Chapter 35. Automatic Implantable Cardioverter-Defibrillator Assessment

The introduction of implantable cardioverter-defibrillator (ICD) technology has revolutionized the fields of cardiology and electrophysiology. More than 100,000 such devices are implanted annually in the United States alone. ICDs allow life-threatening ventricular tachycardia and ventricular fibrillation to be safely controlled and benefit patients at risk for sudden cardiac death. Multiple studies (e.g., CABG patch, MADIT, MADIT II, MUSTT, DINAMIT, AMIOVIRT, COMPANION, SCD-HEFT) have examined the prophylactic indication for ICD therapy in high-risk groups.1–7 The ICD is becoming a more common therapeutic option for the young population with a diagnosis of Brugada syndrome, prolonged QT syndrome, and hereditary cardiomyopathies to name a few.

The Emergency Department is often the initial contact point for these patients. Now more than ever Emergency Physicians must be familiar with the problems that can be encountered by a patient with an ICD. This chapter describes technical aspects, basic interrogation of the device, and a general approach to a patient who presents to the Emergency Department with an ICD.

The ICD has four main functions. It recognizes and records local atrial and ventricular electrogram signals. It then classifies the sensed signals according to programmable heart rate zones. The ICD provides therapy (i.e., a shock) to terminate ventricular tachycardia or ventricular fibrillation. It has a pacing capability for bradycardia and/or cardiac resynchronization therapy. When detection criteria are satisfied, therapy to terminate the arrhythmia is initiated with high-energy shock of up to 40 Joules (J).

ICD technology has progressed exponentially since its introduction by Mirowski and colleagues in the early 1980s.8 The ICD system is comprised of a pulse generator, a battery, and a lead system. The lead system is required for sensing, pacing, and the delivery of therapy. Earlier systems required that the pulse generators be placed abdominally due to their large size (Figure 35-1). Defibrillation was delivered via two epicardial patches positioned anteriorly and posteriorly. Occasionally, a transvenous spring electrode in the superior vena cava was utilized with an epicardial patch. Sensing was achieved through separate epicardial screw-in electrodes. Initial lead placement required either a sternotomy, lateral thoracotomy, or a subxiphoid approach, making early implants quite cumbersome.9

The smaller size of the newer devices allows for superficial implantation of the pulse generator in the anterior chest wall, similar to a pacemaker (Figure 35-2). The current ICD systems are comprised of three main parts. The pulse generator is programmable and capable of analyzing and recording the patient's heart and rhythm. The ICD generator houses the batteries, high-voltage capacitors, and microprocessors necessary to process sensed intrinsic cardiac electrical activity. In essence, the generator is a minicomputer within a hermetically sealed titanium can (aka case) capable of generating shocks. Typical ICDs contain lithium silver vanadium oxide cells that store between 2 and 7 Volts (V).12 The high voltages necessary for defibrillation are generated with the aid of high-voltage capacitors that are able to generate 700 to 800 V of defibrillation energy in under 20 seconds. The lead system or wires are inserted into the right ventricle for single chamber devices or both the right atrium and right ventricle for dual chamber devices. The leads are required for sensing, pacing, and the delivery of therapy. The final component is the battery to power ICD system.

ICD implantation has evolved quite rapidly due to advancements in lead technology, generator technology, and the development of biphasic defibrillation waveforms, which lowered the energy requirements necessary for successful defibrillation.10 The creation of a (bipolar) lead combining pacing and sensing capabilities with a high-voltage electrode coil allowed for nonthoracotomy system implants, which reduced surgical morbidity and mortality.11 The leads were now positioned transvenously via the subclavian vein and fixed to the inside of the right ventricle. However, the leads still had to be tunneled subcutaneously to the abdomen, as the generators remained fairly large. Technology has advanced the development of more compact generators. The smallest commercially available devices today are under 40 cm3 and weigh well under 100 grams. Smaller generators allow for subcutaneous pectoral implantation and simplification of the implantation process (Figure 35-2).12

Current ICDs allow extensive programmability for tiered antitachycardia pacing (ATP), tiered high voltage therapies, single or dual chamber bradycardia pacing, supraventricular tachycardia (SVT) discrimination algorithms, and detailed diagnostics of tachycardic and bradycardic episodes. Implantable loop recorders and home monitoring (HM) functions extend the technical capabilities for automatic detection of arrhythmias that may not be symptomatic. It also allows for a fully automatic and wireless data transmission, including episode counters.9,10 This allows alterations in device programming or medication dose modifications in the outpatient setting, thus avoiding hospitalization.13,14

ATP may be enabled to manage ventricular tachycardia. ATP commonly consists of a burst of pacing (i.e., 6 to 10 beats) at a rate faster than the ventricular tachycardia rate. ATP may be felt by the patient. It is painless and often terminates the ventricular tachycardia before the patient becomes symptomatic. ATP is the initial preferred treatment, even for fast ventricular tachycardias, with shocks programmed as a backup if ATP fails.15 In addition, current ICDs have an antibradycardia function that works like a pacemaker. If the heart rate falls below the programmed rate, the ICD will provide a pacing function with significantly less forceful shocks.

Exercise testing should be performed with the realization that increasing the heart rate above the programmed ventricular tachycardia and ventricular fibrillation detection rate is likely to elicit therapy (i.e., a shock) from the ICD. Consult the patient's Electrophysiologist before any exercise or stress testing. Determine the programmed the ICD arrhythmia detection rate. During the exercise or stress test, if the programmed rate is approached, stop the test to avoid the delivery of inappropriate therapy (i.e., a shock).

A systematic follow-up procedure to assess the integrity of the ICD system is recommended every 1 to 6 months depending upon the device, battery status, and arrhythmia frequency (Table 35-1). The device should be interrogated with the appropriate system analyzer to assure battery voltage, lead and electrode integrity, and capacitor charge times. System integrity test results can reveal malfunctions that may compromise the device's ability to treat an arrhythmia effectively.16,17 Once the system's integrity has been verified, each recorded arrhythmic episode should be analyzed to determine the appropriateness and effectiveness of therapies. This is achieved by examining the stored diagnostic information contained within the ICD, which may include intracardiac electrograms, arrhythmic interval values, classification markers for each interval, episode plots, textual episode descriptions, energy, charge time and impedance values for shocks, and device classifications of therapy success. This diagnostic information can be extremely valuable to determine whether programming changes should be made.18

Battery and Capacitors

Battery longevity is significantly different among manufacturers. Battery voltages and capacitor charge times must be noted. Capacitors with smaller or larger capacities and changes in the initial shock polarity have not significantly improved defibrillation efficacy.19 A variation in the number and position of electrodes can significantly influence the defibrillation threshold.19 Generator replacement is usually recommended when voltages fall below the elective replacement indicator (ERI), which is approximately 2.6 V. When battery voltage falls below 2.2 V, the battery has reached its end of life (EOL). This signifies a more urgent need for battery replacement. The need to replace a generator is also dependent on capacitor charge times. Two consecutive charge times greater than 16 seconds is considered prolonged and may warrant urgent replacement regardless of the battery voltage. Automatic capacitor reformation is usually set to every 6 months to replenish their charge. This function can be programmed at preset time intervals. More frequent reformation is required and performed as the device reaches the battery EOL.

Lead Integrity

With the growing number of prophylactic ICD implantations and generator replacements, concern about the long-term reliability of chronically implanted leads increases. ICD lead failure can be caused by an insulation defect or conductor disruption. Lead failure can affect the high voltage lead or the pace–sense circuit of the lead. Newer ICDs provide lead-impedance monitoring for early detection of lead failure. An abnormal impedance indicates conductor fracture or insulation defect. An audible signal (Patient Alert™) notifies the patient who should then immediately contact their Cardiologist or present to the Emergency Department. Retrospective data from single centers suggest a potential clinical benefit of Patient Alert™ for ICD lead failure detection.20

Various parameters are checked to ensure lead integrity, including pacing thresholds, lead impedance, and the size of intracardiac R waves. At the time of implantation, a pacing threshold of ≤1.0 V at a pulse width of 0.5 milliseconds is desired. An acute rise in threshold may be seen initially; it generally falls with time. Chronic thresholds range between 0.5 and 2.0 V. Epicardial leads generally exhibit higher thresholds. Any change in thresholds must be compared with prior trends to assess its significance. Pacing thresholds may also change with the administration of antiarrhythmic drug therapy. Potential complications of ICD lead failure include oversensing of electrical noise, undersensing of ventricular tachyarrhythmias, inappropriate therapy, and lethal antiarrhythmic therapy.21

Current devices measure lead impedance only once a day. It is unlikely that such discrete measurements will reveal abnormal impedance if lead failure causes sporadic dysfunction. Thus, multiple impedance measurements per day may be necessary to enhance the sensitivity of the Patient Alert™.22 Lead impedances will vary based on the type of lead. The high voltage impedance measurement differs between devices.23 Normal impedances range from 300 to 1200 Ohms (Ω). A sudden change in impedance may signal a problem with lead integrity. A high impedance reading suggests the possibility of lead or patch fracture. A low impedance indicates a problem with the insulation. The impedance of the high-voltage coil must also be evaluated. Epicardial and active can systems demonstrate lower impedance values compared to endocardial systems due to their larger surface area. Normal values generally range between 20 and 80 Ω. High impedance values, as with the pace/sense leads, usually indicate a fracture in the defibrillation system. Any stark changes in impedance levels may suggest a patch problem. This includes crinkling, seroma formation, or migration. Any impedance change in an endocardial system may indicate lead dislodgement.

R-wave amplitude is a direct measure of intracardiac electrogram activity and determines the device's ability to sense. At implantation, an R wave of ≤5 mV is desired to ensure adequate detection of ventricular fibrillation. The most common explanations for a decrease in R-wave amplitude after an implant are lead dislodgment or local factors such as edema or fibrosis. This change may be associated with an increase in lead impedance and pacing thresholds.

Analyzing Appropriateness of Therapy

Current ICDs have simultaneous marker channels with real-time intracardiac electrograms (Figure 35-3). It is important to document intracardiac activity during sinus rhythm. These electrograms can be used as a basis of comparison with tachycardia events. The signals should be examined for evidence of noise, which may be an indicator of sensing problems, a connector issue, or a faulty adaptor. In the event of noise detection, electrograms must be examined during various maneuvers including deep breathing, arm maneuvers, and bending. Intracardiac electrograms should be examined for T-wave oversensing or detection of pacemaker spikes as ventricular signals. In dual-chamber devices, oversensing by the atrial lead should be ruled out. Wide variations in electrogram size may suggest that the lead is not stable or well fixed.

A patient with an implantable defibrillator may present to the Emergency Department for various reasons. Manufacturers offer 24 hour technical support. They also maintain a registry of devices implanted and evaluations performed. A staff of field engineers are often available for help with the assessment, interrogation, and management of ICD function. The US Food and Drug Administration (FDA) is responsible for assessing both the premarket and postmarket safety and effectiveness of medical devices marketed in the United States. They require manufacturers to submit annual reports detailing the number of device implants and malfunctions that have occurred.24

Individuals with recently implanted devices may be quite anxious and may seek medical attention even after a single ICD discharge. Patients may also present after multiple ICD discharges or in full cardiac arrest, requiring cardiopulmonary resuscitation (CPR). The warmth, local redness, and pain associated with a potential infection of the ICD pocket may prompt the individual to seek medical attention. Questions regarding interactions between ICDs and electromagnetic interference (EMI) will arise more frequently. These clinical situations are discussed in more detail below.

Patients with ICDs must be placed on continuous telemetry. Perform a detailed and complete history and physical examination. The patient should be questioned regarding the number of shocks received, symptomatic palpitations, presyncope, symptoms of congestive heart failure, or symptoms consistent with angina. The pocket site must be examined for evidence of local infection, including warmth, tenderness, and discharge. Upper extremity or neck swelling ipsilateral to the inserted endocardial lead suggests the possibility of a subclavian vein or superior vena cava thrombosis.25 Bilateral head edema and neck vein swelling suggest the possibility of a superior vena cava thrombosis or a superior vena cava syndrome. It is important to document any changes in medications or whether antiarrhythmic therapy has recently commenced. Routine blood work should be obtained, including a complete blood count, serum electrolytes, magnesium level, renal function indices, and quantitative levels of measurable medications.

The ICD model should be identified. Patients are generally given identification cards that list the manufacturer, lead system, generator model, and a 24 hour emergency contact number. Often this information is not available. However, in an emergency situation, an overpenetrated chest radiograph showing the generator will demonstrate the radiopaque identifier of the manufacturer.

The chest radiograph may be of diagnostic importance. It is not only important for device identification but can also give useful information about lead integrity. In a recent implant, a chest radiograph is routinely performed to assure lead positioning, slack in the lead, and to rule out other thoracic pathology. In patients with epicardial patches, the chest radiograph is an excellent tool for demonstrating patch crinkling, fracture, or migration. With endocardial systems, the lead can again be assessed for fractures or discontinuities. Fractures can occur anywhere along the lead. They are most commonly seen near the junction of the first rib and the clavicle, a condition referred to as “subclavian crush.”

If the patient presents with a ventricular arrhythmia and is hemodynamically stable, attempts should be made to obtain a 12-lead electrocardiogram (ECG) before any interventions are made to terminate the arrhythmia. Often this may be difficult due to concomitant discharges from the device, an anxious staff, and an anxious patient. Antiarrhythmic medications such as amiodarone, procainamide, and β-blockers may have to be considered in the event of recurrent ventricular tachycardia. Despite the absence of clinical trials examining the efficacy of lidocaine, it is generally considered the treatment of choice in the setting of an acute myocardial infarction or acute ischemia if the patient is experiencing a ventricular arrhythmia.

The ability of an ICD to identify and treat tachyarrhythmias can be temporarily disabled with the use of a magnet. This situation may arise in the setting of multiple ICD discharges where the shocks are not tolerated or prior to a surgical procedure where electrocautery is necessary. Emergency deactivation with a magnet can result in serious complications such as causing a battery indicator to switch to “end of life” and the loss of some antitachycardia therapies.26 The application of a donut-shaped magnet overlying the ICD pulse generator forms a magnetic field that trips a reed switch in the ICD generator circuit. This results in a suspension of tachycardia detection and therapy delivery. Generally, a single magnet will suffice. In obese patients or in the presence of pockets with significant edema, two or more magnets may be required to achieve deactivation.27 The patient must be fully monitored, a cardioverter-defibrillator unit must be readily available, and Advanced Cardiac Life Support (ACLS) medications must be readily available if an ICD is to be deactivated.

The magnet response of an ICD varies subtly from manufacturer to manufacturer. In Medtronic devices, the application of a magnet temporarily disables tachycardia detection and therapy with no effect on bradycardia pacing. Removal of the magnet will resume arrhythmia detection. When activated, newer Medtronic ICDs (Gem II DR, patient alert function) will elicit a continuous beep lasting for 15 seconds if a magnet is placed directly over the ICD. A magnet applied over Guidant (CPI) ICDs also inhibits tachycardia therapy with no effect on bradycardic pacing. These devices will generate beeping tones, which change to a continuous tone. The constant tone indicates that the device is off and will not deliver tachycardia therapy. The device can be turned back on by reapplying the magnet over the ICD for 30 seconds. Tones will now change from continuous to beeping synchronous with R waves, signifying that the device is on again. Newer-generation Guidant ICDs (Prism II) have a built-in electrocautery feature that can be activated by use of the Guidant programmer. This will suspend tachyarrhythmia therapies and pace in the DOO mode. Regular functioning of the ICD is restored by turning this feature off.

Patients who experience a shock but feel unwell after the event or who receive more than one symptomatic ICD therapy within a short period of time (i.e., minutes to hours) require emergent evaluation.27,28 Most ICDs have a limit to the number of shocks it can generate before it stops producing shocks. This number is usually five to seven shocks. However, if the patient's cardiac rhythm reverts to normal sinus rhythm, the ICD can reset and start shocking the patient again to the set maximum number of shocks. Multiple ICD discharges or shocks in a short period of time can cause severe battery depletion. Ongoing arrhythmias not adequately treated by the device, myocardial infarction, electrolytes imbalances, and ICD malfunction are all possible etiologies that warrant medical attention.

Emergency Department patients presenting with multiple ICD discharges require immediate attention. From a psychological perspective, multiple discharges are usually not well tolerated and can be emotionally devastating to the patient.25 Myocardial injury and transient reduction in left ventricular function can occur as a result of multiple shocks.28 This has been associated with a poorer long-term prognosis. Multiple discharges can lead to premature depletion of battery life.

Establishing the Etiology

It is important to establish the etiology of the shocks in order to administer proper and prompt management (Table 35-2). It is of the utmost importance to determine whether the shocks are appropriate for ventricular tachycardia or fibrillation, inappropriate therapy, or phantom shocks.49 More than one-third of patients with a history of ventricular tachycardia or ventricular fibrillation receive a shock within 2 years of the ICD implantation.29 Recurrent ventricular tachyarrhythmia is a common cause of repeated ICD firing. Ineffective termination of a tachyarrhythmia in this situation can be the result of an increase in defibrillation thresholds secondary to concomitant antiarrhythmic drug therapy and lead migration or lead dislodgement. Inefficient termination can occur if inappropriately low amounts of energy are programmed for the initially administered shock. Shocks may also be the result of inappropriate detection of SVTs, the most frequent of which include sinus tachycardia and atrial fibrillation.30 The administration of a low-energy shock may convert a benign SVT into an unstable ventricular arrhythmia resulting in an ICD proarrhythmia. The introduction of an atrial lead in dual-chamber devices has aided in the discrimination process between SVTs and ventricular tachyarrhythmias. Rapid SVTs are particularly a problem in children and athletic individuals in whom exercise or reductions in medications that slow the heart rate, such as β-blockers, are commonly encountered causes of inadequate shocks from sinus or SVT. Inappropriate ICD firing can occur because of the erroneous detection of noise or interference that can be the result of insulation breakdown or a loose set screw (Figure 35-4). Oversensing of T waves, pacing artifacts, R waves, and electromagnetic interference may also lead to inappropriate detection and discharge. Patients who have received painful shocks occasionally suffer from phantom shocks, which are the perception of a shock in the absence of any arrhythmia or therapy from the ICD.49 Finally, random component failure should be considered if all other causes have been ruled out.

Approach to the Patient with Multiple ICD Discharges

These patients must be under constant ECG monitoring. Apply defibrillator pads in anticipation of the development of an unstable cardiac arrhythmia. In devices with limited stored diagnostic capabilities, this may be the only means of establishing a shock-rhythm correlation. Sedation is reasonable in extremely anxious patients. Once the patient is stabilized, obtain an electrophysiology consultation for assistance in interrogating the ICD. The device should be interrogated and stored electrograms obtained for analysis.31 For instance, it is often useful to inquire about the pattern of ICD discharge. Consecutive shocks occurring within a few seconds suggest an inappropriate discharge for SVT, oversensing, or device failure. On the other hand, isolated shocks occurring every few minutes may be indicative of recurrent ventricular tachycardia. Progressive dyspnea on exertion, shortness of breath, orthopnea, or paroxysmal nocturnal dyspnea suggests new onset or worsening heart failure, which can precipitate ventricular arrhythmias. Potential reversible causes, such as electrolyte abnormalities, need to be identified and promptly corrected.

Careful examination of the 12-lead ECG is crucial. Specific ST-segment changes may imply an acute coronary syndrome and determine the need for primary intervention or thrombolytic therapy. An ECG obtained during an actual shock may establish whether the culprit arrhythmia is a supraventricular or ventricular tachycardia.

If the discharges are inappropriate, the ICD should be emergently deactivated, as previously described. Supraventricular tachyarrhythmias should be managed with intravenous drug therapy, such as adenosine, diltiazem, or verapamil. In the situation where discharges are secondary to rapid atrial fibrillation, attempts must be made to control the patient's ventricular response with atrioventricular (AV) nodal blocking agents such as diltiazem, verapamil, β-blockers, and digoxin. Chemical cardioversion or electrical cardioversion may be attempted in the event of hemodynamic instability. Shocks secondary to prolonged episodes of nonsustained ventricular tachycardia can be prevented by adjusting initial detection parameters coupled with the addition of antiarrhythmic drug therapy.

Patients with an ICD can develop “electrical storm.” Patients in electrical storm require immediate attention. This condition involves recurrent, hemodynamically unstable ventricular tachycardia or fibrillation occurring two or more times in a 24-hour period.32 Potential triggers can be found in approximately 66% of patients and include new or worsened heart failure, changes in antiarrhythmic medication, psychological stress, and hypokalemia. In most patients, electrical storm consists of monomorphic ventricular tachycardia indicating the presence of a reentry mechanism. Ventricular fibrillation is rare and may be indicative of acute ischemia. The key intervention in electrical storm is reduction of the elevated sympathetic tone by intravenous β-blockers, benzodiazepines, and amiodarone.33

An Electrophysiologist should be promptly consulted and the patient stabilized prior to transfer to the intensive care unit (ICU). Defibrillator pads should be applied in anticipation of the development of unstable cardiac arrhythmias. Potential reversible causes, such as electrolyte abnormalities, need to be identified and promptly corrected. If torsades de pointes is established, the treatment of choice is magnesium and/or temporary cardiac pacing. Thrombolysis or urgent catheterization/intervention may be needed in the setting of an acute myocardial infarction. If analysis of stored electrograms demonstrates ineffective discharges, the ICD should be deactivated. Attempts to terminate the arrhythmia in the hemodynamically stable patient via ATP is a useful option. Intravenous antiarrhythmic drugs are a necessary adjunct in these situations. The administration of amiodarone in combination with β-blocker therapy has been shown to be successful in the management of electrical storm.34 Often a combination of antiarrhythmic drugs will be required. Bretylium should be reserved until other antiarrhythmic agents have failed.

Patients who have an ongoing arrhythmia when evaluated emergently should be managed according to ACLS guidelines regardless of the presence of an ICD. Although in the vast majority of instances ICD function will be found to be appropriate, this cannot be assumed.18 ACLS guidelines should be followed and the device considered inactive.35,36 If the clinical situation permits and an ICD programmer is readily available, the device should be deactivated. This will prevent the reinduction of ventricular fibrillation or tachycardia due to concomitant ICD discharges that may occur during CPR. There is often some hesitation to initiate resuscitative measures in patients with an ICD for fear of getting shocked. This fear is understandable but unwarranted. Although a mild electric shock might be perceived, these discharges do not pose a risk to persons administering CPR, nor do they damage external monitoring devices.

External defibrillation is permissible, although the paddles or pads should be positioned away from the ICD. Individuals with epicardial patches may require higher energies for defibrillation. Current can be shunted from the myocardium through the patches. Moreover, the insulated portion of the patch serves as a shield from the administered shock. In patients with epicardial patches, an anteroposterior paddle configuration has been suggested for changing the defibrillation vector.37

The ICD may be reset after external defibrillation is delivered, especially if the paddles or pads are located in close proximity to the generator. It is important that ICDs be reinterrogated after the successful completion of a resuscitation to ensure that programmed parameters have not been altered, including the pacing thresholds.

Technological advances have made a gradual reduction in the size of the pulse generator possible. This small size permits superficial implantation of the pulse generator in the anterior chest wall.38 The ICD implantation technique is similar to pacemaker implantation. ICD infections can involve the generator pocket, the leads, or both. Infection is more likely after a recent generator replacement.39 An infected ICD system represents a serious medical situation that should be dealt with urgently. The incidence of infection ranges from 2% to 11% in systems that were implanted via thoracotomy or sternotomy.40 The infection rates for nonthoracotomy implants approach those of pacemakers and range from 0.8% to 1.5%.39–43 Infections generally present clinically within 6 months of the implant, but more typically within the first 3 months. Infection should be suspected when local and systemic signs and symptoms of inflammation are apparent. Prompt referral to a specialist is warranted. In almost all cases, removal of the ICD and all leads is required. Endocarditis prophylaxis with antibiotics is not generally warranted.

Systemic symptoms are seen in up to 50% of patients, especially in those with infections caused by Staphylococcus aureus.44 The pocket and/or incision site is often visibly erythematous, warm, and tender. A fever may be present. Blood work often reveals a leukocytosis. Frank suppuration or device erosion may be seen. Pericarditis may be evident if epicardial patches are infected. Blood cultures may aid in documenting the culprit organism. Infections occurring late are generally indolent and rarely present with fever or leukocytosis; moreover, blood cultures are generally negative.

The most common microorganisms in 50% of infections include S. aureus and coagulase-negative staphylococci. Other common organisms include Escherichia coli, Pseudomonas, Serratia, Corynebacterium, Propionibacterium acnes, Candida spp., Streptococci, and atypical mycobacteria.45 Infection is generally the result of skin contamination during implantation of the ICD. It can also occur due to hematogenous seeding from distant intravenous sites, indwelling catheters, or from concomitant respiratory or urinary tract infections.

Approach to an Infected ICD

The goals of therapy include identifying the culprit organisms, establishing the extent of the infection, and containment of the infection. In the Emergency Department, routine laboratory tests should be obtained that include a complete blood count and differential. Blood for cultures should also be drawn, but it must be kept in mind that they are often negative. Wound cultures and gram stains may be helpful in differentiating an infection from a pocket hematoma, sterile subcutaneous fluid accumulation, or an inflammatory reaction to pacemaker components. Attempts to aspirate the pocket should be performed in consultation with an Electrophysiologist and/or a Surgeon. A sterile pocket or hematoma can often become infected after an aspiration.

There is no single diagnostic modality that can determine the extent of the infection. ICD infections should never be assumed to be localized, as organisms can migrate from the leads into the heart. In patients with epicardial patches, a chest radiograph may reveal patch deformities or wrinkling, suggesting distal migration of the infection. A computed tomography (CT) scan can also detect localized fluid accumulation and patch wrinkling. Echocardiography has been utilized to confirm the presence of vegetations on the leads or coil. Gallium and indium scans may be helpful in localizing an infection.

Although a few reports suggest that infection of a defibrillator system may be controlled, the treatment of choice continues to be removing the entire system followed by the administration of parenteral antibiotic agents. Some localized infections restricted to the ICD generator have been managed with the removal of the generator, debridement, and systemic antibiotic therapy. Vancomycin is frequently used as an empiric agent when cultures are still pending given its good coverage against coagulase-negative staphylococci and methicillin-resistant S. aureus (MRSA). Empiric gram negative and fungal coverage may be necessary in the immunocompromised patient.

The ability of ICDs (and pacemakers) to function is dependent on their ability to sense intrinsic cardiac electrical activity. Hermetic shielding, filtering, interference rejection circuits, and bipolar sensing have safeguarded ICDs (and pacemakers) against the effects of common electromagnetic sources. However, exposure to electromagnetic interference (EMI) may still result in oversensing, asynchronous pacing, ventricular inhibition, and spurious ICD discharges. EMI may also lead to loss of output, increased pacing thresholds, and decreased R-wave amplitude. Common sources of EMI include cellular phones, electronic article surveillance (antitheft) devices, and metal detectors. Occupational sources of EMI include high-voltage power lines, electrical transformers, arc welding, and electric motors. Interference can also be encountered through medical equipment and procedures such as magnetic resonance imaging, electric cautery, spinal cord stimulators, transcutaneous electric nerve stimulator units, radiofrequency catheter ablation, therapeutic diathermy, and lithotripsy.46

Patients with ICDs and pacemakers should be instructed to avoid environments with large magnetic fields. The use of cellular phones is permissible. However, the US Food and Drug Administration (FDA) recommends that direct contact of the phone with the device should be avoided. Use of the contralateral ear while using the phone is suggested. Although inappropriate shocks have been documented through electronic article surveillance systems, recent studies have deemed it safe for patients to walk through these systems as long as they avoid lingering around these devices. New concerns have been raised regarding household appliances. Patients can be reassured that EMI is unlikely to affect their devices if induction ovens are used in their kitchens.47

The strong magnetic fields of MRI can interfere with ICD functioning and induce electrical current flow in the ICD lead that can initiate an arrhythmia or be sensed as an arrhythmia and precipitate spurious therapies. In general, patients with an ICD should not be placed in an MRI field. Ongoing work is aimed at developing MRI-compatible ICDs. There is no such problem with fluoroscopy, CT scanning, or nuclear-based imaging.

There are constant improvements being made in ICD technology. A minimally invasive, entirely subcutaneous ICD (S-ICD®) has been developed (Cameron Health Inc., San Clemente, CA). The advantages include a simplified implantation procedure, it does not require vascular access or intracardiac leads, no imaging or fluoroscopy equipment is required, and the device may be as effective as current ICD systems.48 Potential complications such as deep venous thrombosis, intracardiac thromboses, tricuspid valve disease, as well as those related to central venous access can be eliminated. The electrode is implanted parasternally, and the ICD generator is implanted on the anterolateral chest wall. Small trials have demonstrated that this system can effectively treat ventricular fibrillation.48 Further clinical trials are required before this technology can be routinely implanted.

Expanding clinical indications, advanced new technology, and the increasing number of annual implants require Emergency Physicians to become familiar with problems typically encountered by the patient with an ICD. Complications associated with ICDs are not uncommon. Troubleshooting and programming should ideally be performed in conjunction with a trained Electrophysiologist. Patients presenting with cardiopulmonary arrest may require the device to be deactivated and external defibrillation performed. Do not apply external defibrillation paddles directly over the ICD.