The Implantable Cardioverter Defibrillator: Technology, Complications, and Emergency Management
Authors: Fred F. Tilden MD, FACEP, Department of Traumatology/Emergency
Medicine, Hartford Hospital, Hartford, CT.
Blake A. Spirko MD, Integrated Resident in Emergency Medicine, Hartford
Hospital, Hartford, CT.
Peer Reviewer: Sandra M. Schneider, MD, Professor and Chair, Department of
Emergency Medicine, University of Rochester, Rochester, NY.
Each year, 350,000-400,000 Americans will suffer sudden cardiac death (SCD), with the vast majority of these individuals dying before or shortly after they reach the hospital.1-3 For most of these patients, the initial cardiac rhythm is ventricular tachycardia or fibrillation (VT/VF), which rapidly degenerates to asystolic arrest. Many of these individuals would survive if treated in a timely fashion, but SCD is often the first-and frequently, unpredictable-manifestation of heart disease.4 Nevertheless, it is possible to identify patients who are at high risk for SCD.
In an attempt to reduce SCD, the first human implantation of an indwelling defibrillator was performed in 1980.5 From the outset, there were strong indications that implantable cardioverter defibrillators (ICDs) could improve survival in a subset of patients when compared to medical therapy. But it was not until recently that well-controlled trials confirmed their benefits.6-8 The ineffectiveness of prevention-based medical therapy in combination with expiration of ICD patent protection created the impetus for further development and expanded clinical applications for the device.9-11 Six thousand ICDs were implanted in 1991, rising to 25,000 by 1997. Despite concerns about cost and the ambiguous results of recent trials, about 62,000 new ICDs are expected to be implanted in the year 2000.12-15 (See Figure 1.) "The future of the implantable defibrillator," one industrial publication notes, "looks extremely bright."16
Given their widespread use, the role of the emergency physician (EP) in caring for patients with ICDs is becoming increasingly important. As the number of patients with ICDs grows, EPs will encounter these patients with greater frequency in their practice, both at regional medical centers (where the devices are implanted) and at community hospitals (where most Americans seek emergent care). EPs will be called upon to assess and occasionally treat the patients who have been shocked or have problems related to their device, and they will be expected to discuss their findings with the consulting electrophysiologist. Many of these patients will need resuscitation.
Unfortunately, ICDs have not been widely reviewed in the emergency medicine literature; although, there have been just a few case reports, none of them recent, and a review written in 1994.17-19 Since that time, significant advances have been made.
With these issues in mind, this review will present an update on ICDs and their clinical implications for the current practice of emergency medicine. It will present a pragmatic approach to the emergency department (ED) evaluation and treatment, emphasizing three questions: 1) What are the important elements of ED evaluation of the ICD patient who has been shocked; 2) What are the current complications associated with ICDs; and 3) What are treatment and resuscitation considerations in the ED?
-The Editor
The ICD:A Brief History
The "shock boxes" of the early 1980s were relatively crude devices that were made with off-the-shelf components. Typically, they were implanted deep within or superficial to the rectus muscles, from which long, subcutaneous leads tunneled to the chest. A thoracotomy was required to fix several defibrillation patches to the epicardium. Not surprisingly, these ICDs were bulky, nonprogrammable, and, occasionally, delivered spurious shocks; nevertheless, they were effective in limiting mortality caused by SCD.24-26
Second-generation ICDs, introduced in the mid 1980s, were programmable, and they were better able to detect ventricular arrhythmias (VAs). Moreover, these ICDs could delay electrical therapy if the target rhythm was nonsustained. They produced additional reductions in mortality from SCD, but were not without surgical and/or other complications.27-28 Developed in the late 1980s, third-generation ICDs featured multiple, "tiered" therapies, enhanced programmability, sophisticated and targeted energy delivery, and, perhaps most importantly, permitted retrieval of abnormal rhythms and subsequent defibrillatory currents that were delivered. Along with these improvements, transvenous endocardial leads superceded epicardial patches and the need for thoracotomy, thereby decreasing morbidity and intraoperative mortality.29
By the early 1990s, ICD generators ("cans") were approaching the size of cardiac pacemakers and could be implanted in the chest wall in close proximity to the heart-a feature that decreased complications associated with long leads that had to be tunneled from the abdomen. In 1996, the movement toward simplification took a leap forward with the introduction of the electrically active generator case ("hot can"), where, by acting as a defibrillating pole, it obviated the requirement for a second intracardiac coil. Fourth-generation devices are now emerging, which feature extended, dual chamber bradycardia pacing and sensing and atrial defibrillation capabilities.30-32 The typical battery life in current ICDs is 3-5 years, depending on the manufacturer and frequency of discharges. The generator is replaced when the battery is depleted.
The ICD Patient
Criteria for ICD implantation were established in 1991. Until recently, most candidates for ICDs were those who could not tolerate antiarrhythmic medication and those who had refractory sustained VT/VF, despite pharmacotherapeutic interventions. Although implantation criteria are not based on disease states per se, most patients at risk for sudden cardiac death have severe chronic heart disease and multiple independent risk factors for SCD. More than 70% of ICD patients have significant coronary artery disease, and most have severe left ventricular dysfunction with an ejection fraction of less than 30%.21 A small minority of ICD patients have normal coronary arteries and electrically compromised dilated cardiomyopathy, long QT syndrome, mitral valve prolapse, or hypertrophic cardiomyopathy.21 Most ICD patients are also at risk for "triggers" that precipitate sustained ventricular arrhythmias (VAs) that lead to SCD if left unchecked. These triggers include acute myocardial infarction (AMI), electrolyte abnormalities, and proarrhythmias associated with medications or pacemakers.22
In 1996, the FDA approved ICDs for a new subset of patients. This stemmed from the results of the 1996 Multicenter Automatic Defibrillator Implantation Trial (MADIT), which demonstrated that patients with AMI and significant left ventricular dysfunction accompanied by nonsustained VT had improved survival compared to patients treated medically.23
ICD Technology
There currently are four ICD manufacturers in the United States. Each of these devices differs in its features but they all have two basic components-a generator and a lead system. The generator or "can" consists of micro circuitry, capacitors, and a single battery encased in a hermetically sealed metal box. Most cans weigh about one-half pound and measure about 4 ´ 5 cm.34 There are always two basic circuits, one for low voltage that is used for sensing and analyzing cardiac activity, and one high voltage circuit and battery demand that is used to deliver therapy. The expected battery life in current ICDs is about 3-5 years with longevity, depending on the manufacturer and frequency of discharges. The generator is replaced when the battery is depleted.
A number of different lead systems are available for ICDs in order to meet specific electrical, anatomical, and electrophysiological requirements of each patient. All of these systems have a sensing/pacing electrode fixed to the endocardium of the right ventricular (RV) apex, and a defibrillating coil located just proximal to it. Beyond this consistent configuration, systems vary considerably in their architecture and components. In some systems, the lead is bipolar (See Figure 2a); the sensing lead is accompanied by a second defibrillating coil positioned along the lead and terminates by floating in the right atrium (RA). Some ICDs have unipolar leads with a single RV defibrillating coil and a separate lead inserted into the RA, superior vena cava (SVC), or coronary sinus, or a hot can acting as the second coil (See Figure 2b.) Other lead options, implanted less frequently but still present on chest radiograph of patients with older devices, are subcutaneous or submuscular chest wall patches.35 In 1998, the EP can expect to encounter ICD patients with an array of implanted components, depending on the patient, manufacturer, and the year the device was implanted or upgraded.
Defibrillation Functions. Current ICDs feature tiered therapy and are designed to deliver a number of electrical responses in algorithmic fashion. An electrophysiologist preprograms the algorithm to meet the needs of each patient, which requires adjusting parameters related to sensing heart rhythms and the type of electrical therapy.35 A typical algorithm is presented in Figure 3. Typically, the first tier for VT pacing is overdrive or "antitachycardia" pacing (ATP). Delivering fast, synchronized, low energy bursts (< 1 joule), this form of electro-therapy is designed to interrupt the reentrant circuit and terminate the abnormal rhythm. (See Figure 4.) Patients may sense an abnormal rhythm initially but are generally unaware of ATP. If the ventricular rhythm is not stopped by ATP, the next ICD-medicated therapeutic tier consists of synchronized, low energy cardioversion followed by high energy cardioversion, (0.1-30 joules). (See Figure 5.) High energy cardioversion is variably described by patients as " like being shot" or "kicked by a mule," although many individuals describe less dramatic sensations.36
In order to prevent against treatment of nonsustained VT, ICDs are generally "uncommitted" or programmed to recheck the rhythm after about ten seconds of charging time. If VT is still present at that time, the ICD delivers ATP or, depending on the programming, synchronizes to the R wave and shocks the patient. If the rhythm is nonsustained, the charge is "dumped," sparing the patient a painful shock or proarrhythmia that may result from shocking sinus rhythm. ICDs are programmed to deliver up to five cardioversion shocks per VT event. Because ATP and low energy cardioversion, on occasion, can be proarrhythmic and precipitate accelerated VT or VF, all tiered VT therapy algorithms include defibrillation (DF).37 Fall back VVI bradycardia pacing is necessary to treat DF-induced bradycardia. Except for some fourth-generation devices, ICD bradycardia pacing is not designed for long-term therapy.
Clinical Aspects of Implantation
Until the early 1990s, ICDs required a thoracotomy to facilitate epicardial lead placement and burrowing leads from an abdominal generator. The procedure was performed in the operating room with an electrophysiologist present to perform intraoperative testing. When thoracotomy was no longer necessary, the procedure became the primary domain of the electrophysiologist and was moved to a specialized suite.38,39
Implantation is performed under general anesthesia or with local anesthesia and conscious sedation. A technical representative of the ICD manufacturer often plays an important role in intraoperative testing and programming. The procedure consists of the following steps: the ICD pocket is fashioned and the transvenous lead system is fixed to the RV endocardium under fluoroscopy, and the ICD testing system is then connected. VF is induced using low energy shocks on the T wave, and the minimum energy needed for defibrillation, or the defibrillation threshold (DFT), is ascertained by introducing sequential shocks of decreasing energy.40 A DFT that is suitable to the ICD's output is usually ascertained quickly, after which the generator is connected and placed in the pocket. Occasionally, the DFT is too high and an extra intracardiac coil or chest wall patch is needed to decrease the trigger threshold and make it compatible with the output of the device.41-42
Following implantation, some patients with poor left ventricular (LV) function are anticoagulated, whereas others are started on aspirin. Some centers conduct postoperative electrophysiologic testing using noninvasive programmed stimulation that is available with newer, third-generation ICDs. Postoperatively, patients are monitored for complications associated with the venipuncture (i.e., pneumothorax, hemothorax, air embolism, bleeding, hemoptysis, and brachial plexus injury). Discharge is generally within 24 hours. A few days of oral opiates is usually sufficient for pain control.
Follow-up for ICD patients is essential, and commitment to lifelong surveillance is a requirement for implantation. Follow-up schedules vary, but most patients are seen within 1-2 weeks postoperatively by an electrophysiologist and ICD technician for wound check and device interrogation. At interrogation, a "wand" attached to an ICD-specific programmer is placed over the device; it retrieves information regarding ICD discharges, sensing/pacing parameters, and lead integrity.43 Subsequently, follow-up depends on the patient and electrophysiologist but generally occurs every 1-6 months. All patients are instructed to carry literature that identifies their device and electrophysiologist. If an ICD discharges, the patient is generally instructed to call the clinic or covering electrophysiologist. Depending on the clinical situation, this may lead to an ED visit.
Complications
ICD complications can be function related or nonfunction related, as outlined in Table 1.
Function Related Complications. ICDs can malfunction by delivering inappropriate therapy, ineffective therapy, or no therapy at all. Inappropriate therapy occurs when a shock or ATP is delivered for a nontarget rhythm. For example, the ICD will deliver ATP during supraventricular arrhythmias (SVAs) if the programmed rate criteria for VT/VF are met.45,46 (See Figure 6.) In addition, normal sinus rhythm can be misread as VT if the T or P waves are oversensed as separate R-waves. Third-generation ICDs have programmable sensing "enhancements" that assess such characteristics as QRS width and characteristics of arrhythmia onset, as well as other parameters to distinguish SVAs from VAs. Yet, despite enhancements designed to differentiate among tachyarrhythmias, inappropriate shocks may still be delivered.47 "Device proarrhythmias" occur when low energy ICD discharges such as ATP do not convert VT but instead accelerate VT or induce VF-one reason why all tiered algorithms using ATP must be programmed to include backup high energy shocks.48 (See Figures 7 and 8.) "Device-device interactions" also can occur in ICD patients with permanent pacemakers, in particular when pacing spikes are read as R-waves ("double counting"), fulfilling rate criteria and causing inappropriate ICD discharge.49,50
Spurious ICD discharges also can occur if leads are dislodged, fractured, or fail for other reasons. When this happens, myopotentials from specific movements of chest or arm muscles can be overread by the ICD as VT/VF and cause ICD discharge. (See Figure 9.) Problems with endocardial leads are fewer than with older epicardial leads, but still occur in about 8-9% of patients.51
Electromagnetic interference, which was a problem before sources and ICDs were properly insulated, can cause an intact ICD or an ICD with an occult lead problem to discharge. ICD deactivation secondary to outside interference rarely occurs. Hospital sources of interference include electrocautery, MRI, and lithotripsy. Other sources include slot machines, store surveillance devices, arc welders, and hand-held metal detectors frequently used in airports.52-53 Effects are generally thought to occur only after prolonged exposure at close range (i.e., at least several minutes and within 10 cm of the transmitting source). Cellular telephones, microwaves, and airport walk-through metal detectors do not appear to cause spurious ICD discharges.54,55
Ineffective therapy can occur when there is an increase in the DFT, which can result from deterioration of left ventricular function or concurrent medications, usually amiodarone. Defective lead components can also make generator discharges ineffective.
Failure to deliver the programmed prescription represents another form of ICD failure. These problems are generally related to arrhythmia sensing. The causes of failure are similar to those for inappropriate/ineffective shocks, that is, deterioration of leads or interactions between the ICD and a pacemaker or certain medications. For example, demand pacemakers can be spuriously activated during VF when the R-wave is not detected; the ICD reads the pacemaker spike as a normal R wave and withholds DF.56 "Drug-device" interactions can cause ICDs to withhold therapy by changing the intrinsic cardiac conducting substrate.57 Some patients may require antiarrhythmics (mainly sotalol or amiodarone) to sufficiently suppress and slow VT to prevent frequent, painful, or battery-draining discharges. But, when VT cycle length is slowed, it can go undetected. Combinations of several antiarrhythmics increase cycle length further, slowing VT rate and making the arrhythmia even less likely to be detected. Battery depletion as a cause of sudden failure of ICD therapy is unlikely in patients who are compliant with follow-up.
Nonfunction-Related Complications. Complications that do not threaten ICD function include generator pocket problems, psychological problems, and, rarely, thrombosis and embolization.
The most frequent and devastating complication related to the ICD pocket is infection. As miniaturization of generators allowed the primary implantation site to be the chest wall and transvenous lead systems obviated the need for thoracotomy, the incidence of infections decreased.58 Older studies reported ICD infection rates in the 3-10% range and, in some cases, as high as 20%; these have decreased to less than 2%.59 As with other indwelling devices, it is helpful to consider ICD-related infections in temporal terms. Postoperative complications, such as hematomas and seromas, are usually present within several days of the procedure and are generally sterile but can occasionally become infected. Early infections, or those with an onset of less than 60 days after implantation, are generally caused by intraoperative contamination of the device with Staphylococcus aureus and tend to be more acute in presentation. Late infections, or those after 60 days, tend to be more indolent and usually are caused by transient bacteremia, skin erosion, or delayed intraoperative infection with S. epidermidis. Staphylococcal organisms predominate, but gram negatives, anaerobes, and fungi can also infect the pocket.60,61 Other sources of infection include hematogenous spread from catheters or concomitant urinary or respiratory infections. Patients at increased risk for pocket infections include diabetics, patients on corticosteroids, and those undergoing reimplantation. In addition, the presence of a subcutaneous patch electrodes, added to decrease DFT, seem to be an independent risk for infection.62 As more electrically active generators are implanted, the use of subcutaneous patches will decrease.
In patients with signs of systemic or local infection, it is prudent to consider the ICD as a potential source, especially if implantation is recent. Although ICD infections can be occult, most originate in the pocket. Examination of the pocket site will demonstrate erythema, tenderness, warmth, and fluctuation. Fluctuation, in the absence of other inflammatory findings that present within a few weeks of implantation, is expected and not, in itself, suggestive of infection. It should be stressed that infections of the pocket should never be considered localized, even when there is no fever or leukocytosis; the proximal leads are almost always infected and organisms can migrate to the heart.63 Data is scanty but if an ICD patient presents a systemic infection whose source cannot be confirmed definitively, the EP should consider the ICD or its components to be infected unless an alternative explanation can be found. Although blood cultures do not help guide initial therapy, and are frequently negative with pan-device infection, they should be drawn before empiric, antimicrobial therapy is started. Vancomycin is generally the initial antibiotic used. Most, but not all patients will require removal of the device.64,65 ED physicians should note that patients with ICDs, like those with pacemakers, do not require prophylactic antibiotics for invasive procedures.
Psychological Issues. Psychological problems associated with ICDs have been extensively documented since the early days of implantation.66 Although most patients understand that ICDs are likely to prolong and improve the quality of their lives, they tend to suffer from more anxiety and anger than the general population.67 Many report transient adjustment disorders and about 20% have major depression directly related to the device.68 Fear is the predominant response to anticipated shocks and panic disorders are also quite common. One study found that emotional problems tend to occur within the first six months of implantation and improve thereafter.69 Families of patients with ICDs also have been studied, and they also report anxiety and distress, especially when the patient has been shocked. These anxieties need to be considered in the emergent situation.70
Other ICD-related complications are now rare. Venous thrombosis is well documented in the pacemaker literature, and, although it usually is asymptomatic, it has been reported to cause arm swelling and SVC syndrome.71 Although the larger caliber of ICD leads might be more problematic, the true incidence of thrombus-related symptoms with ICDs seems to be less than 1%.72
However, symptomatic embolic events related to ICDs have occurred in several patients, always after several shocks were delivered spontaneously or during DFT testing.73 Popliteal artery occlusion has occurred, as well as minor ischemic cerebral events, either immediately or within 24 hours. Pulmonary emboli can be associated with pacemaker leads but have not been reported with ICDs.
The "vanishing defibrillator syndrome," or migration of an abdominally placed ICD, can be asymptomatic or cause vague abdominal pain or dysuria.74 The vast majority of generators are now placed in the chest, but depleted abdominal ICDs are replaced by new devices in the same pocket. These, too, can migrate and erode into the peritoneum, if rarely.75 "Twiddlers syndrome" occurs when the patient fondles the ICD, causing the leads to twist around the can. It can cause pocket discomfort, deformity, and, rarely, device malfunction.76,77
Patient Assessment: Emergency Considerations
It should be emphasized that the overwhelming majority of patients with ICDs have significant structural heart disease. Although implantation of an ICD reduces risk of SCD, patients are still at high risk for other cardiac complications. Accordingly, assessment in the ED should focus on identifying patients who are at risk for acute decompensation. This is accomplished by taking a thorough history, a comprehensive physical examination, and performing an appropriate laboratory data base. After the cardiac status has been evaluated, the physician may seek consultation and/or begin troubleshooting the device.
History and Physical Examination. Most patients with ICDs who present to the ED have sustained one or more shocks that prompt an emergency visit. The EP should approach the patient in a calm, reassuring fashion while rapidly establishing the patient's cardiac status. The causes of ICD shocks are listed in Table 1. Generally speaking, shocked patients can be divided into two broad categories: 1) those who have a well-functioning device and have received appropriate shocks, and 2) those who have received inappropriate shocks. Most shocks are appropriate and do not represent new cardiac compromise.78 To help differentiate appropriate from inappropriate discharges, the pattern of the shocks and the events prior to and after the shocks have been delivered should be clearly elucidated. First, compare the patient's current frequency of shocks to those of the prior days, weeks, and months. Many authors consider the characteristics of the shocks important, and it may be useful to consider several patterns:79
· Repetitive shocks, or those shocks that occur within seconds of each other may occur in response to sustained stable VT or salvos of unstable VT ("ventricular storm"). They can also occur after spurious oversensing or SVAs that meet rate criteria.
· Recurrent shocks occur within minutes or hours of each other and are more likely to represent VAs that are being treated appropriately. Such discharges probably reflect a more stable substrate than repetitive shocks, but a thorough search for new problems is essential nonetheless. Some authors use the term "cluster" shocks (variably defined as > 2 shocks within 15 minutes, > 4 shocks per hour, > 7 shocks per 24 hours, or shocks only a few seconds apart), and have found that these patterns are independent predictors of an arrhythmic death.80-82
· Occasional shocks suggest a stable situation and a well functioning device. They may also represent "ghost shocks," which usually occur at night and are sensations that represent anxiety rather than actual ICD discharges.83
Hypotension and signs of heart failure will identify unstable patients, although the primary cause for clinical deterioration may be elusive. Ischemia and cardiac dysfunction must be ruled out in the patients, although ICD malfunction and sepsis from myriad sources, including the ICD, also should be considered.
A description of symptoms prior to or after ICD discharges should also be elicited. If the patient suddenly felt palpitations with significant weakness and diaphoresis, with resolution of symptoms post-shock, the device probably discharged appropriately. SVAs, however, also can present with these symptoms.84 Syncope related to multiple ICD shocks is worrisome and could represent an ineffectively treated and possibly newly malignant arrhythmia. Patients with syncope and no recollection of shocks or with just one shock is also worrisome because the patient may well have experienced multiple discharges without knowing it. On the other hand, if the patient notes no prodromal symptoms, an inappropriate shock is suggested. If he/she relates a story of exercising, hyperventilating, or shivering prior to discharge, myopotentials from the chest wall may have been oversensed by the device. If symptoms of cardiac ischemia preceded or followed the shock, the patient should be worked up accordingly.
Finally, in the ICD patient that presents to the ED having been shocked for whatever reason, consider the possibility of concurrent embolic phenomenon. As mentioned above, emboli, presumably from a diseased LV, can cause ischemia in the extremities and brain. Pain or stroke symptoms can be transient or persistent and occur immediately or hours after a shock.
Clinical Data Base. All patients who have been shocked require a monitored ED bed and a 12-lead electrocardiogram. As a rule, they should also have a chest x-ray and their serum potassium checked and, when appropriate, magnesium and calcium levels should also be checked. After these results are reviewed, a decision can be made regarding an ICD patients disposition. On occasion, a shock is witnessed while the patient is being monitored in the ED; if it is associated with a VA, the administered shock is appropriate and can be worked up accordingly as described below. If the ICD discharges during a nontarget rhythm or if there is no witnessed discharge but the rhythm in the ED is a new atrial arrhythmia, it is also likely that an inappropriate shock was delivered. An EKG of a patient who has been shocked recently is likely to show ischemic changes shortly after discharge, but these should not persist longer than 15 minutes if they are device-related.85
Workup generally includes a chest x-ray to confirm suspected congestive heart failure in patients whose ICD seems to have shocked appropriately, or to check for signs of lead fracture for a suspected inappropriate discharge. To detect lead fracture, a lateral view of the shoulder should be included. Note, however, that a frank break in a lead is rare and that endocardial microdislocation or tiny abnormalities within an intact lead will not be evident on radiograph.86 Lead migration, usually from the coronary sinus, a location that is now used rarely but still present in older ICDs, can be identified on chest x-ray. Such migrations and other intracardiac dislocations are generally asymptomatic and discovered serendipitously. An overpenetrated chest radiograph will identify the ICD if the patient is not carrying relevant paperwork. Serum tests for potassium, magnesium, and calcium should be considered since depletion of electrolytes, especially potassium, can trigger a VA and SCD. Cardiac enzymes should be checked if ischemia is considered, but it should be stressed that elevations in cardiac isoenzymes caused by DF can occur. It is unlikely that ICD shocks damage the myocardium.87
Device Interrogation. After the acute cardiac status has been established in the shocked ICD patient, the device should be "interrogated." All third-generation ICDs have easily accessible event telemetry with which an accurate assessment of the discharge can be made. All programmers are portable and can be wheeled to the ED, or patients can be brought to the ICD clinic.
Management of ICD Patients
Although ED management of the shocked ICD patient does not lend itself easily to an algorithmic approach, guidelines for common situations are helpful. (See Figure 9.) All cases must be discussed with a cardiologist and, preferably, the patient's electrophysiologist. Criteria for admission/discharge and immediate/elective device interrogation are by no means universal.
Patients who can be discharged. Many patients present to the ED looking and feeling well after sustaining a single shock. These patients may report no associated symptoms or only a brief period of weakness or palpitations. If the pattern of shocks is unchanged and the physical examination, monitored rhythm, EKG, chest radiograph, and potassium level are normal (or unchanged from baseline), most electrophysiologists would feel comfortable sending the patient home. In this case, the ICD does not necessarily need interrogation in the ED, as long as arrangements are made for prompt follow-up in the ICD clinic. Patients with multiple shocks generally require admission, though some with just a few or no change in the pattern of discharges and a benign evaluation can be discharged. They should not be discharged, however, without interrogation in the ED or immediately thereafter in the ICD clinic. No ICD patient who has been shocked more than once should be discharged from the hospital without evaluation by a cardiologist.
Patients who require admission. Patients who have had a frank syncopal episode, or a significant near syncopal event, even after sustaining only a single shock, generally require admission, regardless of the ED workup. They may well have been shocked multiple times without realizing it, a pattern suggesting a new cardiac event or ICD malfunction. In addition, except for those patients who have been optimized on medical and electrical therapy and report no change in their previous shock pattern, all patients who have had multiple ICD shocks require admission. In the hospital, cardiac status can be reassessed, and their ICD and medical therapy adjusted accordingly.
Patients who need immediate interrogation. Patients who require immediate interrogation, either in the ED or on the way to a monitored or intensive care unit bed include those whose cardiac status seems to be deteriorating. These include patients who are receiving appropriate shocks in the ED, have unstable vital signs, congestive heart failure, or an AMI. More stable patients will require ICD interrogation as part of their workup but do not need it immediately.
Patients with chest trauma. There are no case reports or data available in the ICD literature concerning the incidence of ICD malfunction as a result of trauma. A pacemaker case report suggests that skin erosion over the generator can occur after sufficient impact, however, and the EP should certainly consider component damage with sufficient or pinpoint trauma. Monitor the cardiac rhythm closely and maintain a low threshold for arranging interrogation.88
Deactivation of the ICD. As a rule, the ICD should not be turned off prior to consultation with the electrophysiologist. There are a few exceptions, however, including the patient who is in cardiopulmonary arrest. Patients in VT/VF or bradysystolic arrest who are 21 years and have absent or nearly absent pulses or blood pressure probably have an inoperative ICD and cardiac life support begun immediately-prior to device deactivation.89 Do not wait for the device to save the patient. Although there are now only a few patients with epicardial patches, these may interfere with external DF; therefore, if external DF is not successful immediately and the ICD is old enough to be implanted in the abdomen, the external defibrillator paddles should be repositioned in an anterior posterior position if initial DF is not successful.90 There are reports of shocks being sustained by those giving chest compressions in patients with functioning ICDs but the sensations are apparently more startling than harmful.91 Patients who have died should undergo device interrogation to assess functionality, and all ICDs should be turned off before leaving the ED.
In hospitals where an electrophysiologist is not available or cannot be reached immediately by phone, there are a few circumstances that should prompt the EP to deactivate the ICD. These circumstances include obvious device malfunction, such as delivering shocks during a rapid SVT or certain motions of the chest or arm that suggest a sensing or lead problem. Atrial fibrillation with rapid ventricular response can also be treated with standard IV medication to reduce the rate below the ICD rate criteria. External DF patches should be applied prior to deactivation. Emergent clinical advice is also available by calling the manufacturer directly.
To deactivate the ICD, place a standard donut magnet on the chest over the device. The magnetic field closes a switch in the generator circuit, causing a programmed response that differs with the model. (See Table 2.)
Summary
The number of patients with ICDs have increased in recent years. As technology advances and data accumulates, it is likely these devices will be used even more widely. Nearly all ICD patients have severe structural heart disease and most have other significant medical problems as well. The approach in the ED usually is straightforward. Obtaining an accurate history is crucial and may suggest a specific problem with the heart or ICD. If there is a new pattern of ICD discharges, an electrophysiologist can assist with rhythm identification and with the patients' final disposition.
References
1. National Center for Health Statistics. Advance report of final mortality. Mortality Vital Statistics Report. 1993;41:1-52
2. Gillum RF. Sudden coronary death in the United States: 1980-85. Circulation. 1989;79:756-765.
3. Davidson T, Van Riper S, et al. Implantable cardioverter defibrillators: A guide to clinicians. Heart Lung 1994;23:205-215.
4. Akhtar M, Jazayer MR, et al. Sudden cardiac death; magnitude of the problem, causes and approaches to prevention. In: Naccarelli GV, Veltri EP, eds. Implantable Cardioverter Defibrillators. Blackwell Scientific Publications; 1993:11.
5. Mirowski M, Reid PR, et al. Mortality in patients with implantable automatic defibrillators. Ann Intern Med 1983;98:585-593.
6. The Antiarrhythmia VS Implantable Defibrillator Investigators, A comparison of antiarrhythmic drug therapy with the implantable defibrillator in patients resuscitated from near fatal ventricular arrhythmia. N Eng J Med 1997;337:1576-1583.
7. Lehmann MH, Steinum, et al. The automatic implantable cardioverter defibrillator as antiarrhythmic treatment modality of choice for survivors of cardiac arrest unrelated to acute myocardial infarction. Am J Cardiol 1988;62:803-805.
8. Myerberg RJ, Castellanos A. Evolution, evaluation and efficacy of the implantable cardioverter defibrillator technology. Circulation 1992;86:691-693.
9. Lavey D, Sakena S. Management of refractory sustained ventricular tachycardia with amiodarone: A reappraisal. Am Heart J 1987;113:49-56.
10. Nisan S, Fogoras RN. Troubleshooting of patients with implantable cardioverter defibrillators. In: Singer I, ed. Interventional Electrophysiology. 1st ed. Baltimore: Williams and Wilkins; 1997: 793-823.
11. Schuder JC. The ICD-Progress, Prospects and Problems. PACE 1997;20: 2367-2370.
12. Owens DK,Sanders GD, et al. Cost effectiveness of implantable cardioverter defibrillators relative to amiodarone for prevention of sudden cardiac death. Ann Intern Med 1997; 126: 1-12.
13. Friedman PL, Stavenson WG. Unsustained ventricular tachycardia-To treat or not to treat. N Engl J Med 1996;335:1984-1985.
14. Bigger JT, et al. Prophylactic use of implantable cardiac defibrillators in patients at high risk for ventricular arrhythmias after coronary-artery bypass graft surgery. N Engl J Med 1997;337:1569-1575.
15. Dillon Read and Co; Cardiovascular Device Update, 1994.
16. American Health Consultants, Cardiovascular Device Update. 1997;3:1-4.
17. Higgins GL. III The automatic implantable cardioverter-defibrillator: Management issues relevant to the emergency car provider. Am J Emerg Med 1990;8:342-347.
18. Craig SA, Hudson AD. Emergency department management of patients with automatic implantable cardioverter-defibrillators. Ann Emerg Med 1990;19: 421-424.
19. Munter DW, Delacey WA. Automatic implanted cardioverter defibrillators. Emerg Med Clin North Am 1994;12:579-594.
20. Lehman MH, Saksera S. Implantable cardioverter defibrillators in cardiovascular practice: Report on the policy conference of the North American Society of Pacing and Electrophysiology. PACE 1991;14:969-972.
21. Akhtar M, Garan H, et al. Sudden cardiac death: Management of high risk patients. Ann Intern Med 1991; 14:499-512.
22. Akhtar M, Garan H, et al. Sudden cardiac death: Management of high risk patients. Ann Intern Med 1991;114:499-512.
23. Moss AJ, Hall WJ, et al. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmias. Multicenter Automatic Defibrillator Implantation Trial Investigators. N Engl J Med 1996;335:1933-1940.
24. Echt DS,Armstrong GK, et al. Clinical experience, complications and survival in 70 patients with the automatic implantable cardioverter/defibrillator. Circulation 1985;71:289-296.
25. Fogoras RN, Fiedler SB, Elson JJ. The automatic implantable cardioverter-defibrillator in drug refractory ventricular tachyarrhythmias. Ann Int Med 1987;107:635-646.
26. Mirowski M, Mower MM, et al. Recent clinical experience with the automatic implantable cardioverter-defibrillator. Cardio Clin 1985;3:623.
27. Kelly PA, Cannom DS, et al. The automatic implantable cardioverter-defibrillator: Efficacy, complications and survival in patients with malignant ventricular arrhythmias. J Am Coll Cardiol 1988;11:1278-1286.
28. Saksena S, Pozzobutt-Johanos M, et al. Long term multicenter experience with a second generation implantable pacemaker-defibrillator in patients with malignant ventricular tachyarrhythmias. J Am Coll Cardiol 1992;19:490 499.
29. Winkle RA, Mead RH, et al. Long term outcome with the automatic implantable cardioverter-defibrillator. J Am Coll Cardiol 1989;13:1353-1361.
30. Brady GH, Johnson G, et al. A simplified, single lead unipolar transvenous cardioverter-defibrillation system. Circulation 1993;88:543-547.
31. Oswald S, Cron TA, et al. First clinical experience with dual chamber sensing for differentiating tachyarrhythmias. PACE 1998;21:292-295.
32. Lavergne T, Daubent JC, et al. Preliminary clinical experiences with the first dual chamber pacemaker defibrillator. PACE 1997;20:182-188.
33. Geelan P, Lorgaf A, at al. The value of DDD pacing in patients with an implantable cardioverter defibrillator. PACE 1997;20:177-181.
34. Austin EH. Nonthoracotomy Approach to the Implantable Cardioverter defibrillator. In: Singer I, ed. Interventional Electrophysiology. 1st ed. Baltimore: Williams and Wilkins; 1997:703-724.
35. Nichols K, Collins J. Update on implantable cardioverter defibrillators: Know the differences in devices and their impact on patient care. AACN Clinical Issues 1995;6:31-43.
36. Cooper D, Luceri R et al. The impact of automatic implantable cardioverter defibrillators on quality of life. Clin Prog Electrophysiol Pacing 1986;4:306-309.
37. Grimm W, Flores BF, Machlinski FE. Electrocardiographically documented unnecessary spontaneously shocks in 241 patients with implanted cardioverter defibrillators. PACE 1992;15:1667-1673.
38. Jones GK, Gust HB. Implantation of ICDs in the electrophysiology laboratory. In: Singer I, ed. Interventional Electrophysiology. 1st ed. Baltimore: Williams and Wilkins; 1997:725-740.
39. Singer I, Dean G, et al. Clinical results with the electrophysiologist-implanted cardioverter defibrillator. PACE 1997;20:189-193.
40. Singer I. Defibrillation threshold testing and intraoperative ICD evaluation. In: Singer I, ed. Interventional Electrophysiology. 1st ed. Baltimore: Williams and Wilkins; 1997:793-823.
41. Soloman AJ, Swartz JF, et al. A second defibrillator chest patch electrode will increase implantation rates for nonthoracotomy defibrillators. PACE 1996;19:1304-1310.
42. Karasik P, Soloman A, et al. A patch in the pectoral position lowers defibrillation threshold. PACE 1997;20:1662-1666.
43. Groh, WJ, Foreman LD, Zipes DP. Advances in treatment of arrhythmias: Implantable cardioverter defibrillators. Am Fam Phy. 1998;57:303.
44. Pinski SL, Fahy GJ. The Proarrhythmic potential of implantable cardioverter defibrillators. Circulation 1995;92:1651-1664.
45. Hutse W, Tavernier R. Inappropriate shocks in a patient treated with a cardioverter defibrillator. Heart 1997;77:386-387.
46. Callans DJ, Hook BG, et al. Unique sensing errora in third-generation implantable cardioverter-defibrillator. J Am Coll Cardiol 1993;22:1135-1140.
47. Johnson NJ, Marchlinski FE. Arrhythmias induced by device antitachycardia therapy due to diagnostic nonspecificity. J Am Coll Cardiol 1991;18: 1418-1425.
48. Roelke M, Ununain S, at al. Ventricular Pacing induced ventricular tachyarrhythmias in patients with implantable cardioverter defibrillators. PACE 1995;18:486-490.
49. Sheahan RG, Dorian P, et al. Dual Device Therapy in the setting of changing ICD technology: device-device interaction revisited. PACE 1997;20: 1704-1707.
50. Geiger MJ, O'Neill P, et al. Interactions between transvenous nonthoracotomy cardioverter defibrillator systems and permanent transvenous endocardial pacemakers. PACE 1997;20:624-630.
51. Lawton JS, Wood MA, et al. Implantable transvenous cardioverter defibrillator leads: The dark side. PACE 1996;19:1273-1278.
52. Madrid A, Sanchez A, et al. Dysfunction of implantable cardioverter defibrillators caused by slot machines. PACE 1997;20:212-214.
53. McIvor ME. Environmental electromagnetic interference from electronic article surveillance devices: Interaction with implantable cardioverter defibrillators. PACE 1995;18:2229-2230.
54. Marco D, Evenger G, Hayes DL. Testing of work environments for electromagnetic interference. PACE 1992;15:2016-2022.
55. Hayes DL, Carrillo RG, et al. State of the science: Pacemakers and defibrillators interference from wireless communications devices. PACE 1996;19: 1419-1430.
56. Brooks R, Garan H, et al. Implantation of transvenous nonthoracotomy cardioverter-defibrillator systems in patients with permanent endocardial pacemakers. Am Heart J 1995;129:45-53.
57. Farre J, Asso A, et al. Antiarrhythmic drugs in patients with as automatic implantable defibrillator. PACE 1995;18:579-591.
58. Pfeiffer D, Jung W, et al. Complications of pacemaker-defibrillator devices: Diagnosis and management Am Heart J 1994;127:1073-1080.
59. Smith PN, Vidaliiet HJ, et al. Infections with nonthoracotomy implantable cardioverter defibrillators: Can these be prevented? PACE 1998;21:42-53.
60. Spratt KA, Emily AB, et al: Infections of implantable cardioverter defibrillators: Approach to management. Clin Infect Dis 1993;17:679-685.
61. Dunn CJ, Ruder M, Deresinski SC. Aspergillus Fumigatus infection of an implantable cardiac defibrillator. PACE 1996;19:2156-2157.
62. Spratt KA, Blumber EF, et al. Infections of implantable cardioverter defibrillators: Approach to management. Clin Infect Dis 1993;17:679-685.
63. Spratt KA, Blumber EF, et al. Infections of implantable cardioverter defibrillators: Approach to management. Clin Infect Dis 1993;17:679-685.
64. Lee JH, Geha AS, et al. Evolving Strategies in the management of postoperative ICD infections [Abstract]. PACE 1994;17:780.
65. Turkasher V, Priel I, Dan M. Successful management of an infected implantable cardioverter defibrillator with oral antibiotics and without removal of the device. PACE 1997;20:2268-2267.
66. Cooper D, Luceri R, et al. The impact of automatic implantable cardioverter defibrillators on quality of life. Clin Prog Electrophysiol Pacing 1986;4:306-309.
67. Vlag SC, Olson LC, et al. Anxiety and anger in patients with ventricular tachycardias. Responses after automatic internal cardioverter defibrillator implantation. Pacing Clin Electrophysiol 1989;12:366-373
68. Morris P, Bayer J, et al. Psychiatric morbidity following implantation of the automatic implantable cardioverter defibrillator. Psychosomatics 1991;32:58-64.
69. May C, Smith P, et al. The impact of the implantable cardioverter defibrillator on quality of life. PACE 1995;18:1411-1418.
70. Dunbar SB, Warner CD, Purcell JA. Internal cardioverter defibrillator device discharge: Experience of patients and family members. Heart Lung 1993;22:494-501.
71. Krug, Zerbe F. Major venous thrombosis: A complication of transvenous pacemaker electrodes. Br Heart J 1980;44:158-161.
72. Block M, Hammel D, et al. Results and realistic expectations with transvenous lead systems. PACE 1992;15:665-670.
73. Benedini G, Marchini A. Implantable defibrillators and thromboembolic events. PACE 1995;18:199-202.
74. Leonelli FM, Salley R, et al. The vanishing defibrillator syndrome. PACE 1997;20:960-965.
75. Chitkara VK, Vlay SC. Erosion of internal cardioverter defibrillators pocket and migration of pulse generators into the peritoneal cavity. PACE 1996;19: 1528-1529.
76. Higgins SL, Suh BD, et al. Recurrent Twiddler's Syndrome in a nonthoracotomy ICD system despite a dacron patch. PACE 1998;21:130-133.
77. Avitall B, Stormo A, et al. Sudden cardiac death: Twiddler's syndrome with an implantable cardioverter defibrillator. Am Heart J 1994;128:833-836.
78. Luceri RM. Implantable defibrillators: Physician's role after hospital discharge. In: Naccarelli GU, Veltri EP, ed. Implantable Cardioverter Defibrillators. Blackwell Scientific Publications, 1993:150-170.
79. Luceri RM. Implantable Defibrillators: Physician's Role After Hospital Discharge. In: Naccarelli GU, Veltri EP, eds. Implantable Cardioverter Defibrillators. Blackwell Scientific Publications, 1993:150-170.
80. Villacastin J, Almendral J, et al. Incidence and clinical significance of multiple consecutive, appropriate, high energy discharge in patients with implantable cardioverter-defibrillators. Circulation 1996;93:753-762.
81. Kluger j, Veronneau J, Fisher J. Cluster shocks: A strong predictor of subsequent mortality in patients with internal cardioverter defibrillator. PACE 1995;18:II-845.
82. Kohno T, Bhandar AK, et al. Cluster shocks in patients with third generation implantable defibrillators: Frequency, mechanics and clinical significance. PACE 1995;18:II-846.
83. Miller JM, Hsia HH. Management of the patient with frequent discharges from implantable cardioverter defibrillator devices. J Cardiovasc Electrophysiol 1996;7:278-285.
84. Grimm W, Flores BF, Marchlinsk FE. Symptoms and electrocardiographically documented rhythm predicting spontaneous shocks in patients with implantable cardioverter defibrillators. Am J Cardiol 1993;71:1415-1418.
85. Eyesmann SB, Machlinski FE. Electrocardiographic changes after cardioversion of ventricular arrhythmias. Circulation 1986;121:231-235.
86. Anderson MH, Wood DE, et al. Radiologic appearance of implantable defibrillator systems. Clin Radiol 1995;50:29-39.
87. Avitall B, Port S, et al. Automatic implantable cardioverter/Defibrillator discharge and acute myocardial injury. Circulation 1990;81:1482-1487.
88. Sakakibara Y. Delayed pinpoint exposure of a pacemaker following seat belt trauma. PACE 1997;20:370-371.
89. Chapman PD. The implantable defibrillator and the emergency physician. Ann Emerg Med 1989;18:578-582.
90. Lerman BB, Deale OC. Effects of epicardial patch electrodes on transthoracic defibrillation. Circulation 1990;81:1409-1414.
91. Lechleuthner A. Electric shocks to paramedic during cardiopulmonary resuscitation of patient with implanted cardioverter defibrillator. Lancet 1995;345:253.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content