INTRODUCTION — The term candidemia describes the presence of Candida species in the blood. Candida in a blood culture should prompt a search for the source; it should never be viewed as a contaminant. Invasive candidiasis encompasses both candidemia and deep-seated (ie, visceral) candidiasis (which may occur with or without candidemia) [1].
The management of candidemia and invasive candidiasis in children will be discussed here. An overview of candidal infections and the clinical manifestations and diagnosis of candidemia and invasive candidiasis in children, candidal infections in neonates, and candidal infections in adults are discussed separately:
●(See "Candida infections in children".)
●(See "Candidemia and invasive candidiasis in children: Clinical manifestations and diagnosis".)
●(See "Epidemiology and risk factors for Candida infection in neonates" and "Clinical manifestations and diagnosis of Candida infection in neonates" and "Treatment of Candida infection in neonates".)
●(See "Candidemia in adults: Epidemiology, microbiology, and pathogenesis" and "Clinical manifestations and diagnosis of candidemia and invasive candidiasis in adults" and "Management of candidemia and invasive candidiasis in adults".)
HIGH SUSPICION FOR INVASIVE CANDIDIASIS — Antifungal therapy may be indicated pending laboratory identification of Candida for patients in whom there is a high suspicion for invasive candidiasis. We initiate antifungal therapy in children with significant risk factors for invasive candidiasis (eg, ongoing fever despite broad-spectrum antibiotics in immunocompromised patients) or characteristic examination findings (eg, retinal (picture 1A-B) or skin lesions (picture 2A-B)).
DOCUMENTED INVASIVE CANDIDIASIS — Treatment of candidemia and invasive candidiasis requires treatment with an antifungal agent and adequate source control.
Choice of antifungal therapy — Antifungal therapy should be initiated as soon as possible after Candida is identified in a blood culture or other appropriate laboratory specimens (eg, urine culture, biopsy specimen culture). Early initiation of antifungal therapy (ie, within 24 hours of obtaining blood cultures) is associated with enhanced survival in adult patients with invasive candidiasis [2-4].
The choice of antifungal agent is guided by clinical features (age, hemodynamic stability/severity of illness, site of infection, baseline renal function, neutropenia) and the infecting Candida species (which affects susceptibility patterns) [5,6]. (See "Management of candidemia and invasive candidiasis in adults".)
Our approach to antifungal therapy in children with invasive candidiasis is generally in keeping with the 2016 Infectious Diseases Society of America (IDSA) and 2012 European Society for Clinical Microbiology and Infectious Diseases (ESCMID) guidelines for the treatment of candidiasis (table 1) [7,8]. The treatment of candidiasis in neonates is discussed separately. (See "Treatment of Candida infection in neonates".)
Pending speciation and susceptibility testing — Pending speciation and results of susceptibility testing, the choice of therapy for Candida is influenced by whether the patient is known to be colonized with resistant Candida species, if the patient has developed candidemia despite antifungal prophylaxis, and the proportion of candidemia isolates due to resistant species within a particular medical center. In addition, a positive germ tube test (if performed) suggests Candida albicans, which is generally susceptible to fluconazole. When the Candida species and susceptibility are known, antifungal therapy can be adjusted as indicated. (See 'Targeted therapy' below.)
●Nonneutropenic patients – Pending speciation and susceptibility testing, for nonneutropenic children with invasive candidiasis that does not involve the eye, central nervous system (CNS), or urinary tract, we suggest treatment with an echinocandin (eg, micafungin, caspofungin) or a lipid formulation of amphotericin B. Fluconazole is an acceptable alternative in patients who are not critically ill and those unlikely to have a fluconazole-resistant infection (eg, positive germ tube test [if performed], lack of prior fluconazole exposure). Echinocandins are not recommended for candidiasis involving the eye, CNS, or urinary tract because they have relatively poor penetration at these sites.
Although experience with caspofungin, micafungin, and voriconazole in children is limited [9-15], these agents have been shown to be efficacious in nonneutropenic adults [16,17]. (See "Management of candidemia and invasive candidiasis in adults", section on 'Non-neutropenic patients'.)
Several large randomized trials in adults have shown that fluconazole is as effective as amphotericin B deoxycholate for the treatment of candidemia [18-21]. Similar comparative trials in children have not been reported, but case reports and noncomparative trials of fluconazole in children suggest that it is safe, well-tolerated, and effective [22-24].
●Neutropenic patients – Pending speciation and susceptibility testing, for neutropenic children with invasive candidiasis that does not involve the eye, CNS, or urinary tract, we suggest treatment with an echinocandin (eg, micafungin, caspofungin) or a lipid formulation of amphotericin B, although use of the latter may be limited by toxicity in this patient population. Fluconazole is an acceptable alternative in patients who are not critically ill and those without prior azole exposure. Echinocandins are not recommended for candidiasis involving the eye, CNS, or urinary tract because they have relatively poor penetration at these sites.
Less information is available to guide antifungal therapy in neutropenic than nonneutropenic patients with candidemia. Limited data from studies in adults support the use of amphotericin B, an echinocandin, or fluconazole in neutropenic adults, but there are few studies in children [14]. We avoid fluconazole in children who are clinically unstable, have disseminated infection, and have received fluconazole for prophylaxis. (See 'Antifungal agents' below.)
Targeted therapy — Antifungal therapy can be adjusted as indicated when the susceptibility pattern is available and/or the species is identified. However, in hospitals where susceptibility testing is performed at an outside reference laboratory, prompt adjustment of antifungal therapy (eg, from an echinocandin to fluconazole) may not be feasible.
C. albicans is the most common cause of candidemia, but nonalbicans species are increasingly isolated. Among nonalbicans species, Candida glabrata and Candida parapsilosis are most prominent, followed by Candida tropicalis and Candida krusei. Antifungal susceptibility patterns and minimal inhibitory concentrations may vary according to species. (See "Candidemia and invasive candidiasis in children: Clinical manifestations and diagnosis", section on 'Epidemiology' and "Management of candidemia and invasive candidiasis in adults".)
●C. albicans, C. parapsilosis, and C. tropicalis – C. albicans, C. parapsilosis, and C. tropicalis generally are susceptible to amphotericin B, azoles, and echinocandins, and initial therapy with any of these is reasonable for patients with invasive candidiasis that does not involve the eye, CNS, or urinary tract. Echinocandins are not recommended for candidiasis involving the eye, CNS, or urinary tract because they have relatively poor penetration at these sites. Voriconazole is also not recommended for patients with urinary candidiasis because the active form is minimally excreted in the urine.
Although there is concern that C. parapsilosis may be less susceptible to echinocandins because of innately higher minimum inhibitory concentrations, clinical data are lacking.
For patients with C. albicans, C. parapsilosis, or C. tropicalis, we suggest transitioning from initial antifungal therapy to fluconazole in patients who are clinically stable, have fluconazole-susceptible isolates, and negative repeat blood cultures.
●C. glabrata – For children with invasive candidiasis due to C. glabrata, we usually treat with amphotericin B. Voriconazole may be an alternative for susceptible isolates but is not recommended for infection involving the urinary tract. Many C. glabrata species are resistant to fluconazole. Although echinocandins are preferred for adult patients, some studies have shown increased rates of echinocandin resistance [25-28].
●C. krusei – For children with invasive candidiasis due to C. krusei, we usually treat with amphotericin B, an echinocandin, or voriconazole, if the isolate is susceptible.
For patients with urinary candidiasis, voriconazole is not recommended because the active form is minimally excreted in the urine. Echinocandins are not recommended for candidiasis involving the eye, CNS, or urinary tract because they have relatively poor penetration at these sites.
C. krusei has decreased susceptibility to amphotericin B, requiring higher doses. C. krusei is intrinsically resistant to fluconazole. Resistance to voriconazole varies geographically.
Monitoring response — We obtain daily or every other day blood cultures after initiating treatment for candidemia to confirm the date of sterilization (which determines the duration of treatment). (See 'Duration of antifungal therapy' below.)
Children with invasive candidiasis without candidemia generally are monitored for resolution of signs and symptoms of infection as appropriate (eg, resolution of fever and characteristic lesions on abdominal imaging).
Persistent candidemia — If blood cultures remain positive for more than three days in a patient receiving appropriate antifungal therapy with adequate source control, then a search for a metastatic focus or endocarditis needs to be undertaken:
●Patients should be examined for thrombophlebitis or other focal evidence of infection (eg, focal muscle pain, abdominal pain). Signs and symptoms of infection guide the next steps in the evaluation, which often includes advanced imaging techniques (eg, ultrasonography, computed tomography, magnetic resonance imaging), especially in immunocompromised hosts.
●An echocardiogram generally is recommended because murmur may be absent in children with endocarditis. (See "Infective endocarditis in children", section on 'Clinical manifestations'.)
In neutropenic patients with persistent fungemia despite appropriate antifungal therapy, central venous catheters (CVCs) should be removed. Removal of CVCs generally is recommended as soon as feasible in the initial management of candidemia in nonneutropenic patients. However, if catheters initially are retained, their removal is necessary once persistent candidemia develops. (See 'Source control' below.)
Duration of antifungal therapy
Suggested duration — The appropriate duration of therapy for invasive candidiasis has not been studied. Our suggested durations are provided below (table 1):
●For children with candidemia and no evidence of metastatic foci of infection, we generally treat for a minimum of two weeks after blood cultures become negative. This is the duration that has been used in most clinical trials in adults [18] and is the recommended duration in the 2016 IDSA and 2012 ESCMID practice guidelines [7,8].
●For patients with CVC-associated candidemia, we generally treat for two weeks after catheter removal and candidemia has cleared.
●For patients with CVC-associated candidemia in which catheter retention is attempted, longer durations of therapy generally are needed.
In such cases, performing an ultrasound of large veins to evaluate for thrombophlebitis may be prudent. Patients with Candida suppurative thrombophlebitis require catheter removal and incision and drainage or resection of the vein, if feasible. Antifungal treatment of Candida suppurative thrombophlebitis is generally continued for at least two weeks after candidemia has cleared and preferably when the thrombus has resolved.
●For children with deep-seated invasive candidiasis (eg, endocarditis or other metastatic focus of infection), longer durations of therapy are necessary and should be determined in consultation with an infectious diseases clinician.
We typically treat endocarditis for a minimum of six weeks in patients with native value disease, but longer durations may be required in those with complications such as perivalvular abscesses. For patients with prosthetic valve endocarditis caused by Candida species, chronic suppressive antifungal therapy usually is recommended to prevent recurrence.
Switching to oral therapy — Data regarding when to switch to oral azole therapy in children are lacking. Switching to oral azole therapy may be reasonable in select patients without disseminated infection who are clinically stable, have azole-susceptible isolates and negative repeat blood cultures, and can tolerate oral medication.
Source control
Central venous catheter — In addition to prompt initiation of antifungal therapy, we recommend that CVCs be removed in nonneutropenic children with candidemia if the catheter is thought to be the source and it can be removed safely. CVCs often are the source of candidemia in patients without neutropenia [29,30]. In patients with neutropenia, the gastrointestinal tract is a frequent source of invasive candidiasis, so the decision regarding catheter removal should be considered on an individual basis.
When appropriate, the catheter should be removed as soon as possible [8,31]. Fungemia clears more quickly when catheters are removed [32,33], and in observational studies, retention of the CVC has been associated with increased mortality in children [30,34,35]. Although successful treatment without catheter removal has been described using antifungal lock therapy in addition to systemic antifungal therapy [36-41], additional studies are necessary before this approach can be routinely recommended.
Other foci of infection — Sources of infection other than a CVC are more common in neutropenic than nonneutropenic patients. Other foci of invasive candidiasis include abscesses or other fluid collections in the abdomen, chest, or bones. In addition to antifungal therapy, successful management of invasive candidiasis requires surgical or radiologic procedures to remove or drain these fluid collections [4].
Outcome — Untreated candidemia has a mortality rate of over 60 percent [42]. With treatment, the overall mortality of candidemia in children is approximately 9 to 40 percent [43-45].
Factors associated with increased mortality include [2-4,43,46,47]:
●Delayed initiation or inadequate dosing of antifungal therapy
●Septic shock
●Inadequate source control (ie, failure to remove a CVC or to drain or surgically remove a deep-seated source)
●Increased severity of illness (eg, as determined by a validated scoring system)
●Infection with an antifungal-resistant strain
●Immunosuppressive therapy
●The infecting Candida species (C. albicans and C. glabrata have increased mortality compared with C. parapsilosis)
EMPIRIC ANTIFUNGAL THERAPY FOR AT-RISK PATIENTS — Empiric antifungal therapy describes provision of antifungal therapy to patients at risk for invasive candidiasis who do not have documented invasive candidiasis, but who may nonetheless have invasive candidiasis, given the presumably low sensitivity of microbiologic tests for Candida. (See "Candidemia and invasive candidiasis in children: Clinical manifestations and diagnosis", section on 'Microbiologic studies'.)
Nonneutropenic patients — We suggest empiric antifungal therapy for nonneutropenic children who are hospitalized in the intensive care unit with risk factors for invasive candidiasis and no other known cause of fever or sepsis. Risk factors for invasive candidiasis in children include immune suppression, damage to the gastrointestinal mucosa, central venous catheter, broad-spectrum antibiotics, parenteral nutrition, renal failure requiring hemodialysis, and mechanical ventilation. (See "Candidemia and invasive candidiasis in children: Clinical manifestations and diagnosis", section on 'Risk factors'.)
The initiation of empiric antifungal therapy should be based on the clinical assessment of the patient in addition to surrogate markers of invasive candidiasis and data regarding the patient's colonization status (ie, Candida isolated in cultures from nonsterile sites), if available. A general approach is to initiate empiric antifungal therapy in critically ill patients with ongoing fever despite antibiotic therapy who have Candida colonization and risk factors for candidemia.
In a 2016 meta-analysis of 17 studies (2024 participants), empiric antifungal treatment reduced the risk of proven invasive fungal infection (35.2 versus 64.8 percent; risk ratio 0.57, 95% CI 0.39-0.83; 17 studies, 2024 participants); however, it had no effect on mortality (approximately 24 percent) [48].
The duration of empiric antifungal therapy in patients without evidence supporting a fungal infection depends upon the clinical response to empiric therapy:
●No clinical response – Empiric antifungal therapy should be stopped after four to five days in patients who have no clinical response and lack supporting evidence for fungal infection (ie, negative cultures, negative blood-based markers of invasive candidiasis).
●Clinical response – In patients who have a clinical response to therapy but have negative cultures and/or surrogate markers of invasive candidiasis, data are lacking about the appropriate duration of antifungal therapy. However, most experts would recommend a treatment course for presumptive candidemia. We agree with this approach.
Neutropenic patients — Empiric antifungal therapy may be given routinely to children with neutropenia and fever that have failed to respond to four to five days of antibiotics because they are at substantially increased risk for invasive candidiasis. This is discussed separately. (See "Management of children with non-chemotherapy-induced neutropenia and fever", section on 'Addition of antifungal therapy' and "Fever in children with chemotherapy-induced neutropenia", section on 'Addition of antifungal therapy'.)
ANTIFUNGAL AGENTS — Amphotericin B and fluconazole have been used in the treatment of candidemia in children because of their long-standing safety and efficacy. However, lipid formulations of amphotericin B, other azoles, and echinocandins are increasingly used in children as information about their safety, efficacy, and pharmacokinetics becomes available [14,49-51].
Amphotericin B deoxycholate — Most of the experience with treating fungal infections in children has been with amphotericin B deoxycholate (conventional amphotericin). However, use of amphotericin B deoxycholate beyond the neonatal period has fallen out of favor because of drug-associated toxicities.
Although very young children tend to tolerate amphotericin B deoxycholate better than adults and have less nephrotoxicity, we suggest lipid amphotericin B for most children, including those with underlying renal insufficiency. Glomerular filtration rate (GFR) can be estimated with the following formula: GFR in mL/min per 1.73 m2 = (0.41 x height in cm) / serum creatinine in mg/dL [52]. The threshold for choosing lipid amphotericin B is individualized according to the need for other nephrotoxic drugs and trends in GFR over time.
●The recommended dose of amphotericin B deoxycholate for infants and children is 1 mg/kg intravenously (IV) per day.
Doses as high as 1.5 mg/kg per day have been used for meningitis or disseminated infection with focal liver or renal disease, but if higher doses are needed, changing to a lipid formulation may be warranted. (See 'Lipid amphotericin B' below.)
During therapy with amphotericin B deoxycholate, we suggest close monitoring of serum creatinine and electrolytes, including magnesium and potassium. Serum potassium should be ≥3 mmol/L (3 mEq/L) before administration of amphotericin B. Daily laboratory monitoring may be required. Lipid formulations of amphotericin or an echinocandin should be used if there is a rise in serum creatinine [49]. (See "Pharmacology of amphotericin B" and "Amphotericin B nephrotoxicity".)
Lipid amphotericin B — Lipid amphotericin B preparations (eg, liposomal amphotericin B, amphotericin B lipid complex [ABLC], amphotericin B cholesteryl sulfate complex [ABCD, not available in the United States]) are preferred for most children, including those who are receiving other nephrotoxic agents (eg, posttransplantation).
●The usual dose is 3 to 5 mg/kg IV once per day; 5 mg/kg per day may be warranted if C. glabrata or C. krusei is suspected or identified.
Although the incidence of decreased renal function and infusion-related adverse events with lipid formulations of amphotericin B are lower than rates observed with amphotericin B deoxycholate, these complications do occur. Therefore, during therapy with lipid formulations of amphotericin B, we suggest close monitoring of serum creatinine and electrolytes, including magnesium and potassium. Serum potassium should be ≥3 mmol/L (3 mEq/L) before administration of lipid formulations of amphotericin B. Daily laboratory monitoring may be required. An alternative antifungal agent (eg, echinocandin, azole) may be needed if there is a rise in serum creatinine or persistent derangement of serum electrolytes.
The pharmacokinetics of lipid formulations in children are similar to those in adults [53]. (See "Pharmacology of amphotericin B", section on 'Lipid-based amphotericin B formulations'.)
Azole antifungal agents — Azole antifungal agents available for the treatment of candidemia and invasive candidiasis in children include fluconazole and voriconazole. The safety and efficacy of other azole antifungal agents (eg, posaconazole, isavuconazole) in the treatment of invasive candidiasis in children have not been adequately studied.
●Fluconazole – Fluconazole is used more often than voriconazole in the treatment of invasive candidiasis in children. Risk factors for fluconazole resistance include neutropenia and recent azole use. Fluconazole should not be used if C. krusei is suspected or identified. C. krusei is intrinsically resistant to fluconazole [54].
•The recommended dose of fluconazole for children is 12 mg/kg IV per day [55].
Children require higher doses of fluconazole than adults because the half-life is shorter in children [55,56].
●Voriconazole – Voriconazole generally is used as step-down therapy in patients with invasive infection due to C. krusei and fluconazole-resistant, voriconazole-susceptible C. glabrata.
Voriconazole is not recommended for patients with urinary candidiasis because the active form is minimally excreted in the urine.
The recommended dose of voriconazole for children varies according to age [57,58]:
•<12 years – 9 mg/kg IV twice daily as a loading dose for the first day and then 8 mg/kg IV every 12 hours
•≥12 years – 6 mg/kg IV twice daily as a loading dose for the first day and then 4 mg/kg IV every 12 hours
Younger children require higher doses of voriconazole than older children and adults because of differing elimination patterns (linear elimination until approximately 12 years of age and nonlinear elimination thereafter) [57].
We suggest that serum voriconazole concentrations be monitored and dose adjusted to maintain trough concentration between 1 and 5 mcg/mL [59-61]. Trough concentrations >5 mcg/mL are associated with toxicity [60,61]. In observational studies, the pharmacokinetics of voriconazole have varied markedly from patient to patient [62-65]. In one study, voriconazole trough concentrations of >1 mcg/mL were associated with increased survival from invasive fungal infection (9 of 46 children had invasive candidiasis) [62]. Pending prospective studies confirming these results, it is prudent to maintain a trough concentration of at least 1 mcg/mL.
Echinocandins — Echinocandins include caspofungin, anidulafungin, and micafungin. Echinocandins are first-line agents for the treatment of invasive candidiasis in adults, based upon randomized trials demonstrating that they are as safe and effective as amphotericin B or fluconazole [16,66-68]. Although there are fewer data than in adults [69-76], echinocandins appear to be safe in children and are frequently used for suspected or confirmed invasive candidiasis.
Echinocandins are not recommended for candidiasis involving the eye, central nervous system (CNS), or urinary tract because they have relatively poor penetration at these sites.
General dosing guidelines for echinocandins in children are as provided below [71,77-82]. Dosing regimens have not been standardized in children. We suggest that echinocandin regimens be discussed with an infectious diseases pharmacist or clinician before starting therapy.
●Caspofungin – Caspofungin is available for the treatment of candidemia and invasive candidiasis in children ≥3 months of age. It is also available for empiric treatment of suspected fungal infection in children ≥3 months of age with fever and neutropenia. Caspofungin is dosed by body surface area rather than weight [70-73]:
•The recommended regimen is a single loading dose of 70 mg/m2 IV per day (maximum dose 70 mg), followed by 50 mg/m2 IV per day (maximum dose 70 mg)
Patients receiving other medications that induce drug clearance (eg, phenytoin, rifampin) may require increased doses (within the maximum dose of 70 mg).
•For patients ≥4 months of age, micafungin is available for the treatment of candidemia, acute disseminated candidiasis, Candida peritonitis and abscesses, esophageal candidiasis, and prophylaxis of Candida infections in patients undergoing hematopoietic cell transplantation [83].
The suggested dose varies according to the indication and clinical response. For the treatment of candidemia and invasive candidiasis, the initial dose is 2 mg/kg IV once daily (maximum 100 mg per day). Increased doses (up to 10 mg/kg per day) may be necessary in children <40 kg or those without an adequate clinical response to a lower dose [8,84,85].
•For infants <4 months of age, micafungin is available for the treatment of candidemia, acute disseminated candidiasis, and Candida peritonitis and abscesses, provided that the infants do not have meningoencephalitis and/or ocular dissemination [83].
The suggested dose is 4 mg/kg IV once daily [83]. Based on limited data, increased doses (up to 10 mg/kg per day) may be necessary for infants who do not respond to lower doses [84,85].
●Anidulafungin – For patients ≥1 month of age, anidulafungin is available for the treatment of candidemia, intra-abdominal abscess, and peritonitis [86].
The recommended regimen is a single loading dose of 3 mg/kg IV (maximum dose 200 mg) on day 1, followed by 1.5 mg/kg IV once daily (maximum dose 100 mg) for at least 14 days after the last positive culture [86].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Candidiasis".)
SUMMARY AND RECOMMENDATIONS
●The term candidemia describes the presence of Candida species in the blood. Candida in a blood culture should prompt a search for the source; it should never be viewed as a contaminant. (See 'Introduction' above.)
●Antifungal therapy should be initiated as soon as possible after Candida is identified in a blood culture. The choice of antifungal agent is guided by clinical features (age, hemodynamic stability/severity of illness, site of infection, baseline renal function, neutropenia) and the infecting Candida species (table 1). (See 'Choice of antifungal therapy' above.)
●We obtain daily or every other day blood cultures after initiating treatment for invasive candidiasis to confirm the date of sterilization, which affects the duration of treatment. (See 'Monitoring response' above.)
●For children with candidemia and no evidence of metastatic foci of infection, we generally treat for a minimum of two weeks after blood cultures become negative (table 1). For children with deep-seated invasive candidiasis (eg, endocarditis or other metastatic focus of infection), longer durations of therapy are necessary and should be determined in consultation with an infectious diseases clinician. (See 'Duration of antifungal therapy' above.)
●In addition to prompt initiation of antifungal therapy, we recommend that central venous catheters be removed in nonneutropenic children with candidemia if the catheter is thought to be the source and it can be removed safely. In patients with neutropenia, the gastrointestinal tract is a frequent source of invasive candidiasis, so the decision regarding catheter removal should be considered on an individual basis. (See 'Source control' above.)
●Even with appropriate therapy, the mortality of candidemia is approximately 9 to 40 percent. Factors associated with increased mortality include delayed initiation or inadequate dosing of antifungal therapy, inadequate source control, increased severity of illness, and immunosuppressive therapy. (See 'Outcome' above.)
ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge Carol A Kauffman, MD, who contributed to an earlier version of this topic review.
