What prompted you to begin studying complication rates in the ICD patient population? Since replacing a defibrillator is generally considered a "minor" procedure, had you noticed an increase in complication rates among your patients? We are a regional centre for lead extraction for patients with infected devices. In the Fall of 2005, we saw three patients who developed significant infection after advisory device replacement over the course of a month. All three of these patients came to London Health Sciences Centre for lead extraction, with death as a complication of lead extraction in one. This prompted us to assess the downside of advisory device replacement and initiate a study to look at the scope of the problem. Describe the study results. How many and what types of complications occurred? How many follow-up surgeries were needed? In addition, how many patients ultimately had their devices replaced? Of 2,915 patients across Canada, a total of 533 patients had their device replaced after discussion with their physician and consideration of their medical status. Of these 533 patients, complications occurred in 43. Ten of these patients (2%) ended up having lead extraction because of infection that did not resolve with medical therapy. Did the higher complication rate from replacement surgery surprise you? Why or why not? We were surprised by the high complication rate since the general quoted rate from minor surgical procedures is 1%. The reasons for the higher than expected incidence of complications are complex. Some potential explanations are a more elderly population with advanced heart disease, multiple comorbidities predisposing them to infection and bleeding complications. The device replacement environment is also an environment where infection is more likely to occur, because of previous scar tissue in the generator pocket. Your report was published near the same time that the Heart Rhythm Society's (HRS) draft recommendations were released. Was there any connection between the two? The publication of our results was in JAMA on April 26th. This coincided with the release of the Heart Rhythm Society's draft recommendations. The decision of the timing to publish was based on the priority of the journal. The JAMA edition included several articles on the topic of advisory devices, as well as a thoughtful editorial by Dr. Bruce Wilkoff. It is conceivable that the journal saw the timing of the draft recommendations, and placed a priority on publication to permit access to this knowledge during the discussion of the draft recommendations. How do the new HRS draft recommendations on performance policies for pacemakers and ICDs address device recalls and device replacements? Do you think they will help improve patient safety? What changes to the guidelines would you recommend? The draft recommendations provide a framework upon which to handle advisories. The largest impact that they will have is to focus efforts in the area of regulatory agencies and device manufacturers in their reporting and monitoring practices. The decision to replace devices will remain in the hands of the physician in discussion with the patient, taking into account patient characteristics and advisory information. This process will undoubtedly enhance patient safety and provide more transparent access to information for all parties involved to monitor and manage device advisories. I read that the risk of failure among recalled devices is estimated at just under 2.6 percent, meaning that not every failure is deadly. What do you recommend to your patients in this situation? Should they be given a choice of whether or not they want their device replaced? The decision to replace the device is based on three factors. The first is the characteristics of the advisory, taking into account the risk of failure and the ability to monitor and program around that failure. The second involves the likelihood that the patient will use their device, and if that use is life-saving. This is influenced by the indication for the device, and previous beneficial therapies. The final deciding factor is the complications of the procedure, which undoubtedly is influenced by patient characteristics as well. Patients on anticoagulants, diabetic patients, or patients with multiple surgical procedures are at higher risk for infection or hematoma. These three factors need to be taken into account when counseling a patient regarding advisory device management. Will this study be continued at a later time? I realize this one in particular went from October 2004 to October 2005. This study was an accelerated look at a period of time when there was a high frequency of advisory device replacements. In principle, ongoing studies such as this should continue to monitor the impact of our decisions on patient outcomes. Is there anything else you'd like to add? Device advisories have been pushed to the forefront by recent controversy, in addition to the unusual media attention focused on some high-profile cases. The principle of device reliability is an old one, and remains an aspect of device practice. In general, careful and thoughtful management of patients and their devices will always take into account the risk of different treatment avenues. Hopefully, this decision rests with the physician and the patient, with transparent access to information and informed consent. For more information, please refer to Complications associated with implantable cardioverter-defibrillator replacement in response to device advisories in JAMA 2006 Apr 26;295(16):1907 - 1911.