Infected Endocardial Pacemaker Electrodes: Successful Open Intracardiac Removal

Infected Endocardial Pacemaker Electrodes:

Successful Open Intracardiac Removal URS N I E D E R H A U S E R , L U D W I G K. VON S E G E S S E R , THIERRY P. CARREL, ANDREAS LASKE, ERWIN BAUER, MARIETTE SCHONBEGK, and MARKO TURINA From the Glinic for Gardiovascular Surgery, University Hospital, Zurich, Switzerland N I E D E R H A U S E R , U . , ET AL.:

Infected Endocardial Pacemaker Electrodes: Successful Open Intracardiac

Removal. The long-term results after open intracardiac removal of infected pacing electrodes are presented. Methods: between 1985 and 1990 open intracardiac removal of 19 infected pacing electrodes was performed in seven patients (six male and one femalej, with a mean age of 56 years. The indications were; persisting bacteremia in three; generator pocket infection in four; endocarditis in one; and ventricular tachycardia caused by retracted electrodes in one. All electrodes were fixed in the right heart and extraction by closed methods failed. Percutaneous catheter techniques were not applied in these seven patients. In five patients two ventricula'r electrodes had to be removed, and in two patients a single one. A total of seven atrial electrodes were removed in six patients (one electrode each in five patients; two electrodes in one patient]. All atrial and two ventricular electrodes could be removed through a pursestring suture without use of a pump oxygenator. For the removal of ten ventricular electrodes in six patients (two electrodes each in four patients; 1 electrode each in two patients] a right-sided atriotomy was necessary with cardiopulmonary bypass (CPB). Simultaneously, five new pacing systems were implanted. Results; there were no early or late mortalities. In January 1991, all seven patients are alive and in a mean New York Heart Association Class 1,3 of heart failure after a mean interval of 33 months. In all cases the infection could be controlled with a simultaneous antimicrobial chemotherapy and the postoperative period was free of major complications. Conclusion; open intracardiac removal of retained pacing electrodes with or without use of CPB is a safe procedure without major complications. It is mandatory for all infected pacing electrodes that cannot be extracted by closed methods. (PACE, Vol. 16, February 1993] pacemaker electrode, infection, electrode explantation, cardiopulmonary bypass

Introduction Infection following pacemaker implantation is a rare complication with an incidence of about 1.5%,^"^ but remains a serious and life-threatening problem. Thete are reports documenting a mortality of up to 66% for infected retained electrodes.'*"'' For the vast majority of these patients, total removal of the infected pacing system combined

Address for reprints; Dr. Urs Niederhauser, Klinik fur Herzgefasschirurgie, Universitatsspital, CH 8091 Zurich, Switzerland. Fax: 01-255-44-46. Received December 27, 1991; revision May 20, 1992; revision July 21, 1992; revision September 16, 1992; accepted September 16, 1992.

PAGE, Vol. 16

with antibiotic chemotherapy is the optimal therapeutic procedure.^ Percutaneous extraction of electrodes may, however, fail. In several reports,^""^^ this situation of open intracardiac removal was found to be a safe surgical option. We report our experience with open intracardiac removal of infected pacing leads in seven patients with and without use of cardiopulmonary bypass (CPB).

Patients and Methods Between 1985 and 1990 seven patients (six men and one woman), with an average age of 56 years (range 22-75 years), underwent operation for infected permanent cardiac pacemakers and re-

February 1993

303

NIEDERHAUSER, ET AL.

tained endocardial electrodes; a chronological review of operative procedures and indications is given in Table I. Indications for the initial implantation of a permanent pacemaker were: total AV block in four; second-degree AV block with syncope in one; sick sinus syndrome in one; and carotid sinus syndrome in one. A total of 28 subsequent operative revisions (average 4) of the pacing systems were performed in all seven patients (9 [1 patient], 8 [1 patient], 3 each [2 patients], 2 each [2 patients], and 1 [1 patient]). These revisions consisted of eight pulse generator replacements for battery depletion, ten for generator pocket infection or skin ulceration, and two for changing the pacing mode. A total of 15 electrode revisions or replacements were performed: eight for skin ulceration or pocket infection, one for insulation defect, two for exit block, one for dislocation, and three for changing of the pacing mode. Before open electrode removal, all but one patient had at least one revision (four in one, two each in four, and one in one) of the generator pocket or electrode(s) for infectious complications. Lead extraction by closed percutaneous traction failed in all seven patients. Indications for complete removal of the pacemaker systems were endocarditis in one patient, and generator pocket infection and septicemia in three patients, respectively. In addition, ventricular tachycardias triggered by a disconnected lead made electrode removal necessary in a patient with generator pocket infection. Open heart surgery was indicated in three patients by the lead type (long helical screw on a bulbous tip) and localization at the hasis of the right ventricle near the tricuspid valve; in two patients by simultaneous cardiac procedures (tricuspid valve reconstruction, vena cava desobliteration); and in two patients because the percutaneous extraction devices were not available. The mean interval between primary implantation and open electrode extraction was 117 months (SD 54.8 months). The mean interval of time from the diagnosis of primary or recurrent infection to the time of open removal was 28 months (range 1-108 months). Surgical access was in all cases through a midsternotomy. Lead extraction was performed in one patient through a purse string suture without use of a pump oxygenator. In the remaining six patients a right-sided atriotomy was necessary with aid of CPB in normothermia and induced ventricu-

304

lar fihrillation. Blood and tissue cultures were taken in all patients.

Results A total of 19 electrodes (7 atrial, 12 ventricular) were removed in seven patients. Prior to open removal, closed percutaneous traction was applied but failed in these patients. The lead fixation was active (screw-in) in 18 electrodes and passive (tined tip) in one. In five patients two ventricular electrodes had to be removed, and in two patients a single one was removed. In one patient two atriai electrodes were explarited, and in five patients a single one was explanted. All atrial and two ventricular electrodes could be extracted through a purse string suture in the right atrium or ventricle without use of a pump oxygenator. In six patients ten ventricular electrodes (two electrodes each in four patients; one electrode each in two-patients) made atriotomy and use of CPB necessary. In one patient a tricuspid valve lesion (leaflet perforation by electrode) was simultaneously treated by surgical valve reconstruction. In a second patient a cava superior thrombectomy was also performed at the same operation. There were no mortalities and the postoperative period was free of major complications. The follow-up through January 1991 of all seven patients has a mean interval of 33 months (SD 32.4 months); at the end of the follow-up, the seven patients are in a mean NYHA Class 1,3 of heart failure. Blood and tissue cultures grew the following organisms: coagulase negative StaphyJococcus, 7; StaphyJococcus aureus, 5; Pseudomonas, 1; Citrobacter, 1; Enterobacter, 1; and KJebsieJJa, 1. Mixed infections with two organisms were seen in two patients. All infections were cured with a combined antimicrobial chemotherapy. The average hospital stay was 12.8 days (range 5-24 days). All infected and draining wounds closed. A simultaneous implantation of a new pacing system from the contralateral side was performed in five patients (VVIR in three, VVI in two) with a total of one endovenous and four epicardial electrodes. In no case was there an infection of a new system, nor was the treatmerit of primary infection prolonged. A 24-hour ECG recording was routinely performed, and did not show an indication

February 1993

PAGE, Vol. 16

GARDIAG PAGING, PAGING ELEGTRODE, INFEGTION, GARDIOPULMONARY BYPASS

Table I. Patients Gharacteristics Implantation Revision

Patient/Age

I.H./61

T.F./22

DDD 1985 PM expl. and EL shortened Nov87 Open expl. of EL Nov 87

VVIR impl. Feb 90 VVI impl. Jan 78 Replacement PM March 78 Replacement PM (DDD) Aug 81 Replacement PM Sept 87 EL revision Nov 87 PM replacement Jan 88 (DDD), Shortening EL EL impl. RVJan88 Replacement PM (VVIR) Shortening EL March 88 Open removal of EL May 88

S.F./66

VDD impi. March 81 EL replacement Apr 81 PM replacement Nov 81 EL impl. Feb 82 PM replacement Aug 82 EL replacement Feb 83

B.G./60

System repiacement (VVi) Jun 83 System change (DDD) Oct84 PM replacement (DDD) Nov 87 Open EL expl. Jun 90 + Cava sup. desobliteration + PM replacement (VVI, epic. EL) VVI impi. June 80 PM replacement June 82 PM expi., shortening EL 84 Open EL expi. + TVR Feb 89

indication

AVB lli Skin perforation. necrosis Poci