Suspected Shunting of Defibrillation Energy in the EP Lab

A Case Study of Multiple-shock Incidents Using Biphasic External Defibrillators Following Induced Ventricular Fibrillation

Defibrillation using biphasic energy delivery is quickly becoming the standard in the acute care setting. The major defibrillator manufacturers are racing to convert their product lines, and the models generating monophasic waveforms are becoming extinct. This shift has been based, in large part, upon a study published in Circulation in 2000 by Schneider et al; their research showed that within the first shock series, 98% of ventricular fibrillation (VF) patients were defibrillated with 150-J biphasic shocks, compared with 69% of patients defibrillated with 200- to 360-J monophasic shocks.¹ The results of this and other studies prompted the American Heart Association to adopt the use of biphasic waveform defibrillation as a Class IIa intervention² and make it part of the most current Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiac Care. In keeping with the AHA recommendations, our electrophysiology lab, which performs over 1,000 studies and device implants each year, upgraded to biphasic defibrillators in August of 2007 as part of a hospital-wide changeover. Like most EP labs, a portion of our procedure volume includes studies to evaluate patients for VT/VF as an indication for ICD implantation. External defibrillation is used as second-line therapy for patients in whom VT or VF is induced during EP studies or during defibrillation threshold testing post ICD implant. Our previous protocol was to deliver 360-J of monophasic energy in situations during EP studies when the patient could not be paced out of VT, VF was induced, or when defibrillation by an ICD was unsuccessful on its second attempt. The 360-J monophasic dose was known to have a negligible failure rate for first-shock defibrillation success in the controlled setting of our EP lab. As stated earlier, biphasic defibrillators demonstrated a 98% success rate, with 96% success on the first shock in the out-of-hospital setting of VF, despite the average time down prior to first attempted defibrillation being 8.9 ± 3.0 minutes. ¹ Because of these results and our previously high success rate with the monophasic defibrillators, the EP staff and physicians were alarmed when the first patient we were forced to externally defibrillate with the new units required three 150-J shocks in immediate succession. The physician on the case called the manufacturer directly to discuss pad placement (which was anterior-posterior) and energy selection, and confirmed they were in accordance with their recommendations. The service representative suggested alternate site pad placement (anterior-apical) and assured the physician that there was no proven increase in efficacy with 200-J shocks, but again offered this as an alternative. When the same physician had a patient that required four shocks at 200-J eight days later, in a different electrophysiology suite using a different unit of the same model biphasic defibrillator, the products were pulled from the department and the manufacturer was contacted for further investigation. Patient #1: The first patient was a 43-year-old male, of medium to large build (weight: 94.9 kg, height: 182 cm, BMI: 27.6) with a medical history of previous myocardial infarction with no evidence of acute ischemia or history of anti-arrhythmic drug therapy. Ventricular fibrillation was induced during an electrophysiology study to evaluate a syncopal episode. The patient required three high-energy shocks at the manufacturer’s recommended dose of 150 joules in immediate succession with defibrillator pads in anterior-posterior configuration in proper position as designated on the packaging for the pads. The third delivered shock, at 150 joules, successfully defibrillated the patient. Rhythm strips printed by the device showed the following information: • Shock #1: 150-J selected, 155-J delivered; impedance 100.6 ohms; 17.1 Amps • Shock #2: 150-J selected, 155-J delivered; impedance 98.5 ohms; 17.8 Amps • Shock #3: 150-J selected, 155-J delivered; impedance 100.1 ohms; 17.7 Amps Patient #2: The second patient was a 66-year-old male of large build (weight: 100 kg, height: 183 cm, BMI: 29.9) with a medical history of coronary artery disease free of acute ischemia and no history taking anti-arrhythmic drugs. Ventricular fibrillation was also induced during stimulation in an electrophysiology study. This patient was given three high-energy shocks (200-J) in immediate succession. Chest compressions were performed briefly by an EP staff member (for less than 1 minute), after which a fourth shock at 200 joules was able to successfully defibrillate patient. Rhythm strips printed by the device showed the following information: • Shock #1: 200-J selected, 204.6-J delivered; impedance 76.2 ohms; 25.6 Amps • Shock #2: 200-J selected, 203.9-J delivered; impedance 74.7 ohms; 26.6 Amps • Shock #3: 200-J selected, 204.0-J delivered; impedance 74.7 ohms; 26.7 Amps • Shock #4: 200-J selected, 203.4-J delivered; impedance 69.9 ohms; 28.3 Amps Follow-up Testing Following each multiple shock episode, the defibrillator was inspected by the hospital’s Biomedical Engineering Department and found to be functioning properly. Unit settings and calibrations were verified using test shocks and measuring equipment in parallel to the defibrillator, but no evidence of device failure could be found. Pads used in these cases were the defibrillator manufacturer’s brand and the recommended model. They were retained and checked, but appeared to be free of defect to our biomed staff and the manufacturer. The electronically recorded event logs in the defibrillator units were downloaded and shipped to the manufacturer for review by their clinical, algorithm, and engineering staff. The manufacturer arranged to send engineers out to our facility in an attempt to determine the cause of the multiple shock episodes we experienced. In September of 2007, engineers from the defibrillator manufacturer visited our EP lab to observe the use of the model and identify all of the equipment being used within the lab in conjunction with the defibrillators. Review of the event logs and the results of the testing that was done showed that the defibrillators were working as designed/intended and performing to specifications. Following this visit, they requested a set of each cable we connected to the patient during these EP studies for impedance measurements and further bench testing, which was indeterminate. At that point they decided that an additional evaluation of the EP lab environment was necessary, and an engineer returned once again in January of 2008 to perform additional tests on the equipment. Their findings, although still inconclusive, indicated that “defibrillation current was likely being shunted away from the patient due to other connections that were incompatible with defibrillation energy.” Discussion Shunting of defibrillation energy is not a phenomenon that is unique to biphasic waveform defibrillators. In this setting, shunting refers to the diversion of energy through non-myocardium on its path from one defibrillator pad to the other. Although it may seem counterintuitive, the lower the chest wall impedance, the less current travels through the ventricles. Instead, the flow of energy diffuses through the entire area of thorax between the pads, which requires a greater peak current to be delivered (Figure 1). Conversely, the more resistance that is offered by these non-cardiac structures, the more current is preferentially directed through the heart, which literally becomes a ‘path of least resistance’ for the energy. However, since the creation of the external defibrillator, high transthoracic impedance has been the enemy. “High energy is necessary for MDS [monophasic damped sinusodial] waveforms, since the waveform shape degrades in the face of high impedance… Traditional escalating energy protocols were based on the notion that a failed defibrillation attempt indicated the presence of high transthoracic impedance.” ³ A hallmark of the biphasic truncated exponential waveforms is the fact that they are not degraded by high impedance, and some of these defibrillators feature algorithms that detect the impedence and compensate for it through customization of the peak current, rate of current decline, and duration of current delivery to ensure the selected energy is delivered. ³ In the electrophysiology lab, our patients are tethered to a myriad of electrical conductors known to lower the natural impedance of the body. An average patient is connected to a hemodynamic monitor, a 12-lead electrocardiographic recording system, and indifferent/return electrodes, in addition to the defibrillator pads and its leads, and any number of intra-cardiac pacing and recording electrodes. In electrophysiology we are also taught to arrange these connections in a manner that helps us reach our goal of generating as little impedance and noise as possible. Some components tested by the engineers were found to have impedences of less than 50 ohms. Although many of these connections are designed to be ‘defibrillation proof’ (as indicated at each point of connection to the patient by the icon of a heart shielded within a box; Figure 2), some are not. We can also be lulled into a false sense of security by the term ‘defibrillation proof,’ since this generally refers to the fact that the receiving product is shielded from damage by an over-voltage situation, such as the inflow of current known to be a concern in the case of defibrillation. It does not guarantee that current is not shunted to the device, only that it “is isolated to such a degree that no current higher than the allowable patient leakage current… flows into it from an application of external voltage source to the patient.” ⁴ Per our hospital policy, this allowable leakage should be no more than 50 microamps per patient connection, which is far less than 1% of the current flow used in defibrillation. The theory of the manufacturer is that the shunting occurred through these multiple low-resistance pathways created by one or more of our external or internal patient connections, including the intra-cardiac pacing and recording electrodes. We believe it is also possible that these alternate paths reduced the functional intra-thoracic impedance at the time of defibrillation below what was detected by the pads, thereby causing those same impedance compensation algorithms that are designed to customize the current delivery pattern to deliver an insufficient current. As seen in Figure 3, “low impedance requires higher peak current to address greater shunting.” ³ If an intra-thoracic environment was created by these numerous low-resistance connections where the impedance was 25 ohms lower than what was detected by the pads on the skin surface, the defibrillator’s algorithm would have delivered the incorrect biphasic waveform. Testing by the defibrillator manufacturer stopped short of tracing the actual path of energy diversion, but we have found several pieces of indirect evidence that are suggestive of excessive current leakage. Following their tests of our equipment, in which 150-J was delivered, the personality module that connects our RF generator to our mapping system became inoperable. We have also had system errors following monophasic cardioversion of patients during electrophysiology studies that have required intervention, from hardware re-booting to exchanging whole systems in order to complete the procedure. In one case, a pin fracture occurred in our electrophysiology recording system’s amplifier following a 100-J monophasic shock, rendering the channel unusable without repair at the factory. Since that event, we have temporarily disconnected the EP recording and ablation systems from the patients prior to monophasic cardioversion without further evidence of damage to equipment or procedure interruption. A search of MedSun, the FDA Medical Product Safety Network for adverse event reporting, found only one reported incident of biphasic defibrillator failure. In that case, ICD defibrillator threshold testing was unsuccessful and rescue shocks were attempted using that facility’s biphasic defibrillator. Several shocks were attempted using escalating energy up to 200-J. These shocks failed to terminate VF. The staff had to bring in a monophasic defibrillator from a separate care unit to successfully defibrillate the patient. As was the case with our equipment, the unit was inspected and tested and found to be functioning normally. Concrete clinical information on this issue is scarce, which may be due in large part to the nature of our patients’ conditions and our experiences in cardiac arrest settings where multiple shocks and chest compressions are often necessary to return spontaneous circulation. However, it is important to remember that VT or VF induced during an EP procedure should terminate with relative ease. Defibrillation fails because the initial cause persists, the energy traveling through the myocardium is insufficient, or both. In the case of induced VT or VF, the initial causative factor is transient and immediately removed. It is initiated by pacing as opposed to tissue hypoxia from an acute MI. The mass depolarization of the myocardium should create a sufficient refractory period in all areas of the heart. This will block any re-entrant pathways that might be present, if that mass depolarization is actually achieved. It is for these reasons that incidents of induced VF not cleanly terminated with external defibrillation should be viewed with some concern and critiqued. In any case of multiple shocks, the EP team should ask themselves “is there a reasonable cause for this patient to have required multiple shocks?” If not, the event should be reported. Our lab has filed a report of each incident with the FDA using the MedWatch Form 3500 for voluntary reporting of adverse events or product problems (available at www.accessdata.fda.gov/scripts/medwatch or found by following the links on the FDA’s webpage at www.fda.gov). They encourage you to report these events “even if you are not certain the product caused the event” or if you “do not have all of the details.” The hope in publishing this case study is to increase awareness of this phenomenon within the electrophysiology community so these incidents will be tracked and reported in greater number to the FDA as appropriate. The manufacturers of all these systems we use need to study this further to analyze what role, if any, their particular product plays in this, and to help identify the cause. By doing so, the EP staff and our industry partners can then take whatever steps are necessary to prevent harm to our patients and the tragic loss of life from the elective induction of a lethal arrhythmia without a properly working safety net.