|
|
||||||||
Ann Thorac Surg 2001;72:725-730
© 2001 The Society of Thoracic Surgeons
a Department of Infectious Disease, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
b Department of Cardiothoracic Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
c Transplant Center, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
d Department of Infection Control and Epidemiology, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
Accepted for publication May 16, 2001.
Address reprint requests to Dr McCarthy, Department of Cardiothoracic Surgery, Kaufman Center For Heart Failure, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195
e-mail: mccartp{at}ccf.org
| Abstract |
|---|
|
|
|---|
Methods. We retrospectively reviewed the medical records of all patients with implantable LVAD at the Cleveland Clinic with 72 hours or longer of LVAD support from January 1992 through June 2000, to determine the attack rate, incidence, and impact of nosocomial BSI in patients with LVAD. A nosocomial BSI was defined using Centers for Disease Control and Prevention definition. An LVAD-related BSI was defined as one where the same pathogen is cultured from the device and the blood with no other obvious source. Two hundred fourteen patients were included in the study (17,831 LVAD-days).
Results. One hundred forty BSI were identified in 104 patients for an attack rate of 49% and incidence of 7.9 BSI per 1000 LVAD-days. Thirty-eight percent of the BSI were LVAD associated. The most common pathogens causing BSI were coagulase-negative staphylococci (n = 33), Staphylococcus aureus, and Candida spp. (19 each), and Pseudomonas aeruginosa (16 each). A Cox proportional hazard model found BSI in patients with LVAD to be significantly associated with death (hazard ratio = 4.02, p < 0.001). Fungemia had the highest hazard ratio (10.9), followed by gram-negative bacteremia (5.1), and gram-positive bacteremia (2.2).
Conclusions. Patients with implantable LVAD have a high incidence of BSI, which are associated with a significantly increased mortality. Strategies for prevention of infection in LVAD recipients should focus on the drive line exit site until technical advances can achieve a totally implantable device.
| Introduction |
|---|
|
|
|---|
Recipients of these implantable LVAD are prone to device-related infections primarily because the extracorporeal drive lines (13.5 to 15 mm in diameter) breaches the normal cutaneous defenses against infection, providing a portal of entry for potential pathogens [2, 3]. The incidence of infection increases with duration of LVAD support, (a mean of 120 days for LVAD patients awaiting heart transplantation at the Cleveland Clinic in 1999). The primary objectives of this study were to determine the attack rate and incidence of nosocomial bloodstream infections (BSI) in patients with LVAD.
| Material and methods |
|---|
|
|
|---|
Data collection
Demographic, clinical, and microbiologic data were retrospectively abstracted from patient charts, pathology reports, and microbiology reports. During the study time period three types of LVAD were implanted in patients by cardiothoracic surgeons at the Cleveland Clinic Foundation: (1) the pneumatic HeartMate 1000 IP air-driven system (TCI); (2) the portable, vented, electrically powered HeartMate VE; and (3) the portable programmable Novacor (World Heath). In most patients the devices were implanted in a preperitoneal pocket on top of the posterior rectus sheath and below the rectus muscle [4].
Case definitions
A nosocomial BSI was defined using criteria from the Centers for Disease Control and Prevention [5]. An LVAD-associated BSI was defined as one where the same pathogen is cultured from purulent drainage around any portion of the device, including drive lines, and the blood with no other obvious source.
Case ascertainment
A computerized database of all LVAD recipients including preoperative, perioperative, and postoperative information is maintained in the Transplant Center at the Cleveland Clinic Foundation. This list was cross-referenced with a computerized database of all nosocomial BSI maintained by the Department of Infection Control and Healthcare Epidemiology.
Statistical analysis
Estimates of infection-free time by type of LVAD device were determined using Kaplan-Meier curves. The impact of BSI on death on device (all patients) and survival to transplantation were assessed using these methods. Two Cox proportional hazards models were run for time until death on device. The occurrence of a BSI was treated as a time-dependent covariate. Infection is set to zero until the model is estimating at a time point after the infection occurred, at which time infection was set to one. Probability values were considered significant if less than 0.05.
| Results |
|---|
|
|
|---|
Almost all (97%) LVADs were implanted as a bridge to transplantation. The most common type of LVAD implanted was the HeartMate VE, 41% (88 patients) followed by the pneumatic HeartMate 1000 IP, 33% (70), and the Novocar device, 26% (56 patents). We identified a total of 140 nosocomial BSI in 104 patients, for an attack rate of 49% and incidence of 7.9 BSI per 1000 LVAD-days. Thirty-eight percent (53 patients) of these BSI were LVAD associated; other sources for BSI in these patients included vascular catheters (16%), lower respiratory tract infections (6%), abdominal infections (6%), urinary tract infections (1%), and unknown (29%). The most common pathogens causing BSI were coagulase-negative staphylococci (23%, 33 episodes), Staphylococcus aureus (13.5%, 19 episodes), and Candida spp. (13.5%, 19 episodes), and Pseudomonas aeruginosa (11.4%, 16 episodes) (Table 1). Bloodstream infections with vancomycin-resistant enterococcemia occurred in 5 patients.
|
|
|
|
|
|
|
|
| Comment |
|---|
|
|
|---|
Infections have been a common complication reported in patients with implantable left ventricular assist devices [1, 6]. Multiple factors probably contribute the pathogenesis of infection in these patients, but first and foremost is that all of the currently available Food and Drug Administration-approved implantable LVAD have a transcutaneous line that carries the electrical cable and air vent to the battery pack and electronic controls. The 13.5- to 15-mm diameter drive line breaches the normal cutaneous barrier against infection and serves as a major portal of entry for pathogens. Recently published studies using pulsed field gel electrophoresis analysis of isolates obtained from the bloodstream and the drive line support the clinical observation that the drive line is a major portal of entry for nosocomial BSI in patients on LVAD support [7, 8]. Second, recipients of LVAD are critically ill by definition, often malnourished with multiple invasive supportive devices. Third, use of broad spectrum antimicrobial therapy is common, often empiric, prophylactic, or for treatment of infections in critically ill patients.
A review of published reports of infections in patients with implantable LVADs with an emphasis on BSI reveal results comparable to our findings (Table 6). The overall attack rates of infection in patients with LVADs was 34% (range, 4.5% [wounds only] to 51.4%) [7, 915]. The overall attack rate for BSI among patients with LVADs was 38% (162 episodes in 425 patients) (range, 16%53%) [7, 9, 12, 1618]. The overall incidence of BSI was 5 per 1000 device-days (range, 1.8 to 7.9) [912, 17, 18].
|
There has been increased interest in the immunologic effects of LVAD implantation and associated infections with progressive defects in cellular immunity that may be the result of T-cell activation. These defects have been associated with increased risk of Candida and other systemic infections [20]. Targeted approaches to downregulate abnormal immune reactivation in LVAD recipients may reduce the infection risk in patients on long-term support.
There were several limitations to our study. The data were collected retrospectively and specific information on antibiotic days and drive line cultures were not abstracted. We also used an epidemiologic definition of LVAD-associated BSI, which may have underestimated the true number of device-associated BSI.
The management of LVAD-associated infections, including BSI, has been challenging. We have classified LVAD-associated infections along a clinical spectrum that includes prosthetic valve endocarditis (involvement of either of the porcine valves or grafts on the inflow or outflow tracts), pocket infections or drive line infections. Some low-grade infections of the drive line or tunnel have been treated successfully with local wound care and treatment with less than 14 days of antibiotics. Other patients have relapsed after stopping antimicrobial therapy and have required chronic suppressive antibiotic treatment until transplantation. Surgical management of the LVAD infections (revisions of the infected pump pocket, drive line tunnel, or device exchange) have also been attempted. In our experience, device replacement carries a higher perioperative mortality and risk of relapse and has been restricted to patients with intractable sepsis syndrome or multiple septic embolic in the setting of maximal medical therapy.
Strategies for prevention of infection in LVAD recipients should focus on the drive line exit site. At our institution ongoing efforts to reduce drive line infections include (1) placement of the transcutaneous exit site on the side of the abdomen contralateral to where the LVAD pump pocket is placed to have a longer subcutaneous course and barrier to infection, (2) postoperative care emphasizing early immobilization of the drive line to promote granulation tissue to occur, (3) use of a silver-impregnated dressing at the exit site (Arglaes film, Maersk Medical Ltd, Mundelein, IL) and plans for use of antimicrobial impregnated drive line [21], (4) judicious use of antimicrobial agents in the perioperative period with no routine continuous antimicrobial prophylaxis, and (5) studying the effect of antimicrobial drive line coating.
Future strategies for prevention of LVAD-associated BSI are directed at eliminating the portal of entry, the transcutaneous drive line. Engineering efforts have produced prototypes for continuous flow (nonpulsatile) LVADs, with smaller transcutaneous drive lines, and transcutaneous energy transmission [22].
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. J. Gordon, A. D. Weinberg, F. D. Pagani, M. S. Slaughter, P. S. Pappas, Y. Naka, D. J. Goldstein, W. P. Dembitsky, J. C. Giacalone, J. Ferrante, et al. Prospective, Multicenter Study of Ventricular Assist Device Infections Circulation, February 12, 2013; 127(6): 691 - 702. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Toda, Y. Yonemoto, T. Fujita, Y. Shimahara, S. Sato, T. Nakatani, and J. Kobayashi Risk Analysis of Bloodstream Infection During Long-Term Left Ventricular Assist Device Support Ann. Thorac. Surg., November 1, 2012; 94(5): 1387 - 1393. [Abstract] [Full Text] [PDF] |
||||
|
|
B. A. Thomas, C. M. Logar, and A. E. Anderson Renal Replacement Therapy in Congestive Heart Failure Requiring Left Ventricular Assist Device Augmentation Perit. Dial. Int., July 1, 2012; 32(4): 386 - 392. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. K. Topkara, S. Kondareddy, F. Malik, I.- W. Wang, D. L. Mann, G. A. Ewald, and N. Moazami Infectious Complications in Patients With Left Ventricular Assist Device: Etiology and Outcomes in the Continuous-Flow Era Ann. Thorac. Surg., October 1, 2010; 90(4): 1270 - 1277. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. A. Genovese, M. A. Dew, J. J. Teuteberg, M. A. Simon, J. Kay, M. P. Siegenthaler, J. K. Bhama, C. A. Bermudez, K. L. Lockard, S. Winowich, et al. Incidence and Patterns of Adverse Event Onset During the First 60 Days After Ventricular Assist Device Implantation Ann. Thorac. Surg., October 1, 2009; 88(4): 1162 - 1170. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. O. Robertson, C. Lober, N. G. Smedira, J. L. Navia, N. Sopko, and G. V. Gonzalez-Stawinski One hundred days or more bridged on a ventricular assist device and effects on outcomes following heart transplantation Eur J Cardiothorac Surg, August 1, 2008; 34(2): 295 - 300. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Shoham, R. Shaffer, L. Sweet, R. Cooke, N. Donegan, and S. Boyce Candidemia in Patients with Ventricular Assist Devices Clinical Infectious Diseases, January 15, 2007; 44(2): e9 - e12. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Tsukui, J. J. Teuteberg, S. Murali, D. M. McNamara, J. R. Buchanan, S. Winowich, E. Stanford, M. A. Mathier, L. M. Cadaret, and R. L. Kormos Biventricular Assist Device Utilization for Patients with Morbid Congestive Heart Failure: A Justifiable Strategy Circulation, August 30, 2005; 112(9_suppl): I-65 - I-72. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Simon, S. Fischer, A. Grossman, C. Downer, B. Hota, A. Heroux, and G. Trenholme Left Ventricular Assist Device--Related Infection: Treatment and Outcome Clinical Infectious Diseases, April 15, 2005; 40(8): 1108 - 1115. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. M. Kojic and R. O. Darouiche Candida Infections of Medical Devices Clin. Microbiol. Rev., April 1, 2004; 17(2): 255 - 267. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. L. Holman, B. K. Rayburn, D. C. McGiffin, B. A. Foley, R. L. Benza, R. C. Bourge, L. J. Pinderski, and J. K. Kirklin Infection in ventricular assist devices: prevention and treatment Ann. Thorac. Surg., June 1, 2003; 75(90060): S48 - 57. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. L. Hawkins and L. M. Baddour Candida lusitaniae Infections in the Era of Fluconazole Availability Clinical Infectious Diseases, January 15, 2003; 36(2): e14 - e18. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Matsumiya, M. Nishimura, Y. Miyamoto, Y. Sawa, and H. Matsuda Successful treatment of Novacor pump pocket infection by omental transposition Ann. Thorac. Surg., January 1, 2003; 75(1): 287 - 288. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. L. Navia, P. M. McCarthy, K. J. Hoercher, N. G. Smedira, M. K. Banbury, and E. H. Blackstone Do left ventricular assist device (LVAD) bridge-to-transplantation outcomes predict the results of permanent LVAD implantation? Ann. Thorac. Surg., December 1, 2002; 74(6): 2051 - 2063. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Mekontso-Dessap, M. Kirsch, E. Vermes, C. B. Buisson, D. Loisance, and R. Houel Nosocomial Infections Occurring during Receipt of Circulatory Support with the Paracorporeal Ventricular Assist System Clinical Infectious Diseases, December 1, 2002; 35(11): 1308 - 1315. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. L. Holman, J. E. Davies, B. K. Rayburn, D. C. McGiffin, B. A. Foley, R. L. Benza, R. C. Bourge, P. Blood, and J. K. Kirklin Treatment of end-stage heart disease with outpatient ventricular assist devices Ann. Thorac. Surg., May 1, 2002; 73(5): 1489 - 1494. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |