By Dr. Payal Kohli, MD, FACC
An electrocardiogram (ECG) can instantly diagnose disease. The surface recording of the electrical forces of the heart can look into the future to risk stratify an athlete for sudden cardiac death and identify channelopathies. They can look into the past to identify a prior heart attack, or they can stay in the present to give an instantaneous picture of an ongoing cardiac arrhythmia or an electrolyte disturbance. Yet, although this tool seems simple, it requires close attention to detail in order to improve diagnostic accuracy.
It was a Monday afternoon consult for an "abnormal ECG" that made me acutely aware of the importance of recognizing the factors that influence ECG signal quality. The primary care provider had sent the patient over for "inferior Q waves." And yet, when I looked at the ECG in my office, I saw beautiful QRS complexes with R waves inferiorly and no suggestion or hint of a Q wave in sight.
It was impossible that the Q waves disappeared within the two weeks between the ECGs—the more likely possibility was lead reversal by the medical assistant or technician performing the ECG. In fact, lead reversals occur commonly in clinical practice, and according to an editorial published in JAMA Internal Medicine, they account for 0.4% to 4% of all performed ECGs.1 I was happy to tell the patient his ECG was normal and no further testing was indicated. That said, I was unable to shake the skeptical look on his face when I tried to explain the technical error.
Electrode Storage and Selection
I took the opportunity to explain the basics to my medical assistant, informing her that a cardiologist's office must have the highest-quality ECG signal quality in order to ensure the highest technical and diagnostic accuracy of the ECG. The beginning of a high-quality ECG acquisition starts with proper electrode storage and selection. For example, technicians should only open the package after completing patient prep, and they should apply electrodes immediately. A dry electrode can compromise the signal acquisition due to reduced conduction.
Similarly, an expired or damaged (due to exposure to excess light or heat) electrode can also compromise the data. Other important steps to ensure ECG signal quality include:
- Selecting electrodes carefully
- Using the correct manufacturer for your ECG machine
- Avoiding mixing of brand manufacturers on a single patient (as that can change the signal strength asymmetrically)
- Making sure not to cut or alter the electrode
- Using the appropriate electrodes for pediatric patients
Patient Preparation and Lead Placement
In addition to storage and selection, patient preparation and lead placement are also key steps to success. Following a standardized patient preparation routine can minimize motion and electrostatic artifact, and can even ensure the accuracy of the comparison of serial ECGs over time or between different healthcare facilities. Technicians should completely remove any hair, lotion, creams, or dead skin cells, and pay careful attention to lead placement to minimize errors.
Once I had reviewed the basics with my medical assistant, I decided to take a "deep dive" into other factors that may affect the accuracy of the ECGs in my office.
With myriad electrical or electromagnetic signals all around us in a world with increasing use of smartphones and wearable transmitting device technologies, electromagnetic signal interference is increasingly common. Such interference can lead to mistakes from the ECG algorithm as well as the diagnosing physician, and they can compromise the accuracy of the ECG and compromise ECG signal quality.
Turning off or removing devices that can cause electromagnetic signal interference can help address this problem. Signal conditioning or removal of the signals that display characteristics that could not be possibly generated by the heart (such as incorrect frequencies below 0.67 Hz) can also address this. Signal conditioning can help remove many types of noise, including AC interferences, baseline wander, high-frequency artifact, and upper cut-off frequency.
Analog Filters: When to Use or Avoid Them
An analog filter is automatically applied to attenuate high-frequency electrical noise that is not part of the physiologic signal. Without this filter, high-frequency signals could get erroneously digitized as physiologic signals. For the GE HealthCare ECG machine, it's called an anti-aliasing filter.
Removing AC Interference
This type of noise is usually continuous and sinusoidal and can occur as a result of devices that are powered by alternating currents (ACs). Selecting a filter frequency that matches the main frequency of the power grid (either 50 or 60 Hz) of the electrocardiogram eliminates this noise. This type of "notch filter" (a filter that eliminates a single frequency from a spectrum of frequencies) removes the amplitude and syncs the shape conferred by that frequency with the AC interference—which is continuous—so it does not attenuate the naturally occurring transient frequencies at 50 or 60 Hz.
Removing Baseline Wander
Baseline wander is a common problem in clinical practice that can occur due to loose electrodes, respiration, perspiration, body movements, or dry electrodes. Even slight degrees of baseline wander can create challenges for the accurate assessment of the ST-T wave. To correct this and increase ECG signal quality, the clinician can apply a "high-pass filter," which allows high frequencies to pass, with a low-frequency signal. To deploy this filter, the clinician must know the lowest possible frequency generated by the heart so as to not filter out any physiologic frequencies. This can be easily calculated based on heart rate (HR). For example, if the HR is 60 beats per minute (BPM), the lowest frequency is one cycle per second, or 1 Hz. Heart rates below 40 BPM (0.67 Hz) are uncommon, and therefore the American College of Cardiology (ACC) and the American Heart Association (AHA) recommend removing frequencies below 0.67 Hz.2
Take caution with using a high-pass filter because they aren't all alike! For example, some filters—especially aggressive ones—can actually shift the low frequencies in time, causing phase distortion and inaccurate signal acquisition. While removing the noise, the last thing you want to do is create noise. With the newer ECG machines that have a ZPD (Zero Phase Distortion) filter, there is less phase distortion and the high-pass filter setting can behave the same way for both the rhythm and 12-lead strips without ST-segment distortion. The 12SL filter has been able to remove low frequencies (<0.32 Hz) without ST-segment distortion by running the same filter forward and backward over the entire acquisition. To know if your electrocardiograph uses a high-pass filter with ZPD, a biomedical engineer can run a square wave and assess the extent of artifact.
Removing High-Frequency Artifact
On one particularly cold and snowy day in Denver, the heat in our clinic building had broken down. While waiting for the engineers to come and repair the heat, all our patients who received ECGs had high-frequency artifact due to shivering, compromising the accuracy of their ECGs. This type of artifact can include noise due to muscle tremor or electrode-motion artifact. Keep in mind, the lower the filter setting, the more aggressively the filter removes high-frequency signals. The low-pass filter (meaning it allows lower frequencies to pass) includes 40 Hz, 100 Hz or 150 Hz settings. The currently recommended settings are the full bandwidth of 150 Hz.
Remember to use caution and have similar filter settings, especially when comparing ECGs across time, as a very aggressive low-pass filter can attenuate the most peaked aspects (the highest frequencies) of the QRS complexes. Filter settings travel with the ECG, which means technicians can configure the ECG to remove frequencies or they can apply additional filters to the underlying signal at any time.
These two experiences—the simple lead reversal by my referring provider's medical assistant and the snowy day in Denver—taught me so much about this complex technology. ECG machines have revolutionized point-of-care diagnosis and have become one of the most fundamental tools in all medical and surgical healthcare specialties. Understanding the tools available to reduce noise and attenuate signal interference, when to use them, and, most importantly, when to avoid them can significantly impact the sensitivity and specificity of ECG diagnostic and prognostic testing.
1. Littmann L. Beware of limb lead reversal. JAMA Internal Medicine. 2018;178(3):435. doi:10.1001/jamainternmed.2017.8636.
2. Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram. Circulation. 2007;115:1306–1324. https://doi.org/10.1161/CIRCULATIONAHA.106.180200.
Dr. Payal Kohli, MD, FACC is a top graduate of MIT and Harvard Medical School (magna cum laude) and, as a practicing noninvasive cardiologist, is the managing partner of Cherry Creek Heart in Denver, Colorado.
The opinions, beliefs, and viewpoints expressed in this article are solely those of the author and do not necessarily reflect the opinions, beliefs, and viewpoints of GE Healthcare. The author is a paid consultant for GE Healthcare and was compensated for creation of this article.