Automatic Holter electrocardiogram analysis in ischaemic stroke patients to detect paroxysmal atrial fibrillation: ready to replace physicians?

Intelligent Algorithm-Based Electrocardiography to Predict Atrial Fibrillation after Coronary Artery Bypass Grafting in the Elderly

The objective of this study was to explore the predictive value of electrocardiogram (ECG) based on intelligent analysis algorithm for atrial fibrillation (AF) in elderly patients undergoing coronary artery bypass grafting (CABG). Specifically, 106 elderly patients with coronary heart disease who underwent CABG in the hospital were selected, including 52 patients with postoperative AF (AF group) and 54 patients without arrhythmia (control group). Within 1-3 weeks after operation, the dynamic ECG monitoring system based on Gentle AdaBoost algorithm constructed in this study was adopted. After the measurement of the 12-lead P wave duration, the maximum P wave duration (Pmax) and minimum P wave duration (Pmin) were recorded. As for simulation experiments, the same data was used as the back-propagation algorithm. The results showed that for the detection accuracy of the test samples, the Gentle AdaBoost algorithm showed 93.7% accuracy after the first iteration, and the Gentle AdaBoost algorithm was 16.1% higher than the back-propagation algorithm. Compared with the control group, the detection rate of arrhythmia in patients after CABG was significantly lower ( $P<0.05$ ). Bivariate logistic regression analysis on Pmax and Pmin showed as follows: Pmax: 95% confidential interval (CI): 1.024-1.081, $P<0.05$ ; Pmin: 95% CI: 1.036-1.117, $P<0.05$ . The sensitivity of Pmax and Pmin in predicting paroxysmal AF was 78.2% and 73.4%, respectively; the specificity of them was 80.1% and 85.6%, respectively; the positive predictive value was 81.2% and 83.4%, respectively; and the negative predictive value was 79.5% and 75.3%, respectively. In conclusion, the generalization ability of Gentle AdaBoost algorithm was better than that of back-propagation algorithm, and it can identify arrhythmia better. Pmax and Pmin were important indicators of AF after CABG.

Atrial Fibrillation Detected After Stroke and Transient Ischemic Attack: A Novel Clinical Concept Challenging Current Views

Atrial fibrillation (AF) can be newly detected in approximately one-fourth of patients with ischemic stroke and transient ischemic attack without previously recognized AF. We present updated evidence supporting that AF detected after stroke or transient ischemic attack (AFDAS) may be a distinct clinical entity from AF known before stroke occurrence (known atrial fibrillation). Data suggest that AFDAS can arise from the interplay of cardiogenic and neurogenic forces. The embolic risk of AFDAS can be understood as a gradient defined by the prevalence of vascular comorbidities, the burden of AF, neurogenic autonomic changes, and the severity of atrial cardiopathy. The balance of existing data indicates that AFDAS has a lower prevalence of cardiovascular comorbidities, a lower degree of cardiac abnormalities than known atrial fibrillation, a high proportion (52%) of very brief (<30 seconds) AF paroxysms, and is more frequently associated with insular brain infarction. These distinctive features of AFDAS may explain its recently observed lower associated risk of stroke than known atrial fibrillation. We present an updated ad-hoc meta-analysis of randomized clinical trials in which the association between prolonged cardiac monitoring and reduced risk of ischemic stroke was nonsignificant (incidence rate ratio, 0.90 [95% CI, 0.71-1.15]). These findings highlight that larger and sufficiently powered randomized controlled trials of prolonged cardiac monitoring assessing the risk of stroke recurrence are needed. Meanwhile, we call for further research on AFDAS and stroke recurrence, and a tailored approach when using prolonged cardiac monitoring after ischemic stroke or transient ischemic attack, focusing on patients at higher risk of AFDAS and, more importantly, at higher risk of cardiac embolism.

Effect of the recording condition on the quality of a single-lead electrocardiogram

Although many wearable single-lead electrocardiogram (ECG) monitoring devices have been developed, information regarding their ECG quality is limited. This study aimed to evaluate the quality of single-lead ECG in healthy subjects under various conditions (body positions and motions) and in patients with arrhythmias, to estimate requirements for automatic analysis, and to identify a way to improve ECG quality by changing the type and placement of electrodes. A single-lead ECG transmitter was placed on the sternum with a pair of electrodes, and ECG was simultaneously recorded with a conventional Holter ECG in 12 healthy subjects under various conditions and 35 patients with arrhythmias. Subjects with arrhythmias were divided into sinus rhythm (SR) and atrial fibrillation (AF) groups. ECG quality was assessed by calculating the sensitivity and positive predictive value (PPV) of the visual detection of QRS complexes (vQRS), automatic detection of QRS complexes (aQRS), and visual detection of P waves (vP). Accuracy was defined as a 100% sensitivity and PPV. We also measured the amplitude of the baseline, P wave, and QRS complex, and calculated the signal-to-noise ratio (SNR). We then focused on aQRS and estimated thresholds to obtain an accurate aQRS in more than 95% of the data. Finally, we sought to improve ECG quality by changing electrode placement using offset-type electrodes in 10 healthy subjects. The single-lead ECG provided 100% accuracy for vQRS, 87% for aQRS, and 74% for vP in healthy subjects under various conditions. Failure for accurate detection occurred in several motions in which the baseline amplitude was increased or in subjects with low QRS or P amplitude, resulting in low SNR. The single-lead ECG provided 97% accuracy for vQRS, 80% for aQRS in patients with arrhythmias, and 95% accuracy for vP in the SR group. The AF group showed higher baseline amplitude than the SR group (0.08 mV vs. 0.02 mV, P < 0.01) but no significant difference in accuracy for aQRS (79% vs. 81%, P = 1.00). The thresholds to obtain an accurate aQRS were a QRS amplitude > 0.42 mV and a baseline amplitude < 0.20 mV. The QRS amplitude was significantly influenced by electrode placement and body position (P < 0.01 for both, two-way analysis of variance), and the maximum reduction by changing body position was estimated as 30% compared to the sitting posture. The QRS amplitude significantly increased when the inter-electrode distance was extended vertically (1.51 mV for vertical extension vs. 0.93 mV for control, P < 0.01). The single-lead ECG provided at least 97% accuracy for vQRS, 80% for aQRS, and 74% for vP. To obtain stable aQRS in any body positions, a QRS amplitude > 0.60 mV and a baseline amplitude < 0.20 mV were required in the sitting posture considering the reduction induced by changing body position. Vertical extension of the inter-electrode distance increased the QRS amplitude.

[Cardiac diagnostics after ischemic stroke or transitory ischemic attack]

Stroke is the most common cause of permanent disability and one of the most common causes of death. Cardio-embolic strokes are associated with a poor prognosis and a high risk of recurrence compared to other stroke etiologies. The most common source of cardiac embolism is atrial fibrillation which must be quickly identified to optimize secondary stroke prevention. A structured evaluation after ischemic stroke includes taking the medical history, a physical examination, 12-lead ECG recording, rhythm monitoring for 72 h, transthoracic echocardiography and transesophageal echocardiography, if an atrial embolic source of stroke is suspected. Extended cardiac work-up (e. g., MRI/CT, prolonged rhythm monitoring) should be performed in selected patients based on diagnostic findings.

Software-based analysis of 1-hour Holter ECG to select for prolonged ECG monitoring after stroke

Objective

Identification of ischemic stroke patients at high risk for paroxysmal atrial fibrillation (pAF) during 72 hours Holter ECG might be useful to individualize the allocation of prolonged ECG monitoring times, currently not routinely applied in clinical practice.

Methods

In a prospective multicenter study, the first analysable hour of raw ECG data from prolonged 72 hours Holter ECG monitoring in 1031 patients with acute ischemic stroke/TIA presenting in sinus rhythm was classified by an automated software (AA) into “no risk of AF” or “risk of AF” and compared to clinical variables to predict AF during 72 hours Holter‐ECG.

Results

pAF was diagnosed in 54 patients (5.2%; mean age: 78 years; female 56%) and was more frequently detected after 72 hours in patients classified by AA as “risk of AF” (n = 21, 17.8%) compared to “no risk of AF” (n = 33, 3.6%). AA‐based risk stratification as “risk of AF” remained in the prediction model for pAF detection during 72 hours Holter ECG (OR3.814, 95% CI 2.024‐7.816, P < 0.001), in addition to age (OR1.052, 95% CI 1.021‐1.084, P = 0.001), NIHSS (OR 1.087, 95% CI 1.023‐1.154, P = 0.007) and prior treatment with thrombolysis (OR2.639, 95% CI 1.313‐5.306, P = 0.006). Similarly, risk stratification by AA significantly increased the area under the receiver operating characteristic curve (AUC) for prediction of pAF detection compared to a purely clinical risk score (AS5F alone: AUC 0.751; 95% CI 0.724‐0.778; AUC for the combination: 0.789, 95% CI 0.763‐0.814; difference between the AUC P = 0.022).

Interpretation

Automated software‐based ECG risk stratification selects patients with high risk of AF during 72 hours Holter ECG and adds predictive value to common clinical risk factors for AF prediction.