Related Experiment Videos
Interobserver variation in interpretation of electrocardiographic signs of atrial infarction
J H Christensen1, F E Nielsen, N Falstie-Jensen
1Department of Cardiology, Aalborg Hospital, Denmark.
Insights
Diagnosing atrial infarction using electrocardiogram (ECG) criteria shows significant interobserver variation. Specific ECG signs for atrial infarction require further validation due to inconsistent detection among physicians.
Area of Science:
- Cardiology
- Diagnostic Imaging
Background:
- The electrocardiogram (ECG) is crucial for diagnosing atrial infarction antemortem.
- PR-segment displacements on ECG have been historically considered indicators of atrial infarction.
Purpose of the Study:
- To evaluate the interobserver variability of proposed ECG criteria for diagnosing atrial infarction in acute myocardial infarction patients.
Main Methods:
- Three physicians independently assessed ECGs from 290 acute myocardial infarction patients.
- Seven specific criteria suggestive of atrial infarction were evaluated, including PR-segment displacements and abnormal P waves.
- Interobserver agreement was quantified using Kappa values.
Main Results:
- Kappa values for the seven criteria ranged from 0.00 to 0.86.
- Only three criteria (PR-segment depressions in leads II/III, PR-segment depressions in V1/V2, and abnormal P waves) were detectable in a sufficient number of patients.
- Considerable interobserver variation was observed for these detectable criteria, with fair to moderate agreement.
Conclusions:
- The interobserver reliability of current ECG criteria for atrial infarction is limited.
- Further refinement and validation of ECG diagnostic criteria for atrial infarction are necessary to improve diagnostic accuracy and consistency.
Abstract:
The electrocardiogram (ECG) is the only means of diagnosing atrial infarction antemortem. Certain ECG changes (PR-segment displacements) have been taken earlier as signs of atrial infarction. The purpose of this study was to assess the interobserver variation on suggested ECG signs of atrial infarction in patients admitted with acute myocardial infarction. The ECGs from 290 patients were evaluated by three physicians with respect to the occurrence of each of the following seven criteria suggestive of atrial infarction: (1) PR-segment elevation > 0.5 mm in lead I; (2) PR-segment depressions > 0.5 mm in leads II and III; (3) PR-segment depressions > 1.2 mm in leads I, II, and III; (4) PR-segment depressions > 0.5 mm in leads V1 and V2; (5) PR-segment elevations > 0.5 mm in leads V5 and V6; (6) PR-segment depressions > 1.5 mm in precordial leads; and (7) abnormal P waves. Kappa values ranged from 0.00-0.86. Of the seven criteria tested only criteria 2, 4, and 7 could be detected in a reasonable number of patients. The interobserver variations of these criteria were considerable, although the strength of agreement could be designated as fair to moderate.