1 What is the definition of disseminated fungal infection?

We define disseminated fungal infection as the presence of a fungal pathogen in the blood (fungemia) and/or any other sterile deep-seated structure because of hematogenous seeding. This distinguishes disseminated infection from superficial infection, which mostly involves the mucocutaneous structures, that is, dermatitis, onychitis, stomatitis, esophagitis, and keratitis, as well as from simple colonization, which is the isolation of a fungal pathogen from a nonsterile site without any sign of infection attributable to the specific pathogen. Invasive fungal infection is a more general term and refers to fungemia and other fungal infections such as disseminated candidiasis, endocarditis, meningitis, and hepatosplenic infection.

2 What are the most clinically important fungal pathogens?

Candida spp., Aspergillus spp., and Cryptococcus spp. are by far the most common fungal pathogens encountered in the hospital setting, with Candida being the leading fungal pathogen. Of note is that Aspergillus spp. have been steadily increasing in the intensive care unit (ICU) setting.

3 What is the epidemiology of fungal infections in hospitalized patients?

In the last 25 years the total number of fungal infections in hospitalized patients has increased from 6% in 1980 to 10.4% in 1990 and currently may be as high as 25%. Candida species are the fourth most commonly recovered blood culture isolates in the United States.

4 Why has the incidence of fungal infection increased so dramatically?

Fungi generally do not cause invasive infection in healthy individuals. Robust cellular and antibody-mediated immunity and intact mucosal barriers play a major role in shielding the human body from opportunists such as Candida spp., which seek a way of passing through tissues and entering the bloodstream and other deep-seated organs. With the numbers of immunosuppressed patients increasing through cancer and chemotherapy, transplantation, and HIV infection, as well as with the increased use of vascular and urinary catheters and broad-spectrum antibacterial agents, an alarming increase of deep-seated fungal infections has been seen in clinical practice.

5 What fungi are responsible for invasive infection in humans?

C. albicans accounts for the majority (mostly > 50%-60%) of all Candida infections, but this percentage is declining to 45%. C. tropicalis, C. glabrata, and C. krusei account for most of the remainder. This is particularly important because certain Candida spp. such as C. krusei can be resistant to fluconazole. On the other hand, Aspergillus spp. account for at least 15% to 20% of all fungal infections; however, the rate can be higher in patients after lung transplantation. Other less-common but increasing mycoses include blastomycoses, coccidioidomycoses, cryptococcosis, histoplasmosis, and sporotrichosis.

6 What are the most important risk factors for disseminated Candida infection?

7 List the diagnostic criteria for disseminated fungal infection

Definitive:

Suggestive: three confirmed colonized sites (should be regarded as a risk factor rather than a definite sign of infection)

8 How reliable are these diagnostic criteria?

The preceding criteria are positive in only 30% to 50% of patients with disseminated fungal infection. Therefore a high index of suspicion must be maintained.

9 Should asymptomatic candiduria be treated?

Among low-risk individuals no treatment is recommended. Amphotericin bladder irrigation is no longer recommended. Usually a change of urinary catheter should suffice. Among high-risk patients (patients with neutropenia or with urologic manipulations, low-birth-weight infants) treatment is similar to the one used for invasive candidiasis.

10 Should a central venous catheter be removed once candidemia is confirmed?

Practice guidelines indicate that all central venous catheters should be removed once candidemia is confirmed (Table 35-1). Of note is that a recent randomized controlled trial and other studies question the benefit of early removal of central venous catheters in the onset of candidemia for some selected patients. We recommend following the standard practice guidelines.

^{Table 35-1} Recommendation on CVC Removal in Patients with Candidemia

CVC, Central venous catheter.

Modified from www.guidelines.gov and Mermel LA, Allon M, Bouza E, et al: Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 update by the Infectious Diseases Society of America [published errata appear in: Clin Infect Dis 50:457, 2010, and Clin Infect Dis 50:1079, 2010]. Clin Infect Dis 49:1-45, 2009.

11 When should you suspect disseminated candidiasis?

Unfortunately, disseminated candidiasis has a wide spectrum of manifestations from a mild fever to a sepsis syndrome with multiorgan failure. On certain occasions the hematogenous spread of Candida produces visible changes throughout the body including muscle, skin, and eyes, making a bloodstream process clinically apparent. However, this is not always the case, and that is why there must be a low threshold for the disease especially in patients with multiple risk factors for candidiasis.

12 If disseminated candidiasis is suspected, where should you look for it?

The first consideration is to perform blood cultures. Then examine the retina for endophthalmitis. However, recent reports show that rate of hematogenous endophthalmitis is considerably lower (2%-3%) compared with older studies that reported a 30% rate of such complication. Keep in mind that patients with neutropenia may have no symptoms until they regain their normal counts. You can also search for Candida in the heart valves for endocarditis; the bone for osteomyelitis; and the liver, spleen, and kidneys for renal abscesses and candiduria. Also make sure to perform a biopsy of skin lesions to add extra yield to the diagnosis along with blood cultures. This is important because blood cultures are only 50% to 60% sensitive.

13 What is the overall mortality associated with candidemia?

The overall mortality associated with candidemia is 40% to 68%, with an attributable mortality of 25% to 40%. However, the earlier the initiation of antifungal agents, the better the prognosis. Early targeted antifungal therapy is difficult to accomplish because cultures take 24 to 48 hours to yield the species and antifungal resistance profiles. That is why empirical therapy should take into account risk factors for antifungal resistance, that is, prolonged exposure to antifungal agents or long length of hospital stays, and also risk factors for non-albicans species intrinsically resistant to fluconazole.

14 Should antifungal therapy be delayed until blood cultures are positive for fungus?

Absolutely not! Early therapy means lower mortality. Blood cultures have been found to be only 40% to 70% sensitive. Systemic antifungal therapy should be strongly considered, especially in a patient who is at high risk for disseminated fungal infection, if:

15 What are the major classes of antifungal drugs in use today?

Antifungal drugs in clinical use today fall into three broad categories: polyene antifungals (amphotericin B), antifungal azoles, and the echinocandins.

16 So, how do we treat?

According to recent guidelines, physicians should choose among an echinocandin, fluconazole, an amphotericin B preparation, or combination of amphotericin B with flucytosine. The therapeutic strategy should take into account any previous use of antifungal agents (because it can select resistant species), the epidemiology of fluconazole-resistant or non-albicans strains in the community, and any comorbid conditions (which could influence drug pharmacokinetics or worsen coexisting conditions such as renal failure). An echinocandin or amphotericin B should be preferred over fluconazole among patients with neutropenia and critically ill patients, as well as those known to be exposed to fluconazole-resistant strains. Of course the physician can always switch to fluconazole whenever an antifungal resistance profile becomes available.

17 How do amphotericin B and flucytosine work?

Amphotericin B, a polyene, is fungicidal. It binds irreversibly to ergosterol (but not to cholesterol, the major sterol in mammalian cell membranes), creating a membrane channel that allows leakage of cytosol leading to cell death. Flucytosine is sometimes used in conjunction with amphotericin B and is synergistic Cryptococcus. Flucytosine acts directly on fungal organisms by competitive inhibition of purine and pyrimidine uptake.

18 What antifungal azoles are available, and how do they work?

Fluconazole, itraconazole, voriconazole, and posaconazole, which are triazoles, are fungistatic against Candida spp. They inhibit C-14 α-demethylase, a cytochrome P-450–dependent fungal enzyme required for synthesis of ergosterol, the major sterol in the fungal cell membrane. This alters cell membrane fluidity, decreasing nutrient transport, increasing membrane permeability, and inhibiting cell growth and proliferation.

19 How do echinocandins work, and are they being used?

The target for echinocandins is the complex of proteins responsible for synthesis of cell wall polysaccharides. Caspofungin, anidulafungin, and micafungin are used in the treatment of candidemia and other forms of disseminated Candida infections.

20 What advantages does fluconazole offer over amphotericin B in the treatment or prevention of disseminated fungal infections?

21 Are there any limitations to the use of fluconazole?

Yes. Fluconazole is not active against Aspergillus spp. or C. krusei and other resistant Candida spp. Also, remember that fluconazole may inhibit the P-450 detoxification system and cause hepatotoxicity, increasing phenytoin and cyclosporin levels and potentiating warfarin’s anticoagulant effects. Fluconazole has fewer overall adverse effects than amphotericin B.

22 What should be done when a Candida infection fails to respond to fluconazole?

An echinocandin agent or an amphotericin B preparation should be considered. Make sure that the diagnosis is confirmed, the dosage is appropriate, and drug-drug interactions are ruled out (for example, rifampin decreases fluconazole levels). Keep in mind that resistance can happen to all antifungal agents. For example, C. krusei and C. glabrata can be resistant to azoles, C. lusitaniae can be resistant to amphotericin B, and C. parapsilosis can have higher minimum inhibitory concentration to echinocandins.

23 Are there less toxic forms of amphotericin B available?

Yes. To reduce the toxicity associated with amphotericin B, lipid formulations have been produced. The earliest and most widely studied of these is AmBisome, which in randomized trials has been shown to be safer than amphotericin B with many fewer side effects. Other lipid-associated, nonliposomal products are amphotericin B lipid complex (Abelcet) and amphotericin B colloidal dispersion (Amphocil). The disadvantage of these alternative forms of amphotericin B is their currently high cost.

24 Can health care providers help prevent the spread of fungal colonization in the ICU?

Easily. Wash hands and wear gloves when working directly with patients. Candida species were found on the hands of 33% to 75% of ICU staff in one study.

25 Does the strategy of presumptive or preemptive treatment of high-risk patients prevent severe candidiasis in critically ill surgical patients?

The effectiveness of fluconazole in treating overt candidiasis has unfortunately provoked its widespread, unjustified use in patients without neutropenia in the ICU setting. This practice has likely led to an increase in non-albicans species, which are resistant to fluconazole. Several studies have shown decreased incidence of colonization and the risk of candidiasis with such empirical treatment but have failed to show decreased mortality in any group other than high-risk patients who have received a transplant. Recent reviews have suggested that targeted preemptive strategy may be of benefit in preventing candidiasis in the ICU. This concept requires further study before continued practice. Of note is that fluconazole should be given as secondary prophylaxis in patients with HIV with CD4 < 200 cells/mm³ who survived cryptococcosis.