Fungalinfections in The Icu: Marya D. Zilberberg,, Andrew F. Shorr
Fungalinfections in The Icu: Marya D. Zilberberg,, Andrew F. Shorr
Fungalinfections in The Icu: Marya D. Zilberberg,, Andrew F. Shorr
in the ICU
Marya D. Zilberberg, MD, MPHa,b,*, Andrew F. Shorr, MD, MPHc,d
KEYWORDS
Critical illness ICU Fungal infection Candida
Aspergillus Invasive
Pulmonologists and intensivists often care for patients at risk for infections caused by
both Aspergillus and Candida. Infection with either can lead to severe life- threatening
disease. Mortality rates for invasive fungal disease often exceed 30%. Furthermore,
immunosuppressed patients are at increased risk for disease caused by both patho-
gens. despite these similarities, however, important differences exist between infec-
tions caused by mold as compared with those related to yeast. For mold, the lungs
(in addition to the central nervous system) remain the key affected organ system, while
yeast rarely are implicated as a cause of pneumonia. Conversely, colonization by
Aspergillus has clearly different implications for the patient than does colonization
with Candida. For both organisms, however, multiple diagnostic challenges remain.
Fortunately, therapeutic paradigms are shifting, and clinicians have many new agents
in their armamentarium for combating fungal infection. Given the rapidly changing
literature in this broad area, it is imperative that physicians caring for the immunosup-
pressed patient and for the critically ill remain abreast of this evolving field.
a
Department of Health Policy and Management, School of Public Health and Health Sciences,
University of Massachusetts, Amherst, MA 01003, USA
b
EviMed Research Group, LLC, Post Office Box 303, Goshen, MA 01032, USA
c
Division of Pulmonary and Critical Care Medicine, Room 2a_68, Washington Hospital Center,
110 Irving Street, Northwest, Washington, DC 20010, USA
d
Georgetown University School of Medicine, Washington, DC 20010, USA
* Corresponding author. EviMed Research Group, LLC, Post Office Box 303, Goshen, MA 01032.
E-mail address: marya@evimedgroup.org (M.D. Zilberberg).
formation, and chronic necrotizing aspergillosis (CNA). In these three conditions, the
immune system is generally intact, and in select syndromes such as ABPA, it is the
host response to infection that leads to a clinical syndrome. Invariably these three
conditions are associated with local impairments of the intrapulmonary immune
system as a result of local airway damage and destruction. Unless leading to hemop-
tysis that compromises the airway, patients who have these non-IPA syndromes
usually do not require treatment in the ICU. Because of the disproportionate burden
of IPA, the remainder of this discussion will focus on IPA.
Epidemiologically, the true incidence of IPA is unclear. Although certainly less
common than infection with yeast, the frequency of IPA may be increasing. For
example, Groll and colleagues1 in examining autopsy data between 1978 and 1992
noted a rise in prevalence of all invasive mycoses from 0.4% to 3.1%, and Aspergillus
as the causative organism increased from 17% to 60% over the same time period.
Certainly one factor driving rates of IPA is the broadening use of immunosuppressive
agents. Both the types of and duration of immunosuppression related to all forms of
transplant are evolving. Attendant with more extensive immunosuppression is
a greater risk for IPA. Not all forms of immunosuppression appear to carry similar risks
for IPA. Both the extent and duration of neutropenia represent long recognized risk
factors for IPA. More recently, use of select monoclonal antibody strategies for induc-
tion of immunosuppression has been thought to heighten the chance for IPA. In hema-
topoietic stem-cell transplant (HSCT) subjects, the development of graft versus host
disease has been linked to a greater frequency of invasive mold infections. Following
any type of transplant, cytomegalovirus (CMV) disease also amplifies the potential for
IPA.
Recently, Patterson and Singh described the disproportionate distribution of IPA as
a function of the type of transplant.2 In a systematic review of over 20,000 cases of IPA,
they reported that the incidence of IPA was greatest among those undergoing lung
transplant, with 8.4% developing IPA.2 Patients receiving kidney transplants face
a substantially lower risk for IPA (0.7%).2 The rate of IPA following either HSCT or heart
transplant was approximately 6.0%. In part, this difference related to transplant type
reflects both the difference in the agents used for immunosuppression (eg, renal trans-
plant patients require much less immunosuppression) and also the fact that Aspergillus
spores are inhaled. Hence, the lungs potentially are exposed to an excessive inoculum.
The impact of IPA on post-transplant mortality was staggering: anywhere from 10% to
17% if all post-transplant deaths were attributed to IPA.2
More recent analyses suggest that the epidemiology of IPA in the ICU may be shift-
ing away from those traditionally considered at risk. Several recent case series have
described IPA in nonimmunocompromised critically ill subjects. Samarakoon and
Soubani reported five cases of IPA among chronic obstructive pulmonary disease
(COPD) patients from their medical center and performed a systematic review of liter-
ature on this topic.3 These investigators found that among the 65 cases examined,
clinical presentation was largely nonspecific, and in most (63%), radiological findings
consisted of infiltrates. Although in 15 patients (23%) diagnosis was made with a bron-
choalveolar lavage (BAL), in a plurality of cases (n528, 43%) Aspergillus was recov-
ered at autopsy. Despite treatment with antifungal agents in most cases, the
mortality rate was 91%. The authors concluded that the likely risk factors in this cohort
of COPD patients for IPA included chronic treatment with corticosteroids and
advanced severe COPD.
Similarly, Vandewoude and colleagues4 described 38 individuals (incidence 4 out of
1000 ICU admissions) who had IPA cared for between 1997 and 1999. Only 17 of them
(45%) had traditional risk factors, while among the remaining 21 without evidence of
Fungal Infections in the ICU 627
immune compromise, invasive disease was a complication of their critical illness.4 The
authors noted that while in the presence of risk factors the actual mortality was similar
to that predicted by Acute Physiology and Chronic Health Evaluation (APACHE) II, in
those lacking the traditional risk factors for IPA, actual mortality exceeded that pre-
dicted by APACHE.4 The same investigators in a case–control analysis calculated
that the attributable mortality of IPA in the ICU approached 20%, finding the presence
of this disease to be an independent predictor of hospital mortality among the critically
ill.5 Finally, Meerssemann and colleagues6 described a 127-patient cohort of ICU
patients who had IPA, of whom 89 did not have an underlying malignancy. Of these
89, 35 had COPD; nine had undergone a solid organ transplantation, and 17 were
on immunosuppressive therapy. Observed mortality in this group (80%) was higher
than that predicted by the Simplified Acute Physiology Score (SAPS) II (48%). Of
particular interest, were five critically ill patients without any predisposition but with
proven IPA. In an additional analysis addressing only critically ill patients colonized
with Aspergillus (n 5 172), Vandewoude and colleagues7 concluded that nearly
50% of such subjects had invasive disease and not simply colonization. Again, despite
the absence of typical risk factors for IPA, they detected the syndrome in patients,
where before one might never have considered it a clinical possibility. Although
thought-provoking, each of these analyses has major limitations because of essen-
tially retrospective, single-center designs. Nonetheless, these findings, taken as
a whole, suggest that the epidemiology of IPA in the ICU is shifting. In response, physi-
cians need to remain vigilant and consider this syndrome in patients not significantly
immunocompromised.
Diagnostically, IPA remains a challenge. Initially, the clinical manifestation of IPA may
be nonspecific. As many as 30% of patients may be asymptomatic. Nonspecific symp-
toms such as dyspnea and fever are common. Radiographic manifestations of IPA also
may be diverse. The classic description of cavitation associated with IPA is a late finding,
and the absence of cavitation does not preclude a diagnosis of IPA. In some cases, the
disease begins as small nodules that eventually coalesce. Alternatively, IPA can lead to
peripheral, wedge-shaped opacities that resemble those seen after pulmonary infarc-
tion. The air crescent sign is a late finding in IPA and is caused by contraction of infarcted
tissue surrounding the site of infection. Conversely, the halo sign, an area of low atten-
uation near the primary lesion, is an early finding that often is transient. These two signs
are somewhat specific for invasive aspergillosis but lack sensitivity.8
To facilitate the diagnosis of IPA, two strategies have been proposed. The first
addresses the risk for IPA based on airway colonization with Aspergillus. In many
syndromes, colonization with a pathogen does not increase the subsequent risk for
infection necessarily. With respect to IPA, this appears not to be the case. In the review
by Paterson and Singh2 described earlier, the authors estimated that colonization with
Aspergillus correlated with an eventual risk for IPA ranging from 16% to 80%, depend-
ing on the type of transplantation. HSCT subjects faced a greater risk for IPA if they
were known to be colonized (60% to 80%), while in lung recipients, a positive airway
culture for Aspergillus conveyed a less than 20% chance of eventual IPA.2 In unse-
lected ICU patients, recovery of Aspergillus in apparently noninfected patients may
also represent a harbinger of IPA. For example, Vandewoude and colleagues7 per-
formed a large cohort study of 172 patients (incidence 6.8/1000 ICU admissions)
with a positive sputum culture for Aspergillus. Based on a predefined algorithm
involving clinical presentation, radiographic signs, and BAL cultures, 83 cases were
deemed to be IPA, while the remaining 89 were thought to be colonized with the
organism. The algorithm was found to have good positive and negative predictive
values for invasive disease in this population of patients.7
628 Zilberberg & Shorr
immunocompromised patients who had proven invasive disease, 201 subjects were
randomized to either a standard (3 mg/kg) or high (10 mg/kg) daily dose of L-AMB
for 14 days followed by 3 mg/kg/d.21 Although efficacy outcomes were similar in
both groups, the high-dose group experienced higher rates of toxicity than observed
with the standard dose.21 Caspofungin, a member of the echinocandin family, repre-
sents an option for IA treatment among those who have refractory disease or for
people unable to tolerate AMB or triazoles. Maertens and colleagues,22 in an open-
label noncomparative multicenter clinical trial in patients who had probable or proven
IA intolerant of treatment with AMBs or triazoles, investigated the use of caspofungin
at the dose of 50 mg daily following a 70 mg loading dose on day 1. A complete or
partial clinical resolution was the primary outcome of the study. Among the 83 effi-
cacy-evaluable patients, 45% responded favorably, and only two patients were forced
to discontinue caspofungin because of drug-related toxicity. The authors concluded
that caspofungin is a viable salvage treatment for IA, and it is recommended as
such by the IDSA.16,22
Before trials with voriconazole, itraconazole, an older azole with activity against
Aspergillus, was studied for IPA. Caillot and colleagues23 reported the results of a trial
of intravenous itraconazole followed by oral treatment for probable or definite IPA in 31
immunocompromised patients. In this single-arm open-label multicenter study, the
primary outcome of complete or partial clinical resolution was met in 18 (58%) of
the subjects, in the face of mostly mild-to-moderate toxicities.23 The largest study
of invasive Aspergillus was conducted by Patterson and colleagues24 in North America
and Western Europe and represented a naturalistic observational analysis examining
case records of 595 patients who had IA, most of whom (56%) had IPA. This study was
completed before the commercial availability of voriconazole. When looking at treat-
ment regimens, 187 were treated with AMB alone, 58 with itraconazole alone, and
93 with a combination of the two (the remaining patients were treated with other
various regimens).24 Itraconazole resulted in a 57% rate of combined complete and
partial response, similar to the combination therapy (54%) and higher than AMB alone
(32%).24 Although it boasts the largest study sample to date, rigorous randomized
trials are lacking to compare the treatment of IA with itraconazole with either AMB
or voriconazole. If using itraconazole, serum level measurements are needed to docu-
ment adequate drug absorption.16 Posaconazole is an additional extended- spectrum
triazole available in the European Union for treating refractory disease. This monother-
apy was investigated in an open-label prospective trial in 107 patients, the results of
which were compared with 86 retrospectively collected historic controls.25 Complete
or partial response was achieved in 42% of posaconazole-treated patients versus only
26% of those on the comparator regimen.25 In a logistic regression model, the authors
noted that the adjusted odds ratio (OR) of response was 4.06 (95% CI, 1.50 to 11.04)
for posaconazole over other salvage therapies.25
The final drug recommended as salvage therapy for IA is micafungin, a newer echi-
nocandin.16 Although a treatment dose has not been conclusively established, two
trials indicate its efficacy in probable or definite IA. In a 225-patient open-label single-
arm study, 35.6% of all patients treated with micafungin achieved a response, with the
rates even higher in the primary (50%) and salvage (41%) groups.26 An additional
open-label study investigated the efficacy of micafungin alone or in combination for
treating refractory or intolerant patients.27 Of the 98 patients enrolled in the study,
eight were treated singly with micafungin, of whom three (38%) achieved a response.27
Clearly, the experience with micafungin is limited and not controlled well enough to
render it a strongly recommended therapy. Similarly, because very little evidence
exists for the efficacy and safety of combination regimens, these are not
630 Zilberberg & Shorr
Risk Factors
Before considering treatment issues, clinicians must appreciate the risk factors for
fungal BSIs. Through understanding the issues and variables that heighten the poten-
tial for candidemia, one can focus on efforts not only at prevention but also on prompt
diagnosis and treatment. Common risk factors predisposing to candidemia include
candidal colonization, prior exposure to antibiotics, renal failure, presence of a CVC,
and need for total parenteral nutrition (TPN).48–51 In a case–control study, Wey and
colleagues48 established that the number of antibiotics administered, candidal coloni-
zation, hemodialysis, and use of a Hickman catheter were strongly associated with the
development of candidemia. A group of Swiss investigators led by Pittet in a study of
29 surgical ICU patients confirmed colonization to be a risk factor for candidal
632 Zilberberg & Shorr
to allow for their broad application in identifying patients who might merit from either
prophylaxis or presumptive therapy.
Because of frustration with risk stratification and diagnostic modalities, efforts
recently have focused on developing surrogate markers to detect the presence of fun-
gemia. Moreover, the sensitivity of blood cultures for this pathogen varies from 8% to
82%, depending on the population in question.56 Additionally, cultures may take
several days to become positive.56 Similar to research addressing galctomannan for
Aspergillus, the candidal cell wall has many components that can be detected with
modern assays. The most promising target appears to be b-D-glucan. Unfortunately,
diagnosis of candidemia remains a challenge. The b-D-glucan test has relatively high
sensitivity, specificity, and positive and negative predictive values, although data in
the ICU are lacking.57 The mostly small studies conducted to date mainly have
enrolled severely immunosuppressed patients. Molecular testing to detect Candida
DNA is in early development, and it is not clear how useful it will be among the critically
ill.57 An additional challenge in the current environment is the proliferation of azole-
resistant non-albicans species.
treatment delay on mortality among 157 consecutive patients with Candida blood
stream infection. In this study most of the isolates (53.5%) were C albicans, and the
cumulative proportion of C glabrata and C krusei was 14%. The investigators found
that only 5 of the 157 patients received prompt, appropriate empiric therapy for can-
didemia. In a multivariate analysis, they reported that a delay of as little as 12 hours
from when the eventually positive blood cultures were drawn more than doubled
the risk for death (OR 2.09, 95% CI, 1.53 to 2.84).62 Readers should note that the delay
was defined, not as time from when the laboratory contacted the provider and re-
ported yeast in the cultures, but the time from which the initial evaluation for infection
commenced. Garey and colleagues63 confirmed the Morrel group’s observation. In
this multicenter cohort, investigators noted increasing mortality based on the delay,
measured in days from when the culture was drawn to antifungal treatment. Similarly,
a recent study by Labelle and colleagues64 expanded the definition of inappropriate
therapy for fungemia to encompass more than just the time to treatment. In their anal-
ysis, they also defined the inadequate dosing of fluconazole as inappropriate. This,
along with a delay in therapy independently raised the risk of death ninefold.64 All
four of these studies underscore the importance of prompt appropriate treatment of
candidemia to impact survival. They also reveal the need for an emphasis on preven-
tion and an urgent need for rapid diagnostic strategies.
Embracing the need to be prompt in treatment of suspected candidemia should not
lead intensivists to endorse presumptive therapy. In this case, presumptive therapy
represents the administration of antifungals purely based on the presence of several
risk factors for candidal infection and the development of a new fever. This approach
represents a standard strategy for prolonged neutropenic fever. For critically ill
subjects, however, a large multicenter, randomized trial demonstrated the futility of
this paradigm.65 In this study, ICU patients who had persistent fever were assigned
randomly to empiric fluconazole or placebo. To enrich the population as to their risk
for fungemia, the protocol required patients to concurrently be receiving broad-spec-
trum antibiotics and have CVCs in place. Additionally, most subjects had been under-
going MV for several days. Overall, outcomes including mortality and fever resolution
were similar.65 The main reason for these negative findings was that the rate of even-
tually documented fungemia was low; approximately 7% of subjects subsequently
were diagnosed with a candidal BSI. This fact underscores the need for better risk
stratification schemes. Also, the list of causes of fever in the ICU is extensive. Because
the investigators did not require an effort to exclude other possible causes for fever,
such as the presence of a new pneumonia, one limitation of the trial may have been
that the entry criteria simply did not reflect the way clinical decision making takes
place at the bedside in the ICU.65 In other words, the potential for fungemia is not
simply a function of the number of risk factors present, but more properly reflects
whether an alternative diagnosis for the nonspecific signs and symptoms that can
be seen in with this condition is more or less likely.
Treatment
A key step in the approach to fungemia, irrespective of the antifungal used, remains
management of the CVC. In many instances, the CVC represents the portal of entry
for the fungus into the patient. Alternatively, the CVC may become seeded as
a secondary point in the disease process. Current guidelines strongly advocate
prompt CVC removal.66 This in part is based on the observation that the biofilm
evolving on a CVC represents a relatively protected site for fungus. Older-generation
antifungals (eg, fluconazole) penetrate biofilm very poorly. Moreover, older analyses
have identified that CVC retention may be an independent predictor of mortality.67,68
Fungal Infections in the ICU 635
Some have criticized this approach of uniform CVC removal, because the initial
reports addressing CVC removal failed to address confounding issues such as
severity of illness and why the CVC was not removed.69 Patients in whom the CVC
is not removed in fact may be described better as subjects for whom the CVC cannot
be removed given comorbid issues, the need for continued central access, and risks
associated with replacing the CVC. For example, Rodriguez and colleagues70 deter-
mined that severity of illness was the most important predictor of mortality in candi-
demic patients and that assessments of timing of removal were confounded by this
overriding issue. They concluded that decisions regarding CVC removal must be indi-
vidualized. Similarly, in a qualitative review of the topic, Pasqualotto and Severo
observed that many analyses were too flawed to allow one to draw meaningful conclu-
sions.71 They advocated for a randomized trial to determine appropriate practice.
More careful and up-to-date analyses, however, clearly show the risks related to
failing to remove the CVC in persons with fungemia. Raad and colleagues72 explored
predictors of mortality in over 400 patients with cancer and candidemia. After adjust-
ing for severity of illness, immune system status, and the source of the candidemia,
failure to remove the CVC increased the probability of death more than fourfold. Label-
le and colleagues64 reached similar conclusions. In addition to taking many variables
into consideration that included severity of illness, comorbid illness, and most impor-
tantly the timing and appropriateness of antifungal treatment, they observed that cath-
eter retention independently heightened the risk for death more than sixfold.64
The past decade has witnessed a proliferation of commercially available antifungals.
At present, the clinician must consider various forms of amphotericin, azoles, and
echinocandins. For many years, amphotericin (AMB) represented the standard of
care for candidemia. Although the utility of AMB is limited by its multiple toxicities,
the newer liposomal AMB (L-AMB) formulations are tolerated better. In general, the
L-AMB options lead to less nausea, vomiting, and fever. They clearly cause less neph-
rotoxicity. Unfortunately, L-AMB preparations still may result in substantial potassium
wasting as is seen with AMB deoxycholate. Clinical trials document that L-AMB has
similar but not better, efficacy relative to conventional AMB preparations.73
Fluconazole represents a well-tolerated alternative for candidemia. In a randomized
trial, fluconazole was shown to result in similar outcomes when compared with AMB.74
In this study, evaluating 206 non-neutropenic patients evenly divided between the two
treatment regimens, mortality rates with fluconazole and AMB were 33% and 40%,
respectively. This difference was not statistically different. Notably, the rates of blood
culture clearance were also similar. The authors concluded that in patients without
neutropenia, fluconazole and AMB had similar effectiveness.74 A similar trial from
Canada confirmed these findings.75 One major limitation of studies employing fluco-
nazole is that they were completed in an era when the prevalence of C glabrata and
C krusei were much lower. For example, in the initial report by Rex and colleagues,74
these pathogens accounted for less than 15% of all isolates, while in modern settings,
the proportion of candidal BSIs caused by these yeast species may be increasing.
Furthermore, dosing of fluconazole can be confusing. Current guidelines recommend
treatment with a greater dose of fluconazole. This theoretically allows one to poten-
tially overcome the dose-dependent resistance seen in some candidal isolates. Unfor-
tunately, observational analyses document that clinicians often underdose this
agent.64
Voriconazole, a triazole with excellent activity against Aspergillus, also has been
examined as an alternative to AMB in non-neutropenic patients who have candide-
mia.76 In a randomized, noninferiority trial enrolling over 400 patients, voriconazole
was found to be similar in effectiveness to AMB.76 Interestingly, despite more adverse
636 Zilberberg & Shorr
daily, or micafungin 150 mg daily. Outcomes were the same with all three strategies. In
general, it therefore appears that the 100 mg dose of micafungin is adequate. None-
theless, clinicians must note that among the patients given either dose of micafungin,
the number of infections caused by C glabrata and C krusei was quite small.82
Anidulafungin is the only echinocandin to be compared directly with fluconazole.
Reboli and colleagues83 randomized 245 subjects to either anidulafungin or flucona-
zole. As with other studies, most patients suffered from candidemia, and few were
neutropenic. Because of the innate resistance of C krusei to fluconazole, patients
found to be infected with this pathogen were dropped from the trial. Overall, 40% of
infections were caused by non-albicans species, including C glabrata. Clinical
success rates were significantly higher at end of therapy in patients treated with ani-
dulafungin. This 15.4% difference in response rates did not arise because of flucona-
zole resistance, as few isolates displayed this. Anidulafungin, however, did not lead to
higher survival rates.83 This report has been criticized, because one site in the study
enrolled a disproportionate number of subjects. Additionally, concern has been ex-
pressed about a potential center effect. Statistical tests for an interaction between
study site and outcome did not confirm this to be an issue.
SUMMARY
Mold and yeast infections remain diagnostic and therapeutic challenges. The preva-
lence of both of these types of infections is likely to grow over the next decade. Unfor-
tunately, mortality rates with either process are exceedingly high. Coupled with the
economic burden of these illness, physicians must remain vigilant when approaching
patients at risk for these fungal diseases. Fortunately, newer diagnostic modalities are
being developed and tested. Likely some combination of sero-diagnosis testing along
with clinical risk stratification will evolve in the coming years. For both conditions,
prompt diagnosis and treatment remain the keys to ensuring favorable outcomes.
Newer agents for treatment are now commercially available. The literature on these
therapies, as is the research into many aspects of fungal infection, is changing rapidly.
Hence clinicians caring for the critically ill must strive to remain abreast of this
information.
REFERENCES
1. Groll AH, Shah PM, Mentzel C, et al. Trends in the postmortem epidemiology of
invasive fungal infections at a university hospital. J Infect 1996;33:23–32.
2. Paterson DL, Singh N. Invasive aspergillosis in transplant recipients. Medicine
(Baltimore) 1999;78:123–38.
3. Samarakoon P, Soubani AO. Invasive pulmonary aspergillosis in patients with
COPD: a report of five cases and systematic review of the literature. Chron Respir
Dis 2008;5:19–27.
4. Vandewoude K, Blot S, Benoit D, et al. Invasive aspergillosis in critically ill
patients: analysis of risk factors fro acquisition and mortality. Acta Clin Belg
2004;59:251–7.
5. Vandewoude KH, Blot SI, Benoit D, et al. Invasive aspergillosis in critically ill
patients: attributable mortality and excesses in length of ICU stay and ventilator
dependence. J Hosp Infect 2004;56:269–76.
6. Meersseman W, Vandecasteele SJ, Wilmer A, et al. Invasive aspergillosis in crit-
ically ill patients without malignancy. Am J Respir Crit Care Med 2004;170:621–5.
7. Vandewoude KH, Blot SI, Depuydt P, et al. Clinical relevance of Aspergillus isola-
tion from respiratory tract samples in critically ill patients. Crit Care 2006;10:31.
638 Zilberberg & Shorr
8. Buckingham SJ, Hansell DM. Aspergillus in the lung: diverse and coincident
forms. Eur Radiol 2003;13:1786–800.
9. Maertens J, Verhaegen J, Lagrou K, et al. Screening for circulating galacto-
mannan as a noninvasive diagnostic tool for invasive aspergillosis in prolonged
neutropenic patients and stem cell transplantation recipients: a prospective vali-
dation. Blood 2001;97:1604–10.
10. Sulahian A, Tabouret M, Ribaud P, et al. Comparison of an enzyme immunoassay
and latex agglutination test for detection of galactomannan in the diagnosis of
aspergillosis. Eur J Clin Microbiol Infect Dis 1996;15:139–45.
11. Swanink CM, Meis JF, Rijs AJ, et al. Specificity of a sandwich enzyme-linked
immunosorbent assay for detecting Aspergillus galactomannan. J Clin Microbiol
1997;35:257–60.
12. Pfeiffer CD, Fine JP, Safdar N. Diagnosis of invasive aspergillosis using a galac-
tomannan assay: a meta-analysis. Clin Infect Dis 2006;42:1417–27.
13. Herbrecht R, Letscher-Bru V, Oprea C, et al. Aspergillus galactomannan detec-
tion in the diagnosis of invasive aspergillosis in cancer patients. J Clin Oncol
2002;20:1898–906.
14. Pinel C, Fricker-Hidalgo H, Lebeau B, et al. Detection of circulating Aspergillus
fumigatus galactomannan: value and limits of the Platelia test for diagnosing inva-
sive aspergillosis. J Clin Microbiol 2003;41:2184–6.
15. Husain S, Paterson DL, Studer SM, et al. Aspergillus galactomannan antigen in
the bronchoalveolar lavage fluid for the diagnosis of invasive aspergillosis in
lung transplant recipients. Transplantation 2007;83:1330–6.
16. Walsh TJ, Anaissie EJ, Denning DW, et al. Treatment of aspergillosis: clinical
practice guidelines of the Infectious Diseases Society of America. Clin Infect
Dis 2008;46:327–60.
17. Herbrecht R, Denning DW, Patterson TF, et al. Voriconazole versus amphotericin
B for primary therapy of invasive aspergillosis. N Engl J Med 2002;347:408–15.
18. Hope WW, Billaud EM, Lestner J, et al. Therapeutic drug monitoring for triazoles.
Curr Opin Infect Dis 2008;21:580–6.
19. Bowden R, Chandrasekar P, White MH, et al. A double-blind, randomized,
controlled trial of amphotericin B colloidal dispersion versus amphotericin B for
treatment of invasive aspergillosis in immunocompromised patients. Clin Infect
Dis 2002;35:359–66.
20. Leenders AC, Daenen S, Jansen RLH, et al. Liposomal amphotericin B compared
with amphotericin B deoxycholate in the treatment of documented and suspected
neutropenia-associated invasive fungal infections. Br J Haematol 1998;103:
205–12.
21. Cornley OA, Maertens J, Bresnik M, et al. Liposomal amphotericin B as initial
therapy for invasive mold infection: a randomized trial comparing a high-loading
dose regimen with standard dosing (AmBiLoad trial). Clin Infect Dis 2007;44:
1289–97.
22. Maertens J, Raad I, Petrikkos G, et al. Efficacy and safety of caspofungin for
treatment of invasive aspergillosis in patients refractory to or intolerant of conven-
tional antifungal therapy. Clin Infect Dis 2004;39:1563–71.
23. Caillot D, Bassaris H, McGeer A, et al. Intravenous itraconazole followed by oral
itraconazole I. The treatment of invasive pulmonary aspergillosis in patients with
hematologic malignancies, chronic granulomatous disease, or AIDS. Clin Infect
Dis 2001;33:e83–90.
24. Patterson TF, Kirkpatrick WR, White M, et al. Invasive aspergillosis: disease spec-
trum, treatment practices, and outcomes. Medicine 2000;79:250–60.
Fungal Infections in the ICU 639
25. Walsh TJ, Raad I, Patterson TF, et al. Treatment of invasive aspergillosis with pos-
aconazole in patients who are refractory to or intolerant of conventional therapy:
an externally controlled trial. Clin Infect Dis 2007;44:2–12.
26. Denning DW, Marr KA, Lau WM, et al. Micafungin (FK463), alone or in combina-
tion with other systemic antifungal agents, for the treatment of acute invasive
aspergillosis. J Infect 2006;53:337–49.
27. Kontoyiannis DP, Ratanatharathorn V, Young JA, et al. Micafungin alone or in
combination with other systemic antifungal therapies in hematopoietic stem cell
transplant recipients with invasive aspergillosis. Transpl Infect Dis 2009;11:
89–93.
28. Marr KA, Boeckh M, Carter RA, et al. Combination antifungal therapy for invasive
aspergillosis. Clin Infect Dis 2004;39:797–802.
29. Singh N, Limaye AP, Forrest G, et al. Combination of voriconazole and caspofun-
gin as primary therapy for invasive aspergillosis in solid organ transplant recipi-
ents: a prospective, multicenter, observational study. Transplantation 2006;15:
320–6.
30. Pfaller MA, Diekema DJ. Epidemiology of invasive candidiasis: a persistent
public health problem. Clin Microbiol Rev 2007;20:133–63.
31. Hidron AI, Edwards JR, Patel J, et al. Antimicrobial-resistant pathogens associ-
ated with healthcare-associated infections: annual summary of data reported to
the National Healthcare Safety Network at the Centers for Disease Control and
Prevention, 2006–2007. Infect Control Hosp Epidemiol 2008;29:996–1011.
32. Banerjee SN, Emori TG, Culver DH. Secular trends in nosocomial primary blood-
stream infections in the United States, 1980–1989. National Nosocomial Infec-
tions Surveillance System. Am J Med 1991;91:86S–9S.
33. Fridkin SK, Kaufman D, Edwards JR, et al. Changing incidence of Candida blood-
stream infections among NICU patients in the United States: 1995–2004. Pediat-
rics 2006;117:1680–7.
34. San Miguel LG, Cobo J, Otheo E, et al. Secular trends of candidemia in a large
tertiary-care hospital from 1988 to 2000: emergence of Candida parapsilosis.
Infect Control Hosp Epidemiol 2005;26:548–52.
35. Trick WE, Fridkin SK, Edwards JR, et al. Secular trend of hospital-acquired can-
didemia among intensive care unit patients in the United States during 1989–
1999. Clin Infect Dis 2002;35:627–30.
36. Martin GS, Mannino DM, Eaton S, et al. The epidemiology of sepsis in the United
States from 1979 through 2000. N Engl J Med 2003;348:1546–54.
37. Shorr AF, Shorr AF, Tabak YP, et al. Burden of early-onset candidemia: analysis of
culture-positive bloodstream infections from a large US database. Crit Care Med,
in press.
38. Zilberberg MD, Shorr AF, Kollef MH. Secular trends in candidemia-related hospi-
talizations in the US, 2000–2005. Infect Control Hosp Epidemiol 2008;29:978–80.
39. Sandven P, Bevanger L, Digranes A, et al. Candidemia in Norway (1991 to 2003):
results from a nationwide study. J Clin Microbiol 2006;44:1977–81.
40. Poikonen E, Lyytikamen O, Anttila VJ, et al. Canididemia in Finland, 1995–1999.
Emerg Infect Dis 2003;9:985–90.
41. Smith PB, Morgan J, Benjamin JD, et al. Excess costs of hospital care associated
with neonatal candidemia. Pediatr Infect Dis J 2007;26:197–200.
42. Zaoutis TE, Argon J, Chu J, et al. The epidemiology and attributable outcomes of
candidemia in adults and children hospitalized in the United States: a propensity
analysis. Clin Infect Dis 2005;41:1232–9.
640 Zilberberg & Shorr
43. Snydman DR. Shifting patterns in the epidemiology of nosocomial Candida infec-
tions. Chest 2003;123:500–3.
44. Nguyen MH, Peacock JE Jr, Morris AJ, et al. The changing face of candidemia:
emergence of non-Candida albicans species and antifungal resistance. Am
J Med 1996;100:617–23.
45. Bassetti M, Righi E, Costa A, et al. Epidemiological trends in nosocomial candi-
demia in intensive care. BMC Infect Dis 2006;6:21.
46. Shorr AF, Lazarus DR, Sherner JH, et al. Do clinical features allow for accurate
prediction of fungal pathogenesis in bloodstream infections? Potential implica-
tions of the increasing prevalence of non-albicans candidemia. Crit Care Med
2007;35:1077–83.
47. Pfaller MA, Diekema DJ, Gibbs DL, et al. Results from the ARTEMIS DISK Global
Antifungal Surveillance study, 1997 to 2005: an 8.5-year analysis of susceptibili-
ties of Candida species and other yeast species to fluconazole and voriconazole
determined by CLSI standardized disk diffusion testing. J Clin Microbiol 2007;45:
1735–45.
48. Wey SB, Mori M, Pfaller MA, et al. Risk factors for hospital-acquired candidemia.
A matched case–control study. Arch Intern Med 1989;149:2349–53.
49. Pittet D, Monod M, Suter PM, et al. Candida colonization and subsequent infec-
tions in critically ill surgical patients. Ann Surg 1994;220:751–8.
50. Fraser VJ, Jones M, Dunkel J, et al. Candidemia in a tertiary care hospital: epide-
miology, risk factors, and predictors of mortality. Clin Infect Dis 1992;15:414–21.
51. Blumberg HM, Jarvis WR, Soucie JM, et al. Risk factors for candidal bloodstream
infections in surgical intensive care unit patients: the NEMIS prospective multi-
center study. The National Epidemiology of Mycosis Survey. Clin Infect Dis
2001;33:177–86.
52. Chow JK, Golan Y, Ruthazer R, et al. Risk factors for albicans and non-albicans
candidemia in the intensive care unit. Crit Care Med 2008;36:1993–8.
53. Leon C, Ruiz-Santana S, Saavedra P, et al. A bedside scoring system (Candida
score) for early antifungal treatment in non-neutropenic critically ill patients with
Candida colonization. Crit Care Med 2006;34:730–7.
54. Ostrosky-Zeichner L, Sable C, Sobel J, et al. Multicenter retrospective develop-
ment and validation of a clinical prediction rule for nosocomial invasive candidi-
asis in the intensive care setting. Eur J Clin Microbiol Infect Dis 2007;26:271–6.
55. Faiz S, Neale B, Rios E, et al. Risk-based fluconazole prophylaxis of Candida
bloodstream infection in a medical intensive care unit. Eur J Clin Microbiol Infect
Dis 2008; in press.
56. Reiss E, Morrison CJ. Nonculture methods for diagnosis of disseminated candi-
diasis. Clin Microbiol Rev 1993;6:311–23.
57. Mean M, Marchetti O, Calandra T. Bench-to-bedside review: Candida infections
in the intensive care unit. Crit Care 2008;12:204.
58. Garbino J, Lew DP, Romand JA, et al. Prevention of severe Candida infections in
non-neutropenic, high-risk, critically ill patients: a randomized, double-blind,
placebo-controlled trial in patients treated by selective digestive decontamina-
tion. Intensive Care Med 2002;28:1708–17.
59. Eggimann P, Francioli P, Bille J, et al. Fluconazole prophylaxis prevents intra-
abdominal candidiasis in high-risk surgical patients. Crit Care Med 1999;27:
1066–72.
60. Shorr AF, Chung K, Jackson WL, et al. Fluconazole prophylaxis in critically ill
surgical patients: a meta-analysis. Crit Care Med 2005;33:1928–35.
Fungal Infections in the ICU 641