Fungal Infections in the ICU: Missed Disease, Colonization, and the Cost of Overcalling Candida

Introduction

Fungal infections in the ICU occupy an uneasy clinical space. Some patients with invasive fungal disease are recognized late because blood cultures are incompletely sensitive for deep-seated invasive candidiasis, classic host factors for aspergillosis may be absent, or early manifestations may be difficult to distinguish from bacterial sepsis. Other patients receive antifungal therapy for colonization, low-probability biomarker results, or nonspecific clinical deterioration that ultimately proves unrelated to fungal infection.

Both errors are consequential. Delayed treatment of documented candidemia or strongly suspected invasive pulmonary aspergillosis can be dangerous. Unnecessary antifungal therapy can also cause harm through toxicity, drug interactions, selection pressure, diagnostic anchoring, cost, and prolonged treatment inertia. ICU clinicians encounter both underdiagnosis and overdiagnosis, usually in different patients and for different reasons.

The practical task is to estimate the probability of invasive fungal disease with discipline. This requires attention to host risk, culture site, specimen quality, imaging, source control, severity trajectory, recent antimicrobial exposure, local epidemiology, and antifungal safety. It also requires restraint when interpreting tests that are clinically useful but imperfect.

Why This Topic Matters Now

Critically ill populations increasingly include patients with complex immunosuppression, transplantation, hematologic malignancy treatment, advanced liver disease, prolonged broad-spectrum antibiotic exposure, renal replacement therapy, parenteral nutrition, central venous access, and postoperative abdominal complications. These factors may increase the probability of invasive candidiasis or invasive mold disease, but none is diagnostic independently.

The 2024 FUNDICU consensus definitions were developed to improve the consistency of invasive fungal disease classification among nonneutropenic ICU populations. They are valuable for research and provide useful diagnostic structure, but they do not replace patient-specific clinical assessment. (Bassetti et al., 2024).

Severe viral pneumonia has also changed the discussion surrounding aspergillosis. Influenza-associated pulmonary aspergillosis and COVID-19-associated pulmonary aspergillosis have demonstrated that severe viral lung injury may create an at-risk state even without classic neutropenic host factors. ICU clinicians should consider aspergillosis in selected patients with viral acute respiratory distress syndrome, corticosteroid exposure, unexplained respiratory deterioration, compatible imaging findings, tracheobronchitis, or supportive lower respiratory tract mycology. Diagnostic criteria, screening practices, and treatment thresholds nevertheless remain heterogeneous. (Feys et al., 2024; Koehler et al., 2021).

Antifungal stewardship has become increasingly important as well. C. auris can cause invasive infection, persistently colonize patients, spread within healthcare facilities, and exhibit resistance to multiple antifungal classes. Colonization should not be treated with antifungal drugs, but it requires appropriate infection-prevention precautions and communication during transitions of care. CDC continues to recommend an echinocandin as initial treatment for clinical C. auris infection in adults, together with susceptibility testing and close assessment of treatment response.

Current sepsis guidance reflects the same balance. The 2026 Surviving Sepsis Campaign suggests against empiric antifungal therapy for adults with sepsis or septic shock as a routine strategy. Empiric treatment may still be considered case by case in selected patients with substantial fungal risk, including immunosuppression, prolonged antimicrobial exposure, prolonged hospitalization, or an intra-abdominal source. Diagnostic workup, source control, daily reassessment, and de-escalation remain central.

The ICU Fungal Differential

For many adult ICUs, the dominant fungal concern is invasive candidiasis, including candidemia and deep-seated candidiasis. Risk may be increased by recent abdominal surgery, gastrointestinal perforation, anastomotic leak, necrotizing pancreatitis, parenteral nutrition, renal replacement therapy, central venous catheters, broad-spectrum antibiotic exposure, immunosuppression, multifocal Candida colonization, and prolonged critical illness.

Candidemia is not a benign culture result. Candida recovered from blood should be managed as clinically significant, with active antifungal therapy, repeat blood cultures, species identification, susceptibility testing, source evaluation, and source control.

In contrast, Candida recovered from nonsterile respiratory specimens in ventilated patients usually represents colonization rather than Candida pneumonia. Candida pneumonia is rare, and a decision to initiate antifungal therapy should not be based on respiratory culture results alone. (Pappas et al., 2016).

Invasive pulmonary aspergillosis is less common than invasive candidiasis in many general ICUs, but it carries serious consequences when present. It should be considered in appropriate patients with hematologic malignancy, transplantation, prolonged or high-dose corticosteroid exposure, severe chronic obstructive pulmonary disease, influenza, COVID-19, advanced liver disease, persistent pulmonary infiltrates, nodules, cavitation, airway plaques, hemoptysis, tracheobronchitis, or unexplained respiratory failure despite appropriate antibacterial treatment.

Other fungal infections require syndrome-specific and geographic reasoning. Mucormycosis should be considered in patients with diabetic ketoacidosis, profound immunosuppression, necrotic sinus or pulmonary disease, or breakthrough mold infection during voriconazole exposure. Histoplasmosis, blastomycosis, coccidioidomycosis, cryptococcosis, and other endemic or opportunistic mycoses depend on geography, exposure history, immune status, and clinical syndrome. They should not become routine empiric targets in every ICU patient, but they should remain in the differential when the epidemiology and presentation fit.

The Problem of Missing Invasive Candidiasis

Invasive candidiasis is not synonymous with candidemia. A positive blood culture is highly specific, but blood cultures are incompletely sensitive for deep-seated invasive candidiasis. A patient with postoperative peritonitis, persistent shock, broad-spectrum antibiotic exposure, and yeast from an appropriately obtained intra-abdominal specimen has a very different probability of invasive disease than a patient with Candida isolated from sputum.

The sampling method matters. Candida recovered from an operative specimen or a newly placed drain in a patient with compatible intra-abdominal infection may support the diagnosis of intra-abdominal candidiasis. Cultures obtained from drains that have been in place for a prolonged period are more difficult to interpret because colonization is common. FUNDICU research definitions similarly distinguish operative, percutaneous, and recently placed drain specimens from specimens obtained through chronic drains. (Bassetti et al., 2024).

The ICU presentation is rarely distinctive. Fever, leukocytosis, vasopressor requirement, renal dysfunction, thrombocytopenia, and increased inflammatory markers do not reliably distinguish bacterial sepsis from fungal infection. Risk factors increase pretest probability, but they do not establish the diagnosis. Colonization may increase concern, particularly when multifocal or accompanied by high-risk abdominal disease, but colonization alone is not invasive infection.

When candidemia is documented, an echinocandin is preferred as initial therapy for most critically ill adults, especially while species identification and susceptibility results are pending. Fluconazole may be an acceptable initial alternative in selected patients who are not critically ill and are considered unlikely to have a fluconazole-resistant isolate. Step-down from an echinocandin to fluconazole generally requires clinical stability, a susceptible isolate, and documented bloodstream clearance. (Cornely et al., 2025; Pappas et al., 2016).

Follow-up blood cultures should be obtained every day or every other day until candidemia has cleared. For uncomplicated candidemia without metastatic complications, treatment is generally continued for 14 days after documented bloodstream clearance and resolution of signs and symptoms attributable to candidemia. Deep-seated infection, endocarditis, endophthalmitis, osteomyelitis, septic thrombophlebitis, abscess, or a persistent source requires a longer and syndrome-specific course. (Pappas et al., 2016).

Source control is essential. Central venous catheter removal should be considered as early as possible when the catheter is the presumed source and can be removed safely, with the decision individualized to the patient. Intra-abdominal candidiasis requires drainage and repair of a leak or perforation when feasible. Persistent candidemia should prompt reassessment for a retained infected catheter, septic thrombophlebitis, endocarditis, deep abscess, urinary source, inadequate drug exposure, resistant species, or another uncontrolled focus.

The Problem of Overcalling Candida

The most common overcall is treatment of Candida recovered from the respiratory tract. Candida is frequently isolated from endotracheal aspirates, bronchoalveolar lavage fluid, or sputum in mechanically ventilated patients. True Candida pneumonia is rare and generally requires histopathologic evidence for a firm diagnosis. Treating respiratory Candida colonization rarely addresses the cause of respiratory failure and may distract from bacterial, viral, inflammatory, aspiration-related, thrombotic, cardiogenic, or ventilator-associated causes.

Another error is treating colonization alone. Multifocal Candida colonization may contribute to a risk assessment, but it should not automatically trigger antifungal therapy. The same principle applies to C. auris colonization. Colonization may require contact or other transmission-based precautions, environmental cleaning, communication during transfer, and public health notification according to local requirements, but antifungal treatment is reserved for clinical infection.

A third overcall is treating a biomarker rather than the patient. Beta-D-glucan may support a diagnosis of invasive fungal disease in an appropriate clinical context, but false-positive and false-negative results occur. Galactomannan, Aspergillus polymerase chain reaction, fungal culture, and histopathology also require interpretation according to host risk, clinical syndrome, specimen quality, sampling site, background prevalence, and prior antifungal exposure.

A positive result in a low-probability patient should not automatically lead to prolonged therapy. A negative result should not override a high-probability syndrome when sampling is inadequate, antifungal therapy preceded testing, or disease is localized. The 2026 Surviving Sepsis Campaign suggests against using Candida biomarkers routinely to determine whether empiric antifungal therapy should be initiated, although individualized use may be considered in selected high-risk patients.

Diagnostic Strategy: Use Probability, Not Reflex Testing

A useful ICU diagnostic strategy begins with pretest probability. Before ordering or acting on fungal testing, clinicians should consider three questions.

First, is there a plausible host or exposure risk? Relevant examples include recent abdominal surgery, gastrointestinal perforation, necrotizing pancreatitis, parenteral nutrition, renal replacement therapy, central venous access, prolonged broad-spectrum antibiotic exposure, transplantation, hematologic malignancy, prolonged corticosteroid use, severe chronic obstructive pulmonary disease, severe influenza or COVID-19, and known colonization with a resistant fungal organism.

Second, is there a compatible syndrome? Candidemia may present as persistent sepsis, catheter-associated bloodstream infection, endophthalmitis, endocarditis, hepatosplenic disease, thrombophlebitis, or deep-seated abdominal infection. Aspergillosis may present with worsening respiratory failure, nodules, cavitation, airway plaques, tracheobronchitis, hemoptysis, pleuritic symptoms, or refractory pulmonary infiltrates.

Third, does the microbiology come from a clinically meaningful site? Candida in blood is clinically significant. Candida recovered from an appropriately obtained normally sterile specimen may support invasive candidiasis when the clinical syndrome is compatible. Candida in sputum usually represents colonization. Aspergillus recovered from bronchoalveolar lavage in a compatible host and syndrome is more persuasive than Aspergillus recovered from a low-quality upper airway specimen.

Serum galactomannan may be less sensitive in nonneutropenic ICU aspergillosis than in classic hematologic populations. Bronchoalveolar lavage galactomannan and Aspergillus polymerase chain reaction may be useful when bronchoscopy is feasible and safe, but neither should be interpreted independently of host factors, imaging, microscopy, culture, and specimen quality. (Hage et al., 2019).

Testing should be paired with reassessment. If empiric treatment is started in a high-risk, unstable patient, the care team should define which findings would support continuation, narrowing, changing, or discontinuing treatment. Persistent severity alone is not proof of fungal disease.

Table 1. Practical Interpretation of Common ICU Findings

Finding	Interpretation and practical response
Candida in blood culture	Treat as candidemia. Start active therapy, repeat blood cultures, identify the species, perform susceptibility testing, assess the source, and pursue source control.
Candida in respiratory culture	Usually colonization. Do not treat solely on this basis. Reassess the pulmonary differential and specimen quality.
Candida from an intra-abdominal specimen	May indicate intra-abdominal candidiasis when obtained operatively, percutaneously, or from a recently placed drain in a compatible syndrome. Interpret chronic-drain cultures cautiously.
Positive beta-D-glucan	Supportive only in context. Reassess pretest probability and seek site-specific microbiologic or radiographic evidence.
Bronchoalveolar lavage galactomannan or Aspergillus PCR	May support pulmonary aspergillosis in a compatible host and syndrome. Integrate with imaging, microscopy, culture, bronchoscopy findings, and prior antifungal exposure.
C. auris screening positivity	Indicates colonization and transmission risk, not necessarily infection. Apply infection-control precautions and treat only if clinical infection is present.

Therapeutic Considerations

Candidemia and Invasive Candidiasis

For most critically ill adults with candidemia, an echinocandin is the preferred initial agent. Micafungin, caspofungin, and anidulafungin provide activity against most clinically important Candida species and generally have fewer cytochrome P450-mediated interactions than triazoles. Species identification and susceptibility testing should guide continued therapy, step-down treatment, and resistance management.

Fluconazole may be reasonable as initial therapy only in selected patients who are not critically ill and are unlikely to have a fluconazole-resistant isolate. It is not the default ICU choice when species and susceptibility are unknown. Step-down from an echinocandin to fluconazole should generally wait until the patient is clinically stable, the isolate is susceptible, and repeat blood cultures have become negative.

Rezafungin is FDA approved for adults with candidemia or invasive candidiasis who have limited or no alternative treatment options. Its labeling should not be generalized to Candida endocarditis, osteomyelitis, meningitis, endophthalmitis, or other syndromes excluded from its clinical trials. It should not be presented as an interchangeable default substitute for established daily echinocandins in every ICU patient.

Ophthalmologic Evaluation

Ophthalmologic evaluation remains an area of guideline discordance. The 2016 IDSA candidiasis guideline recommends a dilated ophthalmologic examination for patients with candidemia, generally within the first week for nonneutropenic patients. The American Academy of Ophthalmology recommends against routine ophthalmologic consultation after systemic Candida septicemia in asymptomatic patients and favors targeted examination when ocular symptoms or signs are present. (Breazzano et al., 2022; Pappas et al., 2016).

A practical approach is to align management with local infectious diseases and ophthalmology policy while accounting for the patient’s ability to report visual symptoms. Targeted consultation is particularly important when ocular symptoms or signs are present, candidemia persists, ocular dissemination is otherwise suspected, or the patient cannot reliably participate in symptom assessment.

Empiric Therapy in Sepsis or Septic Shock

Routine empiric antifungal therapy for all septic ICU patients is not supported. The 2026 Surviving Sepsis Campaign suggests against empirical antifungal therapy in adults with sepsis or septic shock, based on low-certainty evidence. The guideline permits case-by-case treatment in selected patients with substantial fungal risk factors.

The 2025 ATS guideline similarly makes a conditional recommendation against routine prophylactic or empiric Candida-directed antifungal therapy in nonneutropenic, nontransplant critically ill patients. This recommendation is based on low-quality evidence and should not be interpreted as prohibiting treatment when invasive candidiasis is sufficiently likely.

Empiric therapy may be reasonable in selected patients with persistent or worsening sepsis despite appropriate antibacterial therapy and source-control efforts, especially when several risk factors converge. Examples include an abdominal catastrophe, recent major gastrointestinal surgery, necrotizing pancreatitis, parenteral nutrition, renal replacement therapy, central venous access, multifocal Candida colonization, immunosuppression, prolonged antibiotic exposure, or local epidemiology indicating substantial risk.

Even then, therapy should be time-limited and reassessment-driven. Before treatment is started, clinicians should identify what additional information would justify continuation, such as candidemia, Candida from an appropriately obtained sterile-site specimen, a convincing intra-abdominal source, or persistently high clinical probability despite incomplete testing. Clinical improvement after treatment of an alternative source, negative cultures in a low-probability syndrome, or a persuasive nonfungal diagnosis should support discontinuation.

The EMPIRICUS randomized trial evaluated empiric micafungin in adults with ICU-acquired sepsis, Candida colonization, and multiple organ failure. Micafungin did not improve the primary outcome of survival free of proven invasive fungal infection at day 28. It reduced the incidence of newly diagnosed invasive fungal infection, but this secondary finding did not translate into improvement in the primary clinical outcome. The trial argues against broad reflexive treatment of colonized ICU patients but does not establish that empiric therapy is never appropriate. (Timsit et al., 2016).

Candida auris

C. auris requires species-level identification and infection-control awareness. It can cause candidemia and other invasive infections, but it can also colonize skin and other body sites for prolonged periods. Colonization should not be treated with antifungal drugs. It should prompt appropriate infection-control precautions, environmental disinfection, transfer communication, and reporting according to local and public health requirements.

For clinical C. auris infection in adults, CDC recommends an echinocandin as initial therapy while susceptibility testing is pending. Susceptibility testing is important because resistance patterns vary and reports of echinocandin-resistant and pan-resistant isolates are increasing. Persistent candidemia during echinocandin therapy should prompt reassessment of source control, adherence and drug exposure, species identification, susceptibility, anatomic complications, and the need for infectious diseases consultation.

Invasive Pulmonary Aspergillosis

Invasive pulmonary aspergillosis in the ICU may not resemble classic neutropenic disease. Nonneutropenic patients may lack a typical halo sign and instead have nonspecific consolidation, airway disease, nodules, cavitation, or progressive infiltrates in the setting of viral lung injury, corticosteroid exposure, structural lung disease, liver failure, or another critical illness.

Voriconazole has historically been preferred for primary treatment of invasive aspergillosis, supported by a randomized trial against amphotericin B deoxycholate. That trial primarily involved patients with hematologic disease, so its findings should not be overgeneralized without considering differences in ICU populations. (Herbrecht et al., 2002).

Isavuconazole is an accepted alternative for invasive pulmonary aspergillosis. In the SECURE trial, isavuconazole was noninferior to voriconazole for all-cause mortality through day 42 in patients with invasive mold disease and was associated with fewer selected drug-related adverse effects. Patient-specific selection should consider hepatic function, interaction burden, absorption, route of administration, QT effects, prior azole exposure, local resistance, and susceptibility data. (Maertens et al., 2016).

Liposomal amphotericin B is an important alternative when azole resistance, intolerance, severe drug interactions, unreliable absorption, or mucormycosis is a concern. Suspected mucormycosis requires urgent diagnostic assessment, early active therapy, and surgical evaluation when anatomically appropriate. Voriconazole does not provide reliable activity against Mucorales and should not be used as sole treatment when mucormycosis is a meaningful possibility. (Cornely et al., 2019).

Combination therapy with a mold-active triazole plus an echinocandin remains nuanced. The 2025 ATS guideline conditionally supports either initial triazole monotherapy or initial triazole-echinocandin combination therapy for proven or probable invasive pulmonary aspergillosis, based on low-quality evidence. The evidence base is derived predominantly from patients with hematologic malignancy or hematopoietic stem-cell transplantation, so applicability to other ICU populations remains uncertain.

Combination therapy should not be initiated reflexively solely because of a positive Aspergillus test. The decision should integrate whether disease is proven or probable, illness severity, host factors, diagnostic confidence, prior antifungal exposure, resistance concerns, interaction burden, and treatment toxicity. Available evidence does not clearly identify which subgroups derive the greatest benefit from combination treatment.

Therapeutic drug monitoring is important for voriconazole and often for posaconazole because of pharmacokinetic variability, exposure-related toxicity, absorption concerns, and drug interactions. Routine therapeutic drug monitoring is not universally required for isavuconazole, but measurement may be considered in selected situations involving absorption uncertainty, extreme body size, organ support, unexpected toxicity, treatment failure, or major interactions.

Table 2. Initial Therapy and Monitoring by Syndrome

Syndrome	Initial approach and limitations
Documented candidemia	Use an echinocandin for most critically ill adults. Repeat blood cultures, identify the species, obtain susceptibility testing, address source control, and step down only when clinically appropriate.
High-probability invasive candidiasis with sepsis or shock	Consider an empiric echinocandin selectively. Reassess early and daily. Discontinue if probability falls and no supportive evidence emerges.
Clinical C. auris infection	Use an echinocandin initially in most adults while susceptibility results are pending. Do not treat colonization. Maintain infection-control precautions.
Proven or probable pulmonary aspergillosis	Use voriconazole or isavuconazole in appropriate patients. Liposomal amphotericin B is an alternative when azoles are unsuitable. Combination therapy may be considered selectively, but evidence is limited.
Possible mucormycosis	Begin liposomal amphotericin B when clinical probability justifies treatment while urgent diagnostic and surgical evaluation proceeds. Avoid relying on voriconazole monotherapy.

Antifungal Safety and Monitoring

Antifungal selection in the ICU is not solely a microbiology decision. It is also an organ-support, interaction, contraindication, formulation, exposure, and monitoring decision. Safety profiles differ within antifungal classes, so class-wide generalizations should be avoided.

Table 3. Antifungal Safety and Monitoring

Agent or class	Important safety and monitoring considerations
Echinocandins	Hypersensitivity, histamine-mediated infusion reactions, and hepatic laboratory abnormalities can occur. Review the specific product label because interactions and hepatic recommendations differ. Anidulafungin products containing fructose are contraindicated in known or suspected hereditary fructose intolerance.
Fluconazole	Review renal function because dosage adjustment is generally required for repeated dosing in renal impairment. Monitor hepatic function when clinically appropriate and evaluate QT risk and CYP-mediated interactions.
Voriconazole	Monitor hepatic function, clinically significant interactions, QT risk, visual or neurologic effects, photosensitivity, and severe cutaneous reactions. Therapeutic drug monitoring is often appropriate. Correct clinically important electrolyte abnormalities that may increase arrhythmia risk.
Isavuconazonium	Review hepatic function, infusion reactions, pregnancy-related risk, and major CYP3A4 interactions. Isavuconazole shortens the QT interval and is contraindicated in familial short QT syndrome.
Liposomal amphotericin B	Monitor renal function, potassium, magnesium, complete blood count, and infusion reactions. Nephrotoxicity is generally less frequent than with amphotericin B deoxycholate but remains clinically important.
Rezafungin	Reserve use for its labeled population of adults with limited or no alternative options for candidemia or invasive candidiasis. Monitor hypersensitivity and infusion reactions. Do not extrapolate efficacy to Candida endocarditis, osteomyelitis, meningitis, or other excluded syndromes.

Current DailyMed labeling confirms the hereditary fructose intolerance contraindication for anidulafungin, the familial short QT contraindication and CYP3A4 restrictions for isavuconazonium, the major hepatic, QT, visual, neurologic, cutaneous, and interaction concerns for voriconazole, and the limited labeled population and syndrome exclusions for rezafungin.

A Practical ICU Approach

A practical approach begins by assigning the patient to a low-, intermediate-, or high-probability category and updating that assessment as new information emerges.

In low-probability patients, Candida colonization, respiratory Candida, or a single nonspecific biomarker should not drive antifungal therapy. The better approach is to evaluate more likely causes of deterioration, optimize antibacterial stewardship, review devices and medications, and avoid diagnostic anchoring.

In intermediate-probability patients, targeted diagnostics are often more valuable than immediate indefinite therapy. Useful steps may include repeat blood cultures, reassessment of catheter and abdominal sources, abdominal imaging, bronchoscopy when feasible, bronchoalveolar lavage galactomannan or Aspergillus polymerase chain reaction for suspected aspergillosis, and infectious diseases consultation.

In patients with a high probability of invasive fungal disease, especially those with candidemia, yeast from an appropriately obtained sterile-site specimen, a compatible intra-abdominal catastrophe, or strong suspicion of invasive pulmonary aspergillosis, antifungal therapy should not be delayed solely while awaiting perfect microbiologic confirmation. Early treatment should be paired with source control, diagnostic workup, species identification, susceptibility testing, toxicity monitoring, and a de-escalation plan.

The daily ICU question should be explicit: What is the current probability of invasive fungal disease, and what new evidence would change management today? This approach helps prevent both therapeutic delay and antifungal inertia.

Limitations of the Evidence

The ICU fungal evidence base remains imperfect. Randomized trial data are limited for empiric antifungal therapy in heterogeneous sepsis populations. Diagnostic definitions for invasive fungal disease among nonneutropenic ICU patients have evolved, and newer consensus definitions improve research consistency but do not replace clinical judgment. Biomarker performance varies by host population, assay, threshold, prior antifungal exposure, sampling site, and background prevalence.

Viral-associated aspergillosis remains an evolving area. Influenza-associated and COVID-19-associated pulmonary aspergillosis literature supports vigilance in selected patients with severe viral pneumonia, but diagnostic criteria, screening strategies, and treatment thresholds continue to vary among centers.

Observational associations between delayed antifungal treatment and poor outcomes are clinically concerning but vulnerable to confounding by illness severity, diagnostic delay, source control, treatment selection, and immortal-time bias. These associations support urgency when invasive disease is sufficiently likely. They do not justify indiscriminate empiric antifungal therapy for every episode of ICU deterioration.

Several antifungal treatment recommendations are based partly on studies conducted in hematologic malignancy or transplant populations. Extrapolation to general, surgical, hepatic, pulmonary, or viral-associated ICU populations requires caution. Surrogate outcomes such as biomarker clearance, reduction in colonization, or prevention of newly diagnosed fungal infection should not be equated automatically with improved survival or patient-centered benefit.

Future Directions

The next phase of ICU fungal care should move beyond single-test decision-making. Better prediction strategies may combine host factors, colonization burden, local epidemiology, imaging, culture data, fungal biomarkers, molecular testing, antimicrobial exposure, source-control response, and longitudinal clinical trajectory.

Rapid species identification and susceptibility testing will become increasingly important as C. auris and resistant Candida species expand. Diagnostic platforms should be evaluated for their effects on treatment decisions, antifungal exposure, time to appropriate therapy, organ toxicity, length of stay, and patient-centered outcomes rather than analytical performance alone.

Future trials should focus on better-defined high-risk ICU phenotypes rather than broad, undifferentiated sepsis populations. Priority areas include blood-culture-negative deep-seated candidiasis, viral-associated aspergillosis, individualized combination therapy, resistant C. auris infection, and safe antifungal de-escalation.

Stewardship programs should measure more than antifungal days. Relevant outcomes include diagnostic quality, time to source control, species and susceptibility documentation, avoidable toxicity, resistance patterns, delayed treatment of confirmed disease, and clinical outcomes.

References

American Academy of Ophthalmology. (2022). Recommendations on screening for endogenous Candida endophthalmitis.

Astellas Pharma US, Inc. (2025). AmBisome (amphotericin B) liposome for injection: Prescribing information. DailyMed.

Bassetti, M., Giacobbe, D. R., Agvald-Ohman, C., Akova, M., Alastruey-Izquierdo, A., Arikan-Akdagli, S., et al. (2024). Invasive fungal diseases in adult patients in intensive care units (FUNDICU): 2024 consensus definitions from ESGCIP, EFISG, ESICM, ECMM, MSGERC, ISAC, and ISHAM. Intensive Care Medicine, 50(4), 502-515. PMID: 38512399.

Breazzano, M. P., Bond, J. B., Bearelly, S., Kim, D. H., Donahue, S. P., Lum, F., & Olsen, T. W. (2022). American Academy of Ophthalmology recommendations on screening for endogenous Candida endophthalmitis. Ophthalmology, 129(1), 73-76. PMID: 34293405.

Centers for Disease Control and Prevention. (2024). Clinical treatment of C. auris infections.

Centers for Disease Control and Prevention. (2026). Clinical overview of Candida auris.

Centers for Disease Control and Prevention. (2026). Tracking Candida auris.

Cornely, O. A., Alastruey-Izquierdo, A., Arenz, D., Chen, S. C. A., Dannaoui, E., Hochhegger, B., et al. (2019). Global guideline for the diagnosis and management of mucormycosis: An initiative of the European Confederation of Medical Mycology in cooperation with the Mycoses Study Group Education and Research Consortium. The Lancet Infectious Diseases, 19(12), e405-e421. PMID: 31699664.

Cornely, O. A., Sprute, R., Bassetti, M., Chen, S. C. A., Groll, A. H., Kurzai, O., et al. (2025). Global guideline for the diagnosis and management of candidiasis: An initiative of the ECMM in cooperation with ISHAM and ASM. The Lancet Infectious Diseases, 25(5), e280-e293. PMID: 39956121.

Epelbaum, O., Marinelli, T., Haydour, Q., Pennington, K. M., Evans, S. E., Carmona, E. M., et al. (2025). Treatment of invasive pulmonary aspergillosis and preventive and empirical therapy for invasive candidiasis in adult pulmonary and critical care patients: An official American Thoracic Society clinical practice guideline. American Journal of Respiratory and Critical Care Medicine, 211(1), 34-53. PMID: 39556361.

Feys, S., Carvalho, A., Clancy, C. J., Gangneux, J. P., Hoenigl, M., Lagrou, K., Rijnders, B. J. A., Seldeslachts, L., Vanderbeke, L., van de Veerdonk, F. L., Verweij, P. E., & Wauters, J. (2024). Influenza-associated and COVID-19-associated pulmonary aspergillosis in critically ill patients. The Lancet Respiratory Medicine, 12(9), 728-742. PMID: 39025089.

Hage, C. A., Carmona, E. M., Epelbaum, O., Evans, S. E., Gabe, L. M., Haydour, Q., et al. (2019). Microbiological laboratory testing in the diagnosis of fungal infections in pulmonary and critical care practice: An official American Thoracic Society clinical practice guideline. American Journal of Respiratory and Critical Care Medicine, 200(5), 535-550. PMID: 31469325.

Herbrecht, R., Denning, D. W., Patterson, T. F., Bennett, J. E., Greene, R. E., Oestmann, J. W., et al. (2002). Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. The New England Journal of Medicine, 347(6), 408-415. PMID: 12167683.

Koehler, P., Bassetti, M., Chakrabarti, A., Chen, S. C. A., Colombo, A. L., Hoenigl, M., Klimko, N., Lass-Florl, C., Oladele, R. O., & Vinh, D. C. (2021). Defining and managing COVID-19-associated pulmonary aspergillosis: The 2020 ECMM/ISHAM consensus criteria for research and clinical guidance. The Lancet Infectious Diseases, 21(6), e149-e162. PMID: 33333012.

Maertens, J. A., Raad, I. I., Marr, K. A., Patterson, T. F., Kontoyiannis, D. P., Cornely, O. A., et al. (2016). Isavuconazole versus voriconazole for primary treatment of invasive mould disease caused by Aspergillus and other filamentous fungi: The SECURE phase 3 noninferiority trial. The Lancet, 387(10020), 760-769. PMID: 26684607.

Melinta Therapeutics, LLC. (2026). Rezzayo (rezafungin for injection): Prescribing information. DailyMed.

Pappas, P. G., Kauffman, C. A., Andes, D. R., Clancy, C. J., Marr, K. A., Ostrosky-Zeichner, L., Reboli, A. C., Schuster, M. G., Vazquez, J. A., Walsh, T. J., Zaoutis, T. E., & Sobel, J. D. (2016). Clinical practice guideline for the management of candidiasis: 2016 update by the Infectious Diseases Society of America. Clinical Infectious Diseases, 62(4), e1-e50. PMID: 26679628.

Patterson, T. F., Thompson, G. R., III, Denning, D. W., Fishman, J. A., Hadley, S., Herbrecht, R., et al. (2016). Practice guidelines for the diagnosis and management of aspergillosis: 2016 update by the Infectious Diseases Society of America. Clinical Infectious Diseases, 63(4), e1-e60. PMID: 27365388.

Pfizer Inc. (2024). Eraxis (anidulafungin) for injection: Prescribing information. DailyMed.

Pfizer Inc. (2025). Cresemba (isavuconazonium sulfate): Prescribing information. DailyMed.

Prescott, H. C., Antonelli, M., Alhazzani, W., Moller, M. H., Alshamsi, F., Azevedo, L. C. P., et al. (2026). Surviving Sepsis Campaign: International guidelines for management of sepsis and septic shock 2026. Intensive Care Medicine, 52(5), 863-936. PMID: 41870560.

Reboli, A. C., Rotstein, C., Pappas, P. G., Chapman, S. W., Kett, D. H., Kumar, D., Betts, R., Wible, M., Goldstein, B. P., Schranz, J., Krause, D. S., & Walsh, T. J. (2007). Anidulafungin versus fluconazole for invasive candidiasis. The New England Journal of Medicine, 356(24), 2472-2482. PMID: 17568028.

Sandoz Inc. (2025). Caspofungin acetate for injection: Prescribing information. DailyMed.

Timsit, J. F., Azoulay, E., Schwebel, C., Charles, P. E., Cornet, M., Souweine, B., Klouche, K., Jaber, S., Trouillet, J. L., Bruneel, F., Argaud, L., Cousson, J., Meziani, F., Gruson, D., Paris, A., et al. (2016). Empirical micafungin treatment and survival without invasive fungal infection in adults with ICU-acquired sepsis, Candida colonization, and multiple organ failure: The EMPIRICUS randomized clinical trial. JAMA, 316(15), 1555-1564. PMID: 27706483.

Thompson, G. R., III, Soriano, A., Cornely, O. A., Kullberg, B. J., Ostrosky-Zeichner, L., Vazquez, J. A., et al. (2023). Rezafungin versus caspofungin for treatment of candidemia and invasive candidiasis: The ReSTORE phase 3 trial. The Lancet, 401(10370), 49-59. PMID: 36442484.

U.S. National Library of Medicine. (2026). Fluconazole tablets: Prescribing information. DailyMed.

U.S. National Library of Medicine. (2026). Voriconazole tablets: Prescribing information. DailyMed.

Verweij, P. E., Rijnders, B. J. A., Bruggemann, R. J. M., Azoulay, E., Bassetti, M., Blot, S., et al. (2020). Review of influenza-associated pulmonary aspergillosis in ICU patients and proposal for a case definition. Intensive Care Medicine, 46(8), 1524-1535. PMID: 32572532.

Xellia Pharmaceuticals USA LLC. (2026). Micafungin for injection: Prescribing information. DailyMed.