Anesth Analg 2001;92:384-388
© 2001 International Anesthesia Research Society
TECHNOLOGY, COMPUTING, AND SIMULATION
Radiofrequency Transmission to Monitoring Devices in the Operating Room: A Simulation Study
Stephen E. McNulty, DO,
Benjamin Kline, MD,
Joseph Welsh, DO, and
Richard Bartkowski, MD, PhD
Department of Anesthesiology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania
Address correspondence and reprint requests to Stephen E. McNulty, DO, Thomas Jefferson University Hospital, Department of Anesthesiology, 111 South 11th Street, Suite G-8490, Philadelphia, PA 19107-5092.
 |
Abstract
|
|---|
We evaluated radiofrequency (RF) transmission to various monitoring devices using circuits that simulated potentially hazardous conditions for patients in the operating room. Right heart ejection fraction (REF) pulmonary artery catheters, transesophageal atrial pacing stethoscopes, and temperature-sensing esophageal stethoscopes were subjected to RF transmission from an electrosurgery unit. Peak voltage and spark intensity were measured in circuits between the electrocautery dispersive pad and conductive elements of the various medical devices. All monitoring devices with an exposed conductive surface were found to have induced voltages and even spark generation. The ranking for peak voltage from least to most was as follows: disrupted esophageal stethoscope (620 volts), the transesophageal pacemaker (640 volts), and the REF pulmonary artery catheter (PAC) (680 volts). Peak voltage measurements of the REF PAC significantly decreased from 388 ± 23 to 142 ± 22 volts (P < 0.0001, Student’s t-tests) in a fluid medium compared to air. In a fluid medium, peak voltage significantly decreased from 142 ± 22 to 85 ± 15 volts (P < 0.0001, Student’s t-tests) when the REF PAC was connected to the cardiopulmonary monitor.
Implications: Circuit simulations were used to evaluate radiofrequency transmission from the electrosurgical unit (ESU) to monitoring devices in the operating room. A connection between the dispersive electrode and a medical device can occur outside the body, as simulated by current passing through an electrolyte fluid medium; or through higher resistance pathways such as body tissue, as simulated by current passing through water. The variability in voltage and spark intensity among the monitoring devices that we investigated suggests different relative risks for an ESU-induced thermal injury.
|
Introduction
|
|---|
An electrosurgical unit (ESU) functions by producing alternating radio frequency (RF) current that is normally transmitted between a small area at the tip of the active electrode and a dispersive electrode. The returning energy is normally of low current density, and little heat is generated because of the large surface area of the dispersive electrode (1). The hazards resulting from the misdirection of ESU output through various medical devices are well documented (2). Precautions to avoid ESU dispersive electrode faults are generally well established. ESU generators now monitor for faulty connections of the dispersive electrode to the ESU generator as well as the contact between the dispersive electrode and the patient’s skin (3). However, potentially dangerous current paths may occur when the ESU blade electrode is activated yet not in direct contact with the patient (4). In that situation, the dispersive electrode can become a primary output source. Detailed information about RF current flow between an intact grounding pad and patient monitoring devices has not been described. This information could be useful in predicting the relative risk of thermal injury resulting from RF current flow to different medical devices or monitors used in the operating room (OR).
This study was initiated to determine the characteristics of RF current paths between the dispersive electrode and various medical monitoring devices used in the OR. The purpose of this study was to assess these potentially harmful RF current paths by measuring peak voltage and electrical spark production in different simulation models.
|
Methods
|
|---|
Study Part 1
The medical devices selected for study included the following: 1) esophageal stethoscope, intact; 2) esophageal stethoscope, disrupted to expose the temperature sensing wire; 3) pacing esophageal stethoscope, TapscopeTM model F-0220-E (Arzco Medical Systems; Tampa, FL) 4) pacing esophageal stethoscope, TapscopeTM model TAP-550 (Arzco Medical Systems); and 5) a right ventricular ejection fraction/volumetric oximetry pulmonary artery catheter (REF PAC) (Model number 93A-754H-7.5F; Baxter Healthcare Corp., Irvine, CA).
First, a test circuit was configured in vitro in an OR ( Fig. 1) to simulate conditions when a surgical patient becomes wet. Two, 30 cm length and 1 cm diameter plastic tubes were modified by creating a lengthwise opening approximately one-third of the circumference along the top. The ends of each tube were sealed with silicon rubber and the two tubes were aligned in an end-to-end manner. The close ends of the tubes were connected together with a 1 megohm resistor. One of the outside ends had a direct hardwire connection to the ESU dispersive electrode. The other outside end was connected by wire to conductive elements in the medical device being tested, in proximity to the monitor’s cable interface. The two tubes were then filled with normal saline solution thereby creating a 1 megohm, conductive, electrolyte solution path between the ESU dispersive electrode and the monitoring device.
View larger version (40K):
[in this window]
[in a new window]
|
Figure 1. Two, sealed, saline filled, plastic tubes are connected together by a 1-megohm resistor. On one end, the saline channel is connected by wire to an ESU-dispersive electrode. The saline channel on the opposite end is attached by wire to a right heart ejection fraction pulmonary artery catheter at a point where the catheter is connected to the cardiopulmonary monitor cable. An oscilloscope monitors peak voltage (maximum peak to peak amplitude) from the proximal electrode site. The blade electrode was used to initiate electrosurgical unit (ESU) current flow, although it was otherwise isolated from the circuit.
|
|
Peak voltage was measured from the most conductive part of the device that would have patient contact with intended use. A Tektronix 2215 60 MHz oscilloscope (Tektronix Inc.; Beaverton, OR) with 10 megohm input impedance was used to measure peak voltage. Peak voltage was determined from the maximum peak to peak change in waveform amplitude. The ESU blade electrode was used to initiate RF transmission from a remote location by an individual having no contact with the test circuit. The Aspen ExcaliburTM ESU was set to deliver 50 watts in the coagulation mode.
Measurement of electrical spark generation was based on a subjective assessment of intensity. Each monitoring device was attached to the high resistance saline circuit as previously described. The ESU blade electrode was again used to initiate RF transmission at 50 watts in the coagulation mode from a remote location by an individual having no contact with the test circuit. A stainless steel bar mounted to the OR table was used to test for sparking. The OR table was free standing in one of our ORs. There was no direct connection between the stainless steel bar and any part of the test circuit, the medical device, or the ESU components. If no visible spark occurred between the medical device and the isolated steel bar, a "0" grade was noted. If a visible spark was noted at a separation of 1 mm or less, a grade of "1" was recorded. If a visible spark was observed beyond a separation of 1 mm with no audible sound, a grade of "2" was noted. If a visible spark with an audible sound was noted at a separation of more than 1 mm, a grade of "3" was entered. Five measurements were obtained for each medical device and the maximum value reported.
Study Part 2
Because the REF PAC is normally used in a fluid medium, an in vitro test circuit was constructed to compare radiofrequency transmission and the REF PAC in fluid and air. The test circuit was constructed as described below ( Fig. 2). A 61 x 30 x 12 cm plastic basin was modified by having a 7.5 x 1.3 cm opening cut out of a side wall of the container. This opening was then sealed from the outside using an ESU dispersive pad (ConMed®; Aspen Surgical Systems, Englewood, CA). The dispersive electrode was then connected to an Aspen ExcaliburTM ESU (ConMed®; Aspen Labs). The basin was filled with sterile water to a depth that was 2 cm higher than the opening on the side of the basin. The REF PAC was immersed in a water bath to establish the connection between the conductive elements of the catheter and the return pad. The PAC was placed in the center of the basin and was separated from the dispersive electrode by a distance of approximately 15 cm (Fig. 2). The measured resistance of the water bath between the submersed electrode surface and the ESU dispersive electrode was 2.2 megohms. The circuit was activated by means of a hand-controlled electrocautery pencil with blade electrode (ConMed®; Aspen Surgical Systems) that was triggered in a remote location with no connection to the test circuit. The remote pencil electrode was activated at an ESU setting of 50 watts in the coagulation mode for all measurements. A Tektronix 2215 60 MHz oscilloscope (Tektronix Inc.) was used to measure peak voltage from conductive elements within the PAC.
View larger version (88K):
[in this window]
[in a new window]
|
Figure 2. The right heart ejection fraction pulmonary artery catheter is immersed in a water bath to form a connection between the dispersive electrode that is sealing an opening in the side of the basin. Peak voltage measurements (maximum peak to peak amplitude) were made outside and inside the water bath and then repeated with the right heart ejection fraction pulmonary artery catheter connected and unconnected to the cardiopulmonary monitor. ESU = electrosurgical unit.
|
|
Peak RF voltage was measured from the submersed proximal electrode on the REF PAC. A second measurement was obtained directly from a conductive part of the catheter that was outside the water bath (Fig. 2). Conditions were further varied as the REF PAC was alternately connected and unconnected to the cardiopulmonary monitor. Differences in peak voltages measured outside the water bath connected and unconnected to the cardiopulmonary monitor were compared using Student’s paired t-test. Differences in peak voltages measured inside the water bath connected and unconnected to the cardiopulmonary monitor were compared using Student’s paired t-test. Differences in peak voltages between measurements made inside and outside the water bath were also compared using Student’s paired t-test. A P < 0.05 was considered statistically significant.
|
Results
|
|---|
Part 1
Peak voltage was measured from a conductive surface of every device tested. No voltage potential was measured from the intact surface of an esophageal stethoscope. Maximum values are summarized in Table 1. The peak voltage was greatest in the REF PAC, followed by the pacing esophageal stethoscopes and the esophageal stethoscope with a disrupted sheath.
View this table:
[in this window]
[in a new window]
|
Table 1. Peak Voltage and Spark Intensity Associated With Electrosurgical Unit Transmission to Various Medical Devices
|
|
The maximum grades for spark production are also summarized in Table 1. The devices capable of producing a spark from most to least were the REF PAC (intensity = 3), the pacing esophageal stethoscope (F-0220-E) (intensity = 2), followed by the pacing esophageal stethoscope (TAP-550) (intensity = 1) and the esophageal stethoscope with a disrupted sheath (intensity = 1).
Part 2
A total of 80 peak RF voltage measurements were obtained from 20 REF PACs that were connected and unconnected to the cardiopulmonary monitor from a part of the PAC that was inside and a part that was outside the water bath. Peak voltage varied significantly according to whether or not the PAC was connected to the cardiopulmonary monitor. In 40 measurements made from that part of the PAC outside the water bath, the mean ± SD peak RF voltage significantly decreased from 388 ± 23 to 169 ± 11 volts and 142 ± 22 volts (P < 0.0001, Student’s t-tests) when measurements were made under water compared with those made in air. In 40 measurements made from that part of the PAC inside the water bath, peak RF voltage significantly decreased from 142 ± 22 to 85 ± 15 volts (P < 0.0001, Student’s t-tests) when connections to the cardiopulmonary cable were intact compared with those with no connection. Peak voltage also varied significantly according to whether measurements were taken inside or outside the water bath. In 40 measurements made with the PAC not connected to the cardiopulmonary monitor, peak RF voltage was significantly decreased (388 ± 23 vs 142 ± 22 volts) comparing measurements taken outside and those taken inside the water bath (P < 0.0001, Student’s t-tests). The relationship of mean peak voltages made from the 20 PACs under 4 different conditions are summarized in Figure 3.
View larger version (24K):
[in this window]
[in a new window]
|
Figure 3. *Peak voltage (maximum peak to peak amplitude) was significantly increased both inside and outside the water bath when the right heart ejection fraction pulmonary artery catheter was not attached to its cardiopulmonary monitor cable, P < 0.0001, Student’s t-tests. **Peak voltage was significantly increased when measurements were taken outside the water bath compared with inside the water bath, P < 0.0001, Student’s t-tests.
|
|
|
Discussion
|
|---|
The risk of injury associated with RF transmission to conductive elements in medical devices is often attributed to some defect in the ESU return path from the patient. This study demonstrates characteristics and conditions associated with RF transmission to monitoring devices through the dispersive electrode.
The results of our study suggest that different monitoring devices have different levels of risk based on measurements of peak voltage and sparking. One explanation for differences in peak voltages and spark propagation among the different devices may be because of capacitive differences of the conductive elements and varying current densities resulting from variations in the area of the exposed conductive component (5). The increased current density at the exposed REF PA electrode might be part of the explanation why this device had more intense sparking than the other devices.
The finding that ESU transmission to the REF PAC was significantly decreased within a fluid environment could be a result of a dispersive effect from a moderately conductive fluid medium. This agrees with the clinical observation that the energy required for transvenous RF ablation in the heart increases as blood flow increases (6). A water bath was used to simulate a high resistance pathway that could be encountered between the REF PA electrode and the ESU dispersive electrode in an in vivo setting. The REF PA catheter is usually surrounded by blood and that should decrease the risk of RF transmission through the heart. However, as evidenced by measurements taken in air (outside the water bath), the risk of RF transmission through the heart could increase if the REF PA electrode came in contact with air. With routine use of transesophageal echocardiography in cardiac surgery patients we have found that small amounts of IV air can be entrained during IV fluid administration and accumulate in the right ventricular outflow tract region. Increased peak voltages at the proximal electrode site could occur with a small bubble of air in contact with the surface area of the electrode and the endocardium. It is also possible for the proximal electrode to directly interface with the endocardium, especially if the catheter is in a wedged position that limits movement of the catheter. This study also supports the recommendation that the REF PAC should be securely connected to its monitor when the ESU is activated because peak voltage significantly decreased from unconnected values.
Adverse clinical sequelae of transmitted RF energy to monitoring devices include technical interference and thermal injury. Additionally, ESU-induced ventricular dysrhythmias have occurred in patients with transvenous, centrally placed catheters (7). Although frequency demodulation (8) and formation of direct current potentials (9) have been implicated in these events, direct RF transmission alone may have been responsible. Geddes et al. (10), demonstrated that high frequency current could produce ventricular fibrillation and intracardiac burns in dogs with central venous catheters when the ESU-dispersive electrode was disconnected. Also, junctional ectopy may occur during RF catheter ablation of the slow conductive pathway in patients with atrioventricular nodal reentrant tachycardia (11) and may be a response of the atrioventricular node to injury during RF ablation (12).
This study uses in vitro circuits to demonstrate potentially harmful pathways between monitoring devices and the dispersive electrode. One of the potential problems with interpreting the results of this study is determining the clinical applicability of the in vitro circuit models. An ESU setting of 50 watts in the coagulation mode was selected for all measurements and observations. Fifty watts was selected as representative of the higher range of power outputs used in the OR. Coagulation mode was selected because it produces a wider range of frequencies compared with the Cut mode. The first circuit model (Fig. 1) used in this study was constructed to simulate a clinical situation in which the surgical patient becomes sufficiently wet to allow an electrolyte fluid contact to the edges of the dispersive electrode. An electrolyte solution "bridge" from the dispersive electrode to the monitoring device could occur as a result of profuse sweating or possibly a leak from an IV set. In this model, the electrical connection was made to some conductive, external part of the device that would be outside of the patient. A 1-megohm resistor was selected to represent the upper end of resistances likely to be encountered. Assessment of peak voltage and sparking should be valid clinical correlates for possible thermal injury because voltage, current, power, and heat generation are related. Increased arcing and sparking activity at the active electrode are associated with increased direct current voltage potential and power setting of the ESU in watts. Generally, the ESU setting in watts, the intensity of sparking, and the amount of tissue affected are related to each other. Because our voltage assessment was subjective we attempted to calibrate our grading system using the REF PAC in the same circuit model and varying the ESU output. An ESU setting of >30 watts corresponded to a grading of three by a blinded observer. A grade of two corresponded to a setting of 10 to 29 watts and a grade of one corresponded to a setting of 3 to 9 watts. These results are summarized in Table 2. The water bath circuit was used to test a theoretical current path directly from the dispersive electrode through a fluid medium to the REF PA, as it might occur in a patient. We focused on the REF PA catheter because it seemed to be the best receiver for RF current in our models.
View this table:
[in this window]
[in a new window]
|
Table 2. Approximate Equivalency of Spark Grade Using the Right Heart Ejection Fraction Pulmonary Artery Catheter Connected to the Electrosurgical Unit (ESU) Dispersive Electrode
|
|
This study does not establish the likelihood of developing an injury in the clinical setting. Occurrences of RF current-induced hazards of this type are probably uncommon (13). However, there are at least two factors that could be associated with a more frequent than expected incidence of tissue injuries. One factor is that creating a dispersive electrode circuit is not a complicated process. It could be established every time the hand-controlled blade electrode was activated without direct patient contact. The hand-controlled "active" electrode can initiate RF transmission but is otherwise not a part of the circuit. Second, these circuits do not require any obvious fault with the dispersive electrode connection to the ESU generator or with the patient contact site because it is the active electrode that is essentially disconnected from the patient. An electrolyte fluid connection to the dispersive electrode would not necessarily interfere with the normal operation of the ESU and may not be detected as a dispersive electrode fault by the ESU generator. Another factor to consider is that RF thermal injuries involving the devices tested in this study could be difficult to detect or verify by virtue of their in vivo locations.
We conclude that RF energy can be directly transmitted between the ESU dispersive electrode and conductive elements of monitoring devices. This study reinforces previous recommendations for vigilance and avoidance of hazardous situations that could develop after initial inspection of the ESU-dispersive electrode site. The potential for a hazardous circuit between the monitoring devices that we tested and the ESU-dispersive electrode should be negligible under normal conditions. However, the presence of profuse sweating or an electrolyte solution leak could lead to connection between the dispersive electrode and other conductive sites that may not be immediately obvious. Monitoring devices containing conductive elements in contact with the patient should be carefully observed for any possible contacts to conductive surfaces outside the patient.
|
References
|
|---|
-
Pearce J. Current electrosurgical practice: hazards. J Med Eng Technol 1985; 9: 107–11.[ISI][Medline]
-
Litt L, Ehrenwerth J. Electrical safety in the operating room: important old wine, disguised new bottles. Anesth Analg 1994; 78: 417–9.[Free Full Text]
-
Helfman SM, Berry AJ. Review of electrical safety and electrosurgery in the operating room. Am J Anesthesiol 1999; 26: 313–20.
-
Litt L. Electrical safety in the operating room. In: Miller RD, ed. Anesthesia. New York: Churchill Livingstone, 1994: 2625–34.
-
Ehrenwerth J. Electrical safety. In: Barash PG, et al, eds. Clinical anesthesia. Philadelphia: Lippincott-Raven, 1996: 137–55.
-
Panescu D, Whayne JG, Fleischman SD, et al. Three-dimensional finite element analysis of current density and temperature distributions during radio-frequency ablation. IEEE Trans Biomed Eng 1995; 42: 879–90.[ISI][Medline]
-
McNulty SE, Cooper M, Staudt S. Transmitted radiofrequency current through a flow directed pulmonary artery catheter. Anesth Analg 1994; 78: 587–9.[Free Full Text]
-
Tucker RD, Schmitt OH, Sievert CE, Silvis SE. Demodulated low frequency currents from electrosurgical procedures. Surg Gynecol Obstet 1984; 159: 39–43.[ISI][Medline]
-
Tucker RD, Kramolowsky EV, Stasz P. Direct-current potentials created by arcing during monopolar radiofrequency electrosurgery. Biomed Instrum Technol 1990; 24: 212–6.[Medline]
-
Geddes LA, Tacker WA, Cabler P. A new electrical hazard associated with the electrocautery. Med Instrum 1975; 9: 112–13.[Medline]
-
Jentzer JH, Goyal R, Williamson BD, et al. Analysis of junctional ectopy during radiofrequency ablation of the slow pathway in patients with atrioventricular nodal reentrant tachycardia. Circulation 1994; 90: 2820–6.[Abstract/Free Full Text]
-
Thakur RK, Klein GJ, Yee R, Stites HW. Junctional tachycardia: a useful marker during radiofrequency ablation for atrioventricular node reentrant tachycardia. J Am Coll Cardiol 1993; 22: 1706–10.[Abstract]
-
Bruner JMR, Aronow S, Cavicchi RV. Electrical incidents in a large hospital: a 42 month register. J Assoc Adv Med Instrum 1972; 6: 222–30.[Medline]
Accepted for publication September 29, 2000.
Top
Abstract
Introduction
