| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
*From the Hospital das Clinicas (Dr. Pierrotti), University de São Paulo, São Paulo, Brazil; and the University of Tennessee Medical Center at Knoxville (Dr. Baddour), Knoxville, TN.
Correspondence to: Larry M. Baddour, MD, University of Tennessee, Medical Center at Knoxville, 1924 Alcoa Highway U-114; Knoxville, TN 37920-6999; lbaddour{at}mc.utmck.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Key Words: fungal • infective endocarditis • mold-related • yeast-related
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FE has been characterized by excessive mortality (> 50%) and morbidity, regardless of treatment. A combined medical-surgical approach seems to offer an improved outcome. However, there are no clinical trials to support or refute this opinion, largely because of the rarity of the syndrome. We reviewed clinical features, echocardiographic findings, microbiologic data, treatment, and outcome of all FE cases reported in the English-language literature between January 1, 1995, and June 30, 2000, to provide a current characterization of the syndrome.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Case Definition
The Duke criteria were used to define cases that were included
in the analysis. According to the Duke criteria,5
infective endocarditis (IE) can be defined as definite IE,
clinical and echocardiographic criteria or pathologic features
(demonstration of microscopic findings of a vegetation or fungi in
valvular tissue or embolus obtained at autopsy or surgery); or
possible IE, findings consistent with IE that fall short of
a definitive definition and are not rejected.
Clinical Features
Demographic information, echocardiographic findings,
microbiology data, treatment modalities, histopathologic findings, and
clinical outcome, including complications and relapses, were recorded
for all cases.
Prosthetic valve endocarditis (PVE) was defined as early PVE, when infection symptoms and signs developed within 60 days of valve replacement surgery, and late PVE, when infection symptoms and signs occurred > 60 days after surgery. One patient with an intra-atrial pacemaker infection and three patients with Gore-Tex patch (W.L. Gore and Associates; Flagstaff, AZ) infections were categorized as PVE. Complicated endocarditis was defined as the presence of congestive heart failure, embolic phenomenon, valvular insufficiency, or prosthetic valve dehiscence. Relapse was defined as the return of symptoms and signs of endocarditis after an initial response to treatment that was caused by the same microorganism as was initially isolated. Long-term suppressive therapy was defined as treatment after the completion of acute treatment and administered for at least 6 months.
Statistical Analysis
Statistical analysis was performed with software (Epi Info;
Centers for Disease Control and Prevention [in association with the
World Health Organization]; Atlanta, GA). Categorical data were
analyzed using a 
2 or Fisher exact test, and the
unpaired Student t test was used for continuous variables
(age and follow-up). A p value < 0.05 was considered significant.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gender and Age
Gender and age were cited in 119 cases; 80 patients (67.2%) were
men, and 39 patients (32.8%) were women (age range, 1 week to 84
years; mean age, 40.3 years; median age, 44 years). When patients with
PVE and native valve endocarditis (NVE) were compared, there was a
difference in age; the mean and median ages were 49.7 years and 53
years for patients with PVE, and 34.1 years and 33 years for patients
with NVE, respectively (p < 0.001, t test).
Predisposing Factors
Predisposing factors were listed in 148 cases out of the
total of 152. The most common predisposing conditions (Table 1
) included underlying anatomical cardiac conditions in 70 patients
(47.3%), prosthetic cardiac devices in 66 patients (44.6%), central
venous catheters in 45 patients (30.4%), and previous antibiotic
use in 30 patients (20.3%). A variety of other predisposing conditions
were less frequently seen. In four cases, there was no mention of
predisposing factors. Five patients (3.4%) had no predisposing factor
identified.
|
Clinical and Echocardiographic Findings
Previously unrecognized or changing heart murmurs were described
in 24 of 31 patients in whom there was mention of auscultatory findings
at hospital admission. Echocardiograms were performed in 102 patients
and not performed in 5 patients. There was no mention of
echocardiography in 45 patients. Echocardiography demonstrated
vegetations in 83 of 102 patients (81.4%); vegetations were described
as large in 42 patients (50.6%) and mobile in 16 patients (19.3%).
Myocardial abscess (5.9%) was infrequently seen on echocardiographic
examination. At least 44 of 102 patients underwent transesophageal
echocardiography (TEE). The type of echocardiographic procedure was not
described in 32 cases. Three patients had normal transthoracic
echocardiographic results but did not undergo TEE. Both transthoracic
echocardiography (TTE) and TEE results were described in 16 patients,
and in only one patient was a vegetation seen on TEE that was not
visualized on transthoracic study. In that patient, the vegetation was
attached to the mitral valve, a well-known anatomic problem area for
TTE sensitivity.
TTE identified 88.9% of vegetations in patients with NVE and 76.5% in of vegetations in patients with PVE. TEE identified 92.0% of vegetations in patients with NVE and 61.1% of vegetations in patients with PVE. Large vegetations were noted in 35 cases of NVE compared with 7 cases of PVE (p = 0.03, Mantel-Haenszel test).
Microbiology
Table 2
summarizes the types of fungi and the frequency with which they were
identified. Yeasts caused 101 infections, and Candida species were
recovered in 95 patients (94.1%); in 2 of the 95 cases, polymicrobial
IE was caused by two different Candida species. In one other case,
Candida parapsilosis was isolated with a mold,
Fusarium solani. Candida albicans was the pathogen in 46
cases (45.5%).
|
The distribution of yeasts and molds causing prosthetic valve vs native valve infections was not statistically different (data not shown). Similar findings were seen for PVE differentiated into early and late categories.
Almost one half of identified pathogens (46.5%) were recovered in blood cultures; 25.2% were recovered from an intracardiac site, and 28.3% were recovered from both blood cultures and intracardiac sites. The site of microorganism isolation was not mentioned in 24 cases. Positive blood culture results were more frequent in yeast-related IE cases than in mold-related cases (81.2% vs 30.8%; p < 0.001, Mantel-Haenszel test).
Diagnosis
Clinical and microbiologic features of each case were analyzed
according to the Duke criteria. IE cases were defined as either
definite (n = 120) or possible (n = 23) in 143 patients. The IE
category was not specified in the remaining nine
cases,12
88
although the Duke criteria were used to define
these cases, and they were not rejected. Among all cases classified as
definite, 89.2% were based on pathologic criteria.
The interval between valve replacement and onset of infection was available in only 54 cases and ranged from 1 day to 24 years. Twenty-six cases (45.6%) and 31 cases (54.4%) were classified as early PVE and late PVE, respectively; 9 cases could not be classified because of lack of information. There was no statistical difference in the age distributions for the two groups (p = 0.35, t test). Twenty-two patients had mechanical prostheses, 23 patients had biological prostheses, and in 17 patients did not have a specified prosthesis. One patient had an intra-atrial pacemaker, and three patients had cardiac patches, two of which were Gore-Tex.
Complications
Complications were reported in 100 patients (Table 3
). Embolic phenomena were most commonly seen and described in 61
patients. Congestive heart failure and sepsis, the next most common
complications, were seen in 16 patients each. Complications were not
described in 52 patients. Seventy-four patients (48.7%) had
complicated IE, with a similar distribution among yeast endocarditis
and mold endocarditis cases and among NVE and PVE patients.
|
Long-term (
6 months) suppressive therapy was administered to 13
patients after acute treatment. Eleven patients and 2 patients had FE
caused by yeasts and molds, respectively. Eight patients with yeast
infections were administered fluconazole; the remaining three patients
received itraconazole, ketoconazole, or amphotericin B. Itraconazole
was administered to the two patients with FE caused by molds
(Aspergillus niger in both cases). In five other patients,
suppressive therapy was administered but no specific duration of
therapy was mentioned. Instead, the duration of therapy was labeled as
lifelong, prolonged, or long-term.
Surgical intervention was common. Seventy-eight of 119 patients (65.5%) had surgery. No data regarding surgery were available for 33 patients. Intracardiac surgery was the most common type of surgical intervention and was performed in 63 patients. Embolectomy was performed in four patients, and both intracardiac surgery and embolectomy were performed in seven cases.
Among PVE patients, 79.6% patients underwent surgery, whereas only 53.8% of NVE patients underwent surgery (p = 0.03, Mantel-Haenszel test). In addition, surgery was performed more frequently among late PVE than early PVE patients (89.6% vs 65.2%; p = 0.003, Mantel-Haenszel test). Both medical and surgical therapy were used in 83.3% of complicated FE cases and in 54.5% of uncomplicated FE cases (p = 0.002, Mantel-Haenszel test).
A limited number of cases included a gross description of vegetations seen intraoperatively or at postmortem examination. In 17 of 28 cases (60.7%) in which comments were included, the vegetations were described as large. Fungi were seen on histopathologic examination of resected cardiac and/or embolic tissue in 46 of 49 cases (93.9%) that included this information.
Outcome
The mortality rate among 122 patients with IE was 56.6%; outcome
data were not described for 30 patients. The ages of patients who
survived did not differ statistically from those of patients who died
(p = 0.50, t test).
The mortality rate among patients with mold IE was higher than that of patients with yeast IE (82.1% vs 40.3%; p < 0.001, Mantel-Haenszel test). The mortality rates among PVE and NVE cases, stratified by age and microorganism (yeast or mold), were not different (50.0% vs 66.1%; p = 0.23, Mantel-Haenszel test).
Fifty percent of patients who received antifungal therapy died. All seven patients who did not receive antifungal therapy died (p = 0.01, Mantel-Haenszel test).
There was a trend toward higher survival for patients with yeast-related endocarditis who had undergone valve surgery (68.3% of patients who underwent surgery survived, compared with 52.2% of patients who did not have surgery; p = 0.20, Mantel-Haenszel test). The mortality rates for patients who received medical therapy and for patients with combined medical-surgical intervention were both approximately 50%. The mortality rates did not differ for complicated and uncomplicated FE cases, regardless of whether the patients underwent medical or combined medical and surgical therapy.
Follow-Up
Follow-up was mentioned in 36 of 53 patients who survived (range,
42 days to 17 years). Follow-up was longer for PVE cases (mean and
median of 3.6 years and 1.5 years, respectively) than for NVE cases
(mean and median of 1.2 years and 1 years, respectively) [p = 0.03,
t test]. The period of follow-up did not differ for yeast
IE cases vs mold IE cases, for patients who did or did not undergo
surgery, or for patients who did or did not receive lifelong
suppressive therapy.
Relapse occurred in five patients (two yeast-related and three mold-related) after a variable period ranging from 1 week to 7 months after initial response and completion of acute therapy. In one patient, two relapses caused by C albicans occurred in NVE after acute medical therapy with fluconazole was discontinued. After the second relapse and acute treatment, 17 months of suppressive therapy with fluconazole was administered; no additional relapse was noted during a 34-month follow-up. In the other four patients, one relapse was diagnosed. One patient received suppressive therapy, underwent no surgery, and relapse occurred just after suppressive therapy was stopped. In the other three patients, no suppressive therapy was administered, and relapse occurred after valve replacement.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FE involves a younger population. The mean age of adult patients in one review97 was 44.3 years. In the current review, which included both children and adults, the mean and median ages were 40.3 years and 44 years, respectively.
In past years, injection drug use was considered a prominent risk factor for the development of FE.98 In both the review by Ellis and colleagues97 and in the current review, injection drug use was described in a minority (13% and 4.1%, respectively) of patients. In contrast, therapeutic interventions, including intravascular catheter use, valve surgery, immunosuppressive treatment, and broad-spectrum antibiotic use, were much more prevalent as risk factors for the subsequent development of FE.
Several factors positively and negatively impact our ability to diagnose FE. These factors are important because, in years past, the diagnosis was not made until postmortem examination for a sizable portion of patients. The rarity of the syndrome coupled with negative blood culture results in most mold cases and some yeast cases compromised our ability to secure an early diagnosis. However, the large vegetations that characterize this illness seem to increase the sensitivity of TTE and should prompt a consideration of FE in the setting of (blood) culture-negative endocarditis.
Establishing a definitive diagnosis of infective endocarditis is frequently a problem in which blood culture results are negative and histopathologic evidence of endocardial infection is not available. This could impact the findings of a literature review if the cases are not valid. Clinical criteria have been established to satisfy a case definition of infective endocarditis in a setting in which histopathologic evidence is not available, and these criteria have enjoyed widespread application. In the current literature review, Duke clinical criteria were infrequently needed to secure a diagnosis of FE because histopathologic evidence was obtained in 34 of 37 mold cases (91.9%) and in 67 of 77 yeast cases (87.0%). The routine availability of cardiac and/or embolic tissue for histopathologic examination and culture was, in part, due to the generally accepted doctrine that patients with proven or suspected FE undergo valve replacement in combination with medical therapy for attempted cure.
Despite a recognized dismal outcome in patients with mold endocarditis and a less-than-acceptable outcome in patients with yeast endocarditis, the choice of medical therapy has not changed in decades. Amphotericin B, as exemplified in the current literature review, remains the mainstay of medical therapy. Histopathologic and microbiologic findings of resected intracardiac tissue indicate that amphotericin B is slow to clear fungi. In one extreme example,99 Candida species could be cultured from resected valvular tissue after 190 days of amphotericin B. Mortality caused by Aspergillus endocarditis in both recent literature reviews is > 90%, and surgical intervention with valve replacement did not improve mortality rates as compared with rates for patients who received antifungal therapy alone.
It is less clear whether the tenet that FE is a stand-alone indication
for valve surgery in cases of Candida-related endocarditis is correct.
Both literature surveys indicate that there may be a survival benefit
to valve replacement, but they should not be overinterpreted because
the data were collected retrospectively from a series of case reports.
Treatment bias is inevitable in this setting because patients who are
more ill are often deemed nonsurgical candidates and suffer increased
mortality. Data from another investigation100
suggested
that surgical intervention did not improve outcome over that of medical
therapy alone in patients with uncomplicated Candida prosthetic valve
endocarditis. There are, however, potential weaknesses in the
interpretation of these data. First, the number of patients in both
treatment groups was small (n = 10 and n = 3, respectively).
Second, follow-up data were not available in 5 of the 13 cases; in 3
other cases, the duration of follow-up was 
1 year. Thus, late
relapses of FE may have been missed. Third, at least four patients
received long-term antifungal suppressive therapy that could have
prevented FE relapse and death. Considering all of the clinical
information to date, valve replacement should probably be offered with
aggressive medical therapy until prospectively collected data are
generated that direct clinicians to do otherwise.
Relapsing FE is a complication seen in as many as 30 to 40% of patients who have FE develop and who survive to complete short-term therapy.6 9 Relapsing FE was less often seen (3.3%) in the current literature survey for at least two reasons. First, follow-up information was limited. Only 36 of the 53 patients who survived had information regarding follow-up, and for those patients listed, only 12 patients had follow-up of > 2 years. It is well recognized that relapses can occur late in FE cases and that short-term (< 2 years) follow-up will miss episodes of late relapse. Second, the newly adopted long-term suppressive therapy100 101 102 was administered to 13 patients and likely prevented relapsing FE that would have occurred without the use of suppressive treatment.
Long-term suppressive therapy for FE is being used in two clinical scenarios. In one, the patient is deemed a nonsurgical candidate, usually for medical or surgically technical reasons, and suppressive therapy is administered after the patient’s response to acute antifungal treatment. Because the patient is not considered curable without valve replacement, long-term (life-long) suppressive therapy is given. In the second scenario, long-term suppressive therapy is administered to patients who undergo valve replacement and acute treatment. Because the relapse rate is high and cure with suppressive treatment may not be achievable,101 some cardiothoracic surgeons advocate long-term (life-long) suppressive antifungal treatment. Thus, potentially all patients who survive long enough to complete acute treatment, which can include valve replacement, may be considered candidates for chronic suppressive antifungal therapy.
FE has gained more attention in the recent literature. With our continued expansion of medical and surgical techniques, it is expected that an increasing number of these cases will occur. Advances in therapy are also needed to diminish the high mortality rate that currently characterizes FE.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Received for publication July 12, 2001. Accepted for publication September 10, 2001.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Vassiloyanakopoulos, M. E. Falagas, M. Allamani, and A. Michalopoulos Aspergillus fumigatus tricuspid native valve endocarditis in a non-intravenous drug user. J. Med. Microbiol., May 1, 2006; 55(Pt 5): 635 - 638. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. B. Luciani, G. Casali, F. Viscardi, S. Marcora, M. A. Prioli, and A. Mazzucco Tricuspid Valve Repair in an Infant With Multiple Obstructive Candida Mycetomas Ann. Thorac. Surg., December 1, 2005; 80(6): 2378 - 2381. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. El-Hamamsy, N. Durrleman, L.-M. Stevens, L. P. Perrault, and M. Carrier Aspergillus Endocarditis After Cardiac Surgery Ann. Thorac. Surg., July 1, 2005; 80(1): 359 - 364. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Baddour, W. R. Wilson, A. S. Bayer, V. G. Fowler Jr, A. F. Bolger, M. E. Levison, P. Ferrieri, M. A. Gerber, L. Y. Tani, M. H. Gewitz, et al. Infective Endocarditis: Diagnosis, Antimicrobial Therapy, and Management of Complications: A Statement for Healthcare Professionals From the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Cardiovascular Disease in the Young, and the Councils on Clinical Cardiology, Stroke, and Cardiovascular Surgery and Anesthesia, American Heart Association: Endorsed by the Infectious Diseases Society of America Circulation, June 14, 2005; 111(23): e394 - e434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. G. Leshnower and T. G. Gleason Reoperative innominate arterial, ascending aortic, and root replacement for extensive fungal endocarditis J. Thorac. Cardiovasc. Surg., April 1, 2005; 129(4): 941 - 942. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Challa, A. K. Prayaga, L. Vemu, J. Sadasivan, M. K. M. Jagarlapudi, R. Digumarti, and R. Prabhala Fungal Endocarditis: An Autopsy Study Asian Cardiovasc Thorac Ann, June 1, 2004; 12(2): 95 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. El-Hamamsy, N. Durrleman, L.-M. Stevens, R. Cartier, M. Pellerin, L. P. Perrault, and M. Carrier A cluster of cases of Aspergillus endocarditis after cardiac surgery Ann. Thorac. Surg., June 1, 2004; 77(6): 2184 - 2186. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
