Journal of Electrocardiology 39 (2006) 113 119

www.elsevier.com/locate/jelectrocard

Virtual tools for teaching electrocardiographic rhythm analysis

John Michael Criley, MD, MACP, FAHA, FACCa,b,*, William P. Nelson, MD, FACP, FAHA, FACCc
b a Geffen School of Medicine at UCLA, Los Angeles, CA, USA The Los Angeles Biomedical Research Institute, St Johns Cardiovascular Research Center at Harbor-UCLA Medical Center, Torrance, CA 90502, USA c Saint Joseph Hospital, Denver, CO, USA

Received 3 May 2005

Abstract

Electrocardiographic (ECG) rhythm analysis is inadequately taught in training programs, resulting in undertrained physicians reading a large percentage of the 40 million ECGs recorded annually. The effective use of simple tools (calipers, ruler, and magnifier) required for crucial measurements and comparisons of intervals requires considerable time for interactive instruction and is difficult to teach in the classroom. The ECGviewer (Blaufuss Medical Multimedia Laboratories, Palo Alto, Calif) program was developed using virtual tools easily manipulated by computer mouse that can be used to analyze archived scanned ECGs on computer screens and classroom projection. Trainees manipulate the on-screen tools from their seats by wireless mouse while the instructor makes corrections with a second mouse, in clear view of the trainees. An on-screen ladder diagram may be constructed by the trainee and critiqued by the instructor. The ECGviewer program has been successfully used and well received by trainees at medical school, residence, and subspecialty fellow level. 2006 Elsevier Inc. All rights reserved. ECG; ECGViewer; Ladder diagram The tools required for performance of the mechanical aspects of rhythm analysis are simple and inexpensive: ruler, calipers, and magnifying glass. However, the training required for proper use of these tools requires considerable time commitment for both instructor and trainee. The instructor should demonstrate the proper manipulation of

Keywords: 1. Introduction

Electrocardiography (ECG) is the most widely used cardiac diagnostic laboratory procedure and an indispensable tool for detection of arrhythmias. Approximately 40 million ECGs are recorded annually in the United States, which, by law, must be interpreted by physicians. Most ECGs are read by noncardiologists (Table 1). There is concern regarding the accuracy of these readings; unwarranted reliance on computer analysis and lack of adequate training in ECG interpretation can be cited as principal reasons [1-3]. In contemporary training programs, the need to establish competence in interpretation of an expanding plethora of diagnostic imaging technologies has greatly diminished the time formerly devoted to teaching ECG interpretation. _______
* Corresponding author. The Los Angeles Biomedical Research Institute, St Johns Cardiovascular Research Center at Harbor-UCLA Medical Center, Torrance, CA 90502, USA. Tel.: +1 310 222 2532; fax: +1 310 787 0448. 0022-0736/$ see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jelectrocard.2005.07.002

Table 1 Number of ECGs reviewed by specialty Physicians

ECGs

Exposure (ECGs/y)

Family practice 85006 7146324 93 Internal medicine 128709 8461971 66 363 Emergency medicine 22025 8000000a Cardiovascular medicine 20205 14391705 712 This was adapted from the US Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center for Health Statistics, 2001 data. a Estimate, based on number of emergency department visits per year for chest pain.

114

J.M. Criley, W.P. Nelson / Journal of Electrocardiology 39 (2006) 113 119

Fig. 1. The Heart Station ECG reading and teaching environment. A 12-lead ECG is shown with common tools: calipers, a magnifying glass, and a 6-second (150 mm) ruler.

calipers by making only those pertinent measurements or comparisons of measurements that will incrementally lead to the eventual diagnosis. The student cannot readily see the instructors moves, and the instructor in turn does not have an

abundance of time required to observe the fledglings manipulations and measurements. These limitations are exponentially magnified when 1 instructor is responsible for multiple trainees.

Fig. 2. ECGviewer image with 6-second ruler. The rhythm is irregular, and there are approximately 6.1 R-R intervals so the rate is about 61 beats per minute. The toolbox in the upper left corner contains icons representing the other tools while the ruler icon is highlighted. This illustration as well as subsequent ones were prepared directly from screen captures.

J.M. Criley, W.P. Nelson / Journal of Electrocardiology 39 (2006) 113 119

115

Fig. 3. Magnification mode. The field width of the ECG display is 5 seconds. Vertical arrows representing P waves were added to the screen capture.

In many training programs, classroom instruction is performed by projection of ECG transparencies or 35-mm slides on a screen while measurements made with oversized calipers are demonstrated by the instructor and

emulated by students. This procedure is far from ideal because the person manipulating the calipers blocks the audiences line of vision and casts shadows that obscure other portions of the tracing from himself and the

Fig. 4. Measurement of P-P intervals. A multiple exposure image prepared from 5-screen captures reveals that the calipers document constant P-P intervals of 1490 milliseconds (atrial rate, 40 beats per minute). Atrioventricular dissociation is seen, with a ventricular rate of 61 and intermittent AV conduction (beats 3, 6, and 9) when the R-P interval is sufficiently long to overcome the refractory period.

116

J.M. Criley, W.P. Nelson / Journal of Electrocardiology 39 (2006) 113 119

audience. Nevertheless, for lack of better technology, this method is still widely used, not only in medical schools and teaching hospitals, but also in workshops at professional medical meetings.

2. Materials and methods Our goal was to try to overcome these limitations through the use of computer graphic technology to demonstrate conveniently and efficiently the proper use of simple tools required for arrhythmia analysis. We endeavored to reproduce the familiar Heart Station ECG reading environment (Fig. 1) by developing a proprietary computer program, ECGviewer (Blaufuss Medical Multimedia Laboratories, Palo Alto, Calif), to display and permit analysis of electronically scanned ECGs (Figs. 2-6). The program was designed to be widely accessible, with very modest system requirements: XGA screen resolution (1024 768 pixels), 300-MHz processor, and 64-MB memory. As a Web-based program, it is designed to match most computer displays connected to the Internet. It will run on Windows 98, 2000,

and XP (Microsoft Corporation, Redmond, Wash) as well as on Mac OS (Apple Computer, Inc, Cupertino, Calif) computers, or as a Web-based program using Shockwave (Macromedia, Inc, San Francisco, Calif) player. In addition to the calipers, a 2_ magnifying glass and a 6-second (150 mm) ruler may be deployed from the toolbox, and a vertical plumb line can be used as needed to relate events on concurrent leads. A ladder diagram template with horizontal columns for atrial, atrioventricular (AV) nodal, and ventricular depolarization can be constructed. The arrows in these columns may be displayed as normal (arrowhead down) or retrograde (arrowhead up), or designated as ectopic in origin. The ECGviewer program was designed to be electronically projected on a large screen for classroom instruction but can also be clearly seen on a personal computer screen. Manipulation of the virtual tools is performed with a mouse. In our conferences, 1 cordless optical mouse is passed from trainee to trainee, so their performance can be supervised and peer-reviewed while the instructor may override the students manipulations with a second mouse. Moreover, the entire audience has a

Fig. 5. Ladder diagram. A, A basic ladder diagram has been created on the ECGviewer, depicting conducted sinus beats from the second, fourth, and sixth P waves. The plumb line is seen aligned with the last R wave. B, A more detailed ladder diagram depicts junctional origin of the R waves that are not conducted from the sinus P waves (marked by asterisks) and concealed conduction entering the AV node and blocking or delaying the impulses emanating from the sinus P waves. The R-wave arrows can also be diagrammed as ventricular or as fusion beats by repeatedly clicking the mouse in the horizontal column below the V column.

J.M. Criley, W.P. Nelson / Journal of Electrocardiology 39 (2006) 113119

117

Fig. 5 (continued).

clear, unobstructed view of the high resolution image on the screen because the operators stay in their seats while remotely controlling the virtual tools on the screen. The instructor or trainee can activate the transparent 6second ruler, placing it across the rhythm strip so that the zero is centered over a QRS complex. Counting the number of intervals (and fractions thereof) in 6 seconds and multiplying by 10 yields the beats per minute when the rhythm is irregular (Fig. 2). If AV dissociation is present, the atrial rate may be similarly calculated. A 6-second measurement to determine rate is clearly superior in precision to the counting little boxes or measure 2 R-R with a rate ruler techniques when the rhythm is irregular or the rate is rapid. The magnifying glass (Fig. 3) increases the image size 2fold and allows the ECG to be moved relative the grid. By moving the ECG tracing independent of the grid, measurements of duration and magnitude can be readily made by alignment of a bold (5 mm) line at the onset of the signal to be measured. The virtual calipers (Fig. 4) are deployed by clicking on

the icon in the toolbox, moved by click-and-drag, and spread by placement of the cursor over either strut while click-and-dragging laterally. Our requirements called for virtual calipers that were precise, smoothly moved, and easily spread, with fixation of 1 point while the other is moved. The calipers have a span of 0 to 4 seconds in 5millisecond gradations. The intervals represented by the span are displayed in milliseconds and can be interconverted to heart rate in beats per minute by clicking on the numerical display. Instructors and trainees rapidly master these manipulations on their first exposure. The plumb line is stored on the right edge of the tracing and can be moved by click-and-drag to align events such as P waves that are more readily seen in some leads with a rhythm strip in which those events may be obscure. The ladder diagram (Fig. 5) is constructed by clicking on the ladder icon in the toolbox, moving the template vertically to a convenient location, clicking to mark the P and QRS complexes in the appropriate columns, and clickand-dragging anterograde (Fig. 5A) and/or retrograde

118

J.M. Criley, W.P. Nelson / Journal of Electrocardiology 39 (2006) 113 119

Fig. 6. Arrhythmia Conference at Harbor-UCLA Medical Center. A ceiling-mounted digital projector displays the ECG on a large screen as a resident (right foreground) controls the mouse and ladder diagram with the wireless mouse in her left hand. The instructor (left background) selects the tracings from a laptop computer and can override the residents manipulations of the tools when needed with another mouse. The calipers are being used to fine a hidden P wave in a tracing that displays electronic pacemaker rhythm with AV dissociation and fusion.

(Fig. 5B) conduction through the AV node. Each of the tools may be toggled on and off the ECG display by clicking on the appropriate icon in the toolbox. Electrocardiograms are placed in the ECGviewer program by electronic scanning of original paper copies at 200 dpi, and the grid is removed by color subtraction. This resolution was selected to optimize the quality of the reproduction for computer display without consuming excessive storage capacity. Scanning is obviated when access to proprietary digital ECG records is available. Removal of the grid reduces the storage requirements of the individual ECGs and provides a cleaner recording. A standard grid is replaced on the ECGviewer display as shown in Fig. 2. In the magnified mode, the grid is moveable so that it may be used to measure duration, magnitude, and displacements of the waveforms. Our archives contain more than 1000 scanned and formatted ECGs that may be selected from a numerically coded menu. The ECGviewer program and more than 3000 12-lead ECGs can be stored on a single CD-ROM.

An electrophysiology version of the ECGviewer has also been developed for advanced training. This version emulates the features available in the electrophysiology laboratory, with 200-mm/s recordings, H-Bar calipers, alphanumeric readouts of intervals, and a custom ladder diagram feature.

3. Results The ECGviewer has been successfully used in weekly ECG Arrhythmia Conferences (Fig. 6) at Harbor University of California, Los Angeles (UCLA), Medical Center for more than a year and beta tested in a number of other institutions as well (see Acknowledgments). Trainees (as well as computer-impaired instructors) quickly master the use of the tools with manipulation of the cordless mouse. In a survey completed anonymously by trainees exposed to the program, 20 of 21 strongly agreed that the ECGviewer was a very useful teaching tool, superior to

J.M. Criley, W.P. Nelson / Journal of Electrocardiology 39 (2006) 113119

119

anything else I used, and very easy to use. The respondents included fourth-year medical students, internal medicine and family medicine residents, and cardiology fellows.

4. Discussion Advances in cardiac electrophysiology provide greater precision in characterization of arrhythmias and effectiveness of management, but recruitment and appropriate referral of worthy patients require detection by less specialized physicians responsible for the interpretation of the bulk of ECGs recorded in offices, clinics, and hospitals. Although there are many textbooks devoted to electrocardiography containing excellent text and graphics, they lack the critical interaction that is imparted by handson instruction. Existing computer-based ECG tutorial programs cannot display ECGs as functionally as a paper copy, making the use of tools such as calipers and rulers cumbersome. Our goal was to try to overcome the limitations of computer graphic displays of ECGs by providing interactivity with virtual tools that can be manipulated in a manner similar to the manual tools used to interpret hard copies. Critical interactivity can be provided by an instructor in a oneon- one situation or in a classroom by instructors and peers. We are greatly encouraged by the reception of the program by our trainees and instructors in our program as well as several beta-testing centers and are sufficiently encouraged to proceed with the development of stand-alone tutorial and testing programs. Tutorial programs will be designed to track trainees caliper moves and provide feedback regarding their appropriateness. For example, in an ECG demonstrating high-degree AV block, it would be more appropriate to first measure the R-R intervals (to confirm regularity) rather than the P-P intervals (which may or not vary and are inconsequential). We also believe that this program format lends itself well to testing, both during training and for board

certification. The computer could track the number of, as well as the appropriateness of, the caliper manipulations. Although our initial program was specifically designed for arrhythmia analysis, many of the archived examples exhibit important pathology (myocardial ischemia or infarction, chamber hypertrophy or enlargement, electrolyte abnormalities, and others).

Acknowledgments The ECGviewer was designed and programmed by Blaufuss Medical Multimedia Laboratories, with support from National Heart, Lung, and Blood Institute (Bethesda, MD) SBIR (Small Business Innovation Research) 1R43HL073538-01. Electrocardiograms were selected by the first author, and the diagnoses were vetted by the second author. The scans were performed by Daniel Beckmann in the facilities of the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center. The following investigators tested the ECGviewer in their teaching programs and provided valuable feedback: Melvin Scheinman, MD, University of California, San Francisco; Victor Froelicher, MD, Veterans Affairs Palo Alto Health Care System (Palo Alto, CA) and Stanford University, Stanford, Calif; Carole Warde, MD, and Alan J. Hermer, MD, Memorial Medical Center of Long Beach (Long Beach, CA) Hospital and University of California, Irvine; Laura Wexler, MD, University of Cincinnati, Ohio; and Michael M. Laks, Harbor-UCLA Medical Center.

References
[1] Kadish AH, Buxton AE, Mason JW, et al. ACC/AHA clinical competence statement on electrocardiography and ambulatory electrocardiography. J Am Coll Cardiol 2000;38(7):2091. Hurst JW. Current status of clinical electrocardiography with suggestions for the improvement of the interpretive process. Am J Cardiol 2003;92(9):1072. Salerno SM, Alguire PC, Waxman HS. Competency in interpretation of 12-lead electrocardiograms: a summary and appraisal of published evidence. Ann Intern Med 2003;138:751. J.M. Criley, W.P. Nelson / Journal of Electrocardiology 39 (2006) 113119 119

[2]

[3]