Complications of Device Generator Replacements

Abstract & Commentary

By John P. DiMarco, MD, PhD, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville. Dr. DiMarco receives grant/research support from Medtronic, is a consultant for Medtronic, Novartis, and St. Jude, and is a speaker for Boston Scientific.

Source: Poole JE, et al. Complication rates associated with pacemaker or implantable cardioverter-defibrillator generator replacements and upgrade procedures: Results from the REPLACE registry. Circulation. 2010;122:1553-1561.

Implantable cardiac rhythm devices are being used in an increasing number of patients and, in addition, patients who receive them are living longer. Since battery longevity in both pacemakers and defibrillators is limited, the risks of generator replacements must be appreciated by referring and implanting cardiologists. However, relatively few data are available that detail the risks and complications associated with generator replacements. For this reason, the REPLACE investigators performed a prospective, multicenter study designed to collect complication data on patients after replacement of either a pacemaker or ICD generator. They also were particularly interested in patients who were undergoing upgrades which required lead additions.

Data were collected from 72 institutions with an essentially equal mix of academic and private hospitals. Patients were separated into two cohorts. Cohort 1 consisted of patients undergoing generator placement in which no lead revision or addition was planned. Patients in cohort 2 included those in whom the intent at the time of the procedure was to add one or more leads. Patients with an expected heart transplant within six months or a planned lead extraction for any cause were excluded. Patients with any commercially available generator or lead could be included. The decision to perform the generator replacement or upgrade was at the investigator's discretion. After the generator replacement, follow-up included a wound examination performed according to the institution's routine practice followed by 3- and 6-month clinic visits. Definitions of major and minor complications were pre-specified and comprehensive. Major complications were those that placed the patient at significant risk, or required an intervention or subsequent procedure or a hospitalization for management of issues related to the device. Other complications were classified as minor. Reported complications were adjudicated by a clinical events committee blinded to the operator and the institution. Deaths occurring within 30 days of the procedure were reported. Death was considered to be procedure-related if there was an acute mechanical complication during the procedure or, if after the procedure, the death was unexplained and unanticipated.

The REPLACE Registry enrolled 1,715 patients. Six patients were later censored because they met exclusion criteria. As a result, there were 1,031 cohort-1 patients and 713 cohort-2 patients. Patients in cohort 2 were less likely to be female, had a lower ejection fraction, and a higher New York Heart Association class and more prior cardiac surgeries. Both cohorts had essentially equal numbers of ICD and pacemaker patients. About 90% of the device locations were prepectoral, with most of the remainder being subpectoral implants. The reasons for lead addition in cohort 2 were an upgrade to a CRT device in 407 patients, upgrade from a single chamber to a dual chamber pacemaker, or ICD in 114 patients and an upgrade from a CRT pacemaker to a CRT-D in 13 patients. Replacement or evaluation of a suspected malfunctioning lead was the reason for lead replacement in 179 patients. Among the patients who underwent generator replacement only (cohort 1) there were no deaths during the procedure, but two patients had periprocedural complications consisting of hemodynamic instability requiring intervention. The most common post-procedure complications were lead malfunction requiring re-operation in 10 (1%) patients and re-exploration for hematomas in seven (0.7%) patients. In addition, eight patients (0.8%) experienced a major infection, and five of these patients required extraction of their generator or leads. A minor infection occurred in six (0.6%) patients who were treated with outpatient oral antibiotics. The overall major complication rate was 4.0%. The minor complication rate was 7.4%.

Complications were considerably higher in cohort 2. Intraprocedural complications occurred in 13 patients. These included cardiac perforation in five (0.7%) patients, pneumothorax or hemothorax in six (0.8%) patients, and cardiac arrest in two patients. In addition, 56 patients (7.9%) required re-operation because of lead dislodgement or malfunction. Prolonged hospitalizations due to procedural related exacerbation of heart failure or renal failure occurred in 18 patients (2.5%), and hematomas requiring evacuation developed in 11 patients (1.5%). There were 8 procedural-related patient deaths (1.1%) within 30 days of the procedure in cohort 2. Four of these occurred in patients in whom an unsuccessful attempt to place a transvenous left ventricular lead prompted a surgical thoracotomy for lead placement. In addition, there were four unexplained deaths that occurred after hospital discharge but within 30 days. The overall major complication rate in cohort 2 was 15.3%. The minor complication rate was 7.6%. Six patients (0.8%) in cohort 2 experienced a major infection, with five requiring extraction of the generators and leads. Minor cellulitis treated with antibiotics was noted in two additional patients. When complication rates were examined by the type of lead procedure, patients who underwent an upgrade or revision to a CRT had a major complication rate of 18.7%, whereas patients who had an upgrade of a single chamber pacemaker ICD to a dual chamber ICD had a complication rate of 11.1%. In contrast, the cohort-1 patients who did not receive a new lead had a complication rate of 4.4%.

The authors concluded that cardiac-rhythm device generator changes are not benign, and the risk increases dramatically when additional leads are inserted at the time of generator change. They caution that physicians considering device upgrades should be aware of these risks, appreciate them, and make sure that the anticipated benefit from lead additions outweighs the possibility of complications.


For many years, little attention was paid to ICD- and pacemaker-generator changes. It was only several years ago, after some ICD generator recalls led to a large number of elective generator replacements, that the risks associated with generator changes really were appreciated. In most reports, the complication rate at least equaled that seen with new implants. The data from the REPLACE registry confirm these observations and add data about the risks of adding new leads. These data are particularly important since more liberal indications for ICD and CRT therapy now permit device upgrades in many patients with previously implanted devices. The risks reported in REPLACE are quite sobering. Clearly, we need better ways to predict risks and benefits for individual patients. We should neither hesitate to upgrade patients to a CRT device if they have significant heart failure symptoms nor should we without thinking recommend upgrades for all class-I patients with a low ejection fraction.