Ann Thorac Surg 2008;85:50-55. doi:10.1016/j.athoracsur.2007.03.048
© 2008 The Society of Thoracic Surgeons
Original Articles: Cardiovascular
Explantation of Implantable Defibrillator Leads Using Open Heart Surgery or Percutaneous Techniques
Daniele Camboni, MD*,a,*,
Christian G. Wollmann, MD*,b,
Andreas Löher, MDa,
Rainer Gradaus, MDb,
Hans Heinrich Scheld, MDa,
Christof Schmid, MDa
a Department of Thoracic and Cardiovascular Surgery, University Hospital, Muenster, Germany
b Department of Cardiology and Angiology, University Hospital, Muenster, Germany
Accepted for publication March 19, 2007.
* Address correspondence to Dr Camboni, Klinik fuer Thorax-, Herz- und Gefaesschirurgie, Uniklinik Muenster, Albert-Schweitzer Strasse 33, Muenster, D-48129, Germany (Email: dcamboni{at}arcor.de).
 |
Abstract
|
|---|
Background: To remove failing or infected implantable cardioverter defibrillator leads, percutaneous techniques and open heart surgery are two common approaches. However, well-defined indications for either technique are not available. We summarize our experience with implantable cardioverter defibrillator system explantation using open heart surgery and percutaneous lead removal.
Methods: A total of 1,391 transvenously introduced implantable cardioverter defibrillator systems were implanted during the analyzed time interval from January 1995 to June 2005 in our institution. In 21 patients (1.5%), open heart surgery for implantable cardioverter defibrillator lead and generator explantation was applied (group A), and in 53 patients (3.8%), a percutaneous lead removal was possible (group B). The log-rank test was used to calculate differences in survival between both patient groups, and the Student’s t test was applied for differences in nonlethal complications.
Results: The 30-day, 6-month, 12-month, and 5-year survival rates were 91%, 91%, 81%, and 71%, respectively, for group A patients, and 100%, 100%, 94%, and 78%, respectively, for group B patients, which was not statistically different (p = 0.11). After open heart surgery, survival was comparable for cases with lead removal because of lead infection and those with lead malfunction (p = 0.28); however, patients with open heart surgery had a longer hospital stay (p = 0.03). Student’s t test revealed no statistical difference in nonlethal complications between both patient groups (p = 0.37).
Conclusions: As open heart surgery yielded similar results with regard to survival and complications, implantable cardioverter defibrillator lead removal using extracorporeal circulation may be well justified as a last therapeutic option, eg, in case of large bacterial vegetations.
|
Introduction
|
|---|
The introduction of transvenous implantable cardioverter defibrillator (ICD) leads greatly facilitated the implant procedure as thoracotomy and sternotomy were no longer necessary [1]. The indications for device placement progressively broadened, and the number of ICD implantations steadily increased. As a consequence, the number of ICD leads with malfunction, structural failure, and infection rose and is still rising. Moreover, advances in technology additionally have promoted system upgrades and device exchanges.
There are several approaches to remove transvenously introduced ICD leads. If leads have been in place for only a short period, they can frequently be removed by simple traction. If a tight scar tissue withholds the leads during traction, the force applied to the leads is limited by the tensile strength of the insulation and conductor coils. Locking stylets allow a more forceful tension, but successful lead removal can still be hampered by invagination of the myocardium, myocardial rupture, arrhythmia, hypotension, or avulsion of a tricuspid valve leaflet. Failure to retrieve transvenous leads, especially when they are firmly attached to the myocardial wall or tricuspid valve, as well as endocarditic vegetations exceeding 10 mm in size, are commonly accepted indications for open heart surgery using cardiopulmonary bypass [2, 3]. Some centers define any size of vegetations, even far less than 10 mm, as an indication for cardiopulmonary bypass, because of the risk of spreading septic emboli and a possible underestimation by echocardiography [4]. However, sharply defined indications for either technique do not exist. With this present report we summarize our experience with ICD lead explantation using either open heart surgery with cardiopulmonary bypass or percutaneous lead extraction.
|
Patients and Methods
|
|---|
Patient Groups
After institutional review board approval with a waiver of consent, a 10-year retrospective analysis was performed. During the time interval from January 1995 to June 2005, a total of 1,391 patients underwent implantation of a transvenously introduced ICD system at our institution. In 21 patients (1.5%), an ICD lead and generator explantation using open heart surgery and cardiopulmonary bypass had been performed (group A). A group of 53 patients (3.8%) who underwent percutaneous lead removal without cardiopulmonary bypass served as a control (group B).
Mean age of group A (n = 21, 13 male) was 53 ± 24 years. Left ventricular ejection fraction was slightly impaired with a mean of 0.49 ± 0.21. Indications for ICD implantation included survival of cardiac arrests (16 patients) and recurrent episodes of hemodynamically compromising ventricular tachycardia without an event of resuscitation (5 patients). Nine patients had underlying coronary artery disease, 6 patients had dilative cardiomyopathy, and 2 patients presented with arrhythmogenic right ventricular cardiomyopathy (Table 1). In 16 patients, only one lead had been placed, whereas 4 patients had two leads and 1 had three leads implanted. The number of previous ICD-related operations including implantation and surgical revisions amounted to one operation in 6 patients, two operations in 5 patients, and more than two in 10 patients. The average interval between initial lead placement and lead extraction was 6 ± 1 years. The mean interval since the last surgical procedure (without lead replacement) to the final system explantation was 1.6 ± 2.6 years.
The demographic data of group B patients (n = 53, 40 male), who were operated on using percutaneous techniques, was similar with regard to age, sex, cardiac diagnosis, and left ventricular ejection fraction (Table 1). The decision for percutaneous lead removal and replacement of the defective leads was mainly owing to relatively short overall implant duration, which was less than 1 year in one third of patients.
Indications for Surgery
The main indication for ICD lead and generator explantation using open heart surgery and cardiopulmonary bypass was infection with echocardiographically verified large vegetations (11 patients). Three patients of the latter group were in septic condition with high temperatures despite extensive antibiotic treatment requiring hemodynamic support with vasopressors. The most frequently detected causative microorganism was staphylococcus. Coagulase-negative staphylococci were detected in 3 patients (2 Staphylococcus epidermidis, 1 Staphylococcus hominis). The coagulase-positive Staphylococcus aureus was isolated in another 3 patients. Other causative bacteria were Escherichia coli (1 patient), Salmonella enteritis (1 patient), Streptococcus agalactiae (1 patient), and Streptococcus bovis (1 patient). In the remaining patient with echocardiographically verified vegetations and signs of infection, the microbiologic tests did not isolate a microorganism. Two of these patients had previously undergone an unsuccessful attempt of transvenous lead extraction.
Ten patients presented with lead dysfunction, including lead fracture, insulation defect, and undersensing or oversensing without a conductor or insulation problem. In 8 of them, a trial of transvenous extraction had failed before. Open heart surgery was indicated as the defective lead was likely to interfere with the new lead, which would have resulted in an inadequate ICD function. One patient with diagnosed lead dysfunction needed an operative myocardial revascularization in addition.
Surgical Procedures
Open heart surgery (group A)
In all patients, both leads and generator were removed regardless of the indication for surgery. The generator was removed at the site of implantation, either at the beginning of surgery or after termination of extracorporeal circulation. In 17 patients, the generator had been implanted below the pectoral muscle, and in 4 patients, into the posterior sheath of the abdominal rectus muscle. For lead removal, extracorporeal circulation was instituted through a median sternotomy. Bicaval cannulation was used in 7 patients, and cannulation of the inferior cava only was used in the remainder according to the surgeons’ preference. In all cases, both caval veins were provided with a vessel loop and tourniquet to control bleeding after incising the right atrium. Both methods similarly allowed sufficient exposure of the leads. In all patients, the ICD system and all leads could be completely removed. All patients except the one with concomitantly performed coronary artery bypass grafting surgery were operated on with the heart beating. Three patients with device infection and large vegetations presented with involvement of the tricuspid valve, ie, the lead was firmly attached to one of the leaflets. One tricuspid valve could be reconstructed. In the second patient, valve replacement was necessary. In the third patient, part of the tricuspid valve was excised without an option of further reconstructive surgery or valve replacement because of infected and extremely fragile tissue.
Percutaneous lead removal (group B)
In 17 patients, simple traction was used to explant the defective leads. In the remaining 36 patients, various extraction devices were applied, mainly locking stylets and polytetrafluoroethylene sheaths. In general, a stepwise protocol for extraction was followed, using simple traction first. If traction alone was not successful, locking stylets were introduced to protect the integrity of the lead during traction. If this approach failed, polytetrafluoroethylene extraction sheaths were inserted over the lead and forwarded into the subclavian vein. A laser sheath was not available in our hospital. Lead removal was complete in all patients except for 2 in whom a negligible fragment of a lead remained in situ. In all 53 patients, a new lead was implanted during the same surgical intervention through puncture of the subclavian vein.
Statistical Analysis
Statistical analysis was performed with the Statistical Package for Social Sciences (SPSS), version 12.0 (Chicago, IL). Qualitative differences in demographics and results of both evaluated groups were compared using Student’s t test. A value of p less than 0.05 was considered significant. In addition, Kaplan–Meier calculation for cumulative survival was applied. The log-rank test computed differences in survival between both patient groups.
|
Results
|
|---|
All except 2 patients who underwent open heart surgery (group A) survived the perioperative 30-day period (91%). The survival rates after 6 months, 12 months, and 5 years were 91% (19 patients), 81% (17 patients), and 71% (15 patients; Table 2).
Six patients (28%) died after a mean interval of 13 months after surgery, among them 2 patients with tricuspid surgery. Three patients (14%) died in-hospital during the same hospitalization. Four patients (19%) experienced nonlethal complications. One patient experienced acute renal failure. This patient with a primarily infected ICD system also exhibited deep sternal wound infection, which required delayed closure. Two patients had a hemothorax, one of whom had to be reoperated on, and in 1 patient a pneumothorax occurred. Mean hospitalization after ICD system explantation was 37 ± 13 days in the patients without lead infection, and 80 ± 31 days in those with infected systems (p = 0.04).
After a mean interval of 36 ± 26 days after generator and lead explantation, 18 patients (85%) were provided with a new ICD system. In 16 patients, the new ICD leads could be implanted transvenously. After angiographic assessment of the patency of the respective subclavian vein, reimplantation at the primary site of implantation was possible in 5 patients. In the remaining 11 patients, the opposite site had to be used owing to occlusion of the initially entered subclavian vein. Two patients required epicardial ICD leads through a small thoracotomy because of thrombosis of the upper caval vein. Log-rank analysis did not demonstrate a difference in survival in patients with infective leads compared with patients with lead malfunctions (p = 0.28; Fig 1).
View larger version (12K):
[in this window]
[in a new window]
|
Fig 1. Kaplan–Meier curves for cumulative freedom of death for patients with infective leads (continuous line) versus patients with lead malfunction (broken line) in group A (log rank = 1.16; degrees of freedom = 1; p = 0.28; 95% confidence interval for patients with infective leads, 24 to 64 months; 95% confidence interval for patients with lead malfunction, 50 to 90 months).
|
|
All group B patients, who underwent percutaneous lead removal, survived the perioperative period. The survival rates of group B patients after 6 months, 12 months, and 5 years were 100% (53 patients), 94% (50 patients), and 78% (41 patients). Nonlethal complications occurred in 8 patients of group B (15%), including pneumothorax, hemothorax, pericardial effusion, pocket hematoma, and wound infection.
Survival between patients undergoing major open heart surgery and patients who had percutaneous lead removal was comparable (p = 0.11; Fig 2). The incidence of nonlethal complications was also not statistically different when comparing groups A and B (p = 0.37). However, the hospital stay was significantly shorter for patients with percutaneous lead removal (12 ± 8 days) in contrast to patients with major open heart surgery (58 ± 37 days; p = 0.03).
View larger version (11K):
[in this window]
[in a new window]
|
Fig 2. Kaplan–Meier curves for cumulative freedom of death for group A patients after lead extraction using cardiopulmonary bypass (continuous line) versus group B patients after percutaneous lead removal (broken line; log-rank = 2.53; degrees of freedom = 1; p = 0.11; 95% confidence interval for group A, 44 to 78 months; 95% confidence interval for group B, 60 to 74 months).
|
|
|
Comment
|
|---|
The increasing number of ICD implantations after the demonstrated superiority of ICD therapy as compared with antiarrhythmic medication poses a progressive burden of lead malfunction and lead infection on the surgeon [5]. Because device infection can only be cured by complete removal of the ICD system, and placement of multiple intracardiac electrodes may hinder adequate device function, generator and lead removal has become an increasingly frequent task.
Although the exchange of an ICD generator can be performed in a rather simple surgical procedure, withdrawal of malfunctioning leads is more demanding. Malfunctioning leads can be removed by simple traction or by means of steel, polypropylene, polytetrafluoroethylene, or laser sheaths [6]. Complications of these approaches include hemopericardium and pericardial tamponade, hemothorax, pulmonary embolism, and migrating lead fragments [6]. However, there are no well-defined indications for either procedure, and large studies comparing various extraction techniques are lacking. Success and complication rates are usually derived from small series. In one of the largest studies using the rather uncommon laser sheaths for removal of pacemaker and ICD leads, an overall complication rate of 3.6% and a perioperative mortality rate of 0.8% was reported. Similarly, the US Lead Extraction Database showed a fatal or near fatal complication rate of only 2.5% for patients, who underwent intravascular lead extraction with a mortality rate of 0.6% [7]. These results concerning percutaneous lead removal are comparable with our findings; however we observed a slightly higher complication rate.
When the leads are firmly attached to the tricuspid valve or the venous vessel wall, transvenous lead extraction is dangerous or even impossible. How harmful a transvenous extraction could be is depicted in Figure 3, in which an ICD lead with an attached tricuspid valve leaflet is shown. Similarly, in patients with large vegetations or thrombi covering the leads, simple traction is associated with displacement of thrombi or infectious vegetations into the pulmonary vascular bed (Fig 4). The aforementioned patients supposedly are much better off with open lead removal, ie, using cardiopulmonary bypass. The risk of uncontrolled mechanical injury to intracardiac structures (mainly the tricuspid valve) or vessels and dissemination of vegetations in case of infection is lower [8]. The open heart approach has the advantage of direct vision of the field of interest and allows additional surgical procedures to be performed, such as reconstruction of the tricuspid valve or coronary artery bypass grafting surgery. Unfortunately, clearly defined indications for using open heart surgery to remove ICD leads do not exist. Reasons for that include the strong preference of less invasive surgery, ie, of percutaneous extraction techniques.
View larger version (138K):
[in this window]
[in a new window]
|
Fig 3. Picture of a transvenously extracted implantable cardioverter defibrillator lead with an attached tricuspid valve leaflet. Notice the myocardial tissue at the tip of the implantable cardioverter defibrillator lead.
|
|
View larger version (138K):
[in this window]
[in a new window]
|
Fig 4. Intraoperative view into the right atrium (RA) with a large vegetation surrounding an implantable cardioverter defibrillator lead. (EC = extracorporeal circulation access; V = infective vegetation.)
|
|
Studies focusing on the removal of ICD systems using open heart surgery have been restricted to case reports or small series [9, 10]. More data are available from pacemaker lead extraction. Brodmann and colleagues [11] used cardiopulmonary bypass to remove pacemaker leads in 9 patients with one surgery-related complication. Wilhelm and associates [12] published a 30-day mortality of 12.5% with 6 of 8 patients surviving to a follow-up of 18 months. An increase of the complication rate was noticed if cardiopulmonary bypass had to be used before signs of infection had resolved [12]. Starck and coworkers [13] observed no complications and an uneventful postoperative course in 13 patients after thoracotomy and cardiopulmonary bypass. Similar results were obtained by Niederhäuser and colleagues [14] and Cacoub and associates [15]. Our present study comparing the open heart surgery approach with percutaneous techniques for lead removal revealed in part contradictory results in comparison with the above studies. One reason for this discrepancy between the outcome of patients after pacemaker lead extraction and our results is certainly the difference in patient populations. The underlying heart disease is usually much more severe in ICD patients as compared with patients with pacemaker systems. Even if there are great similarities between pacemaker and defibrillator leads, especially with regard to design and components, ICD leads are larger in diameter and usually have a greater propensity toward endocardial attachment [16]. Therefore, the extraction of ICD leads is more complex in contrast to pacemaker lead extraction. Hence, the outcome of patients after pacemaker lead extraction cannot be extrapolated without limitations to ICD lead removal.
We lost 6 of 21 patients after open heart surgery after a mean interval of 13 months, 3 of them dying in-hospital. It is impossible to state whether the observed mortality is in part related to the surgical procedure itself or a consequence of the poor medical condition of the patients. This is also true for the 7 of 53 patients who died after percutaneous lead removal after a mean follow-up of 21 months. Cumulative survival after open heart surgery and percutaneous ICD extraction did not reveal a significant difference. However, the Kaplan–Meier curves displayed a tendency toward a better survival of patients after percutaneous ICD removal. As patients with open heart surgery have a longer necessity of ICD implantation, and usually are much sicker, especially when being burdened with systemic bacterial infection, these mild differences should rather been seen in favor of open heart surgery. In our opinion, the presented data well demonstrate the feasibility of open heart surgery with quite acceptable results, even in septic conditions.
Nonlethal complications in both patient groups were not different. However, as expected, patients after percutaneous lead removal had a faster recovery, which is indicated by a shorter hospital stay, in comparison with patients after open heart surgery. Surprisingly, we observed no significant difference in survival between patients with infected leads and patients with malfunctioning leads. The lack of an increased complication rate in patients with infected leads can also be regarded as a testimony to the efficacy of the open heart technique to completely abolish the infectious focus.
This study is a single-center retrospective analysis. Different devices, leads, and implantation and explantation techniques were used during the retrospective observational interval of up to 10 years.
In conclusion, we found similar results for patients after open heart surgery and patients undergoing percutaneous ICD lead removal. For the reason of less invasiveness and shorter hospital stay, a transvenous approach should remain the first step for ICD lead removal. In case of failed or incomplete transvenous ICD lead removal, as well as in case of large endocarditic vegetations, open heart surgery may well be used with an acceptable risk.
|
Footnotes
|
|---|
* Daniele Camboni, MD, and Christian G. Wollmann, MD, contributed equally to this article. 
|
References
|
|---|
- Block M, Hammel D, Borggerfe M, et al. Erste klinische Erfahrungen mit einem transvenös-subkutanen Defibrillatorsystem Z Kardiol 1991;80:657-664.[Medline]
- Danish Pacemaker and ICD Register1999. Pacing Clin Electrophysiol 2000;23(10 Pt 2):S1-S93.[Medline]
- Love CJ, Wilkoff BL, Byrd CL, et al. Recommendations for extraction of chronically implanted transvenous pacing and defibrillator leads: Indications, facilities, training Pacing Clin Electrophysiol 2000;23:544-551.[Medline]
- Hemmer W, Fröhlig G, Markewitz A. Kommentar zu den NASPE-Empfehlungen zur Entfernung von permanent implantierten, transvenösen Herzschrittmacher und Defibrillatorsonden Z Kardiol 2002;91:956-968.[Medline]
- Moss AJ, Hall WJ, Cannom DS, et al. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmiaMulticenter Automatic Defibrillator Implantation Trial Investigators. N Engl J Med 1996;335:1933-1940.[Abstract/Free Full Text]
- Epstein LM, Byrd CL, Wilkoff BL, et al. Initial experience with larger laser sheaths for the removal of transvenous pacemaker and implantable defibrillator leads Circulation 1999;100:516-525.[Abstract/Free Full Text]
- Smith HJ, Fearnot NE, Byrd CL, Wilkoff BL, Love CJ, Sellers TD. Five years experience with intravascular lead extractionUS Lead Extraction Database. Pacing Clin Electrophysiol 1994;17:2016-2020.[Medline]
- Klug D, Lacroix D, Savoye C, et al. Systemic infection related to endocarditis on pacemaker leads: clinical presentation and management Circulation 1997;95:2098-2107.[Abstract/Free Full Text]
- Renzulli A, Vitale N, D'Onofrio A, Cortrufo M. Implantable cardioverter defibrillator malfunction due to transvenous lead insulation break Pacing Clin Electrophysiol 1994;17245–6.
- Jordaens L, Van Balleghem Y, Herregods L. Removal of endocardial defibrillation leads Pacing Clin Electrophysiol 1996;19:127-129.[Medline]
- Brodmann R, Frame R, Andrew C, Furman S. Removal of infected transvenous leads requiring cardiopulmonary bypass or inflow occlusion J Thorac Cardiovasc Surg 1992;103:1164-1165.[Abstract]
- Wilhelm MJ, Schmid C, Hammel D, et al. Cardiac pacemaker infection: surgical management with and without extracorporal circulation Ann Thorac Surg 1997;64:1707-1712.[Abstract/Free Full Text]
- Starck C, Hemmer W, Böhm JO. Surgical pacemaker lead extraction: 10 years of experience Herschr Elektrophy 2001;12:217-224.
- Niederhäuser U, von Segesser LK, Carrel TP, et al. Infected endocardial pacemaker electrodes: successful open intracardiac removal Pacing Clin Electrophysiol 1993;16:303-308.[Medline]
- Cacoub P, Leprince P, Nataf P, et al. Pacemaker infective endocarditis Am J Cardiol 1998;82:480-484.[Medline]
- Epstein AE, Kay GN, Plumb VJ, et al. Gross and microscopic pathological changes associated with nonthoracotomy implantable defibrillator leads Circulation 1998;98:1517-1524.[Abstract/Free Full Text]
Top
Abstract
Introduction
