Associations Between Electrocardiographic Features and Type 1 and Type 2 Myocardial Infarction in Critically Ill...

Area of Science:

• Cardiology
• Critical Care Medicine
• Medical Diagnostics

Background:

• Electrocardiograms (ECGs) are vital for diagnosing myocardial infarction (MI).
• Distinguishing between Type 1 MI (atherosclerotic plaque rupture) and Type 2 MI (supply-demand mismatch) in critically ill patients using ECGs remains challenging.
• Existing research on ECG utility for MI subtyping in intensive care units (ICUs) is limited.

Purpose of the Study:

• To investigate the association between specific ECG features and MI subtypes (Type 1 vs. Type 2) in critically ill adult patients.
• To evaluate the diagnostic performance of ECG findings in differentiating these MI types within an ICU setting.
• To assess the clinical utility of ECGs for MI subtyping in complex patient populations.

Main Methods:

• Retrospective observational study utilizing the MIMIC-IV database.
• Analysis of ECG interpretation features (ST-segment changes, T-wave abnormalities, QTc prolongation, arrhythmia) in 2,159 adult ICU patients.
• Multivariable logistic regression to assess associations between ECG features and MI subtypes (Type 1 vs. Type 2), defined by ICD-10 codes and elevated troponin.

Main Results:

• ST-segment elevation was more frequent in Type 1 MI (9.4% vs. 3.1%), showing a significant association (aOR 2.48).
• QTc prolongation was highly prevalent, especially in Type 2 MI (66.2% vs. 52.9%), and associated with lower odds of Type 1 MI (aOR 0.63).
• The overall model discrimination was modest (AUC 0.73), with high sensitivity but low specificity for differentiating MI types due to class imbalance.

Conclusions:

• ECG features alone offer limited clinical and mechanistic insight for differentiating Type 1 and Type 2 MI in critically ill patients.
• The diagnostic performance of ECGs for MI subtyping in this population is constrained.
• ECG interpretation should be integrated with the broader clinical and physiological context for accurate MI diagnosis in ICUs.