Recognizing IFI is often difficult because of non-specific signs and symptoms where fever may be the only symptom. Therefore, there should be a high index of suspicion in patients at increased risk of IFI. Mortality is high in those who develop IFI and is influenced by the type of fungus, site of disease and presence of ongoing immunosuppression. Timely diagnosis of IFI is challenging due to difficulty in obtaining clinical sample, need for anaesthesia to perform certain diagnostic procedures, and insufficient robust data on usefulness of fungal biomarkers and molecular detection tests. Definitive diagnosis by tissue microscopy and culture is difficult in the early stages of infection. Using host factor and microbiological and clinical criteria, current definitions exist for proven, probable and possible invasive fungal infection [5]. In practice, IFI is often suspected in those with persistent fever and neutropenia who fail to respond to broad spectrum antibacterial agents within three to five days, and in whom a reasonable attempt has been made to exclude bacterial and viral infections. As mortality is high, early and aggressive treatment of suspected infections with empiric antifungal therapy is common. As many patients with persistent fever and neutropenia on empirical antifungal therapy do not have a fungal infection, there is increasing interest in biological markers and radiological imaging to identify those at highest risk of IFI [6]. There are currently insufficient data to support this approach in children. Conventional amphotericin B deoxycholate is a broad spectrum antifungal agent that has previously been the treatment of choice for most IFI. Its use is complicated by nephrotoxicity and infusion-related reactions. More recently, less toxic lipid formulations of amphotericin B have become available. Triazole derivatives (fluconazole, itraconazole, voriconazole, posaconazole and ravuconazole) and echinocandins (caspofungin, micafungin and anidulafungin) are the other available agents. The role of biological agents e.g. antiHSP90 monoclonal antibodies, Mycograb is being explored. The increasing use of combination antifungal therapy adds further complexity to treatment decisions [7]. Knowledge of predisposing condition, likely organisms, risk factors, relevant clinical signs and symptoms of IFI is important for enabling early diagnosis of IFI and improving disease outcomes [8] (Web Table I ).

Invasive fungal infections cooperative group (IFICG) of European Organization for Research and Treatment for Cancer (EORTC) and Mycology Study Group (MSG) of National Institute of Allergy and Infectious Diseases (NIAID) have published standard definitions of IFI [5].

Proven IFI: The presence of fungal elements either as mould/yeast in deep tissues of biopsy/needle aspirates that is confirmed on culture and histopathological examination.

Probable and Possible IFI: Based on host factor and microbiological and clinical criteria and indirect tests [9].

Fungi having potential to cause IFI include Yeasts (Candida spp, Cryptococcus spp), moulds (Aspergillus spp, Fusarium spp, Scedosporium prolificans, Mucor, Rhizopus, and Rhizomucor Absidia), and dimorphic fungi (Histoplasma capsulatum, Coccidioide simmitis, Blastomyces dermatitidis, Paracoccidioides spp, Sporothrix spp and Penicillium marneffii).

Microscopy (direct and histopathology) and culture are the cornerstones for the diagnosis of IFI. Direct examination implies the use of 10% KOH–calcofluor mount. Microscopic examination may be less sensitive than culture and hence, a negative result does not rule out fungal infection. Positive direct microscopy of yeast or hyphae from a sterile site must be considered significant, even if the laboratory is unable to culture the fungus. Demonstration of septate hyphae from respiratory secretion in immunocompromised patient is suggestive of aspergillus, fusarium and scedosporlum, and broad aseptate hyphae for invasive mucormycosis. India ink staining of cerebrospinal fluid for diagnosing cryptococcal meningitis has a sensitivity of 60%. Pneumocystis jirovecii is detected by microscopy on Giemsa staining, which demonstrates the nuclei of trophozoites and intracystic stages, and silver stains, which display the cyst walls. The finding of 2- to 4 µm, oval, narrow-based budding yeasts allows a tentative diagnosis of histoplasmosis. Other organisms which can mimic the appearance of H. capsulatum in tissues includes Candida glabrata, Penicillium marneffi, Leshmania donovani bodies. Culture represents the gold standard in diagnosing fungal diseases.

Invasive candidiasis (IC) (both albicans and non-albican) is a frequent infection in hospital settings. Blood culture is considered as gold standard for diagnosis of candidemia [10,11] with sensitivity between 21% and 71%; lower in neonates and infants [12]. The European Society of Clinical Microbiology and Infectious Diseases　/ Fungal Infection Study Group ESCMID/EFISG guideline recommends collection of three blood samples with a minimum volume of 2 mL each in children with a bodyweight <2 kg and 6 mL each between 2 and 12 kg to ensure a 50-75% sensitivity to detect Candida [13]. Blood cultures are further limited by slow turn-around time and are negative in deep seated Candida infections including hepatosplenic candidiasis. Aspergillus species are rarely cultured from blood samples, except Apergillus fumigatus and Aspergillus terreus [14]. The sensitivity of blood culture for invasive Aspergillosis is ~10% and ~40% in disseminated fusariosis. The culture positivity of broncho-alveolar lavage (BAL) is 34% - 50% [15,16]. The tissue culture positivity in mucormycosis is ~50% [17]. Histopathological examination helps in differen-tiating different classes of moulds with an accuracy of less than 80% and need of culture for confirmation. The antifungal susceptibility pattern of the isolates guide the therapy.

Lateral flow assay: A point of care assay has been recently FDA approved for diagnosis of invasive aspergillosis. The antigen is Aspergillus specific manoprotein that is detected by monoclonal antibody. It has a rapid turnaround time and does not cross react with other moulds [24]. According to a recent meta-analysis, the overall sensitivity and specificity of 68% and 87% in serum and of 86% and 93% in BAL is observed when compared with GM [25]. The test has not yet been evaluated in pediatric populations.

Mannan antigen (Mn) and anti-Mannan antibody (a-Mn): Detection of mannan antigen may be helpful in serum of hepatosplenic candidiasis and Candida meningitis where blood culture is rarely positive [31]. The sensitivity and specificity of Mn/a-Mn antibodies combination is 83% and 86%, respectively. The sensitivity depends on type of Candida spp (species specific) viz 100% in Candida albicans, 50% in Candida krusei and Candida kefyr and 40% in Candida parapsilosis [32]. Presently, utility of this assay is limited because of several disadvantages like rapid clearance of the Candida Mn from serum, and cross-reactions due to colonization with Candida spp.

Markers for Cryptoccocosis and Histoplasmosis: Diagnosis of Cryptococcal meningitis or disseminated cryptococcosis is possible by demonstration of encapsulated yeast cells on India ink, culture or detection of Cryptococcal antigen (CrAg). The detection of CrAg in serum or CSF can be done by latex agglutination test (LA) and enzyme immunoassay (EIA). The sensitivity and specificity of LA varies from 93%-100% and 93%- 98 %. False-positive findings have been reported in cases of Trichosporon spp., Capnocytophaga spp., or Stomatococcus spp. infections or detergents. Recently,a point of care assay, an immune chromatographic lateral flow assay (CrAg LFA; Immuno-Mycologics, Norman, OK, USA) has been designed which has a sensitivity and specificity of 98% [21]. It is cheaper, has shorter turnaround time (15 min) and can also detect C. gattii, an additional advantage in comparison to other CrAg tests available [33].

Histoplasma takes several weeks to grow on culture. Detection of Histoplasma antigen from serum or urine is non-invasive, and is a rapid test for diagnosis in disseminated histoplasmosis. The sensitivity is highest in disseminated histoplasmosis followed by acute and chronic pulmonary and least in subacute pulmonary histoplasmosis. Rising antigen levels can also be used as early predictors for clinical relapse or treatment failure. A limitation of antigen testing is the significant cross-reactivity of the assay in the presence of other fungal infections, including blastomycosis, paracoccidioido-mycosis, penicilliosis, aspergillosis, and coccidio-idomycosis [34].

Molecular methods are potential alternative options for early diagnosis of IFI, ideal being broad-range/pan-fungal PCR. European Fungal PCR Initiative group (FPCRI) have standardized and validated protocols for Aspergillus PCR which can be used as a screening tool because of its high negative predictive value. SeptiFast PCR, a commercial mutiplex realtime PCR (Roche diagnostics, Germany) available for 20 clinically relevant pathogens including six fungi i.e. five Candida spp and A. fumigatus proved to be helpful where culture was negative. Rate of positivity was 14.6% as compared to culture (10.3%) [35]. PCR positivity depends on site and amount of clinical specimen. The Asper Genius assay detects and differentiates wild type from pathogenic Aspergillus fumigatus with (four) azole resistance associated mutations in the cyp51A gene. FKS1- echinocandin resistance for Candida spp can also be done directly from blood [36].

Recently FDA approved T2 candida method which is a rapid test to detect five species of Candida (C. albicans, C. tropicalis, C. parapsilosis, C. krusei and C. glabrata) directly from blood sample. Principle of the assay is initial nucleic acid extraction amplification and hybridization of the product. The detection limit is 1 CFU/ml (compared 100 -1000 CFU/ml in NAATs). It is fully automated and results provided within 3-5 hours. However, this technique has not yet been evaluated in pediatric population [21].

The role of imaging in invasive fungal infections is multi-dimensional. Imaging helps in identification of the focus of infection, establishing a possible diagnosis, and detection of serial changes. Plain radiographs often prove inconclusive. Other imaging modalities include computed tomography (CT), ultrasound and magnetic resonance imaging (MRI).

Contrast enhanced CT scan remains the most important imaging modality in the diagnosis of invasive fungal infections. Evaluation of the pulmonary pathology requires a reconstruction of images into routine lung window and high resolution lung window images. With the present day fast CT scanners and a volumetric acquisition of data, good quality image acquisition in a very short time frame is made possible. This is especially crucial in acutely ill febrile or dyspneic children. The appearance of new abnormalities on CT chest not responding to broad spectrum antibiotics should be considered as possible IFI in these children.

Detection of any abdominal focus of infection requires a meticulous ultrasound examination with additional high resolution ultrasonography, using high frequency linear transducer of 5-12 MHz, of the solid organs such as liver, spleen and kidneys.Transcranial USG using a small footprint sector probe is useful in neonates and young infants.

MRI is invaluable in the evaluation of infection of the central nervous system and musculoskeletal system. MRI is helpful in follow-up of pulmonary fungal infections as it is non-ionizing radiation, though its sensitivity in the detection of small pulmonary nodules in infection is less than CT.

CNS infections: CNS infections in IFI vary depending on the organism. CNS aspergillosis may be seen on imaging as (i) multiple areas of embolic infarcts secondary to meningitis and involvement of perforating vessels, (ii) multiple ring enhancing lesions or (iii) contiguous involvement of the dura associated with adjacent sinusitis/skull base osteomyelitis [37]. CNS mucor-mycosis is usually associated with sinusitis and bone erosion. Often there is associated cavernous sinus thrombosis and contiguous CNS spread of infection (Web Fig. 1). Disseminated hematogenous candida infection may result in meningoencephalitis. The imaging findings include meningeal enhancement and multiple ring enhancing or nodular lesions in the brain parenchyma.

Screening patients with probable and proven invasive pulmonary aspergillosis by MRI of brain is recommended even in absence of neurologic signs/symptoms. Dissemination of pulmonary aspergillosis to CNS has been reported in 14% and clinical features of CNS infection occur late in the course of disease.

Sinonasal infection (Web Fig. 2): Sinonasal invasive fungal infection on imaging appears as soft tissue opacification of the sinonasal cavity with mucoperiosteal thickening; with variable degree of bony erosion. Skull base osteomyelitis and contiguous CNS spread of infection can occur as a complication.

Pulmonary infection: Pulmonary candidiasis can present as a part of disseminated candidiasis. The imaging features can manifest as lobar consolidation and mimic bacterial pneumonia or may present as multiple nodules similar to aspergillosis. Invasive aspergillosis (Web Fig. 3) usually shows multiple pulmonary nodules often with perinodular ground glass opacity ‘halo sign’. The imaging signs of improvement include cavitation (crescent sign) within the nodule. Pulmonary mucormycosis can present as airspace consolidation or multiple nodules. The nodules may show halo sign or reverse halo sign. Reverse halo sign refers to an area of peripheral consolidation and a central ground glass opacity. Bird’s nest sign is seen as central ground glass opacity with multiple intersecting or irregular lines [38]. Both the signs can be seen in angioinvasive aspergillosis, mucormycosis, cryptogenic organizing pneumonia, Wegener’s granulomatosis and other causes.

Abdominal infection: Disseminated candidemia may show focal lesions in liver, spleen and other abdominal organs. The lesions appear hypoechoic, target lesion or as small abscesses on USG. CT has superior sensitivity than USG in detecting focal hepatosplenic lesions which appear hypodense (Web Fig. 4).

Musculoskeletal: Musculoskeletal involvement in disseminated fungal infection may manifest as osteomyelitis and arthritis. Vertebral and costal　fungal osteomyelitis can develop from contiguous pulmonary infection as in aspergillosis, or by hematogenous dissemination and traumatic inoculation. The imaging findings are similar to other forms of osteomyelitis and include osteopenia, erosions and periosteal reaction.

The risk factors for occurrence of fungal infections must be kept in mind while evaluating a patient of pyrexia of unknown origin or unusual signs/symptoms at usual/unusual sites. If signs /symptoms of localized fungal infection are present (oral ulcers/skin ulcers/pneumonia/sinusitis) sample must be sent as soon as possible for direct microscopy and culture for definitive diagnosis. The treating team must interact with the mycologist and radiologist when evaluating such a case. The investigations must be ordered judiciously and interpreted rationally as definite evidence is often lacking and antifungals have significant side effects and require prolonged administration. They should be chosen depending on the organism identified, host characteristics, site of infection and local epidemiological data of fungal infections including resistance.

A suggested algorithm for evaluation of a suspected case of fungal infection is depicted in Fig 1.

Fig. 1 Evaluation of suspected invasive fungal infection.

Diagnosis of IFIs in children remains challenging. The validation and utility of various currently available fungal diagnostic tools are lacking, particularly in the pediatric group. Although tissue diagnosis remains the gold standard, other tests like Galactomannan assay and PCR can be used as adjunct for diagnosis of IFI in children. Newer methods like T2 candida and lateral flow assay need validation in children with candidemia and invasive aspergillosis. Role of radiology in diagnosing IFI needs further exploration.

1. Rosen GP, Nielsen K, Glenn S, Abelson J, Deville J, Moore TB. Invasive fungal infections in pediatric oncology patients: 11-year experience at a single institution. J Pediatric Hematol Oncol. 2005;27:135-40.

2. Wattier RL, Dvorak CC, Hoffman JA, Brozovich AA, Bin-Hussain I, Groll AH, et al. A prospective, international cohort study of invasive mold infections in children. J Pediatr Infect Dis Soc. 2014;4:313-22.

3. Jordan I, Balaguer M, López-Castilla JD, Belda S, Shuffelman C, Garcia-Teresa MA, et al. for the ERICAP Study Group. Per-species risk factors and predictors of invasive Candida infections in patients admitted to pediatric intensive care units: development of ERICAP scoring systems. Pediatr Infect Dis J. 2014;33:s187-93.

4. Jaworski R, Haponiuk I, Irga-Jaworska N, Chojnicki M, Steffens M, Paczkowski K, et al. Fungal infections in children in the early postoperative period after cardiac surgery for congenital heart disease: A single-centre experience. Interact Cardiovasc Thorac Surg. 2016;23: 431-7.

5. De Pauw B, Walsh TJ, Donnelly JP, Stevens DA, Edwards JE, Calandra T, et al. Revised Definitions of Invasive Fungal Disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) consensus group. Clin Infect Dis. 2008;46:1813-21.

6. W Adilia, Thomas L. Progress in the Diagnosis of Invasive Fungal disease in Children. Curr Fungal Infect Rep. 2017;11:35-44.

7. Mukherjee PK, Sheehan DJ, Hitchcock CA, Ghannoum MA. Combination treatment of invasive fungal infections. Clin Microb Rev. 2005;18:163-94.

8. King J, Pana ZD, Lehrnbecher T, Steinbach WJ, Warris A. Recognition and clinical presentation of invasive fungal disease in neonates and children. J Pediatr Infect Dis Soc. 2017;6:S12-21.

9. Ascioglu S, Rex JH, De Pauw B, Bennett JE, Bille J, Crokaert F, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: An international consensus. Clin Infect Dis. 2002;34:7-14.

10. Oz Y, Kiraz N. Diagnostic methods for fungal infections in pediatric patients: Microbiological, serological and molecular methods. Expert Rev Anti Infect Ther. 2011;9;289-98.

11. Badiee P, Hashemizadeh Z. Opportunistic invasive fungal infections: Diagnosis and clinical management. Indian J Med Res. 2014;139:195-204.

12. Clancy CJ, Nguyen MH. Finding the "missing 50%" of invasive candidiasis: How nonculture diagnostics will improve understanding of disease spectrum and transform patient care. Clin Infect Dis. 2013;56:1284-92.

13. Hope WW, Castagnola E, Groll AH, Roilides E, Akova M, Arendrup MC, et al. ESCMID Guideline for the Diagnosis and Management of Candida Diseases 2012: Prevention and Management of Invasive Infections in Neonates and Children Caused by Candida Spp. Clin Microbiol Infect. 2012;18:38-52.

14. Kontoyiannis DP, Sumoza D, Tarrand J, Bodey GP, Storey R, Raad II. Significance of aspergillemia in patients with cancer: A 10-year study. Clin Infect Dis. 2000;31:188-9.

15. Lass-Florl C, Resch G, Nachbaur D, Mayr A, Gastl G, Auberger J, et al. The value of computed tomography-guided percutaneous lung biopsy for diagnosis of invasive fungal infection in immunocompromised patients. Clin Infect Dis. 2007;45:e101-4.

16. Hoenigl M, Prattes J, Spiess B, Wagner J, Prueller F, Raggam RB, et al. Performance of galactomannan, beta-d-glucan, aspergillus lateral-flow device, conventional culture, and PCR tests with bronchoalveolar lavage fluid for diagnosis of invasive pulmonary aspergillosis. J Clin Microbiol. 2014;52:2039-45.

17. Skiada A, Lass-Floerl C, Klimko N, Ibrahim A, Roilides E, Petrikkos G. Challenges in the diagnosis and treatment of mucormycosis. Med Mycol. 2018;56:93-101.

18. Klont RR, Mennink-Kersten MA, Verweij PE. Utility of aspergillus antigen detection in specimens other than serum specimens. Clin Infect Dis. 2004;39:1467-74.

19. Leeflang MM, Debets Ossenkopp YJ, Wang J, Visser CE, Scholten RJ, Hooft L, et al. Galactomannan detection for invasive aspergillosis in immunocompromised patients. Cochrane Database of Syst Rev. 2015;4:CD007394.

20. Groll AH, Castagnola E, Cesaro S, Dalle JH, Engelhard D, Hope W, et al. Fourth European conference on infections in leukaemia (ECIL-4): Guidelines for diagnosis, prevention, and treatment of invasive fungal diseases in paediatric patients with cancer or allogeneic haemopoietic stem-cell transplantation. Lancet Oncol. 2014;15:e327-40.

21. Arvanitis M, Anagnostou T, Fuchs BB, Caliendo AM, Mylonakis E. Molecular and nonmolecular diagnostic methods for invasive fungal infections. Clin Microbiol Rev. 2014;27:490-526.

22. Mennink-Kersten MA, Ruegebrink D, Klont RR, Warris A, Gavini F, Op den Camp HJ, et al. Bifidobacterial lipoglycan as a new cause for false-positive platelia Aspergillus enzyme-linked immunosorbent assay reactivity. J Clin Microbiol. 2005;43:3925-31.

23. Pana ZD, Vikelouda K, Roilides E. Diagnosis of invasive fungal diseases in pediatric patients. Expert Rev Anti Infect Ther. 2016;14:1203-13.

24. Thornton CR. Development of an immuno-chromatographic lateral-flow device for rapid serodiagnosis of invasive aspergillosis. Clin Vaccine Immunol. 2008;15:1095-105.

25. Pan Z, Fu M, Zhang J, Zhou H, Fu Y, Zhou J. Diagnostic accuracy of a novel lateral-flow device in invasive aspergillosis: a meta-analysis. J Med Microbiol. 2015;64:702-7.

26. Desmet S, Van Wijngaerden E, Maertens J, Verhaegen J, Verbeken E, De Munter P, et al. Serum (1-3)-beta-D-glucan as a tool for diagnosis of Pneumocystis jirovecii pneumonia in patients with human immunodeficiency virus infection or hematological malignancy. J Clin Microbiol. 2009;47:3871-4.

27. Mennink-Kersten MA, Ruegebrink D, Verweij PE. Pseudomonas aeruginosa as a cause of 1,3-beta-D-glucan assay reactivity. Clin Infect Dis. 2008;46:1930-1.

28. Marty FM, Lowry CM, Lempitski SJ, Kubiak DW, Finkelman MA, Baden LR. Reactivity of (1->3)-beta-d-glucan assay with commonly used intravenous antimicro-bials. Antimicrob Agents Chemother. 2006;50:3450-3.

29. Usami M, Ohata A, Horiuchi T, Nagasawa K, Wakabayashi T, Tanaka S. Positive (1->3)-beta-D-glucan in blood components and release of (1->3)-beta-D-glucan from depth-type membrane filters for blood processing. Transfus. 2002;42:1189-95.

30. Smith PB, Benjamin DK, Jr., Alexander BD, Johnson MD, Finkelman MA, Steinbach WJ. Quantification of 1,3-beta-D-glucan levels in children: preliminary data for diagnostic use of the beta-glucan assay in a pediatric setting. Clin Vaccine Immunol. 2007;14:924-5.

31. Kumar J, Singh A, Seth R, Xess I, Jana M, Kabra SK. Prevalence and predictors of invasive fungal infections in children with persistent febrile neutropenia treated for acute leukemia – A prospective study. Indian J Pediatr. 2018;85:1090-5.

32. Mikulska M, Calandra T, Sanguinetti M, Poulain D, Viscoli C. The use of mannan antigen and anti-mannan antibodies in the diagnosis of invasive candidiasis: recommendations from the third European conference on infections in leukemia. Crit Care. 2010;14:R222.

33. Vidal JE, Boulware DR. Lateral flow assay for cryptococcal antigen: An important advance to improve the continuum of HIV care and reduce cryptococcal meningitis-related mortality. Rev Inst Med Trop Sao Paulo. 2015;57:38-45.

34. Azar MM, Hage CA. Laboratory diagnostics for histoplasmosis. J Clin Microbiol. 2017;55:1612-20.

35. Warhurst G, Maddi S, Dunn G, Ghrew M, Chadwick P, Alexander P, et al. Diagnostic accuracy of SeptiFast multi-pathogen real-time PCR in the setting of suspected healthcare-associated bloodstream infection. Intensive Care Med. 2015;41:86-93.

36. Chong GL, van de Sande WW, Dingemans GJ, Gaajetaan GR, Vonk AG, Hayette MP, et al. Validation of a new Aspergillus real-time PCR assay for direct detection of Aspergillus and azole resistance of Aspergillus fumigatus on bronchoalveolar lavage fluid. J Clin Microbiol. 2015;53:868-74.

37. Ashdown BC, Tien RD, Felsberg GJ. Aspergillosis of the brain and paranasal sinuses in immunocompromised patients: CT and MR imaging findings. Am J Roentgenol. 1994;162:155-9.