Electrocardiographic artifact caused by extracorporeal roller pump

Abstract

To the Editor: Monitoring the electrocardiogram of patients connected to extracorporeal tubing circuits may reveal interesting electrocardiographic artifacts. We observed this phenomenon during a routine aortocoronary artery bypass procedure. The appearance of the artifactual signal simulated atrial flutter at a rate of approximately 300 waves per minute (Fig 1). This led us to investigate further the conditions under which this artifact occurs. During routine extracorporeal bypass for aortocoronary artery bypass surgery electrocardiographic lead II was recorded. The electrode used was an Ag-AgCI foam-back column type (3M Red Dot, St. Paul, MN). The recording monitor was a Siemens Sirecust 1281 operating room monitor with a Siemens Siredoc 220 paper recorder (Iselin, NJ). The frequency response range was 0.5 to 40 Hz with a commonmode rejection ratio greater than 90 dB. The patient was perfused with a Sarns (Ann Arbor, MI) roller pump and oxygenated with a membrane oxygenator (Cobe, Lakewood, CO). Decreased contact o f the electrode to the patient's skin was achieved by peeling the electrode's adhesive from the skin just enough to expose the electrode jelly. Povidone-iodine 0.5% (Betadine) was then squirted onto the skin and jelly. This ensured less than perfect skin-to-electrode contact. The electrode was then reattached. Because the bottom half of the electrode's adhesive remained dry, the electrode could still be attached to the patient's skin. The electrocardiogram was then recorded, revealing the artifact while on extracorporeal bypass (Fig 2A). When the pump was briefly turned off, the artifact disappeared (Fig 2B). When the electrocardiographic electrode was secured in the proper fashion the artifactual signal also disappeared (Fig 2C), despite the patient's being on full extracorporeal support. To establish the relationship of pump speed to frequency of artifactual signal, the artifact was recorded at two different pump speeds (Fig. 2D). We have shown that there is a correlation between the periodic artifactual signal and the roller pump. Furthermore, an important condition to recording the artifact is poor electrode skin contact. Our method of inducing poor skin contact simulates clinical conditions, where Betadine often drips down onto and under electrocardiographic electrodes. This signal artifact has been described before [1-3]. Similar artifacts have been seen during use of intravenous infusion pumps [4-7] and during hemodialysis [8]. It has been hypothesized that the artifact originates from the deformation of the polyvinylchloride tubing by the roller pump head [2]. Voltage is generated when the tubing and its conductive fluid (e.g., saline, Ringer's lactate) is compressed by the roller pump [2]. Soltys et al [2] provided evidence that the tubing produces the signal when compressed by the roller pump: They manually turned the pump head and reproduced the signal, thereby showing that the signal is not an artifact of the electrical power source for the pump. The pump heads we use have two rollers. With each complete rotation o f the pump head, each roller compresses the polyvinylchloride tubing once. Therefore, the frequency of the artifactual signal should be twice the rotational frequency of the pump head. This was the case (see Fig 2D). Noted too was a decrease in artifactual signal amplitude with the decrease in rotational frequency of the pump head. Fig 1. Electrocardiographic recording. Patient is on cardiopuhnonary bypass with an empty, spontaneously beating heart. The patient's normal QRS complex (x) often is swamped by artifacts (arrows) (Recording speed is 25 ram~s; 10 mm = I m V.)

DOI: 10.1007/BF02832157

2 Figures and Tables

Cite this paper

@article{Klcinman1990ElectrocardiographicAC, title={Electrocardiographic artifact caused by extracorporeal roller pump}, author={Bruce Klcinman and Kamlcsh Shah and PhD Ronald Belusko and Bradford Blakeman}, journal={Journal of Clinical Monitoring}, year={1990}, volume={6}, pages={258-259} }