The need for stewardship in anti-fungal therapy has become increasingly evident. Anti-fungal overuse in hospitals is primarily driven by empiric therapy. Invasive fungal diseases are challenging to front line clinicians, because they are difficult to diagnose and delays in therapy are associated with increased morbidity and mortality. However, widespread empiric therapy is costly, may expose patients to unnecessary drug toxicity, and exert selective pressure, leading to the emergence of resistance. In addition, misdiagnosis of true fungal infections, in which definitive anti-fungal therapy is indicated, may contribute to the inappropriate use of antibacterials and also contribute to the emergence of drug-resistant bacteria. The increasing prevalence of invasive fungal infections is associated with advances in modern medicine, including the use of invasive devices and immuno-suppression. Candidemia ranks among the most common causes of healthcare associated bloodstream infections in the United States, especially in intensive care units. In this segment, we will review the management of suspected and proven Candidemia and highlight potential stewardship opportunities. First, some background. Invasive candidiasis comprises two broad categories of disease that are not mutually exclusive. Candidemia or the presence of Candida in blood, is more common disease entity and accounts for the majority of patients included in clinical trials. Deep-seated candidiasis arises from either hematogenous dissemination or direct inoculation of a sterile site, such as the peritoneal cavity after bowel perforation. Patients with deep-seated candidiasis may or may not have concombinant Candidemia. There are at least 15 distinct candida species that cause human disease but five cause greater than 90 percent of invasive infections, Candida albicans, Candida glabrata, Candida tropicalis, Candida parapsilosis, and Candida krusei. In recent years, the prevalence of infections caused by non-albican species has increased. However, the predominant non-albicans species varies geographically. For example, Candida glabrata has emerged as an important pathogen in Northern Europe, the United States, and Canada, whereas Candida parapsilosis is more prominent in southern Europe, Asia, and South America. The local prevalence of non-albican species impacts empiric therapy, given susceptibility patterns vary by species. For example, many Candida glabrata and Candida krusei isolates exhibit resistance to azoles, while Candida parapsilosis isolates exhibit higher MICs to echinocandins, although they do remain effective. Flucanizole exposure is associated with an increased risk of infection with a non-albican species. It's also important to remember that candida species differ in the incubation time necessary to grow in culture, with Candida glabrata requiring 72 hours in some studies. Of concern is the recent emergence of Candida auris, which was first reported in Japan in 2009 and since then has been isolated in other nations including the US. This organism is multi-drug resistant, exhibiting high MICs to azoles, amphotericin, and in some isolates, echinocandins, and has been associated with outbreaks in hospitals. Candida are normal human commensals and commonly colonize the gastrointestinal tract, skin, and female genital tract. Although Candida species are frequently cultured from expectorated sputum and urine obtained from indwelling bladder catheters, they rarely cause infections at these sites. However, Candida isolated from blood cultures, should never be considered a contaminant. It should prompt to search for the source, as well as a metastatic foci of infection, such as chorioretinitis. Now, let's focus specifically on the management of Candidemia. The majority of anti-candidal use in the inpatient setting is empiric. This is in part due to the lack of rapid sensitive diagnostics for the detection of Candidemia, as well as non-specific clinical presentation of this infection. Patients with Candidemia may present with the spectrum of disease, ranging from fever alone to severe sepsis, indistinguishable from Gram-negative bacterial sepsis and complicated by multi-organ failure. A number of factors increase a patient's risk of developing Candidemia, including some commonly seen in ICU patients, the receipt of broad-spectrum antibiotics, the presence of central catheters, particularly those used for total parenteral nutrition, and a prolonged length of stay. Other risk factors include abdominal surgery, especially in those with an anastomotic leak or those who required repeated abdominal surgeries, acute necrotizing pancreatitis, cancer, immuno-suppression associated with organ transplantation, chemotherapy, or steroids use. Neonates, particularly those born prematurely or with low birth weight, are also at increased risk of Candidemia. Despite the fact that all of these risk factors have been noted, prediction models have only been modestly beneficial at identifying patients with Candidemia. The gold standard for the diagnosis of Candidemia is positive blood cultures. Historical autopsy studies found that antemortem blood cultures were only positive in half of the patients with proven invasive candidiasis. However, some suggest that the sensitivity of blood cultures for Candidemia may be underestimated in these studies due to the inclusions of patients, they,re likely had invasive candidiasis without active Candidemia. Nevertheless, new diagnostics are needed to complement cultures in detecting patients with Candidemia. A full discussion of the assays available and in development to detect Candidemia is beyond the scope of this segment. However, several of these assays, including beta-D glucan, have high negative predictive values and could provide stewardship programs with the opportunity to coordinate with the clinical laboratory, on anti Candidal de-escalation projects. New rapid diagnostics, such as Viracor Candida Real-Time PCR panel and the T2Candida panel show promise, but their clinical usefulness will depend on whether their negative predictive values are demonstrated in large clinical studies, to allow avoidance or early discontinuation of empiric therapy. Early initiation of therapy and source control are key steps in managing Candidemia. A critical element of source control is the early removal of central venous catheters, which is recommended for all non-neutropinic patients with Candidemia. For neutropinic patients, the guidelines recommend that decisions regarding the removal of lines be made on a case-by-case basis, given alternative sources of infection are common. The 2016 Infectious Disease Society of America guidelines recommend echinocandins for the initial empiric treatment of Candidemia, in both neutropinic and non-neutropinic patients. This recommendation is based in part due to the emergence of azole-resistant candidal infections. For non-neutropinic patients, fluconazole is an acceptable alternative to echinocandins as initial empiric therapy, for patients who are not critically ill and who are considered unlikely to be infected with the fluconazole-resistant candidate species, based upon local epidemiology and or no previous exposure to azoles. For neutropinic patients, the guidelines recommend liposomal formulations of amphotericin as second line alternatives, given the use of azoles as prophylaxis in this population. It's important to remember that echinocandins do not achieve therapeutic levels in the CSF, vitreous fluid, and urine. Although the guidelines recommend starting echinocandins, they encourage the transition to fluconazole or voriconazole for patients who are clinically stable, have cleared Candidemia, and are infected by isolates susceptible to one of these agents. Both fluconazole and voriconazole are available in oral formulation, and this transition obviates the need for continued intravenous administration of anti-fungals. This provides stewardship programs the opportunity to identify cases based upon microbiologic data, and intervene to avoid the unnecessary use of central venous catheters, and to potentially reduce hospital length of stay. The recommended total duration of therapy for Candidemia without metastatic complications is 14 days after documented clearance of Candidemia and resolution of symptoms. In neutropinic patients, therapy is often continued until neutropinia resolves. Here are the key take-home points. The epidemiology of Candidemia is changing with the increasing prevalence of more aggressive medical care in the emergence of anti-fungal resistance. Candidemia is an important target for antimicrobial stewardship programs. It provides an opportunity to collaborate with the microbiology lab to establish diagnostic strategies to improve detection, and also rule out Candidemia thus optimizing empiric anti-fungal use. Although guidelines encourage the use of echinocandins as a first line agents, especially in those who are critically ill or have previous exposure to azoles, they encourage de-escalation to oral agents when indicated. Antimicrobial Stewardship Programs could target these patients and provide de-escalation guidance. Improving the detection and management of Candidemia, may allow ASPs to guide the de-escalation of empiric antibacterials as well. We direct learners to case five from our course, optimizing antibiotics through timeouts, for more information about the management of neutropinic patients with suspected and proven Candidemia
