Clinical Window International web journal for medical professionals
Menu

Functions

 

 

 

Sponsored by GE Healthcare

Clinical Window Educational Program is sponsored by GE Healthcare

 

 

 

This website is accredited by Health On the Net Foundation. Click to verify.

Clinical Window Web Journal complies with the HON code standard for trustworthy health information: verify here.

 

 

 
 
Cardiovascular

Bedside Arrhythmia Monitoring Quick Guide

The article also available in PDF: 236 KB

Multi-lead arrhythmia monitoring

Arrhythmia means any disturbance or irregularity of cardiac rhythm. Stability of the cardiac rhythm is essential for sufficient pumping function of the heart and adequate cardiac output. Maintaining adequate cardiac output is vital for organ perfusion and survival. Arrhythmia can cause a decrease in cardiac output. Therefore fast and accurate detection of arrhythmia is critical.

Each ECG lead views the heart at a unique angle. Multilead monitoring provides continuous viewing of the hearts rhythm from multiple sites. The more leads are used, the more reliable information is available for the arrhythmia analysis.

A multi-lead arrhythmia algorithm uses several ECG leads for detection and analysis of cardiac arrhythmias. The performance of a multi-lead algorithm is superior to a single-lead algorithm. In noisy situations, there might be noise present on some leads, while the signal in other leads might be good enough for reliable detection of cardiac rhythm. Sometimes the ventricular beat can be detected only on couple of leads e.g. on V3 and V4. All the other leads may have no sign at all about a premature ventricular beat.

The recognition of ventricular beats is improved by multi-lead monitoring. The decision between normal and ventricular beats is more reliable when information from several leads is available. Datex-Ohmeda algorithm uses up to 8 signals for arrhythmia detection. Thus the sensitivity to catch PVC is increased but more importantly the number of false alarms is decreased.

How are arrhythmias monitored at the bedside?

The Datex-Ohmeda bedside arrhythmia algorithm is based on template matching. (A template is a group of beats matching the same morphology.) The algorithm detects QRS complexes, generates QRS templates and performs beat labeling. This algorithm is divided into three parts: detector, classifier and labeling. Parallel to this process there is an algorithm for detection of ventricular fibrillation. Detection of ventricular fibrillation is based on waveform analysis.

The detector algorithm detects waves in the ECG signal that could be QRS complexes. Then the algorithm separates true QRS-complexes from T-waves and artifacts. The detector uses two leads; the first lead is I or II, and the second lead is one precordial lead (V1-V6).

The classifier algorithm forms templates of similar QRS complexes. During the learning phase an initial set of QRS templates is built. When a new true QRS complex is detected, it is compared with the existing templates. If no match is found, a new QRS template is added to the template set. The classifier uses three leads: I, II and one precordial lead (V1-V6).

The labeling algorithm analyses all templates. Each template and the beats belonging to it are labeled with one of the following names: normal beats, ventricular beats and paced beats. The labeling uses all possible independent leads of the ECG measurement. With 5-lead cable that is I, II and V, with 10-lead cable that is I, II, V1, V2, V3, V4, V5 and V6.

When using the 3-lead cable, only one lead is measured at a time. With 3-lead cable there is only one lead available for detector, classifier and labeling algorithms as well as the algorithm for detection of ventricular fibrillation.

Detection of ventricular fibrillation is based on waveform analysis. The main criteria for signaling ventricular fibrillation is: irregular waveform with high enough amplitude and rate. The algorithm for detecting ventricular fibrillation uses two leads; the first lead is I or II, and the second lead is one precordial lead (V1-V6).

The more signals the better performance

The bedside arrhythmia algorithm has been tested as recommended by the Association for the Advancement of Medical Instrumentation (AAMI EC-57). The testing has been done using the MIT-BIH and AHA arrhythmia databases.

Performance measures MIT Database AHA Database
QRS sensitivity 99,87% 99,85%
QRS positive predictive accuracy 99,88% 99,89%
PVC sensitivity 94,09% 94,07%
PVC positive predictive accuracy 97,40% 97,69%
PVC false positive rate 0,234% 0,225%


Practical aspects in bedside arrhythmia monitoring

Signal Quality

Careful skin preparation is very important for good results in arrhythmia monitoring. Good skin preparation ensures good signal. The use of high quality electrodes also improves the signal quality. Good signal ensures accurate arrhythmia detection and especially decreases the number of false alarms.

Relearning

When the morphology of the patient’s ECG changes considerably (e.g. due to removal of a pacemaker), relearning should be started manually. This can be done in the ECG menu by selecting Relearn – Start.

Selecting leads for the arrhythmia analysis

The selection of user leads (ECG1, ECG2 and ECG3) on the monitor affects the leads used for bedside arrhythmia analysis. The first lead used for arrhythmia analysis is either lead I or lead II. The algorithm uses the lead appearing first in the user leads. The second lead used for arrhythmia analysis is one of the precordial leads (V1-V6). The algorithm uses the precordial lead appearing first in the user leads. NOTE: after changing the leads, Relearning has to be started manually for changes to take an effect.

Bedside arrhythmia alarm definitions

Alarm Definition
Asystole Cardiac arrest, no QRS complexes for five seconds.
Bradycardia HR below the lower HR alarm limit.
Tachycardia HR over the higher HR alarm limit.
Ventricular Fibrillation Fibrillatory waveform caused by ventricular fibrillation.
Ventricular Tachycardia Five or more consecutive PVCs and rate of successive beats over 100 bpm
Ventricular Run Three or more consecutive PVCs and rate of successive beats over 100 bpm
Ventricular Couplet Two consecutive ventricular beats, preceded and followed by a normal beat.
Ventricular Bigeminy The following pattern is detected: N, V, N, V, N, V where N = normal, V = PVC. In other words, every other beat is a PVC.
Ventricular Trigeminy The following pattern is detected: N, N, V, N, N, V, N, N, V where N = normal, V = PVC. In other words, every third beat is a PVC.
Multifocal PVC's Over the last 15 beats two or more premature ventricular beats with different morphologies are detected.
R on T PVC's Early PVC, beat detected as a PVC and previous beat as not PVC; current RR interval is less than half of the previous RR interval.
Missing Beat Actual RR interval more than 1.8 times the average RR interval.
Frequent PVC's PVCs per minute above the alarm limit.

Graded arrhythmia alarms

Alarms are graded into three levels – red, yellow and white – to identify the priority of each arrhythmia alarm. The arrhythmia alarm profile can be changed for each individual patient. Preconfigured alarm profiles can also be saved to the modes for different patient groups. The most severe alarms, asystole and ventricular fibrillation, are fixed as red alarms.


Last updated: 1 December 2002Created
Legal notice © GE Healthcare 2008
ISSN 1795-6269 (Web)
ISSN 1795-6277 (CD)		 Webmaster