An ICD is the first line of therapy for prevention of sudden cardiac death, which is commonly the result of ventricular fibrillation (VF) or ventricular tachycardia (VT).⁴ The first internal defibrillator was implanted in 1980 by Mirowski and Mower. Since then, the technology and indications have grown enormously. More than 120,000 devices are implanted in the United States every year.⁵ Given this scenario, emergency physicians (EPs) face a growing number of patients with increasing complexity who are arriving at EDs with ICD-related complaints.

General Concepts of Function

Current ICDs correspond to third-generation devices, which are small (40 mL) and reliable and contain sophisticated electrophysiologic analysis algorithms. They can store and report a large number of variables, such as electrocardiograms (ECGs), defibrillation logs, energies, lead impedance, and battery charge.⁶ ICDs are usually implanted in the left infraclavicular area via a transvenous technique.

The ICD unit consists of a case containing the battery, circuitry, and pulse generator; a right ventricular apex lead for sensing and defibrillation; and an atrial lead. Current devices for biventricular pacing can be equipped with a coronary venous lead. The diagnostic and treatment functions are configured during placement of the device, along with determination of the defibrillatory threshold (DFT) necessary for the specific patient.^6,⁷ Typically, the ICD is set to deliver energies 5 to 10 J above the DFT. According to these specifications, the battery life of the modern lithium-silver-vanadium device is approximately 8 years but depends largely on the frequency of the shocks delivered. Most of the devices are equipped with a patient alert system that prevents clinically significant battery-related complications.⁷

The algorithmic criteria for delivering a shock are based largely on the rate, duration, polarity, and waveform of the signal sensed. When the ICD detects atrial and ventricular electrical signals that fulfill the preprogrammed criteria for VT or VF, the device decides the appropriate tier of treatment, which can be:

• Antitachycardia pacing, generally for monomorphic tachycardias

• Low-energy defibrillation (±2 J synchronized with the R wave)

• High-energy defibrillation in which the shocks are delivered as a biphasic wave

These treatments can be felt by the patient as sensations varying from discomfort to frank pain.⁶

Approximately 50% of the patients will experience an ICD discharge in the first 2 years of use. To lower the incidence of ventricular and supraventricular arrhythmias, a considerable proportion of these patients receive adjunctive pharmacologic therapy, usually with amiodarone, sotalol, and statins.⁸

After the ICD delivers a shock, three scenarios are possible: successful defibrillation, continuation of the VT or VF, or conversion to another rhythm—usually pulseless electrical activity (PEA) or asystole. After an efficacious shock, the patient’s heart returns to a previous stable rhythm. If the patient continues in VT or VF, the device delivers five more rescue shocks, after which it reanalyzes the waveform.^6,⁷ In case of postdefibrillation (or primary) bradycardia or asystole, the ICD can display antibradycardiac features similar to those of a VVI PM.

Up to 25% of patients can experience an inappropriate shock, defined as a shock delivered for a rhythm different from sustained VT or VF. The most common causes of inappropriate shocks are supraventricular tachycardias (e.g., atrial fibrillation, paroxysmal supraventricular tachycardia, sinus tachycardia), which are read as VT.⁶ Another common cause is misreading of T waves as part of the QRS complex, thereby duplicating the sensed rate. Occasionally, the leads can suffer mechanical damage, such as insulation defects or lead fracture, and such damage can cause electronic noise to be mistakenly detected as VT or VF. These problems are partially solved in modern “noncommitted” apparatuses, which can reanalyze the appropriateness of the rhythm after charging but before shocking.⁷

Currently, no evidence has shown that the electromagnetic fields from daily life artifacts can interfere significantly with defibrillators. However, it is recommended that patients with ICDs avoid placing their cell phones closer than 15 cm to the device and avoid long exposure to metal detectors or antitheft devices.⁷ Some medically related sources of interference such as electrocauteries can cause significant malfunction; therefore, interrogation of the device after exposure is recommended.⁹ Magnetic resonance imaging (MRI) is contraindicated in patients with ICDs given the risk for mechanical torque, thermal injury, and deprogramming.

Indications for Placement of an Implantable Cardioverter-Defibrillator

The major and commonly accepted indications for use of an ICD are summarized in Box 63.1. In recent years the indications and uses have expanded such that the current published guidelines have been outpaced.¹⁰ Multiple trials showing a significant reduction in sudden cardiac death have been followed by publications with similar results regarding primary prevention in selected populations with structural heart disease and a low ejection fraction, as well as in other populations with specific cardiac abnormalities.

Box 63.1 Commonly Accepted Indications for Placement of an Implantable Cardioverter-Defibrillator

Class I

Structural heart disease and episode of sustained VT

Syncope of unknown source with inducible VT or VF in electrophysiologic studies

Patients with LVEF < 35% (NYHA class II or III)

Patients with LVEF < 40% because of previous myocardial infarction with inducible VT or VF in electrophysiologic studies

Approach to Complications Related to the Implantation Procedure

During the early days of ICDs, with their large abdominal cases and pericardial leads, the morbidity and mortality associated with the procedure were considerable. With later use of the transvenous technique, the perioperative mortality rate for ICD placement is less than 0.8%.⁴ Nevertheless, infectious, lead-related, thromboembolic, and mechanical complications can occur.

The rate of pocket or lead infection has been reported to be between 2% and 7%.¹¹ The most common pathogens are cutaneous flora, usually Staphylococcus aureus and Staphylococcus epidermidis. During the first year after ICD implantation, infections related to the device are primarily due to the procedure; after that period they are probably due to secondary seeding. From a clinical perspective, common infectious signs and symptoms in patients with ICDs are notoriously absent and patients may have only vague complaints. Infections of the case and leads have a wide incidence of about 1% to 12%. Diagnosis of a delayed hardware infection requires a high index of suspicion given the absence of a confirmatory ancillary test. Almost without exception, suspected or proven hardware infections require contact with the patient’s primary electrophysiologist, hospital admission, long-term intravenous antibiotic therapy, and potential removal of the device.^7,¹¹

Modern lead systems are extremely reliable but are still prone to fracture, malposition, dislodgment, and damage to the insulation. These defects commonly lead to electrical noise that can precipitate inappropriate shocks. In the evaluation of a patient with suspected lead malfunction, a chest radiograph is required for confirmation of proper positioning and integrity of the leads.^4,^7,¹¹ Contacting the implant team for replacement of the lead is the only alternative possible for hardware failure. Problems related to the battery, pulse generator, and circuitry are extraordinarily rare.

Thromboembolic complications can be seen in as many as 30% of patients with ICDs; they usually involve the cephalic and subclavian veins and do not generally lead to ICD malfunction. Affected patients exhibit unilateral arm swelling, pain, discoloration, and paresthesias, which require evaluation with ultrasonography, venography, or computed tomography. Standard treatment with heparin and warfarin usually results in a good outcome.¹¹

The many mechanical complications related to placement may be manifested early or in delayed fashion. A considerable number of patients experience some degree of tricuspid regurgitation, with approximately 10% of cases being clinically significant. In addition, there is a theoretic risk for fibrosis of the apical lead, which could increase the DFT and make the shocks ineffective. Later manifestations with life-threatening mechanical complications, such as cardiac perforation, cardiac tamponade, hemothorax, pneumothorax, and air embolism, are very rare.¹¹

Approach to Problems Related to Function and Dysfunction

In a patient with complaints related to functioning of the ICD, the EP must consider that the majority of such patients have severe structural heart disease with a poor ejection fraction and that most of them are in end-stage CHF.⁷ They can have a myriad of symptoms; however, these symptoms can be approached systematically (Box 63.2). Evaluation of a patient with an ICD must start by placing the patient in a monitored setting with external defibrillator capacity.^8,¹²

Box 63.2 Clinical Findings in Patients with Problems Related to Functioning of Implantable Cardioverter-Defibrillators

Cardiac arrest

Unstable with ongoing shocks and arrhythmias

Electrical storm

Stable and complaining about a recent isolated shock

Stable and complaining about other cardiorespiratory symptoms

Cardiac Arrest

Causes of death in this population are PEA after VT or VF (29%), defibrillation failure (26%), primary PEA (16%), and refractory VT or VF (13%).¹² When a patient is seen in VT- or VF-related cardiac arrest, the most likely scenario is that VT or VF occurred and the ICD correctly sensed and delivered the shocks but failed to achieve defibrillation. It is critical for the EP to recognize and treat correctable causes of VT and VF. Common causes in this population are ongoing ischemia, electrolyte disturbances (especially hypokalemia and hypomagnesemia), and the arrhythmic effect of drugs.⁴ Many such patients have non–VT/VF-associated cardiac arrest in the context of end-stage CHF; in these cases, disabling the device could be helpful in the resuscitative efforts. Disabling can be accomplished by placing a magnet over the surface of the case pocket. It is very important to remember that after a magnet is placed over the device, it must be assumed that the device is permanently disabled and reprogramming is needed. Standard advanced cardiac life support (ACLS) protocols must be followed both for VT/VF- and non–VT/VF-related causes of cardiac arrest, the only difference being that the external defibrillator paddles and patches should not be placed directly over the ICD case.

Unstable with Ongoing Shocks

Management of an unstable patient must be based on the functioning of the ICD. Basically, this situation may represent two scenarios:

• Inappropriate shocks leading to hemodynamic compromise

• Refractory VT with a pulse and delivery of multiple appropriate shocks

Patients can experience cardiovascular collapse caused by repetitive inappropriate shocks secondary to true or misdiagnosed supraventricular tachycardias. In such cases, treatment must start by decreasing the rate of the tachycardia with the usual pharmacologic treatment (e.g., beta-blockers, calcium channel blockers). If these measures fail, the next step is to disable the ICD. A considerable proportion of patients arrive at the ED in unstable bradycardia; in this setting the VVI function of the device will be helpful, but prompt recognition of the cause (commonly end-stage CHF) and standard ACLS treatment are mandatory.⁷

Those with sustained VT with a pulse and appropriately shocked by the device represent an in extremis population. The most critical intervention is the delivery of higher-energy shocks with an external defibrillator (regardless of whether the ICD is disabled) and rapid correction of the cause of the VT.^7,⁸

Both populations require mandatory electrophysiologic consultation and interrogation of the device.

Electrical Storm

Electrical storm is commonly defined as more than two therapies (antitachycardia pacing or shocks) delivered in a 24-hour period.¹³ It is believed to affect about 10% of patients with ICDs⁷ and is a common complaint in these patients.⁸ Classically, such patients are rather stable but complain of several shocks delivered in the preceding hours. The significance of electrical storm is that it is usually a herald of life-threatening acute pathology, commonly acute cardiac ischemia, hyperkalemia, and decompensated CHF, thus placing these patients at immediate high risk for death.

Management starts with standard stabilization and cardiac monitoring followed by correction of any obvious abnormalities, evaluation for mechanical failure, device interrogation, electrophysiologic consultation, and aggressive management of CHF and acute ischemia.

Stable with a Recent Isolated Shock

The most important first step is to determine whether the shocks were appropriate.⁴ The patient should be placed in a monitored setting where the heart rhythm can be recorded, followed by a chest radiograph to evaluate for possible hardware failure (e.g., lead fracture) and basic laboratory tests to look for ischemia or electrolyte disturbances. Any signs or symptoms around the moment of the shock (e.g., chest pain, shortness of breath, chest trauma) should be noted. It is also important to inquire about new drugs or changes in dosage (especially for amiodarone). Arrhythmias discovered during monitoring, as well as metabolic causes of VT or VF, must be treated in the usual fashion.

Occasionally, patients complain of hearing a beep from the device. Some models can emit a beep in the event of battery discharge or another cause of malfunction. Stable patients with an isolated shock require interrogation of the ICD so that the underlying rhythms can be evaluated and the ICD can be reprogrammed.⁴ The decision for hospital admission is usually made jointly by the EP and the electrophysiologist based on the appropriateness of the shock, the availability of follow-up, and the overall clinical status of the patient. Patients who have experienced isolated appropriate shocks but have no change in cardiopulmonary status, no evidence of ischemia, and no electrolyte abnormalities can be discharged from the ED for follow-up with an electrophysiologist.¹⁴