Heart Rate Variability to Quantify General Anesthesia Depth

The shortage of anesthetic agents can lead to intraoperative awareness while overdosing can trigger severe intra and postoperative problems. Therefore, monitoring anesthesia's depth (DoA: Depth of Anesthesia) is a crucial but still challenging task. Although some commercial monitors are based on electroencephalogram (EEG), designed to quantify DoA, their use in clinical practice has limitations. On the other hand, heart rate variability (HRV) has valuable information about physiological states, both from the heart and the organism. Classical indices derived from HRV have been shown to be able to differentiate the different stages of anesthesia. In this study, it is proposed to create a model to monitor DoA combining several HRV indices. Patients will be divided into three groups, according to the type of anesthesia to which they will be submitted (inhalation, total or balanced intravenous) and will have the electrocardiogram recorded during the entire surgical procedure. Various HRV indices will be calculated, and machine learning techniques will be used to combine and identify the most relevant index to compose a score that reliably represents DoA. Several commercial devices have been developed to monitor the level of consciousness during anesthesia. Among the most popular tools are included: Narcotrend TM (MonitorTechnik, Bad Bramstedt, Germany); the M-Entropy TM (GE Healthcare, Helsinki, Finland); Nindex SA (Controls, Montevideo, Uruguay) and the Bi-Spectral Index (BIS, TM Medtronic-Covidien, Dublin, Ireland). In this study, BIS or Nindex will also be monitored during the entire period that the patients remain anesthetized and will later be used to compose the DoA score based on HRV. As a result, a computer program will be created to monitor DoA in real-time.