Docstoc

Assessment of the Standardized Reporting Guidelines ECG miocardial infarction

Document Sample
Assessment of the Standardized Reporting Guidelines ECG  miocardial infarction Powered By Docstoc
					Assessment of the Standardized Reporting
Guidelines ECG Classification System: The
Presenting ECG Predicts 30-Day Outcomes


 ANNALS OF EMERGENCY MEDICINE 4 4 : 3 SEPTEMBER 2004
                  INTRODUCTION
 Chest pain at ED: 1/3 cardiac cause
 Inappropriate discharge leads to medical error
 Many studies for better methods to risk-stratify: none
  widespread accepted
 Current draft of guidelines (Standardized Reporting Criteria
    Working Group of the Emergency Medicine Cardiovascular
    Research and Education Group): recommend    ECG classified
  into 1 of 6 categories
 Evaluate recommended classification system to predict
  endpoint of death, MI, revascularization within 30 days
     MATERIALS AND METHODS
Study Design
 Prospective study
Setting
 July 1999 to March 2002: Hospital of the University of
  Pennsylvania, urban tertiary care hospital ED in
  Philadelphia
 Included: >24 y/o or <24 y/o with cocaine use, with
  chest pain prompting an ECG
 Excluded: <24 y/o, without chest pain, chest pain
  without ECG.
     MATERIALS AND METHODS

Data Collection and Processing
 Information: directly from the treating physician
 Demographic characteristics, cardiac risk factors,
  chest pain characteristics, associated symptoms,
  medications, initial vital signs, treatment,
  disposition.
 Daily f/u for death, MI, revascularization for 30
  days
     MATERIALS AND METHODS

Standardized Reporting Criteria
 Standardized Reporting Criteria Working Group of the
  Emergency Medicine Cardiovascular Research and
  Education Group
 47 ‘core characteristics’ organized into 8 major
  reporting categories: inclusion criteria, demographic
  characteristics, ECG interpretation, cardiac risk factors,
  chief complaint, cardiac markers, patient course,
  outcomes.
   Figure 1.
  Electrocardiographic core criteria advocated by the guidelines.
Report who interpreted the ECGs. The investigator should clarify whether the
   ECG interpretation was done by the clinician when patient care was provided
   or by an investigator after patient enrollment in the study.
ECG analysis should be reported in terms of specific findings suggestive of
   acute coronary syndrome.
 Rate
 Rhythm
 ST-segment elevation and depression (and unit of deviation: 1 vs 2 mm/mV).
   Should be reported as ‘‘known to be old’’ or ‘‘not known to be old.’’
 T-wave inversion (should report standardized depth). Should be reported as
   ‘‘known to be old’’ or ‘‘not known to be old.’’
 Left bundle-branch block (presence or absence). Should be reported as
   ‘‘known to be old’’ or ‘‘not known to be old.’’
 Right bundle-branch block (presence or absence). Should be reported as
   ‘‘known to be old’’ or ‘‘not known to be old.’’
Report overall categorization of the ECG
 Normal
 Nonspecific
 Abnormal but not diagnostic of ischemia
 Ischemia known to be old
 Ischemia or previous infarction not known to be old
 ST elevation consistent with acute myocardial infarction
The definitions assigned to each category are as follows:
 Normal: no possible evidence for ischemia
 Nonspecific: accepted deviation from the norm, with the lowest likelihood of ischemia (eg,
   inverted T-wave axis in III or V1)
 Abnormal but not diagnostic of ischemia: prolonged PR, QRS, QTc intervals, bundle
   branch blocks, left ventricular hypertrophy with strain
 Ischemia or previous infarction not known to be old: ST-segment depression >0.1 mV
   measured 80 ms from the J point, inverted T waves >0.3 mV, or Q waves 30 ms in
   duration
 Ischemia or previous infarction known to be old
 Suggestive of acute myocardial infarction: ST elevation >0.1 mV measured 80 ms from
   the J point in 2 contiguous leads, with or without reciprocal ST depressions.
       MATERIALS AND METHODS
Outcome Measures
   Include physicians of all specialties rotating to ED at all levels of
    training
 Assess whether the proposed classification system is generalized
    to any treating physician.
 5 categories in this study:
(1) normal
(2) early repolarization without other abnormalities
(3) abnormal but not diagnostic of ischemia: prolonged PR, QRS, or
    QTc, BBB, LVH with strain, nonspecific intraventricular conduction
    deficits, NSSTTC
(4) ischemia: STD > 0.1 mV, 80 ms from J, TWI > 0.3 mV, Q wave at
    least 30 ms
(5) suggestive of MI: STE > 0.1 mV, 80 ms from J in > 2 contiguous
    leads, with or without reciprocal change
     MATERIALS AND METHODS
 Diagnosis of AMI: troponin I >= 2 ng/mL, CK-MB >=10
  ng/mL
 Revascularization: PCI, bypass
 Death: all-cause mortality.
Primary Data Analysis
 ECG classification category in relation to rate of death,
  MI, revascularization within 30 days
 SAS statistical software
 Patient with multiple ED visits: each visit counted
  separately, evaluated as an individual event
        MATERIALS AND METHODS
   Standardized reporting guidelines not available in beginning:
    ECG classification in study slightly different.
   Standardized: Normal, Nonspecific, Abnormal but not
    diagnostic of ischemia, Ischemia known to be old, Ischemia
    or previous infarction not known to be old, ST elevation
    consistent with acute myocardial infarction
   Normal, ischemia, suggestive of MI: identical to guidelines
   Abnormal nondiagnostic: nonspecific, abnormal
    nondiagnostic
   Further analysis in final: ST, T-wave changes or BBB:
    abnormal nondiagnostic, without: nonspecific.
                                        RESULTS
Demographic characteristics of the patient population.
Characteristic                                                Study Patients,No. (%) (N=4,487 Visits)
Mean age (6SD), y*                                             51.86+-15.9
Ethnicity*
Asian                                                          69 (2)
Black                                                          2,611 (68)
Hispanic                                                      43 (1)
Other/unknown                                                  31 (1)
White                                                         1,060 (28)
Sex*
Male                                                          1,547 (41)
Female                                                        2,267 (59)
Cardiac risk factors
Hypertension                                                   2,192 (49)
Family history of premature coronary artery disease            903 (20)
Diabetes mellitus                                              805 (18)
Tobacco use                                                   1,684 (38)
Hypercholesterolemia/hyperlipidemia                            789 (18)
Cocaine use                                                   104 (2)
Previous myocardial infarction                                497 (11)
Known coronary artery disease                                 878 (20)
*Age, sex, and ethnicity are counted only once per patient (n=3,814), regardless of the number of visits.
                  RESULTS
 1,672 (37%) patients: discharged; 2,147 (48%)
  admitted to telemetry, 496 (11%): ICU, 134 (3%):
  unmonitored beds, 14 (<1%): catheterization; 13:
  AAD, 10: transfer.
 Cardiac markers obtained in 3,262 (73%) patient
  visits.
                                    RESULTS
Details of the presenting ECG.
Characteristic*                                              Study Patients,No. (%) (N=4,487)
Overall ECG impression
Normal                                                      1,912 (43)
Early repolarization only                                   171 (4)
Nonspecific/nondiagnostic changes                           2,081 (50)
Ischemia                                                    224 (5)
Acute myocardial infarction                                 99 (2)
ST-segment elevation >1 mm                                  420 (9)
ST-segment depression >0.5 mm                               338 (8)
T-wave inversion >5 mm                                      885 (20)
Hyperacute T waves                                          115 (3)
Pathologic Q waves                                           261 (6)
Left bundle-branch block                                    90 (2)
Right bundle-branch block                                   125 (3)
* Specific definitions were used for the classification system, ST-segment elevation, ST-segment
     depression, and T-wave inversion. There were no specific definitions onthe case report forms for
     the other items.
                      RESULTS
   Most: nonspecific, nondiagnostic
   Initial hospital course: 318 (7%) AMI, additional 545 (12%)
    unstable angina, 28 died, 11 late MI (>24 hours), 111 PCI,
    26 CABG, 399 meet endpoint (death, MI,
    revascularization)
   98% 30 days follow-up, 446 patients meet endpoint, 52
    deaths, 30 MI, 17 PCI, 11 CABG.
   Significant (P<.0.01) relationship between initial ECG
    classification and 30-day outcome.
   Further analysis in abnormal nondiagnostic: nonspecific,
    abnormal nondiagnostic
   ST, T-wave abnormalities, BBB: higher risk than patients
    without these
                    LIMITATIONS
   ECG classification system in this study: not exactly
    match recommended classification guidelines.
   Subset analysis in abnormal but not diagnostic: different
    prognosis between with and without ST, T-wave
    changes.
   Presenting ECG vs outcome: additional ECGs would be
    better?
   Whether ECGs were obtained during chest pain attack?
   A triple composite outcome not individual: need more
    samples
   Population predominantly black
                 DISCUSSION
 Developing standardized reporting guidelines for risk-
  stratification in potential acute coronary syndromes.
 Include 8 major reporting categories with 47 core
  criteria: 9 of these are electrocardiographic criteria.
 High interrater reliability in this recommend ECG
  classification system: presenting ECG classification
  predicts inhospital, 30-day triple composite endpoint
  (death, MI, revascularization)
 Similar ECG interpretation schemes in predicting short-
  term complication in critical care populations, AMI in
  older ED chest pain populations.
               DISCUSSION
 Normal, nonspecific ECG: low risk short-term
  cardiovascular complications.
 Ischemia: higher risk
 Abnormal but nondiagnostic: intermediate risk,
 Further analysis: lower risk in nonspecific then
  abnormal but nondiagnostic
 Conclusion: 6-item ECG classification system
  recommended in the standardized guidelines
  predicts 30-day rates of death, acute myocardial
  infarction, revascularization.
THNAKS FOR YOUR
   ATTENTION.

				
DOCUMENT INFO