HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Feza Nurozler Michael Argenziano Mehmet C. Oz Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nurozler, F. Articles by Naka, Y. Search for Related Content PubMed PubMed Citation Articles by Nurozler, F. Articles by Naka, Y. Related Collections Mechanical Circulatory Assistance

Ann Thorac Surg 2001;71:614-618
© 2001 The Society of Thoracic Surgeons

---

Original article: cardiovascular

Fungal left ventricular assist device endocarditis

^a Department of Surgery, College of Physicians & Surgeons of Columbia University, New York, New York, USA

Accepted for publication March 22, 2000.

Address reprint requests to Dr Naka, Division of Cardiothoracic Surgery, College of Physicians & Surgeons of Columbia University, Milstein Bldg 7-435, 177 Ft Washington Ave, New York, NY 10032
e-mail: yn33{at}columbia.edu

	Abstract

Top Abstract Introduction Patients and methods Results Comment References

 
Background. Infection remains as the most serious complication and represents a significant threat to patients during long-term mechanical circulatory support. Fungal infection is a particularly worrisome complication and left ventricular assist device (LVAD) endocarditis does pose a serious threat.

Methods. One hundred and sixty-five patients underwent TCI Heartmate LVAD implantation between July 1991 and December 1999 at our institution. Detailed medical records were kept prospectively for all patients, and a variety of infection-related endpoints were analyzed on patients with fungal LVAD endocarditis.

Results. Thirty-seven patients (22%) developed fungal infections during LVAD support. Five (3%) of those met our criteria for the diagnosis of fungal LVAD endocarditis. Microbial portals of entry were identifiable in all cases. Infections were managed successfully in 4 patients (80%).

Conclusions. The successful management of fungal LVAD endocarditis currently requires early recognition of potentially nonspecific signs and symptoms, and timely institution of antifungal therapy. In some cases with device-specific manifestations of LVAD endocarditis, device removal and replacement is necessary. In patients with clinical manifestations of sepsis and fungal driveline site or pocket infections without positive blood culture, urgent transplantation may be the appropriate management. In the setting of shortage in the donor supply, device removal and replacement is necessary.

	Introduction

Top Abstract Introduction Patients and methods Results Comment References

 
Mechanical circulatory support with left ventricular assist devices (LVAD) has been used successfully as a bridge to heart transplantation and to left ventricular recovery in increasing number of centers. The potential for infection to adversely affect the risks of circulatory support in patients with implanted total artificial hearts and left ventricular assist devices has long been recognized [1, 2]. Infection remains as a significant life-threatening complication to patients receiving LVAD despite new technology used in the implantable LVADs [3, 4] and has serious implications for the anticipated application of LVADs as permanent therapy. Recipients of left ventricular assist devices are prone to nosocomial infections due to a number of device-related and host-related predisposing factors [3, 5]. Opportunistic fungal infections, which are characterized by resistance to treatment, also occur frequently in these patients due to a number of predisposing factors, including defects in cell-mediated immunity [5]. Treatment with broad-spectrum antibiotics may also make the patient more susceptible to fungal colonization and infection. Although infections during LVAD support have not been associated with inferior outcomes, a subset of these infections does pose a serious threat [3]. LVAD endocarditis, defined as infection of the blood-contacting LVAD surface, was associated with a high mortality rate [3]. In this report, we focus on the patients with fungal LVAD endocarditis.

	Patients and methods

Top Abstract Introduction Patients and methods Results Comment References

 
One hundred and sixty-five patients underwent TCI Heartmate LVAD implantation between July 1991 and December 1999, at the Columbia Presbyterian Medical Center. Fifty-three patients received pneumatic devices, whereas vented electric device (VE) was used in 112 cases. Detailed medical records were kept prospectively for all patients undergoing LVAD implantation, and for the purposes of the present analysis, this database was utilized to determine the incidence and distribution of infections occurring before and after LVAD implantation. Preoperative infections were defined as a leukocytosis or fever in the presence of a positive culture. Infections during LVAD support were similarly defined and included infections of the driveline and LVAD pocket. Infections were defined as device-related if a positive culture was obtained from the LVAD pocket, driveline sites, or explanted LVADs. Driveline infections were defined as local signs of infection at the site of driveline entry. All explanted LVADs were immediately disassembled during surgery, allowing visual inspection and culturing of all blood-contacting LVAD surfaces (diaphragm, inflow–outflow valves, and inflow–outflow grafts). The diagnosis of LVAD endocarditis was reserved for patients with positive cultures of the blood-contacting LVAD surface at explantation with or without accompanying manifestations of systemic infection. All patients undergoing LVAD insertion received antimicrobial prophylaxis including intravenous fluconazole given preoperatively and continued for 2 days. Additionally, patients with culture-proven fungal colonization received a full 10-day course of fluconazole. Driveline sites were cleansed antiseptically (with hydrogen peroxide and povidone-iodine solutions) on a daily basis and dressed with sterile gauze.

Statistical significance of differences of device-related fungal infection rates between pneumatic and vented electric Heartmate LVAD recipients were tested by Fisher’s exact test.

	Results

Top Abstract Introduction Patients and methods Results Comment References

 
Of the 165 patients receiving LVADs, 5 (3.0%) had positive fungal culture (3 urine, 2 sputum) at the time of LVAD implantation. Four of those were considered colonization and not treated; none of those patients developed clinical fungal infection during the course of LVAD support. Ninety-eight patients (59%) subsequently developed clinical infections confirmed by positive cultures of blood, sputum, urine, stool, central venous catheters, LVAD driveline and pocket sites, abscess cavities, and explanted LVAD. Thirty-seven patients (22.4%) subsequently developed fungal infections confirmed by a positive culture for fungus (34 candida species, 2 aspergillus, 1 syncephalastrum). Eighteen (10.9%) were defined as the device-related fungal infection confirmed by positive fungal culture from LVAD pocket, mediastinum, driveline sites, and explanted LVADs. Fungal driveline infections were more common in patients who received pneumatic devices (10.3% versus 3.5%); on the contrary, fungal LVAD pocket or mediastinal infections were more common in patients who received vented electric devices (6.2% versus 1.8%). However, differences were not statistically significant (Table 1). Five (3.0%) of those patients had positive fungal cultures of the blood-contacting LVAD surfaces at explantation, meeting our criteria for the diagnosis of fungal LVAD endocarditis (Table 2). Microbial portals of entry were identifiable in all cases with fungal LVAD endocarditis and consistent of two driveline infections, two LVAD pocket infections, and one mediastinatis. Of the 5 patients who developed fungal LVAD endocarditis, 2 received pneumatic devices, whereas vented electric devices were used in other 3. The positive fungal cultures of the blood-contacting LVAD surfaces accompanied manifestations of systemic infection in 4 patients. Of the 18 patients who developed device-related fungal infections (driveline, LVAD pocket, mediastinum), 13 (72%) did not developed blood-contacting infections. All fungal LVAD endocarditis patients also developed one or more concomitant bacterial infections confirmed by positive culture and received broad-spectrum antibiotic treatment. In none of the patients, did two-dimensional echocardiography detect any vegetation or pannus formation at the blood-contacting LVAD surfaces. Average length of LVAD support at the time of diagnosis of LVAD endocarditis was 78 ± 56 days. Of the 5 patients, 1 patient expired and 4 patients were bridged to transplantation. LVAD was explanted and replaced with a new device in 1 patient. The survivors received 3 to 6 weeks of intravenous amphotherecin B treatment with or without flucanozole treatment.

In the first of these fungal LVAD endocarditis patients, clinical manifestations of sepsis developed with c parapsilosis fungemia. The LVAD was explanted and replaced with a new device when the infection was found to be unresponsive to antifungal treatment. C parapsilosis grew out from explanted LVAD inflow valve and diaphragm cultures. The patient subsequently was transplanted successfully.

In the second and third patients, development of fungal driveline site and pocket infections with clinical manifestations of sepsis prompted urgent transplantation. In the second patient, c albicans grew out from explanted LVAD inflow valve and LVAD pocket. In the third patient, c albicans grew out from explanted LVAD diaphragm.

Device-specific manifestations of LVAD endocarditis were more acute in nature. In the fourth patient, a mild cerebrovascular accident in the setting of c albicans driveline infection prompted urgent transplantation. Upon removal of LVAD, exploration revealed a large mass, subsequently confirmed as c albicans, blocking the outflow tract of the device despite the fact that there had been no sign of outflow tract obstruction. The patient also experienced systemic multiple thromboembolic events (femoral artery occlusion, spleen and liver mass) immediately after transplantation. He has survived for 6 years.

The postoperative course of the fifth patient was complicated with fungal mediastinatis and clinical manifestations of sepsis. He developed decreased LVAD flow and impending hemodynamic collapse accompanied with LVAD "power limit advisory" alarm. During emergent exploration, the LVAD suddenly ceased functioning, suggesting complete obstruction of the outflow graft. The lumen was lined by an 8 mm-thick, yellowish pannus (subsequently confirmed as Syncephalastrum racemosum) along the entire inner surface of the graft (Fig 1). The graft to aorta anastomosis was completely obstructed by dislodged pannus preventing any backflow from the aorta. The LVAD was explanted in the hope of left ventricular recovery but and the patient subsequently expired.

View larger version (130K): [in this window] [in a new window]   Fig 1. Fungal growth inside the left ventricular assist device (LVAD) outflow graft.

	Comment

Top Abstract Introduction Patients and methods Results Comment References

Furthermore, the presence of blood-contacting prosthetic surfaces, device-related pockets and cavities, and transcutaneous drivelines and power cables increases the risk of microbial contamination and infection. Opportunistic fungal infections also occur frequently in these patients due to a number of device-related and host-related predisposing factors. LVAD implantation leads to progressive defects in cell-mediated immunity and increased risk for opportunistic infections [5]. The risk of opportunistic fungal infections has been deemed to be so high that prophylactic use of antifungal therapy in selected LVAD recipients has been recommended [6]. The pathogen can colonize the inner surfaces of the device and grafts as a bloodstream infection, or the outer surfaces of the device and drivelines as a localized infection [4, 5]. The distinguishing characteristics of prosthesis-related infections include pathogen colonization by adhesion, a decreased minimally infective inoculum size, bacterial resistance to host defense mechanisms and to antibiotic therapy, and the persistence of infection until artificial substrates are removed [7]. Candida species are the most common reported pathogens in fungal infections in LVAD recipients [3, 6].

Nosocomial candidemia has increased dramatically over the last 15 years, currently ranking fourth amongst bloodstream pathogens [8]. Broad-spectrum antibiotics may also make the patient more susceptible to fungal colonization and infection. Widespread use of antibiotics can lead to overgrowth of gastrointestinal candida and high-level candida colonization at multiple mucosal sites is a risk factor for invasive candidiasis [9]. Conversely, antifungal prophylactic measures reduce both colonization and invasive candidiasis [10].

In the absence of empiric data, information about the clinical and potential therapy of infections involving the blood-contacting LVAD surface can only be extrapolated from the fungal prosthetic valve endocarditis literature. However, diagnosing fungal prosthetic valve endocarditis remains difficult. A definitive preoperative diagnosis of fungal endocarditis requires the pathological identification of fungal emboli or isolation of fungus from the blood in the face of prosthetic valve vegetations. Two-dimensional echocardiography provides a detection rate of fungal vegetations from 80% to 85% in the patients with prosthetic valve endocarditis [11]. Detection of fungal vegetations at the blood-contacting LVAD surfaces is difficult because of the reflecting properties and design of the LVAD. Past reports have also noted blood cultures to be positive in only 50% to 70% of affected patients, and fever may not be present in the patients with fungal prosthetic valve endocarditis [12, 13]. Occasionally, fungal prosthetic valve endocarditis is diagnosed only after a patient with persistent fever and constitutional symptoms suggestive of systemic disease had undergone cardiotomy for presumed prosthetic valve endocarditis. The fungus was then identified postoperatively after culturing the vegetation. Finally, the fungal prosthetic valve endocarditis literature supports the notion that operative intervention early in the course of infection is associated with a prohibitively high mortality, and is best deferred until medical management is successful in eradicating active infection, but recognizes that early intervention may be necessary, especially in the setting of mechanical device failure. In our series, only 1 patient (20%) had positive blood culture for fungus. Microbial portals of entry were identifiable with a positive culture of the same pathogen, and consistent with infection of device-related pockets and cavities, and transcutaneous drivelines in all cases. This finding emphasizes the importance of prevention of LVAD contamination. Measures, which may decrease the risk of early device contamination, include the use of perioperative prophylactic antifungals, minimization of perioperative hemorrhage, and secure positioning of the driveline in order to promote early tissue incorporation. Prevention of the late LVAD contamination requires continued vigilance with regard to potential sources of infection, prompt diagnosis of even minor infections, and aggressive treatment of these with appropriate antimicrobial regimens. In LVAD recipients who developed fungal driveline or pocket infection, our current management is institution of potent antifungal therapy and prompt transplantation if a suitable donor can be found. We currently use IV fluconazole in patients without clinical manifestations of sepsis, however, if clinical manifestations of sepsis are observed, amphotherecin B is added to the therapy. We continue with antifungal therapy during the rest of the mechanical circulatory support. Two patients had urgent operative intervention because 1 developed the clinical signs of outflow graft obstruction while the other had cerebral emboli. One patient presented with clinical signs of outflow graft obstruction while the other had a large mass blocking the outflow tract of the device without any clinical sign of outflow tract obstruction. The occlusion of the 20-mm outflow graft by a fungal pathogen is unexpected. However, it should be remembered as part of the differential diagnosis in patients presenting with outflow graft obstruction. Accordingly, during reoperation for LVAD outflow graft obstruction, the graft should be inspected gently because of the risk of systemic embolization of dislodged fragments.

Finally, the successful management of fungal LVAD endocarditis currently requires early recognition of potentially nonspecific signs and symptoms, a high level of suspicion in cases that do not follow an identifiable pattern, and timely institution of potent antifungal therapy. In some cases with device-specific manifestations of LVAD endocarditis, prompt device removal and replacement is necessary. In patients with clinical manifestations of sepsis and fungal driveline site or pocket infections without positive blood culture, urgent transplantation may be the appropriate management if a suitable donor can be found. In the setting of shortage in the donor supply, device removal and replacement is necessary, however, it is important to recognize that some patients may not tolerate device removal and replacement, and may necessitate conservative management until hemodynamic instability and end-organ dysfunction resolve.

	References

Top Abstract Introduction Patients and methods Results Comment References

 

1. Frazier O.H., Rose E.A., Macmanus Q., et al. Multicenter clinical evaluation of the HeartMate 1000IP left ventricular assist device. Ann Thorac Surg 1992;53:1080-1090.[Abstract/Free Full Text]
2. Griffith B.P., Kormos R.L., Hardesty R.L., Armitage J.M., Dummer J.S. The artificial heart: infection-related morbidity and its effect on transplantation. Ann Thorac Surg 1988;45:409-414.[Abstract/Free Full Text]
3. Argenziano M., Catanese K.A., Moazami N., et al. The influence of infection on survival and successful transplantation in patients with left ventricular assist devices. J Heart Lung Transplant 1997;16:822-831.[Medline]
4. Fischer S.A., Trenholme G.M., Costanzo M.R., et al. Infectious complications in left ventricular assist device recipients. Clin Infect Dis 1997;24:18-23.[Abstract/Free Full Text]
5. Ankersmit J.H., Tugulea S., Spanier T., et al. Activation-induced T cell death following left ventricular assist device implantation: a iatrogenic cause of immunodeficiency. Lancet 1999;354:550-555.[Medline]
6. Goldstein D.J., El-Amir N.G., Ashton R.C., et al. Fungal infection in LVAD recipients: incidence, prophylaxis, and treatment. ASAIO J 1995;41:873-875.[Medline]
7. Gristina A.G., Gridhar G., Gabriel B.L., et al. Cell biology and molecular mechanism in artificial device infections. Int J Artif Organs 1993;16:755-764.[Medline]
8. Ellis M. Fungal endocarditis. J Infect 1997;35:99-103.[Medline]
9. Bross J., Talbot G.H., Maislin G., Hurwitzs S., Storm B.L. Risk factors for nosocomial candidamia. Am J Med 1989;87:614-620.[Medline]
10. Ellis M.E., Qudri S.M.H., Clink H., et al. The effect of fluconazole as prophlaxis for neutropenic patients. J Antimicrob Chemother 1994;33:1223-1228.[Free Full Text]
11. McFadden M., Gonzales-Levin L., Remington J.S. Limited reliability of the negative two-dimensional echocardiography in the evaluation of infectious vegetative endocarditis; diagnosis and surgical implications. J Cardiovasc Surg 1985;26:59-63.[Medline]
12. Seeling M.S., Speth C.P., Kozinn P.J., et al. Candida endocarditis after cardiac surgery: clues to earlier detection. J Thorac Cardiovasc Surg 1973;65:583-601.[Medline]
13. Muehrcke D.D. Fungal prosthetic valve endocarditis. Semin Thorac Cardiovasc Surg 1995;7:20-24.[Medline]

This article has been cited by other articles:

				S. R. Wilson, M. M. Givertz, G. C. Stewart, and G. H. Mudge Jr Ventricular Assist Devices: The Challenges of Outpatient Management J. Am. Coll. Cardiol., October 27, 2009; 54(18): 1647 - 1659. [Abstract] [Full Text] [PDF]

				S. Aggarwal, F. Cheema, M. C. Oz, and Y. Naka Long-Term Mechanical Circulatory Support Card. Surg. Adult, January 1, 2008; 3(2008): 1609 - 1628. [Full Text]

				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]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Feza Nurozler Michael Argenziano Mehmet C. Oz Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nurozler, F. Articles by Naka, Y. Search for Related Content PubMed PubMed Citation Articles by Nurozler, F. Articles by Naka, Y. Related Collections Mechanical Circulatory Assistance