                                                                specificity than standard
                                                                                                (98% vs 88% respectively)
                                                                                                Both methods insensitive
                                                                                                to chamber enlargement.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                16
Author: Bernice Redley
Parameter            Drew (1992)           Rautaharju,   Welinder,    Drew (1997)   Sejersten,   Drew (1999)                     Horacek
                                           (2002)        (2004)                     (2003)                                       (2000)
Others               Ventricular pre-                                                            Cardiac intervals for PR,
                     excitation                                                                  QRS, QT and QTC varied
                     patterns 100%                                                               by 0-2ms. The observed
                     agreement                                                                   difference of 1ms in QRS
                                                                                                 was statistically significant
                                                                                                 but not clinically
Artifact             Less muscle                         Similar
                     artifact after                      baseline
                     DCCR and on                         wander,
                     arm movement.                       EASI less
                                                         noise for
                                                         and supine
                                                         to L.
No MI                                      80%

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                   17
Author: Bernice Redley
                                                                Other Variables to consider:
Parameter            Drew (1992)           Rautaharju,      Welinder,       Drew (1997)    Sejersten,       Drew (1999)                 Horacek
                                           (2002)           (2004)                         (2003)                                       (2000)
Electrodes           Not identified        Not identified   Identified                     Not identified   No specified
Method of            Marquette             Standard         Identified   Described         Identified       Identified                  Simultaneou
deriving             computer              measure          Simultaneou ECGs                                                            s recording
                                                                                           Three            Simultaneous monitoring
traditional          program.              ECGs             s collection collected                                                      supine for 15
ECG                  Simultaneous          collected in                  simultaneousl                                                  seconds
                                                            of ECGs                        recordings on
(machine             measurement           time sequence                 y during          same subject.
used)                                                                    procedure
Application          In the ED and         Method of       Identified       I V4 used in   Paramedic        Not identified, presumed    Nor
of electrodes        PCTA procedure        application                      ½ study and    and expert       to be in the ED/ cath lab   specified.
by whom/             Not identified        Not identified,                  EASI for       (Traditional
position                                   position id.                     latter ½       and EASI)
Setting              ED & EP studies,      PCTA, AMI        Laboratory      PCTA.          Experimental     ED and catheter lab         Not specified
                                           pts with                                        conditions at                                but presumed
                                           enzyme ?                                        conference                                   to be a
                                           from location                                                                                clinical
                                           suggests                                                                                     environment
                                           clinical                                                                                     and clinical
                                           situations.                                                                                  diagnostic
                                                                                                                                        criteria used.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                                                                                             18
Author: Bernice Redley

1.       Dower, G.E., The ECGD: a derivation of the ECG from VCG leads. Journal of
         Electrocardiology, 1984. 17(2): p. 189-91.
2.       Dower, G.E., H.B. Machado, and J.A. Osborne, On deriving the
         electrocardiogram from vectorcadiographic leads. Clinical Cardiology, 1980.
         3(2): p. 87-95.
3.       Dower, G.E., et al., Deriving the 12-lead electrocardiogram from four (EASI)
         electrodes. Journal of Electrocardiology, 1988. 21 Suppl: p. S182-7.
4.       Sejersten, M., et al., The relative accuracies of ECG precordial lead
         waveforms derived from EASI leads and those acquired from paramedic
         applied standard leads. Journal of Electrocardiology, 2003. 36(3): p. 179-85.
5.       Drew, B.J., et al., Practice standards for electrographic monitoring in hospital
         settings. Circulation, 2004. 110: p. 2721-2746.
6.       Welinder, A., et al., Comparison of signal quality between EASI and Mason-
         Likar 12-lead electrocardiograms during physical activity. American Journal
         of Critical Care, 2004. 13(3): p. 228-234.
7.       Drew, B.J., M.M. Scheinman, and G.T. Evans, Jr., Comparison of a
         vectorcardiographically derived 12-lead electrocardiogram with the
         conventional electrocardiogram during wide QRS complex tachycardia, and
         its potential application for continuous bedside monitoring. American Journal
         of Cardiology, 1992. 69(6): p. 612-8.
8.       Adams, M.G. and B.J. Drew, Body position effects on the ECG: implication
         for ischemia monitoring. Journal of Electrocardiology, 1997. 30(4): p. 285-91.
9.       Horacek, B.M., et al. Diagnostic accuracy of derived compared to standard
         12-lead electrocardiograms. in International Society for Computerized
         Cardiology. 2000. Yosemite, California: Journal of Electrocardiology.
10.      Drew, B.J., et al., Value of a derived 12-lead ECG for detecting transient
         myocardial ischemia. Journal of Electrocardiology, 1995. 28 Suppl: p. 211.
11.      Drew, B.J. and B. Ide, Diagnosing ischemia from the bedside monitor.
         Prognostic Cardiovascular Nursing, 1996. 11(1): p. 45-6.
12.      Drew, B.J., et al., Accuracy of the EASI 12-lead electrocardiogram compares
         to the standard 12-lead electrocardiogram for diagnosing multiple cardiac
         abnormalities. Journal of Electrocardiology, 1999. 32: p. 38-47.
13.      Drew, B.J., et al., Comparison of standard and derived 12-lead
         electrocardiograms for diagnosis of coronary angioplasty-induced myocardial
         ischemia. American Journal of Cardiology, 1997. 79(5): p. 639-44.
14.      Rautaharju, P.M., et al., Comparability of 12-lead ECGs derived from EASI
         leads with standard 12-lead ECGS in the classification of acute myocardial
         ischemia and old myocardial infarction. Journal of Electrocardiology, 2002.
         35 Suppl: p. 35-9.
15.      Drew, B.J., et al., Derived 12-lead ECG. Comparison with the standard ECG
         during myocardial ischemia and its potential application for continuous ST-
         segment monitoring. Journal of Electrocardiology, 1994. 27 Suppl: p. 249-55.
16.      Drew, B.J., et al., Comparison of a new reduced lead set ECG with the
         standard ECG for diagnosing cardiac arrhythmias and myocardial ischemia.
         Journal of Electrocardiology, 2002. 35 Suppl: p. 13-21.

2bc91db2-70de-490c-92f8-9b3e5f39733c.doc                                              19
Author: Bernice Redley