- Email: [email protected]
The exposed cardiac pacemaker
J
THORAC CARDIOVASC SURG
89:136-141, 1985
The exposed cardiac pacemaker Treatment by subfascial pocket relocation Ten patients were treated for skin erosion and exposure of chronically implanted pacemakers. The pulse generators bad been implanted either subcutaneously or in the subpectoral, submammary location. All patients were treated either by primary or secondary relocation of the pulse generator to deeper subfascial planes. Pacemakers relocated primarily to subfacial planes healed without resurfacing. Those pacemakers relocated to other superficial planes resurfaced but healed in a second subfascial t:ramlocation. Bacteriologic studies on these patients revealed that the orgllllisIm cultured or antibiotics utilized did not influence the fate of the relocated pulse generator. The subfasciallocation was the only determinant of the ultimate healing of the exposed pacemaker. We conclude that subfascial relocation of weU-fwtctioning pacemaker generators should be COIL'iidered as an alternative to complete replacement of the unit
Raul Garcia-Rinaldi, M.D., Ph.D., F.A.C.S., Jose M. Revuelta, M.D., Ph.D., Lucy Bonnington, PA.-C., and Luis Soltero-Harrington, M.D.,
Houston, Texas, and Cayey, Puerto Rico Consulting Statistician: Janine Pyrek, M.S., * St. Paul, Minn.
In 1961, Chardack, Gage, and Greatbatch' implanted the first cardiac pacemaker in the United States for treatment of patients with. life-threatening conduction abnormalities, particularly complete heart block. Currently, as in the past, many surgeons implant the electrode transvenously or epicardially by the subxiphoid, subcostal, or transthoracic approach. The generator is usually placed subcutaneously and occasionally subfascially, particularly if an epicardial pacemaker is implanted at the time of an open cardiac procedure. Regardless of route or technique of implantation, longterm results are excellent. Despite improvements in hardware and techniques, there remain complications associated with pacemaker insertion, such as lead malfunction or displacement. 2,3 Infection of the pulse generator pocket with subsequent
From The Houston Heart Institute and The Memorial Hospital Houston, Texas, and Department of Surgery, Universidad Central del Caribe, Cayey, Puerto Rico. Received for publication Aug. 31, 1983. Accepted for publication March I, 1984. Address for reprints: Raul Garcia-Rinaldi, M.D., Cardiovascular Surgery of Texas, Suite 110, 7500 Beechnut, Houston, Texas 77074. *Supervisor, Clinical Research, Surgical Products Division, 3M, St. Paul, Minn.
136
exposure of the generator are still problems associated with a high morbidity.' Grossly infected pacemaker systems with acute abscess formation require immediate removal and insertion of a new system. The incidence of chronic skin erosion by the generator with low-grade infection varies in different series between 0% and 12.6%. A number of methods have been utilized to combat this problem, with little success. Sowton, Hendrix, and Roy? treated patients with systemic and topical antibiotics with only a 5% cure rate. Chait and Ritchie' reported on a series of five patients treated by debridement, tissue advancement, and flap rotation; administration of systemic antibiotics; and removal of the pulse generator units. After infection was controlled, the units were then sterilized and reimplanted. Although this technique was innovative and successful, two patients had reexposure on three occasions. Chari and Suri" used similar methods in three patients and were successful in two patients. They concluded that the method was cumbersome and "logically the pacemaker should be removed and a new one implanted." Other authors have tried other variations, all of which have resulted in failure." 10 Most authors have arrived at the same conclusion: that the generator should be replaced. Although this premise is acceptable, it must be remembered that cardiac pacemakers are expensive,
Volume 89 Number 1
Exposed cardiac pacemaker 1 3 7
January, 1985
Table I. Results of cases of exposed pacemakers Case No. (age, sex) I (46, M)
2
Interval between first implantation and exposure
Original location
II mo 5 mo
Subcutaneous, submammary Subpectoral, submammary Subpectoral, submammary
9 mo
Subpectoral, submammary
12 mo 36 mo
Subpectoral, submammary Subcutaneous, subclavicular Subcutaneous abdominal wall Subcutaneous abdominal wall Subcutaneous abdominal wall Subpectoral, submammary
60 mo
(34, F)
3 (74, F) 4 (71, F) 5
36 mo
(41, F) 6
5 mo
(69, M) 7 (70, M) 8 (44, M)
9 (36, M) 10 (84, M)
48 mo
12 mo 120 mo 3 mo 37 mo
5 mo
Subcutaneous, submammary Subpectoral, submammary Subpectoral, submammary Subcutaneous, subclavicular
particularly the current multiprogrammable units with long battery life. Consequently, it would be highly desirable to develop a method that could salvage an exposed well-functioning unit.'! Our series consists of 10 patients with pulse generator exposure referred to us for treatment. These 10 patients are the only ones with this problem whom we have treated. Five other patients whose generators were translocated to subfascial planes before exposure occurred are not included. All the pulse generators in the series were implanted in subcutaneous pockets or under the thin pectoral muscle in the submammary region. From February, 1979, to the present, patients referred to us with exposed pulse generator units have been treated by relocation of the exposed pulse generator to deeper subpectoral or subfascial locations, and this report summarizes our experience. Of four pulse generators relocated to a subpectoral site, three (75%) became reexposed. All pulse generators relocated primarily (six) or secondarily (three) to a subfascial plane have remained in situ without reexposure, irrespective of the organism cultured. Follow-up time from the last relocation to the present, or time of death, averaged 24.6 months (3 to 44 months).
Final location
Result
Subpectoral, submammary
Reexposure II mo later
Subxiphoid, subfascial Subpectoral relocation of electrode Subcutaneous abdominal wall relocation of electrode Subxiphoid, subfascial Subpectoral, subclavicular
Recovered 25 mo postop. Reexposure of plate in 9 mo
Inside rectus sheath
Exposure of pacemaker 12 mo later Recovered 23 mo postop. Died 20 mo later with healed pacemaker Recovered 44 mo postop.
Inside rectus sheath
Recovered 40 mo postop.
Subxiphoid, subfascial
Recovered 23 mo postop.
Subxiphoid, subfascial
Recovered 35 mo postop
Subpectoral, submammary
Reexposure in 3 mo
Subxiphoid, subfascial Subxiphoid, subpectoral
Recovered 26 mo postop Recovered 7 mo postop
Subxiphoid, subfascial
Recovered 3 mo postop.
Patients and methods
The study consisted of 10 patients, six men and four women, seen from February, 1979, to July 1, 1983, and referred to us by other surgeons or cardiologists. Over this period we inserted approximately 350 transvenous or epicardial pacemakers in subcutaneous, subpectoral, or subfascial pockets without encountering the problem of skin erosion. Thus, in our experience, pulse generator exposure is uncommon. However, when it occurs, it can be a very challenging problem. All patients were hospitalized and the pulse generator pockets cultured. Empirical intravenous antibiotic coverage was started, depending on the generation of antibiotics in use at the time. After culture and sensitivities were obtained, the specific antibiotic was administered for at least 3 days. The patient was treated by pacemaker relocation, and intravenous antibiotics were continued for at least another 4 days. Surgical techniques
The infected pocket is opened wide and the generator unit freed from all attachments. The electrode is released from surrounding granulation tissue and from the fibrous tract so that it is freely movable. The pulse
138
The Journal of Thoracic and Cardiovascular
Garcia-Rinaldi et al.
Surgery
Table II. Frequency of success or failure in relocation of pacemakers by site of relocation" Beginning Site:
Site Subpectoral Subfascial Total
Success I
Failure 3
Total 4
9
o
9
10
3
13
'This table does not include the intermediate relocation to the subcutaneous tissue (failure) of Patient 2.
Final Site: (Success)
Fig. 1. Flow chart of site of relocation. This chart does not include the intermediate relocation to subcutaneous tissue (failure) of Patient 2. Dashed lines. Intermediate relocation (failure). Solid lines. Final relocation.
generator and the electrode are tested and thoroughly irrigated with saline solution and immersed in antibiotic or povidone-iodine* solution for 5 minutes. The generator is then relocated to a' subfascial or intrafascial location. We prefer the subfascial, subxiphoid area where it lies in the pre-peritoneal space. A vertical incision is made through the linea alba in the the subxiphoid area. A subfascial pocket is made in either side and thus the pulse generator is extraperitoneal. The electrode is tunneled to the new pocket and the pulse generator unit is reconnected immediately. The pocket is closed, without drainage, with absorbable polyglycolic acid sutures. The skin is closed with a subcuticular suture. Once the new pocket site is closed, the fibrous shell of the previous pulse generator is removed. A small drain is placed before closure to avoid the formation of a serous collection. The wound is closed subcuticularly with polyglycolic acid sutures. The drain is removed in 2 days. Results In seven patients the original pulse generators were in a subcutaneous position and were extruded at an average of 44 months (5 to 120 months). In two patients (Patients 1 and 8, Table I) the pulse generators were translocated from a subcutaneous, submammary pocket *Betadine, The Purdue Frederick Company, Norwalk, Conn.
to a subpectoral location. In both patients a second extrusion of the pulse generator necessitated a second translocation to a subfascial plane. Both recovered and are alive 25 and 26 months after operation, respectively. Two patients with subcutaneous, subclavicular pulse generators had generator translocation. One patient's generator (Patient 3) was transferred to a subpectoral, subclavicular position. This patient recovered but died of cardiac causes 20 months after operation. The other patient's pulse generator (Patient 10) was transferred to another subxiphoid, subfascial location. The patient recovered and is alive 3 months after the operation. Three patients (Patients 4, 5, and 6) had subcutaneous placement of their original pulse generator, but in the abdominal wall. The pulse generators were all relocated to subfascial areas under or inside the rectus sheath. All patients recovered and have been followed up for 44, 40, and 23 months, respectively. In three patients the original pulse generators were implanted in a subpectoral, submammary location. The pulse generator became exposed at 5, 12, and 37 months, respectively (average 18 months). One patient's indifferent electrode (Patient 2) was relocated in another subpectoral, submammary pocket and it resurfaced in 9 months. When the electrode finally was translocated to a subfasciallocation, the patient recovered and is alive 23 months postoperatively. In the other two patients (Patients 7 and 9) the pulse generator was transferred to a subfascial plane. Both have recovered and are alive 35 and 7 months after operation, respectively. A summary flow chart of location of pulse generators is shown in Fig. 1. Of the six pulse generators whose first relocation was to a subfascial area, all were successfully relocated with no resurfacing. Of the four whose first relocation was to a subpectoral area, only one was successfully relocated. The three that resurfaced were then transferred to a subfascial area. These three have not resurfaced. The hypothesis that successful relocation of pulse generators is independent of site of implantation was tested statistically. Table II shows the frequency of
Volume 89
Exposed cardiac pacemaker 1 3 9
Number 1 January, 1985
Table Case
no.
2
m. Pathogens and antibiotic coverage in cases of exposed pacemakers Antibiotics used
Bacterial culture
Anaerobe Propionibacterium Anaerobe Clostridium perfringens Staph. epidermidis Proteus morgagni Staph. mirabilis
Staph. epidermidis
No growth 4
Staph. epidermidls
5
Staph. aureus
6 7
No growth No growth Staph. aureus
Route
Length of treatment (days)
Length of hasp. stay (days)
IV
10
12
IV
7
8
IV
10
10
IV
7
7
IV
6 6
10 10
IV
I
3
Healed per primum
PO
6 3
II
Healed per primum
IV
5 10
14
Healed per primum
IV IV
9 7
10 10
Healed per primum Healed per primum
IV
2
Chloramphenicol I gm q8h Cefamandole I gm q6h C1indamycin 600 mg q6h Cefazolin sodium I gm q6h Gentamicin 80 mg q8h Chloramphenicol I gm q8h Cefazolin sodium I gm q8h C1indamycin 600 mg q6h Tobramycin 80 mg q8h Cefazolin sodium I gm q6h Gentamicin 80 mg q8h Clindamycin 600 mg q6h Ampicillin 1 gm q6h then 500 mg 'lid Chloramphenicol I gm q8h Tobramycin 80 mg q8h Cefazolin sodium 1 gm q6h then Clindamycin 300 mg q6h C1indamycin 600 mg q6h Gentamicin 80 mg q8h Cefazolin sodium I gm q6h Cefalothin I gm q6h Methicillin 2 gm q6h Clindamycin 600 mg q8h Aqueous penicillin 1,000,000 Uq4h Methicillin I gm q8h
PO
Result
Healed per pnrnum: reexposure II mo later Healed per primum Healed per primum: reexposure of electrode in 9 rna Healed per primurn; reexposure of pacemaker 12 ma later Healed per primum
Healed per primum: reexposure II rna later
7
14
Staph. aureus
9
10
Staph. epidermidis
Enterobacter sp.
Methicillin 1 gm q6h Gentamicin 80 mg q8h then Oxacillin 500 mg q6h Clindamycin 600 mg q6h Cefazolin sodium I gm q6h Cephradine 250 mg 'lid C1indamycin 300 mg 'lid Gentamicin 80 mgq8h Cefoxitin sodium 2 gm q6h Cefadroxil 500 mg bid Gentamicin 80 mg 'ISh
IV
7
10
Healed per primum
PO
IV
14 3
5
Healed per prirnurn
PO
3
IV
7
12
Healed per prirnum
PO 1M
6
Ltgend: IV, Intravenous. PO, By mouth. 1M. Intramuscular. q8h (q6h), Every 8 hours (6 hours). 'lid, Four times a day.
procedures in which the pulse generator was relocated to subfascial and subpectoral locations by success or failure. Using the chi square test, one can conclude that pulse generators relocated to a subfasciallocation have a significantly higher success rate than those placed in a subpectoral location (x 2 (1) = 6.58; P 0.02).
Bacteriologic studies (Table III) We were unable to culture bacteria or other organisms in only three (Patients 3, 6, and 7) of the 10
patients in the study. However, all were treated empirically with intravenous antibiotics. The antibiotics used were ampicillin for 7 days in Patient 3, cephalothin for 9 days in Patient 6, and methicillin and clindamycin for 7 days in Patient 7. The pulse generator was transferred to a subpectoral, subclavicular location in one and a subfascial position in two. All three patients recovered. Staphylococcus sp. grew in the first culture in five patients (Patients 2, 4, 5, 8, and 9): S. epidermidis in three and S. aureus in two. Those whose cultures grew
The Journal of Thoracic and Cardiovascular Surgery
1 40 Garcia-Rinaldi et al.
S. epidermidis were treated with chloromycetin, gentamicin, and cefazolin for 10 days each. One patient was treated with clindamycin and cefazolin for 3 days. One patient (Patient 2) treated by subpectoral, submammary relocation developed pacer reexposure. Culture at this time grew S. mirabilis. Relocation in another subpectoral, submammary location resulted in pacer reexposure. Culture of the pulse generator pocket grew S. epidermidis again, and the patient was treated with clindamycin and gentamicin, combined with subfascial relocation of the pulse generator. This relocation resolved the problem up to the current 23 months' follow-up. The other two patients whose cultures grew S. epidermidis were treated by subfascial relocation of the pacer. Both pulse generators are still in situ without problems. Of the two patients with S. aureus infection (Patients 5 and 8), one (Patient 5) was treated with clindamycin, gentamicin, and cefazolin for 10 days. Relocation to a subfascial, intrarectus sheath pocket resolved the problem, and the patient is now alive 40 months after the operation. The other patient (Patient 8) was treated by relocation of the pulse generator to a subpectoral, submammary pocket from a subcutaneous pocket. Pacemaker exposure occurred at 3 months postoperatively. S. aureus was cultured again. The patient was treated with methicillin, gentamicin, and oxacillin for 2 weeks. The pacer was relocated to a subfascial location with resolution of the problem. He is now alive 26 months after the operation. In Patient 1, an anaerobic Propionibacterium grew at the time of exposure. The patient was treated with chloromycetin and cefamandole for 10 days. The pulse generator was relocated to a subpectoral, submammary location. Generator reexposure occurred at 11 months and another anaerobe, Clostridium perfringens, was cultured. The patient was treated with clindamycin and cefazolin for 7 days and the pocket was relocated to a subxiphoid, subfascial position. The patient recovered without incident and is now alive 25 months after the operation. Enterobacter sp. grew at the time of exposure in Patient 10. The patient was treated with intravenous gentamicin and cefoxitin for 7 days and intramuscular gentamicin and oral cefadroxil for an additional 6 days. The pulse generator was relocated to a subxiphoid, subfascial location. The relocation site healed and the patient is now alive 3 months after the operation. Thus in this small series the organism, aerobe, or anaerobe, the antibiotic, and the length of antibiotic administration were not determining factors in the eventual outcome of the patients. The common denominator for eventual recovery was the relocation (primary
or secondary) of the pulse generator unit to a subfascial location.
Discussion Since the original description of the implantation of pulse generators in 1961, most surgeons and cardiologists have successfully located the generator in the subcutaneous position.' Almost invariably the unit becomes more prominent. The skin and subcutaneous tissue overlying the unit "thin out," probably as a result of pressure necrosis.!2,13 In a small percentage of patients, the skin breaks and the pulse generator becomes exposed. Surgeons have resorted to various techniques to reduce the incidence of these problems. These have consisted of location of the generator intrapleurally," subcostally," or subdiaphragmatical-
ly." When skin breakage occurs there is usually secondary bacterial invasion of this traumatized tissue. The long interval observed between implantation and exposure of the pulse generator is very strong evidence against contamination at the time of initial implantation. Pacemakers that become infected during implantation usually develop abscesses that necessitate immediate removal. Our series does not propose to deal with infected pulse generators in the absence of long-term exposure. Bacterial infection of a foreign body is usually considered an unresolvable problem if one uses conservative means. The experience with porous arterial grafts is dismal. Invariably the body "rejects" the foreign material. Our experience with exposed pulse generators is different, for reasons not clearly understood. In this series, all wounds healed primarily. All patients were treated with intravenous antibiotics during the preoperative, intraoperative, and postoperative periods to cover specific organisms, for 5 to 10 days. Pulse generator reexposure occurred only in those patients in whom the generator was relocated to the thin subpectoral, submammary pocket. Reexposure in the subpectoral, submammary pockets occurred late (average 10 months), again probably because of pressure necrosis and not infection. Consequently, it would be difficult to ascribe reexposure to bacterial contamination at the time of first relocation. Furthermore, in most cases the organism cultured the second time was different. In all cases in which the pulse generator was relocated to a subfasciallocation and in one case of intrapectoral, subclavicular location, healing occurred. Relocation of the pulse generator to a subfascial pocket is definitely much simpler and less expensive than reimplantation of a new pacing system. The recovery is usually uncomplicated and the hospital stay usually short. Although our
Volume 89 Number 1
Exposed cardiac pacemaker
14 1
January. 1985
series is small, the excellent results should be taken into account when dealing with the exposed pulse generator. Of course, a substantially larger number of cases will be required to corroborate the significance of this observation. However, relocation should be considered as an alternative to the replacement of an exposed but wellfunctioning generator unit. REFERENCES Chardack WM, Gage AA, Greatbatch WA: Correction of completeheart block by self-contained and subcutaneously implanted pulse generator. J THoRAc CARDIOVASC SURG 42:814-830,1961 2 Lawrence GE, Paine RM, Hughes ML: Management of complications associated with the use of implantable electronic cardiac pulse generators for the relief of complete heart block. Am J Surg 110:177-185, 1965 3 Martinis AJ: Incidence and management of problems associated with permanent pulse generator implants. Am Surg 38:612-616, 1972 4 Harstein AI, Jackson J, Gilbert DN: Prophylactic antibioticsand the insertion of transvenous cardiac pulse generators. J THORAC CARDIOVASC SURG 75:219-222, 1978 5 Chait LA, Ritchie B: A method of treating the exposed cardiac pulse generator. Br J Plast Surg 32:281-284, 1979
6 Silverton NP: Prophylaxis of surgical wound sepsis. Cardiac pulse generators. Br Med J 280:1321, 1980 7 Sowton E, Hendrix G, Roy P: Ten year survey in treatment with implanted cardiopacemakers. Br Med J 3:155-160,1974 8 Chari PS, Sury RK: Skin cover for exposed cardiac pulse generators. Br J Plast Surg 30:231-234, 1977 9 Coburn RJ, Blank HI, Campbell PM: Covering an exposed pulse generator. Plast Reconstr Surg 50:622-625, 1972 10 Mittapalli MR: Erosion of pulse generator lead. South Med J 73:528-529, 1980 II Furman S: Reuse of implanted cardiac pulse generators. PACE 3:265-266, 1979 12 Kim GE, Haveson S, Imparato A: Late displacement of cardiac pulse generator electrode due to heavy pulse generator. JAMA 228:74-75, 1974 13 Furman S: Subcutaneous pulse generator placement. JAMA 229:1045-1046, 1974 14 Escano FB, Berroya RB, Gianfrancesco H: Intrapleural pulse generator in children. J THORAC CARDIOVASC SURG 62:454-456, 1971 15 Gibbons JA, Denig PM, Aaron BL: The recessed chest wall pulse generator pocket. PACE 1:55-57, 1979 16 Lindesmith GG, Stiles QR: Experience with implantable synchronous pulse generators in children. Ann Thorac Surg 6:358, 1968
THORAC CARDIOVASC SURG
89:136-141, 1985
The exposed cardiac pacemaker Treatment by subfascial pocket relocation Ten patients were treated for skin erosion and exposure of chronically implanted pacemakers. The pulse generators bad been implanted either subcutaneously or in the subpectoral, submammary location. All patients were treated either by primary or secondary relocation of the pulse generator to deeper subfascial planes. Pacemakers relocated primarily to subfacial planes healed without resurfacing. Those pacemakers relocated to other superficial planes resurfaced but healed in a second subfascial t:ramlocation. Bacteriologic studies on these patients revealed that the orgllllisIm cultured or antibiotics utilized did not influence the fate of the relocated pulse generator. The subfasciallocation was the only determinant of the ultimate healing of the exposed pacemaker. We conclude that subfascial relocation of weU-fwtctioning pacemaker generators should be COIL'iidered as an alternative to complete replacement of the unit
Raul Garcia-Rinaldi, M.D., Ph.D., F.A.C.S., Jose M. Revuelta, M.D., Ph.D., Lucy Bonnington, PA.-C., and Luis Soltero-Harrington, M.D.,
Houston, Texas, and Cayey, Puerto Rico Consulting Statistician: Janine Pyrek, M.S., * St. Paul, Minn.
In 1961, Chardack, Gage, and Greatbatch' implanted the first cardiac pacemaker in the United States for treatment of patients with. life-threatening conduction abnormalities, particularly complete heart block. Currently, as in the past, many surgeons implant the electrode transvenously or epicardially by the subxiphoid, subcostal, or transthoracic approach. The generator is usually placed subcutaneously and occasionally subfascially, particularly if an epicardial pacemaker is implanted at the time of an open cardiac procedure. Regardless of route or technique of implantation, longterm results are excellent. Despite improvements in hardware and techniques, there remain complications associated with pacemaker insertion, such as lead malfunction or displacement. 2,3 Infection of the pulse generator pocket with subsequent
From The Houston Heart Institute and The Memorial Hospital Houston, Texas, and Department of Surgery, Universidad Central del Caribe, Cayey, Puerto Rico. Received for publication Aug. 31, 1983. Accepted for publication March I, 1984. Address for reprints: Raul Garcia-Rinaldi, M.D., Cardiovascular Surgery of Texas, Suite 110, 7500 Beechnut, Houston, Texas 77074. *Supervisor, Clinical Research, Surgical Products Division, 3M, St. Paul, Minn.
136
exposure of the generator are still problems associated with a high morbidity.' Grossly infected pacemaker systems with acute abscess formation require immediate removal and insertion of a new system. The incidence of chronic skin erosion by the generator with low-grade infection varies in different series between 0% and 12.6%. A number of methods have been utilized to combat this problem, with little success. Sowton, Hendrix, and Roy? treated patients with systemic and topical antibiotics with only a 5% cure rate. Chait and Ritchie' reported on a series of five patients treated by debridement, tissue advancement, and flap rotation; administration of systemic antibiotics; and removal of the pulse generator units. After infection was controlled, the units were then sterilized and reimplanted. Although this technique was innovative and successful, two patients had reexposure on three occasions. Chari and Suri" used similar methods in three patients and were successful in two patients. They concluded that the method was cumbersome and "logically the pacemaker should be removed and a new one implanted." Other authors have tried other variations, all of which have resulted in failure." 10 Most authors have arrived at the same conclusion: that the generator should be replaced. Although this premise is acceptable, it must be remembered that cardiac pacemakers are expensive,
Volume 89 Number 1
Exposed cardiac pacemaker 1 3 7
January, 1985
Table I. Results of cases of exposed pacemakers Case No. (age, sex) I (46, M)
2
Interval between first implantation and exposure
Original location
II mo 5 mo
Subcutaneous, submammary Subpectoral, submammary Subpectoral, submammary
9 mo
Subpectoral, submammary
12 mo 36 mo
Subpectoral, submammary Subcutaneous, subclavicular Subcutaneous abdominal wall Subcutaneous abdominal wall Subcutaneous abdominal wall Subpectoral, submammary
60 mo
(34, F)
3 (74, F) 4 (71, F) 5
36 mo
(41, F) 6
5 mo
(69, M) 7 (70, M) 8 (44, M)
9 (36, M) 10 (84, M)
48 mo
12 mo 120 mo 3 mo 37 mo
5 mo
Subcutaneous, submammary Subpectoral, submammary Subpectoral, submammary Subcutaneous, subclavicular
particularly the current multiprogrammable units with long battery life. Consequently, it would be highly desirable to develop a method that could salvage an exposed well-functioning unit.'! Our series consists of 10 patients with pulse generator exposure referred to us for treatment. These 10 patients are the only ones with this problem whom we have treated. Five other patients whose generators were translocated to subfascial planes before exposure occurred are not included. All the pulse generators in the series were implanted in subcutaneous pockets or under the thin pectoral muscle in the submammary region. From February, 1979, to the present, patients referred to us with exposed pulse generator units have been treated by relocation of the exposed pulse generator to deeper subpectoral or subfascial locations, and this report summarizes our experience. Of four pulse generators relocated to a subpectoral site, three (75%) became reexposed. All pulse generators relocated primarily (six) or secondarily (three) to a subfascial plane have remained in situ without reexposure, irrespective of the organism cultured. Follow-up time from the last relocation to the present, or time of death, averaged 24.6 months (3 to 44 months).
Final location
Result
Subpectoral, submammary
Reexposure II mo later
Subxiphoid, subfascial Subpectoral relocation of electrode Subcutaneous abdominal wall relocation of electrode Subxiphoid, subfascial Subpectoral, subclavicular
Recovered 25 mo postop. Reexposure of plate in 9 mo
Inside rectus sheath
Exposure of pacemaker 12 mo later Recovered 23 mo postop. Died 20 mo later with healed pacemaker Recovered 44 mo postop.
Inside rectus sheath
Recovered 40 mo postop.
Subxiphoid, subfascial
Recovered 23 mo postop.
Subxiphoid, subfascial
Recovered 35 mo postop
Subpectoral, submammary
Reexposure in 3 mo
Subxiphoid, subfascial Subxiphoid, subpectoral
Recovered 26 mo postop Recovered 7 mo postop
Subxiphoid, subfascial
Recovered 3 mo postop.
Patients and methods
The study consisted of 10 patients, six men and four women, seen from February, 1979, to July 1, 1983, and referred to us by other surgeons or cardiologists. Over this period we inserted approximately 350 transvenous or epicardial pacemakers in subcutaneous, subpectoral, or subfascial pockets without encountering the problem of skin erosion. Thus, in our experience, pulse generator exposure is uncommon. However, when it occurs, it can be a very challenging problem. All patients were hospitalized and the pulse generator pockets cultured. Empirical intravenous antibiotic coverage was started, depending on the generation of antibiotics in use at the time. After culture and sensitivities were obtained, the specific antibiotic was administered for at least 3 days. The patient was treated by pacemaker relocation, and intravenous antibiotics were continued for at least another 4 days. Surgical techniques
The infected pocket is opened wide and the generator unit freed from all attachments. The electrode is released from surrounding granulation tissue and from the fibrous tract so that it is freely movable. The pulse
138
The Journal of Thoracic and Cardiovascular
Garcia-Rinaldi et al.
Surgery
Table II. Frequency of success or failure in relocation of pacemakers by site of relocation" Beginning Site:
Site Subpectoral Subfascial Total
Success I
Failure 3
Total 4
9
o
9
10
3
13
'This table does not include the intermediate relocation to the subcutaneous tissue (failure) of Patient 2.
Final Site: (Success)
Fig. 1. Flow chart of site of relocation. This chart does not include the intermediate relocation to subcutaneous tissue (failure) of Patient 2. Dashed lines. Intermediate relocation (failure). Solid lines. Final relocation.
generator and the electrode are tested and thoroughly irrigated with saline solution and immersed in antibiotic or povidone-iodine* solution for 5 minutes. The generator is then relocated to a' subfascial or intrafascial location. We prefer the subfascial, subxiphoid area where it lies in the pre-peritoneal space. A vertical incision is made through the linea alba in the the subxiphoid area. A subfascial pocket is made in either side and thus the pulse generator is extraperitoneal. The electrode is tunneled to the new pocket and the pulse generator unit is reconnected immediately. The pocket is closed, without drainage, with absorbable polyglycolic acid sutures. The skin is closed with a subcuticular suture. Once the new pocket site is closed, the fibrous shell of the previous pulse generator is removed. A small drain is placed before closure to avoid the formation of a serous collection. The wound is closed subcuticularly with polyglycolic acid sutures. The drain is removed in 2 days. Results In seven patients the original pulse generators were in a subcutaneous position and were extruded at an average of 44 months (5 to 120 months). In two patients (Patients 1 and 8, Table I) the pulse generators were translocated from a subcutaneous, submammary pocket *Betadine, The Purdue Frederick Company, Norwalk, Conn.
to a subpectoral location. In both patients a second extrusion of the pulse generator necessitated a second translocation to a subfascial plane. Both recovered and are alive 25 and 26 months after operation, respectively. Two patients with subcutaneous, subclavicular pulse generators had generator translocation. One patient's generator (Patient 3) was transferred to a subpectoral, subclavicular position. This patient recovered but died of cardiac causes 20 months after operation. The other patient's pulse generator (Patient 10) was transferred to another subxiphoid, subfascial location. The patient recovered and is alive 3 months after the operation. Three patients (Patients 4, 5, and 6) had subcutaneous placement of their original pulse generator, but in the abdominal wall. The pulse generators were all relocated to subfascial areas under or inside the rectus sheath. All patients recovered and have been followed up for 44, 40, and 23 months, respectively. In three patients the original pulse generators were implanted in a subpectoral, submammary location. The pulse generator became exposed at 5, 12, and 37 months, respectively (average 18 months). One patient's indifferent electrode (Patient 2) was relocated in another subpectoral, submammary pocket and it resurfaced in 9 months. When the electrode finally was translocated to a subfasciallocation, the patient recovered and is alive 23 months postoperatively. In the other two patients (Patients 7 and 9) the pulse generator was transferred to a subfascial plane. Both have recovered and are alive 35 and 7 months after operation, respectively. A summary flow chart of location of pulse generators is shown in Fig. 1. Of the six pulse generators whose first relocation was to a subfascial area, all were successfully relocated with no resurfacing. Of the four whose first relocation was to a subpectoral area, only one was successfully relocated. The three that resurfaced were then transferred to a subfascial area. These three have not resurfaced. The hypothesis that successful relocation of pulse generators is independent of site of implantation was tested statistically. Table II shows the frequency of
Volume 89
Exposed cardiac pacemaker 1 3 9
Number 1 January, 1985
Table Case
no.
2
m. Pathogens and antibiotic coverage in cases of exposed pacemakers Antibiotics used
Bacterial culture
Anaerobe Propionibacterium Anaerobe Clostridium perfringens Staph. epidermidis Proteus morgagni Staph. mirabilis
Staph. epidermidis
No growth 4
Staph. epidermidls
5
Staph. aureus
6 7
No growth No growth Staph. aureus
Route
Length of treatment (days)
Length of hasp. stay (days)
IV
10
12
IV
7
8
IV
10
10
IV
7
7
IV
6 6
10 10
IV
I
3
Healed per primum
PO
6 3
II
Healed per primum
IV
5 10
14
Healed per primum
IV IV
9 7
10 10
Healed per primum Healed per primum
IV
2
Chloramphenicol I gm q8h Cefamandole I gm q6h C1indamycin 600 mg q6h Cefazolin sodium I gm q6h Gentamicin 80 mg q8h Chloramphenicol I gm q8h Cefazolin sodium I gm q8h C1indamycin 600 mg q6h Tobramycin 80 mg q8h Cefazolin sodium I gm q6h Gentamicin 80 mg q8h Clindamycin 600 mg q6h Ampicillin 1 gm q6h then 500 mg 'lid Chloramphenicol I gm q8h Tobramycin 80 mg q8h Cefazolin sodium 1 gm q6h then Clindamycin 300 mg q6h C1indamycin 600 mg q6h Gentamicin 80 mg q8h Cefazolin sodium I gm q6h Cefalothin I gm q6h Methicillin 2 gm q6h Clindamycin 600 mg q8h Aqueous penicillin 1,000,000 Uq4h Methicillin I gm q8h
PO
Result
Healed per pnrnum: reexposure II mo later Healed per primum Healed per primum: reexposure of electrode in 9 rna Healed per primurn; reexposure of pacemaker 12 ma later Healed per primum
Healed per primum: reexposure II rna later
7
14
Staph. aureus
9
10
Staph. epidermidis
Enterobacter sp.
Methicillin 1 gm q6h Gentamicin 80 mg q8h then Oxacillin 500 mg q6h Clindamycin 600 mg q6h Cefazolin sodium I gm q6h Cephradine 250 mg 'lid C1indamycin 300 mg 'lid Gentamicin 80 mgq8h Cefoxitin sodium 2 gm q6h Cefadroxil 500 mg bid Gentamicin 80 mg 'ISh
IV
7
10
Healed per primum
PO
IV
14 3
5
Healed per prirnurn
PO
3
IV
7
12
Healed per prirnum
PO 1M
6
Ltgend: IV, Intravenous. PO, By mouth. 1M. Intramuscular. q8h (q6h), Every 8 hours (6 hours). 'lid, Four times a day.
procedures in which the pulse generator was relocated to subfascial and subpectoral locations by success or failure. Using the chi square test, one can conclude that pulse generators relocated to a subfasciallocation have a significantly higher success rate than those placed in a subpectoral location (x 2 (1) = 6.58; P 0.02).
Bacteriologic studies (Table III) We were unable to culture bacteria or other organisms in only three (Patients 3, 6, and 7) of the 10
patients in the study. However, all were treated empirically with intravenous antibiotics. The antibiotics used were ampicillin for 7 days in Patient 3, cephalothin for 9 days in Patient 6, and methicillin and clindamycin for 7 days in Patient 7. The pulse generator was transferred to a subpectoral, subclavicular location in one and a subfascial position in two. All three patients recovered. Staphylococcus sp. grew in the first culture in five patients (Patients 2, 4, 5, 8, and 9): S. epidermidis in three and S. aureus in two. Those whose cultures grew
The Journal of Thoracic and Cardiovascular Surgery
1 40 Garcia-Rinaldi et al.
S. epidermidis were treated with chloromycetin, gentamicin, and cefazolin for 10 days each. One patient was treated with clindamycin and cefazolin for 3 days. One patient (Patient 2) treated by subpectoral, submammary relocation developed pacer reexposure. Culture at this time grew S. mirabilis. Relocation in another subpectoral, submammary location resulted in pacer reexposure. Culture of the pulse generator pocket grew S. epidermidis again, and the patient was treated with clindamycin and gentamicin, combined with subfascial relocation of the pulse generator. This relocation resolved the problem up to the current 23 months' follow-up. The other two patients whose cultures grew S. epidermidis were treated by subfascial relocation of the pacer. Both pulse generators are still in situ without problems. Of the two patients with S. aureus infection (Patients 5 and 8), one (Patient 5) was treated with clindamycin, gentamicin, and cefazolin for 10 days. Relocation to a subfascial, intrarectus sheath pocket resolved the problem, and the patient is now alive 40 months after the operation. The other patient (Patient 8) was treated by relocation of the pulse generator to a subpectoral, submammary pocket from a subcutaneous pocket. Pacemaker exposure occurred at 3 months postoperatively. S. aureus was cultured again. The patient was treated with methicillin, gentamicin, and oxacillin for 2 weeks. The pacer was relocated to a subfascial location with resolution of the problem. He is now alive 26 months after the operation. In Patient 1, an anaerobic Propionibacterium grew at the time of exposure. The patient was treated with chloromycetin and cefamandole for 10 days. The pulse generator was relocated to a subpectoral, submammary location. Generator reexposure occurred at 11 months and another anaerobe, Clostridium perfringens, was cultured. The patient was treated with clindamycin and cefazolin for 7 days and the pocket was relocated to a subxiphoid, subfascial position. The patient recovered without incident and is now alive 25 months after the operation. Enterobacter sp. grew at the time of exposure in Patient 10. The patient was treated with intravenous gentamicin and cefoxitin for 7 days and intramuscular gentamicin and oral cefadroxil for an additional 6 days. The pulse generator was relocated to a subxiphoid, subfascial location. The relocation site healed and the patient is now alive 3 months after the operation. Thus in this small series the organism, aerobe, or anaerobe, the antibiotic, and the length of antibiotic administration were not determining factors in the eventual outcome of the patients. The common denominator for eventual recovery was the relocation (primary
or secondary) of the pulse generator unit to a subfascial location.
Discussion Since the original description of the implantation of pulse generators in 1961, most surgeons and cardiologists have successfully located the generator in the subcutaneous position.' Almost invariably the unit becomes more prominent. The skin and subcutaneous tissue overlying the unit "thin out," probably as a result of pressure necrosis.!2,13 In a small percentage of patients, the skin breaks and the pulse generator becomes exposed. Surgeons have resorted to various techniques to reduce the incidence of these problems. These have consisted of location of the generator intrapleurally," subcostally," or subdiaphragmatical-
ly." When skin breakage occurs there is usually secondary bacterial invasion of this traumatized tissue. The long interval observed between implantation and exposure of the pulse generator is very strong evidence against contamination at the time of initial implantation. Pacemakers that become infected during implantation usually develop abscesses that necessitate immediate removal. Our series does not propose to deal with infected pulse generators in the absence of long-term exposure. Bacterial infection of a foreign body is usually considered an unresolvable problem if one uses conservative means. The experience with porous arterial grafts is dismal. Invariably the body "rejects" the foreign material. Our experience with exposed pulse generators is different, for reasons not clearly understood. In this series, all wounds healed primarily. All patients were treated with intravenous antibiotics during the preoperative, intraoperative, and postoperative periods to cover specific organisms, for 5 to 10 days. Pulse generator reexposure occurred only in those patients in whom the generator was relocated to the thin subpectoral, submammary pocket. Reexposure in the subpectoral, submammary pockets occurred late (average 10 months), again probably because of pressure necrosis and not infection. Consequently, it would be difficult to ascribe reexposure to bacterial contamination at the time of first relocation. Furthermore, in most cases the organism cultured the second time was different. In all cases in which the pulse generator was relocated to a subfasciallocation and in one case of intrapectoral, subclavicular location, healing occurred. Relocation of the pulse generator to a subfascial pocket is definitely much simpler and less expensive than reimplantation of a new pacing system. The recovery is usually uncomplicated and the hospital stay usually short. Although our
Volume 89 Number 1
Exposed cardiac pacemaker
14 1
January. 1985
series is small, the excellent results should be taken into account when dealing with the exposed pulse generator. Of course, a substantially larger number of cases will be required to corroborate the significance of this observation. However, relocation should be considered as an alternative to the replacement of an exposed but wellfunctioning generator unit. REFERENCES Chardack WM, Gage AA, Greatbatch WA: Correction of completeheart block by self-contained and subcutaneously implanted pulse generator. J THoRAc CARDIOVASC SURG 42:814-830,1961 2 Lawrence GE, Paine RM, Hughes ML: Management of complications associated with the use of implantable electronic cardiac pulse generators for the relief of complete heart block. Am J Surg 110:177-185, 1965 3 Martinis AJ: Incidence and management of problems associated with permanent pulse generator implants. Am Surg 38:612-616, 1972 4 Harstein AI, Jackson J, Gilbert DN: Prophylactic antibioticsand the insertion of transvenous cardiac pulse generators. J THORAC CARDIOVASC SURG 75:219-222, 1978 5 Chait LA, Ritchie B: A method of treating the exposed cardiac pulse generator. Br J Plast Surg 32:281-284, 1979
6 Silverton NP: Prophylaxis of surgical wound sepsis. Cardiac pulse generators. Br Med J 280:1321, 1980 7 Sowton E, Hendrix G, Roy P: Ten year survey in treatment with implanted cardiopacemakers. Br Med J 3:155-160,1974 8 Chari PS, Sury RK: Skin cover for exposed cardiac pulse generators. Br J Plast Surg 30:231-234, 1977 9 Coburn RJ, Blank HI, Campbell PM: Covering an exposed pulse generator. Plast Reconstr Surg 50:622-625, 1972 10 Mittapalli MR: Erosion of pulse generator lead. South Med J 73:528-529, 1980 II Furman S: Reuse of implanted cardiac pulse generators. PACE 3:265-266, 1979 12 Kim GE, Haveson S, Imparato A: Late displacement of cardiac pulse generator electrode due to heavy pulse generator. JAMA 228:74-75, 1974 13 Furman S: Subcutaneous pulse generator placement. JAMA 229:1045-1046, 1974 14 Escano FB, Berroya RB, Gianfrancesco H: Intrapleural pulse generator in children. J THORAC CARDIOVASC SURG 62:454-456, 1971 15 Gibbons JA, Denig PM, Aaron BL: The recessed chest wall pulse generator pocket. PACE 1:55-57, 1979 16 Lindesmith GG, Stiles QR: Experience with implantable synchronous pulse generators in children. Ann Thorac Surg 6:358, 1968