1. Price, Cynthia

Article Content

According to this study:


* Infection of a cardiac device necessitating its removal is commonly associated with episodes of bloodstream infection with Staphylococcus aureus, but not with Gram-negative bacilli.


* Infection rates were higher in patients with defibrillators than in those with pacemakers.



The estimated rate of device infection after implanting a permanent endocardial pacemaker or cardioverter-defibrillator has increased by 124% in the past 10 years (although infection rates vary widely, ranging from 0.13% to 19.9%). Because parenteral antibiotic treatment for an infected cardiac device has been generally unsuccessful, the consensus is that the device should be removed if an infection is detected. The accurate determination of whether infection is present and which organisms are responsible for it can help to avert unnecessary removal of cardiac devices, an important consideration in the many patients who are dependent on them and who would be presented with possibly life-threatening complications of removal.


Now, in what the study authors believe to be the first population-based study comparing the rates of infection of implanted cardiac devices with various pathogens, researchers found that device infection was associated much more commonly with bloodstream infections with Staphylococcus aureus than with Gram-negative bacilli, and that infection rates were higher with implantable cardioverter-defibrillators than with permanent endocardial pacemakers.


In a study conducted from 1975 to 2004, 1,524 Mayo Clinic patients with cardiac devices living in Minnesota were followed for 7,578 years' worth (person-years) of follow-up. The findings revealed that 12 (54.6%) of 22 cases of S. aureus bloodstream infection were either definitively or possibly linked to cardiac device infection, compared with three (12%) of 25 cases of bloodstream infection resulting from Gram-negative bacilli. In addition, the incidence of device-related endocarditis or generator pocket infection in the presence of bloodstream infection was 1.14 per 1,000 device-years, and the incidence of definite device infection was 1.9 per 1,000 device-years. Further, the incidence of defibrillator infection, at 8.9 per 1,000 device-years, was much higher than the incidence of pacemaker infection, at 1 per 1,000 device-years.


The findings led the study authors to conclude that, while patients with S. aureus bloodstream infection most likely have infected cardiac devices and should therefore have them removed, patients with bloodstream infection caused by Gram-negative bacilli generally do not run the risk of such infection and therefore do not need to have theirs removed.


Most of the study limitations are associated with the 30-year retrospective design, including the homogeneity of the sample, possibly precluding the generalizability of the findings to other populations, the use of International Classification of Diseases, ninth revision, codes adapted for the screening of possible cases of bloodstream infection rather than computerized microbiology records (which weren't available until the later years of the study period), the lack of universal diagnostic transesophageal echocardiography, and changes in microbiologic techniques during the study period.


Further study is necessary to determine whether the variance in the rates of infection according to device are caused by underlying comorbidities or demographic factors among patients with defibrillators, differences in the physical properties of devices or leads, or risks associated with implantation.




Uslan DZ, et al. Arch Intern Med 2007;167(7):669-75.