Influence of Prone Positioning on Electrocardiogram in a Patient With COVID-19 | Cardiology | JAMA Internal Medicine | JAMA Network
[Skip to Navigation]
Views 12,924
Citations 0
Challenges in Clinical Electrocardiography
September 28, 2020

Influence of Prone Positioning on Electrocardiogram in a Patient With COVID-19

Author Affiliations
  • 1Rush University Medical Center, Chicago, Illinois
JAMA Intern Med. 2020;180(11):1521-1523. doi:10.1001/jamainternmed.2020.3818

A woman in her 50s with morbid obesity and obstructive sleep apnea presented with 3 days of fever with a temperature up to 39 °C, anosmia, and shortness of breath. Physical examination findings were significant for oxygen saturation of 85% on room air and coarse rales at the lung bases bilaterally. Chest radiography was suggestive of bilateral multifocal pneumonia. Results of a nasopharyngeal swab for coronavirus disease 2019 (COVID-19) were positive by reverse transcriptase–polymerase chain reaction. She was admitted to the intensive care unit for management.

Upon admission, the patient began hydroxychloroquine and azithromycin combination therapy. On hospitalization day 2, the patient developed acute respiratory distress syndrome (ARDS) requiring intubation and mechanical ventilation. Management strategies for refractory hypoxia were not successful until she was placed in a prone position. She was maintained in a prone position for 16 hours followed by supine positioning as tolerated for 4 hours a day. Given the treatment with hydroxychloroquine and azithromycin, serial electrocardiograms (ECGs) were performed to monitor QT interval during treatment. The ECG on hospitalization day 3 (Figure 1A) showed acute changes in comparison to her prior ECG (Figure 1B).

Figure 1.  12-Lead Electrocardiogram (ECG) and Vectorcardiogram Loops in the Prone and Supine Positions
12-Lead Electrocardiogram (ECG) and Vectorcardiogram Loops in the Prone and Supine Positions

A, A prone ECG during normal sinus rhythm with normal frontal QRS axis. The precordial leads were of low amplitude, and Q waves are present in leads V1 to V3. B, Normal ECG in the supine position in the same patient. C and D, The relationship between vector loops and surface ECG is demonstrated. The QRS vector loops initiate from the origin of the x-y-z axes. Arrows indicate sequential 2-ms intervals during ventricular depolarization. The QRS amplitude on surface lead ECG correlates to the perpendicular projection of the vector loop at a given position and time. The horizontal vector loop in C shows how ventricular depolarization is initially directed anteriorly and leftward away from leads V1 to V3, which are placed on the patient’s back in the prone position, resulting in QS appearance and/or low amplitude R waves. In contrast, the horizontal vector loop in D, obtained in the supine position, shows that the initial vectors are positive in leads V1 to V3 with ECG leads placed anteriorly on the chest.

Question: What is the cause of the Q waves and low amplitudes seen in the anterior precordial leads?


The ECG showed normal sinus rhythm. The frontal QRS axis was normal. The precordial leads showed low amplitudes and prominent Q waves in leads V1 to V3, which was interpreted as an anteroseptal infarct. However, the technician who performed the study confirmed unconventional ECG lead placement, with leads placed on the patient’s back, which was thought to be causing the ECG changes.

Clinical Course

No intervention was taken in respect to the interpretation of anteroseptal infarct. Subsequent ECGs with posterior lead placement were annotated to alert the reader to the expected ECG changes. The ECGs recorded when the patient was supine remained completely normal.

The patient gradually improved to the point where she no longer needed prone positioning. She was able to transition to noninvasive ventilation on hospitalization day 30 and subsequently to room air on hospitalization day 45. The patient’s course was also complicated by acute kidney failure requiring hemodialysis. However, her kidney function gradually improved and completely recovered on hospitalization day 25. The patient was discharged to a long-term care facility on hospitalization day 60 to continue her physical and occupational rehabilitation. She was expected to have a full recovery.


Currently, some of the drugs being evaluated as candidate therapies for COVID-19 may prolong the QT interval and potentially cause torsades de pointes.1 Hydroxychloroquine and chloroquine have also been associated with conduction disorders, widened QRS duration and T-wave inversion on ECG, myocardial toxic effects, and exacerbation of underlying myocardial dysfunction.2,3 Therefore, hospitals have been implementing ECG-based monitoring protocols whenever these agents are used for treatment of patients with COVID-19. However, severely ill patients placed in prolonged prone positioning to treat refractory hypoxemia secondary to acute respiratory distress syndrome often will have ECG leads placed on their back side instead, directly posterior to lead placement in the traditional supine position (Figure 2).4 This may affect the interpretation of the ECGs, especially when lead placement is not annotated.

Figure 2.  Standard Anterior and Prone Lead Placement
Standard Anterior and Prone Lead Placement

A, Typical anterior chest lead placement seen from the front. The patient’s right side is on the left of the image. B, Posterior lead placement in the prone position. The patient’s right side is on the right of the image. The leads are placed on the back directly opposite of their typical anterior position. The limb leads retain their left-right direction. Lead V1 is to the right of the spine. Leads V2 to V6 are to the left of the spine.

The expected changes from the combination of prone positioning and posterior lead placement can be intuitively visualized using vectorcardiograms. The vector loop traces the pathway of the action potential (ie, magnitude and direction) during the cardiac cycle. For example, the ventricular activation first spreads from the left bundle branch to the interventricular septum and then from the left and right bundle branches to the left and right ventricular walls. In the frontal plane, the relationship between the vector loops and the limb leads on surface ECG do not change significantly between supine and prone positions despite differences in lead placement.

Of note, the position of the heart in the thorax is relatively fixed at the base but tends to be free at the apex, so the apex will move with body position change or the diaphragm’s movement, resulting in vector changes.5 The position of the heart is relatively horizontal when supine, with the ventricular septum almost parallel to the frontal plane.5 When patients are prone, the heart falls forward (ventrally), abutting the anterior chest wall, and moves caudally.6 It also assumes a more globular shape.6 Viewed from the back side, the heart is farther away from the posteriorly placed leads. These findings may explain the precordial changes observed on the surface ECG with prone positioning. The more distant position of the heart coupled with the greater impedance from more soft tissue and bone in between the displaced myocardium and the electrodes may contribute to the low amplitudes in the precordial recordings. The problem may be exacerbated by the diseased lungs. In the horizontal plane, initial forces are directed anteriorly away from leads positioned on the back, explaining the appearance of Q waves on the surface ECG in the prone position (Figure 1C and D).

To confirm whether the findings of this patient’s case applied to a larger sample, we reviewed prone and traditional supine ECGs from 20 randomly selected patients with COVID-19 who had both performed during their index hospitalization. A comparison of the mean amplitude and waveform characteristics for each lead between the prone and traditional supine ECG showed no significant differences in limb lead waveform characteristics, except for lower overall Q/S wave amplitude in lead aVR. Moreover, there was a significant decrease in mean QRS amplitude in leads V1 to V5 and reduction in R-wave amplitude in leads V1 to V4 on prone ECGs. When ECGs were obtained with patients in the prone position, the precordial leads often demonstrated miniscule P waves and prominent Q waves in leads V1 to V3, which consistently led to interpretation as an anteroseptal infarct, as seen in this patient’s case.

Take-Home Points

  • Body position and ECG lead placement greatly affect ECG tracings and their interpretation; clear annotation of unconventional lead placement should be made.

  • It is critically important to discern a prone ECG from a supine ECG to avoid misinterpretations that could lead to potentially costly and fruitless investigations.

  • Vectorcardiograms are a useful adjunctive tool that provide intuitive and visual confirmation of posterior lead placement, thereby aiding in recognition of a prone ECG.

Back to top
Article Information

Corresponding Author: Henry D. Huang, MD, Rush University Medical Center, 1653 W Congress Pkwy, Chicago, IL 60612 (

Published Online: September 28, 2020. doi:10.1001/jamainternmed.2020.3818

Conflict of Interest Disclosures: Dr Nguyen reported other relationships or activities unrelated to the present work with Medtronic, Boston Scientific, and Medical Device Business Services. Dr Trohman reported grants from Medtronic and Boston Scientific; grants from and other relationships or activities unrelated to the present work with Abbott; and personal fees from Daiichi Sankyo and Alta Thera Pharmaceuticals. Dr Huang reported grants from Medtronic and consultant fees from CardioFocus.

Additional Contributions: We thank the people described and pictured for granting permission to publish this information.

Giudicessi  JR, Noseworthy  PA, Friedman  PA, Ackerman  MJ.  Urgent guidance for navigating and circumventing the QTc-prolonging and torsadogenic potential of possible pharmacotherapies for coronavirus disease 19 (COVID-19).   Mayo Clin Proc. 2020;95(6):1213-1221. doi:10.1016/j.mayocp.2020.03.024PubMedGoogle ScholarCrossref
Page  RL  II, O’Bryant  CL, Cheng  D,  et al; American Heart Association Clinical Pharmacology and Heart Failure and Transplantation Committees of the Council on Clinical Cardiology; Council on Cardiovascular Surgery and Anesthesia; Council on Cardiovascular and Stroke Nursing; and Council on Quality of Care and Outcomes Research.  Drugs that may cause or exacerbate heart failure: a scientific statement from the American Heart Association.   Circulation. 2016;134(6):e32-e69. doi:10.1161/CIR.0000000000000426PubMedGoogle ScholarCrossref
Naksuk  N, Lazar  S, Peeraphatdit  TB.  Cardiac safety of off-label COVID-19 drug therapy: a review and proposed monitoring protocol.   Eur Heart J Acute Cardiovasc Care. 2020;9(3):215-221. doi:10.1177/2048872620922784PubMedGoogle ScholarCrossref
Russi  CS, Myers  LA, Kolb  LJ,  et al.  Prehospital diagnosis of ST-segment elevation myocardial infarction using an “all-posterior” 12-lead electrocardiogram.   Prehosp Emerg Care. 2011;15(3):410-413. doi:10.3109/10903127.2011.561414PubMedGoogle ScholarCrossref
Pan  H, Xu  Z, Yan  H,  et al.  Lying position classification based on ECG waveform and random forest during sleep in healthy people.   Biomed Eng Online. 2018;17(1):116. doi:10.1186/s12938-018-0548-7PubMedGoogle ScholarCrossref
Ball  WS, Wicks  JD, Mettler  FA  Jr.  Prone-supine change in organ position: CT demonstration.   AJR Am J Roentgenol. 1980;135(4):815-820. doi:10.2214/ajr.135.4.815PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words