Approach
Candida species in a blood culture should never be considered as a contaminant and in all cases requires treatment with an antifungal agent.[32] It should not be assumed that removal of a central venous catheter alone is adequate therapy for candidaemia. Given the high mortality associated with systemic candidiasis, treatment should not be delayed. Delay of appropriate antifungal therapy contributes to higher mortality.[61][62][63] Routine and early consultation with a relevant consultant such as an infectious disease physician or a clinical microbiologist is strongly recommended.
Empirical treatment options are discussed below. Once Candida species is determined, antifungal treatment can be tailored further as some species are inherently resistant to specific antifungals. Antifungal choice also depends on location of the infection and ability of the drug to penetrate the affected organ. Antifungal susceptibility testing can be performed, especially when there is lack of response to treatment or when more resistant species are isolated, such as Candida auris.[32]
Antifungal agents
Antifungal therapy has progressed dramatically from the period when the only drug available was the frequently toxic intravenous amphotericin-B. There are now additional classes of antifungal agents (azoles and echinocandins). Clinicians have a wide choice of antifungals, none of which has been shown to be superior to amphotericin-B. However, azoles and echinocandins are easier to use and possess superior safety profiles.
Amphotericin-B deoxycholate was previously the standard drug for systemic candidiasis. It targets extraction of ergosterol from fungal cells. It is rapidly fungicidal in vitro and possesses broad-spectrum activity against most Candida species except Candida lusitaniae, but it is significantly nephrotoxic and associated with frequent infusion toxicity. Lipid and liposomal formulations of amphotericin-B have largely replaced the deoxycholate formulation, with reduced, but not absent, nephrotoxicity. However, these agents are expensive, and although less toxic, not therapeutically superior.
Echinocandins are non-competitive inhibitors of synthesis of beta 1,3-beta-D-glucan, a major component of the fungal cell wall, and include caspofungin, anidulafungin, and micafungin.[64] They have excellent fungicidal activity against all Candida species and a good safety profile.[65] No dose adjustments are required for advanced renal disease. The echinocandins are preferred over azoles for the initial treatment of candidaemia, particularly if Candida krusei, Candida glabrata, or C auris are suspected or documented and in critically ill, haemodynamically unstable patients.[19][32][66][67] The highest minimum inhibitory concentrations are found with Candida parapsilosis (1-2 microgram/mL), although clinical efficacy is comparable to other agents and failures with this organism are extremely rare.[68] Similarly, acquired resistance to all or any of the drugs in this class remains relatively rare, although it is increasing, particularly among C glabrata isolates.[69][70] Caspofungin, anidulafungin, and micafungin are widely considered as therapeutically equivalent.[32][71]
Azoles inhibit the formation of ergosterol in the fungal cell membrane. Fluconazole has been widely used for candidiasis for more than 15 years, and has a good safety record and a favourable toxicity profile. Several large studies have shown equivalent activity and success when compared with polyenes, such as amphotericin-B, and the newer echinocandins in non-neutropenic hosts.[72][73][74] However, growing resistance now limits its use to empirical therapy in settings where fluconazole resistance is low and/or echinocandins are not available. Fluconazole is also preferred over echinocandins if infection is suspected in body compartments where echinocandins have poor penetration (e.g., central nervous system, eye, and urinary tract). Fluconazole can also be used for step-down therapy from echinocandins once patients are clinically stable and infection is controlled. Voriconazole can retain activity against most Candida species and is active against fluconazole-resistant C krusei and some strains of fluconazole-resistant C glabrata. Its efficiency in treatment of invasive candidiasis has been demonstrated in a large prospective, randomised study.[75] However, its lack of superiority, additional cost and less favourable toxicity profile, particularly numerous drug interactions, have resulted in voriconazole being reserved for selected patients with fluconazole-resistant but voriconazole-sensitive strains.
Treatment for suspected systemic candidiasis
Antifungal therapy should be started in patients who are critically ill and have risk factors for systemic candidiasis, and for whom other causes of fever have been excluded. This decision should be based on clinical assessment and results of cultures from non-sterile sites. Local pathogen knowledge is also important when deciding on empirical therapy.
Non-neutropenic patients
First-line choice is an echinocandin. An echinocandin is preferred over azoles for the initial treatment when candidaemia secondary to a Candida species known to have azole resistance is suspected or documented, particularly in critically ill, haemodynamically unstable patients; those who have had recent exposure to an azole; and those colonised with azole-resistant Candida species.[19][32][66][67] Fluconazole is an acceptable first-line alternative for clinically stable patients who have had no recent azole exposure and are not colonised with azole-resistant Candida species such as C krusei, C glabrata, or C auris.[32] If echinocandins are not tolerated or are not available, recommended alternatives include the lipid or liposomal formulation of amphotericin-B.
For patients who have no clinical response to empirical antifungal therapy at 4-5 days and who do not have subsequent evidence of invasive candidiasis after the start of empirical therapy or if cultures and surrogate markers are negative, consideration should be given to stopping antifungal therapy.[32]
Neutropenic patients
First-line choice is an echinocandin.[76] Lipid or liposomal amphotericin B is also effective, although increased toxicity makes it less attractive as initial therapy. Fluconazole and voriconazole are alternative agents in patients who are not critically ill, have not had azole exposure, and are not suspected or known to have an azole-resistant Candida species. An echinocandin is preferred over azoles for the initial treatment of candidaemia when azole-resistant Candida species are suspected or documented, particularly in critically ill, haemodynamically unstable patients; those who have had recent exposure to an azole, and those colonised with azole-resistant Candida species such as C krusei, C glabrata, or C auris.[19][32][66][67] If patients have received an azole for prophylaxis, an azole should not be given as empirical therapy. Prophylactic antifungals are standard of care and routinely used in patients with neutropenia likely to persist longer than 7 days.
Treatment for confirmed systemic candidiasis
The recommended duration of therapy for systemic candidiasis with no complications is 2 weeks beyond clearance of Candida from the blood and when clinical symptoms have resolved. In all cases of candidaemia, repeat blood cultures should be obtained daily until clearance of candidaemia is documented.
Source control is an essential part of therapy. Central venous catheter removal is recommended in all patients who have a confirmed diagnosis, except in some instances in neutropenic patients where translocation from the gastrointestinal tract is suspected to be the source of candidaemia rather than a true catheter infection.[32]
Early drainage of any abscess or removal of infected material should be attempted as guided by the individual circumstances.[77]
Non-neutropenic patients
Initial therapy is with an echinocandin or fluconazole as first-line choices.
An echinocandin is preferred for patients who are critically ill, who have had recent azole exposure, or in whom the infection is caused by a species expected to be azole-resistant, such as C glabrata, C krusei, or C auris.[19][32][66][67] Some experts consider echinocandins to be agents of first choice for all patients as initial therapy.[78]
Fluconazole is an acceptable first-line option for patients who are not critically ill and who are considered unlikely to have fluconazole-resistant Candida species.
Lipid or liposomal formulations of amphotericin-B are considered alternatives if initial antifungal agents are not tolerated or are not available.
Transition from an echinocandin to fluconazole is recommended for patients who are clinically stable, who have documented clearance of Candida from the bloodstream, and who are infected with an organism that is susceptible to fluconazole (e.g., Candida albicans, Candida parapsilosis, and Candida tropicalis).[32]
Voriconazole is effective for candidaemia but offers little advantage over fluconazole as initial therapy. It is recommended as step-down oral therapy for selected cases of candidaemia due to C krusei or for C glabrata that has been shown to be susceptible.[32]
Neutropenic patients
Initial therapy for most patients should be with an echinocandin.
Echinocandins are recommended as initial therapy for C auris species, as most strains in the US are susceptible.[19]
Lipid or liposomal formulations of amphotericin-B are considered an alternative therapy in this patient group.
Fluconazole is an alternative for patients who are not critically ill and who have had no prior azole exposure, if fluconazole resistance is not suspected.
If mould coverage is required, voriconazole can be used. It can also be used as step-down therapy during neutropenia in clinically stable patients who have had documented bloodstream clearance and isolates that are susceptible to voriconazole.
Treatment of complications
If patients have complications (e.g., endocarditis, endophthalmitis, renal candidiasis), more prolonged therapy might be required depending upon the site involved.[32]
For endocarditis, initial treatment should be with a high-dose echinocandin, or lipid or liposomal amphotericin-B with or without flucytosine.[32] Fluconazole can be used as step-down therapy in patients with fluconazole-susceptible Candida, after clinical improvement and clearance of candidaemia is documented. Other azoles (e.g., voriconazole) can be used for step-down therapy if isolates are resistant to fluconazole but susceptible to other azoles. Total duration of treatment is at least 6 weeks, although it can be much longer, especially if there is no surgical intervention or in cases complicated by perivalvular abscesses.[32] Despite the lack of powered randomised studies, most literature suggests improved outcomes and decreased mortality in patients treated with surgical intervention in addition to antifungal therapy. Surgical intervention (valve replacement/repair or vegetectomy) is therefore recommended when possible.[79] Antifungals should be continued 6 weeks after surgery. In patients with native valve endocarditis not able to undergo valve replacement, long-term suppression with fluconazole may be considered after completion of 6 weeks of therapy. In patients with prosthetic valve candida endocarditis, chronic suppressive therapy is recommended.[79]
Endophthalmitis should be treated with lipid or liposomal amphotericin-B plus flucytosine for infections due to azole-resistant isolates, although fluconazole or voriconazole are preferred in azole-susceptible isolates. The echinocandins may not be useful in this context due to inadequate penetration into the vitreous space.[80] Treatment should be continued for 4 to 6 weeks or longer, depending on clinical response as determined by serial ophthalmological examinations. Vitrectomy should be strongly considered as well, along with intravitreal injection of amphotericin-B or voriconazole in cases of vitritis or macular involvement.[32] For cases of chorioretinitis without vitreal involvement, systemic antifungals for 4-6 weeks are recommended.
Candida pyelonephritis can be secondary to ascending infection from the urinary tract or haematogenous spread. Fluconazole is the only azole with good concentration in the urine. Pyelonephritis should be treated with fluconazole for at least 2 weeks, with amphotericin-B deoxycholate as an alternative. Another option is flucytosine in combination with amphotericin-B.[20] Surgical intervention may be indicated for source control, especially in the setting of fungal balls. Removal or replacement of nephrostomy tubes or stents should be attempted when these are present. Longer courses of therapy may be needed if there are additional complications.
Supportive care
If the patient is critically ill with sepsis, he or she may need to be admitted to the intensive care unit and require additional care in the form of vigorous fluid resuscitation. If hypotension persists, vasopressors or inotropes may be required.
Use of this content is subject to our disclaimer