- Email: [email protected]
Endocarditis related to transvenous pacemakers
Endocarditis related to transvenous pacemakers Syndromes and surgical implications Two cases of microbial endocarditis related to transvenous pacemakers illustrate syndromes whose pathogenesis we consider to be distinctive. Acute Pseudomonas aeruginosa endocarditis related to a pacemaker developed in a 75-year-old man, an event which to our knowledge has not been previously described. There was no evidence of generator site infection, and the sequence of events indicated metastatic implantation of bacteria on traumatized endothelium. A 76-year-old woman with a 3 year history of local generator site infection and recurrent fever was found to have extensive vegetative Staphylococcus epidermidis endocarditis at cardiotomy. The sequence of events indicated gradual spread of infection locally, related to the contaminated foreign body. Awareness of these separate pathogenetic mechanisms should facilitate recognition and appropriate management. Removal of the entire pacing system and prolonged antibiotic therapy were considered to be essential to cure of infection in both instances.
Charles S. Bryan, M.D., F.A.C.P.,* John P. Sutton, M.D., F.A.C.S.,** Donald E. Saunders, Jr., M.D., F.A.C.P.,*** Dail W. Longaker, M.D., F.A.C.S., and Claude W. Smith, M.D., Columbia, S. C.
Microbial endocarditis related to transvenous pacemaker implantation is considered to be rare, the literature consisting mainly of single case reports. Management remains unstandardized, especially with regard to the need to remove the pacemaker by cardiotomy when attempts to withdraw the electrode by simple traction fail. Zeller and colleagues! suggested a trial of intensive antibiotic therapy in this setting. On the other hand, Chavez and Conn" outlined the need to remove the infected wire even should cardiotomy be required. Recently, we have treated two elderly patients with pacemaker-associated endocarditis which began 5 and 7 years after the original implantations. The strikingly different presentations suggested to us that a clear distinction, on the basis of pathogenesis, should be made From the University of South Carolina School of Medicine, Columbia, S. C. Received for publication Oct. 14, 1977. Accepted for publication Dec. 20, 1977. Address for reprints: Dr. C. S. Bryan, 3301 Harden Street, ACC II, Columbia, S. C. 29203. *Associate Professor, Department of Medicine. **Clinical Associate Professor, Department of Surgery. ***Clinical Associate Professor, Department of Medicine. 758
between two syndromes of pacemaker-associated endocarditis. Appreciation of these syndromes should facilitate early recognition and appropriate management of such problems.
Case reports CASE I. N. W., a 75-year-old man, was admitted to the hospital on Nov. 25, 1975, with fever and chills of 2 days' duration. A permanent transvenous pacemaker had been inserted in December, 1970, for Stokes-Adams attacks. The pulse generator had last been changed, uneventfully, in December, 1974. His general health had been good, and there was no history of chronic lung disease, diabetes mellitus, heart failure, or recent antibiotic therapy. A faint systolic murmur at the left sternal border had been described on previous physical examinations. Although he appeared acutely ill on admission to the hospital, physical examination disclosed no source for the fever. The previously described murmur was not heard. The initial chest x-ray film revealed mild congestive changes with some haziness along the right border of the heart but no definite infiltrates. Within several days, the following findings developed: substernal pain and pleuritic pain on the right side of the chest; a Grade 2/6, nearly pansystolic murmur along the lower left sternal border; a I em. black nodular lesion on the lower lip; and a cluster of petechiae about the right ankle. The chest x-ray film (Fig. I, A) revealed extensive, bilateral nodular pulmonary infiltrates. Pseudomonas aeruginosa, Fisher-
0022-5223/78/0575-0758$00.50/0 © 1978 The C. V. Mosby Co.
Volume 75 Number 5 May, 1978
Endocarditis related to transvenous pacemakers
759
A Fig. 1. A , Extensive bilateral pulmonary infiltrates early in the course of bacteremic Pseudomonas pneumonia. B, Resolution of infiltrate s after prolonged antibiotic therapy, removal of the transvenous pacing system, and implant ation of epicardial pacing electrodes. Devlin-Gnabasik immunotype 3,* was isolated from three of six blood cultures and from two sputum cultures. Urine cultures were sterile. Therapy with gentamicin sulfate and disodium carbenicillin w~s begun on the third hospital day . The microorganism was shown to be susceptible to tobramycin sulfate at a minimum inhibitory concentration (MIC) of 1.56 meg. per milliliter, to gentamicin sulfate at 6.13 meg. per milliliter, and to carbenicillin at 31 meg. per milliliter. Synergism between tobramycin and carbenicillin was demonstrated by a twodimensional agar dilution checkerboard study ." A combination of tobramycin and carbenicillin was therefore used for subsequent therapy. Because bacteremic Pseudomonas aeruginosa pneumonia has been an almost uniformly fatal disease despite maximum antibiotic therapy (see Discussion), and because no other factors could be identified which might have predisposed the patient to this illness, it was elected to remove the transvenous pacemaker even though there was no direct evidence of infection of the generator pocket or electrode . This was accomplished by traction on the ninth hospital day. The generator pocket appeared to be unremarkable . However, Pseudomonas aeruginosa was isolated from the electrode tip, at a time when cultures of the peripheral blood were becoming sterile. *Typing done by George C. Cole, Ph.D ., Parke, Davis & Company, Detroit, Mich.
Fever and pneumonia resolved with continued therapy (Fig. I, B). Third-degree heart block and a brief period of asystole, occurring on the twenty-eighth hospital day, prompted implantation of a permanent unipolar demand pacemaker with screw-in epicardial electrodes by means of a left thoracotomy . Rectal bleeding occurred and was thought possibly to be due to hemorrhoidal bleeding aggravated by the hemostatic defect known to result from high-dose carbenicillin therapy. Barium enema revealed only diverticulosis, and sigmoidoscopic examination showed no abnormalities. The patient was discharged after 42 days of combined aminoglycoside-carbenicillin therapy . The systolic murmur became longer, sometimes with definite inspiratory augmentation . He remained afebrile after discharge but was mildly anemic and failed to regain his original weight. On March 3, 1976, he was admitted to the hospital again because of the gradual appearance of warmth and erythema near the sixth and seventh costochondral junctions on the right and because of pain over the right ischium. Excisional biopsy of the inflamed area over the chest disclosed a nonspecific cellulitis, from which Pseudomonas aeruginosa was isolated . A biopsy specimen from the right ischium was sterile. Pseudomonas osteomyelitis was thought to be present at both sites , and therapy with tobramycin and carbenicillin was again instituted . Rectal bleeding again occurred during antibiotic therapy. Barium enema and sigmoidoscopic study were essentially unrevealing. On examination of the colon with a fiberoptic col -
The Journal of
760
Bryan et al.
Thoracic and Cardiovascular Surgery
Table I. Syndromes of transvenous pacemaker-related endocarditis
Relative frequency (fromliterature) Source of bacteria Pathogenesis Microorganisms Experimental model
"Metastatic implantation" type (Case /)
"Foreign body" type (Case 2)
Less common Remote site Implantation of microorganisms, in the course of bacteremia, on traumatized endothelium Diverse Abrasion of endothelium followed by induced bacteremia (so-called "rabbit model endocarditis")
More common Pacemaker generator site Direct extension of infection alongpacemaker wire, with or without systemic bacteremia Usually staphylococci Contamination of a foreign body, resulting in local spread of infection
onoscope, an adenocarcinoma was found at the splenic flexure. One of six mesenteric lymph nodes was abnormal at the time of resection. He was discharged from the hospital after 53 days of combined tobramycin-carbenicillin therapy. In 15 months of follow-up, he has enjoyed good health without recurrence of Stokes-Adams attacks, endocarditis, osteomyelitis, other evidence of Pseudomonas infection, or carcinoma. CASE 2. E. F., a 76-year-old woman, was admitted to the hospital on Aug. 13, 1976, for removal of an infected pacemaker wire. A permanent transvenous pacemaker had been implanted for complete heart block of undetermined origin in 1966. In 1973, infection of the generator pocket due to coagulasenegative staphylococci was recognized. The generator was removed, but it was not possible to remove the electrode by traction. It was therefore allowed to remain in place, a second transvenous pacemaker system being implanted with the pulse generator on the opposite side. Between 1973 and 1976, she had an intermittently draining sinus at the site of the original pulse generator. Cultures of this sinus intermittently revealed coagulase-negative staphylococci. Blood cultures were sterile. Intermittent fever and chills responded to courses of oral antibiotic therapy. Several further attempts were made to withdraw the original electrode by traction, but it continued to be densely adherent to the right ventricular endocardium. On Feb. 2, 1976, a final attempt was made to remove the original fractured wire by traction. Again, the electrode appeared to be firmly anchored to the endocardium. The proximal portion was severed. Coagulase-negative staphylococci were isolated from this segment of wire, but blood cultures remained sterile. It was elected to institute intensive parenteral antibiotic therapy. However, a paucity of peripheral veins and the occurrence of a generalized rash while the woman was receiving nafcillin sodium led to termination of this therapy. She completed a 10 week course of oral cephalexin combined with probenecid. Blood levels were monitored during this therapy and appeared to be adequate. On June 17, 1976, recurrent chills prompted additional blood cultures; two of six cultures revealed a coagulasenegative staphylococcus, subsequently confirmed to be Staphylococcus epidermidis, susceptible to cephalothin at an MIC of 0.2 mcg. per milliliter and to cephalexin at 0.8 mcg. per milliliter. Fever and other symptoms persisted during oral antibiotic therapy. It was therefore elected to remove the retained electrode by cardiopulmonary bypass surgery. On Aug. 17, 1976, she underwent median stemotomy and
cardiopulmonary bypass. The superior vena cava was almost totally occluded. Vegetations extended from the entrance of the superior vena cava into the right atrium down to the posterior leaflet of the tricuspid valve. Because the electrode could not be removed by traction, the ventricle was inverted into the atrium, and the electrode was cut out of its position in the apex and septum. All visible vegetations were removed. A new pulse generator and two screw-in epicardial electrodes were implanted. Cultures of the vegetations removed at operation and of the fractured electrode tip revealed Staphylococcus epidermidis. There were no major postoperative complications. She received intravenous cephalothin sodium and then oral cephalexin, which was continued for 8 weeks. One year after discharge from the hospital, she remains afebrile and is active, with no further evidence of pacemaker-related infection or of endocarditis.
Discussion Two syndromes of endocarditis related to transvenous pacemakers should be distinguished on the basis of pathogenesis (Table I). Case I illustrates what might be called the "metastatic implantation" type of pacemaker-related endocarditis. This syndrome has been encountered infrequently.v 5 but it is analogous to the experimental model of endocarditis in which bacteremia is induced after abrading the endocardium with a polyethylene catheter" or pacemaker electrode. 7 The source of bacteremia is thus remote from the pacemaker system, and the pulse generator pocket remains uninfected. Case 2 illustrates what might be called the "foreign body" type of pacemaker-related endocarditis. Local infection at the generator site, presumably acquired during the operation, propagates along the electrode and eventually causes intracardiac infection. This syndrome appears to be more common and usually has been associated with staphylococci. 2 In Case I, the decision to remove the pacemaker and the conclusion that endocarditis was present merit discussion. Bacteremic Pseudomonas aeruginosa pneumonia developed explosively in an elderly patient whose health had been good except for Stokes-Adams
Volume 75 Number 5 May, 1978
attacks. Bacteremic Pseudomonas pneumonia is extremely lethal; in a recent study, Iannini and coworkers" reported no survivors in 20 cases and could find but a single survivor among more than 28 previously reported cases. We therefore reasoned that pacemaker-related endocarditis was the patient's only "treatable" illness and that removal of the pacemaker represented his major chance for survival. To our knowledge, Pseudomonas aeruginosa endocarditis related to transvenous pacemakers has not been reported previously. Recent interest in Pseudomonas endocarditis owes mainly to its prevalence among subjects addicted to heroin in some areas"; it is of interest that Pseudomonas endocarditis can be produced experimentally with relative ease after endothelial abrasion. 10 The portal of entry for Pseudomonas in our patient may well have been the subsequently discovered adenocarcinoma of the colon. The isolation of Pseudomonas aeruginosa from the pacemaker electrode at a time when blood cultures were becoming sterile, the appearance of petechial skin lesions, the definite changes in the systolic murmur so that it became increasingly suggestive of tricupsid regurgitation, and the patient's eventual cure after removal of the transvenous pacemaker system all constitute strong evidence that endocarditis was present. The gradual progression of local infection at the pulse generator site to extensive vegetative endocarditis in Case 2 resembles more nearly the majority of previously published instances of pacemaker-related endocarditis. A local sinus tract and intermittent fever with sterile blood cultures preceded the recognition of bacteremia and endocarditis by 3 years. The decisions to allow the original fractured electrode to be retained and to attempt control of the infection with prolonged oral antibiotic therapy find acceptance in the recommendations of previous authors.!": 12 Our patient, however, became increasingly symptomatic despite prolonged oral antibiotic therapy with documented adequate blood levels. Dense adherence of the pacemaker electrode system to the endocardium, as in Case 2, precludes the withdrawal of the enlarged, bulbous electrode tip through the fibrous ring which surrounds it and attaches it to the tricupsid valve. Although a trial of intensive parenteral antibiotic therapy such as that used by Zeller and colleagues' could not be completed in our patient, we doubt that such therapy would have resulted in cure in view of the previous failures of antibiotic therapy and the unusually extensive intracardiac vegetations. We concur with others" that cardiopulmonary bypass should be carried out in such patients, if at all possible.
Endocarditis related to transvenous pacemakers
76 I
The rarity of both of these syndromes seems remarkable because of the similarity of pacemaker-induced endothelial trauma to experimental models of endocarditis'"" and because of an appreciable incidence, up to 7 percent, of local infection related to pacemaker generators.v 13. 14 The apparent rarity of the "metastatic implantation" type of pacemaker-related endocarditis probably reflects the organization of pacemaker-induced endocardial lesions into dense layers of endothelialized fibrous tissue.!" This histologic composition differs from the endothelial abrasions and thrombus formation found immediately after implantation. However, the potential for such abrasions probably persists, explaining sporadic instances such as our Case I and previously published cases.": 5 We do not believe that prophylactic antibiotic therapy during procedures which are likely to cause bacteremia is justified in such patients on the basis of available data. The apparent rarity of pacemaker-related endocarditis may reflect, in part, a reluctance of some authors to employ the term "endocarditis" in this setting. Thus such infections have been described as "septicemia," even though cultures of the endocardium taken at operation were positive.f or as "sustained bacteremia,"16 even though endocarditis with sterile blood cultures or only occasional positive blood cultures is widely recognized. We suggest that the term' 'endocarditis" is appropriate and serves to emphasize the intracardiac location of disease and the need for both medical and surgical management. Finally, we suggest that the opinion':' that there are no medical contraindications to transvenous pacing requires minor modification. The existence of active endocarditis, especially on the right side, should be deemed at least a relative contraindication, especially since bradyarrythmias can be managed with epicardial pacing systems. REFERENCES
2
3
4
5
Zeller NH, Lusk RH, Palmer DL: Nonsurgical cure of endocarditis associated with a pacemaker. Arch Intern Med 135:580-581, 1975 Chavez CM, Conn JH: Septicemia secondary to impacted infected pacemaker wire. Successful treatment by removal with cardiopulmonary bypass. J THORAC CARDIOVASC SURG 73:796-800, 1977 Anderson EL, Gramling PK, Vestal PR, Farrar WE Jr: Susceptibility of Pseudomonas aeruginosa to tobramycin or gentamicin alone and combined with carbenicillin. Antimicrob Agents Chemother 8:300-304, 1975 Schwartz IS, Pervez N: Bacterial endocarditis associated with a permanent transvenous cardiac pacemaker. JAMA 218:736-737, 1971 Svanbom M, Gastrin B, Rodriguez L: Transvenous car-
762
6
7
8 9
10
Bryan et al.
diac pacemaker as a focus of Salmonella infection in a patient with heart block. Acta Med Scand 196:281-284, 1974 Garrison PK, Freedman LR: Experimental endocarditis. I. Staphylococcal endocarditis in rabbits resulting from placement of a polyethylene catheter in the right side of the heart. Yale J Bioi Med 42:394-410, 1970 Tanphaichitra, D., Ries, K., Levison, M. E.: Susceptibility to Streptococcus viridans endocarditis in rabbits with intracardiac pacemaker electrodes or polyethylene tubing. J Lab Clin Med 84:726-730, 1974 Iannini PB Claffey T, Quintiliani R: Bacteremic Pseudomonas pneumonia. JAMA 230:558-561, 1974 Reyes MP, Palutke WA, Wylin RF, Lerner AM: Pseudomonas endocarditis in the Detroit Medical Center, 1969-1972. Medicine 52: 173-I94, 1973 Archer G, Fekety FR: Experimental endocarditis due to Pseudomonas aeruginosa. I. Description of a model. J Infect Dis 134: 1-7, 1976
The Journal of Thoracic and Cardiovascular Surgery
II Furman S, Escher DJW: Retained endocardial pacemaker electrodes. J THORAC CARDIOVASC SURG 55:737-740, 1968 12 Craig CP: Long-term suppression of foreign body endocarditis with cephalexin. Am Heart J 84:714-715, 1972 13 Conklin EF, Giannelli S Jr, Nealon TF Jr: Four hundred consecutive patients with permanent transvenous pacemakers. J THORAC CARDIOVASC SURG 69:1-7, 1975 14 Grogler PM, Frank G, Greven G, Dragojevic D, Oelert H, Leitz K, Dalichau H, Brinke U, Lohlein D, Rogge D, Hetzer R, Hennersdorf G, Borst HG: Complications of permanent transvenous cardiac pacing. J THORAC CARDIOVASC SURG 69:895-904, 1975 15 Huang T-Y, Baba N: Cardiac pathology of transvenous pacemakers. Am Heart J 83:469-474, 1972 16 Corman LC, Levison ME: Sustained bacteremia and transvenous cardiac pacemakers. JAMA 233:264-266, 1975
Information for authors Most of the provisions of the Copyright Act of 1976 became effective on January I, 1978. Therefore, all manuscripts must be accompanied by the following statement, signed by each author: "The undersigned author(s) transfers all copyright ownership of the manuscript entitled (title of article) to The C. v. Mosby Company in the event the work is published. The author(s) warrants that the article is original, is not under consideration by another journal, and has not been previously published." Authors will be consulted, when possible, regarding republication of their material.
Charles S. Bryan, M.D., F.A.C.P.,* John P. Sutton, M.D., F.A.C.S.,** Donald E. Saunders, Jr., M.D., F.A.C.P.,*** Dail W. Longaker, M.D., F.A.C.S., and Claude W. Smith, M.D., Columbia, S. C.
Microbial endocarditis related to transvenous pacemaker implantation is considered to be rare, the literature consisting mainly of single case reports. Management remains unstandardized, especially with regard to the need to remove the pacemaker by cardiotomy when attempts to withdraw the electrode by simple traction fail. Zeller and colleagues! suggested a trial of intensive antibiotic therapy in this setting. On the other hand, Chavez and Conn" outlined the need to remove the infected wire even should cardiotomy be required. Recently, we have treated two elderly patients with pacemaker-associated endocarditis which began 5 and 7 years after the original implantations. The strikingly different presentations suggested to us that a clear distinction, on the basis of pathogenesis, should be made From the University of South Carolina School of Medicine, Columbia, S. C. Received for publication Oct. 14, 1977. Accepted for publication Dec. 20, 1977. Address for reprints: Dr. C. S. Bryan, 3301 Harden Street, ACC II, Columbia, S. C. 29203. *Associate Professor, Department of Medicine. **Clinical Associate Professor, Department of Surgery. ***Clinical Associate Professor, Department of Medicine. 758
between two syndromes of pacemaker-associated endocarditis. Appreciation of these syndromes should facilitate early recognition and appropriate management of such problems.
Case reports CASE I. N. W., a 75-year-old man, was admitted to the hospital on Nov. 25, 1975, with fever and chills of 2 days' duration. A permanent transvenous pacemaker had been inserted in December, 1970, for Stokes-Adams attacks. The pulse generator had last been changed, uneventfully, in December, 1974. His general health had been good, and there was no history of chronic lung disease, diabetes mellitus, heart failure, or recent antibiotic therapy. A faint systolic murmur at the left sternal border had been described on previous physical examinations. Although he appeared acutely ill on admission to the hospital, physical examination disclosed no source for the fever. The previously described murmur was not heard. The initial chest x-ray film revealed mild congestive changes with some haziness along the right border of the heart but no definite infiltrates. Within several days, the following findings developed: substernal pain and pleuritic pain on the right side of the chest; a Grade 2/6, nearly pansystolic murmur along the lower left sternal border; a I em. black nodular lesion on the lower lip; and a cluster of petechiae about the right ankle. The chest x-ray film (Fig. I, A) revealed extensive, bilateral nodular pulmonary infiltrates. Pseudomonas aeruginosa, Fisher-
0022-5223/78/0575-0758$00.50/0 © 1978 The C. V. Mosby Co.
Volume 75 Number 5 May, 1978
Endocarditis related to transvenous pacemakers
759
A Fig. 1. A , Extensive bilateral pulmonary infiltrates early in the course of bacteremic Pseudomonas pneumonia. B, Resolution of infiltrate s after prolonged antibiotic therapy, removal of the transvenous pacing system, and implant ation of epicardial pacing electrodes. Devlin-Gnabasik immunotype 3,* was isolated from three of six blood cultures and from two sputum cultures. Urine cultures were sterile. Therapy with gentamicin sulfate and disodium carbenicillin w~s begun on the third hospital day . The microorganism was shown to be susceptible to tobramycin sulfate at a minimum inhibitory concentration (MIC) of 1.56 meg. per milliliter, to gentamicin sulfate at 6.13 meg. per milliliter, and to carbenicillin at 31 meg. per milliliter. Synergism between tobramycin and carbenicillin was demonstrated by a twodimensional agar dilution checkerboard study ." A combination of tobramycin and carbenicillin was therefore used for subsequent therapy. Because bacteremic Pseudomonas aeruginosa pneumonia has been an almost uniformly fatal disease despite maximum antibiotic therapy (see Discussion), and because no other factors could be identified which might have predisposed the patient to this illness, it was elected to remove the transvenous pacemaker even though there was no direct evidence of infection of the generator pocket or electrode . This was accomplished by traction on the ninth hospital day. The generator pocket appeared to be unremarkable . However, Pseudomonas aeruginosa was isolated from the electrode tip, at a time when cultures of the peripheral blood were becoming sterile. *Typing done by George C. Cole, Ph.D ., Parke, Davis & Company, Detroit, Mich.
Fever and pneumonia resolved with continued therapy (Fig. I, B). Third-degree heart block and a brief period of asystole, occurring on the twenty-eighth hospital day, prompted implantation of a permanent unipolar demand pacemaker with screw-in epicardial electrodes by means of a left thoracotomy . Rectal bleeding occurred and was thought possibly to be due to hemorrhoidal bleeding aggravated by the hemostatic defect known to result from high-dose carbenicillin therapy. Barium enema revealed only diverticulosis, and sigmoidoscopic examination showed no abnormalities. The patient was discharged after 42 days of combined aminoglycoside-carbenicillin therapy . The systolic murmur became longer, sometimes with definite inspiratory augmentation . He remained afebrile after discharge but was mildly anemic and failed to regain his original weight. On March 3, 1976, he was admitted to the hospital again because of the gradual appearance of warmth and erythema near the sixth and seventh costochondral junctions on the right and because of pain over the right ischium. Excisional biopsy of the inflamed area over the chest disclosed a nonspecific cellulitis, from which Pseudomonas aeruginosa was isolated . A biopsy specimen from the right ischium was sterile. Pseudomonas osteomyelitis was thought to be present at both sites , and therapy with tobramycin and carbenicillin was again instituted . Rectal bleeding again occurred during antibiotic therapy. Barium enema and sigmoidoscopic study were essentially unrevealing. On examination of the colon with a fiberoptic col -
The Journal of
760
Bryan et al.
Thoracic and Cardiovascular Surgery
Table I. Syndromes of transvenous pacemaker-related endocarditis
Relative frequency (fromliterature) Source of bacteria Pathogenesis Microorganisms Experimental model
"Metastatic implantation" type (Case /)
"Foreign body" type (Case 2)
Less common Remote site Implantation of microorganisms, in the course of bacteremia, on traumatized endothelium Diverse Abrasion of endothelium followed by induced bacteremia (so-called "rabbit model endocarditis")
More common Pacemaker generator site Direct extension of infection alongpacemaker wire, with or without systemic bacteremia Usually staphylococci Contamination of a foreign body, resulting in local spread of infection
onoscope, an adenocarcinoma was found at the splenic flexure. One of six mesenteric lymph nodes was abnormal at the time of resection. He was discharged from the hospital after 53 days of combined tobramycin-carbenicillin therapy. In 15 months of follow-up, he has enjoyed good health without recurrence of Stokes-Adams attacks, endocarditis, osteomyelitis, other evidence of Pseudomonas infection, or carcinoma. CASE 2. E. F., a 76-year-old woman, was admitted to the hospital on Aug. 13, 1976, for removal of an infected pacemaker wire. A permanent transvenous pacemaker had been implanted for complete heart block of undetermined origin in 1966. In 1973, infection of the generator pocket due to coagulasenegative staphylococci was recognized. The generator was removed, but it was not possible to remove the electrode by traction. It was therefore allowed to remain in place, a second transvenous pacemaker system being implanted with the pulse generator on the opposite side. Between 1973 and 1976, she had an intermittently draining sinus at the site of the original pulse generator. Cultures of this sinus intermittently revealed coagulase-negative staphylococci. Blood cultures were sterile. Intermittent fever and chills responded to courses of oral antibiotic therapy. Several further attempts were made to withdraw the original electrode by traction, but it continued to be densely adherent to the right ventricular endocardium. On Feb. 2, 1976, a final attempt was made to remove the original fractured wire by traction. Again, the electrode appeared to be firmly anchored to the endocardium. The proximal portion was severed. Coagulase-negative staphylococci were isolated from this segment of wire, but blood cultures remained sterile. It was elected to institute intensive parenteral antibiotic therapy. However, a paucity of peripheral veins and the occurrence of a generalized rash while the woman was receiving nafcillin sodium led to termination of this therapy. She completed a 10 week course of oral cephalexin combined with probenecid. Blood levels were monitored during this therapy and appeared to be adequate. On June 17, 1976, recurrent chills prompted additional blood cultures; two of six cultures revealed a coagulasenegative staphylococcus, subsequently confirmed to be Staphylococcus epidermidis, susceptible to cephalothin at an MIC of 0.2 mcg. per milliliter and to cephalexin at 0.8 mcg. per milliliter. Fever and other symptoms persisted during oral antibiotic therapy. It was therefore elected to remove the retained electrode by cardiopulmonary bypass surgery. On Aug. 17, 1976, she underwent median stemotomy and
cardiopulmonary bypass. The superior vena cava was almost totally occluded. Vegetations extended from the entrance of the superior vena cava into the right atrium down to the posterior leaflet of the tricuspid valve. Because the electrode could not be removed by traction, the ventricle was inverted into the atrium, and the electrode was cut out of its position in the apex and septum. All visible vegetations were removed. A new pulse generator and two screw-in epicardial electrodes were implanted. Cultures of the vegetations removed at operation and of the fractured electrode tip revealed Staphylococcus epidermidis. There were no major postoperative complications. She received intravenous cephalothin sodium and then oral cephalexin, which was continued for 8 weeks. One year after discharge from the hospital, she remains afebrile and is active, with no further evidence of pacemaker-related infection or of endocarditis.
Discussion Two syndromes of endocarditis related to transvenous pacemakers should be distinguished on the basis of pathogenesis (Table I). Case I illustrates what might be called the "metastatic implantation" type of pacemaker-related endocarditis. This syndrome has been encountered infrequently.v 5 but it is analogous to the experimental model of endocarditis in which bacteremia is induced after abrading the endocardium with a polyethylene catheter" or pacemaker electrode. 7 The source of bacteremia is thus remote from the pacemaker system, and the pulse generator pocket remains uninfected. Case 2 illustrates what might be called the "foreign body" type of pacemaker-related endocarditis. Local infection at the generator site, presumably acquired during the operation, propagates along the electrode and eventually causes intracardiac infection. This syndrome appears to be more common and usually has been associated with staphylococci. 2 In Case I, the decision to remove the pacemaker and the conclusion that endocarditis was present merit discussion. Bacteremic Pseudomonas aeruginosa pneumonia developed explosively in an elderly patient whose health had been good except for Stokes-Adams
Volume 75 Number 5 May, 1978
attacks. Bacteremic Pseudomonas pneumonia is extremely lethal; in a recent study, Iannini and coworkers" reported no survivors in 20 cases and could find but a single survivor among more than 28 previously reported cases. We therefore reasoned that pacemaker-related endocarditis was the patient's only "treatable" illness and that removal of the pacemaker represented his major chance for survival. To our knowledge, Pseudomonas aeruginosa endocarditis related to transvenous pacemakers has not been reported previously. Recent interest in Pseudomonas endocarditis owes mainly to its prevalence among subjects addicted to heroin in some areas"; it is of interest that Pseudomonas endocarditis can be produced experimentally with relative ease after endothelial abrasion. 10 The portal of entry for Pseudomonas in our patient may well have been the subsequently discovered adenocarcinoma of the colon. The isolation of Pseudomonas aeruginosa from the pacemaker electrode at a time when blood cultures were becoming sterile, the appearance of petechial skin lesions, the definite changes in the systolic murmur so that it became increasingly suggestive of tricupsid regurgitation, and the patient's eventual cure after removal of the transvenous pacemaker system all constitute strong evidence that endocarditis was present. The gradual progression of local infection at the pulse generator site to extensive vegetative endocarditis in Case 2 resembles more nearly the majority of previously published instances of pacemaker-related endocarditis. A local sinus tract and intermittent fever with sterile blood cultures preceded the recognition of bacteremia and endocarditis by 3 years. The decisions to allow the original fractured electrode to be retained and to attempt control of the infection with prolonged oral antibiotic therapy find acceptance in the recommendations of previous authors.!": 12 Our patient, however, became increasingly symptomatic despite prolonged oral antibiotic therapy with documented adequate blood levels. Dense adherence of the pacemaker electrode system to the endocardium, as in Case 2, precludes the withdrawal of the enlarged, bulbous electrode tip through the fibrous ring which surrounds it and attaches it to the tricupsid valve. Although a trial of intensive parenteral antibiotic therapy such as that used by Zeller and colleagues' could not be completed in our patient, we doubt that such therapy would have resulted in cure in view of the previous failures of antibiotic therapy and the unusually extensive intracardiac vegetations. We concur with others" that cardiopulmonary bypass should be carried out in such patients, if at all possible.
Endocarditis related to transvenous pacemakers
76 I
The rarity of both of these syndromes seems remarkable because of the similarity of pacemaker-induced endothelial trauma to experimental models of endocarditis'"" and because of an appreciable incidence, up to 7 percent, of local infection related to pacemaker generators.v 13. 14 The apparent rarity of the "metastatic implantation" type of pacemaker-related endocarditis probably reflects the organization of pacemaker-induced endocardial lesions into dense layers of endothelialized fibrous tissue.!" This histologic composition differs from the endothelial abrasions and thrombus formation found immediately after implantation. However, the potential for such abrasions probably persists, explaining sporadic instances such as our Case I and previously published cases.": 5 We do not believe that prophylactic antibiotic therapy during procedures which are likely to cause bacteremia is justified in such patients on the basis of available data. The apparent rarity of pacemaker-related endocarditis may reflect, in part, a reluctance of some authors to employ the term "endocarditis" in this setting. Thus such infections have been described as "septicemia," even though cultures of the endocardium taken at operation were positive.f or as "sustained bacteremia,"16 even though endocarditis with sterile blood cultures or only occasional positive blood cultures is widely recognized. We suggest that the term' 'endocarditis" is appropriate and serves to emphasize the intracardiac location of disease and the need for both medical and surgical management. Finally, we suggest that the opinion':' that there are no medical contraindications to transvenous pacing requires minor modification. The existence of active endocarditis, especially on the right side, should be deemed at least a relative contraindication, especially since bradyarrythmias can be managed with epicardial pacing systems. REFERENCES
2
3
4
5
Zeller NH, Lusk RH, Palmer DL: Nonsurgical cure of endocarditis associated with a pacemaker. Arch Intern Med 135:580-581, 1975 Chavez CM, Conn JH: Septicemia secondary to impacted infected pacemaker wire. Successful treatment by removal with cardiopulmonary bypass. J THORAC CARDIOVASC SURG 73:796-800, 1977 Anderson EL, Gramling PK, Vestal PR, Farrar WE Jr: Susceptibility of Pseudomonas aeruginosa to tobramycin or gentamicin alone and combined with carbenicillin. Antimicrob Agents Chemother 8:300-304, 1975 Schwartz IS, Pervez N: Bacterial endocarditis associated with a permanent transvenous cardiac pacemaker. JAMA 218:736-737, 1971 Svanbom M, Gastrin B, Rodriguez L: Transvenous car-
762
6
7
8 9
10
Bryan et al.
diac pacemaker as a focus of Salmonella infection in a patient with heart block. Acta Med Scand 196:281-284, 1974 Garrison PK, Freedman LR: Experimental endocarditis. I. Staphylococcal endocarditis in rabbits resulting from placement of a polyethylene catheter in the right side of the heart. Yale J Bioi Med 42:394-410, 1970 Tanphaichitra, D., Ries, K., Levison, M. E.: Susceptibility to Streptococcus viridans endocarditis in rabbits with intracardiac pacemaker electrodes or polyethylene tubing. J Lab Clin Med 84:726-730, 1974 Iannini PB Claffey T, Quintiliani R: Bacteremic Pseudomonas pneumonia. JAMA 230:558-561, 1974 Reyes MP, Palutke WA, Wylin RF, Lerner AM: Pseudomonas endocarditis in the Detroit Medical Center, 1969-1972. Medicine 52: 173-I94, 1973 Archer G, Fekety FR: Experimental endocarditis due to Pseudomonas aeruginosa. I. Description of a model. J Infect Dis 134: 1-7, 1976
The Journal of Thoracic and Cardiovascular Surgery
II Furman S, Escher DJW: Retained endocardial pacemaker electrodes. J THORAC CARDIOVASC SURG 55:737-740, 1968 12 Craig CP: Long-term suppression of foreign body endocarditis with cephalexin. Am Heart J 84:714-715, 1972 13 Conklin EF, Giannelli S Jr, Nealon TF Jr: Four hundred consecutive patients with permanent transvenous pacemakers. J THORAC CARDIOVASC SURG 69:1-7, 1975 14 Grogler PM, Frank G, Greven G, Dragojevic D, Oelert H, Leitz K, Dalichau H, Brinke U, Lohlein D, Rogge D, Hetzer R, Hennersdorf G, Borst HG: Complications of permanent transvenous cardiac pacing. J THORAC CARDIOVASC SURG 69:895-904, 1975 15 Huang T-Y, Baba N: Cardiac pathology of transvenous pacemakers. Am Heart J 83:469-474, 1972 16 Corman LC, Levison ME: Sustained bacteremia and transvenous cardiac pacemakers. JAMA 233:264-266, 1975
Information for authors Most of the provisions of the Copyright Act of 1976 became effective on January I, 1978. Therefore, all manuscripts must be accompanied by the following statement, signed by each author: "The undersigned author(s) transfers all copyright ownership of the manuscript entitled (title of article) to The C. v. Mosby Company in the event the work is published. The author(s) warrants that the article is original, is not under consideration by another journal, and has not been previously published." Authors will be consulted, when possible, regarding republication of their material.