Real Time ST-segment Deviation Detection in High-risk Patients Detected by Wireless Single-lead ECG

Overview

This study primarily aims to describe the frequency of significant ST-deviations, defined as ECG-ST-deviations \<-0.255 or \>0.245 mV for a minimum duration of 30 minutes as measured by a single-lead ECG in patients admitted with AECOPD or following major abdominal surgery. Secondarily we will describe the frequency of ST-deviations \<-0.255 or \>0.245 mV for a minimum duration of 1, 10 and 20 minutes, respectively, as well as for patients with ST-deviations \<-0.1 or \>0.1 mV for a minimum duration of 1, 10, 20 and 30 minutes, respectively. Lastly, we will investigate the association between ST-deviations and subsequent myocardial injury while adjusting for known risk factors.

Background: Assesment of wireless ECG-monitoring devices is essential for patient monitoring as it enables activity and out-of-hospital monitoring. Former studies have found a random variation for ST-deviations between wireless single-lead ECG and 12-lead ECG, which suggests that clinical monitoring of ST-deviations by single-lead ECG must be corrected for this to minimize false positive cases. Thus, we aimed to assess how a corrected limit of agreement for ST-deviations affects the ability to detect true myocardial ischemia by comparing thresholds to troponin elevations. Methods: This study sought to analyze data from two WARD cohort studies (NCT03491137) and (NCT03660501). Patients were included at Rigshospitalet and Bispebjerg Hospital from February 2018 to September 2020. The Surgery and COPD studies consisted of 505 and 202 patients, respectively, who were hospitalized with AECOPD or scheduled for elective major abdominal surgery lasting \> 2 hours. All patients were monitored with wireless equipment, which continuously measured vital signs. The continuous wireless monitoring used in this study consisted of a single-lead ECG patch (Isansys Life Touch patch) which consists of two ECG electrodes placed on the left side of the patients' thorax. The patch measures respiratory rate (RR), heart rate (HR), heart rate variability (HRV) and electrocardiogram (ECG)s. Another monitoring device was the plasma concentration of high-sensitive cardiac troponin T (hsTnT, Cobas 8000, e801 module, Diagnostics Roche). In this study, patients were divided into a myocardial injury group with TnT elevations (of 20-64 ng/L with an absolute change of ≥ 5 ng/L, or a hsTnT level ≥65 ng/L, both evaluated as due to ischemic etiology) and a group with no myocardial injury. Analysis: Baseline characteristics of patients undergoing major abdominal surgery and patients admitted with AECOPD with and without myocardial injury will be summarized in frequency tables including Using descriptive statistics, we will describe the occurrence of ST-segment deviations in the single-lead (Lifetouch) ECG in patients with and without myocardial injury as well as the association between ST-deviation and subsequent myocardial injury, which will be adjusted for known risk factors; age, gender, BMI, daily smoker (never-, previous-, current smoker), alcohol consumption (none, below- or above recommendations), ASA classification (ASA 1,2,3,4), history of myocardial infarction, history of diabetes mellitus, history of chronic heart failure. Furthermore, the sensitivity and specificity of continuous wireless single-lead ECG with the applied threshold of \<-0.255 mV and \> 0.245 mV will be determined. Receiver operating characteristics (ROC) analyses will be made to determine which type of ST-deviation and which specific threshold value was most sensitive for TnT elevations. Through ROC analyses it will also be examined how different variations of durations affect the thresholds' sensitivity and specificity. Statistical analysis will be performed using statistical software SPSS.