Variability of Automated Intraoperative ST Segment Values Predicts Postoperative Troponin Elevation

Area of Science:

• Anesthesiology
• Cardiology
• Medical Informatics

Background:

• Intraoperative electrocardiographic monitoring is standard, but lacks evidence-based algorithms for predicting adverse cardiac events.
• Current methods do not effectively utilize ST segment data for risk stratification.
• Exploratory study to assess ST segment variability and its association with postoperative cardiac events.

Purpose of the Study:

• To determine if statistical measures of intraoperative ST segment values are associated with adverse postoperative events.
• To investigate the hypothesis that ST segment elevation, depression, and variability predict postoperative troponin elevation.

Main Methods:

• Retrospective analysis of intraoperative ST segment measurements (max, min, mean, SD) from leads I, II, and III in adult noncardiac surgery patients.
• Logistic regression models used to identify associations between ST segment values and myocardial injury (elevated troponin).
• Model performance assessed using receiver operator characteristic curve (AUROC) and net reclassification improvement (NRI).

Main Results:

• Postoperative myocardial injury occurred in 5.6% of 81,011 subjects.
• Maximal ST segment depression, elevation, and SD were significantly associated with myocardial injury.
• Increased SD correlated with decreased risk, even after accounting for maximal ST depression/elevation and patient characteristics.
• The ST segment model demonstrated fair discrimination (AUROC=0.71) and improved prediction (NRI=0.0345).

Conclusions:

• Automated ST segment analysis during anesthesia can enhance prediction of postoperative troponin elevation.
• Intraoperative ST segment data offers valuable insights for perioperative cardiac risk assessment.
• Further research can refine algorithms for utilizing ST segment data in clinical practice.