Pacemaker and Internal Cardioverter Defibrillator Lead Extraction: A Safe and Effective Surgical Approach

Transcription

1 ORIGINAL ARTICLES: SURGERY: The Annals of Thoracic Surgery CME Program is located online at To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal. Pacemaker and Internal Cardioverter Defibrillator Lead Extraction: A Safe and Effective Surgical Approach John M. Kratz, MD, and John M. Toole, MD Division of Cardiothoracic Surgery, Department of Surgery, Medical University of South Carolina, Charleston, South Carolina Background. Need for pacemaker or internal cardioverter defibrillator lead removal is increasing. Removal can be dangerous, difficult, or unsuccessful. Methods. We retrospectively reviewed our results and the techniques we used in 365 patients from 1992 through 2009 for successful complete removal of leads and complications. Various techniques of extraction were analyzed for effectiveness and complications. The eras before (1992 to 1999) and after the availability of laser sheath extraction (2000 to 2009) are compared. Results. Of 365 patients who underwent transvenous lead extraction, of which 235 were infected, and 130 had lead removal for noninfectious indication. Staphylococcus aureus was the infecting organism in 40%, and coagulase-negative Staphylococcus occurred in 33%. One-half of the organisms were methicillin resistant. Preimplant risk factors for infection included more than one device implant procedure in 105 (47%), preimplant Coumadin therapy (Bristol-Myers Squibb, Princeton, NJ) in 74 (31%), and hemodialysis in 9 (4%). Laser extraction became available in The era with the availability of laser extraction was associated with a better complete extraction rate (93% vs 89.55%) a lower bleeding rate (1.9% vs 3.1%), and complete extraction without the additional use of femoral workstation extraction tools. Mortality was 1.1%. No death was due to device removal. All deaths were the result of severe preoperative and continuing postextraction sepsis. Conclusions. A lead extraction protocol that included procedures done in an operating room environment allowing rapid, open intervention for bleeding, a varied choice of extraction tools, arterial line monitoring, transesophageal echocardiography, general anesthesia, and an experienced team yielded complete extraction in more than 90% of patients, with a low complication rate and no procedurally related deaths. (Ann Thorac Surg 2010;90:1411 7) 2010 by The Society of Thoracic Surgeons Many factors have resulted in an increasing number of patients presenting with infections or other complications of pacemaker or internal cardioverter defibrillator systems (ICD). Owing to expanding indications, more patients are undergoing device implant. The aging population presents a higher percentage of patients requiring device implant. The population of patients surviving to require reimplantation for battery depletion is growing, as is the need to implant devices in anticoagulated or otherwise increasingly complicated patients. Pacemaker or ICD lead extraction can be fatal or complicated. Even the lay press has reported the possible hazards of lead extraction [1]. Although patients requiring explantation of pacing Accepted for publication May 10, Presented at the Poster Session of the Forty-sixth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 25 27, Address correspondence to Dr Kratz, Division of Cardiothoracic Surgery, Medical University of South Carolina, 25 Courtenay St, Charleston, SC 29424; kratzj@musc.edu. systems have become more numerous and complicated, techniques for complete and safe removal of pacing leads have also become more diverse and available. We review the experience of a single center with principally 1 surgeon s experience in the management of complicated pacemakers or ICDs requiring revision or removal from 1979 through Material and Methods The Institutional review board of the Medical University of South Carolina approved this study and waived requirement for patient consent for review of records for this study. Between 1992 and 2009, 365 patients were referred for pacemaker or ICD lead extraction (Table 1). We retrospectively reviewed these patient s records for techniques used, complete lead removal, complications, possible causes of infection, causative organism, and methods used to prevent and minimize complications by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 1412 KRATZ AND TOOLE Ann Thorac Surg SAFE SURGICAL PACEMAKER LEAD EXTRACTION 2010;90: Table 1. Demographics of Patients With Lead Extraction Variable All Patients Infectious Noninfectious Total No Sex Male, % Female, % Age, median 62 (3 103) 66 (3 95) 55 (5 103) (range), y ICD, % 25 (91/365) a 76 (69/91) b 24 (22/91) c Pacemaker, % 75 (274/365) a 61 (166/274) d 39 (108/274) e Coumadin, % 22 (80/365) a 31 (74/235) f 5 (6/130) g Dialysis, % 3 (10/365) a 4 (9/235) f 0.8 (1/130) g Multiple implants, % 36 (131/365) a 47 (105/235) f 20 (26/130) g a Percentage of total patients. b Percentage of ICD patients infected. c Percentage of ICD patients not infected. d Percentage of pacemaker patients infected. e Percentage of pacemaker patients not infected. f Percentage of infected patients. g Percentage of not infected patients. ICD internal cardioverter defibrillator. Statistical Analysis Demographics and results are presented as counts. Percentages are reported when appropriate. Outcomes were compared with the Fisher exact one-sided test using NCSS 2001 software (Number Cruncher Statistical Systems, Kaysville, UT). Surgical Procedures All patients with infection received intravenous vancomycin until an organism and its antibiotic sensitivity could be identified. Cultures were obtained from the pocket before the operation whenever possible. Cultures were preformed on tissue from the pocket and lead tips upon extraction. Antibiotics were given for 1 week after extraction, except in instances of possible or certain endocarditis, in which case 6 weeks of antibiotics were given. All procedures were performed in the operating room under general anesthesia with arterial catheter monitoring. Patients were prepared and draped to allow immediate access for cervical, subclavian, femoral, and median sternotomy incisions. In recent years, monitoring with transesophageal echocardiography has provided earlier and better detection of pericardial effusion and possible previously undetected endocarditis. Pacemaker-dependent patients were managed in the following order. Sick sinus syndrome patients or heart block patients with a stable escape rhythm and a satisfactory heart rate were managed with backup cutaneous pacing and observation in the intensive care unit for a few days to allow treatment with antibiotics before placing a new pacing system. Patients with a pocket infection only or noninfected patients were frequently managed by placement of a new pacing system transvenously from the opposite shoulder or using epicardial leads by way of a subxiphoid incision before lead extraction. In patients with more aggressive infection and an unsatisfactory heart rate due to heart block, a permanent transvenous pacing lead with an active fixation screw was placed as a temporary lead through a separate cutaneous venous entry. This lead was securely sutured to the skin and attached to an exteriorized permanent pacemaker. Antibiotics are given for 3 to 4 days to clear any remaining infection before placement of a new pacing system. We prefer, when possible, to treat the infectious process for a few days with appropriate intravenous antibiotics before a new system is implanted. A subxiphoid approach for epicardial lead removal was used in patients where active endocarditis was present or the risk of recurring exposure to blood-borne organisms was present, such as in dialysis patients. Our goal in patients with an infectious process was to remove all elements of the pacing system, including currently used or old retained leads. In patients with soft indications for removal, such as lead malfunction or system upgrade, the planned extraction was used cautiously, recognizing the ever-present risk of a major bleeding complication. Various lead removal techniques were used. The distribution of techniques used is reported in Table 2. The least invasive number of techniques that would result in a complete removal of all lead material was used; that is, if traction alone or traction with a stylet could affect complete removal, sheaths were not used. Excessive traction without a locking stylet in place may cause damage to the stylet channel, however; and therefore, a low threshold for use of a locking stylet was indicated if minimal traction would not easily dislodge the lead. Traction-only has been used since 1989 and continues to be successful in some cases. It was initially the only transvenous method available. This method was occasionally enhanced with the use of a cord tied to the lead and then run over a pulley to a hanging 1-pound weight to provide constant controlled traction to a recalcitrant lead. This method was rarely currently used except in the case of a recently implanted lead. Traction with locking stylets became available to us in We have used locking stylets consistently as part of our lead extraction technique, except in the case of a recently implanted lead that may be easily removed with traction alone. We initially used the Wilkoff stylet (Cook Table 2. Techniques Used Technique No. Traction 32 Traction, stylet 47 Traction, stylet, sheath 50 Traction, stylet, sheath, femoral 12 Traction, stylet, sheath, femoral, jugular forceps 1 Traction, stylet, laser 202 Traction, stylet, laser, femoral 19 Traction, stylet, laser, femoral, jugular 1 Femoral only 1 Open 8

3 Ann Thorac Surg KRATZ AND TOOLE 2010;90: SAFE SURGICAL PACEMAKER LEAD EXTRACTION Catheter Company, Bloomington, IN), which grasped the lead using a hook wire at the distal end of the locking stylet. We have consistently used the lead locking device (LLD; Spectranetics, Colorado Springs, CO) since it became available in Using an expanding coil, this device provides lead lumen contact throughout its entire length and a more secure grasp of the lead. Dissecting sheaths (Cook Catheter Company) became available to us simultaneously with the availability of the locking stylets. We have used semirigid plastic sheaths as well as rigid steel sheaths to dissect scar tissue adhering to transvenous leads. We have used Laser Dissecting Sheaths (Spectranetics) since January These devices consist of a plastic sheath with embedded fiberoptic fibers that apply an excimer laser light to the scar tissues surrounding and entrapping a chronically implanted lead. Femoral workstations in combination with a needle eye snare grasping system (Cook Catheter Co) have been used since In selected patients in whom extraction could not be accomplished by removal of the lead by way of its subclavian entry site, the femoral workstation has been a valuable asset. Plastics sheaths with an electrocautery tip (Cook Catheter Co) have been applied to dissect scar tissue about the lead. On a few trials we were disappointed with our limited success with this device and have not used it since that time. Transvenous grasping forceps were used successfully in a few procedures when no other technique would allow complete lead removal [2]. Open chest incision removal of leads was rarely used except in the instance where a cardiac operation was being preformed for an unrelated condition. Only two of the open procedures were required because transvenous extraction could not be accomplished. In 1 patient, a small subxiphoid incision with retrograde laser extraction was used [3]. Results Patients referred for lead extraction varied in age from 3 to 103 years. Of the 365 patients who underwent extraction, 274 had pacemakers and 91 had ICDs. At the time of extraction, 103 patients were pacemaker dependent. Femoral Workstation Use The femoral workstation was only used when extraction could not be accomplished by other extraction techniques through the prior implantation site of the transvenous lead. The femoral workstation was available throughout our experience. Therefore, the need for the femoral workstation is an indicator of unsuccessful extraction by other techniques. During the prelaser period, use of the femoral workstation was required in 13 of 95 patients (14%), whereas in the postlaser period the workstation was only required in 19 of 270 patients (7%), a statistically significant difference (p 0.043; Fisher exact test). Table 3. Indications for Extraction Indication 1413 No. Infection 235 Lead malfunction 109 Pain 12 Teletronics J lead 1 Device no longer needed 4 Superior vena cava obstruction 2 Upgrade to internal cardioverter defibrillator 1 Tricuspid insufficiency 1 Indication for Extraction Indications for lead extraction are summarized in Table 3. Most extractions (64%) were performed for infection. Many of the remaining patients underwent lead extraction for what might be considered soft indications, meaning not all practitioners would agree extraction was absolutely necessary. These patients were informed preoperatively of the risks, benefits, and controversial nature of their indication for extraction. Owing to the less compelling nature of the indication in these patients, complete extraction was pursued less aggressively. Only one serious complication occurred as a result of these extractions for noninfectious indications. A previously place ventricular lead failed to pace as the result of cardiac perforation, and bleeding occurred on removal. This was quickly repaired without subsequent sequela. No deaths occurred in the noninfectious patients. Organisms Causing Infection The organisms causing infection are reported in Table 4. Of the 235 with infection, 172 (73%) were caused by Staphylococcus aureus or S epidermidis. Of these organisms, 78 of 172 (45%) were methicillin resistant. Predisposing Factors for Infection Certain patient-associated conditions occurred in the infected patients far more frequently than in our noninfected group or the otherwise healthy patients who required pacing or an ICD (Table 1). Hemodialysis patients comprised 4% (9 of 235) of infected patients and 0.8% (1 of 130) of the noninfected group. Preoperative Coumadin (Bristol-Myers Squibb, Princeton, NJ) use was present in 31% (74 of 235) of the infected group and in 5% (6 of 130) of the noninfected group. A history of 2 or more implants was present in 47% (105 of 235) of infected patients and in 20% (26 of 130) of noninfected patients. Results by Techniques Available During a Specific Era Only a few procedures were done in the era when only traction was available; therefore, meaningful conclusions regarding results are not possible. Open removal (8 patients) by way of sternotomy or subxiphoid approach was available throughout the entire time. Extraction using biopsy forceps by way of the right jugular vein (2 patients) was also available during the entire period. Because increasingly aggressive tools for extraction were

4 1414 KRATZ AND TOOLE Ann Thorac Surg SAFE SURGICAL PACEMAKER LEAD EXTRACTION 2010;90: Table 4. Infectious Organisms Organism Frequency Penicillin Sensitive Methicillin Sensitive Methicillin Resistant No. (%) No. (%) No. (%) No. (%) Staphylococcus aureus 95 (40) 0 45 (47) 50 (53) S epidermidis 77 (33) 10 (13) 39 (51) 28 (36) Enterococcus Corynebacterium Propionibacter Pseudomonas Escherichia coli Beta-hemolytic Streptococcus S pneumonia S viridans Torulopsis Citrobacter Lactobacillus Mycobacterium fortuitum Unknown used when lesser techniques were unsuccessful, comparison of each individual technique would not be meaningful. Laser sheaths, however, represent a significant change in available tools for lead extraction. It is therefore useful to examine results for eras before and after the availability of the laser sheath. Traction with locking stylets, dissecting sheaths, and femoral extraction tools were initially used in 1992 and continue to be used. Results before availability of the laser sheath (1992 to 1999) and after laser sheath availability (2000 to 2009) are reported in Table 5. Results are subdivided for infectious patients, noninfectious patients, and all patients during this period. Of the 4 infectious patients who had lead retention, 3 were tips of the pacing leads only and were left in place without further problem. The fourth was a 2-year-old child with retention of 2 inches of distal lead. He received 6 weeks of antibiotics and has also done well without further intervention. Bleeding complications occurred in 3 patients and were treated by repair by way of median sternotomy, without further complication. One patient with preoperative systemic sepsis, who had complete removal of all leads, died of persistent sepsis. Laser sheaths, in addition to locking stylets and femoral extraction, became available in 2000 and have been used through the current time. Results are also reported in Table 5. Of the 8 infected patients with retained leads, 6 had tip-only retentions and were being treated for pocket infection only. They have done well, without intervention or recurrence of infection. Two patients were believed to have active endocarditis, and the retained leads were 3 to 4 inches long. These were removed using an open technique. Removal was incomplete in 11 of the 108 patients (10%) operated on for noninfectious indications, and because of the soft indication for removal, no further intervention was performed. Bleeding occurred in 5 of the 270 patients (1.9%). All bleeding occurred in infected patients and all instances were corrected by way of median sternotomy, without further complication. One patient (0.3%) returned with a Table 5. Results of Treatment Total Complete Extraction Recurrent Infection Bleeding Death Technique Interval Time Indication No. No. (%) No. (%) No. (%) No. (%) Before laser extraction Infectious (93) 0 2 (3) 1 (2.0) a Noninfectious (83) 1 (3) 0 All (89) 0 3 (3) 1 (1) Laser extraction available Infectious (95) 1 (0.5) b 5 (3) 3 (2) c Noninfectious (88) 0 0 All (93) 1 (0.3) 5 (2) 3 (1) All extractions Infectious (95) 1 (0.4) 7 (3) 4 (2) Noninfectious (89) 1 (0.7) 0 All (92) 1 (0.3) 8 (2) 4 (1) a Preoperative sepsis was the cause of death. b Recurring pocket infection. c Preoperative systemic sepsis in 2 patients and preoperative stroke in 1 were the ccauses of death.

5 Ann Thorac Surg KRATZ AND TOOLE 2010;90: SAFE SURGICAL PACEMAKER LEAD EXTRACTION recurring pocket infection. Death occurred in 3 of the 270 patients (1.1%) as the result of persistent preoperative systemic sepsis [2] and stoke due to endocarditis, which occurred preoperatively in 1 patient. All had complete removal of all leads. ICD vs Pacemaker Removal Of the 365 patients undergoing lead extraction, 91 had ICDs and 274 were pacemaker leads. Incomplete removal occurred in 9.1% of pacemaker patients and in 4.4% of ICD patients. Better success in the ICD patients is probably related to a shorter length of implant time before the need for extraction compared with the pacemaker patients. Death occurred in 0.4% of pacemaker patients compared with 3.3% of ICD patients. All deaths occurred as the result of sepsis and not the extraction or a complication of the extraction. The ICD group, having inferior cardiac function, may have been less able to withstand the rigors of severe sepsis. Cure of Infection Our only case of recurrent infection was an instance of placement of a new transvenous system just before the removal of a device with a pocket infection. We would now delay implant of a new device for at least a few days to ensure eradication of active infection. Comment Analysis of factors that may have contributed to causing an infection about the device that required extraction revealed three important factors. Four percent of our infected patients were dialysis patients. These patients frequently have trouble maintaining adequate access for dialysis and end up having long-term plastic percutaneous catheter access rather than an arteriovenous fistula. These catheters inevitably become infected and can lead to systemic sepsis. We have adopted a policy of implanting their new device with epicardial leads in these patients to avoid exposure to the vascular system with recurring bacteremia. Dialysis patients also present challenges for repeated transvenous access because most implanters will avoid implant on the side of a functioning arteriovenous fistula, and previous access procedures may have damaged many of the usual venous pathways. Dialysis patients require pacemaker therapy approximately twice as often (0.68% vs 0.29%) as the general population [4]. This increase would however not explain the large number of dialysis patients presenting for extraction with infection. We cannot know the frequency of dialysis patients in the population from which our referrals for infected pacemakers and ICDs arises because our referrals come from a 3-state area. However, 4% of our infected cases were dialysis patients, whereas only 0.8% of our noninfected cases were dialysis patients. This would suggest a higher incidence of infection in pacing systems in dialysis patients (Table 1). Chang and colleagues [5], studying a population of dialysis patients, found that patients with endocarditis were far more likely to have a history of pacemaker 1415 implantation (15% vs 1.1%, p 0.01). Hayes and colleagues [6] compared complications of pacemakers in dialysis patients with matched controls and found infection in 6 of the 70 dialysis patients and in none of the matched controls. A history of Coumadin therapy was present in 31% of our patients with infection and in only 5% of our noninfected patients (Table 1). With increasing attention to strokes in patients with atrial fibrillation, many physicians have been reluctant to stop anticoagulation or have been insistent in restarting anticoagulation immediately after pacemaker or ICD implantation. We believe this may lead to pocket hematoma, potentiating infection. Cheng and colleagues [7] found that the use of warfarin within 3 days of implantation was associated with a 16 in 51 incidence of pocket hematoma. Avoidance of heparin in the postoperative period decreased hematoma formation in the Coumadin patients by fourfold. Michaud and colleagues [8] also found a 20% incidence of pocket hematoma in patients treated with heparin after implant but only a 4% incidence when Coumadin alone was used. Perioperative stroke or valve thromboses are the most devastating complications, and some hematoma formation may have to be accepted rather than risk these complications. However, it would appear that heparin should be avoided in the perioperative period and that all efforts using homeostasis should be made to avoid hematoma and possible increased infectious complication. A history of repeat procedure for battery depletion, device malfunction, or upgrade was present in 47% of the infected patients and in 20% of our noninfected patients (Table 1). Again, we cannot know the percentage of multiple implants that were present in the population from which our infected patients arose due to our 3-state referral area. However, Millennium Research Group [9] found that 23.9% of single-chamber pacemakers, 20.6% of dual-chamber pacemakers, and 28.3% of biventricular pacemakers represented replacements or repeat implant for upgrade. Thus, a 47% incidence of multiple implants in the infected group would suggest an association of multiple implants and infections. This observation has been made by others. Catanchin and colleagues [10] found an almost fivefold increase in infection after repeat procedures rather than primary implants. Likewise, Harcombe and colleagues [11] found patients undergoing elective unit replacement were at particular risk of complications. The initial implant complication rate was 1.4% but rose to 6.5% at a repeat visit to the pocket [11]. These complications were infection or erosion, a variant of infection. With the availability of the laser sheath in 2000, we initially attempted to remove the lead using locking stylets and traction. Because the laser sheath quickly demonstrated that it offered a more rapid and less aggressive mobilization of the lead than simple plastic sheath dissection, the laser was immediately brought into the field when simple traction with stylets was unable to remove the lead. It is therefore not possible to estimate how many leads could have been removed with plastic

6 1416 KRATZ AND TOOLE Ann Thorac Surg SAFE SURGICAL PACEMAKER LEAD EXTRACTION 2010;90: sheaths alone. In the laser era, complete extraction rose from 89.5% to 93%, and bleeding fell from 3.2% to 1.9%. These results are not statistically significant. However, our impression is that the laser era was associated with leads that had been in place for longer periods and thus as a group were far more challenging extractions. Unfortunately, the records reviewed did not consistently document the duration of lead implantation adequately to know definitely. Although the lack of statistical significance, increasing operator experience over time, and the retrospective nature of this review make scientific conclusions about the benefit of the laser difficult, personnel experience of the operator indicated the availability of the laser made the procedures quicker and more successful in obtaining a complete extraction. Experience reported in the literature supports this conclusion. In the period before laser extraction, Fearnot and colleagues [12] reported a 15% incomplete removal of leads, whereas Brodell and colleagues [13] had a similar 20% incomplete removal. We would not have been able to accomplish our 89.5% complete removal of leads during the prelaser period without the use of the femoral approach workstation or transjugular rigid forceps extraction. These two techniques were only used when other techniques were unsuccessful. The decrease in need for femoral extraction from 14% to 7% (p.043) thus further demonstrates the improved extraction rates during the laser era. The Pacemaker Lead Extraction With the Laser Sheath: Results of the Pacing Lead Extraction With the Excimer Sheath (PLEXES) trial, which was a prospective randomized trial comparing locking stylets and simple plastic sheaths with addition of the laser sheath, resulted in complete removal in 64% with the simple plastic sheath and in 94% with the laser. In the failed nonlaser group, subsequent use of the laser allowed complete removal in 88%. Operating time was also shorter in the laser group [14]. Other recent reports have shown excellent complete lead extraction rates in high-volume experienced centers of 96% for Wazni and colleagues [15], 89% for Kennergren [16, 17], and 97.5% for Jones and colleagues [18]. Our similar rate of 93% for complete extraction in the laser era compares similarly with these reports. Despite advances in tools and techniques, lead extractions continue to have the risk of sudden catastrophic bleeding with possible associated death. Our bleeding rate requiring thoracotomy of 3.2% in the prelaser period but falling to 1.9% in the postlaser era (p NS) is similar to rates reported in other recent series, including 3 of 153 (2.0%) in laser cases for Wilkoff and colleagues [19], 3 of 149 (2.0%) for Kennergren [16], and 5 of 189 (2.6%) for Sohail and colleagues [20]. Bleeding might be slow and the resulting cardiac tamponade might be manageable with volume resuscitation for some period of time. Our experience, however, is that most bleeding episodes are associated with the sudden development of severe cardiac tamponade. Only very rapid sternotomy will salvage these fragile patients in this setting. The involvement of a cardiothoracic surgeon in lead extraction procedures has helped to avoid deaths associated with catastrophic bleeding [21]. The protocol we have developed and followed has prevented any procedurally related death in our series. Our experience managing our 8 patients with bleeding leads us to believe these steps were extremely helpful in managing these potentially morbid events without death or further complications. None of the deaths in our patients were associated with a bleeding event. Our recommended protocol includes: Preoperative type and cross or screen for all patients General anesthesia Arterial catheter monitoring Large-bore intravenous access Temporary cardiac pacing, both transvenous and transcutaneous available All procedures performed in the operating room Sterile preparation and draping to allow immediate median sternotomy and femoral access Equipment available for immediate sternotomy Nursing and surgeon available for immediate sternotomy Transesophageal echocardiogram monitoring High quality fluoroscopy by late generation C arm Complete variety of extraction tools available, including laser Extensively experienced operator performing extractions (minimum, 20 cases per year) Our definition of immediate sternotomy, based on experience with bleeding, is opening the sternotomy and control of bleeding within 2 minutes. Although we do not believe a surgeon must do these procedures, we do believe a surgeon must be available in this extremely short period of time to prevent rare but unnecessary death. Our recommendations closely mirror those from the Heart Rhythm Society consensus published in July 2009 [19]. They allow for the possibility of procedures being performed in a procedural laboratory specifically designed for device implantation procedures that would allow for emergency thoracotomy. Although we would agree with this position, most cardiac catheterization laboratories are not set up to support large open cardiac operations and would not be adequate if an intervention was needed for a massive hemorrhage. Fortunately, modern portable fluoroscopy units provide excellent imaging and allow the procedures to be performed in an operating room. Our short-term procedural death rate was 0.0%. This compares with other similar reports (Wazni and colleagues, 0.27% [15]; Kennergren [17], 0.0%; Jones and colleagues [18], 0.0%; Wilkoff and colleagues [14], 0.3%; and Sohail and colleagues [20], 0.5%). Our series, however, did recognize a 1.1% incidence of late nonprocedurally related deaths. All deaths in this series were related to preoperative severe sepsis that persisted despite complete extraction of all leads without adverse event. An opportunity exists to educate our referring physicians and patients regarding the hazards

7 Ann Thorac Surg KRATZ AND TOOLE 2010;90: SAFE SURGICAL PACEMAKER LEAD EXTRACTION of delayed referral for extraction when infection associated with a pacemaker or ICD occurs. Earlier intervention before advanced endocarditis with multisystem failure occurs may, hopefully, avert these deaths. References 1. Furton T. Medtronic says wires may have had role in 13 deaths. Wall St J 2009;Mar 14:B5. 2. Kratz J, Leman RB, Gillette PC. Forceps extraction of permanent pacing leads. Ann Thorac Surg 1990;49: Khitin L, Kim K, Kratz J. Transxyphoid approach to retrieval of retained pacemaker leads using laser sheath. Pacing Clin Electrophysiol 2005;28: Leman RB, Kratz JM, Gazes PC. Permanent cardiac pacing in patients on chronic renal dialysis. Am Heart J 1985;6: Chang CF, Kuo BI, Chen TL. Infective endocarditis in maintenance hemodialysis patients: fifteen years experience in one medical center. J Nephrol 2004;17: Hayes DL, Hyberger LK, Hodge DO. Pacemaker and ICD complications in patients on hemodialysis: Does Hemodialysis increase risk? Heart Rhythm 2005;2:S Cheng M, Hua W, Chen K. Perioperative anticoagulation for patients with mechanic heart valves undertaking pacemaker implantation. Europace 2009;11: Michaud GF, Pelose F, Noble MD. A randomized trial comparing heparin initiation 6 h or 24 h after pacemaker or defibrillator implantation. J Am Coll Cardiol 2000;35: Millennium Research Group. Global markets for cardiac rhythm management devices. 2009; Exhibit 40, 175 Bloom St, Toronto, Ontario, Canada. 10. Catanchin A, Murdock CJ, Athan E. Pacemaker infections: a 10-year experience. Heart Lung Circ 2007;16: Harcombe AA, Newell SA, Ludman PF. Late complications following permanent pacemaker implantation of elective unit replacement. Heart 1998;80: Fearnot NE, Smith HJ, Goode LB. Intravascular lead extraction using locking stylets, sheaths, and other techniques. Pace 1990;13: Brodell GK, Castle LW, Moloney JD. Chronic transvenous pacemaker lead removal using a unique, sequential transvenous system. Am J Cardiol 1990;66: Wilkoff BL, Byrd CL, Love CJ. Pacemaker lead extraction with the laser sheath: results of the Pacing Lead Extraction With the Excimer Sheath (PLEXES) trial. J Am Coll Cardiol 1999;33: Wazni O, Epstein LM, Ervin CM. The Lexicon study: a multicenter observational retrospective study of consecutive laser lead extractions. Heart Rhythm 2009;6:S Kennergren C. Excimer Laser assisted extraction of permanent pacemaker and ICD leads: present experiences of a European multi-centre study. Eur J Cardiothorac Surg 1999; 15: Kennergern C. A single centre experience of over one thousand lead extractions. Europace 2009;11: Jones JO, Eckart RE, Albert CM, Epstein LM. Large, singlecenter, single-operator experience with transvenous lead extraction: outcomes and changing indications. Heart Rhythm 2008;5: Wilkoff BL, Byrd CJ, Bongiorni MG. Transvenous lead extraction: Heart Rhythm Society expert consensus on facilities, training, indications, and patient management: this document was endorsed by the American Heart Association (AHA). Heart Rhythm 2009;6: Sohail MR, Uslan DZ, Khan AH. Management and outcome of permanent pacemaker and implantable cardioverterdefibrillator infections. J Am Coll Cardiol 2007;49: Gaca JG, Lima B, Milano CA. Laser-assisted extraction of pacemaker and defibrillator leads: the role of the cardiac surgeon. Ann Thorac Surg 2009;87: