Pacemakers and Implantable Cardioverter-Defibrillator Generators Recalls

Abstract & Commentary

Synopsis: This study is a call for increased efforts by manufacturers and the FDA to improve device reliability and safety.

Source: Maisel WH, et al. JAMA. 2001;286:793-799.

Maisel and colleagues reviewed data on recalls of permanent pacemakers and implantable cardioverter defibrillators (ICDs) in the 10-year period from January 1990 to December 2000. All weekly Food and Drug Administration (FDA) enforcement reports that contained recalls or safety alerts (advisories) were surveyed. Advisories dealing with lead systems only were not included. The advisories were grouped by the FDA into 3 classes based on the possibility of patient injury that might be caused by the defect in question with class I recalls, those likely to produce serious injury being the highest class. Maisel et al then took the recommendations for action contained in the advisory and projected the costs of the corrective action for the number of units affected.

Between 1990 and 2000, there was a 22% increase in the number of patients living with pacemakers and an 11-fold increase in the number of patients with ICDs. During this period, there were 52 advisories that affected 108,500 pacemakers and 114,645 ICDs. Three quarters of the advisors were class I or II. Of these, 67% were issued due to hardware malfunctions, 19% due to firmware (integral device computer programming), and the remainder due to miscellaneous causes. There were 7.7 ± 1.1 advisories per 100 patient years of device usage. ICDs were more frequently involved (16.4 ± 1.61 advisories per year vs 6.7 ± 0.81 per year). Some clinical intervention was recommended by the device manufacturer in 79% of the advisories. For advisories involving pacemakers, replacement was often recommended only if the patient was considered to be pacemaker dependent. In only 6% of the advisories was replacement recommended for all patients. In contrast, replacement was recommended for 72% of the ICD advisories. If replacement was not performed, more frequent device checks were usually required. Maisel et al estimate that device advisories cost $870.2 million in increased medical expenses. No injuries related to device malfunctions covered by these advisories are described in this paper, but isolated case reports have appeared in the literature. However, device failure in a completely pacemaker-dependent patient or in an ICD patient with rapid ventricular tachycardia might result in death, and the cause of death might be attributed to the patient’s underlying condition so Maisel et al recognize that they may be underestimating the severity of the problem. Maisel et al call for increased efforts by manufacturers and the FDA to improve device reliability and safety.

Comment by John P. DiMarco, MD, PhD

Pacemakers and ICDs have become routine and important tools for cardiologists. The study by Maisel et al should remind us that these devices, although extraordinarily effective in most patients who receive them, are complex and require careful follow-up after implantation.

Most physicians are aware of the early complications associated with pacemaker or ICD implantation. Bleeding, infection, vascular damage, pneumothorax, and lead dislodgment are the most commonly described early complications. These complications can be minimized by scrupulous attention to implant technique, by use of smaller devices that are easier to implant and by improved lead designs. Pacemaker and ICD leads may also fail during chronic usage. These leads are subject to mechanical and environmental stress, and insulation failures or lead fractures are common. Some leads have a higher failure rate than others and one system, an atrial J-shaped lead manufactured by Telectronics, caused patient deaths and severe injuries when an imbedded wire fractured with the potential for lacerating the right atrium. Late pocket erosions are less frequently seen as devices have become smaller but are still encountered. Since the function of the pacemaker or ICD is tested at the time of implant, we often consider the device itself the most reliable part of the system.

The report by Maisel et al should make us reconsider this complacency. Although it is true that advisories were triggered by defects in only a small proportion of the devices covered, the advisory affected every patient with that device model. Maisel et al’s cost estimate of $870 million is probably too low since they assumed replacement operations to be free of complications. Progressive scarring in a device pocket from multiple procedures may promote infection, erosion, and lead damage.

The solution to this problem is elusive. Increased clinical testing before release of an individual model is unlikely to work. Most device recalls are triggered by a small number of observations and only a small proportion of recalled devices are defective. Only large-scale market release gives enough experience to detect these defects. What manufacturers and the FDA should strive for are better methods to detect potential for failure before a device is implanted. Innovative approaches for nondestructive testing of all device functions should be developed. Only with better methods for preimplant testing before the device leaves the factory, can we decrease the number of advisories without delaying implementation of new technology.