corpuls.science ECGmax

SPECIAL EDITION | ECGmax 3 4 7 8 9 10 13 14 19 20 21 27 28 32 EDITORIAL THE ACS ALGORITHM ECG FILTERING HOW FILTERING INFLUENCES THE ECG MORPHOLOGY HOW THE ECG SIGNAL GETS FROM THE PATIENT TO THE MONITOR corpuls.mission 22-LEAD ECG DERIVATION OF THE 12-LEAD ELECTROCARDIOGRAM AND 3-LEAD VECTORCARDIOGRAM MATHEMATICAL MODELING AND UTILITY OF THE DERIVED 22-LEAD ELECTROCARDIOGRAM VECTOR LOOPS APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE THE CARDIAC ELECTRICAL BIOMARKER – CEB® DETECTION OF ACUTE MYOCARDIAL ISCHEMIC INJURY BY GENDER USING A NOVEL CARDIAC ELECTRICAL BIOMARKER AUTOMATED ANALYSIS OF THE 12-LEAD ECG IN THE EMERGENCY DEPARTMENT: ASSOCIATION BETWEEN HIGH-SENSITIVITY CARDIAC TROPONIN I AND THE CARDIAC ELECTRICAL BIOMARKER Imprint corpuls.science – Special Edition | ECGmax Issue 2 | 2020 Article Number: 76139.65021 | Vers. 1.0 (07/20) Publisher/Editorial department/Manufacturer: corpuls | GS Elektromedizinische Geräte G. Stemple GmbH Hauswiesenstrasse 26 | 86916 Kaufering | Germany Concept & Layout: Photo: Reading Service: corpuls Team corpuls Team marketing@corpuls.com The rights of named brands and trademarks remain with their respective owner.

EDITORIAL 3 Dr. Christian Klimmer, CEO Michael Heller, Director DEAR LADIES AND GENTLEMEN In the second corpuls.science we have chosen one of our most core technologies. This issue will focus on electrocardiograms. It has become the standard in Emergency Medicine to just attached a bunch of electrodes to the patient for monitoring reasons. For at least 25 years, whenever a patient presents with symptoms possibly indicating that something is not quite right with the heart, the patient receives a diagnostic ECG. Nowadays the 12-lead ECG is part of every guideline and recommendation to do with Acute Coronary Syndromes and is usually referred to as the “gold-standard”. In 1992 corpuls pioneered 12-lead ECG in Emergency Medical Services with the corpuls 08/16. The 12-lead ECG method has not changed much during the past few decades and it leaves some vectors of the heart in the dark. There have recently been some ideas on how to overcome this, most of which were never popularised, possibly due to complexity or simply just the time which is required to gather this additional information. In EMS the recommendations regarding whether, and especially where, to trans- port the patient is decided predominately by the patient’s symptoms. In the various different hospital environments patients with symptoms receive several troponin tests. Whilst troponin tests have become available for out of hospital use, it never really broke through. Again, most likely due to time and complexity reasons, but also presumably due to the high costs of disposables. corpuls now has a solution to all of these problems with none of the disadvantages. The ECGmax technology which corpuls has just released, gives you all the additional ECG-leads and more. You can now get up to 22-leads without the need for any additional electrodes or any change in electrode placement. This includes the posterior leads V7-V9, the right cardiac leads V3r to V6r and the orthogonal leads X, Y and Z with the vector loops. This is all carried out by an evidence based algorithm which runs on the corpuls.mission server and can be sent from there to any device on scene, whether this is be a phone, tablet or electronic patient care record device. As this is part of the corpuls telemedicine solution, the diagnostic ECG is also available in real time where it can be measured and forwarded to a receiving hospital or a specialist for consultation. On top of the additional ECG-leads, the ECGmax technology includes something which might just be the next big thing when it comes to diagnosing Acute Coronary Syndromes. The Cardiac Electrical Biomarker (CEB®) promises to give medical practitioners a real time value, comparable to troponin, without requiring any additional interventions and it is as easy to interpret as a traffic light. According to studies the CEB® has a similar sensitivity and specificity to troponin tests. Everything you need to know about this, and a lot more, is what this issue of corpuls.science is all about. Thank you very much and enjoy the articles. Dr. Christian Klimmer CEO Michael Heller Director Medical Research and Application JULY | 2020

4 THE ACS ALGORITHM SPECIAL EDITION | ECGmax THE ACS ALGORITHM The classic 12-lead ECG represents a snapshot of the elec- trical activity of the heart. With this snapshot it is possible to diagnose various rhythmological and structural diseases of the heart. Due to this fact, the 12-lead ECG is the most important diagnostic tool for myocardial ischemia and in- farction and is the basis for further therapeutic and diag- nostic measures (Wagner et al. 2009). Myocardial infarction (MI) leads to muscular injury (scarring) and consequently to a change in the conduction properties in the heart. In the ECG, it is often associated with a T-wave change, a so-called ST segment elevation and/or depression, changes in the QRS complex or an inverted T-wave. The ST segment changes result from the voltage differences at the borders between ischemic and non-ischemic tissue. Due to the bioelectrical principles of the ECG, an ST segment elevation in one lead usually leads to a reciprocal ST segment depression in the lead with opposite polarity. Consequently, an ST segment elevation in the posterior leads V8 and V9 can be seen as an ST segment depression in the reciprocal leads V1 and V2 (Wagner et al. 2009). As the posterior leads are not part of the standard 12-lead ECG, a distinction is normally made between two classes of ACS (acute coronary syndrome), STEMI (ST-elevation myo- cardial infarction) and NSTEMI (non-ST-elevation myocardial infarction). The STEMI typically has an ST segment eleva- tion at the J point, while the NSTEMI shows an ST segment depression in the corresponding leads. (Wagner et al. 2009) In a STEMI (patients with acute chest pain and ST segment elevation) there is an acute and total occlusion of a coronary artery. These patients must immediately receive reperfusion therapy or fibrinolytic therapy (Steg et al. 2012). According- ly, patients presenting with acute chest pain without persis- tent ST segment elevation have NSTEMI. This is a subacute course and must first be confirmed by laboratory diagnostics to then move on to further therapy. In order to make a well-founded therapy decision, the ECG is the most important option for a quick and safe diagnosis, as well as adequate risk assessment. Myocardial ischemia remains the leading cause of death worldwide. Even though this trend has decreased in Europe over the past three decades, heart attacks still caused 1.8 million deaths in 2018. This represents 20% of all deaths (Ibanez et al. 2018). In Germany alone, 60,000 people die of an acute heart attack each year. For this reason, the guide- lines recommend an ECG is preformed and interpreted no later than 10 minutes after first medical contact (Kelm et al. 2018; Ibanez et al. 2018). JULY | 2020 To be able to offer the patient the fastest and best possi- ble therapy, the corpuls3 contains a computer-aided ECG analysis. With the corpuls ACS algorithm, the ECG can be analysed according to the STEMI or NSTEMI criteria (see Figure 1). corpuls has based this on the current European Society of Cardiology guidelines. Fundamentally, the change in the ST segment is measured from the J point. The following criteria are considered to indicate sustained acute occlusion of a coronary artery (STEMI) (Thygesen et al. 2012). in men < 40 years in men > 40 years • Two adjacent leads with ST elevation of 0.25 mV • Two adjacent leads with ST elevation of 0.2 mV • Or two adjacent leads with ST elevation of 0.15 mV • And/ or ST elevation of 0.1 mV in all other leads • In the absence of LV hypertrophy or LBBB in women in leads V2-V3 According to the guidelines, an NSTEMI is characterized by an ST depression of 0.05 mV in two adjacent leads. Howev- er, as a depression of 0.05 mV (0.5 mm) is usually difficult to diagnose, a segment depression of more than 0.1 mV in a single lead, or even more relevantly by 0.2 mV, is also a clear characteristic (Thygesen et al. 2012; Achenbach et al. 2012; Deutsche Gesellschaft für Kardiologie – Herz- und Kreislauf- forschung e.V. 2011, 2012). For the corpuls ACS algorithm, a segment depression of 0.15 mV was chosen as the NSTEMI limit value. This value should avoid too many false positives yet still recognize an actual NSTEMI. Furthermore, the T-wave inversion above 0.1 mV is a clear characteristic for a NSTEMI (Thygesen et al. 2012; Deutsche Gesellschaft für Kardiologie – Herz- und Kreislaufforschung e.V. 2012). The corpuls ACS algorithm also checks this char- acteristic, as shown on the flowchart (Figure 1). Nevertheless, NSTEMI is a disease that is very difficult to de- tect in a 12-lead ECG. Over half of all NSTEMI display either a normal or a diagnostically unusable ECG (pacemaker, bundle branch block). ST segment depressions only occur in approx. 20% of patients, whereas a T-wave inversion can be found in around 25% of all NSTEMI patients (Arastéh et al. 2012). Accordingly, the guidelines recommend that a measurement of the posterior and the right-precordial leads should also be integrated into everyday clinical practice as routine (Kelm et al. 2018).

THE ACS ALGORITHM 5 FLOWCHART CORPULS ACS ALGORITHM Version 1.7 START corpuls ACS Algorithm ECG measurement not reliable Therapy Suggestion „No advice (invalid ECG)“ Male Patient Female Patient Patient’s age or gender not submitted ≥ 40 years < 40 years S T E M I C H E C K STEMI-Criteria STEMI-Criteria STEMI-Criteria STEMI-Criteria ST-elevation > 200µV in V2/V3; ST-elevation > 250µV in V2/V3; ST-elevation > 150µV in V2/V3; ST-elevation > 150µV in V2/V3; ST-elevation > 100µV in other leads ST-elevation > 100µV in other leads ST-elevation > 100µV in other leads ST-elevation > 100µV in other leads NO YES YES NO YES NO NO YES YES Abnormal ECG? YES NO Abnormal ECG? YES NO NO Therapy Suggestion “ECG meets specific STEMI criteria, consider abnormal ECG” Therapy Suggestion “ECG meets specific STEMI criteria” Therapy Suggestion “No age and/or gender available. ECG shows acute ischemic changes, STEMI possible, consider abnormal ECG” Therapy Suggestion “No age and/or gender available. ECG shows acute ischemic changes, ECG meets STEMI criteria” ALARM STEMI possible ALARM STEMI ALARM STEMI possible ALARM STEMI NSTEMI check activated at NO corpuls ACS settings? YES Abnormal ECG? YES NO Therapy Suggestion “No advice, ECG doesn’t meet specific STEMI criteria, consider abnormal ECG” Therapy Suggestion “No advice, ECG doesn’t meet specific STEMI criteria” ALARM Abnormal ECG Figure 1: Flowchart corpuls ACS Algorithm JULY | 2020

K C E H C I M E T S N 6 THE ACS ALGORITHM SPECIAL EDITION | ECGmax NSTEMI-Criteria 1/2 ST-depression > 150µV NO Abnormal ECG? YES YES NO Abnormal ECG? YES NO NSTEMI-Critera 2/2 Inverted T-wave > 100µV NO in Wilson Leads Therapy Suggestion “ECG shows acute ischemic changes, ECG meets NSTEMI criteria, consider abnormal ECG” Therapy Suggestion “ECG shows acute ischemic changes, ECG meets NSTEMI criteria” ALARM NSTEMI possible ALARM NSTEMI Therapy Suggestion “No advice, ECG doesn’t meet specific STEMI criteria” Therapy Suggestion “No advice, ECG doesn’t meet specific STEMI or NSTEMI criteria, consider abnormal ECG” ALARM Abnormal ECG NOTIFICATION Consider that each STEMI- / NSTEMI-Criterion has to be met by at least two leads of the same lead group LEAD GROUPS • V1, V2, V3, V4, V5, V6 • • II, III, aVF I, aVL VARIATIONS OF „ABNORMAL ECG“ • Wolff-Parkinson-White Syndrom • Left Bundle Branch Block • Right Bundle Branch Block • QRS-Duration >120ms • Pacemaker Rhythm* * during Pacemaker Rhythm no reliable measurement possible! References Achenbach, S.; Szardien, S.; Zeymer, U.; Gielen, S.; Hamm, C. W. (2012): Kommentar zu den Leitlinien der Europäischen Gesellschaft für Kardiologie (ESC) zur Diagnostik und Therapie des akuten Koronarsyndroms ohne persistierende ST-Streckenhebung. In: Kardiologe 6 (4), S. 283–301. DOI: 10.1007/s12181-012- 0436-5. Arastéh, K.; Baenkler, H.-W.; Bieber, C.; Brandt, R.; Chatterjee, T. T.; Dill, T. et al. (2010): Innere Medizin. Thieme, 3. Auflage Deutsche Gesellschaft für Kardiologie – Herz- und Kreislaufforschung e.V. (2011): ESC POCKET GUIDELINES 2011. Akutes Koronarsyndrom ohne ST-Hebung (NSTE-ACS). In: ESC POCKET GUIDELINES, S. 1–56. Deutsche Gesellschaft für Kardiologie – Herz- und Kreislaufforschung e.V. (2012): ESC POCKET GUIDELINES 2012. 3. Allgemeine Definition des Myokardinfarktes. In: ESC POCKET GUIDELINES, S. 1–24. Ibanez, B.; James, S.; Agewall, S.; Antunes, M. J.; Bucciarelli-Ducci, C.; Bueno, H. et al. (2018): 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST segment elevation. The Task Force for the management of acute myocardial infarction in patients presenting with ST segment elevation of the European Society of Cardiology (ESC). In: European heart journal 39 (2), S. 119–177. DOI: 10.1093/eurheartj/ehx393 Kelm, M.; Kastrati, A.; Nef, H.; Richardt, G.; Zeymer, U.; Bauersachs, J.; Kommission für Klinische Kardiovaskuläre Medizin (2018): Kommentar zu den Leitlinien 2017 der Europäischen Gesellschaft für Kardiologie (ESC) zur Therapie des akuten Herzinfarktes bei Patienten mit ST-Streckenhebung. In: Der Kardiologe 12 (2), S. 145–149. DOI: 10.1007/s12181-018-0237-6. Steg, Ph G.; James, S. K.; Atar, D.; Badano, L. P.; Blömstrom-Lundqvist, C.; Borger, M. A. et al. (2012): ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST segment elevation. In: European heart journal 33 (20), S. 2569–2619. DOI: 10.1093/eurheartj/ehs215. Thygesen, K.; Alpert, J. S.; Jaffe, A. S.; Simoons, M. L.; Chaitman, B. R.; White, H. D. et al. (2012): Third universal definition of myocardial infarction. In: Circu- lation 126 (16), S. 2020–2035. DOI: 10.1161/CIR.0b013e31826e1058. Wagner, G. S.; Macfarlane, P. W.; Wellens, H.; Josephson, M.; Gorgels, A. P. M.; Mirvis, D. M. et al. (2009): AHA/ACCF/HRS Recommendations for the Standardization and Interpretation of the Electrocardiogram. Part VI: Acute Ischemia/Infarction A Scientific Statement From the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; JULY | 2020

ECG FILTERING With the increase in diseases of the cardiovascular system, the electrocardiogram (ECG) is an essential tool for preclini- cal diagnostics in modern emergency medicine. It is recom- mended to continuously monitor patients with such diseases so to quickly detect life-threatening arrhythmias. [1] In the preclinical setting, the recorded ECG may show artifacts due to various environmental factors. These can affect the in- terpretation of the ECG signal. [5] Artifacts can be caused by the patient’s muscle activity, transport movements or electrical coupling of the mains frequency, for example. To reduce the quantity of different artifacts in the ECG recording, corpuls devices offer the possibility to use different ECG filters. The human ECG spectrum covers the frequency range 0.05 Hz - 100 Hz, the main part of the QRS complex being in the range 0.5 Hz - 45 Hz. [9] Most of the time the ECG signal is overlaid by the mains frequency of 50 Hz/60 Hz. The mains frequency portion of an ECG can be reduced with a notch filter that can selectively filter individual frequencies. Low-pass filters are used to reduce high-frequency interference such as muscle tremors. These artifacts lie within the range from 5 Hz to 450 Hz. [10] High-pass filters in turn dim low-frequency interference signals that arise, for example, from breathing movements. These are between 0.05 Hz and 1 Hz. [11] This means that high and low frequency interference overlaps the frequency range of the ECG signal. Due to these technical rea- sons, filtering not only reduces interference, but also changes the ECG morphology. [3] The combination of low- and high- pass filters is known as a bandpass filter. References [1] Ibanez B, James S, Agewall S, et al. 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST segment elevation: The Task Force for the management of acute myocardial infarction in patients presenting with ST segment elevation of the European Society of Cardiology (ESC). Eur Heart J. 2018;39(2):119-177. doi:10.1093/eurheartj/ehx393 [2] Morley AP. Prehospital monitoring of trauma patients: experience of a helicopter emergency medical service. Br J Anaesth. 1996;76(5): 726-730. doi:10.1093/bja/76.5.726 [3] Satija U, Ramkumar B, Manikandan MS. A Review of Signal Processing Techniques for Electrocardiogram Signal Quality Assessment. IEEE Rev Biomed Eng. 2018;11:36-52. doi:10.1109/ RBME.2018.2810957 [4] Thakor NV. From Holter monitors to automatic defibrillators: developments in ambulatory arrhythmia monitoring. IEEE Trans Biomed Eng. 1984;31(12):770-778. doi:10.1109/ TBME.1984.325237 [5] Levkov C, Mihov G, Ivanov R, Daskalov I, Christov I, Dotsinsky I. Removal of power-line interference from the ECG: a review of the subtraction procedure. Biomed Eng Online. 2005;4:50. Published 2005 Aug 23. doi:10.1186/1475-925X-4-50 [6] T. M. Chieng, Y. W. Hau and Z. Omar, “The Study and Comparison Malaysia, 2018, pp. 226-232, doi: 10.1109/IECBES.2018.8626661. between Various Digital Filters for ECG Denoising,” 2018 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES), Sarawak, ECG FILTERING 7 The principle of ECG filtering is shown in Figure 1. First, the mains frequency in the ECG is reduced with a notch filter. The notch filter must be set to the local mains frequency. The intended use plays a decisive role in the selection of the bandpass filter. For diagnostic interpretation, the ECG signal must not be affected by the filtering. Therefore, it is rec- ommended for diagnostic purposes to use a high-pass filter with 0.05 Hz and a low-pass filter with 150 Hz. [4] With this bandpass filter setting, the ECG is displayed with the maxi- mum available frequency bandwidth. If the ECG is displayed with a reduced frequency bandwidth, the ECG morphology will be changed. Adjustment to the low-pass filter affects the amplitude of the R wave. Adjusting the high-pass filter can result in shifts in the ST segment. [3] If the bandpass filter impairs the interpretation ability of the ECG signal, it will be shown on the ECG printout. The aim of ECG monitoring is the detection of life-threat- ening arrhythmias such as ventricular fibrillation, ventricular tachycardia or asystole. [1] The detection of such rhythms can be impaired by various interferences during patient transport. [7] If rhythm detection is no longer possible, the bandpass filter with the smallest frequency bandwidth should be used. With this filter setting, the maximum pos- sible number of interference frequencies is filtered from the ECG signal. The ECG is shown with less noise but can only be used for rhythm analysis. [8] [7] U. Satija, B. Ramkumar and M. S. Manikandan, “A Review of Signal Processing Techniques for Electrocardiogram Signal Quality Assessment,” in IEEE Reviews in Biomedical Engineering, vol. 11, pp. 36-52, 2018, doi: 10.1109/RBME.2018.2810957. [8] Buendía-Fuentes F, Arnau-Vives MA, Arnau-Vives A, et al. High-Bandpass Filters in Electrocardiography: Source of Error in the Interpretation of the ST Segment. ISRN Cardiol. 2012;2012:706217. doi:10.5402/2012/706217 [9] Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram: part I: the electrocardiogram and its technology a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2007;49(10): 1109-1127. doi:10.1016/j.jacc.2007.01.024 [10] Yun JG, Jeung KW, Lee BK, et al. Performance of an automated external defibrillator in a moving ambulance vehicle. Resuscitation. 2010;81(4):457-462. doi:10.1016/j.resuscitation.2009.12.031 [11] N. V. Thakor and Y. -. Zhu, “Applications of adaptive filtering to ECG analysis: noise cancellation and arrhythmia detection,” in IEEE Transactions on Biomedical Engineering, vol. 38, no. 8, pp. 785-794, Aug. 1991, doi: 10.1109/10.83591. JULY | 2020

HOW THE ECG SIGNAL GETS FROM THE PATIENT TO THE MONITOR 9 HOW THE ECG SIGNAL GETS FROM THE PATIENT TO THE MONITOR A change in electrical potential arising from the heart’s myocardial tissue can be detected on the skin Electrodes link the human body to the corpuls3 TOUCH / corpuls3 CLASSIC • Calculation of augmented ECG leads • Digital filtering (notch filter, high-pass and low-pass filter) • General signal processing (VF Detection, QRS Detection, Heart Rate calculation) Analog amplification Analog filtering Detection of pacemaker spikes Analog to digital conversion • • • • Transmission via infrared or wireless to corpuls3 TOUCH / corpuls3 CLASSIC monitor/printer/cf-card Transmission to corpuls.mission via WLAN or GSM * IEC Color coding (left) AHA Color coding (right) JULY | 2020

10 corpuls.mission SPECIAL EDITION | ECGmax corpuls.mission is a medical communication platform designed with the patient as the main focus. It enables knowledge to be collected where it can be best utilised: during a mission. Above anything else, adequate patient treatment requires specific knowledge, often stemming from different disci- plines and selected experts. However, this complexity of medicine cannot rest on the shoulders of a single specialist. By combining medical data, chat, video and documentation, all relevant information is found in one place – one software, intuitive and central. Available for web, iOS and Android. Designed for the unique demands of preclinical missions, corpuls.mission stands out from previous classic commu- nication channels. It offers the possibility to view diagnostic ECGs and photos of medication plans or the incident scene in addition to the live curves and vital parameters of the corpuls3. All this in the usual corpuls quality. corpuls.mission consists of the three combinable tools. LIVE, CONFERENCE and REPORT LIVE, approved as medical-device software, is responsible for telemedical connection. The live connection provides vi- tal parameters and curves in real time. It also displays trends and events from the connected device. The Live Board fea- ture makes it possible to monitor multiple devices and for the visual display of events. Automated forwarding, flexible interfaces for data export and notification options round out the functions. Naturally, LIVE is also suitable for measuring the resting ECG. CONFERENCE ensures patient-oriented communication. Here you can get live advice from doctors or specialists via video and audio calls and chat. The ability to send pictures, videos and voice messages completes a modern communica- tion solution. Tactical units such as the dispatch centre, am- bulance and duty doctors, rather than individuals allow easy selection of the communication partners that are required. REPORT ensures true and correct mission documentation. So that everything from patient medical history to handover is documented correctly and according to the standards. If requested, the documentation can be collaborative, so that there is only one set of mission documentation for a patient. In this instance, the report naturally indicates who contrib- uted which content. Interfaces to third-party systems round out the possibilities. JULY | 2020

12 corpuls.mission SPECIAL EDITION | ECGmax Posterior part of a 22-lead ECG being assessed by a telemedical doctor via corpuls.mission LIVE Emergency physicians, cardiologists or any other specialist that is required, can precisely access the mission via LIVE. With the direct transmission of the cell phone number, team ID, radio ID, as well as the device ID, there is no chance for confusion. For better communication possibilities, CONFERENCE offers the option of scanning a QR code on the corpuls3 using a smartphone. With this QR code, both de- vices are automatically associated to the same mission, en- suring correct assignment. With the options that CONFER- ENCE offers in addition to LIVE, the emergency physician or the future treating physician in the hospital can view all vital parameters, ECG data and trends. In addition, the rescue team can exchange information via chat and video confer- ence, transmit photos of medication plans and wounds or breathing sounds and much more. Even before arriving at the hospital, patient treatment measures can be initiated by the emergency services and specifically prepared. In addition to the 12-lead ECG, ECGmax offers the option of displaying posterior leads, the precordial leads on the right and the orthogonal X, Y and Z leads. With ECGmax you get 22 leads for an all-round view of the entire heart. Vector loops are also displayed. This rarely used yet diagnostical- ly valuable method of vectorcardiogram (VCG) can now be used at the same time and without any additional effort. With ECGmax, the CEB® is also introduced as a new feature. A Cardiac Electrical Biomarker that measures the extent of myocardial ischemia based on the multipolar parts of the actual dipolar electrical field of a heart. The CEB® delivers a single numerical value in almost real time, requiring only 10 seconds of resting ECG measurement. If the CEB® ex- ceeds 95, this strongly indicates myocardial problems. Using a tachometer display directly in the software, the CEB® provides quick and easy indication of the patient’s myocardial health. But corpuls.mission can do even more. LIVE now also offers the ECGmax feature via a telemedical solution. As part of the resting ECG, a holistic picture can be presented by means of an electrocardiological measurement. corpuls.mission supports the user so that they can effectively and professionally help the patient throughout the mission. JULY | 2020

22-LEAD ECG 13 22-LEAD ECG ECG diagnostics must be well trained. But despite good knowledge of the peaks illustrating potential differences of the heart, some diseases are still quite difficult to recognize on a standard 12-lead ECG. A typical marker for an acute transmural myocardial infarc- tion is the characteristic ST segment elevation. For an isolat- ed posterior myocardial infarction however, this ST segment elevation is not seen on standard 12-lead ECG (Brown et al. 2003), as no specific leads represent this area (Levis 2015). Nevertheless, up to 20% of cases of acute myocardial in- farction are acute posterior wall myocardial infarction (Levis 2015). The inevitable result is frequently missing the diag- nosis of posterior myocardial infarctions – at least at first. Using additional leads, such as the posterior leads (V7–V9), increases the ability to identify a posterior myocardial infarc- tion significantly (Brown et al. 2003; Levis 2015). Therefore, the ESC Guidelines recommend the routine recording of ad- ditional leads such as the posterior and the right precordial leads, despite the additional effort of new electrode posi- tioning. Nevertheless, this is surely not reality in Germany. (Kelm et al. 2018; Ibanez et al. 2018) Equally in Canada, only 11% of cardiologist and just 9% of emergency physicians use the right precordial and posterior leads (Brown et al. 2003). An explanation for this may be the additional effort, time and materials needed for application. Based on a factor analysis Schreck et al. (1998) showed that three leads are theoretically enough to calculate a 22-lead ECG. All additional leads are more or less redundant and give very little information to the system. Consequently, in 2004, Schreck et al. described a method to derive a 12-, 19- or 22-lead ECG from 3 measured standard leads using a universal patient coefficient matrix. Based on this technique we are able to derive 22-leads out of a standard 12-lead set. Without the necessity of any additional electrodes and without losing the representation of any other area of the heart while observing another. Furthermore, the calcula- tions are based on leads I, II and V2, which are the leads with the least variability in electrode positioning (Schreck and Fishberg 2013). This promises high reliability of the de- rived leads. However, to keep the routine and compatibility to currently used systems, the ECGmax by corpuls, which provides 22-lead ECG, will keep the standard 10 electrodes. Consequently, the 12-lead ECG will still be displayed as the measured ECG. The right precordial, the posterior, as well as the X, Y and Z leads, including the vector loops, will be additionally available. References Brown, L.; Sims, J.; Conforto, A. (2003): Posterior myocardial infarction with isolated ST elevations in V8 and V9. Is this an “ST elevation MI”? In: CJEM 5 (2), S. 115–118. Ibanez, B.; James, S.; Agewall, S.; Antunes, M. J.; Bucciarelli-Ducci, C.; Bueno, H. et al. (2018): 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST segment elevation. The Task Force for the management of acute myocardial infarction in patients presenting with ST segment elevation of the European Society of Cardiology (ESC). In: European heart journal 39 (2), S. 119–177. DOI: 10.1093/eurheartj/ehx393. Kelm, M.; Kastrati, A.; Nef, H.; Richardt, G.; Zeymer, U.; Bauersachs, J.; Kommission für Klinische Kardiovaskuläre Medizin (2018): Kommentar zu den Leitlinien 2017 der Europäischen Gesellschaft für Kardiologie (ESC) zur Therapie des akuten Herzinfarktes bei Patienten mit ST-Streckenhebung. In: Der Kardiologe 12 (2), S. 145–149. DOI: 10.1007/s12181-018-0237-6. Levis J. T. (2015): ECG Diagnosis: Isolated Posterior Wall Myocardial Infarction. In: permj (19(4)), e143-e144. DOI: 10.7812/TPP/14-244. Schreck, D. M.; Tricarico, V. J.; Frank, J. D.; Thielen, L. E.; Chhibber, P.; Brotea, C.; Leber, I. B. (1998): Statistical methodology. VI. Mathematical modeling of the electrocardiogram using factor analysis. In: Academic emergency medicine : official journal of the Society for Academic Emergency Medicine 5 (9), S. 929–934. DOI: 10.1111/j.1553-2712.1998.tb02825.x. Schreck, D. M.; Fishberg, R. D. (2013): Derivation of the 12-lead electrocardiogram and 3-lead vectorcardiogram. In: The American journal of emergency medicine 31 (8), S. 1183–1190. DOI: 10.1016/j.ajem.2013.04.037. JULY | 2020

14 DERIVATION OF THE 12-LEAD ELECTROCARDIOGRAM AND 3-LEAD VECTORCARDIOGRAM SPECIAL EDITION | ECGmax DERIVATION OF THE 12-LEAD ELECTROCARDIOGRAM AND 3-LEAD VECTORCARDIOGRAM (shortened version) David M. Schreck MD, MS a, b, *, Robert D. Fishberg MD c a Departments of Emergency Medicine and Internal Medicine, Summit Medical Group, b Department of Medicine, Atlantic Health System, Overlook Medical Center, Summit, NJ Berkeley Heights, NJ c Department of Medicine and Division of Cardiology, Atlantic Health System, Overlook Medical Center, Summit, NJ * Corresponding author. Department of Emergency Medicine, Summit Medical Group, 1 Diamond Hill Road, Berkeley Heights, NJ 07922. Tel.: +1 908 273 4300. E-mail address: schreckdm@verizon.net (D.M. Schreck). ABSTRACT Objective: The cardiac dipolar field is represented by the measured 12-lead electrocardiogram (ECG) and 3-lead vectorcardio- gram (VCG). The objective is to derive the 12-lead ECG and 3-lead VCG from 3 measured leads acquired from only 5 electrodes. Methods: This is a retrospective blinded study comparing measured and derived ECG and VCG tracings. A nonlinear optimi- zation model was used to synthesize the derived 12-lead ECG and 3-lead derived VCG from leads I, II, and V2. A total of 367 measured 12-lead electrocardiograms and 3-lead vectorcardiograms of varying morphologies were acquired from archived digital ECG databases. All tracings were interpreted by 2 blinded physician reference standards. The derived vs measured tracings were compared quantitatively using Pearson correlation and root mean square error. Qualitative comparisons were determined by physician percent agreement analysis and adjudication. Results: The correlations between the measured and derived ECGs and VCGs were high (r = 0.867). No clinically significant differences were noted in 98.1% of cases. Electrocardiographic rate, rhythm, segment, axis, and acute myocardial infarction interpretations showed 100% correlation. Root mean square error compared favorably against other synthesis techniques. Overall percent agreements for the various ECG morphologies were noted to be 98.4% to 100%. Conclusions: The 12-lead ECG and 3-lead VCG can be derived accurately from 3 measured leads with high quantitative and qualitative correlations. These derived tracings can be acquired instantaneously and displayed in real time from a cardiac rhythm monitor. This will allow for immediate, on-demand, convenient, and cost-effective acquisition and analysis of the 12- lead ECG and 3-lead VCG in areas of acute patient care. 1. INTRODUCTION Cardiac electrical activity is reported to be highly dipolar [1,2] and is clinically represented by both the scalar measured 12- lead electrocardiogram (ECG) (mECG) and 3-lead spatial measured vectorcardiogram (VCG) (mVCG) [3]. In theory, if the cardiac electrical field is dipolar, then only 3 measured leads should be necessary to actually derive this composite 15-lead ECG [4] from just 5 body surface electrodes that are connected to a cardiac rhythm monitoring device. This will allow usual and customary continuous cardiac rhythm mon- itoring with the added simultaneous advantage of acquiring the derived 12-lead ECG (dECG) and scalar 3-lead derived VCG (dVCG), a composite 15-lead ECG, instantaneously and in real time using 1 cardiac rhythm monitoring device. It would be highly desirable to record these 15 scalar ECG leads, along with the corresponding cardiac rhythm data, from a reduced lead set, thereby making the acquisition pro- cess more timely and efficient. The objective of this study is to derive with accuracy the standard 12-lead dECG plus 3-lead dVCG, from just 3 meas- ured leads using 5 body surface electrodes that are a subset of the 12-lead mECG, connected to a new continuous car- diac rhythm monitoring device (VectraplexECG System with JULY | 2020 VectraplexAMI; VectraCor, Inc, Totowa, NJ) using a nonlinear optimization (NLO) constructed universal patient transfor- mation matrix (UPTM). 2. METHODS This study is a blinded retrospective observational design comparing the quantitative and qualitative correlations be- tween the mECGs and mVCGs with their corresponding dECG and dVCG tracings. The standard 12-lead ECGs and 3-lead VCGs were acquired using a Marquette MAC-15 machine (GE Healthcare, Wauke- sha, WI). These ECG records were retrospectively reviewed. Two independent physician readers, a board-certified emer- gency physician (EP) and board-certified cardiologist, served as the blinded reference standards. The physicians were blinded to (1) the ECG acquisition and signal processing, (2) the patient’s disease state, (3) to each other’s interpretation, and (4) to whether the 12-lead ECG and 3-lead VCG were measured or derived. The basis measured 3-lead set (I, II, V2) is used to derive the remaining leads being studied. Leads I, II, and V2 are meas- ured but are not an orthogonal set of lead vectors. However,

DERIVATION OF THE 12-LEAD ELECTROCARDIOGRAM AND 3-LEAD VECTORCARDIOGRAM 15 substituting lead aVF for lead II yields an orthogonal set of leads. Lead aVF can be calculated from leads I and II using accepted geometric formulae [7]. The derived ECG results from the multiplication of the NLO [8] constructed UPTM by the basis 3-lead measured voltage-time data matrix. This UPTM was constructed a priori from an independent set of men and women with variable ECG and VCG morphologies. The quantitative measures of similarity between the original mECG and mVCG and the corresponding dECG and dVCG were determined using Pearson r correlation [4,9] and root mean square error (RMSE – not shown) analysis [10,11] for each derived lead. The Pearson r was considered to show high positive correlation [12] at r ≥ 0.7. Statistical signifi- cance for Pearson r was calculated at α = .025 (1-tailed, positive correlation only). The RMSE is a parameter that indi- cates the average voltage error (microvolts) across the ECG leads studied. Qualitative measures of similarity and reproducibility be- tween the measured and derived ECG and VCG were deter- mined using percent agreement analysis [13] of the interpre- tations by the blinded physician reference standards. The primary outcomes were the quantitative correlations of the measured and derived ECG and VCG tracings based on the Pearson r correlation and RMSE statistics described above and the qualitative methods based on blinded physi- cian percent agreement methodology. Table 2 Pearson r correlation for various 12-lead ECG morphologies (I, II, V2 not shown, because they are the measured lead set) Pearson r All cases AMI Anteroseptal Inferior Lateral Posterior Non_STEMI/ischemia Prior MI (recent or old) Control/non-AMI LBBB RBBB LAFB IRBBB IVCD PVC Paced rhythm NSSTT changes LVH n 366 124 39 46 21 21 63 133 243 19 16 26 21 41 11 4 84 35 III 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 aVR 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 aVL 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 aVF 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 V1 0.87 0.84 0.89 0.78 0.84 0.78 0.86 0.87 0.89 0.91 0.72 0.81 0.62 0.79 0.86 0.80 0.88 0.96 V3 0.90 0.88 0.90 0.85 0.86 0.83 0.88 0.90 0.91 0.94 0.81 0.88 0.89 0.88 0.92 0.89 0.90 0.88 V4 0.71 0.62 0.68 0.56 0.62 0.60 0.57 0.64 0.75 0.53 0.59 0.64 0.85 0.56 0.74 0.74 0.70 0.68 V5 0.71 0.66 .068 0.68 0.53 0.58 0.65 0.63 0.74 0.52 0.66 0.46 0.81 0.58 0.69 0.49 0.69 0.80 V6 0.75 0.68 0.75 0.66 0.60 0.49 0.67 0.65 0.78 0.81 0.69 0.54 0.76 0.74 0.65 0.79 0.78 0.89 12-lead ECG 95% CI 0.867 0.855 0.889 0.842 0.808 0.809 0.847 0.846 0.873 0.881 0.852 0.850 0.869 0.866 0.845 0.913 0.857 0.891 0.010 0.017 0.018 0.033 0.045 0.056 0.024 0.016 0.012 0.032 0.042 0.032 0.043 0.026 0.084 0.009 0.022 0.021 3. RESULTS A total of 367 cases were enrolled in this study. Table 2 describes the Pearson r correlation analysis for a vari- ety of different morphologic ECGs using the NLO method. It should be emphasized that the correlation is only performed on the derived leads. Leads I, II, and V2 are not analyzed because they comprise the measured lead set from which all derivations are constructed and have no error. This is an advantage because these leads are a subset of the 12- lead ECG. As such, additional education on an unfamiliar basis lead set is not required. In addition, the accuracy of the placement of the electrodes for these leads has minimal error. All NLO-synthesized limb leads showed perfect corre- lations (r = 1.00) suggesting the UPTM should have high ac- curacy in synthesizing the remaining derived leads. Pearson r for the 12-lead ECG showed high correlations (r = 0.867) for all cases. In addition, all subtype ECG morphologies demon- strated high correlations ranging from 0.823 to 0.913. All derived leads independently show good correlation with the corresponding measured leads across all cases. All correla- tions were statistically significant at P < 0.001. Table 3 describes the Pearson r correlations for the derived leads at specific points in the ECG and VCG waveforms. All correlations were statistically significant at each portion of the ECG cycle at P < 0.001. For the RMSE analysis for the NLO 12-lead dECG derivation method used by the VectraplexECG System is important to note that there is essentially no error in 7 leads (I, II, III, aVR, aVL, aVF, and V2) using the NLO method. This is due to the NLO method using leads I, II, and V2 as the basis measured lead set from which all other leads are derived. The consensus interobserver agreement for both blinded physician reference standards reached by adjudication also showed high agreement in diagnostic accuracy between the mECG and dECG. The cardiologist intraobserver agreement between the mECG and dECG for various ECG morpholog- JULY | 2020

16 DERIVATION OF THE 12-LEAD ELECTROCARDIOGRAM AND 3-LEAD VECTORCARDIOGRAM SPECIAL EDITION | ECGmax ic conditions demonstrates high agreement in diagnostic accuracy between the mECG and dECG. The EP intraobserv- er agreement between the mECG and dECG for various ECG morphologic conditions also demonstrates high agreement in diagnostic accuracy between the mECG and dECG. Figure 4A-4D show comparisons of the scalar mECGs and mVCGs with the corresponding dECGs and dVCGs for several clinical states. Table 3 Pearson r correlation of measured vs derived leads V1, V3, V4, V5, V6, X, Y and Z at specific waveform peak voltages (P, Q, R, S, J, ST20, ST60, ST80, T+, T-) mP vs dP mQ vs dQ mR vs dR mS vs dS mJ vs dJ mST20 vs dST20 mST60 vs dST60 mST80 vs dST80 mT+ vs dT+ mT- vs dT- V1 0.41 0.69 0.79 0.69 0.86 0.89 0.90 0.89 0.81 0.79 V3 0.73 0.66 0.82 0.75 0.83 0.89 0.90 0.90 0.88 0.89 V4 0.71 0.76 0.63 0.62 0.71 0.73 0.73 0.73 0.76 0.83 V5 0.73 0.78 0.64 0.60 0.71 0.71 0.74 0.75 0.78 0.72 V6 0.75 0.80 0.73 0.72 0.75 0.77 0.80 0.81 0.86 0.73 X 0.84 0.81 0.68 0.80 0.88 0.85 0.84 0.86 0.85 0.77 Y 0.92 0.93 0.88 0.88 0.92 0.92 0.88 0.92 0.91 0.94 Z 0.57 0.84 0.74 0.76 0.81 0.81 0.83 0.83 0.89 0.83 D.M. Schreck, R.D. Fishberg / American Journal of Emergency Medicine 31 (2013) 1183–1190 B Normal 15-Lead ECG: mECG (blue) vs. dECG (red) Acute inferior wall STEMI: mECG (blue) vs. dECG (red) ) V m ( e g a t l o V Measured Derived Measured Derived Time (msec) Time (msec) LBBB 15-Lead ECG: mECG (blue) vs. dECG (red) RBBB 15-Lead ECG: mECG (blue) vs. dECG (red) D ) V m ( e g a t l o V A C ) V m ( e g a t l o V ) V m ( e g a t l o V Measured Derived Measured Derived Time (msec) Time (msec) ECG mECG dECG Electrocardiogram Measured 15-lead ECG (12-lead ECG plus 3-lead Vectrocardiogram) Derived 15-lead ECG MI STEMI LBBB RBBB Myocardial infarction ST elevation MI Left bundle branch block Right bundle branch block Figure 4. A, Normal 15-lead ECG: mECG (blue) vs dECG (red). B, Acute inferior wall ST-elevation myocardial infarction: mECG (blue) vs dECG (red). C, Left bundle-branch block 15-lead ECG: mECG (blue) vs dECG (red). D, Right bundle-branch block 15-Lead ECG: mECG (blue) vs dECG (red). MI, myocardial infarction; STEMI, ST-elevation MI; LBBB, left bundle-branch block; RBBB, right bundle-branch block. JULY | 2020

DERIVATION OF THE 12-LEAD ELECTROCARDIOGRAM AND 3-LEAD VECTORCARDIOGRAM 17 4. DISCUSSION This is the first study describing the process for synthesiz- ing the dECG and dVCG from just 3 measured leads that are a subset of the standard 12-lead ECG, using an NLO technique. The concept of the dECG has been previously re- ported [15-19]. Schreck et al [20] reported that factor anal- ysis [21] demonstrated that the 12-lead ECG can be derived from just 3 measured lead vectors using a “patient-specific” transformation matrix. However, Schreck et al [22] subse- quently demonstrated that a “universal patient” transfor- mation matrix can be computed using an NLO technique that can be used to derive the 12-lead ECG independent of sex, race, age, body habitus, and timing, thus eliminating the need for constructing a patient-specific transformation matrix for each patient. The NLO constructed UPTM allows a “one-size-fits-all” approach to using only 1 UPTM for all patients at any time to derive any patient 12-lead ECG and 3-lead VCG. The device used in this study uses 3 measured leads, with the least lead placement variability (limb leads and V2), which are a subset of the standard 12-lead ECG and are well known and understood. The spatial 3-lead VCG was reported by Frank [23] over 50 years ago, and its clinical value has been reported in nu- merous publications and textbooks [24-28]. In addition, the VCG has been reported to be more sensitive [29-32] in the diagnosis of cardiac pathology, but it is becoming a lost art due to lack of readily available machinery and lack of educa- tion of this technology. It would be clinically advantageous in acute care settings to implement the VCG concepts to enhance diagnostic capabilities if the acquisition of the VCG could be accomplished through an accurate derivation pro- cess directly from a standard cardiac rhythm monitor. The clinical application of this research will be advantageous in areas of acute care. The ability to derive a standard 12- lead ECG and spatial 3-lead VCG directly from a standard cardiac rhythm monitor using only 5 electrodes will make the acquisition of these tests much faster, less costly, more easily accessible, and more efficient if clinical accuracy can be demonstrated between the measured and derived ECG and VCG tracings. The main limitation of this study is its retrospective design. As such, several sources of bias may be present. Another limitation in this study is that the physician refer- ence standards, although blinded, are an imperfect “truth.” In addition, the mECG by itself is also an imperfect truth in that electrode placement variability is very high for the chest leads [33]. One more limitation of this study is that a direct comparison of the NLO method to EASI method of ECG derivation was not performed. The results of this study demonstrate that the dECG and de- rived spatial 3-lead VCG using the NLO methodology have high correlation with the corresponding measured scalar leads based on quantitative and qualitative comparisons. This derived composite 15-lead ECG can be constructed from just 3 acquired leads using 5 body surface electrodes connected to a cardiac rhythm monitor. These derived 15 scalar ECG leads are obtained in real-time or on demand directly from the cardiac rhythm monitor. This minimizes the need to obtain frequent serial measured ECGs requiring staff time and numerous electrodes. Application of this device will be advantageous in acute care settings by identifying poten- tial ECG morphology changes in real time in a more conven- ient and efficient manner. References [1] Geselowitz DB. Dipole theory in electrocardiography. Am J Cardiol 1964;14:301–6. [2] Gabor D, Nelson CV. Determination of the resultant dipole of the heart from measurements on the body surface. J Appl Phys 1954;25:413–6. [3] Malmivuo J, Plonsey R. 12-Lead ECG system. In: Malmivuo J, Plonsey R, editors. Bioelectromagnetism: principles and applications of bioelec- tric and biomagnetic fields. London: Oxford University Press; 1995. p. 277–89. [4] Nelwan SP, Kors JA, Meij SH, et al. Reconstruction ofthe 12-lead electrocardiogram from reduced lead sets. J Electrocardiol 2004;37:11–8. [5] Goldberger AL, Amaral LAN, Glass L, et al. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. Circulation 2000;101:e215–20. [6] Frank E. General theory of heart-vector projection. Circ Res 1954;2:258–70. [7] Goldman MJ. Introduction to electrocardiography. In: Goldman MJ, editor. Principles of clinical electrocardiography. Eighth Edition. Los Altos: Lange Medical Publications; 1973. p. 13–4. [8] Shavers CL, Parsons ML, Deming SN. Simplex optimization of chemical systems. J Chemical Education 1979;56:307–9. [9] Chantad D, Krittayaphong R, Komoltri C. Derived 12-lead electro- cardiogram in the assessment of ST segment deviation and cardiac rhythm. J Electrocardiol 2006;39: 7–12. [10] Nelwan SP, Crater SW, Meij SH, et al. Simultaneous comparison of three derived 12-lead ECGs with standard ECG at rest and during percutane ous coronary occlusion. J Electrocardiol 2008;41:230–7. [11] Drew BJ, Dempsey ED, Joo TH, et al. Pre-hospital synthesized 12-lead ECG ischemia monitoring with trans-telephonic transmission in acute coronary syndromes: pilot study results of the ST SMART trial. J Electro cardiol 2004; 37(Suppl):214–21. [12] Munro BH. Correlation. In: Munro BH, editor. Statistical methods for healthcare research. Fifth Edition. Philadelphia: Lippincott Williams & Wilkins; 2005. p. 239–58. [13] Clinical and Laboratory Standards Institute document EP12-A2: User Protocol for Evaluation of Qualitative Test Performance: Approved Guideline. Second Edition. Wayne, Pennsylvania, 2008. [14] Dower GE, et al. Deriving the 12-lead electrocardiogram from four (EASI) electrodes. J Electrocardiol 1988;21:182. [15] Scher AM, Young AC, Meredith WM. Factor analysis of the electro cardiogram. Test of electrocardiographic theory: normal hearts. Circ Res 1960;8:519–26. [16] Man SC, Maan AC, Kim E, et al. Reconstruction of standard 12-lead electrocardiograms from 12-lead electrocardiograms recorded with the Mason-Likar electrode configuration. J Electrocardiol 2008;41:211–9. [17] Horacek BM, Warren JW, Wang JJ. On designing and testing trans- formations and derivation of standard 12-lead/18-lead electro cardiograms and vectorcardiograms from reduced sets of predictor leads. J Electrocardiol 2008;41:220–9. JULY | 2020

18 DERIVATION OF THE 12-LEAD ELECTROCARDIOGRAM AND 3-LEAD VECTORCARDIOGRAM SPECIAL EDITION | ECGmax [18] Nelwan SP, Carey MG. Reduced and alternative lead sets: clinical and technical challenges. J Electrocardiol 2008;41:451–3. [26] Chou TC, Helm RA, Kaplan S. Clinical vectorcardiography. . Second Edition.New York: Grune & Stratton; 1974. [19] Scherer JA, Jenkins JM, Nicklas JM. Synthesis ofthe12-lead electro- cardiogram from a 3 lead subset using patient-specific transformation vectors: an algorithmic approach to computerized signal synthesis. J Electrocardiol 1989; 22:128. [20] Schreck DM, Tricarico VJ, Frank JD, et al. Statistical methodology: VI. Mathematical modeling ofthe electrocardiogram using factor analysis. Acad Emerg Med 1998;5: 929–34. [21] Malinowski ER. Factor analysis in chemistry. . Second Edition.New York: John Wiley & Sons; 1991. [22] Schreck DM, Brotea C, Shah S. Derivation of the 12-lead electro- cardiogram using abstract factor analysis and simplex optimization. Int J Bioelectromagnetism 2002;4:337–8. [23] Frank E. An accurate, clinically practical system for spatial vector- cardiography. Circulation 1956;13:737–49. [24] Benchimol A. Vectorcardiography. Huntington, NY: Robert E. Krieger; 1979. [25] Friedman H. Diagnostic and clinical electrocardiography and vector- cardiography. . Second Edition.New York: McGraw-Hill; 1977. [27] Cooksey JD, Dunn M, Massie E. Clinical vectorcardiography and electro cardiography. . Second Edition.Chicago: Year Book Medical Publishers; 1977. [28] Carson W. Emergency vectorcardiography. London: Wangdon Carson, MD; 1998. [29] Orlando J, Aronow WS. Value of the vectorcardiogram in assessing tall R waves in right precordial leads. Chest 1976;6:540–2. [30] Chou TC. When is the vectorcardiogram superior to the scalar electrocardiogram? J Am Coll Cardiol 1986;8:791–9. [31] Edenbrandt L, Pahlm O, Lyttkens K, et al. Vectorcardiogram more sensitive than 12-lead ECG in the detection of inferior myocardial infarction. Clin Physiol 1990;10:551–9. [32] Riera RAP, Uchida AH, Filho CF. Significance ofvectorcardiogram in the cardiologic diagnosis of the 21st century. Clin Cardiol 2007;30:319–23. [33] Schijvenaars BJA, van Herpen G, Kors JA. Intraindividual variability in electrocardiograms. J Electrocardiol 2008;41:190–6. JULY | 2020

MATHEMATICAL MODELING AND UTILITY OF THE DERIVED 22-LEAD ELECTROCARDIOGRAM 19 MATHEMATICAL MODELING AND UTILITY OF THE DERIVED 22-LEAD ELECTROCARDIOGRAM David M. Schreck, Robert D. Fishberg, Summit Medical Group, Berkeley Heights, NJ RESULTS The dECGs showed high correlation with mECGs overall by Pearson correlations (0.84-0.88). No clinically significant differences were noted in 98.1% of the dECGs. ECG rate, rhythm, segment, and axis interpretations showed 100% correlation. Acute MI differentiation showed 100% corre- lation. Kappa analysis of the mECG vs. dECG showed high overall correlations (0.73-1.00). CONCLUSIONS The 22-lead ECG can be derived from just 3 measured leads using the SOP technique. The comparison of the mECGs and dECGs shows high quantitative and qualitative correla- tions. Using this technology a 22-lead derived ECG can be displayed instantaneously in real-time to enhance patient observation capabilities and will allow for a convenient and cost effective acquisition and analysis of the ECG in teleme- try and critical care areas of health care. BACKGROUND The cardiac electrical field is dipolar and is measured by the electrocardiogram (ECG) which may be described by a 3 lead-vector space. There are 22 leads used in clinical practice including the standard 12-lead ECG, right heart leads V3R- V6R, posterior leads V7-V9, and the vectorcardiographic (VCG) leads X, Y, Z. It would be advantageous to derive these 22 ECG leads from just 3 measured leads using a universal patient coefficient matrix (UPCM) that can be computed us- ing simplex optimization (SOP). The objective is to derive the ECG (dECG) from 3 measured leads using a SOP-computed UPCM and calculate the quantitative and qualitative correla- tions with the measured ECG (mECG). METHODS A total of 371 mECGs of varying morphology for both men and women age 18 and older were acquired including 371 standard 12-lead ECGs, 353 VCGs, 75 right heart ECGs, and 34 posterior ECGs. The ECG morphologies included nor- mals, acute MIs, LVH, bundle branch blocks, paced beats, PVCs, and non-specific STT types. Each ECG was interpreted by 2 physicians who were blinded reference standards. The SOP technique was used to derive a UPCM from an addi- tional training set of 20 ECGs. Leads I, aVF, and V2 from the mECGs were chosen as the 3 lead-vector basis orthog- onal lead set from which the dECGs were synthesized. The derived vs. measured test case ECGs were compared using Pearson and Kappa statistics. JULY | 2020

20 VECTOR LOOPS SPECIAL EDITION | ECGmax VECTOR LOOPS The vectorcardiogram (VCG) is a useful method for cardi- ac diagnosis. The cardiac cycle is represented in the VCG as three loops. Each loop corresponds to the P, QRS and T wave activities (Bhattacharyya et al. 2020). It contains more information with possible diagnostic significance, it is more sensitive than the conventional ECG and it is useful for the detection and localization of acute myocardial infarction, right ventricular hypertrophy and Tawar arm blockade (Jaros et al. 2019; Bhattacharyya et al. 2020). A healthy myocardi- um leads to a regular and smooth VCG, but a left ventricular dysfunction might lead to irregular, disorganized or criss- cross propagation. An abnormal ventricular repolarization sequence results in an abnormal morphology of the T wave. (Bhattacharyya et al. 2020) The diagnostic potential of the VCG is high, but the difficulty and cumbersomeness in constructing a VCG loop prevents its popularity in routine use. In 1956, Ernest Frank introduced the VCG lead-system based on 7 electrodes (A, C, E, I, M, H and F) to derive three equal- length, orthogonal image vectors which are stable for indi- vidual anatomical variations, called the VCG. The orthogonal image vectors represent the right-to-left axis, the anteropos- terior and the caudal to cranial axis of the human body. The electrodes A, C and I measure the potential difference Vx, which corresponds to the right-to-left axis. The z-axis (anteroposterior lead) is based on the electrodes A, C, E, I, and M. All five are necessary for the reliability of the lead and for stability against dipole-location changes. The last axis from cranial to caudal, the y-axis, is derived from electrodes H, M and F. For measurement of the VCG in addition to the ECG, you needed to attach these 7 electrodes together with the standard 12-lead ECG, which means 17 electrodes for each patient (Schreck and Fishberg 2013). Consequently, the VCG is not commonly used in clinical practice due the effort required in its application and the complex, specific interpre- tation (Jaros et al. 2019). Nowadays these vectors can be derived from a reduced set of the standard 12-lead ECG electrode set (Schreck and Fish- berg 2013). There is no need for any additional electrodes or to change current routines, yet all benefits of this method can be reaped. It can easily be compared to the ECG at the same time, resulting from the same electrode set. This is the chance to implement the VCG into clinical routine and to learn more about the loops, their meaning and their potential. The literature cited and the following review article by Carlos Alberto Pastore and colleagues offers a good overview of the VCG and its diagnostic potential. Visualization of the electrical activity of the three ortogonal leads (X,Y,Z) in the shape of vector loops References Bhattacharyya, Sukanti; Goswami, Damodar Prasad; Sengupta, Arnab (2020): Spatial velocity of the dynamic vectorcardiographic loop provides crucial insight in ventricular dysfunction. In: Medical hypotheses 135, S. 109484. DOI: 10.1016/j.mehy.2019.109484. Jaros, Rene; Martinek, Radek; Danys, Lukas (2019): Comparison of Different Electrocardiography with Vectorcardiography Transformations. In: Sensors 19 (14), S. 3072. DOI: 10.3390/s19143072. Schreck, David M.; Fishberg, Robert D. (2013): Derivation of the 12-lead electrocardiogram and 3-lead vectorcardiogram. In: The American journal of emergency medicine 31 (8), S. 1183–1190. DOI: 10.1016/j.ajem.2013.04.037. JULY | 2020

APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE 21 APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE (Review article - shortened version) Carlos Alberto Pastore, Nelson Samesima, Horacio Gomes Pereira Filho, Nancy Maria Martins de Oliveira Tobias, Bruna Affonso Madaloso, Mirella Esmanhotto Facin Instituto do Coração (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (FMUSP), São Paulo, SP – Brazil ABSTRACT The electrocardiogram (ECG) has been reinvigorated by the identification of electrical alterations that were not definite- ly clarified before. In this context, and mainly regarding the definition of arrhythmogenic substrates, the association of the ECG with the vectorcardiogram (VCG) has gathered much more information about the cardiac electrical phenom- ena, thus allowing us to differentiate potentially fatal cases from benign ones. Obtaining a VCG concomitantly with the performance of an ECG has led to a significant gain in the definition of extremely sophisticated pathologies, which function suffer some type of structural or dynamic alterations, involving either the reduction or enhancement of ionic channels and currents. The classic aspects of the ECG/VCG association in the differential diagnosis of myocardial infarctions, conduction dis- orders, atrial and ventricular hypertrophies, and the correlations between these electrical disorders are still valid and assertive. The association of these pathologies is further clarified when they are seen through the ECG/VCG dyad. The three-dimensional spatial orientation of both the atrial and the ventricular activity provides a far more complete observation tool than the ECG linear form. The modern analysis of the ECG and its respective VCG, simultaneously obtained by the recent technique called electro-vectorcardiogram (ECG/VCG), brought a significant gain for the differ- ential diagnosis of some pathologies. Therefore, we illustrate how this type of analysis can elucidate some of the most important diagnoses found in our daily clinical practice as cardiologists. INTRODUCTION The study of vectorcardiography began during the 1940’s and publications reached a peak between the 1960’s and 1970’s, correlating this method with the heart diseases best known at that time. The great difficulty then was linked to the fact that the VCG device could not be easily moved around. The images were thus not immediately obtained, and so the vectorcardiogram was a tool to be used a pos- teriori to resolve doubts about specific electrocardiograms in some special situations. A temporal gap, associated to a diminished interest in electrovectorcardiography, resulted in a significant decrease in the number of cardiology centers capable of performing and interpreting a vectorcardiogram. However, with the development of invasive electrophysiolo- gy (electroanatomic mapping), genetics and molecular bio- logy, many electrical conditions have been unveiled, result- ing in the identification of their clinical/electrocardiographic patterns, since such conditions can lead to sudden death.1-7 The technological developments seen during the 1990s also affected electrovectorcardiography. The sophistication brought by the use of computers, algorithmic systems and Fourier transforms allowed us to obtain vectorcardiographic information in a much simpler and quicker form, in color, and as three-dimensional images. The electrocardiogram (ECG) was therefore reinvigorated by the identification of electrical alterations. Mainly regard- ing the definition of arrhythmogenic substrates, it was ob- served that the association of the ECG and the vectorcardio- gram (VCG) methods could provide much more information about the cardiac electrical phenomena, thus increasing its employment and allowing us to differentiate potentially fatal cases from benign ones.8-10 Our team is involved with the teaching of electro-cardiog- raphy to undergraduate and postgraduate students in the medical area. Therefore, we feel there is an urgent need to teach vectorcardiography, considering that its spatial visuali- zation of the cardiac electrical activation makes it a lot easier to understand and memorize the basic and more complex electrocardiographic notions.11 Obtaining a VCG concomitantly with the performance of an ECG has led to a significant gain in the definition of ex- tremely sophisticated pathologies, of which genetic muta- tions cause their function to suffer some type of structural or dynamic alterations, involving either the reduction or en- hancement of ionic channels and currents. The classic aspects of the ECG/VCG association in the dif- ferential diagnosis of myocardial infarctions, conduction disorders, atrial and ventricular hypertrophies, and the cor- relations between these electrical disorders are still valid and assertive.12,13 The association of these pathologies is further clarified when they are seen through the ECG/VCG dyad. (Figure 1) Based on the abovementioned facts, during the last decade we were able to develop the performance of the binomial electrovectorcardiogram in the context of the most varied JULY | 2020

22 APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE SPECIAL EDITION | ECGmax pathologies. This binomial can add sophistication to the al- ready known clinical entities, in addition to a greater accu- racy of the recent electrocardiographic definitions (such as Brugada, early repolarization, etc.) Our experience, both academic and scientific, led us to join these new ECG/VCG acquisitions, and to open a window into the observation of the electrical phenomena of the heart. The literature has shown that the more sophisticated vectorcardiogram makes it easier for us to observe punctual phenomena that are not defined by the ECG. Figure 1: Normal Electrocardiogram and Vectorcardiogram Electrocardiology Electrocardiogram Vectorcardiogram Figure 1: Electrocardiogram vs. Vectorcardiogram. The electrovectorcardiography binomial The experience with the VCG during these last decades shows the greater specificity and sensitivity of this method to detect the subtleties of these diagnoses. In comparison with the ECG, the VCG shows some advantages; however, when in association, they can help us differentiate between some very ordinary situations in clinical practice. The three-dimensional spatial orientation of both the atrial and the ventricular activity provides a far more complete ob- servation tool than the linear form of the ECG. The modern analysis of the ECG and its respective VCG, simultaneously obtained by the recent technique called electro-vectorcar- diogram (ECG/VCG), brought a significant gain for the dif- ferential diagnosis of some pathologies.1,3,4,8,14,15 (Figure 2) The electro-vectorcardiographic analysis is very rich and con- sistent for the diagnosis of myocardial infarctions (MI), since the difficulties in defining pathological Q waves or the loss of R waves in the ECG can be very clearly visualized in the ECG/ VCG. This association helps us to define the real changes in the direction and orientation of the vectorcardiographic loops created by the areas of myocardial infarction, in both the transverse and the frontal planes.16-18 (Figure 3) Another important differential aspect obtained by the elec- JULY | 2020 Figure 2: Orthogonal planes

APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE 23 trovectorcardiogram is the investigation of the presence of a myocardial infarction area in the inferior wall, of a left su- perior fascicular block, or the association of both patholo- gies. The association of MIs with the presence of fascicular or troncular blocks can be fully characterized by the ECG/ VCG association. The inferior MIs with a left anterosuperior fascicular block (LASFB), and the anterior MIs with a right bundle-branch block (RBBB) are typical examples of the im- portance of the ECG/VCG association for a differential diag- nosis.19,20 (Figure 3A and 4A) Figure 3A: Old inferior myocardial infraction (MI) Electrocardiology Electrocardiogram Vectorcardiogram Figure 3: (A) Presence of myocar-dial infarction area in the inferior wall, with QRS vectorcardiographic loop onset oriented upwardly and to the left. (B) not shown here (see original article) path marked by the ventricular activation loops can identify the blocks, as well as other associations.21,23,28 (Figure 7) The end-conduction delays, previously denominated incom- plete right bundle-branch blocks, are neatly defined by the ECG/VCG. These findings can be mistaken for the left phe- nomena and also can mimic a myocardial infarction area. Thus, the association of ECG/VCG solves the doubts that arise from the presence of these delays, which can be either the variants from the normal, or even suggest a conduction disorder in specific areas of the right ventricle. The presence of the end-conduction delay (ECD) is clar- ified in the ECG/VCG by the S1 S2 S3 pattern, with the S wave in D2 greater than the one in D3, qR in aVR and pres- ence of S wave from V1 to V6. The ECG/VCG confirms the ECD position backwardly and to the right in the transverse plane, and up-wardly and to the right in the frontal plane.25 (Figure 8 not shown) One of the most recent diagnostic achievements of the ECG/ VCG refers to the criteria to establish Brugada syndrome patterns, as well as the early repolarization (ER) phenome- non. It is important to emphasize that, in typical cases, there are no difficulties to make the electro-vectorcardiographic identification of both conditions. Due to the severity of the JULY | 2020 The spatial orientation of the fascicular blocks can be bet- ter understood through the electro-vectorcardiogram. The septal vector orientation and the direction of the vectorcar- diographic loop activation neatly characterize the fascicular blocks and their associations through the ECG/VCG, since they define the electrical path of this phenomenon, thus characterizing exactly the position of the blocks.21-24 (Figure 4A and 5(A and B not shown) The VCG complements the ECG in the analysis of acute myocardial infarctions and makes the differential diagnosis of the associations with blocks and chamber hypertrophies. Furthermore, the ECG/VCG is capable of characterizing the presence of: 22-27 (Figure 5B, 6[A, B, C (not shown) and D]) a) Left anteromedial fascicular block (LMFB) b) Right ventricular hypertrophy (RVH) c) d) Ventricular pre-excitation syndrome (WPW) e) Brugada syndrome Lateral infarction The ECG/VCG is the gold standard to identify com-plete and fascicular blocks, because it can differen-tiate them either in isolation or in association with other blocks. The electrical

24 APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE SPECIAL EDITION | ECGmax first, to the variability of the second between normal cases and other pathologies, and also to the possibility of having an early repolarization pattern located in a more anterior area, it became essential to make an adequate distinction between them. Specific ECG/VCG patterns of the J-wave abnormalities, namely the Brugada syndrome (BrS), and the early repolarization pattern (RP) were studied by our research team. An important qualitative and quantitative analysis of the ECG/VCG was carried out in all the study population, specifically regarding aspects of the area comprising the terminal portion of the QRS loop, the J point and the ST segment. This analysis showed a neat end-conduction delay (ECD) in all the individuals (BrS and RP). This ECD is charac- terized by a conduction delay greater than 10 ms at the final portion of the QRS in all the planes, either to the right or to the left. In the transverse plane, the QRS loops showed a counter- clockwise rotation, with the ECD beginning posteriorly and ending anteriorly, with the main difference between the groups being the ECD position. In the ER pattern 9,25 (Figure 10 not shown) we see a clock- wise rotation (terminal portion of the QRS) of the same seg- ments, resembling a “fishhook”. In all the ER cases the ECD position was in the left quadrant, with a shorter duration. In 100% of the ER patients, the final portion of the QRS loop showed a fishhook pattern. A very important arrhythmogenic pathology, the arrhythmo- genic right ventricle cardiomyopathy (ARVC), has an almost definitive assessment tool in the electrovectorcardiographic diagnosis. The end-conduction delay with low voltage and long duration to the right (forward or slightly backwards) characterizes the phenomenon with great accuracy, with the differential diagnosis being very important, since this pathol- ogy may lead to severe arrhythmias. (Figure 11 not shown) ARVC sometimes presents with an aspect similar to the RBBB, although with a very low voltage that is different from that block. It can also show an ECD aspect on the right and be slightly backwards. The presence of a negative T wave in V1, V2, V3 and left posteriorly located in the transverse plane of the ECG/VCG is crucial for an accurate diagnosis. The Brugada syndrome has shown to be very dynamic regarding its arrhythmogenic substrate, and the ECG/VCG follow-up can be very useful to define this process.29,30 (Figure 12 not shown). Figure 4A: LASFB Electrocardiology Electrocardiogram Vectorcardiogram Figure 4: Aspects of the VCG loops in the Inferior MI (in Figure 3A of the previous item, note the QRS loop onset upward-ly in the frontal plane, with more than 30 ms du- ration [15 comets]); in Figure 4A – LASFB, observe the QRS loop onset downwardly (through the posterior-inferior division), with counterclockwise rotation and the major portion of the loop oriented up- wardly and to the left; B not shown JULY | 2020

APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE 25 Figure 6: Burgada Syndrome Electrocardiology Electrocardiogram Vectorcardiogram Figure 6: Illustrations of differen-tial diagnoses identified by the ECG/VCG in different pathologies: Brugada syndrome - Figure 6D. Figure 7: LPIFIB + RBBB Vectorcardiographic aspects of the associ-ation of LPIFB and RBBB: axis to the right in the frontal plane (LPIFB), with most of the QRS loop in the frontal plane, oriented downwardly and to the left, and the QRS loop in the transverse plane, slowly ending forwardly and to the right (RBBB). D1 D2 D3 aVR aVL aVF V1 V2 V3 V4 V6 Frontal Sagittal Transverse JULY | 2020

26 APPLICABILITY OF THE ELECTRO-VECTORCARDIOGRAM IN CURRENT CLINICAL PRACTICE SPECIAL EDITION | ECGmax 23. Pastore CA, Moffa PJ, Grinberg M, Verginelli G, Ebaid M, Tranchesi J. Bloqueio da divisão posteroinferior em pós-operatório de comis- surotomia aórtica e ressecção de anel fibroso subvalvar. Correlação com bloqueio atrioventricular total (relatos de casos). Arq. Bras. Cardiol. 1979;32(5):303-7. 24. Moffa PJ, Ferreira BMA, Sanches PCR, Tobias NMM, Pastore CA, Bellotti G. Bloqueio divisional anteromedial intermitente em paciente com insuficiência coronária. Arq Bras Cardiol. 1997;68(4):293-6. 25. Pastore CA, Samesima N, Kaiser E. Vectorcardiographic patterns for differential diagnosis in right end-conduction delay, early repolari-sation and Brugada syndrome. J Electrocardiol. 2012; 46:e1-e37. 26. de Micheli A, Iturralde P, Medrano GA. On the electrical manifes- tations of some heart diseases associated with ventricular preexcitation. Arch Cardiol Mex. 2006;76(Suppl 4):S137-43. 27. Titomir LI, Trunov VG, Aidu EA, Sakhnova TA, Blinova EV. New approaches to the diagnosis of left and right ventricular hypertrophy by means of dipolar electrocardiotopography. Anadolu Kardiyol Derg. 2007;7(Suppl 1):29-31. 28. Pérez-Riera AR, Barbosa-Barros R, Daminello-Raimundo R, de Abreu LC, Tonussi Mendes JE, Nikus K. Left posterior fascicular block, state-of-the- art review: a 2018 update. Indian Pacing Electrophysiol J. 2018;18(6):217-30. 29. Samesima N, Pastore CA, Facin ME. Brugada syndrome ablation: what the vectorcardiogram can tell us? ICE/2017. Portland, EUA [Abstract]. Disponível em: www.electrocardiology.org/images/conferences/ ICE_2017/ sched-ule_2017-06-28.pdf. 30. Pastore CA, Samesima N. Brugada and J wave patterns: the role of vectorcardiography in distinguishing them. In: Brugada after RF ablation: what the vectorcardiogram can tell us? [citado 25 mar. 2019]. Disponível em: www.electrocardiology.org/images/Presentations/VCG-Brugada-Pastore.pdf. References 1. Moffa PJ, Sanches PCR. Eletrocardiograma normal e patológico. 7ª. ed. São Paulo: Rocca; 2001. 2. Strauss DG, Olson CW, Wu KC Heiberg E, Persson E, Selvester RH, et al. Vectorcardiogram synthesized from the 12-lead electrocardiogram to image ischemia. J Electrocardiol. 2009;42(2):190-7. 3. Bayes de Luna A. Clinical electrocardiography: a textbook. 4th. ed. New Jersey: Wiley-Blackwell; 2012. 4. Pastore CA, Samesima N, Munerato R. ABC do ECG. 3ª. ed. São Paulo: Medcel; 2013. 5. Gussak I, Antzelevich C, Wilde AAM, Powell BD, Ackerman MJ, Shen WK. Electrical diseases of the heart. London: Springer-Verlag; 2008. v.1. 6. Moss AJ. Can long QT syndrome be diagnosed by vectorcardiography when the corrected QT on the resting 12-lead electrocardiogram is of normal duration? Heart Rhythm. 2017;14(6):900-1. 7. Cortez D, Schlegel TT, Ackerman MJ, Bos JM. ECG-derived spatial QRS-T angle is strongly associated with hypertrophic cardiomyopathy. J Electrocardiol. 2017;50(2):195-202. 8. 9. Pastore CA, Samesima N, Tobias N, Pereira Filho HG. Eletrocardio- grafia atual: curso do serviço de eletrocardiografia do inCor. 3ª. ed. São Paulo: Atheneu; 2016. Pastore CA, Samesima N, Pereira Filho HG, Madaloso BA. Controver-sial and similar aspects of the Brugada and J wave patterns: the vector- cardiogram point of view. J Electrocardiol. 2016;49(3):439-45. 10. Lingman M, Hartford M, Karlsson T, Herlitz J, Rubulis A, Caidahl K, et al. Value of the QRS-T area angle in improving the prediction of sudden cardiac death after acute coronary syndromes. Int J Cardiol. 2016 Sep 1;218:1-11. 11. Hurst JW. Methods used to interpret the 12-lead electrocardiogram: pattern memorization versus the use of vector concepts. Clin Cardiol. 2000;23(1):4-13. 12. Chou TC. When is the vectorcardiogram superior to the scalar electrocardiogram? J Am Coll Cardiol. 1986;8(4):791-9. 13. Man S, Maan AC, Schalij MJ, Swenne CA. Vectorcardiographic diagnostic & prognostic information derived from the 12-lead electro- cardiogram: historical review and clinical perspective. J Electrocardiol. 2015;48(4):463-75. 14. Sedaghat G, Ghafoori E, Waks JW, Kabir MM, Shvilkin A, Josephson ME, et al. Quantitative assessment of vectorcardiographic loop morphology. J Electrocardiol. 2016;49(2):154-63. 15. Carvalho Filho ET, Moffa PF, Pastore CA, Curiati JÁ et al. Vectorcar- diographic study of normal aged subjects. Arq. Bras. Cardiol. 1987;48(4):211-6. 16. Sederholm M. The origin of monitoring of acute myocardial infarction with continuous vectorcardiography. J Electrocardiol. 2014; 47(4):418-24. 17. Correa R, Arini PD, Correa LS, Valentinuzzi M, Laciar E. Identification of patients with myocardial infarction. Vectorcardiographic and electro- cardiographic analysis. Methods Inf Med. 2016;55(3):242-9. 18. Warner RA, Battaglia J, Hill NE, Mookherjee S, Smulyan H. Importance of the terminal portion of the QRS in the electrocardio-graphic diagnosis of inferior myocardial infarction. Am J Cardiol. 1985;55(8):896-9. 19. Pastore CA, Moffa PJ, Tobias NMO, Moraes AP, Nishioka SA, Chierighini JEC, et al. Bloqueios divisionais do ramo direito e áreas eletricamente inativas. Diagnóstico diferencial eletro-vetocardiográfico. Arq. Bras. Cardiol. 1985;45(5):309-17. 20. Tobias NMMO, Pastore CA, Moffa PJ, Moraes AP, Lima EV, Del Nero Jr E, et al. Bloqueios divisionais do ramo direito na miocardi- opatia chagásica. Arq. Bras. Cardiol. 1986;47(6):387-91. 21. Acunzo RS, Konopka IV, Sanchéz RA, Pizzarelli N, Wells FC, Baranchuk A, et al. Right bundle branch block and middle septal fiber block with or without left anterior fascicular block manifested as aberrant conduction in apparent healthy individuals: electrovectorcardi- ographic characterization. J Electrocardiol. 2013;46(2):167-72. 22. Tranchesi J, Moffa PJ, Pastore CA, Carvalho Filho ET, Tobias NMM, Scalabrini Neto A, et al. Bloqueio da divisão anteromedial do ramo esquerdo do feixe de His nas coronariopatias: caracterização vector-cardiográfica. Arq. Bras. Cardiol. 1979;32(6):355-60. JULY | 2020

THE CARDIAC ELECTRICAL BIOMARKER – CEB® THE CARDIAC ELECTRICAL BIOMARKER – CEB® 27 A heart beat results in a dipolar electrical field (Strebel et al. 2018; Abaecherli et al. 2017; Chattopadhyay et al.) which is measurable as a vector between electrodes on the body surface. If acute myocardial ischemia occurs, this dipolar field is disturbed by multipolar parts (Strebel et al. 2018). Vectra- Cor developed a method to use these multipolar vectors as a marker for myocardial injury. This cardiac electrical biomark- er CEB® is constructed from three ECG leads of the stand- ard ECG electrode set. In addition to the ECG, the CEB® is a tool for detection of acute myocardial ischemic injury with high diagnostic accuracy by analyzing ECG changes (Schreck and Fishberg 2013). The current gold standard for a specific diagnosis of myocardial injury is the highly sensitive cardi- ac troponin I (Hs-cTnI). Hs-cTnI is a marker for myocardial necrosis. It should be measured at clinical presentation and 3h after admission for a 100% sensitivity and negative pre- dictive value (Thygesen et al. 2012). Therefore, Tereshchenko et al. (2014) analyzed the correlation of troponin and CEB® in 3h-steps. They showed a significant correlation between troponin and CEB® for the second and third 3h-period after presentation of the patient. Strebel et al. (2018) stated a weak but significant correlation between troponin and CEB® also at the time of presentation. Consequently, the CEB® is a marker which correlaes with Hs-cTnI with one major benefit. The CEB® is instantly available the moment an ECG is meas- ured (Schreck and Fishberg 2013; Tereshchenko et al. 2014). Rather than 3h, only 10 seconds is required until an inter- pretable numerical index is available. The higher the num- ber, the higher the multipolar activity and consequently the CEB indicator showing acute myocardial problems probability of acute myocardial ischemic injury (Strebel et al. 2018). A receiver-operating-characteristic analysis showed that a value of 65 and smaller indicates a normal dipolar electrical heart, while a value of 95 and higher is a clear indication to suggest acute myocardial injury. The zone be- tween 66 and 94 is indeterminate, but these cases occurred in only 10% of the patients (Schreck and Fishberg 2013). Even though there were a number of indeterminate CEB®s included as false positive or false negative results, analyzing the correlation of a diagnosis with the CEB® and the diagno- sis of different physicians still shows a noninferiority of the CEB®. (Schreck and Fishberg 2013) In summary, the CEB® is simultaneously available with the ECG, it shows high diagnostic accuracy, troponin I correla- tion and correlation to physicians’ diagnostic ECG interpre- tation. This technology allows immediate identification of patients with AMII on scene (Schreck and Fishberg 2015). References Abaecherli, R.; Strebel, I.; Twerenbold, R.; Boeddinghaus, J.; Osswald, S.; Müller, C.; Reichlin, T. (2017): Abstract 13562: Cardiac Electrial Biomarker, a Novel Marker Diagnosing Myocardial Injury in Patients With Symptoms Suggestive for NSTEMI. Online verfügbar unter https://ahajournals.org/doi/ abs/10.1161/circ.136.suppl_1.13562, zuletzt aktualisiert am 14.11.2017, zuletzt geprüft am 11.12.2019. Chattopadhyay, S.; Adjei, F.; Kardos, A.: Cardiac electrical biomarker retains “memory” of myocardial ischemia not severe enough to cause myocardial necrosis. ESC Congress Paris 2019. Online verfügbar unter https://www. vectracor.com/milton-keynes/, zuletzt geprüft am 03.01.2020. Schreck, D. M.; Fishberg, R. D. (2013): Diagnostic Accuracy of a New Cardiac Electrical Biomarker for Detection of Electrocardiogram Changes Suggestive of Acute Myocardial Ischemic Injury. In: Annals of Non-invasive Electrocardi- ology 19 (2), S. 129–144. DOI: 10.1111/anec.12109. Schreck, D. M.; Fishberg, R. D. (2015): Detection of acute myocardial ischemic injury by gender using a novel cardiac electrical biomarker. In: The American journal of emergency medicine 33 (3), S. 383–390. DOI: 10.1016/j.ajem.2014.12.029. Strebel, I.; Twerenbold, R.; Boeddinghaus, J.; Abächerli, R.; Rubini Giménez, M.; Wildi, K. et al. (2018): Diagnostic value of the cardiac electrical bio- marker, a novel ECG marker indicating myocardial injury, in patients with symptoms suggestive of non-ST-elevation myocardial infarction. In: Annals of non-invasive electrocardiology : the official journal of the International Society for Holter and Non-invasive Electrocardiology, Inc 23 (4), e12538. DOI: 10.1111/anec.12538. Tereshchenko, L. G.; Gatz, D.; Feeny, A.; Korley, F. K. (2014): Automated analysis of the 12-lead ECG in the emergency department. Association between high-sensitivity cardiac troponin I and the cardiac electrical bio- marker. In: Critical pathways in cardiology 13 (1), S. 25–28. DOI: 10.1097/ HPC.0000000000000006. Thygesen, K.; Mair, J.; Giannitsis, E.; Mueller, C.; Lindahl, B.; Blankenberg, S. et al. (2012): How to use high-sensitivity cardiac troponins in acute cardiac care. In: European heart journal 33 (18), S. 2252–2257. DOI: 10.1093/eu- rheartj/ehs154. JULY | 2020

DETECTION OF ACUTE MYOCARDIAL ISCHEMIC INJURY BY GENDER USING A NOVEL CARDIAC ELECTRICAL BIOMARKER 28 SPECIAL EDITION | ECGmax DETECTION OF ACUTE MYOCARDIAL ISCHEMIC INJURY BY GENDER USING A NOVEL CARDIAC ELECTRICAL BIOMARKER (shortened version) David M. Schreck MD, MS a, b, *, Robert D. Fishberg MD c a Departments of Emergency Medicine and Internal Medicine, Summit Medical Group, Berkeley Heights, NJ b Department of Medicine, Atlantic Health System, Overlook Medical Center, Summit, NJ c Department of Medicine and Division of Cardiology, Atlantic Health System, Overlook Medical Center, Summit, NJ ABSTRACT Objective: The objective of this study us to stratify by gender a new cardiac electrical biomarker (CEB) diagnostic accuracy for detection of acute myocardial ischemic injury (AMII). Methods: This is a noninferiority retrospective, case-control, blinded study of 310 archived measured electrocardiograms (ECGs) acquired from 218 men and 92 women. The CEB is constructed from the derived ECG (dECG) synthesized from 3 leads. Electrocardiograms were included if acquired less than or equal to 1 day from patient presentation. Electrocardiograms were interpreted by 2 blinded physicians and adjudicated by consensus. Standard ST analyses and computerized ECG interpreta- tions were active controls. Electrocardiograms were excluded for noise and baseline wander, age younger than 18 years, and ectopic beats in the 10-second ECG acquisition. Diagnostic accuracy measures of sensitivity, specificity, positive and negative predictive values, and likelihood ratios were stratified by gender. Measured vs derived ECG correlations were quantitatively compared using Pearson correlation and qualitatively by percent agreement methodology. Results: The CEB sensitivities for AMII detection in men and women were 93.9% and 90.5%, respectively, and CEB specificities were 90.7% and 95.2%, respectively, and were superior to active controls. Derived and measured ECGs showed high correlation for both men and women with r =0.857 and r = 0.893, respectively. Reference standard intra-agreement analysis for meas- ured ECGs and dECGs with AMII was 99.4%. Conclusions: The CEB demonstrates high diagnostic accuracy for detection of AMII in men and women. The ECG can be derived with accuracy from 3 leads. This technology is an efficient real-time method of identifying patients with AMII who are being monitored in acute care settings. © 2014 Elsevier Inc. All rights reserved. 1. INTRODUCTION The concepts of a new and novel cardiac electrical biomark- er (CEB) and the derived 15-lead electrocardiogram (ECG) have been recently reported by Schreck and Fishberg [1,2]. Briefly, cardiac electrical activity is reported to be highly dipolar [3,4], and as such, only 3 measured orthogonal leads should be needed to actually derive this composite 15-lead ECG. This will allow continuous cardiac rhythm monitoring with the added simultaneous advantage of acquiring the de- rived 12-lead ECG (dECG) and scalar 3-lead derived vector- cardiogram, a composite 15-lead ECG, instantaneously and in real-time using 1 cardiac rhythm monitoring device. The objective of this study is to identify the CEB diagnostic accu- racy, stratified by gender, compared to active controls (ACs). 2. MATERIALS AND METHODS This is a noninferiority, retrospective, blinded, case-control, paired comparator [5] study. These 310 measured ECGs (mECGs) of various morphologies were obtained from 2 da- tabases including an archived National Institutes of Health– funded Physiobank PTBDB database [6] and a database from Muhlenberg Regional Medical Center (Plainfield, NJ). The study ECGs represent a gender stratification subanalysis of a recent prior study by Schreck and Fishberg [1]. Electrocardi- ograms were included if acquired less than or equal to 1 day from patient presentation. The dECGs were constructed from 3 measured leads I, II, and V2, which were converted to a 3-lead orthogonal basis set of {I, aVF, V2} [7] using Einthoven triangular geometric relationships. The 15-lead dECGs were synthesized from this orthogonal lead set. The CEB is constructed from the medi- an beat from each dECG using the VectraplexECG System (VectraCor, Inc, Totowa, NJ) and is a measure of the dipolar energy in the cardiac electrical field. The CEB was compared to each AC that included the 12- lead ECG computer interpretation and the ST analysis volt- age parameters ST0 (J point) and ST area under curve (ST- SUM). The STSUM points included the lead voltages at the J point and at 20, 60, and 80 milliseconds after the J point JULY | 2020

DETECTION OF ACUTE MYOCARDIAL ISCHEMIC INJURY BY GENDER USING A NOVEL CARDIAC ELECTRICAL BIOMARKER 29 gender and AC for the EP reference standard. The CEB was noted to be noninferior by hypothesis testing, but superiority was statistically demonstrated using the actual data com- pared to the ACs. Figure 4 (not shown) shows the false-pos- itive rate (FPR), also known as 1 − specificity. The CEB was noted to be noninferior by hypothesis testing, but superiori- ty was statistically demonstrated using the actual and worst case data compared to the ACs. The CEB is shown to be noninferior by hypothesis testing for the TPR analyses. The CEB is shown to superior to ACs in the FPR analyses. Figures 7 to 8 (not shown) show the same TPR and FPR analyses as adjudicated by the reference standards to mimic the real-world situation where the EP and cardiol- ogist both collaborate on the ECG interpretation. The CEB is again shown to be noninferior by hypothesis testing in both the TPR and FPR analyses, but superiority was demonstrated in the actual data TPR analysis. Superiority was also demon- strated for both actual and worst case data FPR analysis. All CEB diagnostic accuracy measures were significant (P<.025). Derived vs measured 12-lead ECGs showed high correlation for both men and women with r=0.857 and r=0.893, respectively. Because the CEB is constructed from the dECG, it is important to demonstrate the correlation be- tween the mECG and dECG to support the diagnostic accu- racy of the dECG for AMII cases. Table 3 (not shown) shows the quantitative Pearson r correlations of the mECGs vs the corresponding dECGs by each derived lead. Table 4 (not shown) shows reference standard percent inter-agreement and intra-agreement analysis for mECGs and dECGs with AMII. (ST0, ST20, ST60, and ST80, respectively). There were 2 reference standards including 1 board-certified emergency physician (EP) and 1 board-certified cardiologist who independently interpreted the ECGs and were blinded to each other’s interpretations. The 2 reference standards also adjudicated the results at study completion. The criteria reported by Thygesen et al [8] were used by the reference standards to interpret ECGs for the presence of ST changes consistent with ST segment elevation myocardial infarction (STEMI) and non-STEMI. Measured vs derived ECGs correlations were quantitatively compared using Pearson correlation coefficient (r) and qual- itatively by reference standard percent agreement method- ology [9]. The CEB diagnostic accuracy parameters including sensitiv- ity, specificity, negative and positive predictive values (NPV and PPV), likelihood ratios (LR+ and LR−), and odd ratios (ORs) were identified and compared to the ACs in a nonin- feriority design [5] and stratified by gender using a 1-sided (α < .025, 1 − β = 0.90) interval with 95% confidence sta- tistical analysis. 3. RESULTS Table 2 shows the CEB sensitivities, specificities, NPV, PPV, LR+, LR−, and OR for acute myocardial ischemic injury (AMII) detection in men and women, also stratified by reference standard and adjudication. The detailed CEB sensitivity anal- yses, also known as the true-positive rate (TPR), and the 1 − specificity analyses, also known at the false-positive rate, are shown in Figures. 3 to 8.These figures show the analyses of the CEB/AC ratios stratified by gender, AC, and reference standard interpretation with adjudication. The CEB is consid- ered a positive test for AMII if greater than 94 and negative for AMII if less than 66. The CEB is considered indeterminate from 66 to 94. As such, the sensitivity is also stratified by “actual” and “worst” case scenarios. The worst-case sce- nario is such that any CEB in the indeterminate region is considered a false-positive or false-negative result. Figure 3 shows the sensitivity (TPR) analyses stratified by Table 2 Cardiac electrical biomarker diagnostic accuracy parameters EP Men Women Total Cardiology Men Consensus Women Total Men Women Total AMII CEB CEB Prevalence Sensitivity Specificity 18.5% 36.6% 20.1% 15.4% 32.1% 16.3% 20.9% 25.0% 20.6% 92.7% 95.5% 93.7% 82.9% 94.4% 86.8% 93.9% 90.5% 92.2% 91.1% 91.9% 91.3% 82.8% 80.0% 82.0% 90.7% 95.2% 92.9% CEB NPV 98.1% 98.3% 98.1% 96.1% 98.2% 96.7% 97.5% 98.3% 97.7% n 222 92 314 228 98 326 218 92 310 CEB PPV 71.7% 80.8% 74.7% 48.3% 54.8% 50.5% 79.6% 76.9% 78.7% CEB LR(+) 10.38 11.84 10.77 4.83 4.72 4.82 10.09 19.00 13.02 CEB LR(-) 0.08 0.05 0.07 0.21 0.07 0.16 0.07 0.10 0.08 CEB OR 129.2 239.4 154.9 23.2 68.0 29.9 149.2 190.0 154.9 CEB Utility 92.2% 91.3% 93.3% 94.3% 89.8% 92.9% 94.5% 91.3% 93.5% JULY | 2020

DETECTION OF ACUTE MYOCARDIAL ISCHEMIC INJURY BY GENDER USING A NOVEL CARDIAC ELECTRICAL BIOMARKER 30 SPECIAL EDITION | ECGmax Figure 3: Emergency physician ECG interpretation: CEB:AC actual and worst case sensitivities for men and women. Abbreviations: TPR, true-positive rate (sensitivity); δ,effect margin; ST0,J-point ST voltage; STSUM, area under curve of ST segment voltages at J-point and 20, 60, and 80 milliseconds after J point. 5. CONCLUSIONS The CEB demonstrates high diagnostic accuracy for detec- tion of AMII for men and women when compared to stand- ard ST segment analysis and ECG computer interpretation ACs. The 12-lead ECG can be derived with accuracy from just 3 leads directly from the cardiac monitor. The measured and derived 12-lead ECGs show high qualitative and quan- titative correlation. This technology will allow an immediate, cost effective, and efficient means of identifying patients with AMII who are being monitored in acute care settings. 4. DISCUSSION The concept of the derived ECG is not new and has been reported by several investigators [11-17]. Now that recent guidelines for STEMI [18] and non-STEMI [19] care have called for frequent serial ECGs to be acquired, the utilization of the derived ECG technology may now become more ad- vantageous and clinically useful. It was very important to note that this study demonstrated high correlations between them ECG and dECG. The high interagreement and intraagreement analysis also lends sup- port for the use of the dECG in the clinical setting. The concept of the CEB allows a more efficient method for observing patients being evaluated for chest pain equiva- lents in any acute care setting, particularly the ED. The CEB is obtained continuously and displayed on the cardiac monitor in real time allowing immediate identification of a potential AMII in the proper clinical setting. It was interesting to note that the CEB diagnostic accuracy parameters were very favorable in both men and women. JULY | 2020

References [1] Schreck DM, Fishberg RD. Diagnostic accuracy of a new cardiac electrical biomarker for detection of electrocardiogram changes suggestive of acute myocardial ischemic injury. Ann Non-invasive Electrocardiol 2014;19(2):129–44. [2] Schreck DM, Fishberg RD. Derivation of the 12-lead electrocardiogram and 3-lead vectorcardiogram. Am J Emerg Med 2013:1183–90. [3] Geselowitz DB. Dipole theory in electrocardiography. Am J Cardiol 1964;14:301–6. [4] Gabor D, Nelson CV. Determination of the resultant dipole of the heart from measurements on the body surface. J Appl Phys 1954;25:413–6. [5] Alonzo TA, PepeMS, Moskowitz CS. Sample size calculations for comparative studies of medical tests for detecting presence of disease. Stat Med 2002;21:835–52. [6] Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG, et al. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. Circulation 2000;101:e215–20. [7] GoldmanMJ. Introduction to electrocardiography. In:Goldman MJ, editor. Principles of clinical electrocardiography. 8 ed. Los Altos: Lange Medical Publications; 1973. p. 13–4. [8] Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD. Third universal definition of myocardial infarction. Circulation 2012;126:2020–35. [9] Clinical and Laboratory Standards Institute document EP12-A2: user protocol for evaluation of qualitative test performance: approved guideline. 2 ed. Pennsylvania: Wayne; 2008. [10] Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, et al. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. Ann Intern Med 2003;138:40–4. [11] Dower GE, Yakush A, Nazzal SB, Jutzy RV, Ruiz CE. Deriving the 12-lead electrocardiogram from four (EASI) electrodes. J Electrocardiol 1988;21:182. [12] Nelwan SP, Kors JA, Meij SH, van Bemmel JH, Simoons ML. Reconstruction of the 12-lead electrocardiogram from reduced lead sets. J Electrocardiol 2004;37:11–8. [13] Drew BJ, Dempsey ED, Joo TH, Sommargren CE, Glancy JP, Benedict K, et al. Prehospital synthesized 12-lead ECG ischemia monitoring with trans-telephonic transmission in acute coronary syndromes: pilot study results of the ST SMART trial. J Electrocardiol 2004(37 Suppl.):214–21. DETECTION OF ACUTE MYOCARDIAL ISCHEMIC INJURY BY GENDER USING A NOVEL CARDIAC ELECTRICAL BIOMARKER 31 [14] Man SC, Maan AC, Kim E, Draisma HHM, Schalij MJ, van der Wall EE, et al. Reconstruction ofstandard 12-lead electrocardiograms from 12-lead electrocardiograms recorded with the Mason-Likar electrode configuration. J Electrocardiol 2008;41:211–9. [15] Horacek BM, Warren JW, Wang JJ. On designing and testing transformations and derivation of standard 12-lead/18-lead electro- cardiograms and vectorcardiograms from reduced sets of predictor leads. J Electrocardiol 2008;41:220–9. [16] Nelwan SP, Carey MG. Reduced and alternative lead sets: clinical and technical challenges. J Electrocardiol 2008;41:451–3. [17] Scherer JA, Jenkins JM, Nicklas JM. Synthesis ofthe 12-lead electro- cardiogram from a 3 lead subset using patient-specific transformation vectors: an algorithmic approach to computerized signal synthesis. J Electrocardiol 1989;22:128. [18] Antman EM,Anbe DT, Armstrong PW,Bates ER, Green LA,Hand M,et al. ACC/AHA Guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. Circulation 2004;110: 588–636. [19] Amsterdam EA, Wenger NK, Brindis RG, Casey Jr DE, Ganiats TG, Holmes Jr DR, et al. 2014 AHA/ACC Guideline for the management of patients with non–ST-elevation acute coronary syndromes: executive summary: a report of the American College of Cardiology/ American Heart Association Task Force on practice guidelines. Circulation 2014;130(25):2354–94. [20] Redberg RF, Shaw LJ. Diagnosis of coronary artery disease in women. Prog Cardiovasc Dis 2003;46(3):239–58. [21] Mosca L, Manson JE, Sutherland SE, Langer RD, Manolio T, Barret-Connor E. Cardiovascular disease in women. Circulation 1997;96:2468–82. [22] Litwin SE, Priester TC. Diagnosing coronary artery disease in women: an update on cardiac imaging modalities. Obstet Gynecol 2010;115(1):156–9. [23] Jekel JF, Elmore JG, Katz DL. Understanding and reducing errors in clinical medicine. In: Jekel JF, editor. Epidemiology biostatistics and preventative medicine. 1st ed. Philadelphia: W. B. Saunders Company; 1996. p. 86–97. JULY | 2020

AUTOMATED ANALYSIS OF THE 12-LEAD ECG IN THE EMERGENCY DEPARTMENT: ASSOCIATION BETWEEN HIGH-SENSITIVITY CARDIAC TROPONIN I AND THE CARDIAC ELECTRICAL BIOMARKER 32 AUTOMATED ANALYSIS OF THE 12-LEAD ECG IN THE EMERGENCY DEPARTMENT: ASSOCIATION BETWEEN HIGH-SENSITIVITY CARDIAC TROPONIN I AND THE CARDIAC ELECTRICAL BIOMARKER (shortened version) Larisa G. Tereshchenko, MD, PhD, David Gatz, BS, Albert Feeny and Frederick K. Korley, MD ABSTRACT Timely detection of myocardial injury is essential for appropriate management of patients in emergency department (ED) evaluated for acute myocardial infarction. A novel electrocardiogram (ECG) metric, the Cardiac Electrical Biomarker (CEB), uses eigenvalue modeling of the 12-lead ECG and quantifies dipolar vs. multipolar forces. The goal of this project was to study association between the CEB and high-sensitivity troponin I (HsTnI). We conducted a retrospective study of patients, evaluated in the ED for acute myocardial infarction [n = 411; 57.6 ± 13.2 years; 186 (45%) men; 266 (64%) African-Ameri- cans]. Resting 12-lead ECG and HsTnI were measured at presentation and at 3, 6, and 9 hours after the initial measurement. The CEB was measured by the VectraplexECG System (VectraCor, Totowa, NJ). Patient-specific longitudinal analysis was performed to study association between the CEB with HsTnI changes over time. The CEB indicated myocardial injury in 116 (28.2%) study participants. HsTnI was significantly elevated during ED observation period in patients with myocardial injury, diagnosed by the CEB [median (interquartile range), 10.3 (5.2–31.4) vs. 6.3 (3.5–16.5) ng/L; P = 0.002]. In a mixed-effects linear regression adjusted for age, race, and sex, increasing HsTnI was associated with the CEB elevation [β-coefficient, 0.071 (95% confidence interval, 0.008–0.134); P = 0.027]. In conclusion, in patients in ED evaluated for acute myocardial injury, increasing values of HsTnI were associated with increasing values of the CEB, suggesting that myocardial injury is the mech- anism that underlines acute changes in the CEB. Key Words: electrocardiogram, emergency medicine, acute coronary syndrome, cardiac electrical biomarker (Crit Pathways in Cardiol 2014;13: 25–28) Copyright © 2014 by Lippincott Williams & Wilkins Each year, a large number of patients are evaluated in emergency departments (EDs) for acute myocardial infarc- tion (AMI). AMI is diagnosed based on the history of pres- ent illness, 12-lead electrocardiogram (ECG), and cardiac biomarkers. Failure to identify patients in the early stages of AMI can result in failure to provide beneficial therapies. The third universal definition of AMI1 requires a rise and/ or a fall in cardiac troponin, with at least 1 troponin val- ue above the 99th percentile. It is recommended that blood samples for troponin measurement are drawn at presentation and repeated 3–6 hours later to optimize clini- cal sensitivity for ruling in AMI. Traditional 12-lead ECG is less informative very early in the course of ST-elevation myocardial infarction (STEMI) and especially in non-STEMI. Recently, new ECG technology was proposed to detect myocardial injury. Development of the novel Vectraplex-ECG System with Cardiac Electrical Biomarker (CEB) or VectraplexAMI was based on factor analysis2 with simplex optimization. Acute myocardial injury may result in heterogeneous electrical field.3,4 VectraplexAMI index was developed by Schreck,5 who post- ulated that CEB measures the degree of the multipolarity of the cardiac electrical field. We conducted a retrospective study of adult patients in ED with available serial digital 12-lead ECGs and high-sensitivity troponin I (HsTnI) results, with the goal of determining whether myocardial injury is the mechanism that underlines acute changes in CEB. We hypothesized that the CEB correlates with HsTnI in patients admitted to ED with suspected AMI. METHODS Study Population We retrospectively analyzed data from an ongoing prospec- tive observational cohort study of patients in ED evaluated for AMI. Serial ECG and troponin measurements were per- formed at presentation and at 3, 6, and 9 hours after pres- entation at the discretion of treating clinicians. The exclusion of patients with STEMI also allows us to evaluate the CEB in patients in whom standard ECG is nondiagnostic. ECG Analysis Each 12-lead ECG was reviewed and clinically evaluated by 2 investigators (D.G. and L.G.T.), blinded to all other clinical data. For each ECG, investigators evaluated cardiac rhythm and determined the presence or absence of left bundle branch block, right bundle branch block, pathological Q wave, ST segment elevation or depression, and nonspecific ST-T changes. The first ECG, recorded at the JHH ED, was com-pared with the previously recorded ECG (if available), and observed ECG abnormalities were categorized as “new,” or “old.” Each ECG was adjudicated and included into 1 of 5 categories: new STEMI, new non-STEMI, new non-specific ST-T changes, unchanged abnormal ECG, and normal ECG. Interreader agreement was evaluated, and in case of disa- greement, the final ECG diagnosis was based on the third ECG reader (JHH attending cardiologist) assessment. The CEB for each ECG was calculated automatically by VectraplexECG System, as previously described5 and provided by Vectracor, JULY | 2020

AUTOMATED ANALYSIS OF THE 12-LEAD ECG IN THE EMERGENCY DEPARTMENT: ASSOCIATION BETWEEN HIGH-SENSITIVITY CARDIAC TROPONIN I AND THE CARDIAC ELECTRICAL BIOMARKER 33 inc. (Totowa, NJ). A predefined threshold was used. We con- sidered that values of the CEB greater than 94 units indicat- ed myocardial injury, as recommended by the manufacturer. Outcomes All clinical, laboratory, and ECG data were reviewed by an independent endpoints adjudication committee, blinded to the results of the VectraplexECG analysis. AMI was defined according to the guidelines,1 clinical setting consistent with myocardial ischemia. Statistical Analysis All statistics were computed using Stata 13 (StataCorp LP, College Station, TX). Results are presented as mean ± SD for normally distributed variables and as median and interquar- tile range for skewed continuous variables. Normally distrib- uted continuous variables were compared using the Student t test. The Wilcoxon rank-sum test was applied to skewed continuous variables—HsTnI and the CEB. Dichotomized variables were compared by Pearson χ2 test. Spearman rank correlation coefficient rs was calculated to quantify relations between HsTnI and the CEB. Then HsTnI and the CEB varia- bles were log-10-transformed to normalize distribution, for subsequent regression analysis. To determine whether the patient-specific changes in the CEB are associated with the HsTnI changes during observation in the ED, we ran the gen- eralized least squares random-effects linear regression anal- ysis. We accounted for correlation of the repeated troponin and ECG observations by including a random intercept for each patient and control for the patient’s CEB (centered). Adequate fitting of the model was checked to ensure that the specified quadrature has adequately approximated the likelihood. The model was adjusted by age, sex, and race. RESULTS Study Population The study population consisted of middle-aged adults (57.6 ± 13.2 years), 65% were African-Americans (n = 266), and 55% were women (n = 225). Among the African-Amer- icans, there were more women than men [159 (60%) vs. 107 (40%); P = 0.006], whereas an opposite sex compo- sition was observed in non-African-Americans [66 (45.5%) women vs. 79 (54.5%) men; P = 0.006]. Risk factors of cor- onary heart disease were frequently observed: 263 patients (64%) had hypertension, 131 patients (32%) had diabetes mellitus, and 263 patients (64%) were current or former smokers. Approximately a quarter of study participants (n = 106; 25.8%) were current or former cocaine users. ECG Analysis: The VectraplexAMI Index The CEB indicated myocardial injury in 116 (28.2%) study participants. Clinical characteristics of patients with and without myocardial injury are shown in Table 1. Patients with myo-cardial injury as detected by the CEB were old- er, more likely to have hypertension, hypercho-lesterolemia, and heart failure. HsTnI was significantly elevated during ED observation period in patients with myocardial injury, diag- nosed by the CEB (Table 1). Table 1 Clinical and ECG Characteristics of Patients with and Without Myocardial Injury as Determined by VectraplexAMI Index VectraplexAMI ≤94 VectraplexAMI >94 Age (SD), yr Men, n (%) African-Americans, n (%) Hypertension, n (%) Diabetes Hx, n (%) Heart failure, n (%) Current or former smokers, n (%) Current or former cocaine users, n (%) Hypercholesterolemia, n (%) Family Hx CHD, n (%) HsTnI #1, median (IQR) HsTnI #2, median (IQR) HsTnI #3, median (IQR) HsTnI #4, median (IQR) Pathological Q wave on ECG, n (%) Normal ECG, n (%) Right bundle brunch block, n (%) Left Bundle branch block, n (%) CHD, coronary heart disease; Hx, history; IQR, interquartile range. Bold indicates statistical findings. (N=295) 56.7 (12.5) 136 (46.1) 186 (63.1) 180 (61.0) 95 (32.2) 66 (22.4) 186 (63.1) 72 (24.4) 126 (42.7) 105 (35.6) 6.3 (3.5–16.5) 6.2 (3.2–14.8) 6.55 (3.5–16.0) 6.55 (3.55–13.9) 38 (12.9) 122 (41.4) 5 (1.7) 6 (2.0) (N=116) 59.9 (14.7) 50 (43.1) 80 (69.0) 83 (71.6) 36 (31.0) 38 (33.8) 77 (66.4) 34 (29.3) 64 (55.2) 44 (37.9) 10.3 (5.2–31.4) 9.2 (4.6–28.6) 12 (4.8–30.4) 12.45 (4.7–32.7) 40 (34.5) 15 (12.9) 15 (12.9) 3 (2.6) P 0.039 0.583 0.259 0.045 0.819 0.029 0.527 0.306 0.023 0.657 0.002 0.021 0.018 0.076 <0.0001 <0.0001 <0.0001 0.731 JULY | 2020

AUTOMATED ANALYSIS OF THE 12-LEAD ECG IN THE EMERGENCY DEPARTMENT: ASSOCIATION BETWEEN HIGH-SENSITIVITY CARDIAC TROPONIN I AND THE CARDIAC ELECTRICAL BIOMARKER 34 There were no patients with typical STEMI ECG presentation among study participants. Pathological Q wave and right bundle branch block were more frequently observed, where- as a normal 12-lead ECG was less frequent in the myocardial injury group (Table 1). Association Between High-sensitivity Cardiac troponin I and the Cardiac Electrical Biomarker During the first 3 hours of observation at the ED, HsTnI and the CEB did not correlate. However, statistically significant correlation was found and strengthened during the next 6 hours of observation (r = 0.163; P = 0.036 and rs = 0.179; P = 0.018), and further at the fourth 3-hour period (r = 0.227; P = 0.026 and rs = 0.217; P = 0.034). Figure 1 illustrates correlation between HsTnI and the CEB. In the univariate generalized least squares random-effects linear regression model, changes in HsTnI were associated with the changes in the CEB: β-coefficient, 0.083 (95% confidence interval, 0.022–0.144); P = 0.008. Thus, increas- ing HsTnI by an order of magnitude (10-fold in-crease) was associated with 8.3% increase of the CEB value. After ad- justment for age, race, and sex, association between HsTnI and the CEB remained significant: β-coefficient, 0.071 (95% confidence interval, 0.008–0.134); P = 0.027. DISCUSSION This study showed that in patients evaluated for AMI in the ED, increasing by an order of magnitude HsTnI is associated with simultaneously increasing value of the CEB. Therefore, an underlying myocardial injury is an important mechanism of acute changes in the CEB in this study population. Multipolar vs. Dipolar Forces in the Cardiac Electrical Field The cardiac electrical field of a healthy subject is primarily dipolar.7 The vector of myocardial injury current differs from the heart vector. Thus, occurrence of myocardial injury leads to the appearance of multipolar cardiac electrical field.9 The CEB quantifies the quality of the cadiac electrical field, x e d n i I l M A x e p a r t c e V d e m r o f s n a r t – 0 1 g o L Log10–transformed HsnTn I Figure 1 Correlation between high-sensitivity cardiac troponin I and CEB in patients in emergency department. JULY | 2020 and in particular, whether the cardiac field is predominantly dipolar, or whether multipolar forces in the cardiac electrical field are present, and how much. VectraplexAMI is provided for the end-user as a single number. In this study, we used the CEB threshold of 94 units, as recommended by the Vec- traCor, Inc., for detection of myocardial injury.5 Additional studies are needed to define optimal threshold for discrimi- nating between no AMI and AMI cases. Simplex optimization (nonlinear optimization technique) was applied to the ECG signal to obtain and reconstruct the ECG signal,2,10–13 and to calculate a proprietary CEB. This is the first study to examine whether myocardial injury (as deter- mined by changes in HsTnI) is associated with acute changes in the CEB. Our findings suggest that in patients with high pretest probability of AMI, an underlying myocardial injury, as detected by an increasing HsTnI, is associated with eleva- tion of the CEB. Clinical Implications of Association Between HsTnI and the CEB Surface ECG is traditionally used for the diagnosis of an acute myocardial injury.1,14 However, with the advent of cardiac tro- ponins, it has been demonstrated that the sensitivity and spec- ificity of the acute changes in the ST segments on an ECG are modest, with sensitivity ranging from 50% to 100%, and specificity ranging from 71% to 91%,15,16 and inconsistent among individual readers. Traditional ECG in-terpretation not only suggests the presence or absence of myocardial injury but at the same time determines localization and extent of the myocardial injury, predicting the possible culprit vessel and complications. This was especially important in the era of conservative AMI management. However, it does not make any difference for current therapeutic strategies.17 In addi- tion, it is worth noting that while cardiac troponin quantifies myocardial injury by a single number, ECG AMI diagnosis is dichotomized (yes or no), and requires complex knowledge of pattern recognition. In the current realm of clinical practice, the CEB offers a simple 1-number quantification of the myo- cardial injury on the surface ECG. Ease of use, low cost, wide availability, and a possibility of the continuous monitoring of the surface ECG in the ED make the CEB promising for future use.23 CONCLUSIONS In conclusion, in patients in ED evaluated for acute myocardi- al injury, increasing values of HsTnI were associated with in- creasing values of the CEB, suggesting that myocardial injury is associated with acute changes in the CEB in the population of patients with high pretest probability of acute myocardial injury.

SPECIAL EDITION | ECGmax 35 References 1. Thygesen K, Alpert JS, Jaffe AS, et al; Joint ESC/ACCF/AHA/WHF Task Force for the Universal Definition of Myocardial Infarction. Third universal definition of myocardial infarction. Circulation. 2012;126:2020–2035. 2. Schreck DM, Tricarico VJ, Frank JD, et al. Statistical methodology: Vi. Mathematical modeling of the electro-cardiogram using factor analysis. Acad Emerg Med. 1998;5:929–934. 3. Han J, Moe GK. Nonuniform recovery of excitability in ventricular muscle. Circ Res. 1964;14:44–60. 4. Allessie MA, Bonke FI, Schopman FJ. Circus movement in rabbit atrial muscle as a mechanism of tachycardia. ii. the role of nonuniform recovery of excitability in the occurrence of unidirectional block, as studied with multiple microelectrodes. Circ Res. 1976;39:168–177. 5. Schreck DM, Fishberg RD. Diagnostic accuracy of a new cardiac electrical biomarker for detection of electrocardiogram changes suggestive of acute myocardial ischemic injury. Ann Non-invasive Electrocardiol. 2013 October 7. [Epub ahead of print] 6. Omland T, Pfeffer MA, Solomon SD, et al; PEACE investi-gators. Prognostic value of cardiac troponin I measured with a highly sensitive assay in patients with stable coronary artery disease. J Am Coll Cardiol. 2013;61:1240–1249. 7. Schmitt OH, Levine RB, Simonson E. Electrocardiographic mirror pattern studies. I: experimental validity tests of the dipole hypothesis and of the central terminal theory. Am Heart J. 1953;45:416–428. 8. Kléber ag, Janse MJ, van Capelle FJ, et al. Mechanism and time course of S-T and T-Q segment changes during acute regional myocardial ischemia in the pig heart determined by extracellular and intracellular recordings. Circ Res. 1978;42:603–613. 9. Tysler M, Svehlikova J. non-invasive finding of local repolarization changes in the heart using dipole models and simplified torso geometry. J Electrocardiol. 2013;46:284–288. 10. Schreck DM, Terndrup TE, Bosco SF, et al. Correlation of nonlinear mathematical transformation of the normal electro- cardiogram with the severity of coronary artery disease. Crit Care Med. 1989;17:269–273. 11. Schreck DM, Fishberg RD. Derivation of the 12-lead electrocardiogram and 3-lead vectorcardiogram. Am J Emerg Med. 2013;31:1183–1190. 12. Schreck DM, Ng L, Schreck BS, et al. Detection of coronary artery disease from the normal resting ECG using nonlinear mathematical transformation. Ann Emerg Med. 1988;17:132–134. 13. Schreck DM, Ng L, Schreck BS, et al. Nonlinear transformation of the resting electrocardiogram in the diagnosis of coronary artery disease. Ann Emerg Med. 1986;15:897–900. 14. Wagner GS, Macfarlane P, Wellens H, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part Vi: acute ischemia/infarction: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2009;119:e262–e270. 15. Tran V, Huang HD, Diez JG, et al. Differentiating Stelevation myocardial infarction from nonischemic ST elevation in patients with chest pain. Am J Cardiol. 2011;108:1096–1101. 16. Jayroe JB, Spodick DH, Nikus K, et al. Differentiating ST elevation myocardial infarction and nonischemic causes of St elevation by analyzing the presenting electrocardiogram. Am J Cardiol. 2009;103:301–306. 17. Writing committee, Jneid H, Anderson JL, Wright RS, et al. 2012 ACCF/ AHA focused update of the guideline for the management of patients with unstable angina/non-ST-elevation myocardial infarction (updating the 2007 guide-line and replacing the 2011 focused update): a report of the American College of Cardiology Foundation/American Heart As sociation Task Force on Practice Guidelines. Circulation. 2012;126:875–910. 18. Anderson CI, Harrison DG, Stack NC, et al. Evaluation of serial QRS changes during acute inferior myocardial infarction using a QRS scoring system. Am J Cardiol. 1983;52:252–256. 19. Anderson WD, Wagner nB, lee Kl, et al. Evaluation of a QRS scoring system for estimating myocardial infarct size. Vi: identification of screening criteria for non-acute myocardial infarcts. Am J Cardiol. 1988;61:729–733. 20. Hindman NB, Schocken DD, Widmann M, et al. Evaluation of a QRS scoring system for estimating myocardial infarct size. V. Specificity and method of application of the complete system. Am J Cardiol. 1985;55(13 Pt 1):1485–1490. 21. Palmeri St, Harrison Dg, Cobb FR, et al. A QRS scoring system for assessing left ventricular function after myocardial infarction. N Engl J Med. 1982;306:4–9. 22. Selvester RH, Wagner GS, Hindman NB. The Selvester QRS scoring system for estimating myocardial infarct size. The development and application of the system. Arch Intern Med. 1985;145:1877–1881. 23. Lippi G, Cervellin G. Choosing troponin immunoassays in a world of limited resources. J Am Coll Cardiol. 2013;62:647–648. JULY | 2020 JULY 2020

SPECIAL EDITION | ECGmax Products may not be available in all markets as product availability depends on the regulatory and/or medical processes in individual markets. For availability please contact info@corpuls.com. Printing errors as well as construction and design modification subject to change. All mentioned product names are registered trademarks of the respective owners. Article No. 76139.65021 Vers. 1.0 (07/20) Manufacturer: corpuls | GS Elektromedizinische Geräte G. Stemple GmbH Hauswiesenstrasse 26 | 86916 Kaufering | Germany Telephone +49 8191 65 722-0 E-mail Web info@corpuls.com www.corpuls.world D L R O W . S L U P R O C . W W W