Be aware of false positive treadmill traces

Sharon Jebamalar B¹, Selva Maheswari M¹, Usnish Adhikari², Deep Chandh Raja S³

¹Physician Assistant, Kauvery Heart Rhythm Services, Kauvery hospital, Alwarpet and Vadapalani

²Associate Consultant, Kauvery Heart Rhythm Services, Kauvery hospital, Alwarpet and Vadapalani

³Senior consultant and Director of Cardiac Electrophysiology, Kauvery Heart Rhythm Services, Kauvery hospital, Alwarpet and Vadapalani

Introduction

A treadmill test (TMT, also known as a Stress Test or Exercise ECG) is a non-invasive diagnostic procedure used to evaluate the heart’s response to physical stress, typically by measuring the electrical activity of the heart during exercise.¹ It was Einthoven in 1908 who documented changes in ST segment of ECG with exercise. In 1932 Goldhammer and Scherf proposed exercise electrocardiography as a diagnostic tool for angina. Since then, Exercise tolerance test (ETT) has played a central role in the diagnostic workup of coronary artery disease. This test is often performed to identify underlying heart conditions, such as coronary artery disease (CAD), arrhythmias, or other cardiac abnormalities^2,3

The primary objective of TMT is to increase the myocardial oxygen demand, thereby revealing any limitations in coronary blood flow. This reduced blood flow can result in myocardial ischemia, which is typically identified through electrocardiographic abnormalities, most commonly in the form of ST-segment changes.⁴Exercise increases myocardial oxygen consumption by elevating heart rate, intramyocardial tension, and the contraction velocity of the myocardium. Increases in heart rate are generally linked to a relatively linear rise in myocardial oxygen consumption, and the exercise heart rate serves as an easily monitored indicator of myocardial oxygen demands. ST segment depression, which occurs as a result of subendocardial myocardial ischemia, is caused by local alterations in the cellular membrane potential at rest and changes in the shape of the action potential. These modifications lead to a flow of current that results in TQ segment elevation and ST segment depression. Both of these changes are reflected on the body’s surface electrocardiogram (ECG) as ST depression.^5,6

However, there are instances of misinterpretation of TMT ECGs, as demonstrated in the following cases.

Key-words: Computerized treadmill test, Raw data, Linked medians, Synthesized ECG

Case Presentation

In this case series, two patients underwent TMT as part of their clinical evaluation. Both individuals were found to have structurally normal hearts on comprehensive cardiac assessment, including echocardiography.

Case 1

A 49-year-old female presented with a history of atypical chest pain and intermittent palpitations. The chest discomfort was non-exertional, not associated with dyspnea. She also had complaints of palpitations which was episodic, without clear precipitating factors. As part of her diagnostic evaluation, the patient underwent a TMT, to assess for inducible myocardial ischemia. The report of TMT stated “Wolf-Parkinson-White syndrome and significant exercise-induced ischemia”. A detailed review of the TMT tracings was undertaken to clarify these findings.

At Stage 2 of the TMT, the computerized ECG tracing (Figure 1A) demonstrated occasional premature ventricular complexes (PVCs), a pre-excitation pattern evident in leads II, III, aVF, and V3–V6, along with significant ST-segment depression in the same leads (II, III, aVF, V3–V6). In contrast, the raw ECG tracing from lead II at the same stage did not reveal any significant ST-T segment changes or evidence of pre-excitation, apart from occasional PVCs.

Further analysis of another computerized tracing at Stage 2 (Figure 1B) again showed occasional PVCs, no clear evidence of pre-excitation, and only minimal ST-T segment depression in the inferior and anterior leads (II, III, aVF, V3–V6). However, the corresponding raw ECG data consistently showed no significant ST-T abnormalities or pre-excitation patterns.

These discrepancies can be attributed to signal-processing artifacts in the computerized ECG system. Specifically, the treadmill system employed a filter-based signal-averaging algorithm that used a PVC beat as the reference template. This led to an averaged QRS morphology mimicking the PVC pattern, despite its absence in the actual rhythm strip, resulting in false-positive ST-segment depression and spurious conduction findings.

Figure (1A): Pre-excitation pattern in form of delta waves seen in the 12-leaed ECG. However, a close look at the Linked Median rhythm strip (Lead II ^$) at the bottom, reveals that the computer has generated a false pattern of dealt waves based on the PVCs noted.

Figure (1B): In the same patient, significant ST depressions are noted in the leads II, III, aVF, V5, V6 in the 12-lead ECG. This too is a false pattern as no true ST-depressions are noted in the Linked Median rhythm strip (Lead II ^$) at the bottom

Case 2

A 50-year-old male with a known history of dyslipidemia presented with complaints of significant exertional palpitations persisting over the past two months. Transthoracic echocardiography revealed normal left ventricular systolic function with no regional wall motion abnormalities. Baseline 12-lead electrocardiography (ECG) demonstrated normal sinus rhythm. As part of his diagnostic evaluation, the patient underwent a treadmill exercise test (TMT). The test revealed broad QRS complexes at peak exercise, and the result was initially reported as “exercise-induced ventricular tachycardia”. A detailed analysis of the TMT tracings was performed to clarify the findings.

The baseline TMT ECG (00:00 min) (Figure 2A) showed normal sinus rhythm without PVCs. At 3:00 minutes (Figure 2B), the computerized ECG continued to show sinus rhythm with no significant ST-T changes. However, the corresponding raw ECG tracing from lead II revealed occasional PVCs that were not visible in the averaged data.

At 4:32 minutes (Figure 2C), the computerized ECG displayed a “sustained ventricular tachycardia (VT)” with a rate of 140 bpm and a left bundle branch block (LBBB) morphology, along with ST-segment depression in the inferior leads. In contrast, the raw ECG tracing from lead II at the same time revealed a pattern consistent with ventricular bigeminy, rather than sustained VT.At 4:33 minutes (speed 0 km/h) (Figure 2D), the computerized tracing showed non-sustained ventricular tachycardia with LBBB morphology and associated ST-T depression in leads II, III, aVF, V5, and V6. However, the raw ECG tracing from lead II at that moment showed normal sinus rhythm with occasional PVCs only, without evidence of VT or significant ST changes.

These discrepancies can be attributed to the filter-based signal-averaging algorithm of the treadmill system, which had used a PVC as the reference beat. This resulted in an averaged QRS complex morphology resembling that of a PVC, erroneously interpreted as ventricular tachycardia with associated ST-segment depression, thus representing a false-positive finding.

Figure 2A: Normal sinus rhythm without PVC was reported based on this 12-lead ECG trace

Figure 2B: The computerized ECG continued to show sinus rhythm with no significant ST-T changes. However, the corresponding raw ECG tracing from (Lead II ^$) revealed occasional PVCs that were not visible in the averaged data

Figure 2C: “Ventricular tachycardia” was reported based on this 12-lead ECG trace. However, a close look at the Raw Rhythm strip (Lead II^$) at the bottom, reveals that the computer has generated a false pattern of delta waves based on the Ventricular bigeminy in the patient.

Figure 2D: In the same patient, Computerized ECG suggested NSVT with LBBB morphology and ST depression, but raw (Lead II^$) showed sinus rhythm with occasional PVCs, indicating likely algorithmic misinterpretation without true VT or ischemia.

Discussion

The above cases illustrate the potential for misinterpretation, that can arise from reliance on ‘computerized’ treadmill ECG interpretations, particularly in the context of signal-averaging artifacts. Although the initial analysis of the treadmill test suggested sustained and non-sustained ventricular tachycardia along with significant ST-segment depression—findings that would typically raise concern for serious arrhythmia or myocardial ischemia—closer inspection of the raw ECG data revealed a different picture. The patient’s actual rhythm was limited to occasional premature ventricular complexes and ventricular bigeminy, without true ST-T segment abnormalities.

The apparent discrepancies were attributed to the ECG system’s signal-averaging algorithm, which had erroneously used a PVC as the reference beat. This led to a distorted QRS complex morphology in the processed tracings, simulating ventricular tachycardia and associated ischemic changes. Such false-positive findings may result in misdiagnosis, unnecessary additional testing, and undue patient anxiety.

Other causes for positive TMT

ST segment depression, in the absence of myocardial ischemia, may result from changes in the action potential due to electrolyte imbalances, cardioactive medications, pericarditis, or non-ischemic myocardial diseases. Catecholamine and autonomic influences can affect the duration of repolarization, primarily impacting the T wave. However, these factors have not been shown to cause isolated ST segment abnormalities. The electrophysiologic mechanisms underlying stress-induced ST segment depression, in the absence of heart disease or other known factors that influence the action potential, remain unclear.^7,8

Positive vs false positive TMT

A positive test is characterized by greater than 1 mm of ST-segment depression occurring 80 milliseconds after the J point in any lead of the unprocessed, or “true,” 12-lead electrocardiogram. A false-positive test is defined by the presence of greater than 1 mm of ST-segment depression 80 milliseconds after the J point in any lead of the signal-averaged 12-lead electrocardiogram, with no corresponding abnormality evident on the raw ECG. A negative test is indicated by less than 1 mm of ST-segment depression 80 milliseconds after the J point in the true 12-lead electrocardiogram, provided that at least 90% of the target heart rate was achieved during the test. An indeterminate test refers to a negative test result in which less than 90% of the target heart rate was attained, thereby limiting diagnostic reliability.

Marked ST-segment depression can be observed during exercise testing even in the absence of angiographically confirmed coronary artery disease. These false-positive results frequently prompt additional, often unnecessary, diagnostic procedures and may contribute to significant patient anxiety. Less commonly recognized are false-positive findings generated by computerized electrocardiographic systems. In such cases, apparent ST-segment depression arises from signal-averaging algorithms, while the abnormality is not present on the unprocessed, or “raw,” ECG.⁹

Conclusion

Treadmill ‘constructs’ a 12-lead ECG, which is basically Medians of the Raw Data, which is conventionally mentioned at the bottom of the trace as ‘Linked Median* or *Raw Rhythm’. Despite the computerized report, one should always have the true or raw electrocardiogram to review before interpreting a test as positive, as most centers report these false-positive tests as abnormal. These cases emphasize the importance of critically evaluating treadmill ECG findings, especially when the results appear inconsistent with the patient’s clinical context and baseline investigations. Signal-averaging artifacts introduced by computerized ECG systems can mimic serious arrhythmias and ischemic changes, leading to false-positive interpretations. Thorough review of the raw ECG data is essential to avoid misdiagnosis, prevent unnecessary investigations, and reduce patient anxiety. Clinicians should remain vigilant for such technical limitations in exercise testing, particularly in patients with structurally normal hearts and low pre-test probability of cardiac disease.

References

Jay Furst, September 6, 2023 Stress test abnormalities reveal more than just cardiovascular risks, Mayo Clinic study finds
Abdul Wajid Khan Faisal, Abdul Rehman Abid, Muhammad Azhar, December 2004 Exercise tolerance test: a comparison between true positive and false positive test results
Ashley E A, Myers J, Froelicher V. Exercise testing in clinical medicine. Lancet 2000;356:1592–7.
James V. Faris, M.D, Paul L. McHenry, M.D, Stephen N. Morris, M.D. January 1978 Concepts and applications of treadmill exercise testing and the exercise electrocardiogram
Peter M. Sapin, MD, Gary Koch, PhD, Mary Beth Blauwet, BS, James J. Mccarthy, BS, Spencer W. Hinds, MD, Leonard S. Gettes, MD, FACC, July 1991 Identification of False Positive Exercise Tests With Use of Electrocardiographic Criteria: A Possible Role for Atrial Repolarization Waves, JACC
Holland RP, Brooks H. TQ-ST segment mapping: critical review and analysis of current concepts. Am J CardioI1977;40: 110-29.
Kuo CS, Surawicz B. Ventricular monophasic action potential changes associated with neurogenic T wave changes and isoproterenol administration in dogs. Am J Cardio 1976;38:170-7.
Yanowitz F, Preston JB, Abildskov JA. Functional distribution of right and left stellate innervation to the ventricles: production of neurogenic electrocardiographic changes by unilateral alteration of sympathetic tone.Circulation Res 1966;43:416-27.
John A. Milliken, MD, Hosiar Abdollah, MB, and Gary W. Burggraf, MD. False-Positive Treadmill Exercise Tests due to Computer Signal Averaging, The American Journal of Cardiology Volume 65.

Kauverian Bookshelf

Applied Medical Statistics

Be aware of false positive treadmill traces