Irritable Contemporary Risk of Implantable Cardioverter Defibrillators

Judith A. Mackall


AccessMedicine from McGraw-Hill 

The risk associated with implantation of the implantable cardioverter-defibrillator (ICD) is often focused on the immediate complications of the procedure. The long-term risk of having an ICD, including lead- or device-related malfunction, infection, and inappropriate shock, should be considered as well. While these risks are felt to be low when compared to the benefits of a life-saving therapy, they are not necessarily the same for every patient. Factors which predispose patients to increased risk include patient characteristics, type of procedure, and operator characteristics.

Procedural Complications

For new ICD implant procedures, peri-operative complications reported in the National ICD Registry (NCDR) from 2006-2008 were 3.36%.[1] The most common complications included lead dislodgment (1.07%), hematoma (0.97%), and hemo/pneumothorax (0.57%). Catastrophic complications of death in the lab (0.02%), cardiac perforation (0.08%), and transient ischemic attack (TIA)/stroke (0.06%) were also reported. Data from the NCDR registry represented 90% of all ICD implants in the country. Although useful as a performance measure, the registry did not report long-term complications or clinical outcome.

The Ontario ICD Database was a prospective, multi-center registry of 3340 new ICD implants between February 2007-May 2009.[2] Major complications occurred in 3.8% of primary prevention patients and 4.8% of secondary prevention patients. Only a small proportion of complications (0.8%) occurred in the hospital. Complications were classed as either mechanical (lead dislodgment, replacement, or extraction, device revision, cardiac perforation, pneumo/hemothorax, infection, erosion, hematoma, coronary venous dissection, or diaphragmatic stimulation) or clinical (pulmonary edema, hypotension requiring resuscitation, stroke or death). Patient characteristics which predicted increased risk included female gender, secondary prevention indication, nonischemic cardiomyopathy, dilated left ventricle (LV), and anti-arrhythmic drug use other than amiodarone ( Table 1 ). The risk of a complication was also associated with the type of device implant with cardiac resynchronization therapy device (CRT-D) implants having the highest risk. Importantly the presence of any complication was associated with an increased mortality over the following 6 months (Figure 1).

Figure 1.

Mortality Rates for Defibrillator Recipients With and Without Complications

Mortality rates at 45, 90, and 180 days for defibrillator recipients with clinical complications, any complications (implant-related or clinical), and no complications. The p values for comparison with no complication group are shown. Any = any complication (major or minor); compl = complication.

From Lee DS. J Am Coll Cardiol 2010;55:774-82.

The observation that female gender is associated with increased procedural risk has been noted in other cardiac procedures such as angioplasty and coronary artery bypass surgery. Data from the NCDR registry noted that women had almost double the risk of a major complication. Women also were more likely to have congestive heart failure, a worse functional class, non-ischemic cardiomyopathy (44% vs. 27%), and more frequently undergo implantation of CRT-D (39% vs. 34%).[3] The reasons for the gender disparity have been attributed to the older age of women, the presence of co-morbidities and the fact that smaller body habitus presents more challenging anatomy at implant.

The presence of renal failure has also been shown to increase in-hospital complication rates with major complications occurring at a rate of 2.2% compared to 1.3% in patients without renal disease.[4] It has also been reported that patients with renal insufficiency are at increased risk of infection.

Most of the significant procedural complications reported in the Ontario Database occurred following defibrillation threshold (DFT) testing.[2] Birnie et al reported serious complications resulting from defibrillation testing, including death (0.016%), prolonged resuscitation (0.14%), and stroke (0.026%).[5] All of the strokes occurred in patients with atrial fibrillation at the time of DFT testing.[5] When compared to the original ICD systems, the current ICD systems have a high likelihood of low DFT based largely on technical improvements in increased energy delivery as well as improved ICD shock delivery: biphasic shocks, dual coil leads, active can configuration, and the ability to tailor waveform morphology.

Since DFT testing can be associated with significant complications and DFT does not predict first shock efficacy, serious thought should be given as to which patients should undergo DFT testing. Swerdlow et al present a comprehensive evaluation of DFT testing.[6] The absolute and relative contraindications to testing are presented in Table 2 . DFT testing should be performed if R-wave sensing is absent (no intrinsic R) or less than 5 mV to ensure detection of ventricular fibrillation, in patients who will be using another electrical device to exclude device interaction, and in patients with risk for high DFT. Deferring testing until after LV reverse remodeling occurs in CRT-D patients or utilizing the upper limit of vulnerability may also be prudent.

Data from the Replace Registry which enrolled 1000 patients undergoing device replacement and 750 patients undergoing lead replacement/revision suggest that these procedures have a higher complication rate than new implants (4.4% vs. 3.3%). The addition or revision of a lead in the RA or RV had an 11.1% complication rate and the addition or revision of an LV lead had an 18.7% complication rate.[7]

Complications: Beyond the Implant

Inappropriate shocks occur because of supraventricular arrhythmias with rapid ventricular rates, T wave oversensing, lead malfunction, or electromagnetic interference. Inappropriate shocks occur in 12-36% of patients, with children having a higher likelihood (up to 47% in some series). Although rare, it has been associated with death. Recent analysis of the SCD-Heft trial data (a primary prevention trial enrolling 2521 patients with Class II-III heart failure) have shown that ICD shocks, both appropriate and inappropriate, were an important predictor of mortality. Most patients died from progressive heart failure.[7]

Long-term complications have been more difficult to quantify. Alter et al reported on their single-center experience following 440 ICD patients for 46 ± 37 months.[8] The complications were due to the implantation, ICD generator, lead(s), inappropriate shocks, and appropriate shocks in 10%, 6 %, 12%, 10%, and 24%, respectively. Most of these patients had a secondary prevention indication for ICD.

Operator Characteristics

For many procedures, there is an association between the physician experience and the patient outcome. Al-Khatib et al analyzed data from 1999-2001 in which 1,672 physicians implanted 9,854 ICDs.[9] Physicians who implanted less than 11 ICDs/year had higher complications and more infections. In addition, review of physician board certification demonstrates that cardiac electrophysiologists had a lower complication rate (3.5%) compared to cardiologists (4.0%) and thoracic surgeons (5.8%).[10] Device type also impacted the complication rate with CRT-D implants having the highest complication rates.

Mitigating the Risk

It is important to realize that the risk associated with ICD implant is patient, procedure and operator specific. Mitigation of the risk involves attention to both the patient and the type of procedure. The quality data available from the NCDR registry can be used to identify areas where complications are excessive.