Article, Cardiology

Agreement between actual and synthesized right-sided and posterior electrocardiographic leads in identifying ischemia

a b s t r a c t

Objective: A 12-lead electrocardiogram is the standard of care for chest pain patients. However, 12-lead ECGs have difficulty detecting ischemia of the right ventricle or posterior wall of the heart. New technology exists to mathematically synthesize these leads from a 12-lead ECG; however, this technology has not been evaluated in the emergency department (ED). We assessed the level of agreement between synthesized 18-lead ECGs and actual 18-lead ECGs in identifying ST elevations, ST depressions, and T wave inversions in ED patients.

Methods: Actual 12- and 18-lead ECGs were acquired and synthesized 18-lead ECGs were produced based on waveforms from 12-lead ECGs. A blinded cardiologist interpreted the actual and synthesized 18-lead ECGs to identify the presence of abnormalities. Using actual 18-lead ECGs as the reference, the sensitivity, specificity, pos- itive predictive value (PPV), negative predictive value (NPV), and kappa of synthesized 18-lead ECGs in identify- ing abnormalities were determined.

Results: Data from 295 patients were analyzed. There was 100% agreement between synthesized 18-lead ECGs and actual 18-lead ECGs in identifying ST elevations and ST depressions (sensitivity, specificity, PPV, and NPV of 100%, and kappa of 1.00). Synthesized 18-lead ECGs had 95% sensitivity, 80% specificity, 97% PPV, and 70% NPV in identifying T wave inversions, when compared with actual 18-lead ECGs (kappa: 0.70).

Conclusion: Synthesized 18-lead ECGs demonstrated 100% agreement with actual 18-lead ECGs in the identifica- tion of ST elevations and ST depressions and good agreement in the identification of T wave inversions in a sam- ple of patients ED patients with complaints suspicious of Cardiac origin.

(C) 2019

Introduction

A 12-lead electrocardiogram (ECG) is considered the standard of care in the evaluation of acute chest pain or Suspected acute coronary syndrome [1]. A 12-lead ECG is non-invasive, can be acquired relatively quickly, and records electrical activity from 12 angles of the heart. Al- though information from the inferior, lateral, anterior, and septal re- gions of the heart is shown on a 12-lead ECG, ischemia affecting the right ventricle or posterior wall of the heart is more difficult to observe from the standard 12-lead ECG because none of the standard leads di- rectly represent these areas [2].

When right ventricular or posterior wall ischemia are suspected, right-sided (V3R, V4R, and V5R) and posterior (V7, V8, and V9) leads should be obtained [1,3]. In order to obtain these right-sided and poste- rior leads with a conventional 12-lead ECG machine, manipulation of the electrodes is necessary [3-5]; the electrodes must be repositioned

* Corresponding author at: 300 Community Drive, Manhasset, NY 11030, United States.

E-mail address: [email protected] (T. Li).

on the patient’s right side of the chest and left side of the back, which re- quires the patient to be in a right lateral recumbent position or upright position leaning forward. This may be uncomfortable for patients with chest pain and is a relatively time-consuming procedure in the emer- gency department (ED) setting [4]. The person acquiring the ECG must also keep track of the numerous ECG cables and manually re- label the leads on the ECG printout.

New technology has been developed that can mathematically esti- mate and create synthesized right-sided (V3R, V4R, V5R) and posterior (V7, V8, V9) leads, based on the waveforms from a standard 12-lead ECG [6]. The technology uses a mathematical algorithm to derive the three right-sided and three posterior lead waveforms based on wave- forms from the standard 12 leads, producing an 18-lead ECG (standard 12-leads with leads V3R, V4R, V5R, V7, V8, and V9). Therefore, the tech- nology produces an 18-lead ECG that does not require the patient to be in an uncomfortable position and does not require manipulation of the electrodes on a standard 12-lead ECG machine. This technology is par- ticularly useful in the evaluation of patients with acute chest pain or pa- tients with suspected cardiac ischemic conditions in the ED setting

https://doi.org/10.1016/j.ajem.2019.10.044

0735-6757/(C) 2019

because the acquisition procedure is the same as the standard 12-lead ECG procedure.

Although the technology is able to synthesize right-sided and poste- rior lead waveforms from standard 12-lead waveforms, the level of agreement between the mathematically synthesized leads and actual right-sided/posterior leads has not been evaluated on patients in the ED setting. Therefore, the primary objective of this study was to deter- mine the level of agreement between synthesized and actual right- sided and posterior leads in the identification of ischemia in a sample of ED patients with complaints that are suspicious of cardiac origin. The secondary objective was to determine the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of synthesized 18-lead ECGs in the identification of right ventricular and/or posterior wall ischemia, using actual 18-lead ECGs as the refer- ence for comparison.

Materials and methods

Study design and population

We conducted a single-center prospective study of a convenience sample of patients arriving to our ED at North Shore University Hospital in Manhasset, New York between February 2018 and August 2018. Part of the Northwell Health system, North Shore University Hospital is a 738-bed tertiary care facility with an ED that cares for nearly 90,000 pa- tients per year. The Northwell Health system’s Institutional Review Board approved the conduct of this study. This study was registered with clinicaltrials.gov with identifier NCT03426436.

We enrolled all English-speaking patients age >=18 years with a chief

complaint of either chest pain, chest pressure, or chest discomfort, had a troponin test ordered by their ED provider, and had decisional capacity to provide informed consent. We excluded patients with known ST ele- vation myocardial infarction (STEMI), patients without a troponin test ordered, pregnant patients, and prisoners. Patients with known STEMI were not approached for participation in this study due to potential in- terference with timely clinical care. Patients without a troponin test or- dered indicates that the ED provider was not suspicious that the symptoms were of cardiac origin and thus, these patients were not approached for participation. We did not approach pregnant patients because acquisition of the actual 18-lead ECG may be uncomfortable for these patients. Lastly, we did not approach prisoners due to potential interference with law enforcement officers’ duties.

Study protocol

Trained research assistants monitored our hospital’s electronic med- ical record system to track all patients in the ED. Patients appearing to meet eligibility criteria were approached by a research assistant after obtaining permission from the patient’s provider to approach the pa- tient. The research assistant confirmed eligibility and obtained informed consent.

After written informed consent was obtained, a trained ED techni- cian used the ECG machine provided by Nihon Kohden Corporation to acquire two consecutive 15-lead ECGs. The first 15-lead ECG was a stan- dard 12-lead ECG with right sided-leads (V3R, V4R, and V5R) and the second 15-lead ECG was a standard 12-lead ECG with posterior leads (V7, V8, and V9). These two research 15-lead ECGs were printed and shown to the patient’s provider, and any further clinical testing was left to the discretion of the patient’s provider. These two 15-lead ECGs are collectively termed “actual 18-lead ECGs” in this study.

Following the patient’s ED visit, we performed a medical record re- view to obtain demographic information, past medical history, present- ing signs and symptoms, test results, final diagnoses, and ED disposition information. Data abstracted from the medical record were directly en- tered into a Research Electronic Data Capture (REDCap) database. At the conclusion of the study enrollment period, we used proprietary

software from Nihon Kohden Corporation to generate “synthesized 18-lead ECGs” for each patient based on the 12-lead ECG waveforms. The study coordinator labeled actual and synthesized 18-lead ECGs from all enrolled patients with a random code. After all ECGs were la- beled with a random code, the ECGs were sorted in numeric order to avoid actual and synthesized ECGs from the same patient being in se- quential order.

The actual and synthesized 18-lead ECGs for all enrolled patients were given to a cardiologist, who was not involved with data collection. The cardiologist was blinded to all clinical information regarding the pa- tients and the cardiologist was unaware of which ECGs were actual vs. synthesized. The cardiologist was asked to identify the presence or ab- sence of ST elevations, ST depressions, and T wave inversions on each of the 18 leads.

Outcome measures

The primary outcome measure of this study was the level of agree- ment between synthesized and actual right-sided (V3R, V4R, V5R) and posterior (V7, V8, V9) leads. The actual ECG leads were used as the ref- erence to which the synthesized leads were compared. If the cardiolo- gist indicated that ST elevations, ST depressions, or T wave inversions were present on a lead, that lead was considered “positive”. Likewise, if the cardiologist indicated that ST elevations, ST depressions, and T wave inversions were all absent on a lead, that lead was considered “negative”.

The secondary outcome measures of this study were the sensitivity, specificity, PPV, and NPV of synthesized 18-lead ECGs in the identifica- tion of ischemia (ST elevations, ST depressions, and T wave inversions), using actual 18-lead ECGs as the reference for comparison. An ECG was considered “positive” if the cardiologist indicated the presence of either ST elevations, ST depressions, or T wave inversions on any of the 18 leads. An ECG was considered “negative” if the cardiologist indicated the absence of ST elevations, ST depressions, and T wave inversions on all 18 leads.

Data integrity

Explicit safeguards were enacted to ensure data integrity. Employees of Nihon Kohden Corporation only conducted training sessions with ED technicians on how to operate the ECG machine used for this study and assisted with troubleshooting technical issues that aroused in operating the machine. Patient identification and consent, ECG acquisition, and data collection, entry, analysis, and interpretation were all performed independently by the study team. Nihon Kohden Corporation em- ployees and LBB did not have access to patient data and did not play a role in the conduct of the study.

Statistical analysis

We used descriptive statistics to characterize the study sample, in terms of demographics, presenting signs and symptoms, past cardiac medical history, Laboratory test results, cardiac tests ordered, and dispo- sition. To address our primary objective, we calculated individual Cohen’s kappa coefficients and 95% confidence intervals (CIs) for each of the six leads (V3R, V4R, V5R, V7, V8, and V9) to assess the level of agreement between actual and synthesized leads in the identification of ST elevation, ST depression, and T wave inversion. We used guidelines from Altman et al. to interpret the kappa coefficients: values b0.20 indi- cate poor agreement; values between 0.21 and 0.40 indicate fair agree- ment; values between 0.41 and 0.60 indicate moderate agreement; values between 0.61 and 0.80 indicate good agreement; and values be- tween 0.81 and 1.00 indicate very good agreement [7]. To address our secondary objective, we determined the sensitivity, specificity, PPV, and NPV of synthesized 18-lead ECGs in the identification ST elevation, ST depression, and T wave inversion, using actual 18-lead ECGs as the

reference for comparison. We also calculated Cohen’s kappa coeffi- cients, with 95% CIs, to assess the level of agreement between synthe- sized and actual 18-lead ECGs in the identification of ST elevation, ST depression, and T wave inversion.

Results

During the study enrollment period, a total of 627 patients met the basic inclusion criteria of age >=18 years and having an ED chief com- plaint of chest pain, chest pressure, or chest discomfort (Fig. 1). Of these patients, 229 (36.5%) were not approached and 398 (63.5%) were approached by a research assistant for participation in the study. Of the 398 patients approached, 314 (78.9%) were enrolled, 56 (14.1%) refused to participate, 16 (4.0%) were discharged prior to completing study procedures, and 12 (3.0%) were determined to be ineligible. Of the 314 enrolled patients, 19 (6.1%) were excluded from the analysis due to various reasons detailed in Fig. 1. Therefore, our analytic sample comprised 295 patients.

Characteristics of the study sample are presented in Table 1. Mean age was 59 years and 50.9% were female. The most common symptoms experienced by patients were chest pain (85.8%), shortness of breath (40.7%), and chest pressure (36.3%), and 7.8% of patients had a final di- agnosis of acute coronary syndrome. Most patients were either discharged from the ED (38.3%) or admitted to a hospital floor (34.2%), but 2.0% were admitted to the intensive care unit. Cardiac tests ordered and relevant laboratory results are shown in Table 2; 7.5% of patients had positive troponin levels, and 35.3% and 16.6% of pa- tients underwent computed tomography angiography and stress test- ing, respectively.

Individual Cohen’s kappa coefficients between synthesized and ac- tual posterior and right-sided leads are presented in Table 3. There were no patients with ST elevation in lead V9; otherwise, there was 100% agreement between synthesized and actual leads in the

Fig. 1. subject enrollment.

Table 1

Characteristics of the study sample (n = 295).

Characteristic n (%)

Age (years)

Mean (+-standard deviation) 59 (+-15)

Sex

Male, n (%) 145 (49.2%)

Female, n (%) 150 (50.9%)

Symptoms (not mutually exclusive)

Chest pain, n (%) 253 (85.8%)

Shortness of breath, n (%) 120 (40.7%)

Chest pressure, n (%) 107 (36.3%)

Chest discomfort, n (%) 60 (20.3%)

Nausea, n (%) 55 (18.6%)

Dizziness, n (%) 33 (11.2%)

Chest tightness, n (%) 37 (12.5%)

Vomiting, n (%) 16 (5.4%)

Abdominal pain, n (%) 11 (3.7%)

Chest pain quality (not mutually exclusive)

Radiating, n (%) 103 (34.9%)

Intermittent, n (%) 101 (34.2%)

Constant, n (%) 54 (18.3%)

Sharp, n (%) 52 (17.6%)

Dull, n (%) 13 (4.4%)

Smoking status

Never, n (%) 153 (51.9%)

Former, n (%) 69 (23.4%)

Current, n (%) 35 (11.9%)

Unknown, n (%) 38 (12.9%)

History of hypertension

Yes, n (%) 158 (53.6%)

No, n (%) 137 (46.4%)

History of diabetes mellitus

Yes, n (%) 61 (20.7%)

No, n (%) 234 (79.3%)

History of hyperlipidemia

Yes, n (%) 125 (42.4%)

No, n (%) 170 (57.6%)

Pacemaker

Yes, n (%) 15 (5.1%)

No, n (%) 280 (94.9%)

Acute coronary syndrome diagnosis

Yes, n (%) 23 (7.8%)

No, n (%) 272 (92.2%)

Emergency department disposition

Discharged, n (%)

113 (38.3%)

Admitted to floor, n (%)

101(34.2%)

Admitted to clinical decision unit, n (%)

75 (25.4%)

Admitted to intensive care unit, n (%)

6 (2.0%)

identification of ST elevation (kappa coefficients of 1.00 for leads V7, V8, V3R, V4R, and V5R). There were no patients with ST depression in the right-sided leads; otherwise, kappa coefficients between synthe- sized and actual leads in the identification of ST depression ranged from 0.45 to 0.74. Similarly, kappa coefficients between synthesized and actual leads in the identification of T wave inversion ranged from

0.42 to 0.78.

As shown in Table 4, the reference actual 18-lead ECGs showed ST el- evations in four patients, and the ST elevations were also observed in the synthesized 18-lead ECGs. Synthesized 18-lead ECGs demonstrated 100% sensitivity, specificity, PPV, and NPV in the identification of ST el- evation, when using actual 18-lead ECGs as the gold standard (kappa: 1.00; 95% CI: 1.00, 1.00). A total of 21 patients had ST depressions in both the synthesized and actual 18-lead ECGs; synthesized 18-lead ECGs demonstrated 100% sensitivity, specificity, PPV, and NPV in the identification of ST depression, when using actual 18-lead ECGs as the gold standard (kappa: 1.00; 95% CI: 1.00, 1.00). However, the actual 18-lead ECGs showed T wave inversions in 14 patients and these T wave inversions were not observed in the synthesized 18-lead ECGs; T wave inversions were also not observed in the actual 18-lead ECGs for 8 patients, but were observed in the synthesized 18-lead ECGs. There- fore, synthesized 18-lead ECGs had 95% sensitivity, 80% specificity, 97% PPV, and 70% NPV in the identification of T wave inversion, when

Table 2

Cardiac and Laboratory Tests Ordered.

Test n (%)

Troponin

Positive, n (%) 22 (7.5%)

Not positive, n (%) 273 (92.5%)

Computed tomography angiography

Ordered, n (%) 104 (35.3%)

Not ordered, n (%) 191 (64.8%)

stress echocardiography

Ordered, n (%) 3 (1.0%)

Not ordered, n (%) 292 (99.0%)

Stress test

Ordered, n (%) 49 (16.6%)

Not ordered, n (%) 246 (83.4%)

Stress test type (n = 49)

exercise ECG, n (%) 3 (6.1%)

Exercise Echocardiography, n (%) 2 (4.1%)

Exercise nuclear, n (%) 21 (42.9%)

Pharmacologic nuclear, n (%) 23 (46.9%) cardiac ultrasound

Ordered, n (%) 49 (16.6%)

Not ordered, n (%) 246 (83.4%)

Cardiac catheterization

Ordered, n (%) 45 (15.3%)

Not ordered, n (%) 250 (84.8%)

Ejection fraction (n = 47)

Median (IQR) 63 (55, 70)

Creatine kinase (n = 112)

Median (25th, 75th percentile) 105 (68, 164) Cholesterol (n = 92)

Median (25th, 75th percentile)

169 (139, 198)

High density lipoprotein (n = 92)

Median (25th, 75th percentile)

50 (37, 62)

Brain natriuretic peptide (n = 88)

Median (25th, 75th percentile)

206 (56, 895)

compared with actual 18-lead ECGs, and the associated kappa coeffi- cient was 0.70 (95% CI: 0.58, 0.82).

Discussion

In this study of patients presenting to the emergency department with chief complaints suspicious of cardiac origin, we found that there was 100% agreement between synthesized and actual right-sided and posterior leads in the identification of ST elevations (kappa coefficients of 1.00). There was also moderate to good agreement between synthe- sized and actual right-sided and posterior leads in the identification of ST depression and T wave inversion (kappa coefficients ranging from 0.42 to 0.78). When synthesized 18-lead ECGs were interpreted as a whole, 18-lead ECGs had a sensitivity, specificity, PPV, and NPV of 100%, compared with actual 18-lead ECGs, in the identification of ST el- evation and ST depression. However, classification performance of syn- thesized 18-lead ECGs was lower in the identification of T wave inversion (sensitivity: 95%; specificity: 90%; PPV: 97%; NPV: 70%; kappa: 0.70). ST elevations and ST depressions are generally considered to be more clinically significant than T wave inversions. Therefore, the 100% agreement between synthesized and actual 18-lead ECGs in the identification of ST elevations and ST depressions is reassuring. How- ever, 22 out of 295 (7.5%) of the 18-lead ECGs were discordant with re- spect to identifying T wave inversions.

In our study, 12-lead ECG waveforms were used to mathematically derive waveforms of the right chest and posterior leads, producing syn- thesized 18-lead ECGs. The technology continuously measures instanta- neous cardioelectric vectors from the standard 12-lead ECG data and synthesizes six extra leads from the data. Tamura et al. assessed the re- liability of ST-segment abnormalities in right-sided leads by the synthe- sis technology [6]. They evaluated the technology after coronary balloon inflations during Percutaneous coronary interventions and observed ex- cellent correlation (r = 0.96) between the actual and synthesized leads

[6]. The study by Katoh et al. assessed the correlation between the actual and synthesized leads in 30 patients admitted to the coronary care unit [4]. They investigated the clinical significance in ST elevation and ob- served correlations of 0.87 to 0.98 in leads V3R to V5R and V7 to V9 [4]. To the best of our knowledge, our study is the first study demon- strating moderate to perfect agreement between the actual and synthe- sized leads in the ED setting.

Our findings indicate that synthesized 18-lead ECGs can be used as an alternative to actual 18-lead ECGs to identify ST elevations and ST de- pressions in the ED. Our study protocol mimicked the actual clinical workflow in the ED, where an initial ECG is obtained by an ED technician and then interpreted by a physician. We relied on a cardiologist to iden- tify the presence or absence of ST elevations, ST depressions, and T wave inversions on the ECGs in this study, rather than relying on a mechanical algorithm. Instead of using mechanical codes such as the Minnesota Code Classification to determine ECG abnormalities, our study protocol followed the actual clinical workflow and therefore, our methods and analysis will be reproducible in actual clinical settings, as manual ECG interpretation is commonplace in clinical practice. In addition, the ECGs were reviewed in an unbiased and objective fashion, where the study cardiologist was blinded to patients’ symptoms and clinical ap- pearances. Given that our evaluation of the ECGs involved a physician, the 100% agreement in the identification of ST elevations and ST depres- sions is promising, and these findings can yield high fidelity in the de- tection of myocardial infarction in the ED setting when the technology is applied.

ST elevation in right-sided leads due to Right ventricular infarction is clinically important to identify [8,9]. The study by Ashida et al. examined whether synthesized leads could improve the diagnostic accuracy of in- farction site diagnosis among STEMI patients [5]. The authors found that synthesized 18-lead ECGs were useful in identifying the infarction site in patients with infarction affecting the right ventricular wall or poste- rior wall of the left ventricular, which the standard 12-lead ECG often fails to show [5]. In our study, the number of patients with ST elevation was limited due to the exclusion of STEMI patients; therefore, further studies are warranted to assess the reliability of synthesized 18-lead ECGs among STEMI patients, particularly in patients with right ventricle myocardial infarction from the right coronary Arterial occlusion, which may concomitantly occur with inferior left ventricle myocardial infarc- tion [10].

Although we found 100% agreement between synthesized and ac- tual 18-lead ECGs in the identification of ST elevations and ST depres- sion, we found lower level of agreement in the identification of T wave inversions. We reviewed the 22 patients in which the synthesized and actual 18-lead ECGs were discordant and found that flattened T waves more frequently occurred in right-sided or posterior leads than the other standard 12 leads. These flattened T waves were generally ac- companied by low amplitude R waves and these might have interfered with the cardiologist’s review of T wave inversions. Chest leads includ- ing right-sided and posterior leads use a theoretical reference point lo- cated approximately in the center of thorax, termed Wilson’s central terminal (WCT), to generate an electrical amplitude for each of the

Table 3

Cohen’s Kappa Coefficients between Synthesized and Actual Posterior and Right-Sided Leads (n = 295).

Lead

ST Elevation

ST Depression

T Wave Inversion

Kappa (95% CI)

Kappa (95% CI)

Kappa (95% CI)

V7

1.00 (1.00, 1.00)

0.70 (0.45, 0.95)

0.77 (0.64, 0.91)

V8

1.00 (1.00, 1.00)

0.74 (0.50, 0.99)

0.58 (0.43, 0.73)

V9

No positives

0.45 (0.11, 0.79)

0.42 (0.27, 0.56)

V3R

1.00 (1.00, 1.00)

No positives

0.75 (0.68, 0.83)

V4R

1.00 (1.00, 1.00)

No positives

0.77 (0.70, 0.84)

V5R

1.00 (1.00, 1.00)

No positives

0.78 (0.71, 0.85)

Note: “No positives” indicate that there were no patients with positive ST elevation or ST depression in that lead.

CI: confidence interval

Table 4

Agreement between Synthesized and Actual 18-lead ECGs in the Identification of ST Elevation, ST Depression, and T Wave Inversion (n = 295).

ST Elevation Actual 18-lead ECG (n = 295) Sensitivity (95%

Specificity (95%

PPV (95% CI) NPV (95% CI) Kappa (95%

Positive (n = 4)

Negative (n = 291)

CI)

CI)

CI)

Synthesized 18-lead ECG

Positive (n = 4) 4 0 100% (100%,

100% (100%,

100% (100%,

100% (100%,

1.00 (1.00,

(n = 295)

Negative (n = 291)

0 291

100%)

100%)

100%)

100%)

1.00)

ST Depression Actual 18-lead ECG (n = 295) Sensitivity (95%

Specificity (95%

PPV (95% CI) NPV (95% CI) Kappa (95%

Positive

Negative

CI)

CI)

CI)

(n = 21)

(n = 274)

Synthesized 18-lead ECG

Positive

21

0

100% (100%,

100% (100%,

1.00 (1.00,

100% (100%,

1.00 (1.00,

(n = 295)

(n = 21)

Negative (n = 274)

0

274

100%)

100%)

1.00)

100%)

1.00)

T Wave Inversion

Actual 18-lead ECG

(n = 295)

Sensitivity (95%

Specificity (95%

PPV (95% CI)

NPV (95% CI)

Kappa (95%

Positive (n = 255)

Negative (n = 40)

CI)

CI)

CI)

Synthesized 18-lead ECG

Positive

241

8

95% (92%, 97%)

80% (68%, 92%)

97% (95%, 99%)

70% (56%, 83%)

0.70 (0.58,

(n = 295)

(n = 249)

Negative (n = 46)

14

32

0.82)

ECG: electrocardiogram; CI: confidence interval; PPV: positive predictive value; NPV: negative predictive value.

leads [11,12]. The amplitude is affected by the amount of the electrical activity that is measured between the WCT and the electrode. Therefore, it is plausible that the thin muscle layers of the right ventricle and the posterior wall may not generate enough amplitude, when it is measured with WCT that is optimized for the standard 12 leads. The optimization of WCT has been debated [11,12]. Therefore, future studies may wish to investigate how to maximize the amplitude of the right-sided and pos- terior leads and whether it affects the synthesized leads.

Findings of this study have substantial clinical implications. As previ- ously discussed, standard 12-lead ECGs have difficulty detecting ische- mia that affects the right ventricle and posterior wall of the heart and thus, 18-lead ECGs are recommended. However, the process of acquir- ing an 18-lead ECG is complicated; it requires the patient to be in an un- comfortable position and requires the person acquiring the ECG to keep track of numerous ECG cables and manually re-label the leads on the ECG printout. Therefore, technology that is able to mathematically syn- thesize an 18-lead ECG based on waveforms from a standard 12-lead ECG is highly valuable in the ED setting, where prompt evaluation of acute chest pain patients is important.

There are a few limitations of our study to acknowledge. First, our study sample comprised patients who were relatively stable, as we ex- cluded known STEMI patients due to potential interference with timely clinical care. Therefore, our study findings have limited generalizability to unstable patients with cardiac complaints. Future studies may wish to include more acute patients. Second, this study only enrolled patients in the emergency department and thus, results may not be generaliz- able to other clinical settings, such as cardiac care units or the prehospital ambulance setting. Third, we excluded pregnant patients, due to the potential discomfort associated with being in the right lateral recumbent or upright positions needed to obtain the actual 18-lead ECGs, which further limits our generalizability. Because myocardial ox- ygen demand during pregnancy is increased, pregnant patients may be an interesting study population for future studies.

Conclusion

We found that synthesized 18-lead ECGs demonstrated 100% agree- ment with actual 18-lead ECGs in the identification of ST elevations and ST depressions and good agreement in the identification of T wave

inversions in a sample of patients with emergency department patients with complaints suspicious of cardiac origin.

Declaration of Competing Interest

LBB is a compensated member of the Scientific Advisory Board of Ni- hon Kohden Corporation. All other authors report no relevant conflicts of interest.

Acknowledgements

We would like to acknowledge Nihon Kohden Corporation for funding this study and thank Kota Saeki and Takashi Kaiami for their technical assistance with this study.

Appendix A. Supplementary material

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2019.10.044.

References

  1. Amsterdam EA, Wenger NK, Brindis RG, et al. 2014 AHA/ACC guideline for the man- agement of patients with non-ST-elevation acute coronary syndromes: a report of the American College of Cardiology/American Heart Association Task Force on Prac- tice Guidelines. J Am Coll Cardiol 2014;64(24):e139-228.
  2. Rich MW, Imburgia M, King TR, Fischer KC, Kovach KL. Electrocardiographic diagno- sis of remote posterior wall myocardial infarction using unipolar posterior lead V9. Chest 1989;96(3):489-93.
  3. Provinse J, Harris C, Stauss M, Gallagher K, Evangelista-Hoffman E. Right-sided and posterior Electrocardiograms . emergency nurses Association. Available at: https://www.ena.org/docs/default-source/resource-library/practice-resources/tips/ right-side-ecg.pdf?sfvrsn=836f00e6_8 [accessed March 11, 2019].
  4. Katoh T, Ueno A, Tanaka K, Suto J, Wei D. Clinical significance of synthesized posterior/right-sided chest lead electrocardiograms in patients with acute chest pain. J Nippon Med Sch 2011;78(1):22-9.
  5. Ashida T, Tani S, Nagao K, Yagi T, Matsumoto N, Hirayama A. Usefulness of synthe- sized 18-lead electrocardiography in the diagnosis of ST-elevation myocardial in- farction: a pilot study. Am J Emerg Med 2017;35(3):448-57.
  6. Tamura A, Torigoe K, Goto Y, et al. Reliability of ST-segment shifts in the synthesized V(3)R-V(5)R leads after coronary balloon inflations during percutaneous coronary intervention. Am J Cardiol 2014;114(8):1187-91.
  7. Altman DG. Practical statistics for medical research. London, UK: Chapman & Hall; 1991.
  8. Zalenski RJ, Rydman RJ, Sloan EP, et al. Value of posterior and right ventricular leads in comparison to the standard 12-lead electrocardiogram in evaluation of ST- segment elevation in suspected acute myocardial infarction. Am J Cardiol 1997;79 (12):1579-85.
  9. Correale E, Battista R, Martone A, et al. Electrocardiographic patterns in acute inferior myocardial infarction with and without right ventricle involvement: classification, Diagnostic and prognostic value, masking effect. Clin Cardiol 1999;22(1):37-44.
  10. Wagner GS, Macfarlane P, Wellens H, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part VI: acute ische- mia/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. En- dorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol 2009;53(11):1003-11.

    Moeinzadeh H, Bifulco P, Cesarelli M, et al. Minimization of the Wilson’s Central Ter- minal voltage potential via a genetic algorithm. BMC Res Notes 2018;11(1):915.

  11. Gargiulo GD. True unipolar ECG machine for Wilson Central Terminal measure- ments. Biomed Res Int 2015;2015:586397.

Leave a Reply

Your email address will not be published. Required fields are marked *